Background: Though the benefits of healthy lifestyle choices are well-established among the general population, less is known about how developing and adhering to healthy lifestyle habits benefits obese versus normal weight or overweight individuals. The purpose of this study was to determine the association between healthy lifestyle habits (eating 5 or more fruits and vegetables daily, exercising regularly, consuming alcohol in moderation, and not smoking) and mortality in a large, population-based sample stratified by body mass index (BMI).
Methods: We examined the association between healthy lifestyle habits and mortality in a sample of 11,761 men and women from the National Health and Nutrition Examination Survey III; subjects were ages 21 and older and fell at various points along the BMI scale, from normal weight to obese. Subjects were enrolled between October 1988 and October 1994 and were followed for an average of 170 months.
Results: After multivariable adjustment for age, sex, race, education, and marital status, the hazard ratios (95% CIs) for all-cause mortality for individuals who adhered to 0, 1, 2, or 3 healthy habits were 3.27 (2.36–4.54), 2.59 (2.06–3.25), 1.74 (1.51–2.02), and 1.29 (1.09–1.53), respectively, relative to individuals who adhered to all 4 healthy habits. When stratified into normal weight, overweight, and obese groups, all groups benefited from the adoption of healthy habits, with the greatest benefit seen within the obese group.
Conclusions: Healthy lifestyle habits are associated with a significant decrease in mortality regardless of baseline body mass index.
In an effort to reduce premature morbidity and mortality, public health officials and health care providers have stressed the critical importance of adherence to a healthy lifestyle that includes exercising regularly, eating a diet high in fruits and vegetables, avoiding or quitting smoking, and consuming alcohol in moderation. Though the benefits of healthy lifestyle choices are well established among the general population, less is known about the relative benefit of healthy habits among obese versus normal weight and overweight individuals.1,2
A large body of epidemiologic evidence suggests that obesity is an independent risk factor for early mortality.3,4 Though the evidence for the risk of being overweight is mixed, most studies find no increased risk of mortality among overweight people.3–5 The increased risk of mortality associated with obesity is of considerable concern because more than 33% of American adults are categorized as obese (body mass index [BMI] ≥30).6 By even the most conservative estimates, obesity is responsible for more than 80,000 deaths annually.3 Obesity increases the risk of illnesses such as coronary artery disease, diabetes mellitus, hypertension, sleep apnea, and several types of cancer.7,8 In addition to increasing morbidity and mortality, obesity is a major financial strain on both individuals and society, with the direct annual medical costs estimated to be in excess of $147 billion dollars.9
Fortunately, regardless of an individual’s BMI, adherence to healthy lifestyle habits seems to decrease the risk of premature mortality.1,10,11 These findings may be of particular importance given that fewer than 1 in 5 Americans who attempt to lose weight are successful in maintaining weight loss over time.12,13
The purpose of this study was to determine the association between healthy lifestyle habits (consumption of ≥5 servings of fruits or vegetable/day, regular exercise >12 times per month, moderate alcohol consumption [up to 1 drink per day for women and up to 2 drinks per day for men], and not smoking) and all-cause mortality in US adults 21 years of age and older. The study focuses on this age range of the adult population because alcohol consumption is illegal in the United States among individuals younger than 21 years of age.
This study was reviewed by the institutional review board of the Medical University of South Carolina and was found to be exempt.
The National Health and Nutrition Examination Survey (NHANES) III is a nationally representative survey based on a complex sampling design that allows for weighted population estimates of the noninstitutionalized US population. The NHANES III was conducted in the United States between October 1988 and October 1994 by the National Center for Health Statistics. The sample was composed of approximately 40,000 noninstitutionalized US civilians, with a lower age limit of 2 months and no upper age limit. The survey included a physical examination, numerous laboratory studies, and an extensive questionnaire about lifestyle, diet, and medical history. The NHANES III survey and sampling procedure has been described previously in the literature.14
Definitions of Obesity, Overweight, and Normal Weight
For the purposes of this study, obesity was defined as a BMI ≥30, overweight as a BMI ≥25 and <30, and normal weight as a BMI ≥18.5 and <25, according to the definition provided by the National Institutes of Health.15 BMI was determined by taking actual measurements of the height and weight of participants then computing BMI based on weight (kg) divided by height (m2).
