Estimating differences in life satisfaction as a function of health-related behaviours: cross-sectional evidence from the Health Survey for England Authors: Linda Ng Fat, Robert West, Sarah E Jackson Published: 2026 ABSTRACT AbstractBackgroundStopping tobacco smoking, reducing alcohol consumption from high-risk levels, increasing physical activity and improving diet can improve mental well-being. To gain a preliminary estimate of the potential benefits of healthy lifestyle changes at a population level, this study estimated differences in life satisfaction among middle-aged adults in England as a function of whether they met recommended guidelines for smoking, alcohol consumption, physical activity and diet.MethodsUsing the Health Survey for England, a cross-sectional nationally representative household survey conducted in 2017 and 2018 (6978 adults aged 35–64 years), linear and logistic models were conducted predicting life satisfaction scores (0–10 scale) or odds of ‘high life satisfaction’ (score ≥7), from smoking status (current vs ex-), alcohol intake, physical activity and fruit and vegetable consumption categorised according to whether participants met recommended guidelines.ResultsAfter adjusting for other health behaviours and socio-demographic characteristics, the odds of high life satisfaction were 1.23 (95% CI 1.04 to 1.45) times higher for those eating ≥5 versus <5 portions/day, 1.47 (95% CI 1.20 to 1.79) times higher for people who have quit smoking versus people who smoke currently and 2.13 (95% CI 1.86 to 2.44) times higher for those with high versus low physical activity. These effects were additive: meeting all three of these guidelines was associated with the greatest increase in the likelihood of high life satisfaction (+32%; 89% vs 67% not meeting any guidelines). No significant difference was observed as a function of meeting alcohol consumption guidelines (OR=0.88 (95% CI 0.74 to 1.04)). Meeting all three of these behavioural guidelines was associated with an 18% increase in predicted life satisfaction (an average life satisfaction score of 8.0 vs 6.8).ConclusionsMeeting English government guidelines for physical activity, smoking cessation and fruit and vegetable intake is associated with a 32 percentage point higher likelihood of reporting high life satisfaction than meeting none of them. People drinking above the English government recommended level have similar life satisfaction to those drinking below it. WHAT IS ALREADY KNOWN ON THIS TOPIC Stopping smoking, reducing alcohol intake from high-risk levels, improving diet and greater physical activity have been found to lead to relatively rapid improvements in mental well-being. Stopping smoking, reducing alcohol intake from high-risk levels, improving diet and greater physical activity have been found to lead to relatively rapid improvements in mental well-being. WHAT THIS STUDY ADDS Adults aged 35–64 years who met government guidelines on physical activity, smoking cessation and diet had an 89% likelihood of reporting high life satisfaction compared with 67% who reported not meeting any of them. Adults aged 35–64 years who met government guidelines on physical activity, smoking cessation and diet had an 89% likelihood of reporting high life satisfaction compared with 67% who reported not meeting any of them. Meeting all three of these behavioural guidelines was associated with an 18% increase in predicted life satisfaction (an average life satisfaction score of 8.0 vs 6.8 out of 10). Meeting all three of these behavioural guidelines was associated with an 18% increase in predicted life satisfaction (an average life satisfaction score of 8.0 vs 6.8 out of 10). While at least some of the difference may be due to residual confounding from unmeasured characteristics, this provides a preliminary estimate of the potential improvements to mental well-being at population level of making changes to health behaviours. While at least some of the difference may be due to residual confounding from unmeasured characteristics, this provides a preliminary estimate of the potential improvements to mental well-being at population level of making changes to health behaviours. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Highlighting improvements in life satisfaction as benefits of health behaviour change could motivate the millions of people not currently meeting behavioural guidelines to attempt improvements in health behaviours. Highlighting improvements in life satisfaction as benefits of health behaviour change could motivate the millions of people not currently meeting behavioural guidelines to attempt improvements in health behaviours. Introduction Stopping tobacco smoking, reducing alcohol consumption from high-risk levels, increasing amount of physical activity and moving to a diet rich in fruit and vegetables are established as key factors in improving longer-term health and life expectancy.13 Public health messaging in England has traditionally focused on these benefits. Official guidelines recommend that adults do not smoke, limit alcohol consumption to no more than 14 units per week,4 engage in at least 150 min of moderate physical activity or at least 75 min of vigorous physical activity per week5 and eat at least five 80 g portions of fruit and vegetables