healthy lifestyle research paper conclusion

• Research article
• Open access
• Published: 29 September 2022

A healthy lifestyle is positively associated with mental health and well-being and core markers in ageing

• Pauline Hautekiet   ORCID: orcid.org/0000-0003-3805-3004 1 , 2 ,
• Nelly D. Saenen 1 , 2 ,
• Dries S. Martens 2 ,
• Margot Debay 2 ,
• Johan Van der Heyden 3 ,
• Tim S. Nawrot 2 , 4 &
• Eva M. De Clercq 1  

BMC Medicine volume  20 , Article number:  328 ( 2022 ) Cite this article

12k Accesses

17 Citations

61 Altmetric

Metrics details

Studies often evaluate mental health and well-being in association with individual health behaviours although evaluating multiple health behaviours that co-occur in real life may reveal important insights into the overall association. Also, the underlying pathways of how lifestyle might affect our health are still under debate. Here, we studied the mediation of different health behaviours or lifestyle factors on mental health and its effect on core markers of ageing: telomere length (TL) and mitochondrial DNA content (mtDNAc).

In this study, 6054 adults from the 2018 Belgian Health Interview Survey (BHIS) were included. Mental health and well-being outcomes included psychological and severe psychological distress, vitality, life satisfaction, self-perceived health, depressive and generalised anxiety disorder and suicidal ideation. A lifestyle score integrating diet, physical activity, smoking status, alcohol consumption and BMI was created and validated. On a subset of 739 participants, leucocyte TL and mtDNAc were assessed using qPCR. Generalised linear mixed models were used while adjusting for a priori chosen covariates.

The average age (SD) of the study population was 49.9 (17.5) years, and 48.8% were men. A one-point increment in the lifestyle score was associated with lower odds (ranging from 0.56 to 0.74) for all studied mental health outcomes and with a 1.74% (95% CI: 0.11, 3.40%) longer TL and 4.07% (95% CI: 2.01, 6.17%) higher mtDNAc. Psychological distress and suicidal ideation were associated with a lower mtDNAc of − 4.62% (95% CI: − 8.85, − 0.20%) and − 7.83% (95% CI: − 14.77, − 0.34%), respectively. No associations were found between mental health and TL.

Conclusions

In this large-scale study, we showed the positive association between a healthy lifestyle and both biological ageing and different dimensions of mental health and well-being. We also indicated that living a healthy lifestyle contributes to more favourable biological ageing.

Peer Review reports

According to the World Health Organization (WHO), a healthy lifestyle is defined as “a way of living that lowers the risk of being seriously ill or dying early” [ 1 ]. Public health authorities emphasise the importance of a healthy lifestyle, but despite this, many individuals worldwide still live an unhealthy lifestyle [ 2 ]. In Europe, 26% of adults smoke [ 3 ], nearly half (46%) never exercise [ 4 ], 8.4% drink alcohol on a daily basis [ 5 ] and over half (51%) are overweight [ 5 ]. These unhealthy behaviours have been associated with adverse health outcomes like cardiovascular diseases [ 6 , 7 , 8 ], respiratory diseases [ 9 ], musculoskeletal diseases [ 10 ] and, to a lesser extent, mental disorders [ 11 , 12 ].

Even though the association between lifestyle and health outcomes has been extensively investigated, biological mechanisms explaining these observed associations are not yet fully understood. One potential mechanism that can be suggested is biological ageing. Both telomere length (TL) and mitochondrial DNA content (mtDNAc) are known biomarkers of ageing. Telomeres are the end caps of chromosomes and consist of multiple TTAGGG sequence repeats. They protect chromosomes from degradation and shorten with every cell division because of the “end-replication problem” [ 13 ]. Mitochondria are crucial to the cell as they are responsible for apoptosis, the control of cytosolic calcium levels and cell signalling [ 14 ]. Living a healthy lifestyle can be linked with healthy ageing as both TL and mtDNAc have been associated with health behaviours like obesity [ 15 ], diet [ 16 ], smoking [ 17 ] and alcohol abuse [ 18 ]. Furthermore, as biomarkers of ageing, both TL and mtDNAc have been associated with age-related diseases like Parkinson’s disease [ 19 ], coronary heart disease [ 20 ], atherosclerosis [ 21 ] and early mortality [ 22 ]. Also, early mortality and higher risks for the aforementioned age-related diseases are observed in psychiatric illnesses, and it is suggested that advanced biological ageing underlies these observations [ 23 ].

Multiple studies evaluated individual health behaviours, but research on the combination of these health behaviours is limited. As they often co-occur and may cause synergistic effects, assessing them in combination with each other rather than independently might better reflect the real-life situation [ 24 , 25 ]. Therefore, in a general adult population, we combined five commonly studied health behaviours including diet, smoking status, alcohol consumption, BMI and physical activity into one healthy lifestyle score to evaluate its association with mental health and well-being and biological ageing. Furthermore, we evaluated the association between the markers of biological ageing and mental health and well-being. We hypothesise that individuals living a healthy lifestyle have a better mental health status, a longer TL and a higher mtDNAc and that these biomarkers are positively associated with mental health and well-being.

Study population

In 2018, 11611 Belgian residents participated in the 2018 Belgian Health Interview Survey (BHIS). The sampling frame of the BHIS was the Belgian National Register, and participants were selected based on a multistage stratified sampling design including a geographical stratification and a selection of municipalities within provinces, of households within municipalities and of respondents within households [ 26 ]. The study population for this cross-sectional study included 6054 BHIS participants (see flowchart in Additional file 1 : Fig. S1) [ 27 , 28 , 29 , 30 , 31 ]. Minors (< 18 years) and participants not eligible to complete the mental health modules (participants who participated through a proxy respondent, i.e. a person of confidence filled out the survey) were excluded ( n  = 2172 and n  = 846, respectively). Furthermore, of the 8593 eligible participants, those with missing information to create the mental health indicators, the lifestyle score or the covariates used in this study were excluded ( n  = 1642, 788 and 109, respectively).

For the first time in 2018, a subset of 1184 BHIS participants contributed to the 2018 Belgian Health Examination Survey (BELHES). All BHIS participants were invited to participate except for minors (< 18 years), BHIS participants who participated through a proxy respondent and residents of the German Community of Belgium, the latter representing 1% of the Belgian population. Participants were recruited on a voluntary basis until the regional quotas were reached (450, 300 and 350 in respectively Flanders, Brussels Capital Region and Wallonia). These participants underwent a health examination, including anthropological measurements and completed an additional questionnaire. Also, blood and urine samples were collected. Of the 6054 included BHIS participants, 909 participated in the BELHES. Participants for whom we could not calculate both TL and mtDNAc were excluded ( n  = 170). More specifically, participants were excluded because they did not provide a blood sample ( n  = 91) or because they did not provide permission for DNA research ( n  = 32). Twenty samples were excluded from DNA extraction because either total blood volume was too low ( n  = 7), samples were clothed ( n  = 1) or tubes were broken due to freezing conditions ( n  = 12). Twenty-seven samples were excluded because they did not meet the biomarker quality control criteria (high technical variation in qPCR triplicates). This was not met for 3 TL samples, 20 mtDNAc samples and 4 samples for both biomarkers. For this subset, we ended up with a final number of 739 participants. Further in this paper, we refer to “the BHIS subset” for the BHIS participants ( n  = 6054) and the “BELHES subset” for the BELHES participants ( n  = 739).

As part of the BELHES, this project was approved by the Medical Ethics Committee of the University Hospital Ghent (registration number B670201834895). The project was carried out in line with the recommendations of the Belgian Privacy Commission. All participants have signed a consent form that was approved by the Medical Ethics Committee.

Health interview survey

The BHIS is a comprehensive survey which aims to gain insight into the health status of the Belgian population. The questions on the different dimensions of mental health and well-being were based on international standardised and validated questionnaires [ 32 ], and this resulted in eight mental health outcomes that were used in this study. Detailed information on each indicator score and its use is addressed in Additional file 1 : Table. S1. Firstly, the General Health Questionnaire (GHQ-12) provides the prevalence of psychological and severe psychological distress in the population [ 27 ]. On the total GHQ score, cut-off points of + 2 and + 4 were used to identify respectively psychological and severe psychological distress.

Secondly, we used two indicators for the positive dimensions of mental health: vitality and life satisfaction. Four questions of the short form health survey (SF-36) indicate the participant’s vital energy level [ 28 , 33 ]. We used a cut-off point to identify participants with an optimal vitality score, which is a score equal to or above one standard deviation above the mean, as used in previous studies [ 34 , 35 ]. Life satisfaction was measured by the Cantril Scale, which ranges from 0 to 10 [ 29 ]. A cut-off point of + 6 was used to indicate participants with high or medium life satisfaction versus low life satisfaction.

Thirdly, the question “How is your health in general? Is it very good, good, fair, bad or very bad?” was used to assess self-perceived health, also known as self-rated health. Based on WHO recommendations [ 36 ], the answer categories were dichotomised into “good to very good self-perceived health” and “very bad to fair self-perceived health”.

Fourthly, depressive and generalised anxiety disorders were defined using respectively the Patient Health Questionnaire (PHQ-9) and the Generalised Anxiety Disorder Questionnaire (GAD-7). We identified individuals who suffer from major depressive syndrome or any other type of depressive syndrome according to the criteria of the PHQ-9 [ 37 ]. A cut-off point of + 10 on the total sum of the GAD-7 score was used to indicate generalised anxiety disorder [ 31 ]. Additionally, a dichotomous question on suicidal ideation was used: “Have you ever seriously thought of ending your life?”; “If yes, did you have such thoughts in the past 12 months?”. Finally, the BHIS also includes personal, socio-economic and lifestyle information. The standardised Cronbach’s alpha coefficients for the PHQ-9, GHQ-12, GAD-7 and questions on vitality of the SF-36 ranged between 0.80 and 0.90.

Healthy lifestyle score

We developed a healthy lifestyle score based on five different health behaviours: body mass index (BMI), smoking status, physical activity, alcohol consumption and diet (Table 1 ). These health behaviours were defined as much as possible according to the existing guidelines for healthy living issued by the Belgian Superior Health Council [ 38 ] and the World Health Organisation [ 39 , 40 , 41 ]. Firstly, BMI was calculated as a person’s self-reported weight in kilogrammes divided by the square of the person’s self-reported height in metres (kg/m 2 ). BMI was classified into four categories: underweight (BMI < 18.5 kg/m 2 ), normal weight (BMI 18.5–24.9 kg/m 2 ), overweight (BMI 25.0–29.9 kg/m 2 ) and obese (BMI ≥ 30.0 kg/m 2 ). Due to a J-shaped association of BMI with the overall mortality and multiple specific causes of death, obesity and underweight were both classified as least healthy [ 42 ]. BMI was scored as follows: obese and underweight = 0, overweight = 1 and normal weight = 2.

Secondly, smoking status was divided into four categories. Participants were categorised as regular smokers if they smoked a minimum of 4 days per week or if they quit smoking less than 1 month before participation (= 0). Occasional smokers were defined as smoking more than once per month up to 3 days per week (= 1). Participants were classified as former smokers if they quit smoking at least 1 month before the questionnaire or if they smoked less than once a month (= 2). The final category included people who never smoked (= 3).

Thirdly, physical activity was assessed by the question: “What describes best your leisure time activities during the last year?”. Four categories were established and scored as follows: sedentary activities (= 0), light activities less than 4 h/week (= 1), light activities more than 4 h/week or recreational sport less than 4 h/week (= 2) and recreational sport more than 4 h or intense training (= 3). Fourthly, information on the number of alcoholic drinks per week was used to categorise alcohol consumption. The different categories were set from high to low alcohol consumption: 22 drinks or more/week (= 0), 15–21 drinks/week (= 1), 8–14 drinks/week (= 2), 1–7 drinks/week (= 3)and less than 1 drink/week (= 4).

Finally, in line with the research by Benetou et al., a diet score was calculated using the frequency of consuming fruit, vegetables, snacks and sodas [ 43 ]. For fruit as well as vegetable consumption, the frequency was scored as follows: never (= 0), < 1/week (= 1), 1–3/week (= 2), 4–6/week (= 3) and ≥ 1/day (= 4). The frequency of consuming snacks and sodas was scored as follows: never (= 4), < 1/week (= 3), 1–3/week (= 2), 4–6/week (= 1) and ≥ 1/day (= 0). The diet score was then divided into tertiles, in line with the research by Benetou et al. [ 43 ]. A diet score of 0–9 points was classified as the least healthy behaviour (= 0). A diet score ranging from 10 to 12 made up the middle category (= 1), and a score from 13 to 16 was classified as the healthiest behaviour (= 2).

All five previously described health behaviours were combined into one healthy lifestyle score (Table 1 ). The sum of the scores obtained for each health behaviour indicated the absolute lifestyle score. To calculate the relative lifestyle score, each absolute scored health behaviour was given equal weight by recalculating its maximum absolute score to a relative score of 1. The relative lifestyle scores were then summed up to achieve a final continuous lifestyle score, ranging from 0 to 5, with a higher score representing a healthier lifestyle.

Telomere length and mitochondrial DNA content assay

Blood samples were collected during the BELHES and centrifuged for 15 min at 3000 rpm before storage at − 80 °C. After extracting the buffy coat from the blood sample, DNA was isolated using the QIAgen Mini Kit (Qiagen, N.V.V Venlo, The Netherlands). The purity and quantity of the sample were measured with a NanoDrop spectrophotometer (ND-2000; Thermo Fisher Scientific, Wilmington, DE, USA). DNA integrity was assessed by agarose gel electrophoresis. To ensure a uniform DNA input of 6 ng for each qPCR reaction, samples were diluted and checked using the Quant-iT™ PicoGreen® dsDNA Assay Kit (Life Technologies, Europe).

Relative TL and mtDNAc were measured in triplicate using a previously described quantitative real-time PCR (qPCR) assay with minor modifications [ 44 , 45 ]. All reactions were performed on a 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) in a 384-well format. Used telomere, mtDNAc and single copy-gene reaction mixtures and PCR cycles are given in Additional file 1 : Text. S1. Reaction efficiency was assessed on each plate by using a 6-point serial dilution of pooled DNA. Efficiencies ranged from 90 to 100% for single-copy gene runs, 100 to 110% for telomere runs and 95 to 105% for mitochondrial DNA runs. Six inter-run calibrators (IRCs) were used to account for inter-run variability. Also, non-template controls were used in each run. Raw data were processed and normalised to the reference gene using the qBase plus software (Biogazelle, Zwijnaarde, Belgium), taking into account the run-to-run differences.

Leucocyte telomere length was expressed as the ratio of telomere copy number to single-copy gene number (T/S) relative to the mean T/S ratio of the entire study population. Leucocyte mtDNAc was expressed as the ratio of mtDNA copy number to single-copy gene number (M/S) relative to the mean M/S ratio of the entire study population. The reliability of our assay was assessed by calculating the interclass correlation coefficient (ICC) of the triplicate measures (T/S and M/S ratios and T, M and S separately) as proposed by the Telomere Research Network, using RStudio version 1.1.463 (RStudio PBC, Boston, MA, USA). The intra-plate ICCs of T/S ratios, TL runs, M/S ratios, mtDNAc runs and single-copy runs were respectively 0.804 ( p  < 0.0001), 0.907 ( p  < 0.0001), 0.815 ( p  < 0.0001), 0.916 ( p  < 0.0001) and 0.781 ( p  < 0.0001). Based on the IRCs, the inter-plate ICC was 0.714 ( p  < 0.0001) for TL and 0.762 ( p  < 0.0001) for mtDNAc.

Statistical analysis

Statistical analyses were performed using the SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA). We performed a log(10) transformation of the TL and mtDNAc data to reduce skewness and to better approximate a normal distribution. Three analyses were done: (1) In the BHIS subset ( n  = 6054), we evaluated the association between the lifestyle score and the mental health and well-being outcomes (separately). These results are presented as the odds ratio (95% CI) of having a mental health condition or disorder for a one-point increment in the lifestyle score. (2) In the BELHES subset ( n  = 739), we evaluated the association between the lifestyle score and both TL and mtDNAc (separately). These results are presented as the percentage difference in TL or mtDNAc (95% CI) for a one-point increment in the lifestyle score. (3) In the BELHES subset ( n  = 739), we evaluated the association between the mental health and well-being outcomes and both TL and mtDNAc (separately). These results are presented as the percentage difference in TL or mtDNAc (95% CI) when having a mental health condition or disorder compared with the healthy group.

For all three analyses, we performed multivariable linear mixed models (GLIMMIX; unstructured covariance matrix) taking into account a priori selected covariates including age (continuous), sex (male, female), region (Flanders, Brussels Capital Region, Wallonia), highest educational level of the household (up to lower secondary, higher secondary, college or university), country of birth (Belgium, EU, non-EU) and household type (single, one parent with child, couple without child, couple with child, others). To capture the non-linear effect of age, we included a quadratic term when the result of the analysis showed that both the linear and quadratic terms had a p -value < 0.1. For the two analyses on TL and mtDNAc, we additionally adjusted for the date of participation in the BELHES. As multiple members of one household participated, we added household numbers in the random statement.

Bivariate analyses evaluating the associations between the characteristics and TL, mtDNAc, the lifestyle score or psychological distress as a parameter of mental health and well-being are evaluated based on the same model. The chi-squared tests (categorical data) and t -tests (continuous data) were used to evaluate the characteristics of included and excluded participants. The lifestyle score was validated by creating a ROC curve and calculating the area under the curve (AUC) of the adjusted association between the lifestyle score and self-perceived health. Adjustments were made for age, sex, region, highest educational level of the household, country of birth and household type.

In a sensitivity analysis, to evaluate the robustness of our findings, we additionally adjusted our main models separately for perceived quality of social support (poor, moderate, strong) and chronic disease (suffering from any chronic disease or condition: yes, no). The third model, evaluating the biomarkers with the mental health outcomes, was also additionally adjusted for the lifestyle score.

Population characteristics

The characteristics of the BHIS and BELHES subset are presented in Table 2 . In the BHIS subset, 48.8% of the participants were men. The average age (SD) was 49.9 (17.5) years, and most participants were born in Belgium (79.5%). The highest educational level in the household was most often college or university degree (53.3%), and the most common household composition was couple with child(ren) (37.7%). The proportion of participants in different regions of Belgium, i.e. Flanders, Brussels Capital Region and Wallonia, was respectively 41.1%, 23.3% and 35.6%. For the BELHES subset, we found similar results except for region and education. We noticed more participants from Flanders and more participants with a high educational level in the household. The mean (SD) relative TL and mtDNAc were respectively 1.04 (0.23) and 1.03 (0.24). TL and mtDNAc were positively correlated (Spearman’s correlation = 0.21, p  < 0.0001).

We compared (1) the characteristics of the 6054 eligible BHIS participants that were included in the BHIS subset with the 2539 eligible participants that were excluded from the BHIS subset (Additional file 1 : Table S2) and (2) the 739 participants from the BHIS subset that were included in the BELHES subset with the 5315 participants that were excluded from the BELHES subset (Additional file 1 : Table S3). Except for sex and nationality in the latter, all other covariates showed differences between the included and excluded groups. On the other hand, population data from 2018 indicates that the average age (SD) of the adult Belgian population was 49.5 (18.9) with a distribution over Flanders, Brussels Capital Region and Wallonia of respectively 58.2%, 10.2% and 31.6% and that 48.7% were men. The distribution of our sample according to age and sex thus largely corresponds to the age and sex distribution of the adult Belgian population figures. The large difference in the regional distribution is due to the oversampling of the Brussels Capital Region in the BHIS.

Bivariate associations evaluating the characteristics with TL, mtDNAc, the lifestyle score or psychological distress as a parameter of mental health are presented in Additional file 1 : Table S4. Briefly, men had a − 6.41% (95% CI: − 9.10 to − 3.65%, p  < 0.0001) shorter TL, a − 8.03% (95% CI: − 11.00 to − 4.96%, p  < 0.0001) lower mtDNAc, lower odds of psychological distress (OR = 0.59, 95% CI: 0.53 to 0.66, p  < 0.0001) and a lifestyle score of − 0.28 (95% CI: − 0.32 to − 0.24, p  < 0.0001) points less compared with women. Furthermore, a 1-year increment in age was associated with a − 0.64% (− 0.73 to − 0.55%, p  < 0.0001) shorter TL and a − 0.19% (95% CI: − 0.31 to − 0.08%, p  = 0.00074) lower mtDNAc.

