Abstract
The objective of this study was to measure change in obesity prevalence among New York City (NYC) adults from 2004 to 2013–2014 and assess variation across sociodemographic subgroups. We used objectively measured height and weight data from the NYC Health and Nutrition Examination Survey to calculate relative percent change in obesity (≥ 30 kg/m2) between 2004 (n = 1987) and 2013–2014 (n = 1489) among all NYC adults and sociodemographic subgroups. We also examined changes in self-reported proxies for energy imbalance. Estimates were age-standardized and statistical significance was evaluated using two-tailed T tests and multivariable regression (p < 0.05). Between 2004 and 2013–2014, obesity increased from 27.5 to 32.4% (p = 0.01). Prevalence remained stable and high among women (31.2 to 32.8%, p = 0.53), but increased among men (23.4 to 32.0%, p = 0.002), especially among non-Latino White men and men age ≥ 65 years. Black adults had the highest prevalence in 2013–2014 (37.1%) and Asian adults experienced the largest increase (20.1 to 29.2%, p = 0.06), especially Asian women. Foreign-born participants and participants lacking health insurance also had large increases in obesity. We observed increases in eating out and screen time over time and no improvements in physical activity. Our findings show increases in obesity in NYC in the past decade, with important sociodemographic differences.
Keywords: Obesity, Disparities, Trends, Physical activity, Diet, Surveillance
Introduction
In the USA, the prevalence of obesity among adults significantly increased from 25.3% in 1976–1980 to 33.2% in 2003–2004 [1–4], with meaningful differences by race/ethnicity and gender and especially large increases among non-Latino Black women [5]. Later findings from the National Health and Nutrition Examination Survey (NHANES) revealed that between 2005 and 2014, the prevalence of obesity continued to increase for women only, regardless of race/ethnicity, but remained stable for men [2]. As obesity is linked to adverse health outcomes, including hypertension, diabetes, heart disease, stroke, and cancer [2, 6, 7], monitoring obesity is critical in shaping national and local prevention strategies [8].
The risk factors for obesity are multi-factorial and likely involve an interplay of genetics, environmental factors, social determinants, and lifestyle factors [9]. The literature suggests that several proxies for energy imbalance—such as sedentary behavior, inadequate physical activity, low fruit and vegetable intake, high eating out frequency, and excess total energy intake—may be linked to obesity [10–14], but results are highly mixed and inconsistent [15, 16]. Results from analyses of NHANES data have shown improvements in some of these indicators and worsening in others. For example, diet quality score (i.e., a diet pattern where higher score predicts lower cardiovascular disease risk) increased slightly and mean energy intake decreased between 1999 and 2012 [17–19], yet sedentary behaviors and the proportion of adults reporting no leisure-time physical activity increased [20]. In contrast, intake of total fruits and vegetables did not significantly change over this time period, and improvements in self-reported diet behaviors and leisure-time physical activity were not observed for racial and ethnic subgroups (e.g., non-Latino Black and Mexican American adults) [17]. These findings underscore the importance of collecting surveillance data, which can be used to track multiple components of obesity risk, including diet, physical activity, and sedentary behaviors.
National trends in obesity and associated risk factors are carefully tracked by population-based health surveys using objective measures, such as NHANES, but most state and local jurisdictions have limited data to monitor change over time, at best having self-reported height and weight data from telephone surveys. In New York City (NYC), a number of major initiatives to improve population health surveillance of chronic diseases and risk factors were launched in the past decade, with an aim to guide decision-making and evaluate progress achieved. As part of its monitoring and surveillance portfolio, the NYC Health Department launched the New York City Health and Nutrition Examination Survey (NYC HANES) in 2004, a population-based interview and physical exam survey with similar methods, labs, and standards as NHANES. In 2013–2014, a second NYC HANES was performed as an academic-government partnership, providing an opportunity to examine objectively measured changes in obesity prevalence at the municipal level in the context of local initiatives. During this time interval, a number of policies, programs, and educational campaigns were implemented in NYC to promote healthy living. Examples include calorie labeling in chain restaurants, zoning and tax incentives for supermarket creation and expansion (i.e., FRESH program), NYC Agency Food Standards, improvements in cycling infrastructure [21], the Shape Up NYC fitness program [22], and the increase in the number of farmer’s markets accepting Health Bucks [23].
Here, we describe changes in prevalence of obesity in NYC adults from 2004 to 2013–2014 using objective measures of height and weight collected from the NYC HANES survey. We examined changes in the distribution of obesity, BMI, and diet, physical activity, and screen time behaviors in NYC adults over the past 10 years, during which several interventions to support healthy living were implemented. We compared changes in obesity prevalence in NYC to corresponding changes at the national level using data from NHANES 2003–2004 and 2011–2012. We also examined local and national changes in obesity prevalence and physical activity behaviors within gender and racial/ethnic subgroups.
Methods
Study Sample
We used data from NYC HANES, a population-based, cross-sectional survey of non-institutionalized adult NYC residents (aged 20 years or older) [24, 25]. A three-stage cluster household sampling plan was used to obtain representative citywide estimates of health status indicators, including body mass index (BMI) and diet and physical activity behaviors. Data collection for study participants consisted of three components: a two-part interview, a brief physical examination, and a biologic specimen collection. Overall response rates were 55% (n = 1999) and 36% (n = 1527) in 2004 and 2013–2014, respectively. Participants with valid height and weight measurements were included in the current study, for final sample sizes of 1987 and 1489 in 2004 and 2013–2014, respectively. More information on survey design, study population, and quality of NYC HANES are published elsewhere [24, 25].
To obtain national estimates of obesity change, we used data from NHANES 2003–2004 and 2011–2012 [26], with a sample size of 10,927 and 7408, respectively, after excluding those without valid height and weight measurements. Comparison of estimates allows us to compare local progress to national secular change in population prevalence of obesity, which can inform policymakers as they tailor their public health strategies.
Outcome Variable
We calculated our outcome variable of obesity (≥ 30 kg/m2) using height and weight measurements from the physical examination. Based on World Health Organization’s recommendations [27], we used a modified cut-off point of ≥ 27.5 kg/m2 to categorize obesity for Asian participants. Trained staff collected weight to the nearest 0.1 kg and height to the nearest 0.5 cm, using protocols adapted from the NHANES Medical Examination Component manuals [28].