a day.6 However, many struggle to adopt or maintain these behaviours,79 suggesting that knowledge of future health risks alone may not be a sufficient motivator. Importantly, even if individuals do not fully meet these guidelines, making incremental changes—such as increasing physical activity levels or improving diet quality—can still confer significant health benefits.10 11 Research also suggests that these health behaviours may also influence subjective well-being, including life satisfaction, over the short to medium term. Well-being is increasingly recognised as a valuable outcome in its own right, closely linked to mental health, productivity and social functioning.12 13 Unlike the delayed benefits of disease prevention, improvements in well-being can be experienced relatively quickly following positive lifestyle changes, offering more immediate and personally relevant rewards. Specifically, stopping cigarette smoking,1417 reducing alcohol consumption from high-risk levels,1822 moving from an unhealthy to a healthy diet2327 and from a low to at least moderate level of physical activity112831 have been found to result in relatively rapid overall increases in mental well-being that persist over time.32 These results are observed consistently across experimental and observational studies, suggesting relationships between lifestyle changes and improved well-being are likely to be causal.32 Highlighting the short-term gains in life satisfaction associated with healthier behaviours may therefore provide a powerful motivational tool. This could encourage greater uptake of behaviour change and increase engagement with support services by linking everyday lifestyle choices with enhanced quality of life. This may be particularly relevant for people in midlife (35–64 years), who often face competing demands across work and personal life, making well-being an important aspect of daily functioning. It also captures the period prior to the age when many non-communicable diseases typically emerge,2 and when lifestyle changes are more likely to have a meaningful, long-term impact on health outcomes.33 This study aimed to estimate the associations between meeting recommended guidelines for four key health behaviours—smoking status, alcohol consumption, physical activity and fruit and vegetable intake—and self-reported life satisfaction in a nationally representative sample of adults in England. The findings can be used to obtain a preliminary estimate of potential gains in well-being that could be achieved by making healthy lifestyle changes (based on the established causal relationships between making changes to these behaviours and achieving changes in well-being32). They can also help to inform public health strategies that incorporate well-being benefits alongside traditional health outcomes to better motivate such changes. Methods Pre-registration The study protocol and analysis plan were pre-registered on Open Science Framework (https://osf.io/gzsjn/). Design This study was a cross-sectional analysis of data from the Health Survey for England,34 an annual, nationally representative survey designed to monitor the health and health-related behaviours of people living in private households across England. The Health Survey for England employs a stratified, multistage probability sampling method to select participants, ensuring representativeness of the population. It is one of the few surveys that collects extensive information on multiple health behaviours in a representative sample of adults. Data are collected through a combination of face-to-face interviews, self-completion questionnaires and physical measurements that take place within participants’ homes. Household response rates were 59–60% in 2017–2018. For this study, we analysed data from adults aged 35–64 years who responded to the 2017 and 2018 surveys; the most recent years to capture information on life satisfaction and the four health behaviours. Data can be accessed from the UK Data Service (UKDS) subject to their End User License Agreement.35 36 Ethical approval is sought ahead of data collection by the relevant Ethics Committee at the time. No further ethical approval was needed for these secondary analyses. Measures Outcome variable Life satisfaction was assessed via a self-completion booklet using the single-item question from the Office for National Statistics well-being module: ‘Overall, how satisfied are you with your life nowadays?’ Participants rated their satisfaction on a scale from 0 (‘not at all satisfied’) to 10 (‘completely satisfied’). This measure has been found to accurately reflect overall well-being and UK government guidance recommends its use to evaluate well-being effects of policies and programmes.37 We analysed scores both as a continuous variable and as a dichotomous variable reflecting low/medium (0–6) versus high life satisfaction (7–10), according to thresholds used by the ONS.38 Input variables Cigarette smoking was assessed by asking participants whether they smoke cigarettes nowadays. We categorised responses as current, ex- and never cigarette smoking, but were primarily interested in the comparison between people who smoke cigarette currently and people who have quit cigarette smoking, as this provides the best indication of the