Mental health prevalence and lifestyle characteristics

Within the BHIS subset, 32.3% and 18.0% of the participants had respectively psychological and severe psychological distress. 86.7% had suboptimal vitality, 12.0% indicated low life satisfaction and 22.0% had very bad to fair self-perceived health. The prevalence of depressive and generalised anxiety disorders was respectively 9.0% and 10.8%, respectively. 4.4% of the participants indicated to have had suicidal thoughts in the past 12 months. Similar results were found for the BELHES subset (Table 3 ).

Within the BHIS subset, the average lifestyle score (SD) was 3.1 (0.9) (Table 4 ). A histogram of the lifestyle score is shown in Additional file 1 : Fig. S2. 16.6% were regular smokers, and 4.9% reported 22 alcoholic drinks per week or more. 29.7% reported that their main leisure time included mainly sedentary activities, and 18.6% were underweight or obese. 29.2% were classified as having an unhealthy diet score. The participants of the BELHES subset were slightly more active, but no other dissimilarities were found (Table 4 ). The ROC curve shows an area under the curve (AUC) of 0.74, indicating a 74% predictive accuracy for the lifestyle score as a self-perceived health predictor (Additional file 1 : Fig. S3).

Healthy lifestyle and mental health and well-being

Living a healthier lifestyle, indicated by having a higher lifestyle score, was associated with lower odds of all mental health and well-being outcomes (Table 5 ). After adjustment, a one-point increment in the lifestyle score was associated with lower odds of psychological (OR = 0.74, 95% CI: 0.69, 0.79) and severe psychological distress (OR = 0.69, 95% CI: 0.64, 0.75). Similarly, for the same increment, the odds of suboptimal vitality, low life satisfaction and very bad to fair self-perceived health were respectively 0.62 (95% CI: 0.56, 0.68), 0.62 (95% CI: 0.56, 0.68) and 0.56 (95% CI: 0.52, 0.61). Finally, the odds of having depressive disorder, generalised anxiety disorder or suicidal ideation were respectively 0.57 (95% CI: 0.51, 0.63), 0.63 (95% CI: 0.57, 0.69) and 0.63 (95% CI: 0.55, 0.72) for a one-point increment in the lifestyle score.

The biomarkers of ageing

After adjustment, living a healthy lifestyle was positively associated with both TL and mtDNAc (Table 6 ). A one-point increment in the lifestyle score was associated with a 1.74 (95% CI: 0.11, 3.40%, p  = 0.037) higher TL and a 4.07 (95% CI: 2.01, 6.17%, p  = 0.00012) higher mtDNAc.

People suffering from severe psychological distress had a − 4.62% (95% CI: − 8.85, − 0.20%, p  = 0.041) lower mtDNAc compared with those who did not suffer from severe psychological distress. Similarly, people with suicidal ideation had a − 7.83% (95% CI: − 14.77, − 0.34%, p  = 0.041) lower mtDNAc compared with those without suicidal ideation. No associations were found for the other mental health and well-being outcomes, and no associations were found between mental health and TL (Table 6 ).

Sensitivity analysis

Additional adjustment of the main analyses for perceived quality of social support, chronic disease or lifestyle score (in the association between the mental health outcomes and the biomarkers of ageing) did not strongly change the effect of our observations (Additional file 1 : Tables S5-S7). However, we noticed that most of the associations between severe psychological distress or suicidal ideation and mtDNAc showed marginally significant results.

In this study, we evaluated the associations between eight mental health and well-being outcomes, a healthy lifestyle score and 2 biomarkers of biological ageing: telomere length and mitochondrial DNA content. Having a healthy lifestyle was positively associated with all mental health and well-being indicators and the markers of biological ageing. Furthermore, having had suicidal ideation or suffering from severe psychological distress was associated with a lower mtDNAc. However, no association was found between mental health and TL.

In the first part of this research, we evaluated the association between lifestyle and mental health and well-being and showed that living a healthy lifestyle was positively associated with better mental health and well-being outcomes. Similar trends were found in previous studies for each of the health behaviours separately [ 11 , 12 , 46 , 47 , 48 ]. Although evaluating these health behaviours separately provides valuable information, assessing them in combination with each other rather than independently might better reflect the real-life situation as they often co-occur and may exert a synergistic effect on each other [ 24 , 25 , 49 ]. For example, 68% of the adults in England engaged in two or more unhealthy behaviours [ 25 ]. Especially, smoking status and alcohol consumption co-occurred, but half of the studies in the review by Noble et al. indicated clustering of all included health behaviours [ 24 ].

To date, the number of studies evaluating the combination of multiple health behaviours and mental health and well-being in adults is limited, and most of them use a different methodology to assess this association [ 50 , 51 , 52 , 53 , 54 , 55 , 56 ]. Firstly, differences are found between the included health behaviours. Most studies included the four “SNAP” risk factors, i.e. smoking, poor nutrition, excess alcohol consumption and physical inactivity. Other health behaviours that were sometimes included were BMI/obesity, sleep duration/quality and psychological distress [ 50 , 53 , 54 , 56 ]. Secondly, differences are found in the scoring of the health behaviours and the use of the lifestyle score. Whereas in this study the health behaviours were scored categorically, studies often dichotomised the health behaviours and/or the final lifestyle score [ 50 , 52 , 53 , 56 ]. Also, two studies performed clustering [ 54 , 55 ]. Health behaviours can cluster together at both ends of the risk spectrum, but less is known about the middle categories. This is avoided by using the cluster method where participants are clustered based on similar behaviours. On the other hand, a lifestyle score can be of better use and more easily be interpreted when aiming to compare healthy versus unhealthy lifestyles, as was the case for this study.

Despite these different methods, all previously mentioned studies show similar results. Together with our findings, which also support these results, this provides clear evidence that an unhealthy lifestyle is associated with poor mental health and well-being outcomes. Important to notice is that, like our research, most studies in this field have a cross-sectional design and are therefore not able to assume causality. Therefore, mental health might be the cause or the consequence of an unhealthy lifestyle. Further prospective and longitudinal studies are warranted to confirm the direction of the association.

Healthy lifestyle and biomarkers of ageing

How lifestyle affects our health is not yet fully understood. One possible pathway is through oxidative stress and biological ageing. An unhealthy lifestyle has been associated with an increase in oxidative stress [ 57 , 58 , 59 ], and in turn, higher concentrations of oxidative stress are known to negatively affect TL and mtDNAc [ 60 ]. In this study, we showed that living a healthy lifestyle was associated with a longer TL and a higher mtDNAc. Our results showed a stronger association of lifestyle with mtDNAc compared with TL. TL is strongly determined by TL at birth [ 61 ]. On the other hand, mtDNAc might be more variable in shorter time periods. Although mtDNAc and TL were strongly correlated, this could explain why lifestyle is more strongly associated with mtDNAc. However, we can only speculate about this, and further research is necessary to confirm our results.

Similar as for the association with mental health, in previous studies, the biomarkers have been associated with health behaviours separately rather than combined [ 62 , 63 , 64 , 65 ]. To our knowledge, we are the first to evaluate the associations between a healthy lifestyle score and mtDNAc. Our results are in line with our expectations. As TL and mtDNAc are known to be correlated [ 60 ], we would expect similar trends for both biomarkers. In the case of TL, few studies included a combined lifestyle score in association with this biomarker. Consistent with our results, in a study population of 1661 men, the sum score of a healthier lifestyle was correlated with a longer TL [ 66 ]. Similar results were found by Sun et al. where a combination of healthy lifestyles in a female study population was associated with a longer TL compared with the least healthy group [ 67 ]. Also, improvement in lifestyle has been associated with TL maintenance in the elderly at risk for dementia [ 68 ], and a lifestyle intervention programme was positively associated with leucocyte telomere length in children and adolescents [ 69 ]. These results suggest that on a biological level, a healthy lifestyle is associated with healthy ageing. Within this context, a study on adults aged 60 and older showed that maintaining a normal weight, not smoking and performing regular physical activity were associated with slower development of disability and a reduction in mortality [ 70 ]. Similarly, midlife lifestyle factors like non-smoking, higher levels of physical activity, non-obesity and good social support have been associated with successful ageing, 22 years later [ 71 ].

Mental health and well-being and biomarkers of ageing

Finally, we evaluated the association between the biomarkers of ageing and the mental health and well-being outcomes. The hypothesis that biological ageing is associated with mental health has been supported by observations showing that chronically stressed or psychiatrically ill persons have a higher risk for age-related diseases like dementia, diabetes and hypertension [ 23 , 72 , 73 ]. Important to notice is that, like our research, the majority of studies on this topic have a cross-sectional design and therefore are unable to identify causality. Therefore, it is currently unknown whether psychological diseases accelerate biological ageing or whether biological ageing precedes the onset of these diseases [ 74 ].

Our results showed a lower mtDNAc for individuals with suicidal ideation or severe psychological distress but not for any of the other mental health outcomes. Evidence on the association between mtDNAc and mental health is inconsistent. Women above 60 years old with depression had a significantly lower mtDNAc compared with the control group [ 75 ]. Furthermore, individuals with a low mtDNAc had poorer outcomes in terms of self-rated health [ 76 ]. In contrast, Otsuka et al. showed a higher peripheral blood mtDNAc in suicide completers [ 77 ], and studies on major depressive syndrome [ 78 ] and self-rated health [ 79 ] showed the same trend. Finally, Vyas et al. showed no significant association between mtDNAc and depression status in mid-life and older adults [ 80 ]. These differences might be due to the differences in study population and methods. For example, the two studies indicating lower mtDNAc in association with poor mental health both had an elderly study population, and one study population consisted of psychiatrically ill patients. Next to that, differences were found in the type of samples, mtDNAc assays and questionnaires or diagnostics. The inconsistency of these studies and our results calls for further research on this association and for standardisation of methods within studies to enable clear comparisons.

As for TL, we did not find an association with any of the mental health and well-being outcomes. Previous studies in adults showed a lower TL in association with current but not remitted anxiety disorder [ 81 ], depressive [ 82 ] and major depressive disorder [ 73 , 83 ], childhood trauma [ 84 ] suicide [ 77 , 85 ], depressive symptoms in younger adults [ 86 ] and acculturative stress and postpartum depression in Latinx women [ 87 ]. Also, in a meta-analysis, psychiatric disorders overall were associated with a shorter leucocyte TL [ 88 ]. However, other studies failed to demonstrate an association between TL and mental health outcomes like major depressive disorder [ 89 ], late-life depression [ 90 ] and anxiety disorder [ 91 ]. Again, this could be due to a different method to assess the mental health outcomes, a different study design, uncontrolled confounding factors and the type of telomere assay. For example, a meta-analysis showed stronger associations with depression when using southern blot or FISH assay compared with qPCR to measure telomere length [ 92 ].

Strengths and limitations

An important strength of this study is the use of a validated lifestyle score that can easily be reproduced and used for other research on lifestyle. Secondly, we were able to use a large sample size for our analyses in the BHIS subset. Thirdly, by assessing multiple dimensions of mental health and well-being, we were able to give a comprehensive overview of the mental health status. To our knowledge, we are the first to evaluate the associations between a healthy lifestyle score and mtDNAc.

Our results should however be interpreted with consideration for some limitations. As mentioned before, the study has a cross-sectional design, and therefore, we cannot assume causality. Secondly, for the lifestyle score, we used self-reported data, which might not always represent the actual situation. For example, BMI values tend to be underestimated due to the overestimation of height and underestimation of weight [ 93 ], and also, smoking behaviour is often underestimated [ 94 ]. Also, equal weights were used for each of the health behaviours as no objective information was available on which weight should be given to a specific health behaviour. Thirdly, there is a distinct time lag between the completion of the BHIS questionnaire and the collection of the BELHES samples. The mean (SD) number of days is 52 (35). This is less than the period for suicidal ideation, assessed over the 12 previous months, but there might be a more limited overlap with the period for assessment of the other mental health variables, such as vitality and psychological distress, assessed over the last few weeks, and depressive and generalised anxiety disorders, assessed over the last 2 weeks. Fourthly, due to a non-response bias, the lowest socio-economic classes are less represented in our study population. This will not affect our dose–response associations but might affect the generalisability of our findings to the overall population. Finally, we do not have data on blood cell counts, which has been associated with mtDNAc [ 95 ].

In this large-scale study, we showed that living a healthy lifestyle was positively associated with mental health and well-being and, on a biological level, with a higher TL and mtDNAc, indicating healthy ageing. Furthermore, individuals with suicidal ideation or suffering from severe psychological distress had a lower mtDNAc. Our findings suggest that implementing strategies to incorporate healthy lifestyle changes in the public’s daily life could be beneficial for public health, and might offset the negative impact of environmental stressors. However, further studies are necessary to confirm our results and especially prospective and longitudinal studies are essential to determine causality of the associations.

Availability of data and materials

The dataset used for this study is available through a request to the Health Committee of the Data Protection Authority.

Abbreviations

Area under the curve

Body mass index

Confidence intervals

Generalised Anxiety Disorder Questionnaire

General Health Questionnaire

Inter-run calibrator

• Mitochondrial DNA content

Patient Health Questionnaire

Relative operating characteristic curve

Short Form Health Survey

• Telomere length

World Health Organization. Healthy living: what is a healthy lifestyle? Copenhagen: WHO Regional Office for Europe; 1999.

Google Scholar  

World Health Organization. Tackling NCDs: ‘best buys’ and other recommended interventions for the prevention and control of noncommunicable diseases. Geneva: World Health Organization; 2017.

World Health Organization. WHO global report on trends in prevalence of tobacco smoking 2000–2025. 3rd ed. Geneva: World Health Organisation; 2019.

World Health Organization. 2021 Physical Activity Factsheets for the European Union Member States in the WHO European Region. Copenhagen: WHO Regional Office for Europe; 2021.

Eurostat. Database on health determinants 2021 [Available from: https://ec.europa.eu/eurostat/web/health/data/database?node_code=hlth_det .

Whitman IR, Agarwal V, Nah G, Dukes JW, Vittinghoff E, Dewland TA, et al. Alcohol abuse and cardiac disease. J Am Coll Cardiol. 2017;69(1):13–24. https://doi.org/10.1016/j.jacc.2016.10.048 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Koliaki C, Liatis S, Kokkinos A. Obesity and cardiovascular disease: revisiting an old relationship. Metabolism. 2019;92:98–107. https://doi.org/10.1016/j.metabol.2018.10.011 .

Article   CAS   PubMed   Google Scholar  

Freisling H, Viallon V, Lennon H, Bagnardi V, Ricci C, Butterworth AS, et al. Lifestyle factors and risk of multimorbidity of cancer and cardiometabolic diseases: a multinational cohort study. BMC Med. 2020;18(1):5. https://doi.org/10.1186/s12916-019-1474-7 .

Liu Y, Pleasants RA, Croft JB, Wheaton AG, Heidari K, Malarcher AM, et al. Smoking duration, respiratory symptoms, and COPD in adults aged ≥ 45 years with a smoking history. Int J Chron Obstruct Pulmon Dis. 2015;10:1409. https://doi.org/10.2147/COPD.S82259 .

Kirsch Micheletti J, Bláfoss R, Sundstrup E, Bay H, Pastre CM, Andersen LL. Association between lifestyle and musculoskeletal pain: cross-sectional study among 10,000 adults from the general working population. BMC Musculoskelet Disord. 2019;20(1):609. https://doi.org/10.1186/s12891-019-3002-5 .

Bowe AK, Owens M, Codd MB, Lawlor BA, Glynn RW. Physical activity and mental health in an Irish population. Ir J Med Sci. 2019;188(2):625–31. https://doi.org/10.1007/s11845-018-1863-5 .

Article   PubMed   Google Scholar  

Richardson S, McNeill A, Brose LS. Smoking and quitting behaviours by mental health conditions in Great Britain (1993–2014). Addict Behav. 2019;90:14–9. https://doi.org/10.1016/j.addbeh.2018.10.011 .

Article   PubMed   PubMed Central   Google Scholar  

Levy MZ, Allsopp RC, Futcher AB, Greider CW, Harley CB. Telomere end-replication problem and cell aging. J Mol Biol. 1992;225(4):951–60. https://doi.org/10.1016/0022-2836(92)90096-3 .

Shaughnessy DT, McAllister K, Worth L, Haugen AC, Meyer JN, Domann FE, et al. Mitochondria, energetics, epigenetics, and cellular responses to stress. Environ Health Perspect. 2014;122(12):1271–8. https://doi.org/10.1289/ehp.1408418 .

Cui Y, Gao YT, Cai Q, Qu S, Cai H, Li HL, et al. Associations of leukocyte telomere length with body anthropometric indices and weight change in Chinese women. Obesity. 2013;21(12):2582–8. https://doi.org/10.1002/oby.20321 .

Crous-Bou M, Fung TT, Prescott J, Julin B, Du M, Sun Q, et al. Mediterranean diet and telomere length in Nurses’ Health Study: population based cohort study. BMJ. 2014;349:g6674. https://doi.org/10.1136/bmj.g6674 .

Janssen BG, Gyselaers W, Byun H-M, Roels HA, Cuypers A, Baccarelli AA, et al. Placental mitochondrial DNA and CYP1A1 gene methylation as molecular signatures for tobacco smoke exposure in pregnant women and the relevance for birth weight. J Transl Med. 2017;15(1):5. https://doi.org/10.1186/s12967-016-1113-4 .

Navarro-Mateu F, Husky M, Cayuela-Fuentes P, Álvarez FJ, Roca-Vega A, Rubio-Aparicio M, et al. The association of telomere length with substance use disorders: a systematic review and meta-analysis of observational studies. Addiction. 2020;116(8):1954–72. https://doi.org/10.1111/add.15312 .

Pyle A, Anugrha H, Kurzawa-Akanbi M, Yarnall A, Burn D, Hudson G. Reduced mitochondrial DNA copy number is a biomarker of Parkinson’s disease. Neurobiol Aging. 2016;38:216.e7-.e10. https://doi.org/10.1016/j.neurobiolaging.2015.10.033 .

Article   CAS   Google Scholar  

Ashar FN, Zhang Y, Longchamps RJ, Lane J, Moes A, Grove ML, et al. Association of mitochondrial DNA copy number with cardiovascular disease. JAMA Cardiol. 2017;2(11):1247–55. https://doi.org/10.1001/jamacardio.2017.3683 .

Chen S, Lin J, Matsuguchi T, Blackburn E, Yeh F, Best LG, et al. Short leukocyte telomere length predicts incidence and progression of carotid atherosclerosis in American Indians: the Strong Heart Family Study. Aging. 2014;6(5):414–27. https://doi.org/10.18632/aging.100671 .

Mons U, Müezzinler A, Schöttker B, Dieffenbach AK, Butterbach K, Schick M, et al. Leukocyte telomere length and all-cause, cardiovascular disease, and cancer mortality: results from individual-participant-data meta-analysis of 2 large prospective cohort studies. Am J Epidemiol. 2017;185(12):1317–26. https://doi.org/10.1093/aje/kww210 .

Lindqvist D, Epel ES, Mellon SH, Penninx BW, Révész D, Verhoeven JE, et al. Psychiatric disorders and leukocyte telomere length: underlying mechanisms linking mental illness with cellular aging. Neurosci Biobehav Rev. 2015;55:333–64. https://doi.org/10.1016/j.neubiorev.2015.05.007 .

Noble N, Paul C, Turon H, Oldmeadow C. Which modifiable health risk behaviours are related? A systematic review of the clustering of Smoking, Nutrition, Alcohol and Physical activity (‘SNAP’) health risk factors. Prev Med. 2015;81:16–41. https://doi.org/10.1016/j.ypmed.2015.07.003 .

Poortinga W. The prevalence and clustering of four major lifestyle risk factors in an English adult population. Prev Med. 2007;44(2):124–8. https://doi.org/10.1016/j.ypmed.2006.10.006 .

Demarest S, Van der Heyden J, Charafeddine R, Drieskens S, Gisle L, Tafforeau J. Methodological basics and evolution of the Belgian Health Interview Survey 1997–2008. Arch Public Health. 2013;71(1):24. https://doi.org/10.1186/0778-7367-71-24 .