Covariates and Secondary Outcomes
Sociodemographic characteristics included age, gender, race/ethnicity, foreign-born status, duration of residence in the USA, education (less than high school, high school, some college, college degree, or more), neighborhood poverty level of census tract (0–< 10%, 10–< 20%, 20–< 30%, and 30–100% of population living below the federal poverty line in a census tract), health insurance status, and having a primary care provider.
To capture potential changes in consumption of high-calorie foods away from home and intake of low-calorie, nutrient-dense foods, we included self-reported weekly frequency of eating meals that were prepared in a restaurant and weekly intake of fruits and vegetables (times per day) as proxies for diet quality. Total screen time was calculated using responses to questions about how much time the participant spent sitting and watching TV or videos or using a computer outside of work. Using Healthy People 2020, participants were defined as physically active if they reported 150 min of moderate physical activity or 75 min of vigorous physical activity or 150 min of a combination of moderate and vigorous physical activity a week in at least 10 min episodes [29]. All questions related to covariates and secondary outcomes were identical across survey years.
Statistical Analysis
To account for the complex sampling design and non-response, we statistically weighted all analyses, re-adjusting weights to account for missing height and weight values; thus, estimates are representative of the NYC population. We calculated the relative percent change in primary and secondary outcome variables by dividing the observed change in values between survey years (2013–2014 and 2004) by the baseline value (2004), and statistical significance of change was tested using a T test with a two-tailed p value < 0.05. Prevalence estimates were age-standardized to the 2000 US population. The primary outcome variable (obesity) was stratified by gender, age, and racial/ethnic subgroups. We also calculated change in obesity prevalence among foreign-born participants by duration of residence and health insurance status. We examined changes in secondary behavioral outcomes in the overall NYC adult population and by gender and obesity status. We flagged as imprecise and potentially unreliable any estimates with relative standard errors greater than 30%, 95% confidence interval half-widths greater than 10, or sample denominators less than 50. We used the “proc descript” statement and the “t_pct” option in SAS 9.4 (SAS Institute Inc., Cary, NC) and SUDAAN version 9.0 (Research Triangle Institute, Research Triangle Park, NC).
In sensitivity analyses, we calculated the relative percent change in obesity without using the modified cut-off point for Asian participants; and we calculated the percent change in physical activity behavior using guidelines from Healthy People 2010, which have stricter criteria for classifying individuals as meeting physical activity recommendations than current guidelines [30]. To determine whether the distribution of BMI in our samples shifted between survey years, we also plotted BMI values at 2004 and 2013–2014 using nonparametric density curves.
To explore whether sociodemographic characteristics and behaviors explained secular changes in obesity prevalence, we constructed two log-binomial regression models to examine the differential magnitude in association between obesity prevalence and survey year, with models either adjusting for (1) age, racial/ethnic origin, foreign-born status, and education, or (2) age, diet, physical activity, and screen time behaviors. The potential for interaction between gender and other variables was assessed but interaction terms lacked statistical significance and were excluded from the final model. We used a change-in-estimate approach to determine the extent to which including these variables in our model accounted for the observed secular change across the two time periods; specifically, we evaluated the magnitude with which inclusion of sociodemographic or behavioral variables in the model changed the survey year beta estimate greater than 10% relative to a crude model [31].
To compare changes in local and national estimates, we calculated the relative percent change in primary outcome variables between NHANES survey years 2011–2014 and 2001–2004 (the modified cut-off point for Asian participants in NYC HANES was not used to be consistent with NHANES), and tested statistical significance of change using a T test with a two-tailed p value < 0.05. We also compared the relative percent change in leisure-time physical activity using both data sources. However, we acknowledge that a new physical activity questionnaire was introduced in NHANES 2007–2008, so estimates of change at the national level should be interpreted with caution (NYC HANES physical activity questions match baseline NHANES physical activity questions).
Results
Between 2004 and 2013–2014, the age-standardized prevalence of obesity in the NYC adult population increased significantly, from 27.5 to 32.4% (p = 0.01) (Table 1). Overall, the distribution of BMI values in the general adult population shifted towards higher values by 2013–2014 (Fig. 1). Prevalence of obesity remained stable and high among women (31.2% in 2004 and 32.8% in 2013–2014, p = 0.53), while we observed a statistically significant increase among men (23.4% in 2004 and 32.0% in 2013–14; p = 0.002). Non-Latino White adults had the lowest prevalence of obesity in 2013–2014 (27.8%), while non-Latino Black adults had the highest prevalence of obesity (37.1%). Relative to other racial/ethnic groups, Asian adults experienced the largest increase in obesity over time (20.1 to 29.2%, p = 0.06), although changes over time by race/ethnicity were not statistically significant.
Table 1.
