Association of Healthy Lifestyle Factors and Obesity-Related Diseases in Adults in the UK

Key Points Question Is a healthy lifestyle more important for lowering the risk of obesity-related disease than obesity itself? Findings In this cohort study of 438 583 UK Biobank participants, adults with obesity were at a higher risk of various diseases compared with those who maintained a healthy body weight and lifestyle. For individuals with obesity, engaging in regular exercise, refraining from smoking, consuming alcohol in moderation or not at all, and eating a healthy diet were associated with a decreased risk of cardiovascular disease, kidney failure, gout, sleep disorders, and mood disorders. Meaning These findings suggest that although adherence to a healthy lifestyle is associated with a reduced risk of several adverse health outcomes in adults with obesity, it does not entirely eradicate the risk of obesity-related diseases.


Introduction
Obesity is the most prevalent chronic disease worldwide.It is estimated that more than 1 billion people will be living with obesity by 2030. 1 The higher rates of obesity-associated mortality and comorbidities, such as diabetes, cardiovascular disease, chronic kidney disease, and several types of cancer, are equally staggering, with an average of 5 million deaths and 160 million disabilityadjusted life-years. 1 As a result, trends toward reduced productivity and greater intensity of health care service use are expected to result in a major economic burden. 2,3A need exists for the treatment of obesity as a chronic, progressive, and relapsing disease to improve outcomes for people living with this disease.A suboptimal lifestyle is a major preventable cause of obesity and its associated comorbidities. 4 As such, interventions to improve lifestyle are an opportunity to optimize the management of obesity.
][7][8][9][10] Despite nutritional guidelines, less is known about such correlations across the body mass index (BMI) spectrum, particularly in adults with obesity.[13][14] Behavioral factors are often mutually linked, and people tend to follow interrelated lifestyle patterns. 15Lifestyle factors should therefore be analyzed jointly to better evaluate their health impact.7][18][19] It remains unclear to what extent specific combinations of healthy lifestyle factors are associated with a reduced risk of obesity-related diseases other than cardiovascular disease.Furthermore, whether a healthy lifestyle offsets the risk of obesity remains to be elucidated.The aim of this study is to estimate the association between healthy lifestyle factors, such as physical activity, never smoking, no or moderate alcohol consumption, and a healthy diet, and the incidence of major obesity-related diseases in adults with obesity compared with those with a normal weight.

Study Population
Data for this cohort study were obtained from the UK Biobank.The UK Biobank is a large prospective cohort study that recruited more than 500 000 volunteers aged 40 and 69 years in 2006-2010.Details of the UK Biobank cohort are described elsewhere. 20For this study, demographic data were collected on age, sex, socioeconomic status (Townsend deprivation index), and lifestyle factors.In the UK Biobank cohort study, participants self-identified their ethnicity via a touchscreen questionnaire, but we did not collect or analyze data on race and ethnicity for this study.The North West Multi-centre Research Ethics Committee approved the collection and use of UK Biobank data in accordance with the principles of the Declaration of Helsinki.The UK Biobank study was approved by the North West Multi-centre Research Ethics Committee, and written informed consent was already provided by each participant.This research was conducted using the UK Biobank Resource under application number 23476, and the study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Assessment of Lifestyle Factors
Four lifestyle factors were investigated, including smoking, alcohol consumption, physical activity, and diet.Data about these lifestyle factors were obtained from touchscreen baseline questionnaires at the UK Biobank assessment center.The codes used to define the variables included in this study are reported in eTable 1 in Supplement 1.For tobacco smoking, participants were grouped into never, former, and current smokers based on their responses to the questions, "Do you smoke tobacco now?" and "In the past, how often have you smoked tobacco?"For alcohol consumption, the number of drinks per week was computed by summing responses to the average weekly consumption of champagne, white wine, beer, cider, spirits, fortified wine, and other types of alcohol consumption.
For physical activity, the number of days per week and duration of vigorous physical activity, moderate physical activity, and walking were estimated via validated short International Physical Activity Questionnaire. 21For dietary assessment, participants were asked to complete a validated questionnaire that included questions on food consumption frequency over the past year and the avoidance of some foods. 22r each lifestyle factor, participants were assigned a score of 1 if they met the criterion for a healthy lifestyle and 0 otherwise.For smoking status, a score of 1 was assigned to never smoker and 0 to current or former smoker.For alcohol consumption, a score of 1 was assigned to nonheavy drinking (men, <10 drinks per week; women, <5 drinks per week) and 0 otherwise (1 drink contains 8 g of ethanol in the UK).For physical activity, a score of 1 was assigned to those who met the World Health Organization recommendations (Ն2.5 hours of moderate-intensity activity per week, Ն1.25 hours of vigorous-intensity activity per week, or an equivalent combination of moderate-and vigorous-intensity activity). 23For diet, a score of 1 was assigned to those who met at least 5 of the dietary recommendations for cardiovascular health 24 (increased consumption of fruits, vegetables, whole grains, fish or shellfish, dairy products, and vegetable oils and reduced or no consumption of refined grains, processed and unprocessed meats, and sugar-sweetened beverages) and 0 otherwise.The definitions and variables used for dietary components were based on 2 previous UK Biobank studies. 25,26The healthy lifestyle score was computed by summing the 4 scores.It ranged between 0 (lowest healthy score) and 4 (highest healthy score).Furthermore, lifestyle profiles were created based on the number of healthy lifestyle factors, as described in Nyberg et al 9