Goldberg DP. User’s guide to the General Health Questionnaire. Windsor: NFER-Nelson; 1988.

Ware JE Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220–33. https://doi.org/10.1097/00005650-199603000-00003 .

Cantril H. Pattern of human concerns. New Brunswick: Rutgers University Press; 1965.

Kroenke K, Spitzer RL, Williams JB, Löwe B. The patient health questionnaire somatic, anxiety, and depressive symptom scales: a systematic review. Gen Hosp Psychiatry. 2010;32(4):345–59. https://doi.org/10.1016/j.genhosppsych.2010.03.006 .

Löwe B, Decker O, Müller S, Brähler E, Schellberg D, Herzog W, et al. Validation and standardization of the Generalized Anxiety Disorder Screener (GAD-7) in the general population. Med Care. 2008;46(3):266–74. https://doi.org/10.1097/MLR.0b013e318160d093 .

Gisle L, Drieskens S, Demarest S, Van der Heyden J. Enquête de santé 2018: Santé mentale. Bruxelles: Sciensano; 2018. ( https://www.enquetesante.be . Available from. Numéro de rapport: D/2020/14.440/3).

Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Medical Care. 1992;30(6):473–83.

Article   Google Scholar  

Braunholtz S, Davidson S, Myant K, O’Connor R. Well? What do you think?: the Third National Scottish Survey of Public Attitudes to Mental Health, Mental Wellbeing and Mental Health Problems. Scotland: Scottish Government Edinburgh; 2007.

Van Lente E, Barry MM, Molcho M, Morgan K, Watson D, Harrington J, et al. Measuring population mental health and social well-being. Int J Public Health. 2012;57(2):421–30. https://doi.org/10.1007/s00038-011-0317-x .

de Bruin A, Picavet HSJ, Nossikov A. Health interview surveys: towards international harmonization of methods and instruments. Copenhagen: World Health Organization. Regional Office for Europe; 1996.

Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–13.

Superior Health Council. Dietary guidelines for the Belgian adult population. Report 9284. Brussels: Superior Health Council; 2019.

World Health Organization. WHO guidelines on physical activity and sedentary behaviour. Geneva: World Health Organization; 2020.

World Health Organization. HEARTS Technical package for cardiovascular disease management in primary health care: healthy-lifestyle counselling. Geneva: World Health Organization; 2018.

World Health Organization. Country profiles on nutrition, physical activity and obesity in the 28 European Union Member States of the WHO European Region. Copenhagen: WHO Regional Office for Europe; 2013.

Bhaskaran K, dos-Santos-Silva I, Leon DA, Douglas IJ, Smeeth L. Association of BMI with overall and cause-specific mortality: a population-based cohort study of 3.6 million adults in the UK. Lancet Diabetes Endocrinol. 2018;6(12):944–53. https://doi.org/10.1016/S2213-8587(18)30288-2 .

Benetou V, Kanellopoulou A, Kanavou E, Fotiou A, Stavrou M, Richardson C, et al. Diet-related behaviors and diet quality among school-aged adolescents living in Greece. Nutrients. 2020;12(12):3804. https://doi.org/10.3390/nu12123804 .

Article   PubMed Central   Google Scholar  

Martens DS, Plusquin M, Gyselaers W, De Vivo I, Nawrot TS. Maternal pre-pregnancy body mass index and newborn telomere length. BMC Med. 2016;14(1):148. https://doi.org/10.1186/s12916-016-0689-0 .

Janssen BG, Munters E, Pieters N, Smeets K, Cox B, Cuypers A, et al. Placental mitochondrial DNA content and particulate air pollution during in utero life. Environ Health Perspect. 2012;120(9):1346–52. https://doi.org/10.1289/ehp.1104458 .

Jacka FN, O’Neil A, Opie R, Itsiopoulos C, Cotton S, Mohebbi M, et al. A randomised controlled trial of dietary improvement for adults with major depression (the ‘SMILES’ trial). BMC Med. 2017;15(1):23. https://doi.org/10.1186/s12916-017-0791-y .

Pavkovic B, Zaric M, Markovic M, Klacar M, Huljic A, Caricic A. Double screening for dual disorder, alcoholism and depression. Psychiatry Res. 2018;270:483–9. https://doi.org/10.1016/j.psychres.2018.10.013 .

De Wit LM, Van Straten A, Van Herten M, Penninx BW, Cuijpers P. Depression and body mass index, a u-shaped association. BMC Public Health. 2009;9:14. https://doi.org/10.1186/1471-2458-9-14 .

Meader N, King K, Moe-Byrne T, Wright K, Graham H, Petticrew M, et al. A systematic review on the clustering and co-occurrence of multiple risk behaviours. BMC Public Health. 2016;16:657. https://doi.org/10.1186/s12889-016-3373-6 .

Saneei P, Esmaillzadeh A, Hassanzadeh Keshteli A, Reza Roohafza H, Afshar H, Feizi A, et al. Combined healthy lifestyle is inversely associated with psychological disorders among adults. PLoS ONE. 2016;11(1):e0146888. https://doi.org/10.1371/journal.pone.0146888 .

Bonnet F, Irving K, Terra J-L, Nony P, Berthezène F, Moulin P. Anxiety and depression are associated with unhealthy lifestyle in patients at risk of cardiovascular disease. Atherosclerosis. 2005;178(2):339–44. https://doi.org/10.1016/j.atherosclerosis.2004.08.035 .

Loprinzi PD, Mahoney S. Concurrent occurrence of multiple positive lifestyle behaviors and depression among adults in the United States. J Affect Disord. 2014;165:126–30. https://doi.org/10.1016/j.jad.2014.04.073 .

Yang H, Gao J, Wang T, Yang L, Liu Y, Shen Y, et al. Association between adverse mental health and an unhealthy lifestyle in rural-to-urban migrant workers in Shanghai. J Formos Med Assoc. 2017;116(2):90–8. https://doi.org/10.1016/j.jfma.2016.03.004 .

Oftedal S, Kolt GS, Holliday EG, Stamatakis E, Vandelanotte C, Brown WJ, et al. Associations of health-behavior patterns, mental health and self-rated health. Prev Med. 2019;118:295–303. https://doi.org/10.1016/j.ypmed.2018.11.017 .

Conry MC, Morgan K, Curry P, McGee H, Harrington J, Ward M, et al. The clustering of health behaviours in Ireland and their relationship with mental health, self-rated health and quality of life. BMC Public Health. 2011;11(1):692. https://doi.org/10.1186/1471-2458-11-692 .

Buttery AK, Mensink GB, Busch MA. Healthy behaviours and mental health: findings from the German Health Update (GEDA). The European Journal of Public Health. 2015;25(2):219–25. https://doi.org/10.1093/eurpub/cku094 .

Ahmed NJ, Husen AZ, Khoshnaw N, Getta HA, Hussein ZS, Yassin AK, et al. The effects of smoking on IgE, oxidative stress and haemoglobin concentration. Asian Pac J Cancer Prev. 2020;21(4):1069–72. https://doi.org/10.31557/APJCP.2020.21.4.1069 .

Langley MR, Yoon H, Kim HN, Choi C-I, Simon W, Kleppe L, et al. High fat diet consumption results in mitochondrial dysfunction, oxidative stress, and oligodendrocyte loss in the central nervous system. Biochim Biophys Acta Mol Basis Dis. 2020;1866(3):165630. https://doi.org/10.1016/j.bbadis.2019.165630 .

Tan HK, Yates E, Lilly K, Dhanda AD. Oxidative stress in alcohol-related liver disease. World J Hepatol. 2020;12(7):332–49. https://doi.org/10.4254/wjh.v12.i7.332 .

Martens DS, Nawrot TS. Air pollution stress and the aging phenotype: the telomere connection. Curr Environ Health Rep. 2016;3(3):258–69. https://doi.org/10.1007/s40572-016-0098-8 .

Martens DS, Van Der Stukken C, Derom C, Thiery E, Bijnens EM, Nawrot TS. Newborn telomere length predicts later life telomere length: tracking telomere length from birth to child- and adulthood. EBioMedicine. 2021;63:103164. https://doi.org/10.1016/j.ebiom.2020.103164 .

Hang D, Nan H, Kværner AS, De Vivo I, Chan AT, Hu Z, et al. Longitudinal associations of lifetime adiposity with leukocyte telomere length and mitochondrial DNA copy number. Eur J Epidemiol. 2018;33(5):485–95. https://doi.org/10.1007/s10654-018-0382-z .

Gu Y, Honig LS, Schupf N, Lee JH, Luchsinger JA, Stern Y, et al. Mediterranean diet and leukocyte telomere length in a multi-ethnic elderly population. Age. 2015;37(2):24. https://doi.org/10.1007/s11357-015-9758-0 .

Sellami M, Al-Muraikhy S, Al-Jaber H, Al-Amri H, Al-Mansoori L, Mazloum NA, et al. Age and sport intensity-dependent changes in cytokines and telomere length in elite athletes. Antioxidants. 2021;10(7):1035. https://doi.org/10.3390/antiox10071035 .

Savela S, Saijonmaa O, Strandberg TE, Koistinen P, Strandberg AY, Tilvis RS, et al. Physical activity in midlife and telomere length measured in old age. Exp Gerontol. 2013;48(1):81–4. https://doi.org/10.1016/j.exger.2012.02.003 .

Mirabello L, Huang WY, Wong JY, Chatterjee N, Reding D, David Crawford E, et al. The association between leukocyte telomere length and cigarette smoking, dietary and physical variables, and risk of prostate cancer. Aging Cell. 2009;8(4):405–13. https://doi.org/10.1111/j.1474-9726.2009.00485.x .

Sun Q, Shi L, Prescott J, Chiuve SE, Hu FB, De Vivo I, et al. Healthy lifestyle and leukocyte telomere length in US women. PLoS ONE. 2012;7(5):e38374. https://doi.org/10.1371/journal.pone.0038374 .

Sindi S, Solomon A, Kåreholt I, Hovatta I, Antikainen R, Hänninen T, et al. Telomere length change in a multidomain lifestyle intervention to prevent cognitive decline: a randomized clinical trial. J Gerontol A. 2021;76(3):491–8. https://doi.org/10.1093/gerona/glaa279 .

Paltoglou G, Raftopoulou C, Nicolaides NC, Genitsaridi SM, Karampatsou SI, Papadopoulou M, et al. A comprehensive, multidisciplinary, personalized, lifestyle intervention program is associated with increased leukocyte telomere length in children and adolescents with overweight and obesity. Nutrients. 2021;13(8):2682. https://doi.org/10.3390/nu13082682 .

Chakravarty EF, Hubert HB, Krishnan E, Bruce BB, Lingala VB, Fries JF. Lifestyle risk factors predict disability and death in healthy aging adults. Am J Med. 2012;125(2):190–7. https://doi.org/10.1016/j.amjmed.2011.08.006 .

Bosnes I, Nordahl HM, Stordal E, Bosnes O, Myklebust TÅ, Almkvist O. Lifestyle predictors of successful aging: a 20-year prospective HUNT study. PLoS ONE. 2019;14(7):e0219200. https://doi.org/10.1371/journal.pone.0219200 .

Epel ES, Prather AA. Stress, telomeres, and psychopathology: toward a deeper understanding of a triad of early aging. Annu Rev Clin Psychol. 2018;14:371–97. https://doi.org/10.1146/annurev-clinpsy-032816-045054 .

Verhoeven JE, Révész D, Epel ES, Lin J, Wolkowitz OM, Penninx BW. Major depressive disorder and accelerated cellular aging: results from a large psychiatric cohort study. Mol Psychiatry. 2014;19(8):895–901. https://doi.org/10.1038/mp.2013.151 .

Han LK, Verhoeven JE, Tyrka AR, Penninx BW, Wolkowitz OM, Månsson KN, et al. Accelerating research on biological aging and mental health: current challenges and future directions. Psychoneuroendocrinology. 2019;106:293–311. https://doi.org/10.1016/j.psyneuen.2019.04.004 .

Kim MY, Lee JW, Kang HC, Kim E, Lee DC. Leukocyte mitochondrial DNA (mtDNA) content is associated with depression in old women. Arch Gerontol Geriatr. 2011;53(2):e218–21. https://doi.org/10.1016/j.archger.2010.11.019 .

Mengel-From J, Thinggaard M, Dalgard C, Kyvik KO, Christensen K, Christiansen L. Mitochondrial DNA copy number in peripheral blood cells declines with age and is associated with general health among elderly. Hum Genet. 2014;133(9):1149–59. https://doi.org/10.1007/s00439-014-1458-9 .

Otsuka I, Izumi T, Boku S, Kimura A, Zhang Y, Mouri K, et al. Aberrant telomere length and mitochondrial DNA copy number in suicide completers. Sci Rep. 2017;7(1):3176.

Chung JK, Lee SY, Park M, Joo E-J, Kim SA. Investigation of mitochondrial DNA copy number in patients with major depressive disorder. Psychiatry Res. 2019;282:112616. https://doi.org/10.1016/j.psychres.2019.112616 .

Takahashi PY, Jenkins GD, Welkie BP, McDonnell SK, Evans JM, Cerhan JR, et al. Association of mitochondrial DNA copy number with self-rated health status. Appl Clin Genet. 2018;11:121–7. https://doi.org/10.2147/TACG.S167640 .

Vyas CM, Ogata S, Reynolds CF 3rd, Mischoulon D, Chang G, Cook NR, et al. Lifestyle and behavioral factors and mitochondrial DNA copy number in a diverse cohort of mid-life and older adults. PLoS ONE. 2020;15(8):e0237235. https://doi.org/10.1371/journal.pone.0237235 .

Verhoeven JE, Révész D, van Oppen P, Epel ES, Wolkowitz OM, Penninx BW. Anxiety disorders and accelerated cellular ageing. Br J Psychiatry. 2015;206(5):371–8. https://doi.org/10.1192/bjp.bp.114.151027 .

Pisanu C, Vitali E, Meloni A, Congiu D, Severino G, Ardau R, et al. Investigating the role of leukocyte telomere length in treatment-resistant depression and in response to electroconvulsive therapy. J Pers Med. 2021;11(11):1100. https://doi.org/10.3390/jpm11111100 .

da Silva RS, de Moraes LS, da Rocha CAM, Ferreira-Fernandes H, Yoshioka FKN, Rey JA, et al. Telomere length and telomerase activity of leukocytes as biomarkers of selective serotonin reuptake inhibitor responses in patients with major depressive disorder. Psychiatr Genet. 2022;32(1):34–6. https://doi.org/10.1097/ypg.0000000000000305 .

Aas M, Elvsåshagen T, Westlye LT, Kaufmann T, Athanasiu L, Djurovic S, et al. Telomere length is associated with childhood trauma in patients with severe mental disorders. Transl Psychiatry. 2019;9(1):97. https://doi.org/10.1038/s41398-019-0432-7 .

Birkenæs V, Elvsåshagen T, Westlye LT, Høegh MC, Haram M, Werner MCF, et al. Telomeres are shorter and associated with number of suicide attempts in affective disorders. J Affect Disord. 2021;295:1032–9. https://doi.org/10.1016/j.jad.2021.08.135 .

Phillips AC, Robertson T, Carroll D, Der G, Shiels PG, McGlynn L, et al. Do symptoms of depression predict telomere length? Evidence from the West of Scotland Twenty-07 Study. Psychosom Med. 2013;75(3):288–96. https://doi.org/10.1097/PSY.0b013e318289e6b5 .

Incollingo Rodriguez AC, Polcari JJ, Nephew BC, Harris R, Zhang C, Murgatroyd C, et al. Acculturative stress, telomere length, and postpartum depression in Latinx mothers. J Psychiatr Res. 2022;147:301–6. https://doi.org/10.1016/j.jpsychires.2022.01.063 .

Darrow SM, Verhoeven JE, Révész D, Lindqvist D, Penninx BW, Delucchi KL, et al. The association between psychiatric disorders and telomere length: a meta-analysis involving 14,827 persons. Psychosom Med. 2016;78(7):776–87. https://doi.org/10.1097/PSY.0000000000000356 .

Simon NM, Walton ZE, Bui E, Prescott J, Hoge E, Keshaviah A, et al. Telomere length and telomerase in a well-characterized sample of individuals with major depressive disorder compared to controls. Psychoneuroendocrinology. 2015;58:9–22. https://doi.org/10.1016/j.psyneuen.2015.04.004 .

Schaakxs R, Verhoeven JE, Voshaar RCO, Comijs HC, Penninx BW. Leukocyte telomere length and late-life depression. Am J Geriatr Psychiatry. 2015;23(4):423–32. https://doi.org/10.1016/j.jagp.2014.06.003 .

de Baumont AC, Hoffmann MS, Bortoluzzi A, Fries GR, Lavandoski P, Grun LK, et al. Telomere length and epigenetic age acceleration in adolescents with anxiety disorders. Sci Rep. 2021;11(1):7716. https://doi.org/10.1038/s41598-021-87045-w .

Schutte NS, Malouff JM. The association between depression and leukocyte telomere length: a meta-analysis. Depress Anxiety. 2015;32(4):229–38. https://doi.org/10.1002/da.22351 .

Drieskens S, Demarest S, Bel S, De Ridder K, Tafforeau J. Correction of self-reported BMI based on objective measurements: a Belgian experience. Archives of Public Health. 2018;76:10. https://doi.org/10.1186/s13690-018-0255-7 .

Gorber SC, Schofield-Hurwitz S, Hardt J, Levasseur G, Tremblay M. The accuracy of self-reported smoking: a systematic review of the relationship between self-reported and cotinine-assessed smoking status. Nicotine Tob Res. 2009;11(1):12–24. https://doi.org/10.1093/ntr/ntn010 .

Knez J, Winckelmans E, Plusquin M, Thijs L, Cauwenberghs N, Gu Y, et al. Correlates of peripheral blood mitochondrial DNA content in a general population. Am J Epidemiol. 2016;183(2):138–46. https://doi.org/10.1093/aje/kwv175 .

Download references

Acknowledgements

We are grateful to all BHIS and BELHES participants for contributing to this study.

The HuBiHIS project is financed by Sciensano (PJ) N°: 1179–101. Dries Martens is a postdoctoral fellow of the Research Foundation—Flanders (FWO 12X9620N).

Author information

Authors and affiliations.

Sciensano, Risk and Health Impact Assessment, Juliette Wytsmanstraat 14, 1050, Brussels, Belgium

Pauline Hautekiet, Nelly D. Saenen & Eva M. De Clercq

Centre for Environmental Sciences, Hasselt University, 3500, Hasselt, Belgium

Pauline Hautekiet, Nelly D. Saenen, Dries S. Martens, Margot Debay & Tim S. Nawrot

Sciensano, Epidemiology and Public Health, Juliette Wytsmanstraat 14, 1050, Brussels, Belgium

Johan Van der Heyden

Centre for Environment and Health, Leuven University, 3000, Leuven, Belgium

Tim S. Nawrot

You can also search for this author in PubMed   Google Scholar

Contributions

PH drafted the paper. PH, NS, MD and ED set up the design of the study. NS, DM, JvdH, TN and ED reviewed and commented on the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Pauline Hautekiet .

Ethics declarations

Ethics approval and consent to participate, consent for publication.

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: text s1..