Changes in prevalence of obesity among NYC adults (aged ≥ 20 years) by sociodemographic characteristics, NYC HANES 2004 and 2013–2014
| 2004 | 2013–2014 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total sample | Obesity prevalence* | Total sample | Obesity prevalence* | Change over time | ||||||||
| Characteristic | N | % | N | % | 95% CI | N | % | N | % | 95% CI | Relative change in obesity | p value |
| Total | 1987 | NA | 538 | 27.5 | 25.0–30.0 | 1489 | NA | 455 | 32.4 | 29.7–35.2 | 17.8 | 0.01 |
| Age group | ||||||||||||
| 20–44 | 1196 | 52.8 | 278 | 25.0 | 21.3–29.2 | 810 | 51.0 | 200 | 29.3 | 25.1–33.8 | 17.2 | 0.16 |
| 45–64 | 611 | 31.7 | 197 | 30.1 | 25.6–34.9 | 477 | 32.6 | 180 | 36.3 | 31.4–41.5 | 20.6 | 0.07 |
| ≥ 65 | 180 | 15.4 | 63 | 34.3 | 27.8–41.4 | 202 | 16.4 | 75 | 40.4 | 33.2–48 | 17.8 | 0.23 |
| Gender | ||||||||||||
| Male | 829 | 46.1 | 196 | 23.4 | 20.3–26.8 | 622 | 46.6 | 180 | 32.0 | 27.7–36.5 | 36.8 | 0.002 |
| Female | 1158 | 53.9 | 342 | 31.2 | 28.0–34.5 | 867 | 53.4 | 275 | 32.8 | 29.2–36.6 | 5.1 | 0.53 |
| Race/ethnicity | ||||||||||||
| Non-Latino White | 611 | 38.3 | 134 | 22.6 | 19.1–26.5 | 499 | 35.0 | 123 | 27.8 | 23.4–32.7 | 23.0 | 0.09 |
| Non-Latino Black | 431 | 23.1 | 146 | 33.7 | 28.6–39.2 | 328 | 21.3 | 124 | 37.1 | 31.4–43.2 | 10.1 | 0.40 |
| Latino | 655 | 26.1 | 202 | 32.0 | 27.7–36.6 | 385 | 27.1 | 136 | 36.9 | 31.9–42.2 | 15.3 | 0.15 |
| Asian | 259 | 10.9 | 51 | 20.1 | 14.6–26.9 | 198 | 14.0 | 51 | 29.2 | 22.3–37.2 | 45.3 | 0.06 |
| Education | ||||||||||||
| Less than high school | 570 | 27.1 | 198 | 34.4 | 29.9–39.1 | 311 | 18.9 | 130 | 40.7 | 35.1–46.5 | 18.3 | 0.09 |
| High school | 382 | 19.1 | 100 | 25.9 | 21.2–31.2 | 231 | 23.1 | 87 | 36.5 | 29.9–43.6 | 40.9 | 0.01 |
| Some college | 416 | 20.5 | 109 | 27.8 | 23.2–33.1 | 329 | 22.7 | 111 | 34.8 | 29.4–40.7 | 25.2 | 0.07 |
| College degree or more | 612 | 33.3 | 128 | 21.9 | 17.9–26.6 | 616 | 35.3 | 127 | 24.0 | 20.0–28.5 | 9.6 | 0.50 |
| Neighborhood poverty level | ||||||||||||
| 0–< 10% | 487 | 28.0 | 105 | 22.5 | 18.1–26.9 | 399 | 27.6 | 109 | 29.9 | 24.3–36.2 | 32.3 | 0.06 |
| 10–< 20% | 569 | 32.4 | 133 | 24.3 | 19.8–28.7 | 502 | 32.4 | 149 | 32.1 | 27.3–37.3 | 31.6 | 0.03 |
| 20–< 30% | 273 | 12.6 | 86 | 33.4 | 25.9–40.8 | 324 | 21.2 | 89 | 28.1 | 23.8–32.8 | − 16.4 | 0.21 |
| 30–100% | 658 | 27.0 | 214 | 32.6 | 28.3–36.8 | 264 | 18.8 | 108 | 41.4 | 35.5–47.6 | 22.1 | 0.05 |
| Country of birth | ||||||||||||
| US born | 967 | 51.2 | 299 | 31.3 | 27.7–35.2 | 831 | 54.0 | 247 | 32.9 | 29.4–36.6 | 5.1 | 0.57 |
| Non-US born | 1014 | 48.8 | 237 | 23.4 | 20.4–26.7 | 651 | 46.0 | 205 | 31.6 | 27.3–36.2 | 35.0 | 0.002 |
| Health insurance | ||||||||||||
| Has health insurance | 1507 | 79.2 | 428 | 28.6 | 25.8–31.6 | 1240 | 83.0 | 390 | 33.1 | 30.2–36 | 15.7 | 0.04 |
| No health insurance | 464 | 20.8 | 107 | 21.6 | 17.0–27.1 | 246 | 17.0 | 64 | 33.4 | 26–41.7 | 54.6 | 0.01 |
| Have a primary care provider | ||||||||||||
| Yes | 1488 | 78.1 | 436 | 29.1 | 26.4–32.0 | 1218 | 81.8 | 398 | 33.7 | 30.7–36.9 | 15.8 | 0.04 |
| No | 483 | 21.9 | 101 | 21.5 | 16.4–27.8 | 270 | 18.2 | 57 | 26.6 | 19.9–34.6 | 23.7 | 0.29 |
| BMI | ||||||||||||
| < 25 | 794 | 38.6 | NA | NA | NA | 558 | 35.2 | NA | NA | NA | NA | NA |
| 25–29 | 682 | 35.4 | NA | NA | NA | 500 | 34.4 | NA | NA | NA | NA | NA |
| ≥ 30 | 511 | 26.1 | NA | NA | NA | 431 | 30.4 | NA | NA | NA | NA | NA |
| Non-Latina White, female | 321 | 18.7 | 76 | 25.1 | 20.2–30.7 | 268 | 17.1 | 56 | 23.4 | 18.5–29 | − 6.8 | 0.67 |
| Non-Latino White, male | 290 | 19.6 | 58 | 19.8 | 15.3–25.4 | 231 | 17.9 | 67 | 32.0 | 25–39.9 | 61.6 | 0.01 |
| Black, female | 269 | 13.4 | 109 | 40.3 | 34.1–47.0 | 208 | 12.5 | 86 | 40.0 | 32.2–48.4 | − 0.7 | 0.95 |
| Black, male | 162 | 9.8 | 37 | 24.4 | 18.2–32.0 | 120 | 8.7 | 38 | 33.0 | 24.1–43.4 | 35.2 | 0.16 |
| Latina, female | 406 | 15.3 | 129 | 33.9 | 28.4–40.0 | 231 | 15.0 | 85 | 37.8 | 31–45.2 | 11.5 | 0.39 |
| Latino, male | 249 | 10.9 | 73 | 30.1 | 23.5–37.6 | 154 | 12.2 | 51 | 35.5 | 27.4–44.4 | 17.9 | 0.34 |
| Asian, female | 145 | 5.7 | 26 | 19.8 | 13.1–28.8 | 117 | 7.5 | 33 | 32.7 | 24.6–41.9 | 65.2 | 0.03 |
| Asian, male | 114 | 5.2 | 25 | 20.5 | 13.3–30.3 | 81 | 6.5 | 18 | 24.1† | 13.6–38.9 | 17.6 | 0.65 |
BMI, body mass index (kg/m2)
*All estimates were age-standardized to the 2000 US population
†Estimate should be interpreted with caution as relative standard error was greater than 30%, 95% confidence interval half-width was greater than 10, or sample denominator was less than 50
Fig. 1.
Distribution of BMI (kg/m2) in NYC HANES 2004 and 2013–2014
In subgroup analyses, observed increases among men were largely concentrated in non-Latino White men, with a 61.6% increase between survey years (p = 0.01). We also observed a statistically significant increase among Asian women (19.8 to 32.7%; p = 0.03). The magnitude and direction of changes were consistent with and without using the modified obesity cut-off point for Asian participants (data not shown). The prevalence of obesity also increased among men age 65 years and older (25.9 to 41.3%; p = 0.04), though the precision is limited due to small sample size.