potential improvement in life satisfaction associated with quitting smoking. Alcohol consumption categories were derived based on reported average weekly consumption in the past year. We distinguished between people with high-risk alcohol use (drinking more than the UK recommended weekly guidelines; ie, >14 units a week),4 people with low-risk alcohol use (up to 14 units/week) and people who do not drink alcohol (0 units/week). People who do not drink alcohol were treated as a separate category because studies have consistently observed poorer health outcomes in this group compared with those drinking at low-risk levels39; likely because many people stop drinking when they experience health problems. Combining them with people with low-risk alcohol use may create a negative health selection bias (decreasing average life satisfaction among this group) due to the pre-existing poor health bias among people who do not drink alcohol. Physical activity was assessed with the International Physical Activity Questionnaire.40 We dichotomised responses to distinguish between those meeting the recommended levels of aerobic activity (ie, ≥150 min of moderate intensity activity or ≥75 min of vigorous intensity activity5) versus those not meeting this recommendation. As a secondary physical activity outcome, we also created a variable identifying those reporting ≥30 min of moderate-to-vigorous physical activity (MVPA) per week (ie, were somewhat physically active) versus <30 min (inactive). Fruit and vegetable consumption was categorised based on self-reported number of portions per day. We dichotomised responses to distinguish between those consuming the recommended five or more portions of fruit and vegetables per day6 versus those not meeting this recommendation. Covariates To adjust for potential confounding by socio-demographic characteristics and survey year, we adjusted for the following covariates: age group (35–44, 45–54, 55–64), sex (male, female), ethnicity (white, black, Asian, mixed or other), marital status (single, married, separated/divorced/widowed), area-level deprivation (Index of Multiple Deprivation quintile) and survey year. We also ran models with additional adjustment for body mass index (BMI; treated as a continuous variable) and the presence of long-standing health conditions (yes/no), which may be confounders but could also plausibly mediate associations between lifestyle behaviours and well-being. Sample selection The analytical sample included 6798 adults aged 35–64 years. They were drawn from an eligible sample of 7312 participants who completed the measure of life satisfaction, with 514 participants excluded due to missing data on health behaviours or socio-demographic covariates. A further 820 participants were excluded from analyses that included BMI and long-standing illness on account of missing data on these variables. A sample selection flowchart is provided in online supplemental figure S1. Statistical analysis The data were analysed in Stata V.18.5. Non-response and complex survey design weighting were applied to all analyses to account for the stratified sampling design and improve generalisability of the estimates to the adult population of England. Primary analyses Descriptive analyses summarised sample characteristics and mean life satisfaction scores across categories of health behaviours. Age-sex adjusted linear regression models were used to estimate simple associations between each health behaviour and life satisfaction separately (Model 1). Multivariable linear regression models then estimated the association between life satisfaction and each health behaviour simultaneously in one model, adjusting for covariates (Model 2 just socio-demographic covariates; Model 3 socio-demographic and health-related covariates). Marginal effects were calculated to quantify predicted differences in life satisfaction associated with a unit change in specified predictor variables (eg, being a person who has quit smoking vs person who smokes or meeting the recommended physical activity guideline vs not), holding other variables constant. This was done based on Model 2, given that BMI and long-standing conditions may lie on the causal pathway between health behaviours and life satisfaction, thereby distorting the true effect.41 Predicted values were computed for each possible combination of behaviours, relative to a baseline scenario of four unhealthy behaviours (person who smokes currently, people with high-risk alcohol use, low physical activity, <5 fruit and vegetables/day). Secondary analyses We conducted three sets of pre-registered secondary analyses. First, we repeated the analyses substituting physical inactivity for meeting physical activity recommendations. Next, we repeated the multivariable models using binary logistic regression with high life satisfaction as the outcome variable. Finally, we repeated the multivariable models with the addition of two-way interactions between each health behaviour and age group and gender. After seeing the results, we also conducted some additional exploratory analyses to examine differences in life satisfaction by a more granular categorisation of alcohol consumption (categorised as 0, 1–13, 14–21, 22–35, 35–50 and ≥50 units/week). Patient and public involvement None. Results