TL, mtDNAc and single copy-gene reaction mixture and PCR cycling conditions. Table S1. The mental health indicators with their scores and uses. Table S2. Comparison of the characteristics of the 6,054 eligible BHIS participants that were included in the BHIS subset compared to the 1,838 eligible participants that were excluded from the BHIS subset. Table S3. Comparison of the characteristics of the 739 participants from the BHIS subset that were included in the BELHES subset compared to the 5,315 participants that were excluded from the BELHES subset. Table S4. Bivariate associations between the characteristics and telomere length (TL), mitochondrial DNA content (mtDNAc), the lifestyle score or psychological distress. Table S5. Results of the sensitivity analysis of the association between lifestyle and mental health. Table S6. Results of the sensitivity analysis of the association between lifestyle and the biomarkers of ageing. Table S7. Results of the sensitivity analysis of the association between mental health and the biomarkers of ageing. Fig. S1. Exclusion criteria. The BHIS subset consisted of 6,055 BHIS participants and the BELHES subset consisted of 739 BELHES participants. Fig. S2. Histogram of the lifestyle score. Fig. S3. Validation of the lifestyle score. ROC curve for the lifestyle score as a predictor for good to very good self-perceived health. The model was adjusted for age, sex, region, highest educational level in the household, household composition and country of birth.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Hautekiet, P., Saenen, N.D., Martens, D.S. et al. A healthy lifestyle is positively associated with mental health and well-being and core markers in ageing. BMC Med 20 , 328 (2022). https://doi.org/10.1186/s12916-022-02524-9

Download citation

Received : 04 February 2022

Accepted : 10 August 2022

Published : 29 September 2022

DOI : https://doi.org/10.1186/s12916-022-02524-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Mental health
• Biological ageing

BMC Medicine

ISSN: 1741-7015

• Submission enquiries: [email protected]
• General enquiries: [email protected]

• Search Menu
• Advance Articles
• Editor's Choice
• Supplements
• E-Collections
• Virtual Roundtables
• Author Videos
• Author Guidelines
• Submission Site
• Open Access Options
• About The European Journal of Public Health
• About the European Public Health Association
• Editorial Board
• Advertising and Corporate Services
• Journals Career Network
• Self-Archiving Policy
• Terms and Conditions
• Explore Publishing with EJPH
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Introduction, acknowledgements.

• < Previous

Living longer and feeling better: healthy lifestyle, self-rated health, obesity and depression in Ireland

• Article contents
• Figures & tables
• Supplementary Data

Janas Harrington, Ivan J. Perry, Jennifer Lutomski, Anthony P. Fitzgerald, Frances Shiely, Hannah McGee, Margaret M. Barry, Eric Van Lente, Karen Morgan, Emer Shelley, Living longer and feeling better: healthy lifestyle, self-rated health, obesity and depression in Ireland, European Journal of Public Health , Volume 20, Issue 1, February 2010, Pages 91–95, https://doi.org/10.1093/eurpub/ckp102

• Permissions Icon Permissions

Background: The combination of four protective lifestyle behaviours (being physically active, a non-smoker, a moderate alcohol consumer and having adequate fruit and vegetable intake) has been estimated to increase life expectancy by 14 years. However, the effect of adopting these lifestyle behaviours on general health, obesity and mental health is less defined. We examined the combined effect of these behaviours on self-rated health, overweight/obesity and depression. Methods: Using data from the Survey of Lifestyle Attitudes and Nutrition (SLÁN) 2007 (), a protective lifestyle behaviour (PLB) score was constructed for 10 364 men and women (>18 years), and representative of the Republic of Ireland adult population (response rate 62%). Respondents scored a maximum of four points, one point each for being physically active, consuming five or more fruit and vegetable servings daily, a non-smoker and a moderate drinker. Results: One-fifth of respondents (20%) adopted four PLBs, 35% adopted three, 29% two, 13% one and 2% adopted none. Compared to those with zero PLBs, those with four were seven times more likely to rate their general health as excellent/very good [OR 6.8 95% CI (3.64–12.82)] and four times more likely to have better mental health [OR 4.4 95% CI (2.34–8.22)]. Conclusions: Adoption of core protective lifestyle factors known to increase life expectancy is associated with positive self-rated health, healthier weight and better mental health. These lifestyles have the potential to add quality and quantity to life.

It has been known for some time that adoption of a number of core protective/health promoting lifestyle behaviours at an individual level has a potentially large positive influence on population health. There is increasing recognition of the value of these behaviourally defined protective behaviours for health promotion and population health monitoring, 1–8 and advice on smoking cessation, healthy diet, physical exercise and moderation in alcohol consumption has been a pillar of health education for many years. While anecdotally a perception exists that adoption of a healthy lifestyle may impair quality of life as evidenced by the admonition ‘You won’t live forever, it will just feel like it’, recent evidence suggests that quality as well as quantity can be added to life through the adoption of relatively minor lifestyle changes. 5

Results from the Nurse's Health Study 9 reported the positive effects of a limited number of core protective lifestyle behaviours (PLBs) [body mass index (BMI) < 25 kg m −2 ; a diet high in cereal fibre and polyunsaturated fat and low in trans fat and glycaemic load; engagement in moderate-to-vigorous physical activity for at least half an hour per day; no current smoking and the consumption of an average of at least half a drink of an alcoholic beverage per day] in relation to the decreased risk of type 2 diabetes. This work has been replicated in a cross-sectional study with markers of cardiovascular risk including hypertension, dyslipidaemia and insulin resistance. 4 , 5 , 10 More recently, Khaw et al. , 1 in their work from the European Prospective Investigation into Cancer (EPIC) study, focused on behaviourally defined measures. They identified four lifestyle behaviours: being physically active, a non-smoker, having a moderate alcohol consumption and an adequate fruit and vegetable intake and found that the combined effect of these health behaviours predicted a 4-fold difference in total mortality in men and women, 1 equating to a 14-year difference in life expectancy between individuals practising none of these behaviours relative to those practising all four of them. In further work from the EPIC study, Myint et al. 11 concluded that behavioural factors were associated with substantial differences in age-related decline in functional health and the prevalence of those in good and poor functional health in the community.

Examining the effects of individual risk factors for chronic disease and poor physical and mental health is not a new concept; however, their combined effect on general health, obesity and mental health is less well defined. The aim of this study was to examine the combined effect of practising four non-clinically defined lifestyle behaviours (being a non-smoker, being physically active, being a moderate drinker, and consuming five portions of fruit and vegetables daily) on self-rated health, overweight/obesity and mental health.

Based on the work by Khaw et al. , 1 we constructed a PLB score. Participants scored one point for each of the following health behaviours: being a non-smoker, being physically active (moderate/high activity score), being a moderate drinker (1–14 alcohol units per week) and consuming five or more servings of fruit and vegetables daily. Respondents could score from zero to four on protective health behaviours.

General study design

The study was the third national Survey of Lifestyle, Attitudes and Nutrition (SLÁN) in Ireland conducted in 2007, 12–14 involving a nationally representative sample of 10 364 respondents (62% response rate) to whom a detailed health and lifestyle questionnaire was administered by face-to-face interview. In addition, 9223 (89%) completed a Willett Food Frequency Questionnaire (FFQ). The FFQ was an adapted version of the EPIC study, 15 validated for use in the Irish population. 16 Participants who did not complete a FFQ were excluded from this analysis.

The population for the survey was defined as adults aged 18 years and over living in residential households in Ireland (residents of institutions, nursing homes, hospitals, prisons and homeless hostels were not included). Full details of the sampling frame and weighting can be found elsewhere. 12 In summary, the sampling frame used for the survey was the GeoDirectory, a list of all addresses in the Republic of Ireland, which distinguishes between residential and commercial establishments. The sample was a multi-stage probability sample, where each dwelling has a known probability of selection. The sample was weighted to closely approximate the Census 2006 figures for gender, age, marital status, education, occupation, region, household size and ethnicity.

Health and lifestyle questionnaire

A single question was included on self-rated health, respondents were asked to rate their health on a 5-point scale ranging from ‘excellent’ to ‘poor’. Being a current smoker was defined as smoking either ‘every day’ or ‘some days’. Non-smokers were classified as those who had never smoked; former smokers were those who had smoked ‘at least 100 cigarettes in their lifetime’ but do not currently smoke. For the purpose of this article, current smokers are compared with non-smokers. Average alcohol consumption was estimated as the units of alcohol consumed per week. For the purpose of this article, a moderate drinker was defined as someone who consumed between 1 and 14 units a week. A unit is defined as either ‘a half pint of beer; a single measure of spirits; or as a single glass of wine, sherry or port’. Respondents were also asked if they had experienced any chronic illness from a pre-defined list in the previous 12 months.

International physical activity questionnaire (IPAQ)

Respondents were asked a series of questions relating to the time they spent being physically active. The responses were used to calculate a physical activity score (IPAQ score) for each respondent. These scores were classified as high (over 10 000 steps per day), moderate (5000–10 000 steps per day) or low (less than 5000 steps per day). For this analysis, a binary variable was created; ‘low’ or ‘moderate/high’, ‘low’ was defined as being physically inactive.

Composite international diagnostic interview (CIDI)

Respondents were asked a series of questions pertaining to their mental health status. The CIDI-SF (short form) Version 1.1 health interview survey, part of which was incorporated in the main SLÁN interview, provides a probable diagnosis (CIDI-SF yields a likelihood of having a major depression rather than a full diagnosis; hence, the term ‘probable Major Depressive Disorder’ is used throughout this article) of major depressive disorder. 17 Full details of the mental health measures have been reported elsewhere. 18

Food frequency questionnaire

The dietary habits of respondents who completed a FFQ were analysed in relation to food groups. Full details of the FFQ have been documented elsewhere. 19 For this analysis, fruit and vegetable intake was collapsed to a binary variable with participants categorized as consuming ‘five or more servings daily’ or ‘less than five servings daily’.

SLÁN 2007 respondents were also asked to self-report their own height and weight. BMI was calculated based on the standard formula [height (m)/weight (kg) × weight (kg)], they were classified as overweight or obese based on a BMI score of ≥25 or 30 kg m −2 , respectively.

Statistical analysis

Data were analysed using SPSS TM (Version 15.0). Logistic regression was used to examine the relationship between PLB score, self-rated health, probable depressive disorder and obesity levels after adjusting for age, sex, education and social class. Additionally, we examined the relationship between PLB score and past diagnoses of medically diagnosed chronic illness.

Table 1 shows a breakdown of the relevant participant characteristics differentiated by gender. Higher proportions of women were of normal weight and consumed five or more daily servings of fruit and vegetables compared with men. Men were more likely to be smokers, to consume more alcohol and to be physically active compared with women. Women were more likely to have adopted more of the PLBs. Table 2 shows the age, gender, social demographic profile and the distribution of key outcome variables in five groups of study participants defined on the basis of number of PLBs. Clear and highly significant trends were seen for age, gender, education and social classification status. Those with three and four PLBs were more likely to be female, in the younger/middle age group to have tertiary education and to be in the ‘large employers/professional/manager’ socioeconomic classification group. Respondents with a lower PLB score were significantly more likely to have a depressive disorder ( P < 0.01).

Distribution of variables for SLÁN 2007 participants included in this analysis (participants who did not complete a FFQ were excluded from the analysis)

a: Smoker was classified as someone who smokes either everyday or some days

*Significant gender difference P < 0.01; ***Significant gender difference P < 0.05

Demographic breakdown by number of protective lifestyle behaviours practised

Associations between PLBs and feeling healthy

The association between PLB score, self-rated health, healthy weight and better mental health adjusted for age, sex, education and social class is shown in table 3 . For self-rated health and depressive state, clear and highly significant trends in odds ratios were observed across the five groups of study participants. These trends were not as obvious for body weight. Relative to those with zero PLBs, those with four were almost seven times more likely to rate their general health as excellent/very good [OR 6.8, 95% CI (3.64–12.82)]. These trends persisted even when the model was adjusted for depressive disorders. Those with four PLBs were also four times more likely to have better mental health [OR 4.4, 95% CI (2.34–8.22)] indicating a better overall general health and well-being. While similar trends were not as obvious in relation to BMI status, those with four PLBs had an elevated likelihood of being normal weight (BMI < 25 kg m −2 ) than overweight/obese (BMI > 25 kg m −2 ) compared with those with fewer PLBs.

Respondent's likelihood of self-rated general health being excellent/very good/good; likelihood of BMI <25 kg m −2 and the likelihood of not having depressive disorder compared with having depressive disorder by number of protective lifestyle behaviours adjusted for age, gender, education and social class

*For trend significant P < 0.01

We know from longitudinal studies that PLBs increase longevity 1 ; this article shows that they are also associated with better self-rated health, better mental health and healthier body weight; conversely, those who had fewer PLBs were ‘not only’ leading unhealthier lifestyles, but they also perceived their overall health to be poorer, had a higher likelihood of having depression and were heavier than those with higher numbers of PLBs. Higher scores were also less likely to be associated with being diagnosed with a cardiovascular event and being diagnosed with any illness by a doctor in the last 12 months. While our results are congruent with the work by Khaw et al. 1 and Myint et al. 11 who examined the relationship between PLBs and mortality 1 and PLBs and functional health, 11 this is one of the first studies to look at self-rated health, depression and overweight/obesity in relation to PLBs.

Limitations of the study include the cross-sectional design, and the relatively low response rate (62%). However, this is similar to response rates seen in other major National Health and Lifestyle Surveys. 13 , 14 It is increasingly difficult to get high response rates from national general population surveys due to the sociodemographic trends in the modern society including longer working days and the phenomenon of gated communities, particularly in urban areas. Unfortunately, data on non-participation are not available. However, sample weights were used derived from the most recent Census. 20 Interpretation of the data must be cautious; since exposure and outcome were measured at the same time, it is not possible to ascertain which is the cause and which is the effect. It can be argued that persons with better than average self-rated health and better mental health are more likely to engage in health seeking behaviour. The issue of reverse causation cannot be resolved in this study; however, it is likely that the causal effects of these health seeking behaviours flow in both directions are mutually beneficial: better mental health and better self-rated health leading to increased health seeking behaviours and vice versa. What is clear is that there is no evidence to suggest that the presence of health seeking behaviours is associated with poorer mental health and well-being.

Our findings add to the evidence that we can achieve progress to address the ‘causes of the causes’ of all-cause mortality, mental ill health and cardiovascular disease through small achievable lifestyle behaviour modifications. A key challenge for future research is to better understand the individual and societal determinants of health-seeking behaviour. For instance, there is emerging data highlighting the importance of adverse childhood experiences as a determinant of health-related behaviour in adult life. 21 Data from the USA 22–24 show that children with low rates of childhood adversity not only have better mental health in adult life but better physical health with lower rates of high-risk behaviours and conditions e.g. obesity.

Given the association between self-rated health, better mental health and higher numbers of PLBs, we propose that the four lifestyle behaviours detailed in this article be used as outcome measures from which effectiveness of public health policy can be gauged.

SLÁN was funded by the Department of Health and Children.

Conflicts of interest : None declared.

Being a non-smoker, being physically active, having a moderate alcohol intake and consuming five portions of fruit and vegetables daily are associated with better self-rated health, better mental health and a healthier weight.

We would propose that the four lifestyle behaviours detailed in this article be used as outcome measures from which effectiveness of public policy can be gauged.

The authors thank other SLÁN 2007 Consortium members for their contribution to this research. Consortium members: Professor Hannah McGee (Project Director)(RCSI), Professor Ivan Perry (PI)(UCC), Professor Margaret Barry (PI)(NUIG), Dr. Dorothy Watson (PI)(ESRI), Dr Karen Morgan (Research Manager, RCSI), Dr. Emer Shelley (RCSI), Professor Ronan Conroy (RCSI), Professor Ruairí Brugha (RCSI), Dr. Michal Molcho (NUIG), Ms. Janas Harrington (UCC) and Professor Richard Layte (ESRI), Ms Nuala Tully (RCSI), Ms Jennifer Lutomski (UCC), Mr Mark Ward (RCSI) and Mr Eric Van Lente (NUIG). Also Jan van den Broeck for his helpful comments during the drafting of the paper. SLÁN 2007 was approved by the Ethics Committee of the Royal College of Surgeons of Ireland.

Google Scholar

Google Preview

• physical activity
• depressive disorders
• mental health

Email alerts

Citing articles via.

• Contact EUPHA
• Recommend to your Library

Affiliations

• Online ISSN 1464-360X
• Print ISSN 1101-1262
• Copyright © 2024 European Public Health Association
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Healthy lifestyle behaviours are positively and independently associated with academic achievement: An analysis of self-reported data from a nationally representative sample of Canadian early adolescents

Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Writing – original draft

* E-mail: [email protected]

Affiliation School of Public Health, University of Alberta, Edmonton, Alberta, Canada

Roles Writing – review & editing

Affiliation Department of Elementary Education, University of Alberta, Edmonton, Alberta, Canada

Roles Data curation, Investigation, Writing – review & editing

Affiliation Departments of Public Health Sciences and Emergency Medicine, Queens University, Kingston, Ontario, Canada

Roles Conceptualization, Funding acquisition, Methodology, Supervision, Writing – review & editing

• Erin L. Faught, 
• Doug Gleddie, 
• Kate E. Storey, 
• Colleen M. Davison, 
• Paul J. Veugelers

• Published: July 28, 2017
• https://doi.org/10.1371/journal.pone.0181938
• Reader Comments

Introduction

The lifestyle behaviours of early adolescents, including diet, physical activity, sleep, and screen usage, are well established contributors to health. These behaviours have also been shown to be associated with academic achievement. Poor academic achievement can additionally contribute to poorer health over the lifespan. This study aims to characterize the associations between health behaviours and self-reported academic achievement.

Data from the 2014 Canadian Health Behaviour in School-Aged Children Study (n = 28,608, ages 11–15) were analyzed. Students provided self-report of academic achievement, diet, physical activity, sleep duration, recreational screen time usage, height, weight, and socioeconomic status. Multi-level logistic regression was used to assess the relationship of lifestyle behaviours and body weight status with academic achievement while considering sex, age, and socioeconomic status as potential confounders.

All health behaviours exhibited independent associations with academic achievement. Frequent consumption of vegetables and fruits, breakfast and dinner with family and regular physical activity were positively associated with higher levels of academic achievement, while frequent consumption of junk food, not meeting sleep recommendations, and overweight and obesity were negatively associated with high academic achievement.

Conclusions

The present findings demonstrate that lifestyle behaviours are associated with academic achievement, potentially identifying these lifestyle behaviours as effective targets to improve academic achievement in early adolescents. These findings also justify investments in school-based health promotion initiatives.

Citation: Faught EL, Gleddie D, Storey KE, Davison CM, Veugelers PJ (2017) Healthy lifestyle behaviours are positively and independently associated with academic achievement: An analysis of self-reported data from a nationally representative sample of Canadian early adolescents. PLoS ONE 12(7): e0181938. https://doi.org/10.1371/journal.pone.0181938

Editor: David Meyre, McMaster University, CANADA

Received: March 16, 2017; Accepted: July 10, 2017; Published: July 28, 2017

Copyright: © 2017 Faught et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data underlying this study are third party data that are available upon request from the Principal Investigator and Project Manager for the Health Behaviour School-Aged Children (HBSC) Study Canada—Dr. John Freeman ( [email protected] ) and Matthew King ( [email protected] ).Others aiming to use HBSC Canada data outside the scope of this manuscript may request it via standard application subject to approval by the HBSC Canada team. This process can be found here. http://www.uib.no/en/hbscdata/94226/access-other-hbsc-survey-data

Funding: The Public Health Agency of Canada provided funding for Cycle 7 of the Health Behaviour in School-aged Children in Canada. The present work was supported by a Collaborative Research and Innovation Opportunities (CRIO) Team program from Alberta Innovates – Health Solutions (AIHS) [201300671]. ELF was supported by the Women and Children’s Health Research Institute through the generous support of the Stollery Children’s Hospital Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

The physical activity, diet, sleep, and screen time of children and youth are an important concern both to public health professionals and to society today. Consistently, population-level evaluations demonstrate that both children and adolescents in Canada are failing to achieve established healthful recommendations for all of these behaviours [ 1 – 3 ]. Consequently, children and adolescents are experiencing adverse health consequences at unprecedented rates, including obesity [ 4 ] and type 2 diabetes [ 5 ], putting them at risk for ill health and chronic disease across their lifespan.

In addition to consequences to health, these behaviours (diet, physical activity, sleep, and screen time) have been shown to be associated with children and adolescents’ academic achievement. Reaching optimal nutrition, physical activity, and sleep levels have demonstrated importance for academic achievement [ 6 – 8 ] while excess recreational screen time has been shown to negatively influence academic achievement [ 9 ]. In addition, overweight and obesity have been associated with poorer academic achievement [ 10 ], although recent reviews have noted that studies including potential mediators and moderators of this relationship, such as physical activity, diet, sleep, screen time, and socioeconomic status, have rarely been considered in these analyses [ 11 , 12 ]. The high prevalence of unhealthy behaviours and their negative associations with academic achievement is concerning for children and adolescents as school engagement and education are demonstrated protective factors against the development of adverse health over the entire lifespan [ 13 , 14 ]. Thus, these unhealthy behaviours contribute to adverse health consequences through direct physiological effects and by negatively influencing the likelihood of succeeding in school, resulting in lower socioeconomic status later in life [ 15 ]. The health-education relationship also suggests that effective interventions which improve the physical activity levels, diets, sleep, and screen use of youth, such as school-based health promotion, can have direct benefits on health as well as improve educational attainment, resulting in a healthier, more prosperous, and productive next generation [ 16 ]. Several school-based health promotion programs have been shown to achieve demonstrated improvements in lifestyle behaviours and academic achievement, though studies assessing this are few and further evidence is needed [ 17 , 18 ].