Increases in obesity prevalence were greatest among adults with no health insurance and among foreign-born participants. The change in obesity among foreign-born participants who were uninsured (20.0 to 33.7%, p = 0.02) was more than twice the increase among foreign-born participants who were insured (24.3 to 31.9%, p = 0.02) (data not shown). We also observed a statistically significant increase in obesity prevalence among those residing in the USA for more than 10 years (26.6 to 34.3%; p = 0.03), but not among those residing in the USA for 10 or fewer years (19.3 to 27.4%; p = 0.15). We did not observe meaningful differences in obesity across levels of neighborhood poverty or educational attainment over time.
We observed several meaningful changes in proxies for energy imbalance among NYC adults over time, including dietary behaviors, physical activity, and screen time (Table 2). Between 2004 and 2013–2014, the level of self-reported weekly intake of fruits and vegetables decreased by 28.2% for fruits (p < 0.001) and 27.7% for vegetables (p < 0.001). The weekly frequency of eating out increased from a citywide average of 2.7 meals per week in 2004 to 3.8 meals per week in 2013–2014 (40.7%; p < 0.001).
Table 2.
Changes in obesity risk factors (mean/%) among (a) all NYC adults, (b) obese NYC adults, and (c) non-obese NYC adults (aged 20 years and older), NYC HANES 2004 and 2013–2014.
| 2004 | 2013–2014 | |||||
|---|---|---|---|---|---|---|
| N | %/mean | N | %/mean | % change | p value | |
| (a) All NYC adults | ||||||
| Physical activity | ||||||
| Recommended physical activity | 785 | 43.1 | 585 | 42.4 | − 1.6 | 0.71 |
| Some physical activity | 459 | 26.0 | 377 | 25.4 | − 2.3 | 0.72 |
| No physical activity | 516 | 30.9 | 439 | 32.2 | 4.2 | 0.52 |
| Diet^ | ||||||
| Weekly eating out frequency | 1972 | 2.7 | 1486 | 3.8 | 40.7 | < 0.001 |
| Weekly vegetable intake | 1971 | 13.7 | 1488 | 9.9 | − 27.7 | < 0.001 |
| Weekly fruit intake | 1971 | 7.8 | 1488 | 5.6 | − 28.2 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 933 | 48 | 946 | 63.5 | 32.3 | < 0.001 |
| 0– < 3 h/day | 1039 | 52 | 543 | 36.5 | − 29.8 | < 0.001 |
| (b) Obese NYC adults | ||||||
| Physical activity | ||||||
| Recommended physical activity | 198 | 41.6 | 155 | 39.7 | − 4.6 | 0.61 |
| Some physical activity | 124 | 26.3 | 105 | 22.1 | − 16.0 | 0.17 |
| No physical activity | 152 | 32.2 | 166 | 38.3 | 18.9 | 0.09 |
| Diet^ | ||||||
| Weekly eating out frequency | 537 | 2.5 | 453 | 3.6 | 44.0 | 0.00 |
| Weekly vegetable intake | 536 | 13.1 | 454 | 9.3 | − 29.0 | < 0.001 |
| Weekly fruit intake | 536 | 8.3 | 454 | 5 | − 39.8 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 286 | 53.8 | 311 | 67.1 | 24.7 | < 0.001 |
| 0– < 3 h/day | 251 | 46.2 | 144 | 32.9 | − 28.8 | < 0.001 |
| (c) Non-obese NYC adults | ||||||
| Physical activity | ||||||
| Recommended physical activity | 587 | 43.7 | 430 | 44.4 | 1.6 | 0.77 |
| Some physical activity | 335 | 26.0 | 272 | 26.5 | 1.9 | 0.83 |
| No physical activity | 364 | 30.3 | 273 | 29.2 | − 3.6 | 0.63 |
| Diet^ | ||||||
| Weekly eating out frequency | 1435 | 2.7 | 1033 | 3.9 | 44.4 | < 0.001 |
| Weekly vegetable intake | 1435 | 13.9 | 1034 | 10.3 | − 25.9 | < 0.001 |
| Weekly fruit intake | 1435 | 7.8 | 1034 | 5.9 | − 24.4 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 647 | 46 | 635 | 61.6 | 33.9 | < 0.001 |
| 0– < 3 h/day | 788 | 54 | 399 | 38.4 | − 28.9 | < 0.001 |
*All estimates were standardized to the 2000 US population
^Diet questions were analyzed as continuous variables
Using Healthy People 2020 guidelines, we did not observe any meaningful changes in the percentage of adults meeting physical activity recommendations (p = 0.71) or the percentage of adults reporting no leisure-time physical activity (p = 0.52). The percentage of NYC adults reporting screen time of 3 or more hours per day increased by 32.3% (p < 0.001). Secular changes in diet, physical activity, and screen time were comparable in magnitude between obese adults and non-obese adults (Appendix Table 4).
Table 4.
