YPMED-03880; No. of pages: 6; 4C: Preventive Medicine xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County Brenda Robles a, Stephanie Frost b, Lucas Moore c, Carole V. Harris b, Andrew S. Bradlyn b, Tony Kuo a,d,⁎ a
Division of Chronic Disease and Injury Prevention, Los Angeles County Department of Public Health, 3530 Wilshire Blvd, 8th Floor, Los Angeles, CA 90010, USA ICF International, 3 Corporate Square, Atlanta, GA 30329, USA College of Education and Human Services, West Virginia University, P.O. Box 9102, Morgantown, WV 26506-9102, USA d Department of Family Medicine, David Geffen School of Medicine at UCLA, 10880 Wilshire Blvd, Suite 1800, Los Angeles, CA 90024-4142, USA b c
a r t i c l e
i n f o
Available online xxxx Keywords: Obesity Population density Women's health Dietary behaviors Nutrition interventions
a b s t r a c t We described the prevalence of overweight and obesity among low-income women in rural West Virginia (WV) and urban Los Angeles County (LA County). Both communities participated in the national Communities Putting Prevention to Work program during 2010–2012. In each community, we completed health assessments on adult women recruited from public-sector clinics serving low-income populations. All participants answered survey questions regarding socio-demographics and diets. In both jurisdictions, we assessed obesity using objectively measured height and weight (calculated BMI). As part of each community case study, we performed multivariable regression analyses to describe the relationships between overweight and obesity and selected covariates (e.g., dietary behaviors). Overweight and obesity were prevalent among low-income women from WV (73%, combined) and LA County (67%, combined). In both communities, race and ethnicity appeared to predict the two conditions; however, the associations were not robust. In LA County, for example, African American and Hispanic women were 1.4 times (95% CI = 1.12, 1.81) more likely than white women to be overweight and obese. Collectively, these subpopulation health data served as an important guide for further planning of obesity prevention efforts in both communities. These efforts became a part of the subsequent Community Transformation Grants portfolio. © 2014 Elsevier Inc. All rights reserved.
Introduction The strain that overweight and obesity place on the nation's health and economy is well documented (Ogden et al., 2012; Wang and Beydoun, 2007). In response to the growing obesity epidemic, recent public health efforts in the U.S. have sought to reduce the obesity burden across various at-risk populations by addressing the physical and social determinants of health (Sallis et al., 2011; Story et al., 2008). The national Communities Putting Prevention to Work (CPPW)1 program recently invested more than $300 million in 50 communities to establish a myriad of system and environmental changes designed to reduce the prevalence of chronic diseases, including those caused by overweight and obesity (Bunnell et al., 2012). Nutrition interventions topped the list of practice-based strategies implemented by this
⁎ Corresponding author at: Division of Chronic Disease and Injury Prevention, Los Angeles County Department of Public Health, 3530 Wilshire Blvd, 8th Floor, Los Angeles, CA 90010, USA. Fax: +1 213 351 2713. E-mail addresses: [email protected] (B. Robles), stephanie.frost@icfi.com (S. Frost), [email protected] (L. Moore), charris5@icfi.com (C.V. Harris), abradlyn@icfi.com (A.S. Bradlyn), [email protected] (T. Kuo). 1 CPPW = Communities Putting Prevention to Work
program, including: institutional nutrition standards and sustainability guidelines for food procurement; retail food establishment practices that encouraged healthy eating; health marketing campaigns that educated the public about the harmful effects of excess calorie intake; and venue-specific health education aimed at empowering individuals to make better food choices (Table 1). In a number of CPPW communities, these interventions targeted low-income women and their families (e.g., spouses, children). Tailoring interventions for women and recruiting them as champions of change in their households are two public health approaches that are informed by prior research. Literature suggests that women frequently play the role of nutrition ‘gatekeepers’ for their households, influencing family eating behaviors (Charles and Kerr, 1988; Wild et al., 1994). Women also represent an important priority population, given that prior research has also shown that children from singleparent households are at increased risk of developing obesity and cardiovascular disease later in life (Huffman et al., 2010; PRB, 2011). Women themselves are a prime target group for intervention. Across age groups and by health status, they are at increased risk for overweight and obesity. Women of childbearing age, for example, are disproportionately affected by overweight and obesity, especially postpartum (Gore et al., 2003). In pregnancy, obese women are more
http://dx.doi.org/10.1016/j.ypmed.2014.02.016 0091-7435/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
2
West Virginia (rural)
Los Angeles County (urban)
Intervention category
Targeted Setting
Intervention
Targeted Setting
Institutional nutrition standards and sustainability guidelines for food procurement.