Although each of diet, physical activity, sleep, screen time, and body weight status have established relationships with academic achievement, few studies [ 19 – 22 ] have considered all of these health behaviours simultaneously in an analysis to determine their independent effects on academic achievement. Findings from these limited studies indicate that exhibiting a healthy diet, adequate physical activity and sleep, and reduced screen time, have individual, positive associations with academic achievement regardless of body weight status with the exception of one study [ 19 – 22 ]. This evidence is supportive of school health approaches that are multi-componential, as focusing on singular behaviours may not have as substantial as an effect on academic achievement as would considering multiple healthy behaviours. In addition, this evidence supports the idea that health promotion to improve physical activity, diet is beneficial for the academic achievement all students, not simply those who are overweight or obese.

Our objective is to complement and expand on the limited existing studies that have aimed to investigate the independent associations of physical activity, diet, sleep, screen time, and body weight status on academic achievement using a large, population-based sample of early adolescents (age 11–15) from all provinces in Canada. This is the largest study to date and the first to use a representative sample of early adolescents in Canada. These findings can be used to inform population-level interventions to improve the physical activity, diet, sleep, and screen use of children and youth and consequently reduce their likelihood of adverse health and academic achievement outcomes.

This work is a secondary data analysis of data from the Canadian version (Cycle 7) of the 2014 Health Behaviour in School-aged Children (HBSC) study [ 23 ]. This questionnaire is conducted in collaboration with the World Health Organization (WHO) [ 23 ]. Data was requested for this secondary analysis using a standard procedure that can be found at http://www.uib.no/en/hbscdata/94226/access-other-hbsc-survey-data . In Canada, consistent with all participating countries, this survey was carried out among a representative sample of Grade 6–10 (focused on 11–15 year old) students in all 13 provinces and territories excluding students who were on First Nation or Indian reserves, private and home schooled students, youth not in a school setting, or incarcerated youth [ 24 ]. All provinces and territories invited to participate in the survey consented to participation. A two-stage cluster sampling approach was used in most provinces: school jurisdictions were identified and categorized based on key characteristics (language of instruction, public/separate designation, and size of community). Upon jurisdictional consent, schools were randomly selected to participate within each jurisdiction. Schools had the opportunity to decide if the survey would be completed online or using paper and pen. Surveys were administered during school time (45–70 minute single session) and overseen by a teacher. Surveys asked a wide variety of questions about health and lifestyle behaviours as well as socio-demographic information. Further information about the HBSC survey can be found at http://healthycanadians.gc.ca/publications/science-research-sciences-recherches/health-behaviour-children-canada-2015-comportements-sante-jeunes/index-eng.php#c1a2 . There was a 77% student response rate for the HBSC survey, resulting in 29,837 student participants [ 24 ]. After excluding students without complete information for academic achievement, 28,608 (96%) students were considered in the analysis. Sampling weights were applied to the sample in order to achieve representativeness of Canadian youth by grade, gender and province or territory. Ethical approval for the HBSC study in Canada was obtained from the Queen’s University General Research Ethics Board (Approval GMISC-062-13) and from Health Canada and the Public Health Agency of Canada.

Academic achievement

Students self-reported their academic achievement by responding to the following question: ‘Which of the following best describes your marks during the past year?’ Possible responses were: ‘Mostly A’s/above 85%/or level 4’, ‘Mostly A’s and B’s/between 70 and 84%/or level 3 and 4’, ‘Mostly B’s and C’s/between 60 and 69%/or level 3’, ‘Mostly C’s/between 50 and 59%/ or level 2’, and ‘Mostly letter grades below C/below 50%/or level 1’. For ease of readibility, these categories are henceforth referred to by their letter categories. These categories were collapsed into two categories: ‘Excellent’ (Mostly A’s, Mostly A’s and B’s) and Fair (Mostly B’s and C’s and below).

Physical activity

Physical activity was assessed using the question: “Over a typical or usual week, on how many days are you physically active for a total of at least 60 minutes per day?” Possible responses were 0, 1, 2, 3, 4, 5, 6, or 7 days per week. This question corresponds with the Canadian 24-hour Movement Guidelines for Children and Youth [ 25 ] which recommend 60 minutes of physical activity per day for children 11–17. Because several studies have found the relationship between physical activity and academic achievement to have an inverse-U shape rather than a positive dose-response shape [ 26 , 27 ], we divided the days per week achieving 60 minutes of physical activity into three categories: 0–2 days, 3–5 days, and 6–7 days.

Dietary aspects

Diet was assessed using a short food frequency questionnaire [ 28 , 29 ] and several free-standing questions about dietary habits. In order to reduce the number of variables and identify essential groupings of data from the short food frequency questionnaire, we conducted exploratory factor analysis with oblique rotation to allow for correlation between factors. This method was used to identify foods and behaviours from the diet-related questions that frequently occur together, such as children reporting frequently eating ‘vegetables’ being also more likely to frequently eat ‘orange vegetables (carrots, squash, sweet potato, etc.),’ which are two separate items in the questionnaire. The factor scores that were generated were used in regression analyses to quantify each factor’s association with academic achievement.

Our factor analysis of diet-related variables identified three factors from 16 variables. We named these: (1) Junk Food, (2) Vegetables, Pulses, and Fruit, and (3) Healthy Eating Habits. The Healthy Eating Habits factor comprised of responses to questions about the frequency of eating breakfast and consuming meals in the presence of family. Although the food frequency questionnaire item ‘Game from hunting (moose, caribou, venison, etc.)’ was included in the factor analysis, it did not load onto any factor using the specified cutoff (0.4) and as such was not included. Table 1 lists all food frequency questionnaire items and their loadings onto respective factors. All factor loadings indicate that the higher the item is reported being consumed, the associated factor score increases. Conversely, if the item was reported to be consumed infrequently, the factor score decreases.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0181938.t001

Students reported their bedtime (“turned out the light and gone to sleep”) and wake-up time on school days and weekend days over the past week. Sleep duration was organized into categories of meeting and not meeting recommendations in reference to age group-specific thresholds in the Canadian 24-hour Movement Guidelines for Children and Youth [ 25 ]. Children between the ages of 8–13 are recommended to get between 9 and 11 hours of sleep per night, whereas youth between the ages of 14–17 are recommended to get 8 and 10 hours of sleep per night.

Screen time

Average daily screen time was assessed using four questions about typical usage of various screens on weekdays and weekends. Students were asked for both a typical weekday and a typical weekend day: How many hours a day, in your free time, do you usually spend watching TV, videos (including YouTube or similar services), DVDs, and other entertainment on a screen? Possible responses were: ‘None at all’, ‘About half an hour a day’, ‘About one hour’, etc, until ‘About 7 or more hours a day.’ This question is also asked for ‘time spent playing games on a computer, console, tablet (like iPad), smartphone, or other electronic device ( not including moving or fitness games)?, and ‘time spent using electronic devices such as computers, tablets (like iPad), or smartphones for other purposes (e.g., homework, emailing, tweeting, Facebook, chatting, surfing the internet)?’. Responses were totaled for all devices for weekdays and weekends; the average of these two totals was calculated to represent average daily screen time during a typical week. Categories were organized in reference to the Canadian 24-hour Movement Guidelines for Children and Youth [ 25 ] where <2 hours per day of screen time is recommended.

Body weight status

Height and weight were self-reported by students in the unit of their choice. Reponses for height were converted to centimeters if reported in inches and responses for weight were converted to kilograms if reported in pounds. These values were used to calculate body mass index (kg/m 2 ), which became a measure of body weight status as per age- and sex- specific WHO Child Growth Standards [ 30 ]. Children were classified as being thin if they were more than 2 standard deviations below the mean and severely thin if they were more than 3 standard deviations below the mean. These two categories were collapsed into one because of small cell sizes. Children were classified as overweight if they were more than 1 standard deviation above the mean and obese if they were more than two standard deviations above the mean [ 30 ].

Socioeconomic status

Socioeconomic status was determined using the Family Affluence Scale (FAS). The FAS is a validated scale that is used across all countries using the HBSC survey, and has been used in aggregate analyses focusing on the relationship between SES and adolescent health [ 31 ]. The FAS comprises of questions that ask about material goods (vehicles, individual bedrooms, and computers) and vacations to assess family wealth. The FAS generates a score between 0–9, with values of 0–2 corresponding with low affluence, 3–5 corresponding with medium affluence, and 6–9 corresponding with high affluence [ 31 ]. These categories were used in this analysis with low affluence as the reference category.

Statistical analysis

Mixed-effects logistic regression was employed given that the proportional odds assumption required for ordinal logistic regression was violated. Bootstrapping techniques were employed to address the complex sampling design of the survey. As participating students are nested within school environments, while schools are nested within provinces, students are more likely to be similar to other students within their school environment, and in their province or territory, as education is under the jurisdiction of provinces and territories in Canada. As such, both schools and provinces were treated as levels of clustering within the sample. The ICCs for clustering at the school and provincial level, respectively, were 0.11 and 0.06 respectively. In order to determine if the fit of the clustered models were significantly better than a model that did not consider the clustering, likelihood ratio tests were performed. Results indicated that the clustered model was a statistically significantly better fit than one that ignored clustering (p<0.001). Odds ratios are interpreted as the likelihood of achieving at an Excellent level (Mostly A’s or Mostly A’s and B’s) compared to a ‘Fair’ level (Mostly B’c and C’s and below). Lifestyle behaviours (physical activity, diet, sleep, and screen time) and confounders (age, sex, and SES) were first considered individually in a univariable analysis to assess unadjusted effects, and then were included together in a fully adjusted model.

Students with incomplete data who were excluded from analysis had significantly lower socioeconomic status, were more likely to have the recommended levels of physical activity, higher Junk Foods and Drinks scores, and more screen time than those with complete data. Table 2 describes the demographic characteristics of participating students within the HBSC survey who were included in the present analysis. Most students (76.1%) reported having Excellent grades. Most students (45.6%) reported getting 60 minutes of physical activity 3–5 days per week. Approximately two-thirds of children met recommendations for sleep (66.3%). Only 11.6% of students reported meeting recommendations for screen time (<2 hours per day), while 47.0% of students reported getting 7+ hours of screen time per day. Finally, the majority of students reported heights and weights that resulted in a normal body mass index (69.4%). Three percent of students were severely thin or thin, 18.9% were categorized as overweight, and 8.7% were obese.

https://doi.org/10.1371/journal.pone.0181938.t002

Table 3 shows the associations of physical activity, dietary factor scores, sleep, screen time, and body mass index with academic achievement. The univariable column represents unadjusted odds ratios between individual factors and likelihood of achieving Excellent grades. Fewer days where adequate physical activity was achieved, increasing junk food score, not meeting sleep recommendations, a high number of hours a day spent using screens, and being overweight or obese were negatively associated with likelihood of achieving a higher academic grading level. Youth who reported achieving 60 minutes of physical activity for 0–2 days per week had half the odds of achieving Excellent grades compared to children who had the recommended 60 minutes of physical activity 6–7 days per week ( Table 3 , Univariable Model, OR: 0.49: 95%CI [0.45, 0.54]). Children who reported 7+ hours of screen time per day had 40% reduced odds of achieving Excellent grades compared to those who met the recommended <2 hours per day ( Table 3 , Univariable Model: OR: 0.60 [0.54, 0.67]). Having either too short or too long of a sleep duration compared to the recommended hours per night was associated with 0.67 times the odds of achieving Excellent grades ( Table 3 , Univariable Model: OR: 0.67 [95%CI: 0.62, 0.72]). Higher consumption of Vegetables, Pulses, and Fruits higher scores in Healthy Eating Habits and moderate levels of screen time were positively associated with achieving Excellent grades. Every one unit increase in Healthy Eating Habits score resulting in 1.37 times the odds of achieving at a level of Excellent. Students who reported 2–4 hours per day of screen time had 1.23 times the odds of achieving Excellent grades compared to students who met the recommended amount of screen time of <2 hours per day ( Table 3 , Multivariable Model: OR: 1.23 [95%CI: 1.08, 1.41]).

https://doi.org/10.1371/journal.pone.0181938.t003

The multivariable results in Table 3 represent odds ratios that are fully adjusted for all variables of interest and potential confounders considered in the analysis. The negative association between fewer days of physical activity and academic achievement continued, although the effects were slightly attenuated. Compared to youth who reported getting 60 minutes per day of physical activity 6 or 7 days a week, youth who reported getting 3–5 days or 0–2 days had 0.89 and 0.65 times the odds of achieving Excellent grades, respectively ( Table 3 , Multivariable Model: OR: 0.89, [95%CI 0.80, 0.99], and OR: 0.65, [95%CI 0.56, 0.77]). Associations between dietary aspects and academic achievement remained consistent with univariable results. For every unit increase in score for Junk Foods and Drinks, students had 0.74 times the odds of achieving Excellent grades ( Table 3 , Multivariable Model: OR: 0.74 [95% CI: 0.70, 0.77]). For every unit increase in Vegetables, Pulses, and Fruit score and Healthy Eating Habits score, students had 1.22 and 1.55 times the odds of achieving Excellent grades, respectively. Students with inadequate sleep duration had 0.85 times the odds of achieving Excellent grades compared to those with adequate sleep duration ( Table 3 , Multivariable Model: OR: 0.85 [0.77, 0.93]). Significant univariable associations between screen time and academic achievement were completely attenuated following adjustment for other health behaviours and confounders.

In early adolescents from Canada, we found that all of physical activity, diet, sleep, and screen time had independent effects on academic achievement. We observed that increased consumption of vegetables, pulses, and fruits, and more regular healthy eating habits, were positively associated with higher academic achievement. Fewer days of achieving adequate physical activity, increased consumption of junk foods and drinks, not meeting recommendations for sleep duration, and overweight and obesity were negatively associated with higher academic achievement. This is the largest study to date, of the few that have been conducted, to consider all of these predictors in their independent relationships with academic achievement. The findings about the independent associations of lifestyle behaviours with academic achievement are consistent with the limited studies that have previously investigated this objective.

Students who had fewer days of achieving the recommended amounts of physical activity in a week had decreased likelihood of achieving Excellent grades. Students achieving the fewest days of adequate physical activity had 33% reduced odds of achieving Excellent grades compared to students achieving the highest number of days. Studies investigating the importance of physical activity and physical fitness for academic achievement have been predominantly indicative of a positive relationship between higher levels of physical activity and academic achievement, although there have been some inconsistencies [ 32 – 34 ]. This study supports previous findings that suggest a positive, linear relationship between physical activity and academic achievement [ 34 – 36 ]. The positive influence of physical activity on academic achievement, independent of other lifestyle behaviours, is consistent with findings from Ickovics et al (2014), Martinez-Gomez et al (2012), and Vassiloudis et al (2014), but inconsistent with those of our team’s previous work in Nova Scotia, Canada, which found a null relationship between physical activity and academic achievement once other lifestyle behaviours were considered [ 19 – 22 , 37 ].

The findings from the present study about diet are largely congruent with previous findings about the relationship between diet and academic achievement. Consumption of vegetables and fruit has consistently had a positive association with academic achievement [ 38 , 39 ] and junk foods, including beverages and snacks high in sugar and fat and frequent consumption of fast foods, have a similar negative association with academic achievement [ 39 , 40 ]. In addition, breakfast consumption has been highly studied in its relationship with academic achievement; the present findings about Healthy Eating Habits, which comprise regular breakfast consumption, are consistent with the positive findings of previous work [ 41 , 42 ]. However, although regular consumption of meals with the family has been shown to be beneficial for children’s diet and reduce likelihood of overweight and obesity [ 43 , 44 ], few studies have considered this in relation to academic achievement. In the present study, of all measures of diet included in this study (Vegetables, Pulses, and Fruit, Healthy Eating Habits, and Junk Foods and Beverages), Healthy Eating Habits, which include regularly consuming breakfast and dinner, and doing so in the presence of family, had the strongest positive relationship (50% increased odds with more frequent healthy habits) with academic achievement when considered simultaneously with other aspects of the diet, similar to findings from Ickovics et al (2014) who also looked at diet, physical activity, sleep, screen time and body weight status with academic achievement in a sample of urban American youth [ 20 ]. Frequently eating meals in the presence of family is associated with better psychosocial well-being among children which may be a contributing factor to better academic success [ 45 ].

Not meeting recommendations for sleep (representing having a sleep duration that is either too short or too long compared to recommendations) was negatively associated with academic achievement in youth. These findings are consistent with previous literature [ 46 ]. Because increased screen time has been shown to negatively affect sleep as well as academic achievement [ 47 ], it is possible that negative relationships between poor sleep and academic achievement are actually attributed to increased screen time [ 22 ]. However, the present findings demonstrate an independent, negative association of sleep with academic achievement, while screen time was found to have no association with academic achievement following adjustment for sleep and other. The lack of association of high levels of screen time with academic achievement is inconsistent with the conclusions recent systematic reviews, in particular for observational studies [ 9 ]. However, it has been acknowledged that self-reported screen time is of lower validity for the measurement of screen time or sedentary behavior, and though the majority of existing results indicate a positive association, there is still substantial inconsistency among cross-sectional studies and very few longitudinal studies using objective measures [ 9 ]. Future studies using objective measurements of sedentary behavior, taking into consideration sleep habits, and longitudinal analyses would benefit this area of the literature and clarify this relationship.

Overweight and obesity had strong, significant relationships with academic achievement independent of lifestyle behaviours. Although there are many studies investigating the relationship between body weight status and academic achievement, there are few that have taken the lifestyle behaviours that contribute to both weight status and academic achievement into account [ 12 ]. Of the existing studies [ 19 – 21 , 37 ] three of the four have found that overweight or obesity no longer had an association with academic achievement following the inclusion of lifestyle behaviours into an associative model. The results from the present study are inconsistent with the findings of these three studies and consistent with the fourth that did find an association. The present study used self-reported measures of academic achievement and body weight status, the work of Martinez-Gomez et al. (2012) used self-reported grades and measured body weight status [ 19 ], and the work of Vassiloudis et al (2014) used teacher-assessed grades and measured body weight status [ 21 ]. The studies by Wang et al. (2008) and Faught et al. (2017) used both objectively measured academic achievement and body weight status [ 22 , 37 ]. The present study categorized body weight status using the WHO Growth Reference standards [ 30 ], while the other four studies used the International Obesity Task Force (IOTF) reference standards [ 48 ]. The distinction between these two methods is significant–the WHO Growth Reference standards are generated in reference to an international cohort raised in strictly controlled, idealistic conditions, while the IOTF is meant to be representative of how an international cohort of children and youth grow under average circumstances [ 31 , 48 , 49 ]. Using a nationally representative sample of Canadian children and youth, Shields et al (2010) calculated childhood obesity rates using both IOTF and WHO reference cutoffs, and found that the WHO estimates of overweight and obesity were consistently and substantially higher than IOTF estimates. For boys 6–11, estimates of overweight including obesity using the WHO references were 14.0% higher than estimates using the IOTF cutpoints [ 49 ]. Results from the comparisons conducted by Shields et al and those from the present study are indicative that the use of different cutpoints across studies of the same objective may contribute to different results. The study by Vassiloudis et al (2014) was comprised of a sample of students with a much higher prevalence of obesity than those considered by the other students that was more comparable with the findings of the present study [ 21 ]. Consideration must be taken in deciding which cutpoints to employ, as well as prevalence of obesity, and how these may affect results and comparisons with existing literature. Regardless, the health promotion messages remain the same: the academic achievement of students of any body weight status will benefit from more frequent physical activity, a better diet, adequate sleep, and reduced screen time.