Changes in obesity risk factors (mean/%) among (a) all NYC adults, (b) obese NYC adults, and (c) non-obese NYC adults (aged 20 years and older), by gender, NYC HANES 2004 and 2013–2014
| 2004 | 2013–2014 | |||||
|---|---|---|---|---|---|---|
| N | %/mean | N | %/mean | % change | p value | |
| (a) All NYC adults, males | ||||||
| Physical activity | ||||||
| Recommended physical activity | 366 | 47.4 | 296 | 50.2 | 5.9 | 0.35 |
| Some physical activity | 175 | 24.9 | 152 | 24.5 | − 1.6 | 0.89 |
| No physical activity | 197 | 27.8 | 139 | 25.3 | − 9.0 | 0.40 |
| Diet^ | ||||||
| Weekly eating out frequency | 823 | 3.4 | 622 | 4 | 17.6 | 0.09 |
| Weekly vegetable intake | 822 | 12.4 | 622 | 9.5 | − 23.4 | < 0.001 |
| Weekly fruit intake | 822 | 7.1 | 622 | 5.2 | − 26.8 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 409 | 49.9 | 404 | 65 | 30.3 | < 0.001 |
| 0–< 3 h/day | 413 | 50.1 | 218 | 35 | − 30.1 | < 0.001 |
| (b) Obese NYC adults | ||||||
| Physical activity | ||||||
| Recommended physical activity | 82 | 46.8 | 86 | 51.8 | 10.7 | 0.41 |
| Some physical activity | 45 | 27.9 | 29 | 14.7 | − 47.3 | 0.01 |
| No physical activity | 48 | 25.2 | 53 | 33.4 | 32.5 | 0.13 |
| Diet^ | ||||||
| Weekly eating out frequency | 196 | 3.5 | 180 | 4.1 | 17.1 | 0.30 |
| Weekly vegetable intake | 195 | 12 | 180 | 8.6 | − 28.3 | 0.04 |
| Weekly fruit intake | 195 | 7.8 | 180 | 4.6 | − 41.0 | 0.00 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 111 | 57.3 | 117 | 63.1 | 10.1 | 0.35 |
| 0–< 3 h/day | 85 | 42.7 | 63 | 36.9 | − 13.6 | 0.35 |
| (c) Non-obese adults, males | ||||||
| Physical activity | ||||||
| Recommended physical activity | 284 | 47.4 | 210 | 49.8 | 5.1 | 0.47 |
| Some physical activity | 130 | 24.1 | 123 | 28.6 | 18.7 | 0.15 |
| No physical activity | 149 | 28.5 | 86 | 21.6 | − 24.2 | 0.03 |
| Diet^ | ||||||
| Weekly eating out frequency | 627 | 3.4 | 442 | 3.9 | 14.7 | 0.18 |
| Weekly vegetable intake | 627 | 12.6 | 442 | 10 | − 20.6 | 0.00 |
| Weekly fruit intake | 627 | 6.9 | 442 | 5.5 | − 20.3 | 0.01 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 298 | 47.6 | 287 | 65.8 | 38.2 | < 0.001 |
| 0–< 3 h/day | 328 | 52.4 | 155 | 34.2 | − 34.7 | < 0.001 |
| (d) All NYC adults, females | ||||||
| Physical activity | ||||||
| Recommended physical activity | 419 | 39.3 | 289 | 35.6 | − 9.4 | 0.14 |
| Some physical activity | 284 | 27.0 | 225 | 26.2 | − 3.0 | 0.72 |
| No physical activity | 319 | 33.7 | 300 | 38.2 | 13.4 | 0.08 |
| Diet^ | ||||||
| Weekly eating out frequency | 1149 | 2.1 | 864 | 3.7 | 76.2 | < 0.001 |
| Weekly vegetable intake | 1149 | 14.7 | 866 | 10.3 | − 29.9 | < 0.001 |
| Weekly fruit intake | 1149 | 8.5 | 866 | 5.8 | − 31.8 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 524 | 46.6 | 542 | 62.1 | 33.3 | < 0.001 |
| 0–< 3 h/day | 626 | 53.4 | 325 | 37.9 | − 29.0 | < 0.001 |
| (e) Obese adults, females | ||||||
| Physical activity | ||||||
| Recommended physical activity | 116 | 38.4 | 69 | 29.2 | − 24.0 | 0.03 |
| Some physical activity | 79 | 25.4 | 76 | 28.6 | 12.6 | 0.44 |
| No physical activity | 104 | 36.1 | 113 | 42.2 | 16.9 | 0.18 |
| Diet^ | ||||||
| Weekly eating out frequency | 341 | 1.8 | 273 | 3.3 | 83.3 | 0.00 |
| Weekly vegetable intake | 341 | 13.9 | 274 | 9.9 | − 28.8 | < 0.001 |
| Weekly fruit intake | 341 | 8.7 | 274 | 5.2 | − 40.2 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 175 | 51.6 | 194 | 71.1 | 37.8 | < 0.001 |
| 0–< 3 h/day | 166 | 48.4 | 81 | 28.9 | − 40.3 | < 0.001 |
| (f) Non-obese adults, females | ||||||
| Physical activity | ||||||
| Recommended physical activity | 303 | 40.4 | 220 | 39.7 | − 1.7 | 0.82 |
| Some physical activity | 205 | 27.5 | 149 | 24.5 | − 10.9 | 0.27 |
| No physical activity | 215 | 32.1 | 187 | 35.8 | 11.5 | 0.23 |
| Diet^ | ||||||
| Weekly eating out frequency | 808 | 2.1 | 591 | 3.9 | 85.7 | < 0.001 |
| Weekly vegetable intake | 808 | 15.2 | 592 | 10.5 | − 30.9 | < 0.001 |
| Weekly fruit intake | 808 | 8.7 | 592 | 6.1 | − 29.9 | < 0.001 |
| Screen time (total) | ||||||
| ≥ 3 h/day | 349 | 44.8 | 348 | 57.8 | 29.0 | < 0.001 |
| 0–< 3 h/day | 460 | 55.2 | 244 | 42.2 | − 23.6 | < 0.001 |
Adjustment for age, gender, race/ethnicity, foreign-born status, and educational attainment [β = 0.15 (SE = 0.06)] did not meaningfully lower the magnitude of the observed shift in obesity prevalence between survey years, when compared to a model adjusting for age only [β = 0.17 (SE = 0.06)]. However, adjusting for age, diet, physical activity, and screen time behavior accounted for approximately 22.5% of the increase in obesity prevalence [β = 0.13 (SE = 0.07)], compared to the model adjusting for age only.
Compared to the increase in obesity prevalence among NYC adults, change in the prevalence of obesity among adults nationwide was substantially larger (31.5% increase nationally, p < 0.001; and 17.1% increase in NYC, p = 0.02) (Table 3). Nationally, the increase in obesity was greater for women (34.5%, p < 0.001) than men (28%, p < 0.001).
Table 3.