–
–
County of Los Angeles government.
–
–
Jurisdiction-wide.
Adoption of healthy convenience store practices to increase access to healthy foods in under resourced communities. Adoption of farmers market incentives to waive the cost of a food permit for all vendors. Increase the number of farmers markets and the availability of EBT machines. Dissemination of paid media advertisements targeting parents (i.e., “Where are the Veggies”) to promote healthy eating and to educate the public about the risk factors associated with unhealthy eating among children.
Food retail practices to encourage healthy eating.
Jurisdiction-wide.
Health marketing.
Jurisdiction-wide.
Venue-specific health education.
Select at-risk community settings. Select grocery stores located in at-risk communities. Select grocery stores located in at-risk communities. Select farmers markets located in at-risk communities.
Intervention
Board of Supervisors motion requiring Public Health review of new and renewing food service and vending contracts in all County of Los Angeles departments. The motion allowed each department the opportunity to incorporate nutrient limits and other Institute of Medicine recommendations for healthy meals, snacks, beverages, etc. in food service contracts. This motion affects County food venues, ranging from public hospitals and clinics to probation camps. Select low-income cities experiencing high rates of City-level resolutions requiring the adoption and obesity. implementation of nutrition standards and other healthy procurement practices across all city vending and food concessions. Select cities with high rates of obesity and low density Adoption of healthy corner store practices to increase of stores offering fresh fruits and vegetables. access to healthy foods in under-resourced communities. – –
Jurisdiction-wide.
– Brownbag “lunch and learn” lecture series to promote healthy eating among lecture attendees. Healthy checkout aisles and grocery stores signage to – increase awareness and promote healthy food options in the environment. Promotion of fresh fruits and vegetables via grocery – store “taste test” booths. Cooking demonstrations to highlight the use of fresh – fruits and vegetables at local farmers markets.
Dissemination of multi-pronged health marketing campaigns designed to promote healthy eating in the community. Communication methods employed included mass media approaches (e.g., billboards, public transit ads, videos, website) and social media channels (e.g., Twitter, Facebook). –
–
– –
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
Table 1 Nutrition interventions in two CPPW communities in the U.S., 2010–2012.
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
likely than their non-obese counterparts to develop gestational diabetes, experience medical complications from pre-eclampsia, require induced early labor, and undergo a cesarean section (Sebire et al., 2001). Among older women, poor health outcomes including early development of heart disease, stroke, and type 2 diabetes are seen for those who are obese (Dennis, 2007; Manson et al., 1990). While an extensive body of empirical evidence supports gender as a strong determinant of health (Krieger, 2003; Sen and Östlin, 2008), other determinants of obesity risk contribute to a more complex picture; the effects of these determinants are difficult to disentangle (Verbrugge, 1985). In health disparities research, obesity risk is often attributed to racial and ethnic differences (Cossrow and Falkner, 2004; Wang and Beydoun, 2007). However, socioeconomic factors and population density (rural, urban) also play important roles (Wang and Beydoun, 2007; Zhang and Wang, 2004). In the literature, unique differences in community resiliency, culture, and geography have been found to be associated with attenuated obesity risk, especially among particular subpopulations (Wang and Beydoun, 2007). Although studying complex causal pathways to disease development is of significant value to obesity research, public health practice often necessitates more applied science, requiring data that can enumerate specific subpopulation needs. At this more granular level, subpopulation health data can aid program planning and fieldwork by tailoring interventions to specifically address key geo-social factors that influence obesity risk (Frieden, 2010). Information on key attributes of targeted populations (e.g., subgroup obesity prevalence, health profiles and/or health behaviors) can be used to plan programs that address group- or culturally-specific covariates including food preparation style, social norms surrounding eating, etc. Such data provides validation of agency decisions to invest federal funds in obesity prevention. Unfortunately, for most communities, access to subpopulation health data is sparse. In this article, we contribute to public health practice by presenting two case studies of CPPW communities that collected subpopulation health data to document community needs. We specifically described the prevalence of overweight and obesity, and the health risk profiles of low-income women in a clinic setting in rural West Virginia (WV, Case-Community No. 1)2 and urban Los Angeles County (LA County, Case-Community No. 2).3 We chose these two specific communities because surveillance of obesity by population density (rural and urban) were key focus areas in the national CPPW program during 2010–2012.