The present study’s strengths include a large, representative sample of Canadian children and youth with extensive data on their lifestyle behaviours, academic achievement, and socioeconomic status. Although statistical significance for small effect sizes can be found in studies using large datasets, the findings from this thesis provide compelling evidence of a strong relationship between lifestyle behaviours and academic achievement. However, this study is cross-sectional in nature so no statement of causality can be made from these results. In addition, there is potential for health behaviours to moderate the association between socioeconomic status and academic achievement. Further studies of this objective using longitudinal datasets are required to provide stronger evidence about relationships rather than associations, and to assess mediating and moderating effects. While measures are validated, the questions are brief to reduce participant burden and do not provide the depth of information that more intensive measurement would provide for each item. In addition, the collection of health and academic achievement data simultaneously may results in some differential bias as both health and academic achievement data may be overestimated by adolescents.

The present study contributes to the literature investigating the relationship among lifestyle behaviours, body weight status, and academic achievement. Findings from this study, and in the context of other studies with a similar objective, indicate that healthier lifestyle behaviours, regardless of body weight status, SES, and gender, contribute to better academic achievement. As such, equitable, accessible interventions to improve lifestyle behaviours to improve academic achievement and reduce the likelihood of adverse subsequent health outcomes are needed. The Ottawa Charter for Health Promotion (1986) states that, “health is created and lived by people within the context of their everyday life; where they learn, work, play and love [ 50 ].” In light of the present study’s findings, health promoting interventions within children and youth’s key settings, like schools and family environments, present a potential opportunity to support both health and academic achievement among children and early adolescents. Future studies investigating these associations using objective measures and longitudinal approaches would be valuable to better understand the potential causal relationship between health behaviours and academic achievement as well as investigate potential mediation and moderation of relationships between various factors and academic achievement.

Acknowledgments

Thank you to all of the participants who contributed their information to the Health Behaviours of School-Aged Children (HBSC) Study. We would like to acknowledge the International Coordinator for the HBSC Survey, Dr. Jo Inchley, University of St. Andrews, Scotland, and the international databank manager, Dr. Oddrun Samdal, University of Bergen, Norway. We would also like to acknowledge the Canadian Principal Investigators of HBSC, Drs. John Freeman and William Pickett, Queen's University, and the HBSC national coordinator, Matthew King. Thank you to Dr. John Freeman for overseeing our usage of this data and for your review of this manuscript. Thank you to Matthew King for generating this dataset for our use and assistance in interpreting the variables. Thank you to Dr. John Paul Ekwaru for your advice on the methodology used for this analysis.

• View Article
• PubMed/NCBI
• Google Scholar
• 4. Report of the Commission on Ending Childhood Obesity [Internet]. Geneva: World Health Organization; 2016 Jan. Available: http://apps.who.int/iris/bitstream/10665/204176/1/9789241510066_eng.pdf?ua=1
• 5. Public Health Agency of Canada. Diabetes in Canada: Facts and Figures from a public health perspective. Available: http://www.phac-aspc.gc.ca/cd-mc/publications/diabetes-diabete/facts-figures-faits-chiffres-2011/chap5-eng.php#endnote5
• 30. Growth Reference 5–19 years. World Health Organization. Available: http://www.who.int/growthref/who2007_bmi_for_age/en/

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• My Account Login
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 06 December 2017

Healthy food choices are happy food choices: Evidence from a real life sample using smartphone based assessments

• Deborah R. Wahl 1   na1 ,
• Karoline Villinger 1   na1 ,
• Laura M. König   ORCID: orcid.org/0000-0003-3655-8842 1 ,
• Katrin Ziesemer 1 ,
• Harald T. Schupp 1 &
• Britta Renner 1  

Scientific Reports volume  7 , Article number:  17069 ( 2017 ) Cite this article

132k Accesses

54 Citations

262 Altmetric

Metrics details

• Health sciences
• Human behaviour

Research suggests that “healthy” food choices such as eating fruits and vegetables have not only physical but also mental health benefits and might be a long-term investment in future well-being. This view contrasts with the belief that high-caloric foods taste better, make us happy, and alleviate a negative mood. To provide a more comprehensive assessment of food choice and well-being, we investigated in-the-moment eating happiness by assessing complete, real life dietary behaviour across eight days using smartphone-based ecological momentary assessment. Three main findings emerged: First, of 14 different main food categories, vegetables consumption contributed the largest share to eating happiness measured across eight days. Second, sweets on average provided comparable induced eating happiness to “healthy” food choices such as fruits or vegetables. Third, dinner elicited comparable eating happiness to snacking. These findings are discussed within the “food as health” and “food as well-being” perspectives on eating behaviour.

Similar content being viewed by others

Effects of single plant-based vs. animal-based meals on satiety and mood in real-world smartphone-embedded studies

Eating Style and the Frequency, Size and Timing of Eating Occasions: A cross-sectional analysis using 7-day weighed dietary records

Interindividual variability in appetitive sensations and relationships between appetitive sensations and energy intake

Introduction.

When it comes to eating, researchers, the media, and policy makers mainly focus on negative aspects of eating behaviour, like restricting certain foods, counting calories, and dieting. Likewise, health intervention efforts, including primary prevention campaigns, typically encourage consumers to trade off the expected enjoyment of hedonic and comfort foods against health benefits 1 . However, research has shown that diets and restrained eating are often counterproductive and may even enhance the risk of long-term weight gain and eating disorders 2 , 3 . A promising new perspective entails a shift from food as pure nourishment towards a more positive and well-being centred perspective of human eating behaviour 1 , 4 , 5 . In this context, Block et al . 4 have advocated a paradigm shift from “food as health” to “food as well-being” (p. 848).

Supporting this perspective of “food as well-being”, recent research suggests that “healthy” food choices, such as eating more fruits and vegetables, have not only physical but also mental health benefits 6 , 7 and might be a long-term investment in future well-being 8 . For example, in a nationally representative panel survey of over 12,000 adults from Australia, Mujcic and Oswald 8 showed that fruit and vegetable consumption predicted increases in happiness, life satisfaction, and well-being over two years. Similarly, using lagged analyses, White and colleagues 9 showed that fruit and vegetable consumption predicted improvements in positive affect on the subsequent day but not vice versa. Also, cross-sectional evidence reported by Blanchflower et al . 10 shows that eating fruits and vegetables is positively associated with well-being after adjusting for demographic variables including age, sex, or race 11 . Of note, previous research includes a wide range of time lags between actual eating occasion and well-being assessment, ranging from 24 hours 9 , 12 to 14 days 6 , to 24 months 8 . Thus, the findings support the notion that fruit and vegetable consumption has beneficial effects on different indicators of well-being, such as happiness or general life satisfaction, across a broad range of time spans.

The contention that healthy food choices such as a higher fruit and vegetable consumption is associated with greater happiness and well-being clearly contrasts with the common belief that in particular high-fat, high-sugar, or high-caloric foods taste better and make us happy while we are eating them. When it comes to eating, people usually have a spontaneous “unhealthy = tasty” association 13 and assume that chocolate is a better mood booster than an apple. According to this in-the-moment well-being perspective, consumers have to trade off the expected enjoyment of eating against the health costs of eating unhealthy foods 1 , 4 .

A wealth of research shows that the experience of negative emotions and stress leads to increased consumption in a substantial number of individuals (“emotional eating”) of unhealthy food (“comfort food”) 14 , 15 , 16 , 17 . However, this research stream focuses on emotional eating to “smooth” unpleasant experiences in response to stress or negative mood states, and the mood-boosting effect of eating is typically not assessed 18 . One of the few studies testing the effectiveness of comfort food in improving mood showed that the consumption of “unhealthy” comfort food had a mood boosting effect after a negative mood induction but not to a greater extent than non-comfort or neutral food 19 . Hence, even though people may believe that snacking on “unhealthy” foods like ice cream or chocolate provides greater pleasure and psychological benefits, the consumption of “unhealthy” foods might not actually be more psychologically beneficial than other foods.

However, both streams of research have either focused on a single food category (fruit and vegetable consumption), a single type of meal (snacking), or a single eating occasion (after negative/neutral mood induction). Accordingly, it is unknown whether the boosting effect of eating is specific to certain types of food choices and categories or whether eating has a more general boosting effect that is observable after the consumption of both “healthy” and “unhealthy” foods and across eating occasions. Accordingly, in the present study, we investigated the psychological benefits of eating that varied by food categories and meal types by assessing complete dietary behaviour across eight days in real life.

Furthermore, previous research on the impact of eating on well-being tended to rely on retrospective assessments such as food frequency questionnaires 8 , 10 and written food diaries 9 . Such retrospective self-report methods rely on the challenging task of accurately estimating average intake or remembering individual eating episodes and may lead to under-reporting food intake, particularly unhealthy food choices such as snacks 7 , 20 . To avoid memory and bias problems in the present study we used ecological momentary assessment (EMA) 21 to obtain ecologically valid and comprehensive real life data on eating behaviour and happiness as experienced in-the-moment.

In the present study, we examined the eating happiness and satisfaction experienced in-the-moment, in real time and in real life, using a smartphone based EMA approach. Specifically, healthy participants were asked to record each eating occasion, including main meals and snacks, for eight consecutive days and rate how tasty their meal/snack was, how much they enjoyed it, and how pleased they were with their meal/snack immediately after each eating episode. This intense recording of every eating episode allows assessing eating behaviour on the level of different meal types and food categories to compare experienced eating happiness across meals and categories. Following the two different research streams, we expected on a food category level that not only “unhealthy” foods like sweets would be associated with high experienced eating happiness but also “healthy” food choices such as fruits and vegetables. On a meal type level, we hypothesised that the happiness of meals differs as a function of meal type. According to previous contention, snacking in particular should be accompanied by greater happiness.

Eating episodes

Overall, during the study period, a total of 1,044 completed eating episodes were reported (see also Table  1 ). On average, participants rated their eating happiness with M  = 77.59 which suggests that overall eating occasions were generally positive. However, experienced eating happiness also varied considerably between eating occasions as indicated by a range from 7.00 to 100.00 and a standard deviation of SD  = 16.41.

Food categories and experienced eating happiness

All eating episodes were categorised according to their food category based on the German Nutrient Database (German: Bundeslebensmittelschlüssel), which covers the average nutritional values of approximately 10,000 foods available on the German market and is a validated standard instrument for the assessment of nutritional surveys in Germany. As shown in Table  1 , eating happiness differed significantly across all 14 food categories, F (13, 2131) = 1.78, p  = 0.04. On average, experienced eating happiness varied from 71.82 ( SD  = 18.65) for fish to 83.62 ( SD  = 11.61) for meat substitutes. Post hoc analysis, however, did not yield significant differences in experienced eating happiness between food categories, p  ≥ 0.22. Hence, on average, “unhealthy” food choices such as sweets ( M  = 78.93, SD  = 15.27) did not differ in experienced happiness from “healthy” food choices such as fruits ( M  = 78.29, SD  = 16.13) or vegetables ( M  = 77.57, SD  = 17.17). In addition, an intraclass correlation (ICC) of ρ = 0.22 for happiness indicated that less than a quarter of the observed variation in experienced eating happiness was due to differences between food categories, while 78% of the variation was due to differences within food categories.

However, as Figure  1 (left side) depicts, consumption frequency differed greatly across food categories. Frequently consumed food categories encompassed vegetables which were consumed at 38% of all eating occasions ( n  = 400), followed by dairy products with 35% ( n  = 366), and sweets with 34% ( n  = 356). Conversely, rarely consumed food categories included meat substitutes, which were consumed in 2.2% of all eating occasions ( n  = 23), salty extras (1.5%, n  = 16), and pastries (1.3%, n  = 14).

Left side: Average experienced eating happiness (colour intensity: darker colours indicate greater happiness) and consumption frequency (size of the cycle) for the 14 food categories. Right side: Absolute share of the 14 food categories in total experienced eating happiness.

Amount of experienced eating happiness by food category

To account for the frequency of consumption, we calculated and scaled the absolute experienced eating happiness according to the total sum score. As shown in Figure  1 (right side), vegetables contributed the biggest share to the total happiness followed by sweets, dairy products, and bread. Clustering food categories shows that fruits and vegetables accounted for nearly one quarter of total eating happiness score and thus, contributed to a large part of eating related happiness. Grain products such as bread, pasta, and cereals, which are main sources of carbohydrates including starch and fibre, were the second main source for eating happiness. However, “unhealthy” snacks including sweets, salty extras, and pastries represented the third biggest source of eating related happiness.

Experienced eating happiness by meal type

To further elucidate the contribution of snacks to eating happiness, analysis on the meal type level was conducted. Experienced in-the-moment eating happiness significantly varied by meal type consumed, F (4, 1039) = 11.75, p  < 0.001. Frequencies of meal type consumption ranged from snacks being the most frequently logged meal type ( n  = 332; see also Table  1 ) to afternoon tea being the least logged meal type ( n  = 27). Figure  2 illustrates the wide dispersion within as well as between different meal types. Afternoon tea ( M  = 82.41, SD  = 15.26), dinner ( M  = 81.47, SD  = 14.73), and snacks ( M  = 79.45, SD  = 14.94) showed eating happiness values above the grand mean, whereas breakfast ( M  = 74.28, SD  = 16.35) and lunch ( M  = 73.09, SD  = 18.99) were below the eating happiness mean. Comparisons between meal types showed that eating happiness for snacks was significantly higher than for lunch t (533) = −4.44, p  = 0.001, d  = −0.38 and breakfast, t (567) = −3.78, p  = 0.001, d  = −0.33. However, this was also true for dinner, which induced greater eating happiness than lunch t (446) = −5.48, p  < 0.001, d  = −0.50 and breakfast, t (480) = −4.90, p  < 0.001, d  = −0.46. Finally, eating happiness for afternoon tea was greater than for lunch t (228) = −2.83, p  = 0.047, d  = −0.50. All other comparisons did not reach significance, t  ≤ 2.49, p  ≥ 0.093.

Experienced eating happiness per meal type. Small dots represent single eating events, big circles indicate average eating happiness, and the horizontal line indicates the grand mean. Boxes indicate the middle 50% (interquartile range) and median (darker/lighter shade). The whiskers above and below represent 1.5 of the interquartile range.

Control Analyses

In order to test for a potential confounding effect between experienced eating happiness, food categories, and meal type, additional control analyses within meal types were conducted. Comparing experienced eating happiness for dinner and lunch suggested that dinner did not trigger a happiness spill-over effect specific to vegetables since the foods consumed at dinner were generally associated with greater happiness than those consumed at other eating occasions (Supplementary Table  S1 ). Moreover, the relative frequency of vegetables consumed at dinner (73%, n  = 180 out of 245) and at lunch were comparable (69%, n  = 140 out of 203), indicating that the observed happiness-vegetables link does not seem to be mainly a meal type confounding effect.

Since the present study focuses on “food effects” (Level 1) rather than “person effects” (Level 2), we analysed the data at the food item level. However, participants who were generally overall happier with their eating could have inflated the observed happiness scores for certain food categories. In order to account for person-level effects, happiness scores were person-mean centred and thereby adjusted for mean level differences in happiness. The person-mean centred happiness scores ( M cwc ) represent the difference between the individual’s average happiness score (across all single in-the-moment happiness scores per food category) and the single happiness scores of the individual within the respective food category. The centred scores indicate whether the single in-the-moment happiness score was above (indicated by positive values) or below (indicated by negative values) the individual person-mean. As Table  1 depicts, the control analyses with centred values yielded highly similar results. Vegetables were again associated on average with more happiness than other food categories (although people might differ in their general eating happiness). An additional conducted ANOVA with person-centred happiness values as dependent variables and food categories as independent variables provided also a highly similar pattern of results. Replicating the previously reported analysis, eating happiness differed significantly across all 14 food categories, F (13, 2129) = 1.94, p  = 0.023, and post hoc analysis did not yield significant differences in experienced eating happiness between food categories, p  ≥ 0.14. Moreover, fruits and vegetables were associated with high happiness values, and “unhealthy” food choices such as sweets did not differ in experienced happiness from “healthy” food choices such as fruits or vegetables. The only difference between the previous and control analysis was that vegetables ( M cwc  = 1.16, SD  = 15.14) gained slightly in importance for eating-related happiness, whereas fruits ( M cwc  = −0.65, SD  = 13.21), salty extras ( M cwc  = −0.07, SD  = 8.01), and pastries ( M cwc  = −2.39, SD  = 18.26) became slightly less important.

This study is the first, to our knowledge, that investigated in-the-moment experienced eating happiness in real time and real life using EMA based self-report and imagery covering the complete diversity of food intake. The present results add to and extend previous findings by suggesting that fruit and vegetable consumption has immediate beneficial psychological effects. Overall, of 14 different main food categories, vegetables consumption contributed the largest share to eating happiness measured across eight days. Thus, in addition to the investment in future well-being indicated by previous research 8 , “healthy” food choices seem to be an investment in the in-the moment well-being.

Importantly, although many cultures convey the belief that eating certain foods has a greater hedonic and mood boosting effect, the present results suggest that this might not reflect actual in-the-moment experiences accurately. Even though people often have a spontaneous “unhealthy = tasty” intuition 13 , thus indicating that a stronger happiness boosting effect of “unhealthy” food is to be expected, the induced eating happiness of sweets did not differ on average from “healthy” food choices such as fruits or vegetables. This was also true for other stereotypically “unhealthy” foods such as pastries and salty extras, which did not show the expected greater boosting effect on happiness. Moreover, analyses on the meal type level support this notion, since snacks, despite their overall positive effect, were not the most psychologically beneficial meal type, i.e., dinner had a comparable “happiness” signature to snacking. Taken together, “healthy choices” seem to be also “happy choices” and at least comparable to or even higher in their hedonic value as compared to stereotypical “unhealthy” food choices.

In general, eating happiness was high, which concurs with previous research from field studies with generally healthy participants. De Castro, Bellisle, and Dalix 22 examined weekly food diaries from 54 French subjects and found that most of the meals were rated as appealing. Also, the observed differences in average eating happiness for the 14 different food categories, albeit statistically significant, were comparable small. One could argue that this simply indicates that participants avoided selecting bad food 22 . Alternatively, this might suggest that the type of food or food categories are less decisive for experienced eating happiness than often assumed. This relates to recent findings in the field of comfort and emotional eating. Many people believe that specific types of food have greater comforting value. Also in research, the foods eaten as response to negative emotional strain, are typically characterised as being high-caloric because such foods are assumed to provide immediate psycho-physical benefits 18 . However, comparing different food types did not provide evidence for the notion that they differed in their provided comfort; rather, eating in general led to significant improvements in mood 19 . This is mirrored in the present findings. Comparing the eating happiness of “healthy” food choices such as fruits and vegetables to that of “unhealthy” food choices such as sweets shows remarkably similar patterns as, on average, they were associated with high eating happiness and their range of experiences ranged from very negative to very positive.

This raises the question of why the idea that we can eat indulgent food to compensate for life’s mishaps is so prevailing. In an innovative experimental study, Adriaanse, Prinsen, de Witt Huberts, de Ridder, and Evers 23 led participants believe that they overate. Those who characterised themselves as emotional eaters falsely attributed their over-consumption to negative emotions, demonstrating a “confabulation”-effect. This indicates that people might have restricted self-knowledge and that recalled eating episodes suffer from systematic recall biases 24 . Moreover, Boelsma, Brink, Stafleu, and Hendriks 25 examined postprandial subjective wellness and objective parameters (e.g., ghrelin, insulin, glucose) after standardised breakfast intakes and did not find direct correlations. This suggests that the impact of different food categories on wellness might not be directly related to biological effects but rather due to conditioning as food is often paired with other positive experienced situations (e.g., social interactions) or to placebo effects 18 . Moreover, experimental and field studies indicate that not only negative, but also positive, emotions trigger eating 15 , 26 . One may speculate that selective attention might contribute to the “myth” of comfort food 19 in that people attend to the consumption effect of “comfort” food in negative situation but neglect the effect in positive ones.

The present data also show that eating behaviour in the real world is a complex behaviour with many different aspects. People make more than 200 food decisions a day 27 which poses a great challenge for the measurement of eating behaviour. Studies often assess specific food categories such as fruit and vegetable consumption using Food Frequency Questionnaires, which has clear advantages in terms of cost-effectiveness. However, focusing on selective aspects of eating and food choices might provide only a selective part of the picture 15 , 17 , 22 . It is important to note that focusing solely on the “unhealthy” food choices such as sweets would have led to the conclusion that they have a high “indulgent” value. To be able to draw conclusions about which foods make people happy, the relation of different food categories needs to be considered. The more comprehensive view, considering the whole dietary behaviour across eating occasions, reveals that “healthy” food choices actually contributed the biggest share to the total experienced eating happiness. Thus, for a more comprehensive understanding of how eating behaviours are regulated, more complete and sensitive measures of the behaviour are necessary. Developments in mobile technologies hold great promise for feasible dietary assessment based on image-assisted methods 28 .