Sample characteristics Table 1 summarises the socio-demographic, health-related and health behaviour characteristics of the overall sample. Stratified summaries by each health behaviour are presented in online supplemental table S1. There was a similar split between age groups 35–44 (33%), 45–54 (36%), 55–64 (31%) and sex (51% female). Table 1Socio-demographic and health-related characteristics of those not meeting recommended guidelines for smoking, physical activity, alcohol and fruit and vegetable consumption and the total sample (N=6798)Person who smokes currently<150-minute MVPA>14 alcohol units/week<5 fruit and veg/dayTotalN*11662616170347396798Age group, % 35–443530283233 45–544038363836 55–642532363031 Female, %4752354951Ethnicity, % White8882958785 Black34134 Asian611389 Mixed21111 Other11011Marital status, % Single2314121312 Married5974777677 Separated/widowed/divorced1812101111Area-level deprivation % 1 (least deprived)1017232020 21520232121 31919222021 42622182120 5 (most deprived)2921151918Body mass index†, mean (SD)27.6 (5.41)29.2 (6.4)29.1 (6.1)28.5 (5.8)28.3 (5.7)Long-standing illness†, %4951414443Smoking status, % Person who smokes currently10019232118 Person who has quit smoking–27342727 Person who has never smoked–53435356Alcohol consumption, % Person with high-risk alcohol use: >14units/week33221002725 Person with low-risk alcohol use: ≤14 units/week5259–5960 Person who does not drink alcohol: 0 units/week1520–1415Meeting MVPA guideline %58–675862Physically active vs inactive %6940817477≥5 fruit and vegetables/day, %192426–31*Sample sizes are the unweighted sample; prevalence estimates are weighted based on sample weights.†Estimates of BMI and long-standing illness were derived from the subset who provided data on both variables (n=5978).BMI, body mass index; MVPA, moderate-to-vigorous physical activity. Socio-demographic and health-related characteristics of those not meeting recommended guidelines for smoking, physical activity, alcohol and fruit and vegetable consumption and the total sample (N=6798) Sample sizes are the unweighted sample; prevalence estimates are weighted based on sample weights. Estimates of BMI and long-standing illness were derived from the subset who provided data on both variables (n=5978). BMI, body mass index; MVPA, moderate-to-vigorous physical activity. Overall, 18% of participants were people who smoke currently, 25% reported drinking more than 14 units of alcohol per week, 38% did not meet the recommended guideline for aerobic activity, 23% were physically inactive and 69% consumed fewer than five portions of fruit and vegetables per day. There was some overlap across the four behaviours: 2% of participants did not meet any of the behavioural guidelines, 12% did not meet three of the guidelines, 35% did not meet two of the guidelines and 36% did not meet one guideline. Collectively, 85% were not meeting at least one guideline. The mean life satisfaction score in the sample was 7.6 (SD=2.0), with 80% of participants reporting high satisfaction (scores 7–10). With the exception of people with high-risk alcohol use—who generally had a more affluent and healthier profile—participants not meeting behavioural guidelines were more likely to live in the most deprived quintile, have a long-standing illness and less likely to meet physical activity recommendations compared with the total sample (p<0.05). Across all four behaviours, those not meeting the guidelines had a higher prevalence of smoking and were less likely to consume five or more portions of fruit and vegetables daily (table 1, see online supplemental table S1 for a more detailed breakdown). Participants excluded due to missing covariate data (n=514) were more likely to live in the most deprived areas, be of Asian ethnicity, be people who smoke currently and be people who do not drink alcohol compared with those included in the analysis (online supplemental table S2). Associations between health behaviours and mean life satisfaction scores Differences in mean life satisfaction were observed across health behaviours (table 2). In age-sex adjusted linear regression models (Model 1), participants who met health behaviour guidelines generally reported higher life satisfaction scores, with the exception of alcohol consumption, for which no significant difference was found. Table 2Associations between health behaviours and life satisfaction (N=6798)Life satisfaction*Mean (SD) score*Model 1:Model 2:B (95% CI)Badj (95% CI)Smoking status Person who smokes currently6.8 (2.0)11 Person who has quit smoking7.5 (2.0)0.69 (0.49 to 0.89)0.35 (0.17 to 0.54) Person who has never smoked7.8 (1.8)1.03 (0.84 to 1.21)0.65 (0.47 to 0.82)Alcohol consumption Person with high-risk alcohol use: ≥14units/week7.7 (1.9)11 Person with low-risk alcohol use: ≤14 units/week7.6 (1.9)−0.03 (−0.15 to 0.08)−0.10 (−0.21 to 0.01) Person who does not drink alcohol: 0 units/week7.0 (2.5)−0.66 (−0.88 to −0.44)−0.63 (−0.86 to −0.41)Physical activity† Low: <150-minute MVPA/week7.0 (2.4)11 High: ≥150-minute MVPA/week7.9 (1.7)0.81 (0.70 to 0.92)0.64 (0.54 to 0.75) Inactive: <30-minute MVPA/week6.7 (2.6)11 Active: ≥30-minute MVPA/week7.8 (1.7)1.16 (1.01 to 1.32)0.89 (0.75 to 1.04)Fruit and vegetable consumption <5/day7.4 (2.1)11 ≥5 /day7.8 (1.7)0.42 (0.32 to 0.53)0.24 (0.14 to 0.34)Model 1: models testing the association between each health behaviour and life satisfaction, adjusting for age and sex.Model 2: multivariable model testing the association between each health behaviour and life satisfaction, adjusting for other health behaviours, age, sex, ethnicity, marital status, area-level deprivation and survey year.