Ascertainment of Outcomes
The outcomes of interest were classified using the International Statistical Classification of Diseases, Tenth Revision (ICD-10).The outcomes included the incidence of the major diseases correlated with obesity 27,28 : infections (bacterial and viral), diabetes, cancer (esophageal adenocarcinoma, gastric cardia, pancreatic, liver, kidney, colorectal, ovarian [in women], endometrial [in women], and breast cancer [in women]), cardiovascular disease (hypertension, ischemic heart disease, pulmonary embolism, arrhythmias, heart failure, cerebrovascular diseases, arteriosclerosis, and deep vein thrombosis), asthma, liver disease, kidney failure, gout, osteoarthritis, sleep disorders, and mood disorders.The diagnosis of the diseases during follow-up was obtained from primary and secondary hospital inpatient records and reported causes of death. 29The ICD-10 codes used to define diseases are reported in eTable 3 in Supplement 1.

Statistical Analysis
The data analysis was performed between December

Results
Of the 502 616 participants in the UK Biobank, 5820 with a BMI less than 18; 40 146 extreme outliers; and 16 340 with missing information on age, sex, socioeconomic status, and lifestyle factors were excluded from the analysis.Furthermore, 1727 participants with a record of any of the studied diseases at baseline were excluded.The final analytic sample included 438 583 participants (eFigure 1 in Supplement 1).
The baseline characteristics of participants in the UK Biobank are presented in  for the profile that included a healthy diet and never smoking and 0.74 (95% CI, 0.69-0.79)for the profile that included a healthy diet, physical activity, and never smoking.The HRs for all-cause mortality according to the lifestyle profiles are presented in eTable 5 in Supplement 1.In adults with obesity, all the studied lifestyle profiles were associated with a reduced risk of all-cause mortality, except for the profiles that included never smoking and no or moderate alcohol consumption alone or combined (eTables 5 and 6 in Supplement 1).

Discussion
The findings from this population-based cohort study suggest that adherence to a healthy lifestyle as a composite score is associated with a reduced risk of several health outcomes in adults with obesity but does not entirely offset the negative effects of obesity.Studies that estimate the risk of a wide range of outcomes according to adherence to lifestyle factors among adults with obesity seem to be lacking.Most of the research has focused on cardiovascular diseases and mortality and individual lifestyle factors.
The novelty of our study is the examination of the association between combinations of lifestyle factors (adequate physical activity, never smoking, no or moderate alcohol consumption) and the incidence of several comorbid diseases.Our results confirmed the lower risk of mortality and cardiovascular disease in adults with obesity meeting the 4 healthy lifestyle factors 6,17,19 and extended the findings to kidney failure, gout, sleep disorders, and mood disorders.The association between a healthy lifestyle and diseases was independent of other potential confounders, such as age, sex, and socioeconomic status.In addition, we observed a reduced risk of diabetes, liver disease, and osteoarthritis, though the interactions were not statistically significant after the Bonferroni correction.The lifestyle profiles that were associated with the lowest risk of developing diseases included a healthy diet and at least 1 of the 2 healthy behaviors of adequate physical activity and never smoking.Tobacco affects health mainly via increasing DNA damage and oxidative stress, and smokers with obesity have been shown to be at higher risk of morbidity and premature mortality compared with nonsmokers without obesity. 11[32] Although adhering to a healthy lifestyle was associated with a reduced risk of health outcomes, adults with obesity still had an increased risk of diseases compared with those with a normal BMI and meeting the 4 lifestyle factors.Obesity is a strong risk factor for comorbidities, whereas not smoking, reducing alcohol intake, exercising regularly, and eating a healthy diet may not be sufficient to attenuate the risk.Many studies have highlighted the potential benefits of weight loss in preventing comorbidities among adults with obesity. 33Thus, supporting people to reduce their body weight in addition to promoting healthy behaviors may bring additional benefits to reduce the risk of developing comorbid diseases and extend disease-free life expectancy in obesity.
This study is the first in our knowledge to assess the association of combined healthy lifestyle factors with a wide range of obesity-related outcomes in adults with obesity.Our findings have important clinical and public health implications.Individual counseling by clinicians and national-level