Healthy Lifestyle Factors
Smoking status was determined by self-report of whether the individual currently smoked cigarettes, pipes, or cigars. Intake of at least 5 servings of fruits and vegetables a day was obtained by using the healthy eating index dataset, which contains data on the number of servings of food consumed by survey participants. Level of physical activity was determined according to the frequency of participation in leisure-time physical activities within the previous month. These activities included walking, jogging or running, riding a bicycle or exercise bicycle, swimming, aerobic exercise, aerobic dancing, regular dancing, calisthenics, garden or yard work, weightlifting, or other similar activities.16 The average time spent exercising and level of intensity of participants’ exercise were not analyzed. Physical activity was divided into 2 frequency groups (0–12 and >12 times per month), a division consistent with national recommendations at the time of the NHANES III.17 Moderate alcohol consumption was defined as more than 0 but no more than 1 drink a day for women and up to 2 drinks a day for men, according to current guidelines of the US Department of Agriculture.18 To be consistent with previous studies, the 4 healthy habits were defined as engaging in physical activity >12 times a month, being a nonsmoker, consuming ≥5 servings of fruit and vegetables a day, and drinking alcohol in moderation.19,13
The NHANES III Mortality Study was a prospective cohort study comprised of individuals 17 years of age and older at the time of study enrollment. Deaths were determined by linking information from the National Death Index to matching personal identifiers in the NHANES III. Participants were followed from the time of study enrollment through December 31, 2006.
Demographic variables (age, sex, race, education, and marital status) were included as control variables because of their known impact on healthy habits.1,20 Age was treated as a continuous variable. Sex was categorized as either female or male. Race included three categories: non-Hispanic white, non-Hispanic black, and Hispanic. Marital status was dichotomized into married or living with a partner and nonmarried. Education was categorized by years of education completed as follows: 0 to 11 years, 12 years, 12 to 16 years, and ≥17 years.
To account for the complex sample design used in the NHANES, SUDAAN software (RTI International, Research Triangle Park, NC) was used for all analyses. The analyses incorporated both the stratification and clustering aspects of the sampling design. The proper weighting procedures included adjustments for basic probability of selection and adjustments for nonresponse. Because minorities were oversampled and the sampling design was complex, sampling weights provided in the NHANES public use dataset were used to compute population estimates based on weighted parameter estimates and standard errors.
Descriptive statistics for the sample stratified by BMI were computed (Table 1). A Cox proportional hazard ratio then was calculated for the entire sample, comparing the risk of all-cause mortality for individuals who adhered to 0, 1, 2, or 3 healthy habits versus individuals who adhered to all 4 healthy habits, controlling for age, race, and sex (Table 2). A survival analysis stratified by BMI then was performed using normal-weight individuals who adhered to all 4 healthy habits as the reference group (Table 3). Finally, a survival analysis was performed, examining each individual healthy habit (ie, engaging in physical activity, not smoking, consuming ≥5 fruits or vegetables daily, and drinking alcohol in moderation). All individuals were included in this analysis (Table 4). The proportionality assumption was tested for all survival analyses using Schoenfeld residuals. A Wald test also was performed to determine whether an interaction existed between healthy habits and BMI. No interaction between these was found, with P ranging between .1 and .9.
Using the inclusion criteria outlined in the Methods section, the unweighted sample of participants included a total of 11,761 individuals, which was equivalent to a weighted sample size of 133,700,723. Subjects were enrolled between October 1988 and October 1994 and were followed for an average of 170 months. A total of 2281 individuals (weighted, N = 18,437,659) died during the follow-up period.