*Scores range from 0 to 10, with higher scores indicating greater life satisfaction.†The multivariable model (Model 2) substituted physical activity (low/high) with physical inactivity (inactive/active). Model 2 results for smoking status, alcohol consumption, and fruit and vegetable consumption are from the version that included physical activity (low/high).MVPA, moderate-to-vigorous physical activity. Associations between health behaviours and life satisfaction (N=6798) Model 1: models testing the association between each health behaviour and life satisfaction, adjusting for age and sex. Model 2: multivariable model testing the association between each health behaviour and life satisfaction, adjusting for other health behaviours, age, sex, ethnicity, marital status, area-level deprivation and survey year. Scores range from 0 to 10, with higher scores indicating greater life satisfaction. The multivariable model (Model 2) substituted physical activity (low/high) with physical inactivity (inactive/active). Model 2 results for smoking status, alcohol consumption, and fruit and vegetable consumption are from the version that included physical activity (low/high). MVPA, moderate-to-vigorous physical activity. After additional adjustment for other health behaviours and socio-demographic covariates (Model 2), similar patterns were found, though effect sizes were attenuated. Compared with people who smoke currently, mean life satisfaction scores were 0.35 points higher (95% CI 0.17 to 0.54) among people who have quit smoking and 0.65 points higher (95% CI 0.47 to 0.82) among people who have never smoked. Participants reporting high physical activity (compared with those with low physical activity) had a mean score 0.64 points higher (95% CI 0.54 to 0.75), while those classified as physically active (vs inactive) had scores that were 0.89 points higher (95% CI 0.75 to 1.04). Those consuming five or more portions of fruit and vegetables daily had scores 0.24 points higher (95% CI 0.14 to 0.36) than those consuming fewer than five portions. In contrast, people who do not drink alcohol reported lower life satisfaction than people with high-risk alcohol use, with a difference of −0.63 points (95% CI −0.86 to −0.41), while no significant difference was observed for people with low-risk alcohol use. Exploratory analyses showed mean life satisfaction scores followed an inverse U-shaped association with level of alcohol consumption, with those drinking slightly above the recommended limit (15–21 units/week; mean=7.88 (95% CI 7.76 to 8.00)) reporting the highest life satisfaction and people who do not drink alcohol (mean=7.01 (95% CI 6.82 to 7.20)) and those drinking at very high levels (≥50 units/week; mean=7.07 (95% CI 6.77 to 7.36)) reporting the lowest levels (online supplemental table S3). Associations between meeting health behaviour guidelines and mean life satisfaction scores remained statistically significant, although reduced in magnitude, after further adjustment for BMI and long-standing illness (Model 3; online supplemental table S4). Associations between health behaviours and the odds of high life satisfaction A similar pattern was observed in logistic regression models assessing the odds of reporting high life satisfaction (table 3). Compared with those not meeting health behaviour guidelines—and after adjustment for other health behaviours and socio-demographic covariates (Model 2)—people who have quit smoking had 1.47 times higher odds of high life satisfaction (OR=1.47; 95% CI 1.20 to 1.79), while those consuming five or more portions of fruit and vegetables daily had 1.23 times higher odds (95% CI 1.04 to 1.45). Participants with high physical activity had 2.13 times higher odds (95% CI 1.86 to 2.44), and those classified as physically active (vs inactive) had 2.55 times higher odds (95% CI 2.18 to 2.99). Table 3Associations between health behaviours and high life satisfaction (N=6798)High life satisfaction*Row %Model 1:Model 2:OR (95% CI)ORadj (95% CI)Smoking status Person who smokes currently6611 Person who has quit smoking792.03 (1.69 to 2.44)1.47 (1.20 to 1.79) Person who has never smoked842.93 (2.46 to 3.49)2.15 (1.78 to 2.59)Alcohol consumption Person with high-risk alcohol use: >14units/week8211 Person with low-risk alcohol use: ≤14 units/week820.95 (0.82 to 1.12)0.88 (0.74 to 1.04) Person who does not drink alcohol: 0 units/week680.45 (0.37 to 0.55)0.46 (0.36 to 0.58)Physical activity† Low: <150-minute MVPA/week7111 High: ≥150-minute MVPA/week852.43 (2.14 to 2.76)2.13 (1.86 to 2.44) Inactive: <30-minute MVPA/week6311 Active: ≥30-minute MVPA/week853.22 (2.79 to 3.71)2.55 (2.18 to 2.99)Fruit and vegetable consumption <5/day7111 ≥5 /day851.50 (1.28 to 1.76)1.23 (1.04 to 1.45)Model 1: models testing the association between each health behaviour and life satisfaction, adjusting for age and sex.Model 2: multivariable model testing the association between each health behaviour and life satisfaction, adjusting for other health behaviours, age group, sex, ethnicity, marital status, area-level deprivation and survey year.