Changes in obesity prevalence and physical activity behavior among all US and NYC adults (aged 20 years and older), NHANES 2003–2004 and 2011–2012 and NYC HANES 2004 and 2013–2014
| NYC HANES | NHANES | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2004 | 2013–2014 | Change | 2001–2004 | 2011–2014 | Change | |||||||
| N | % | N | % | % | p value | N | % | N | % | % | p value | |
| Total population | ||||||||||||
| Total obesity prevalence* | 511 | 26.3 | 431 | 30.8 | 17.1% | 0.02 | 4852 | 27.6 | 3956 | 36.3 | 31.5% | < 0.001 |
| Physical activity | ||||||||||||
| Recommended physical activity | 785 | 43.1 | 585 | 42.4 | − 1.6% | 0.71 | 9500 | 62.8 | 3568 | 37.5 | − 40.3% | < 0.001 |
| Some physical activity | 459 | 26.0 | 377 | 25.4 | − 2.3% | 0.72 | 2957 | 18.3 | 1738 | 17.0 | − 7.1% | 0.1578 |
| No physical activity | 516 | 30.9 | 439 | 32.2 | 4.2% | 0.52 | 3932 | 18.9 | 5425 | 45.5 | 140.7% | < 0.001 |
| Female | ||||||||||||
| Total obesity prevalence* | 330 | 30.3 | 261 | 31.2 | 3.0% | 0.73 | 2577 | 28.4 | 2261 | 38.2 | 34.5% | < 0.001 |
| Physical activity | ||||||||||||
| Recommended physical activity | 419 | 39.3 | 289 | 35.6 | − 9.4 | 0.14 | 4379 | 58.3 | 1619 | 34.2 | − 41.3% | < 0.001 |
| Some physical activity | 284 | 27.0 | 225 | 26.2 | − 3.0 | 0.72 | 1668 | 20.9 | 987 | 18.6 | − 11.0% | 0.043 |
| No physical activity | 319 | 33.7 | 300 | 38.2 | 13.4 | 0.08 | 2170 | 20.8 | 2921 | 47.3 | 127.4% | < 0.001 |
| Male | ||||||||||||
| Total obesity prevalence* | 181 | 21.9 | 170 | 30.2 | 37.9% | 0.00 | 2275 | 26.8 | 1695 | 34.3 | 28.0% | < 0.001 |
| Physical activity | ||||||||||||
| Recommended physical activity | 366 | 47.4 | 296 | 50.2 | 5.9 | 0.35 | 5121 | 67.2 | 1949 | 41.1 | − 38.8% | < 0.001 |
| Some physical activity | 175 | 24.9 | 152 | 24.5 | − 1.6 | 0.89 | 1289 | 15.7 | 751 | 15.4 | − 1.9% | 0.781 |
| No physical activity | 197 | 27.8 | 139 | 25.3 | − 9.0 | 0.40 | 1762 | 17.0 | 2504 | 43.5 | 155.9% | < 0.001 |
All estimates were standardized to the 2000 US population
*Asian specific cutoffs were not used in this table for the purpose of comparison with NHANES
Discussion
Using objective measures of height and weight of participants drawn from a rigorous and representative sampling scheme, we characterized change in obesity prevalence among adults living in NYC from 2004 to 2013–2014, providing local objective information on the obesity epidemic not available in other municipalities. Overall, we found that obesity increased by 17.8% in the NYC adult population, which is sizable yet significantly smaller than national trends. With an understanding that a long-term objective is to reduce the burden of obesity, a relatively smaller increase compared to national trends is notable. Our findings are also consistent with self-reported changes from the NYC Community Health Survey (CHS), a cross-sectional telephone survey of behavioral risk factors and chronic disease, with an annual sample of approximately 10,000 randomly selected adults aged 18 and older from all five boroughs of New York City. The survey results indicated that the prevalence of obesity increased from 21.7% in 2004 to 24.7% in 2014 (p < 0.001) [28]. Although absolute values for obesity prevalence were lower in NYC CHS compared to NYC HANES data, previous work has shown that BMI from self-reported measures is systematically lower than BMI from objective height and weight measures [32].
In NYC, we observed increases in obesity for men but not in women, a finding that is inconsistent with the recent national pattern of increased obesity prevalence among women only [2, 33]. We also observed a meaningful increase in obesity among older men, which is corroborated by recent national estimates; however, small sample size limited our ability to confirm the magnitude of change with precision [34]. Relative changes in the diet and physical activity behaviors examined in this paper do not explain the local gender differences between men and women, though differences in other unmeasured factors—such as total energy intake, diet quality, or body composition—may be responsible for the discrepancy. Consistent with national estimates and other studies, Black women in NYC had the highest prevalence of obesity [2, 35]. We also observed increases in obesity among Asian women, a group with a historically low prevalence of obesity. Our findings highlight the potential need to tailor interventions for specific demographic subgroups.
In addition to changes in obesity prevalence, our findings show that consumption of fruits and vegetables and frequency of eating out worsened among NYC adults, and sedentary behaviors such as hours spent using a computer and watching TV increased; while, physical activity behavior did not change meaningfully. Further, results from the log-binomial regression model suggest that these proxies for energy balance partially explain the increase in obesity between exam years. Though we lacked data related to total energy intake and additional food groups, these findings may indicate that potentially important changes in energy balance (increases in calories consumed and increases in sedentary behaviors) may have contributed to increases in obesity in NYC over time [11, 15, 36].
We found that the prevalence of obesity increased by 35.0% for foreign-born adults, and that this increase was largely driven by those residing in the USA for 10 or more years. This is consistent with literature showing that increasing duration of residence in the USA is linked to higher BMI [37]. We also found that the prevalence of obesity increased by 57.4% for those with no health insurance, and a 68.5% increase in obesity for foreign-born adults lacking health insurance (who are more likely to lack health insurance in our sample). This finding is disconcerting because a lack of health insurance could lead to poor management of obesity complications and co-morbidities (e.g., diabetes) among obese individuals, with a potential loss in work-related income, disability, and premature mortality [38, 39]. From this examination survey, it is difficult to determine how a lack of health insurance may impact obesity risk. While such an association may be driven by underlying socioeconomic differences, lack of health insurance might also be associated with work status and non-standard work hours, which have been shown to increase risk of non-optimal sleep, greater recreational screen time, worse dietary practices, and depression [40]. Lastly, we found that obesity change did not vary by neighborhood poverty level. While we cannot assess from the data collected in NYC HANES, one possibility is that in addition to neighborhood poverty, factors such as work hours, culture, food options, and preferences may be important.
Our study had several limitations, including the lack of longitudinal data on study participants. The cross-sectional nature of the study design does not allow us to capture individual change in weight status and obesity risk factors. Although we were able to measure changes in fruit and vegetable intake and eating out frequency at each year, we were not able to investigate changes in energy intake or potential substitution effects via consumption of other food groups and beverages. Further, direct comparisons between local and national physical activity behavior was not possible due to slight differences in physical activity survey questions. Data related to diet, physical activity, and screen time behaviors were also self-reported, potentially introducing imprecision and bias. In addition, geographic data in NHANES were not available in the public-use files, so we were not able to restrict the national sample to urban core areas and compare changes in NYC to changes in other urban areas. Despite these limitations, our study also had several strengths, including using unique, local objective data related to height and weight, allowing us to track the obesity epidemic in NYC.