3
Case-Community No. 2 In LA County, CPPW funded interventions for 9.8 million adults and children countywide. LA County is largely urban with a land area of 4058 square miles and a population density of 2419 persons per square mile. The population is racially and ethnically diverse. LA County is similar to WV in that its northwest and south-central regions have high rates of poverty and chronic disease (LACDPH, 2011; U.S. Census Bureau, 2012b). Data were collected from February–April, 2011 in five public health centers operated by the Los Angeles County Department of Public Health (LACDPH).5 These centers provide a range of services (e.g., immunizations, treatment of tuberculosis and sexually transmitted diseases, community programming, and other public/social services) to low-income residents. We selected them because they are located within the most impoverished areas of the county. Participant Recruitment for the Health Assessments WV participants (total n = 630; women with children ages 0–5 years, n = 553) were recruited from the waiting rooms of the two selected WIC clinics. To be eligible, they had to meet the following criteria: 1) demonstrated interest in the project; 2) be at least 18 years of age; 3) read and spoke English; 4) lived in one of six county jurisdictions in WV; 5) not be pregnant; 6) be at least eight weeks postpartum; and 7) agreed to return for follow-up visits (i.e., at three and six months post-initial encounter). All WIC clients had incomes that fell at or below 185% of the U.S. Poverty Income Guidelines (USDA, 2003). In LA County, low-income participants (total n = 720; women, n = 408) were recruited from the waiting rooms of five large public health centers using a systematic approach to selection, accounting (when feasible) for each center's clientele volume, time of day, variation in the types of services provided, and variation in clinic flow on the specified recruitment days. Trained staff utilized multi-stage, systematic procedures on pre-specified days of the survey period to recruit and enroll eligible participants. To be eligible, LA County participants had to meet the following criteria: 1) be at least 18 years of age; 2) spoke English or Spanish; 3) be a client (patient) of the health center; 4) not be pregnant; and 5) agreed to complete a battery of anthropometric and self-administered assessments on a scheduled weekend day at a designated center location. Standardized recruitment and measurement protocols were used in both communities. Differences in eligibility criteria between the two health assessments were based on regional demographic differences (e.g., in LA County there was a larger population of Spanish speaking adults). Written informed consents were obtained from all participants in each community. All assessment protocols and materials were reviewed and approved by each jurisdiction's respective Institutional Review Boards. Data collection in each case community
Methods We analyzed cross-sectional data from health assessments conducted during the first 15 months of the national CPPW program in rural WV and urban LA County. Both communities participated in several local CPPW interventions and enhanced evaluation activities, including interval assessments of body mass index (BMI) and self-reported dietary behaviors of low-income community-dwelling adults.
Settings Case-Community No. 1 In WV, CPPW funded interventions in a six-county area. This region is largely rural (U.S. Census Bureau, 2012a; USDA, 2003) with a land area of 1944 square miles and a population density of 68 persons per square mile. Predominantly white, the area is characterized by high rates of unemployment, poverty, and chronic disease. Data collection took place from February-August, 2011, in two Women, Infants, and Children (WIC)4 clinics (USDA, 2011). These two sites were selected because they served the largest proportion of low-income residents in the region. 2 3 4
WV = West Virginia LA County = Los Angeles County WIC = Women, Infants, and Children (program/clinic)
Trained staff collected anthropometric measurements and employed standard procedures for administering participant surveys. In WV, height and weight measurements were measured twice using calibrated Health-O-Meter 50KL scales with built-in height rods (Jarden Corporation, Rye, NY). In LA County, height and weight measurements were collected at least two times using a stadiometer (Seca 213, seca Precision for health, United Kingdom) and a digital scale (Seca 876, seca Precision for health, United Kingdom), respectively. The final recorded heights and weights represented the average of repeated measurements. In both communities, demographic information, and information on dietary behaviors, was collected using self-administered surveys. In WV, an eight-page English-only paper questionnaire was developed and administered (an