*Defined as a score ≥7/10 on the life satisfaction measure.†The multivariable model (Model 2) substituted physical activity (low/high) with physical inactivity (inactive/active). Model 2 results for smoking status, alcohol consumption, and fruit and vegetable consumption are from the version that included physical activity (low/high).MPVA, moderate-to-vigorous physical activity. Associations between health behaviours and high life satisfaction (N=6798) Model 1: models testing the association between each health behaviour and life satisfaction, adjusting for age and sex. Model 2: multivariable model testing the association between each health behaviour and life satisfaction, adjusting for other health behaviours, age group, sex, ethnicity, marital status, area-level deprivation and survey year. Defined as a score ≥7/10 on the life satisfaction measure. The multivariable model (Model 2) substituted physical activity (low/high) with physical inactivity (inactive/active). Model 2 results for smoking status, alcohol consumption, and fruit and vegetable consumption are from the version that included physical activity (low/high). MPVA, moderate-to-vigorous physical activity. People who do not drink alcohol, on the other hand, had significantly lower odds of high life satisfaction (OR=0.46; 95% CI 0.36 to 0.58), while no significant association was observed for people with low-risk alcohol use compared with people with high-risk alcohol use. As we observed for mean life satisfaction scores, exploratory analyses showed an inverse U-shaped pattern by level of alcohol consumption. The proportion reporting high life satisfaction was highest among those drinking slightly above the recommended limit (15–21 units/week; 86% (95% CI 83% to 89%)) and lowest among people who do not drink alcohol (68% (95% CI 64% to 71%)) and those drinking at very high levels (≥50 units/week; 71% (65% to 77%); online supplemental table S5). These associations remained statistically significant after adjusting for BMI and long-standing illness, though effect sizes were somewhat reduced (Model 3; online supplemental table S6). Combined effects of health behaviours on life satisfaction Using the fully adjusted multivariable model incorporating all four health behaviours, predicted life satisfaction was estimated for all possible combinations of healthy and unhealthy behaviours, controlling for socio-demographic covariates. Figure 1A displays predicted life satisfaction scores for each combination relative to a baseline scenario in which no behavioural guidelines were met. Compared with this baseline of ‘unhealthy’ behaviours, scores were generally higher for those with a greater number of healthy behaviours. For example, the relative increase in predicted life satisfaction was 4% for those who only met the guideline for fruit and vegetable consumption, 5% for those who only met the guideline for smoking and 10% for those who only met the guideline for physical activity. By contrast, the relative increase for those meeting two guidelines was 9% for the combination of fruit and vegetable consumption and smoking, 13% for fruit and vegetable consumption and physical activity, and 15% for physical activity and smoking. Meeting all three of these behavioural guidelines was associated with an 18% increase in predicted life satisfaction (a score of 8.0 vs 6.8). Similarly, meeting all four behavioural guidelines was associated with a 17% increase in predicted life satisfaction (7.9 vs 6.8). Predicted life satisfaction by different combinations of healthy and unhealthy behaviours. S, A, P and F indicate, respectively, meeting guidelines for smoking (person who has quit smoking), alcohol (≤14 units/week), physical activity (≥150-minute moderate-to-vigorous physical activity/week) and fruit and vegetable consumption (≥5/day). 0 indicates not meeting the respective guideline. Similarly, figure 1B presents predicted probabilities of having high life satisfaction for each combination of health behaviours, showing a consistent pattern. Relative to those not meeting any behavioural guidelines, the increase in the predicted probability of having high life satisfaction was 6% for those who only met the guideline for fruit and vegetable consumption, 12% for those who only met the guideline for smoking and 21% for those who only met the guideline for physical activity. The relative increase for those meeting two guidelines was 17% for the combination of fruit and vegetable consumption and smoking, 26% for fruit and vegetable consumption and physical activity and 29% for physical activity and smoking. Meeting all three of these behavioural guidelines was associated with a 32% increase in the predicted probability of high life satisfaction (89% vs 67%), and meeting all four behavioural guidelines was associated with a 30% increase (87% vs 67%). Interactions with age and sex Significant two-way interactions were observed between sex and alcohol consumption, and between age group and smoking status, physical activity and inactivity (online supplemental tables S7 and S8). No other significant age or sex interactions were found for the remaining health