After multivariable adjustment for age, sex, race, education, and marital status, the hazard ratios (95% CIs) for all-cause mortality for individuals who adhered to 0, 1, 2, or 3 healthy habits was 3.27 (2.36–4.54), 2.59 (2.06–3.25), 1.74 (1.51–2.02), and 1.29 (1.09–1.53), respectively, relative to individuals who adhered to all 4 healthy habits (Table 2). When a stratified survival analysis of the entire group was performed, with normal-weight individuals who reported adherence to all 4 healthy habits serving as the reference group, survival improved each time a new healthy habit was added in the overweight and obese groups (Table 3) (Figure 1).
” data-icon-position=”” data-hide-link-title=”0″>
In a survival analysis of individual healthy habits stratified by BMI (Table 4), smoking was associated with increased mortality regardless of baseline weight. Consumption of ≥5 servings of fruits and vegetables was associated with decreased mortality in normal-weight and overweight individuals but not in obese individuals. Regular exercise was associated with decreased mortality in normal-weight and obese individuals but failed to reach statistical significance in overweight individuals. Moderate alcohol consumption was associated with decreased mortality only in normal-weight individuals. The proportionality assumption was tested for all survival analyses using Schoenfeld residuals, which indicated that the proportional assumption was upheld.
The results of this study reinforce the association between healthy lifestyle habits and decreased mortality risk regardless of baseline BMI. This finding is of great importance to both patients and health care providers, whose perceptions about BMI may lead them to believe only obese and/or overweight patients require regular counseling about lifestyle adjustments. Although the evidence suggests that patients across the BMI spectrum should adhere to a healthy lifestyle to optimize health, many patients with a normal-weight BMI may believe exercise and healthy eating, for example, are less important for them as long as they maintain a low BMI.1,21,22 In fact, this patient view may be reinforced by health care providers who also are influenced by BMI and are, therefore, significantly more likely to counsel obese patients regarding a healthy lifestyle than normal-weight patients.1,22,23
The lack of time spent counseling patients to adopt a healthy lifestyle is reason for concern, and research suggests that this failure to counsel about healthy behaviors is common. For example, one study found primary care providers spent minimal time addressing nutrition (2.7% of total time), exercise (2.0% of time), or smoking (1.3% of time).24 Such a lack of focus on counseling is particularly troubling because counseling has been shown to be effective in decreasing smoking, increasing fruit and vegetable consumption, moderating alcohol consumption, and increasing exercise frequency.25–28 In the pooled analysis that included all individuals in the cohort (normal weight, overweight, and obese), the adoption of each additional healthy habit decreased all-cause mortality between 29% and 85% (Table 2). To put this in perspective, statins decrease all-cause mortality by 12% in individuals at high risk for cardiovascular disease.29 Given the tremendous benefits of a healthy lifestyle, policies and programs that encourage adherence to healthy lifestyles should be encouraged both locally and at a national level.
This study has several important strengths and weaknesses. The major strengths include the complex sampling design of the NHANES, which makes the findings generalizable to the population of the United States at large, and the accuracy of the National Death Index at assessing mortality.30,14 The study also has several noteworthy weaknesses. First, the cross-sectional nature of the interview portion of the survey did not account for changes in lifestyle habits over time. This may be considered a limitation even though lifestyle habits seem to be relatively stable in adulthood.31,32 Second, the survey was dependent on the accuracy of self-report of healthy habits. Third, although healthy habits were associated with decreased mortality, association does not prove causation. Finally, control variables were limited to age, race, and sex in the survival analysis because of uncertainty about how lifestyle interacted with other commonly used control variables such as socioeconomics or marriage.
Overall, the study reinforces previous findings about the critical importance of adhering to healthy habits. Future studies might focus on effective ways to increase adherence to healthy habits and the cost-effectiveness of such efforts.
This article was externally peer reviewed.
Conflict of interest: none declared.
Disclaimer: All authors had access to the data and a role in writing this manuscript. The National Health and Nutrition Examination Survey (NHANES) is conducted and supported by the National Center for Health Statistics (NCHS), which is part of the Centers for Disease Control and Prevention (CDC). This manuscript was prepared using the NHANES public use dataset and does not necessarily reflect the opinions or views of the NCHS or CDC.
- Received for publication May 3, 2011.
- Revision received August 26, 2011.
- Accepted for publication September 2, 2011.