online version was also available). The dietary behavior module of the instrument was adapted from the University of California, Davis Food Behavior Checklist (used with permission). In LA County, a seven-page paper questionnaire was developed and administered in English or Spanish; the instrument was developed using previously validated question items from national as well as local population health surveys, including the National Health and Nutrition Examination Survey (NHANES)6 (NCHS, 2011) and the Los Angeles County Health Survey (LACDPH, 2011). The Spanish version was translated from the English version using standardized forward–backward language translation protocols. In contrast to WV, a Spanish version of the questionnaire in LA County was 5 6
LACDPH = Los Angeles County Department of Public Health NHANES = National Health and Nutrition Examination Survey
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
4
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
Table 2 Socio-demographic characteristics of women who participated in the health assessments in West Virginia and Los Angeles County, 2011. Characteristics
Age (years)a 18–24 25–44 45–64 65+ Race/ethnicitya African American/Black Hispanic White Asian/Pacific Islander Other Educationa Less than high school High school graduate or GED Some college or college degree Employmenta Unemployed Retired Disabled Employed a b c
West Virginiab (rural) n (%)
Los Angeles Countyc (urban) n (%)
553
408
181 (32.8) 321(58.0) 48 (8.7) 1 (b1.0)
83 (20.3) 196 (48.0) 117 (28.7) 12 (2.9)
9 (1.6) 10 (1.8) 525 (94.9) 4 (b1.0) 3 (b1.0)
171 (42.0) 130 (31.9) 37 (9.1) 44 (10.8) 25 (6.1)
55 (9.9) 261 (47.2) 231 (41.8)
71 (17.6) 76 (18.6) 256 (62.7)
276 (49.9) 3 (b1.0) 39 (7.1) 129 (23.3)
162 (39.7) 12 (3.0) 29 (7.2) 198 (48.5)
Percentages may not add up to 100% because of missing data and rounding. Recruited from Women, Infants, and Children (WIC) clinics in West Virginia. Recruited from ‘safety net’, multi-purpose public health centers in Los Angeles County.
developed because a large proportion of the LA County population is of Hispanic origin and speaks Spanish.
beverages and fruits and vegetables), prevalence rate ratios (PRRs) and 95% confidence intervals (CIs) were generated for each community, using robust Poisson regression methods (Deddens and Peterson, 2008). BMI was the outcome variable of interest used in the multivariable models, where overweight (BMI ≥25 and BMI ≤29.9) and obesity (BMI ≥ 30) were collapsed. All data analyses were conducted using Stata/SE 12.1 (StataCorp LP, College Station, Texas, USA).
Results Of the 2092 parents approached in the WIC clinics, 33% refused and 30% were enrolled by the WV trained staff (total n = 630; women, n = 553). Of the 1393 patients approached in the designated public health centers, 26% refused and 74% were enrolled by the LA County trained staff (total n = 720; women, n = 408). Compared to women in LA County, WV participants were generally younger (Table 2). Women in the WV sample were predominately white (95%), whereas women in the LA County sample were predominately African American and Hispanic (74%, combined). Of the WV women, 73% were overweight and obese, as compared to 67% among LA County women (Fig. 1). In general, women in the LA County sample were more educated than women in the WV sample (63% versus 42%). They also reported consuming less soda (28% versus 37%) but more sugary drink alternatives (41% versus 32%) than their counterparts in WV. In both communities, race and ethnicity appeared to predict overweight and obesity; the associations to covariates, however, were not robust. In LA County, for instance, African American and Hispanic women were 1.4 times (95% CI = 1.12, 1.81) more likely than white women to be overweight and obese (Table 3). Discussion
Data management and statistical analysis For each community, common dietary behavior variables were identified. Due to sample variations and differences in some of the variable response categories, common categorical anchors were generated for key variables in each of the datasets from WV and LA County. For example, using Centers for Disease Control and Prevention (CDC)7 guidelines, both communities converted objectively measured heights and weights to a standard indicator — BMI (weight [kg] / height squared [m2]) (CDC, 2012), with BMI b 24.9, normal or nonobese; 25.0-29.9, overweight; ≥ 30.0, obese. We performed descriptive analyses to describe frequencies and differences in participant characteristics (e.g., demographic characteristics, eating behaviors) by community. To explore the relationships between the prevalence of overweight and obesity and covariates such as race/ethnicity, age, education and dietary behaviors (consumption frequency of sodas/sugar-sweetened
Fig. 1. Prevalence of overweight and obesity by population density (rural and urban).