behaviours. Among age-stratified results, the positive association between ex-smoking and life satisfaction was greatest for the youngest age group (35–44), with a 10% increase in predicted life satisfaction, compared with just 3% in the older age groups (45–64; based on predicted life satisfaction scores in online supplemental table S7). The association between high physical activity and life satisfaction was most pronounced in older adults aged 55–64, with a 13% increase in predicted life satisfaction, compared with 10% in those aged 45–54, and 5% in those aged 35–44. Similarly, the benefit of being physically active rather than inactive was greater among older adults (45–64 years: 16% increase) than among younger adults (35–44 years: 9%). In terms of alcohol consumption, women who were people with low-risk alcohol use had 4% lower life satisfaction scores than those classified as people with high-risk alcohol use, whereas a very small/no difference in life satisfaction scores was observed among men who were high-risk versus low-risk alcohol users (0%). The association between alcohol consumption and the odds of high life satisfaction did not vary significantly by sex in logistic regression models (online supplemental table S8). Our exploratory analyses by a more detailed breakdown of alcohol consumption suggested a similar pattern among men and women (online supplemental tables S3 and S5). Discussion This study examined associations between meeting health behaviour guidelines—for smoking, alcohol consumption, physical activity and fruit and vegetable intake—and life satisfaction in a large, nationally representative sample of adults in England aged 35–64 years. Most participants (85%) were not meeting the guideline for at least one of the four behaviours, equating to approximately 19 million people (out of 22.3 million adults in England aged 35–64 years42). Across both continuous and dichotomous outcomes, we found that not smoking, engaging in regular physical activity and eating at least five portions of fruit and vegetables per day were each independently associated with higher life satisfaction, even after adjustment for socio-demographic and health-related factors. Associations were additive, with progressively higher life satisfaction among those meeting a greater number of behavioural guidelines. In contrast, alcohol consumption showed a more complex relationship with well-being, with no clear benefit associated with drinking within the recommended guideline of 14 weekly units. Our findings align with and extend prior research showing that health-promoting behaviours are associated not only with long-term physical health benefits, but also with improved subjective well-being. As highlighted in previous reviews, eg, 14 29 32 behaviours such as quitting smoking and increasing physical activity may causally contribute to improvements in well-being through both physiological and psychosocial pathways. For example, smoking cessation can lead to improvements in mood and stress regulation,14 while physical activity has well-established effects on mood, self-esteem and quality of life.43 44 Similarly, higher fruit and vegetable consumption has been linked to better mental health,25 potentially through mechanisms involving micronutrient intake and inflammation reduction.45 These associations were strongest for physical activity, particularly among individuals classified as active versus inactive. This supports prior evidence suggesting that even modest increases in activity levels can substantially enhance well-being,10 with benefits potentially accruing through both physiological effects (eg, neurochemical changes) and psychological factors (eg, increased self-efficacy or social contact).10 Importantly, our secondary analysis confirmed that even being ‘somewhat active’ (≥30 min/week) was associated with higher well-being, suggesting that incremental changes—well below guideline thresholds—may still yield meaningful subjective rewards. The inverse association between non-drinking and life satisfaction contrasts with the generally positive associations observed for other health behaviours. This likely reflects residual confounding, as people who do not drink alcohol often include individuals with poorer health or those who have stopped drinking due to illness.39 Our exploratory analysis revealed an inverse U-shaped relationship between alcohol consumption and life satisfaction, with those drinking slightly above the recommended limit reporting the greatest life satisfaction. On average, this group had a more affluent and healthier profile than those not meeting other behavioural guidelines, so this is likely to be driven by some extent to reverse causality (ie, those with better health and life satisfaction choosing to drink above recommended guidelines, as opposed to this level of drinking leading to better well-being). As has been observed in previous research,46 those drinking well above the recommended limit had poorer life satisfaction; more similar to those who did not drink at all. Although our results do not provide clear evidence that drinking below recommended guidelines improves life satisfaction, they suggest that reducing alcohol intake does not diminish it, addressing a common perception that cutting down may reduce