As fruits and vegetables evoked high in-the-moment happiness experiences, one could speculate that these cumulate and have spill-over effects on subsequent general well-being, including life satisfaction across time. Combing in-the-moment measures with longitudinal perspectives might be a promising avenue for future studies for understanding the pathways from eating certain food types to subjective well-being. In the literature different pathways are discussed, including physiological and biochemical aspects of specific food elements or nutrients 7 .

The present EMA based data also revealed that eating happiness varied greatly within the 14 food categories and meal types. As within food category variance represented more than two third of the total observed variance, happiness varied according to nutritional characteristics and meal type; however, a myriad of factors present in the natural environment can affect each and every meal. Thus, widening the “nourishment” perspective by including how much, when, where, how long, and with whom people eat might tell us more about experienced eating happiness. Again, mobile, in-the-moment assessment opens the possibility of assessing the behavioural signature of eating in real life. Moreover, individual factors such as eating motives, habitual eating styles, convenience, and social norms are likely to contribute to eating happiness variance 5 , 29 .

A key strength of this study is that it was the first to examine experienced eating happiness in non-clinical participants using EMA technology and imagery to assess food intake. Despite this strength, there are some limitations to this study that affect the interpretation of the results. In the present study, eating happiness was examined on a food based level. This neglects differences on the individual level and might be examined in future multilevel studies. Furthermore, as a main aim of this study was to assess real life eating behaviour, the “natural” observation level is the meal, the psychological/ecological unit of eating 30 , rather than food categories or nutrients. Therefore, we cannot exclude that specific food categories may have had a comparably higher impact on the experienced happiness of the whole meal. Sample size and therefore Type I and Type II error rates are of concern. Although the total number of observations was higher than in previous studies (see for example, Boushey et al . 28 for a review), the number of participants was small but comparable to previous studies in this field 20 , 31 , 32 , 33 . Small sample sizes can increase error rates because the number of persons is more decisive than the number of nested observations 34 . Specially, nested data can seriously increase Type I error rates, which is rather unlikely to be the case in the present study. Concerning Type II error rates, Aarts et al . 35 illustrated for lower ICCs that adding extra observations per participant also increases power, particularly in the lower observation range. Considering the ICC and the number of observations per participant, one could argue that the power in the present study is likely to be sufficient to render the observed null-differences meaningful. Finally, the predominately white and well-educated sample does limit the degree to which the results can be generalised to the wider community; these results warrant replication with a more representative sample.

Despite these limitations, we think that our study has implications for both theory and practice. The cumulative evidence of psychological benefits from healthy food choices might offer new perspectives for health promotion and public-policy programs 8 . Making people aware of the “healthy = happy” association supported by empirical evidence provides a distinct and novel perspective to the prevailing “unhealthy = tasty” folk intuition and could foster eating choices that increase both in-the-moment happiness and future well-being. Furthermore, the present research lends support to the advocated paradigm shift from “food as health” to “food as well-being” which entails a supporting and encouraging rather constraining and limiting view on eating behaviour.

The study conformed with the Declaration of Helsinki. All study protocols were approved by University of Konstanz’s Institutional Review Board and were conducted in accordance with guidelines and regulations. Upon arrival, all participants signed a written informed consent.

Participants

Thirty-eight participants (28 females: average age = 24.47, SD  = 5.88, range = 18–48 years) from the University of Konstanz assessed their eating behaviour in close to real time and in their natural environment using an event-based ambulatory assessment method (EMA). No participant dropped out or had to be excluded. Thirty-three participants were students, with 52.6% studying psychology. As compensation, participants could choose between taking part in a lottery (4 × 25€) or receiving course credits (2 hours).

Participants were recruited through leaflets distributed at the university and postings on Facebook groups. Prior to participation, all participants gave written informed consent. Participants were invited to the laboratory for individual introductory sessions. During this first session, participants installed the application movisensXS (version 0.8.4203) on their own smartphones and downloaded the study survey (movisensXS Library v4065). In addition, they completed a short baseline questionnaire, including demographic variables like age, gender, education, and eating principles. Participants were instructed to log every eating occasion immediately before eating by using the smartphone to indicate the type of meal, take pictures of the food, and describe its main components using a free input field. Fluid intake was not assessed. Participants were asked to record their food intake on eight consecutive days. After finishing the study, participants were invited back to the laboratory for individual final interviews.

Immediately before eating participants were asked to indicate the type of meal with the following five options: breakfast, lunch, afternoon tea, dinner, snack. In Germany, “afternoon tea” is called “Kaffee & Kuchen” which directly translates as “coffee & cake”. It is similar to the idea of a traditional “afternoon tea” meal in UK. Specifically, in Germany, people have “Kaffee & Kuchen” in the afternoon (between 4–5 pm) and typically coffee (or tea) is served with some cake or cookies. Dinner in Germany is a main meal with mainly savoury food.

After each meal, participants were asked to rate their meal on three dimensions. They rated (1) how much they enjoyed the meal, (2) how pleased they were with their meal, and (3) how tasty their meal was. Ratings were given on a scale of one to 100. For reliability analysis, Cronbach’s Alpha was calculated to assess the internal consistency of the three items. Overall Cronbach’s alpha was calculated with α = 0.87. In addition, the average of the 38 Cronbach’s alpha scores calculated at the person level also yielded a satisfactory value with α = 0.83 ( SD  = 0.24). Thirty-two of 38 participants showed a Cronbach’s alpha value above 0.70 (range = 0.42–0.97). An overall score of experienced happiness of eating was computed using the average of the three questions concerning the meals’ enjoyment, pleasure, and tastiness.

Analytical procedure

The food pictures and descriptions of their main components provided by the participants were subsequently coded by independent and trained raters. Following a standardised manual, additional components displayed in the picture were added to the description by the raters. All consumed foods were categorised into 14 different food categories (see Table  1 ) derived from the food classification system designed by the German Nutrition Society (DGE) and based on the existing food categories of the German Nutrient Database (Max Rubner Institut). Liquid intake and preparation method were not assessed. Therefore, fats and additional recipe ingredients were not included in further analyses, because they do not represent main elements of food intake. Further, salty extras were added to the categorisation.

No participant dropped out or had to be excluded due to high missing rates. Missing values were below 5% for all variables. The compliance rate at the meal level cannot be directly assessed since the numbers of meals and snacks can vary between as well as within persons (between days). As a rough compliance estimate, the numbers of meals that are expected from a “normative” perspective during the eight observation days can be used as a comparison standard (8 x breakfast, 8 × lunch, 8 × dinner = 24 meals). On average, the participants reported M  = 6.3 breakfasts ( SD  = 2.3), M  = 5.3 lunches ( SD  = 1.8), and M  = 6.5 dinners ( SD  = 2.0). In comparison to the “normative” expected 24 meals, these numbers indicate a good compliance (approx. 75%) with a tendency to miss six meals during the study period (approx. 25%). However, the “normative” expected 24 meals for the study period might be too high since participants might also have skipped meals (e.g. breakfast). Also, the present compliance rates are comparable to other studies. For example, Elliston et al . 36 recorded 3.3 meal/snack reports per day in an Australian adult sample and Casperson et al . 37 recorded 2.2 meal reports per day in a sample of adolescents. In the present study, on average, M  = 3.4 ( SD  = 1.35) meals or snacks were reported per day. These data indicate overall a satisfactory compliance rate and did not indicate selective reporting of certain food items.

To graphically visualise data, Tableau (version 10.1) was used and for further statistical analyses, IBM SPSS Statistics (version 24 for Windows).

Data availability

The dataset generated and analysed during the current study is available from the corresponding authors on reasonable request.

Cornil, Y. & Chandon, P. Pleasure as an ally of healthy eating? Contrasting visceral and epicurean eating pleasure and their association with portion size preferences and wellbeing. Appetite 104 , 52–59 (2016).

Article   PubMed   Google Scholar  

Mann, T. et al . Medicare’s search for effective obesity treatments: Diets are not the answer. American Psychologist 62 , 220–233 (2007).

van Strien, T., Herman, C. P. & Verheijden, M. W. Dietary restraint and body mass change. A 3-year follow up study in a representative Dutch sample. Appetite 76 , 44–49 (2014).

Block, L. G. et al . From nutrients to nurturance: A conceptual introduction to food well-being. Journal of Public Policy & Marketing 30 , 5–13 (2011).

Article   Google Scholar  

Renner, B., Sproesser, G., Strohbach, S. & Schupp, H. T. Why we eat what we eat. The eating motivation survey (TEMS). Appetite 59 , 117–128 (2012).

Conner, T. S., Brookie, K. L., Carr, A. C., Mainvil, L. A. & Vissers, M. C. Let them eat fruit! The effect of fruit and vegetable consumption on psychological well-being in young adults: A randomized controlled trial. PloS one 12 , e0171206 (2017).

Article   PubMed   PubMed Central   Google Scholar  

Rooney, C., McKinley, M. C. & Woodside, J. V. The potential role of fruit and vegetables in aspects of psychological well-being: a review of the literature and future directions. Proceedings of the Nutrition Society 72 , 420–432 (2013).

Mujcic, R. & Oswald, A. J. Evolution of well-being and happiness after increases in consumption of fruit and vegetables. American Journal of Public Health 106 , 1504–1510 (2016).

White, B. A., Horwath, C. C. & Conner, T. S. Many apples a day keep the blues away – Daily experiences of negative and positive affect and food consumption in young adults. British Journal of Health Psychology 18 , 782–798 (2013).

Blanchflower, D. G., Oswald, A. J. & Stewart-Brown, S. Is psychological well-being linked to the consumption of fruit and vegetables? Social Indicators Research 114 , 785–801 (2013).

Grant, N., Wardle, J. & Steptoe, A. The relationship between life satisfaction and health behavior: A Cross-cultural analysis of young adults. International Journal of Behavioral Medicine 16 , 259–268 (2009).

Conner, T. S., Brookie, K. L., Richardson, A. C. & Polak, M. A. On carrots and curiosity: Eating fruit and vegetables is associated with greater flourishing in daily life. British Journal of Health Psychology 20 , 413–427 (2015).

Raghunathan, R., Naylor, R. W. & Hoyer, W. D. The unhealthy = tasty intuition and its effects on taste inferences, enjoyment, and choice of food products. Journal of Marketing 70 , 170–184 (2006).

Evers, C., Stok, F. M. & de Ridder, D. T. Feeding your feelings: Emotion regulation strategies and emotional eating. Personality and Social Psychology Bulletin 36 , 792–804 (2010).

Sproesser, G., Schupp, H. T. & Renner, B. The bright side of stress-induced eating: eating more when stressed but less when pleased. Psychological Science 25 , 58–65 (2013).

Wansink, B., Cheney, M. M. & Chan, N. Exploring comfort food preferences across age and gender. Physiology & Behavior 79 , 739–747 (2003).

Article   CAS   Google Scholar  

Taut, D., Renner, B. & Baban, A. Reappraise the situation but express your emotions: impact of emotion regulation strategies on ad libitum food intake. Frontiers in Psychology 3 , 359 (2012).

Tomiyama, J. A., Finch, L. E. & Cummings, J. R. Did that brownie do its job? Stress, eating, and the biobehavioral effects of comfort food. Emerging Trends in the Social and Behavioral Sciences: An Interdisciplinary, Searchable, and Linkable Resource (2015).

Wagner, H. S., Ahlstrom, B., Redden, J. P., Vickers, Z. & Mann, T. The myth of comfort food. Health Psychology 33 , 1552–1557 (2014).

Schüz, B., Bower, J. & Ferguson, S. G. Stimulus control and affect in dietary behaviours. An intensive longitudinal study. Appetite 87 , 310–317 (2015).

Shiffman, S. Conceptualizing analyses of ecological momentary assessment data. Nicotine & Tobacco Research 16 , S76–S87 (2014).

de Castro, J. M., Bellisle, F. & Dalix, A.-M. Palatability and intake relationships in free-living humans: measurement and characterization in the French. Physiology & Behavior 68 , 271–277 (2000).

Adriaanse, M. A., Prinsen, S., de Witt Huberts, J. C., de Ridder, D. T. & Evers, C. ‘I ate too much so I must have been sad’: Emotions as a confabulated reason for overeating. Appetite 103 , 318–323 (2016).

Robinson, E. Relationships between expected, online and remembered enjoyment for food products. Appetite 74 , 55–60 (2014).

Boelsma, E., Brink, E. J., Stafleu, A. & Hendriks, H. F. Measures of postprandial wellness after single intake of two protein–carbohydrate meals. Appetite 54 , 456–464 (2010).

Article   CAS   PubMed   Google Scholar  

Boh, B. et al . Indulgent thinking? Ecological momentary assessment of overweight and healthy-weight participants’ cognitions and emotions. Behaviour Research and Therapy 87 , 196–206 (2016).

Wansink, B. & Sobal, J. Mindless eating: The 200 daily food decisions we overlook. Environment and Behavior 39 , 106–123 (2007).

Boushey, C., Spoden, M., Zhu, F., Delp, E. & Kerr, D. New mobile methods for dietary assessment: review of image-assisted and image-based dietary assessment methods. Proceedings of the Nutrition Society , 1–12 (2016).

Stok, F. M. et al . The DONE framework: Creation, evaluation, and updating of an interdisciplinary, dynamic framework 2.0 of determinants of nutrition and eating. PLoS ONE 12 , e0171077 (2017).

Pliner, P. & Rozin, P. In Dimensions of the meal: The science, culture, business, and art of eating (ed H Meiselman) 19–46 (Aspen Publishers, 2000).

Inauen, J., Shrout, P. E., Bolger, N., Stadler, G. & Scholz, U. Mind the gap? Anintensive longitudinal study of between-person and within-person intention-behaviorrelations. Annals of Behavioral Medicine 50 , 516–522 (2016).

Zepeda, L. & Deal, D. Think before you eat: photographic food diaries asintervention tools to change dietary decision making and attitudes. InternationalJournal of Consumer Studies 32 , 692–698 (2008).

Stein, K. F. & Corte, C. M. Ecologic momentary assessment of eating‐disordered behaviors. International Journal of Eating Disorders 34 , 349–360 (2003).

Bolger, N., Stadler, G. & Laurenceau, J. P. Power analysis for intensive longitudinal studies in Handbook of research methods for studying daily life (ed . Mehl, M. R. & Conner, T. S.) 285–301 (New York: The Guilford Press, 2012).

Aarts, E., Verhage, M., Veenvliet, J. V., Dolan, C. V. & Van Der Sluis, S. A solutionto dependency: using multilevel analysis to accommodate nested data. Natureneuroscience 17 , 491–496 (2014).

Elliston, K. G., Ferguson, S. G., Schüz, N. & Schüz, B. Situational cues andmomentary food environment predict everyday eating behavior in adults withoverweight and obesity. Health Psychology 36 , 337–345 (2017).

Casperson, S. L. et al . A mobile phone food record app to digitally capture dietary intake for adolescents in afree-living environment: usability study. JMIR mHealth and uHealth 3 , e30 (2015).

Download references

Acknowledgements

This research was supported by the Federal Ministry of Education and Research within the project SmartAct (Grant 01EL1420A, granted to B.R. & H.S.). The funding source had no involvement in the study’s design; the collection, analysis, and interpretation of data; the writing of the report; or the decision to submit this article for publication. We thank Gudrun Sproesser, Helge Giese, and Angela Whale for their valuable support.

Author information

Deborah R. Wahl and Karoline Villinger contributed equally to this work.

Authors and Affiliations

Department of Psychology, University of Konstanz, Konstanz, Germany

Deborah R. Wahl, Karoline Villinger, Laura M. König, Katrin Ziesemer, Harald T. Schupp & Britta Renner

You can also search for this author in PubMed   Google Scholar

Contributions

B.R. & H.S. developed the study concept. All authors participated in the generation of the study design. D.W., K.V., L.K. & K.Z. conducted the study, including participant recruitment and data collection, under the supervision of B.R. & H.S.; D.W. & K.V. conducted data analyses. D.W. & K.V. prepared the first manuscript draft, and B.R. & H.S. provided critical revisions. All authors approved the final version of the manuscript for submission.

Corresponding authors

Correspondence to Deborah R. Wahl or Britta Renner .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Supplementary table s1, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Wahl, D.R., Villinger, K., König, L.M. et al. Healthy food choices are happy food choices: Evidence from a real life sample using smartphone based assessments. Sci Rep 7 , 17069 (2017). https://doi.org/10.1038/s41598-017-17262-9

Download citation

Received : 05 June 2017

Accepted : 23 November 2017

Published : 06 December 2017

DOI : https://doi.org/10.1038/s41598-017-17262-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Financial satisfaction, food security, and shared meals are foundations of happiness among older persons in thailand.

• Sirinya Phulkerd
• Rossarin Soottipong Gray
• Sasinee Thapsuwan

BMC Geriatrics (2023)

Enrichment and Conflict Between Work and Health Behaviors: New Scales for Assessing How Work Relates to Physical Exercise and Healthy Eating

• Sabine Sonnentag
• Maria U. Kottwitz
• Jette Völker

Occupational Health Science (2023)

The value of Bayesian predictive projection for variable selection: an example of selecting lifestyle predictors of young adult well-being

• A. Bartonicek
• S. R. Wickham
• T. S. Conner

BMC Public Health (2021)

Smartphone-Based Ecological Momentary Assessment of Well-Being: A Systematic Review and Recommendations for Future Studies

• Lianne P. de Vries
• Bart M. L. Baselmans
• Meike Bartels

Journal of Happiness Studies (2021)

Exploration of nutritional, antioxidative, antibacterial and anticancer status of Russula alatoreticula: towards valorization of a traditionally preferred unique myco-food

• Somanjana Khatua
• Surashree Sen Gupta
• Krishnendu Acharya

Journal of Food Science and Technology (2021)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• U.S. Department of Health & Human Services

• Virtual Tour
• Staff Directory
• En Español

You are here

Nih research matters.

May 8, 2018

Healthy habits can lengthen life

At a glance.

• Researchers found that people who maintained five healthy lifestyle factors lived more than a decade longer than those who didn’t maintain any of the five.
• The results suggest that Americans can increase the length of their lives and lower their disease risk by adopting a healthier lifestyle.

American don’t live as long as people in most other high-income countries. Heart disease and cancer are two of the most common preventable chronic diseases in the United States. An unhealthy lifestyle increases your risk for these and other chronic diseases that can lead to an early death.

To explore the effects of healthy habits on Americans’ health and lifespan, a team of scientists led by Frank Hu at the Harvard T.H. Chan School of Public Health analyzed data from more than 78,000 women and 44,000 men who participated in two nationwide surveys: the Nurses’ Health Study (NHS) and the Health Professionals Follow-up Study (HPFS). They used other data from the Centers for Disease Control and Prevention to estimate the distribution of lifestyle choices and death rates across the U.S. population. The research was supported in part by NIH’s National Heart Lung and Blood Institute (NHLBI) and National Cancer Institute (NCI). Results were published online in Circulation on April 30, 2018 .

The team collected data on five different low-risk lifestyle factors and compared health outcomes for those who adopted all five with those who didn’t adopt any. The five factors included maintaining a healthy eating pattern (getting the daily recommended amounts of vegetables, fruit, nuts, whole grains, polyunsaturated fatty acids, and omega-3 fatty acids and limiting red and processed meats, beverages with added sugar, trans fat, and sodium); not smoking; getting at least 3.5 hours of moderate to vigorous physical activity each week; drinking only moderate amounts of alcohol (one drink or less per day for women or two drinks or less per day for men); and maintaining a normal weight (body mass index between 18.5 and 24.9). The researchers also collected information about the participants’ medical history, such as heart disease, cancer, and diabetes, as well as when they died.

At age 50, women who didn’t adopt any of the five healthy habits were estimated to live on average until they were 79 years old and men until they were 75.5 years. In contrast, women who adopted all five healthy lifestyle habits lived 93.1 years and men lived 87.6 years.