Our findings show increases in the prevalence of obesity, eating out in restaurants, and sedentary behaviors, as well as decreases in fruit and vegetable consumption, among NYC adults between 2004 and 2013–2014. We observed important differences in trends by gender and race/ethnicity, with relatively larger increases in obesity among men and Asian populations. For NYC men, this rise in obesity was consistent with the national rise in obesity among men during the same time period; whereas, obesity among NYC women remained stable, while the national prevalence increased. Awareness of how obesity and diet, physical activity, and screen time behaviors are changing will assist local policy makers in tracking the impact of efforts to decrease the prevalence of obesity.
Public Health Implications
Obesity prevalence increased in NYC between 2004 and 2013–2014, but less so than observed changes in national prevalence over the same time period. We found that several populations were at greater risk, including non-Latino White men, men age ≥ 65 years older men, Asian adults, Black women, foreign-born adults, and those lacking health insurance coverage. A number of interventions to sustainably support healthy living—which can help prevent diet-related chronic disease, such as type-2 diabetes and hypertension—have been implemented in NYC over the past decade [21, 22]. Whether such efforts contributed to the slower increase in NYC cannot be discerned from these findings. Regardless, it is important to address observed increases in obesity among specific subgroups (e.g., older men, foreign-born adults, those lacking health insurance), potentially with targeted resources. Since it is difficult for adults to lose weight and maintain weight loss [41], and because a substantial proportion of the NYC population is currently obese, it may take decades before obesity prevalence decreases meaningfully.
Acknowledgments
Support for NYCHANES 2013–2014 was provided by the de Beaumont Foundation with additional support from the Robert Wood Johnson Foundation, Robin Hood, the NYS Health Foundation, Quest Diagnostics, and the Doris Duke Charitable Foundation, NYC Health Department, Hunter College Office of the Provost, CUNY Vice Chancellors Office of Research, and CUNY School of Public Health Dean’s Office.
Appendix
References
- 1.Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA. 2002;288(14):1723–1727. doi: 10.1001/jama.288.14.1723. [DOI] [PubMed] [Google Scholar]
- 2.Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. Trends in obesity among adults in the United States, 2005 to 2014. JAMA. 2016;315(21):2284–2291. doi: 10.1001/jama.2016.6458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL. Increasing prevalence of overweight among US adults: the National Health and Nutrition Examination Surveys, 1960 to 1991. JAMA. 1994;272(3):205–211. doi: 10.1001/jama.1994.03520030047027. [DOI] [PubMed] [Google Scholar]
- 4.Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA. 2006;295(13):1549–1555. doi: 10.1001/jama.295.13.1549. [DOI] [PubMed] [Google Scholar]
- 5.Romero CX, Romero TE, Shlay JC, Ogden LG, Dabelea D. Changing trends in the prevalence and disparities of obesity and other cardiovascular disease risk factors in three racial/ethnic groups of USA adults. Adv Prev Med. 2012;2012:1–8. doi: 10.1155/2012/172423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.National Heart, Lung, and Blood Institute. Managing overweight and obesity in adults: aystematic evidence review from the obesity expert panel, 2013. US Department of Health and Human Services. National Institutes of Health. Bethesda, Maryland; 2013.
- 7.Executive summary: guidelines (2013) for the management of overweight and obesity in adults. Obesity. 2014;(22 Suppl 2):S5–39. [DOI] [PubMed]
- 8.Pratt CA, Loria CM, Arteaga SS, Nicastro HL, Lopez-Class M, de Jesus JM, Srinivas P, Maric-Bilkan C, Schwartz Longacre L, Boyington JEA, Wouhib A, Gavini N. A systematic review of obesity disparities research. Am J Prev Med. 2017;53(1):113–122. doi: 10.1016/j.amepre.2017.01.041. [DOI] [PubMed] [Google Scholar]
- 9.Caprio S, Daniels SR, Drewnowski A, Kaufman FR, Palinkas LA, Rosenbloom AL, Schwimmer JB, Kirkman MS. Influence of race, ethnicity, and culture on childhood obesity: implications for prevention and treatment. Obesity (Silver Spring, Md) 2008;16(12):2566–2577. doi: 10.1038/oby.2008.398. [DOI] [PubMed] [Google Scholar]
- 10.Asghari G, Mirmiran P, Yuzbashian E, Azizi F. A systematic review of diet quality indices in relation to obesity. Br J Nutr. 2017;117(8):1055–1065. doi: 10.1017/S0007114517000915. [DOI] [PubMed] [Google Scholar]
- 11.de Rezende LFM, Lopes MR, Rey-López JP, Matsudo VKR, do Carmo Luiz O. Sedentary behavior and health outcomes: an overview of systematic reviews. PLoS One. 2014;9(8):e105620. doi: 10.1371/journal.pone.0105620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Shaw KA, Gennat HC, O’Rourke P, Del Mar C. Exercise for overweight or obesity. Cochrane Libr. 2006; [DOI] [PMC free article] [PubMed]
- 13.Ledoux T, Hingle M, Baranowski T. Relationship of fruit and vegetable intake with adiposity: a systematic review. Obes Rev. 2011;12(5):e143–e150. doi: 10.1111/j.1467-789X.2010.00786.x. [DOI] [PubMed] [Google Scholar]
- 14.Nago ES, Lachat CK, Dossa RA, Kolsteren PW. Association of out-of-home eating with anthropometric changes: a systematic review of prospective studies. Crit Rev Food Sci Nutr. 2014;54(9):1103–1116. doi: 10.1080/10408398.2011.627095. [DOI] [PubMed] [Google Scholar]
- 15.Mytton OT, Nnoaham K, Eyles H, Scarborough P, Mhurchu CN. Systematic review and meta-analysis of the effect of increased vegetable and fruit consumption on body weight and energy intake. BMC Public Health. 2014;14(1):886. doi: 10.1186/1471-2458-14-886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Thorp AA, Owen N, Neuhaus M, Dunstan DW. Sedentary behaviors and subsequent health outcomes in adults a systematic review of longitudinal studies, 1996–2011. Am J Prev Med. 2011;41(2):207–215. doi: 10.1016/j.amepre.2011.05.004. [DOI] [PubMed] [Google Scholar]
- 17.Rehm CD, Peñalvo JL, Afshin A, Mozaffarian D. Dietary intake among US adults, 1999–2012. JAMA. 2016;315(23):2542–2553. doi: 10.1001/jama.2016.7491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang DD, Leung CW, Li Y, Ding EL, Chiuve SE, Hu FB, Willett WC. Trends in dietary quality among adults in the United States, 1999 through 2010. JAMA Intern Med. 2014;174(10):1587–1595. doi: 10.1001/jamainternmed.2014.3422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ford ES, Dietz WH. Trends in energy intake among adults in the United States: findings from NHANES. Am J Clin Nutr. 2013;97(4):848–853. doi: 10.3945/ajcn.112.052662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ladabaum U, Mannalithara A, Myer PA, Singh G. Obesity, abdominal obesity, physical activity, and caloric intake in US adults: 1988 to 2010. Am J Med. 2014;127(8):717–727.e12. doi: 10.1016/j.amjmed.2014.02.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kelly PM, Davies A, Greig AJ, Lee KK. Obesity prevention in a city state: lessons from New York City during the Bloomberg Administration. Front Public Health. 2016;4:60. doi: 10.3389/fpubh.2016.00060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.New York City Department of Parks and Recreation. Shape Up NYC. Date Accessed: 1/29/2018. Available from: https://www.nycservice.org/initiatives/index.php?bitinitiative_id=15.