enjoyment.47 In addition, the well-established long-term health risks of excessive alcohol consumption48 underscore the importance of interpreting these findings cautiously in the context of broader public health goals. Subgroup analyses indicated that the benefits of healthier behaviours were not uniform across age and sex groups. The well-being benefits of physical activity were more pronounced among older adults, suggesting that maintaining an active lifestyle may become increasingly important for sustaining life satisfaction with age. In contrast, the well-being gains from quitting smoking were greater among younger adults, highlighting the potential for early cessation to yield greater improvements in life satisfaction. These variations point to the value of tailoring behaviour change messaging to different demographic groups based on what may most strongly influence well-being. Taken together, our findings have two important implications. First, they reinforce the potential benefits of promoting healthy behaviours not only for their long-term disease prevention benefits but also for their short-term impact on quality of life. Most participants (85%) were not meeting at least one behavioural guideline, suggesting scope to improve not only their physical health but also their well-being by making healthy changes. The immediate subjective rewards associated with changes in smoking, diet and physical activity may serve as more tangible and motivating targets for individuals, particularly those who are otherwise disengaged by abstract or distant health risks. Integrating well-being messaging into health promotion strategies—especially those aimed at midlife adults juggling work and caregiving responsibilities—may enhance relevance and uptake. Second, the additive nature of these effects suggests that promoting even small improvements across multiple behaviours could yield meaningful gains in population well-being. For instance, combining improvements in physical activity, smoking status and diet was associated with an 18–32% relative increase in predicted life satisfaction in this sample. This supports the growing movement in public health towards holistic, multi-behavioural interventions that reflect how behaviours interact in real-world settings and could be targeted towards those with poorer well-being or low adherence to health behaviour recommendations. Key strengths of this study include the use of a large, nationally representative sample and robust adjustment for socio-demographic and health-related confounders. The inclusion of multiple behaviours in a single model allowed us to isolate independent associations while exploring their combined effects. There were also a number of limitations. First, the cross-sectional survey design prevents us from drawing causal conclusions about the directionality of observed associations. It is possible, for example, that higher life satisfaction increases motivation to engage in healthier behaviours. However, prior longitudinal and experimental studies suggest there is likely a causal relationship from health behaviour change to improved well-being for each of the four health behaviours explored in this study.32 The relationship between life satisfaction and healthy behaviours such as physical activity may be reciprocal and reinforcing.49 Second, all key variables, including health behaviours and life satisfaction, were self-reported, which may introduce reporting bias due to social desirability or recall inaccuracies. This may affect our estimations of relative effect sizes; however, self-report is the most cost-effective way to collect information on complex health behaviours such as physical activity in a large-scale survey; the use of this survey enabled us to generalise findings to the population. Third, smoking status was limited to cigarette use and did not account for other forms of tobacco smoking. Similarly, alcohol consumption was categorised based on weekly units and did not capture patterns such as heavy episodic drinking or drinking context, which may have different associations with well-being.46 We did not differentiate between people who drank alcohol formerly and lifetime abstainers in the non-drinking group; studies suggest that both groups may be prone to a pre-existing poor health bias.50 Fourth, while adjustments were made for a range of socio-demographic and health-related covariates, residual confounding—particularly in the alcohol analyses—cannot be ruled out. Fifth, although the Health Survey for England uses rigorous sampling and weighting to ensure representativeness, there were some differences between analysed versus excluded participants, and more broadly, exclusion of institutionalised populations and potential non-response bias could limit generalisability. Finally, life satisfaction was measured using a single-item question, which, while widely accepted and validated, may not capture the full complexity of subjective well-being. In conclusion, this study found that life satisfaction was higher in middle-aged adults in England who met government guidelines for cigarette smoking, physical activity and diet compared with those who did not. Life satisfaction was similar in those who consumed alcohol below versus above the guidelines. Supplementary material