Independently, each of the five healthy lifestyle factors significantly lowered the risk of total death, death from cancer, and death from heart disease.

“This study underscores the importance of following healthy lifestyle habits for improving longevity in the U.S. population,” Hu says. “However, adherence to healthy lifestyle habits is very low. Therefore, public policies should put more emphasis on creating healthy food, built, and social environments to support and promote healthy diet and lifestyles.”

—by Tianna Hicklin, Ph.D.

Related Links

• Calorie Restriction Slows Age-Related Epigenetic Changes
• Drugs Extend Healthy Lifespan in Mice
• Longevity Gene Linked to Better Brain Skills
• A Little Exercise Might Lengthen Life
• Take Charge of Your Health: A Guide for Teenagers
• Health Tips for Adults
• Health Tips for Older Adults
• Healthy Habits, Healthy Families
• Changing Your Habits for Better Health
• Can You Lengthen Your Life?
• Creating Healthy Habits: Make Better Choices Easier

References:  Impact of Healthy Lifestyle Factors on Life Expectancies in the US Population. Li Y, Pan A, Wang DD, Liu X, Dhana K, Franco OH, Kaptoge S, Di Angelantonio E, Stampfer M, Willett WC, Hu FB. Circulation . 2018 Apr 30. pii: CIRCULATIONAHA.117.032047. doi: 10.1161/CIRCULATIONAHA.117.032047. [Epub ahead of print] PMID: 29712712.

Funding:  NIH’s National Heart, Lung and Blood Institute (NHLBI) and National Cancer Institute (NCI); British Heart Foundation; UK Medical Research Council; National Key Research and Development Program of China; and American Heart Association.

Connect with Us

• More Social Media from NIH

Get science-backed answers as you write with Paperpal's Research feature

How to Write a Conclusion for Research Papers (with Examples)

The conclusion of a research paper is a crucial section that plays a significant role in the overall impact and effectiveness of your research paper. However, this is also the section that typically receives less attention compared to the introduction and the body of the paper. The conclusion serves to provide a concise summary of the key findings, their significance, their implications, and a sense of closure to the study. Discussing how can the findings be applied in real-world scenarios or inform policy, practice, or decision-making is especially valuable to practitioners and policymakers. The research paper conclusion also provides researchers with clear insights and valuable information for their own work, which they can then build on and contribute to the advancement of knowledge in the field.

The research paper conclusion should explain the significance of your findings within the broader context of your field. It restates how your results contribute to the existing body of knowledge and whether they confirm or challenge existing theories or hypotheses. Also, by identifying unanswered questions or areas requiring further investigation, your awareness of the broader research landscape can be demonstrated.

Remember to tailor the research paper conclusion to the specific needs and interests of your intended audience, which may include researchers, practitioners, policymakers, or a combination of these.

Table of Contents

What is a conclusion in a research paper, summarizing conclusion, editorial conclusion, externalizing conclusion, importance of a good research paper conclusion, how to write a conclusion for your research paper, research paper conclusion examples.

• How to write a research paper conclusion with Paperpal? 

Frequently Asked Questions

A conclusion in a research paper is the final section where you summarize and wrap up your research, presenting the key findings and insights derived from your study. The research paper conclusion is not the place to introduce new information or data that was not discussed in the main body of the paper. When working on how to conclude a research paper, remember to stick to summarizing and interpreting existing content. The research paper conclusion serves the following purposes: 1

• Warn readers of the possible consequences of not attending to the problem.
• Recommend specific course(s) of action.
• Restate key ideas to drive home the ultimate point of your research paper.
• Provide a “take-home” message that you want the readers to remember about your study.

Types of conclusions for research papers

In research papers, the conclusion provides closure to the reader. The type of research paper conclusion you choose depends on the nature of your study, your goals, and your target audience. I provide you with three common types of conclusions:

A summarizing conclusion is the most common type of conclusion in research papers. It involves summarizing the main points, reiterating the research question, and restating the significance of the findings. This common type of research paper conclusion is used across different disciplines.

An editorial conclusion is less common but can be used in research papers that are focused on proposing or advocating for a particular viewpoint or policy. It involves presenting a strong editorial or opinion based on the research findings and offering recommendations or calls to action.

An externalizing conclusion is a type of conclusion that extends the research beyond the scope of the paper by suggesting potential future research directions or discussing the broader implications of the findings. This type of conclusion is often used in more theoretical or exploratory research papers.

Align your conclusion’s tone with the rest of your research paper. Start Writing with Paperpal Now!  

The conclusion in a research paper serves several important purposes:

• Offers Implications and Recommendations : Your research paper conclusion is an excellent place to discuss the broader implications of your research and suggest potential areas for further study. It’s also an opportunity to offer practical recommendations based on your findings.
• Provides Closure : A good research paper conclusion provides a sense of closure to your paper. It should leave the reader with a feeling that they have reached the end of a well-structured and thought-provoking research project.
• Leaves a Lasting Impression : Writing a well-crafted research paper conclusion leaves a lasting impression on your readers. It’s your final opportunity to leave them with a new idea, a call to action, or a memorable quote.

Writing a strong conclusion for your research paper is essential to leave a lasting impression on your readers. Here’s a step-by-step process to help you create and know what to put in the conclusion of a research paper: 2

• Research Statement : Begin your research paper conclusion by restating your research statement. This reminds the reader of the main point you’ve been trying to prove throughout your paper. Keep it concise and clear.
• Key Points : Summarize the main arguments and key points you’ve made in your paper. Avoid introducing new information in the research paper conclusion. Instead, provide a concise overview of what you’ve discussed in the body of your paper.
• Address the Research Questions : If your research paper is based on specific research questions or hypotheses, briefly address whether you’ve answered them or achieved your research goals. Discuss the significance of your findings in this context.
• Significance : Highlight the importance of your research and its relevance in the broader context. Explain why your findings matter and how they contribute to the existing knowledge in your field.
• Implications : Explore the practical or theoretical implications of your research. How might your findings impact future research, policy, or real-world applications? Consider the “so what?” question.
• Future Research : Offer suggestions for future research in your area. What questions or aspects remain unanswered or warrant further investigation? This shows that your work opens the door for future exploration.
• Closing Thought : Conclude your research paper conclusion with a thought-provoking or memorable statement. This can leave a lasting impression on your readers and wrap up your paper effectively. Avoid introducing new information or arguments here.
• Proofread and Revise : Carefully proofread your conclusion for grammar, spelling, and clarity. Ensure that your ideas flow smoothly and that your conclusion is coherent and well-structured.

Write your research paper conclusion 2x faster with Paperpal. Try it now!

Remember that a well-crafted research paper conclusion is a reflection of the strength of your research and your ability to communicate its significance effectively. It should leave a lasting impression on your readers and tie together all the threads of your paper. Now you know how to start the conclusion of a research paper and what elements to include to make it impactful, let’s look at a research paper conclusion sample.

How to write a research paper conclusion with Paperpal?

A research paper conclusion is not just a summary of your study, but a synthesis of the key findings that ties the research together and places it in a broader context. A research paper conclusion should be concise, typically around one paragraph in length. However, some complex topics may require a longer conclusion to ensure the reader is left with a clear understanding of the study’s significance. Paperpal, an AI writing assistant trusted by over 800,000 academics globally, can help you write a well-structured conclusion for your research paper. 

• Sign Up or Log In: Create a new Paperpal account or login with your details.  
• Navigate to Features : Once logged in, head over to the features’ side navigation pane. Click on Templates and you’ll find a suite of generative AI features to help you write better, faster.  
• Generate an outline: Under Templates, select ‘Outlines’. Choose ‘Research article’ as your document type.  
• Select your section: Since you’re focusing on the conclusion, select this section when prompted.  
• Choose your field of study: Identifying your field of study allows Paperpal to provide more targeted suggestions, ensuring the relevance of your conclusion to your specific area of research. 
• Provide a brief description of your study: Enter details about your research topic and findings. This information helps Paperpal generate a tailored outline that aligns with your paper’s content. 
• Generate the conclusion outline: After entering all necessary details, click on ‘generate’. Paperpal will then create a structured outline for your conclusion, to help you start writing and build upon the outline.  
• Write your conclusion: Use the generated outline to build your conclusion. The outline serves as a guide, ensuring you cover all critical aspects of a strong conclusion, from summarizing key findings to highlighting the research’s implications. 
• Refine and enhance: Paperpal’s ‘Make Academic’ feature can be particularly useful in the final stages. Select any paragraph of your conclusion and use this feature to elevate the academic tone, ensuring your writing is aligned to the academic journal standards. 

By following these steps, Paperpal not only simplifies the process of writing a research paper conclusion but also ensures it is impactful, concise, and aligned with academic standards. Sign up with Paperpal today and write your research paper conclusion 2x faster .  

The research paper conclusion is a crucial part of your paper as it provides the final opportunity to leave a strong impression on your readers. In the research paper conclusion, summarize the main points of your research paper by restating your research statement, highlighting the most important findings, addressing the research questions or objectives, explaining the broader context of the study, discussing the significance of your findings, providing recommendations if applicable, and emphasizing the takeaway message. The main purpose of the conclusion is to remind the reader of the main point or argument of your paper and to provide a clear and concise summary of the key findings and their implications. All these elements should feature on your list of what to put in the conclusion of a research paper to create a strong final statement for your work.

A strong conclusion is a critical component of a research paper, as it provides an opportunity to wrap up your arguments, reiterate your main points, and leave a lasting impression on your readers. Here are the key elements of a strong research paper conclusion: 1. Conciseness : A research paper conclusion should be concise and to the point. It should not introduce new information or ideas that were not discussed in the body of the paper. 2. Summarization : The research paper conclusion should be comprehensive enough to give the reader a clear understanding of the research’s main contributions. 3 . Relevance : Ensure that the information included in the research paper conclusion is directly relevant to the research paper’s main topic and objectives; avoid unnecessary details. 4 . Connection to the Introduction : A well-structured research paper conclusion often revisits the key points made in the introduction and shows how the research has addressed the initial questions or objectives. 5. Emphasis : Highlight the significance and implications of your research. Why is your study important? What are the broader implications or applications of your findings? 6 . Call to Action : Include a call to action or a recommendation for future research or action based on your findings.

The length of a research paper conclusion can vary depending on several factors, including the overall length of the paper, the complexity of the research, and the specific journal requirements. While there is no strict rule for the length of a conclusion, but it’s generally advisable to keep it relatively short. A typical research paper conclusion might be around 5-10% of the paper’s total length. For example, if your paper is 10 pages long, the conclusion might be roughly half a page to one page in length.

In general, you do not need to include citations in the research paper conclusion. Citations are typically reserved for the body of the paper to support your arguments and provide evidence for your claims. However, there may be some exceptions to this rule: 1. If you are drawing a direct quote or paraphrasing a specific source in your research paper conclusion, you should include a citation to give proper credit to the original author. 2. If your conclusion refers to or discusses specific research, data, or sources that are crucial to the overall argument, citations can be included to reinforce your conclusion’s validity.

The conclusion of a research paper serves several important purposes: 1. Summarize the Key Points 2. Reinforce the Main Argument 3. Provide Closure 4. Offer Insights or Implications 5. Engage the Reader. 6. Reflect on Limitations

Remember that the primary purpose of the research paper conclusion is to leave a lasting impression on the reader, reinforcing the key points and providing closure to your research. It’s often the last part of the paper that the reader will see, so it should be strong and well-crafted.

• Makar, G., Foltz, C., Lendner, M., & Vaccaro, A. R. (2018). How to write effective discussion and conclusion sections. Clinical spine surgery, 31(8), 345-346.
• Bunton, D. (2005). The structure of PhD conclusion chapters.  Journal of English for academic purposes ,  4 (3), 207-224.

Paperpal is a comprehensive AI writing toolkit that helps students and researchers achieve 2x the writing in half the time. It leverages 21+ years of STM experience and insights from millions of research articles to provide in-depth academic writing, language editing, and submission readiness support to help you write better, faster.  

Get accurate academic translations, rewriting support, grammar checks, vocabulary suggestions, and generative AI assistance that delivers human precision at machine speed. Try for free or upgrade to Paperpal Prime starting at US$19 a month to access premium features, including consistency, plagiarism, and 30+ submission readiness checks to help you succeed.  

Experience the future of academic writing – Sign up to Paperpal and start writing for free!  

Related Reads:

• 5 Reasons for Rejection After Peer Review
• Ethical Research Practices For Research with Human Subjects

7 Ways to Improve Your Academic Writing Process

• Paraphrasing in Academic Writing: Answering Top Author Queries

Preflight For Editorial Desk: The Perfect Hybrid (AI + Human) Assistance Against Compromised Manuscripts

You may also like, academic editing: how to self-edit academic text with..., measuring academic success: definition & strategies for excellence, phd qualifying exam: tips for success , ai in education: it’s time to change the..., is it ethical to use ai-generated abstracts without..., what are journal guidelines on using generative ai..., quillbot review: features, pricing, and free alternatives, what is an academic paper types and elements , should you use ai tools like chatgpt for..., 9 steps to publish a research paper.

• Share full article

Advertisement

Supported by

Study Suggests Genetics as a Cause, Not Just a Risk, for Some Alzheimer’s

People with two copies of the gene variant APOE4 are almost certain to get Alzheimer’s, say researchers, who proposed a framework under which such patients could be diagnosed years before symptoms.

By Pam Belluck

Scientists are proposing a new way of understanding the genetics of Alzheimer’s that would mean that up to a fifth of patients would be considered to have a genetically caused form of the disease.

Currently, the vast majority of Alzheimer’s cases do not have a clearly identified cause. The new designation, proposed in a study published Monday, could broaden the scope of efforts to develop treatments, including gene therapy, and affect the design of clinical trials.

It could also mean that hundreds of thousands of people in the United States alone could, if they chose, receive a diagnosis of Alzheimer’s before developing any symptoms of cognitive decline, although there currently are no treatments for people at that stage.

The new classification would make this type of Alzheimer’s one of the most common genetic disorders in the world, medical experts said.

“This reconceptualization that we’re proposing affects not a small minority of people,” said Dr. Juan Fortea, an author of the study and the director of the Sant Pau Memory Unit in Barcelona, Spain. “Sometimes we say that we don’t know the cause of Alzheimer’s disease,” but, he said, this would mean that about 15 to 20 percent of cases “can be tracked back to a cause, and the cause is in the genes.”

The idea involves a gene variant called APOE4. Scientists have long known that inheriting one copy of the variant increases the risk of developing Alzheimer’s, and that people with two copies, inherited from each parent, have vastly increased risk.

The new study , published in the journal Nature Medicine, analyzed data from over 500 people with two copies of APOE4, a significantly larger pool than in previous studies. The researchers found that almost all of those patients developed the biological pathology of Alzheimer’s, and the authors say that two copies of APOE4 should now be considered a cause of Alzheimer’s — not simply a risk factor.

The patients also developed Alzheimer’s pathology relatively young, the study found. By age 55, over 95 percent had biological markers associated with the disease. By 65, almost all had abnormal levels of a protein called amyloid that forms plaques in the brain, a hallmark of Alzheimer’s. And many started developing symptoms of cognitive decline at age 65, younger than most people without the APOE4 variant.

“The critical thing is that these individuals are often symptomatic 10 years earlier than other forms of Alzheimer’s disease,” said Dr. Reisa Sperling, a neurologist at Mass General Brigham in Boston and an author of the study.

She added, “By the time they are picked up and clinically diagnosed, because they’re often younger, they have more pathology.”

People with two copies, known as APOE4 homozygotes, make up 2 to 3 percent of the general population, but are an estimated 15 to 20 percent of people with Alzheimer’s dementia, experts said. People with one copy make up about 15 to 25 percent of the general population, and about 50 percent of Alzheimer’s dementia patients.

The most common variant is called APOE3, which seems to have a neutral effect on Alzheimer’s risk. About 75 percent of the general population has one copy of APOE3, and more than half of the general population has two copies.

Alzheimer’s experts not involved in the study said classifying the two-copy condition as genetically determined Alzheimer’s could have significant implications, including encouraging drug development beyond the field’s recent major focus on treatments that target and reduce amyloid.

Dr. Samuel Gandy, an Alzheimer’s researcher at Mount Sinai in New York, who was not involved in the study, said that patients with two copies of APOE4 faced much higher safety risks from anti-amyloid drugs.

When the Food and Drug Administration approved the anti-amyloid drug Leqembi last year, it required a black-box warning on the label saying that the medication can cause “serious and life-threatening events” such as swelling and bleeding in the brain, especially for people with two copies of APOE4. Some treatment centers decided not to offer Leqembi, an intravenous infusion, to such patients.

Dr. Gandy and other experts said that classifying these patients as having a distinct genetic form of Alzheimer’s would galvanize interest in developing drugs that are safe and effective for them and add urgency to current efforts to prevent cognitive decline in people who do not yet have symptoms.

“Rather than say we have nothing for you, let’s look for a trial,” Dr. Gandy said, adding that such patients should be included in trials at younger ages, given how early their pathology starts.

Besides trying to develop drugs, some researchers are exploring gene editing to transform APOE4 into a variant called APOE2, which appears to protect against Alzheimer’s. Another gene-therapy approach being studied involves injecting APOE2 into patients’ brains.

The new study had some limitations, including a lack of diversity that might make the findings less generalizable. Most patients in the study had European ancestry. While two copies of APOE4 also greatly increase Alzheimer’s risk in other ethnicities, the risk levels differ, said Dr. Michael Greicius, a neurologist at Stanford University School of Medicine who was not involved in the research.

“One important argument against their interpretation is that the risk of Alzheimer’s disease in APOE4 homozygotes varies substantially across different genetic ancestries,” said Dr. Greicius, who cowrote a study that found that white people with two copies of APOE4 had 13 times the risk of white people with two copies of APOE3, while Black people with two copies of APOE4 had 6.5 times the risk of Black people with two copies of APOE3.

“This has critical implications when counseling patients about their ancestry-informed genetic risk for Alzheimer’s disease,” he said, “and it also speaks to some yet-to-be-discovered genetics and biology that presumably drive this massive difference in risk.”

Under the current genetic understanding of Alzheimer’s, less than 2 percent of cases are considered genetically caused. Some of those patients inherited a mutation in one of three genes and can develop symptoms as early as their 30s or 40s. Others are people with Down syndrome, who have three copies of a chromosome containing a protein that often leads to what is called Down syndrome-associated Alzheimer’s disease .

Dr. Sperling said the genetic alterations in those cases are believed to fuel buildup of amyloid, while APOE4 is believed to interfere with clearing amyloid buildup.

Under the researchers’ proposal, having one copy of APOE4 would continue to be considered a risk factor, not enough to cause Alzheimer’s, Dr. Fortea said. It is unusual for diseases to follow that genetic pattern, called “semidominance,” with two copies of a variant causing the disease, but one copy only increasing risk, experts said.

The new recommendation will prompt questions about whether people should get tested to determine if they have the APOE4 variant.

Dr. Greicius said that until there were treatments for people with two copies of APOE4 or trials of therapies to prevent them from developing dementia, “My recommendation is if you don’t have symptoms, you should definitely not figure out your APOE status.”

He added, “It will only cause grief at this point.”

Finding ways to help these patients cannot come soon enough, Dr. Sperling said, adding, “These individuals are desperate, they’ve seen it in both of their parents often and really need therapies.”

Pam Belluck is a health and science reporter, covering a range of subjects, including reproductive health, long Covid, brain science, neurological disorders, mental health and genetics. More about Pam Belluck

The Fight Against Alzheimer’s Disease

Alzheimer’s is the most common form of dementia, but much remains unknown about this daunting disease..

How is Alzheimer’s diagnosed? What causes Alzheimer’s? We answered some common questions .

A study suggests that genetics can be a cause of Alzheimer’s , not just a risk, raising the prospect of diagnosis years before symptoms appear.

Determining whether someone has Alzheimer’s usually requires an extended diagnostic process . But new criteria could lead to a diagnosis on the basis of a simple blood test .

The F.D.A. has given full approval to the Alzheimer’s drug Leqembi. Here is what to know about i t.

Alzheimer’s can make communicating difficult. We asked experts for tips on how to talk to someone with the disease .