- 23.New York City Department of Health and Mental Hygiene. Health Bucks. Available from: https://www1.nyc.gov/site/doh/health/health-topics/health-bucks.page. Accessed 22 May 2018.
- 24.Thorpe LE, Greene C, Freeman A, Snell E, Rodriguez-Lopez JS, Frankel M, Punsalang A, Jr, Chernov C, Lurie E, Friedman M, Koppaka R, Perlman SE. Rationale, design and respondent characteristics of the 2013–2014 New York City Health and Nutrition Examination Survey (NYC HANES 2013–2014) Prev Med Rep. 2015;2:580–585. doi: 10.1016/j.pmedr.2015.06.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Thorpe LE, Gwynn RC, Mandel-Ricci J, Roberts S, Tsoi B, Berman L, Porter K, Ostchega Y, Curtain LR, Montaquila J, Mohadjer L, Frieden TR. Study design and participation rates of the New York City Health and Nutrition Examination Survey, 2004. Prev Chronic Dis. 2006;3(3):A94. [PMC free article] [PubMed] [Google Scholar]
- 26.Centers for Disease Control and Prevention (CDC). National Center for Health Statistics (NCHS). National Health and Nutrition Examination Survey Data. Hyattsville, MD: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. Available from: https://www.cdc.gov/nchs/nhanes/nhanes_questionnaires.htm. Accessed 2 Feb 2018.
- 27.Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies Lancet (London, England) 2004;363(9403):157–163. doi: 10.1016/S0140-6736(03)15268-3. [DOI] [PubMed] [Google Scholar]
- 28.National Health and Nutrition Examination Survey (NHANES): Physician Examination Procedures Manual. Available from: https://www.cdc.gov/nchs/data/nhanes/nhanes_13_14/Phys_Exam_Manual_2013.pdf. Accessed 2 Feb 2018.
- 29.Prevention OoD, Promotion H. Healthy People 2020. Washington, DC: US Department of Health and Human Services. 2013.
- 30.U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Healthy People 2010. Washington, DC Date Accessed: 7/27/2017. Available from: https://www.cdc.gov/nchs/healthy_people/hp2010.htm. [PubMed]
- 31.Budtz-Jorgensen E, Keiding N, Grandjean P, Weihe P. Confounder selection in environmental epidemiology: assessment of health effects of prenatal mercury exposure. Ann Epidemiol. 2007;17(1):27–35. doi: 10.1016/j.annepidem.2006.05.007. [DOI] [PubMed] [Google Scholar]
- 32.Shields M, Connor Gorber S, Tremblay MS. Estimates of obesity based on self-report versus direct measures. Health Rep. 2008;19(2):61–76. [PubMed] [Google Scholar]
- 33.Black JL, Macinko J. The changing distribution and determinants of obesity in the neighborhoods of New York City, 2003-2007. Am J Epidemiol. 2010;171(7):765–775. doi: 10.1093/aje/kwp458. [DOI] [PubMed] [Google Scholar]
- 34.Fakhouri TH, Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of obesity among older adults in the United States, 2007–2010. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics. Hyattsville, MD; 2012.
- 35.Fryar CD, Carroll MD, Ogden CL. Prevalence of overweight, obesity, and extreme obesity among adults: United States, trends 1960–1962 through 2009–2010. Hyattsville, MD: National Center for Health Statistics; 2012. [Google Scholar]
- 36.Seguin RA, Aggarwal A, Vermeylen F, Drewnowski A. Consumption frequency of foods away from home linked with higher body mass index and lower fruit and vegetable intake among adults: a cross-sectional study. J Environ Public Health. 2016;2016:1–12. doi: 10.1155/2016/3074241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Park Y, Neckerman KM, Quinn J, Weiss C, Rundle A. Place of birth, duration of residence, neighborhood immigrant composition and body mass index in New York City. Int J Behav Nutr Phys Act. 2008;5(1):19. doi: 10.1186/1479-5868-5-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Dee A, Kearns K, O'Neill C, Sharp L, Staines A, O'Dwyer V, et al. The direct and indirect costs of both overweight and obesity: a systematic review. BMC Res Notes. 2014;7:242. doi: 10.1186/1756-0500-7-242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Doucette ED, Salas J, Scherrer JF. Insurance coverage and diabetes quality indicators among patients in NHANES. Am J Manag Care. 2016;22(7):484–490. [PubMed] [Google Scholar]
- 40.Winkler MR, Mason S, Laska MN, Christoph MJ, Neumark-Sztainer D. Does non-standard work mean non-standard health? Exploring links between non-standard work schedules, health behavior, and well-being. SSM Popul Health. 2018;4:135–143. doi: 10.1016/j.ssmph.2017.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Wing RR, Phelan S. Long-term weight loss maintenance. Am J Clin Nutr. 2005;82(1):222S–225S. doi: 10.1093/ajcn/82.1.222S. [DOI] [PubMed] [Google Scholar]