Limitations
This study has several limitations.First, the study population may not be completely representative of the UK population because participants are more likely to have a healthy lifestyle, 34 which might lead to an underestimation of health hazards in individuals with the lowest healthy lifestyle score. 34cond, we used measures of lifestyle behavior at baseline and did not consider changes over the follow-up period.Third, although lifestyle factors may have different effects on obesity-related diseases, we did not weigh them by their association with the outcomes in our analysis.We wanted to evaluate overall lifestyle as a cluster, and the weighted score cannot fully account for the complex interactions between lifestyle factors.Fourth, although our study included data on infectious diseases, we did not differentiate among them, including SARS-CoV-2 infection (COVID-19), and we did not specifically evaluate the impact of healthy lifestyle on COVID-19 morbidity and mortality.
Obesity is associated with an increased risk of COVID-19 complications and mortality, 35 and investigating the specific association of a healthy lifestyle with COVID-19 outcomes in patients with a high BMI is of particular interest.Fifth, despite the exclusion of participants with diseases diagnosed within the first 2 years of follow-up, reverse causality and residual confounding remain limitations because of the nature of observational studies.

Conclusions
In : 1 profile with any healthy lifestyle factor, 4 profiles with 1 healthy lifestyle factor, 6 profiles with 2 healthy lifestyle factors, 4 profiles with 3 healthy lifestyle factors, and 1 profile with 4 healthy lifestyle factors.The definition and scoring of lifestyle factors are reported in eTable 2 in Supplement 1.

Downloaded From: https://jamanetwork.com/ on 05/30/2023 different
levels of BMI (as measured by weight in kilograms divided by height in meters squared).

Table 2
provides the adjusted HRs (95% CIs) for obesity-related diseases according to16lifestyle profiles in the overall study population and in each BMI category.In the overall population, the lifestyle profiles that were associated with the lowest HRs included a healthy diet and at least 2 healthy behaviors of physical activity and never smoking, with adjusted HRs less than 0.80 (0.79 [95% CI, 0.76-0.82],0.70 [95% CI, 0.64-0.76],and 0.76 [95% CI, 0.71-0.81] in the overall, normal weight, and overweight cohorts, respectively).In adults with obesity, the HR was 0.87 (95% CI, 0.82-0.93)for the profile that included a healthy diet and never smoking, 0.84 (95% CI, 0.78-0.89)for the profile that included a healthy diet and physical activity, and 0.80 (95% CI, 0.75-0.85)for the profile that included a healthy diet, never smoking, and physical activity.The unadjusted HRs for incident diseases by lifestyle profile in each BMI category were comparatively lower (eTable 4 in Supplement 1).In individuals with obesity, the HRs before adjustment were 0.80 (95% CI, 0.75-0.86)

Table 2 .
Cox Proportional Hazards Models for Incident Diseases by Lifestyle Profile in Each BMI CategoryThe association between obesity and the incidence of diseases at each level of the healthy lifestyle score is shown in Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HR, hazard ratio; NA, not applicable.a Hazard ratios adjusted for age, sex, BMI, and Townsend deprivation index.b The HRs are for all obesity-associated diseases.JAMA Network Open | Nutrition, Obesity, and Exercise Healthy Lifestyle and Obesity Risks in Adults in the UK JAMA Network Open.2023;6(5):e2314741.doi:10.1001/jamanetworkopen.2023.14741(Reprinted) May 26, 2023 6/13 Downloaded From: https://jamanetwork.com/ on 05/30/2023

Table 3 .
For each lifestyle score, adults with obesity were at higher risk of several primary outcomes compared with those with normal weight and the healthiest lifestyle score (ie, meeting all 4 healthy lifestyle factors) (HRs adjusted for age, sex, and Townsend deprivation index ranged from 1.41 [95% CI, 1.27-1.56]for arrhythmias to 7.16 [95% CI, 6.36-8.05]for diabetes in adults with obesity and 4 healthy lifestyle factors).The HRs for diseases in adults with overweight at each level of the healthy lifestyle score compared with those with normal weight and the healthiest

Table 3 .
Associations of Obesity as Measured by BMI With Incident Obesity-Related Diseases by Healthy Lifestyle Score Downloaded From: https://jamanetwork.com/

on 05/30/2023 public
policies to decrease smoking, moderate alcohol consumption, increase exercise frequency, adhere to a healthy diet, and reduce body weight are essential to reduce the burden of obesity.

Table 3 .
Associations of Obesity as Measured by BMI With Incident Obesity-Related Diseases by Healthy Lifestyle Score (continued) Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); HR, hazard ratio; NA, not applicable.aHealthylifestyle score included 4 lifestyle factors (never smoking, no or moderate alcohol consumption, physical activity, and healthy diet).Participants scored 1 if they met the criterion for a healthy lifestyle and 0 otherwise.bTheHRs are adjusted for age, sex, and Townsend deprivation index.

JAMA Network Open | Nutrition, Obesity, and Exercise Healthy
this cohort study of UK Biobank participants, adherence to a healthy lifestyle as a composite score, including not smoking, exercising regularly, consuming no or moderate amounts of alcohol, and eating a healthy diet, was associated with a reduced risk of several health outcomes among adults with obesity.Although a healthy lifestyle appeared to be beneficial, it did not entirely offset the health risks associated with obesity.Codes Used to Define the Variables Included in the Study eTable 2. Definition and Scoring of Lifestyle Factors eTable 3. International Classification of Diseases, Tenth Revision (ICD-10) Codes for Diseases Included in the Study eTable 4. Hazard Ratios for Incident Diseases by Lifestyle Profile in Each Body Mass Index Category eTable 5. Hazard Ratios for All-Cause Mortality by Lifestyle Profile in Each Body Mass Index Category eTable 6. Adjusted Hazard Ratios for All-Cause Mortality by Lifestyle Profile in Each Body Mass Index Category eTable 7. Associations of Overweight With Incident Diseases by Healthy Lifestyle Score eTable 8. Associations of Body Mass Index With All-Cause Mortality by Healthy Lifestyle Score eFigure 1. Flowchart of Sample Selection eFigure 2. Hazard Ratios for Incident Diseases by Body Mass Index Category eFigure 3. Adjusted Hazard Ratios for Incident Diseases by Body Mass Index CategoryeFigure 4. A. Hazard Ratios for Incident Diseases by Lifestyle Profile in Individuals With Normal Weight (BMI 1.5-24.9),B. Overweight (BMI 25-29.9),and C. Obesity (BMI Ն30) eFigure 5. Adjusted Hazard Ratios for Incident Diseases by Lifestyle Profile in Individuals With Overweight (BMI 25-29.9)Lifestyle and Obesity Risks in Adults in the UK Hazard Ratios for All-Cause Mortality by Lifestyle Profile in Individuals With Normal Weight (BMI 18.5-24.9kg/m 2 ), Overweight (BMI 25-29.9),and Obesity (BMI Ն30) Before (A) and After Adjustment (B) for Age, Sex, and Townsend Deprivation Index eTable 9. Associations of Body Mass Index With Incident Diseases by Healthy Lifestyle Score After Exclusion of Participants With Diseases Diagnosed Within the First 2 Years of Follow-up eTable 10.Hazard Ratios for Incident Diseases by Lifestyle Profile in Each Body Mass Index Category After Exclusion of Participants With Diseases Diagnosed With the First 2 Years of Follow-up eTable 11.Adjusted Hazard Ratios for Incident Diseases by Lifestyle Profile in Each Body Mass Index Category After Exclusion of Participants With Diseases Diagnosed With the First 2 Years of Follow-up eFigure 7. Hazard Ratios for Incident Diseases by Body Mass Index Category After Exclusion of Participants With Diseases Diagnosed Within the First 2 Years of Follow-up eFigure 8. Adjusted Hazard Ratios for Incident Diseases by Body Mass Index Category After Exclusion of Participants With Diseases Diagnosed Within the First 2 Years of Follow-up eFigure 9. Hazard Ratios for Incident Diseases by Lifestyle Profile in Individuals With Normal Weight (BMI 18.5-24.9)(A), Overweight (BMI 25-29.9)(B), and Obesity (BMI Ն30) (C) After Exclusion of Participants With Diseases Diagnosed Within the First 2 Years of Follow-up eFigure 10.Adjusted Hazard Ratios for Incident Diseases by Lifestyle Profile in Individuals With Normal Weight (BMI 18.5-24.9)(A), Overweight (BMI 25-29.9)(B), and Obesity (BMI Ն30) (C) After Exclusion of Participants With Diseases Diagnosed Within the First 2 Years of Follow-up eFigure 11.Association of Healthy Lifestyle Score With Incident Diseases in Adults With a Healthy Weight eFigure 12. Association of Healthy Lifestyle Score With Incident Diseases in Adults With Obesity JAMA Network Open.2023;6(5):e2314741.doi:10.1001/jamanetworkopen.2023.14741(Reprinted) May 26, 2023 12/13 Downloaded From: https://jamanetwork.com/ on 05/30/2023 eFigure 6.