The present case examples by population density (rural WV and urban LA County) highlight the burden of overweight and obesity among low-income women in two communities supported by CPPW during 2010–2012. Although the health assessment methods and data collection protocols differed somewhat from one another, both communities showed impressive magnitudes of obesity prevalence in this subpopulation, suggesting that federal investments in obesity prevention for these geographic regions were relatively well-aligned with the needs of these communities. Closer examination of each case example suggests that this burden may be greater than it appears in each setting. For example, we found obesity rates among LA County women to exceed 50%; this contrasts county-wide estimates of 30% for this same gender group (LACDPH-OWH, 2013). Similarly, when comparing health behaviors, approximately 27% of women in LA County reported consuming one soda or sugar-sweetened beverage per day whereas in the overall county population, this self-reported behavior was closer to 35% (LACDPH, 2011). Findings from our case studies aligned with those found in the literature, including: 1) low socioeconomic status is strongly associated with a variety of risk factors (e.g., diet) for obesity and obesity-related chronic diseases (Larson et al., 2009; Zhang and Wang, 2004); 2) lowincome individuals are less likely to consume nutritious foods (Lynch et al., 2004) and more likely to consume calorie-dense foods such as soda, sugar-sweetened beverages, and other processed foods (Cohen et al., 2010); and 3) fruit and vegetable consumption, a proxy for healthy eating, is disproportionately lower among low-income subgroups (Drewnowski, 2009). In LA County, African American and Hispanic women were more likely than white women to be overweight or obese. This observation, however, may be due to the higher representation of African Americans in the LA County sample. In contrast to recent U.S. Census estimates — African Americans accounted only for approximately 9% of the total county population (U.S. Census Bureau, 2012b), African Americans represented 42% of the LA case study sample. In WV, racial
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
5
Table 3 Overweight and obesity among low-income rural and urban women by socio-demographic characteristics and dietary behaviors, 2011a.
Race White African American or Hispanic Age 18–24 N25 Education Some college or college degree High school graduate or less Reported frequency of sugar-sweetened beverage consumption Does not drink/drinks sometimes Drinks often/everyday Reported frequency of soda consumption Does not drink/drinks sometimes Drinks often/everyday Average reported consumption of vegetables each day Four or more servings 3 servings or less Average reported consumption of fruits each day Four or more servings 3 servings or less
West Virginia (rural)
Los Angeles County (urban)
Adj. PRR (95% CI)b
Adj. PRR (95% CI)b
Referent 1.18 (1.00, 1.40)
Referent 1.42 (1.12, 1.81)
Referent 1.18 (1.05, 1.33)
Referent 1.44 (1.19, 1.74)
Referent 1.04 (0.94, 1.15)
Referent 0.93 (0.84, 1.03)
Referent 0.92 (0.82, 1.03)
Referent 0.95 (0.84, 1.07)
Referent 0.99 (0.89, 1.09)
Referent 1.04 (0.92, 1.18)
Referent 0.95 (0.83, 1.10)
Referent 1.13 (0.97, 1.33)
Referent 0.95 (0.80, 1.12)
Referent 0.82 (0.70, 0.97)
a Model: overweight and obese body mass index stratified by community, adjusted for other covariates, including socio-demographic characteristics (race, age, education) and dietary behaviors (e.g., soda and sugar-sweetened beverage consumption, fruits and vegetable intake); robust Poisson regression analyses were performed to estimate the adjusted prevalence rate ratios. b Adj. PRR = adjusted prevalence rate ratio, CI = confidence intervals.
differences were difficult to assess because more than 90% of health assessment participants were white. Although case studies provide important insights into regional differences in overweight and obesity — WV (rural) versus LA County (urban), inferences about the root causes of these regional disparities cannot be fully explained given the dissimilar methods used to collect the data. While it is possible that such factors as sparse open space, unsafe neighborhoods, an inefficient public transit system, limited access to grocery stores, and non-competitive food pricing (CPSTF, 2011; French et al., 2001; Larson et al., 2009; Moore et al., 2008; NPC, 2011) may all present important challenges to healthy eating and active living in both communities, the magnitude of how these factors differentially impact overweight and obesity prevalence across the two regions remain unclear and warrant further study. Unique regional preferences for soda and customs in preparing food, for example, may have differential impact on overweight and obesity prevalence across the various subgroups in each jurisdiction. Barriers to healthy eating (e.g., access to fresh fruits and vegetables) that are thought to be similar may actually be dissimilar, as the solutions to the obesity epidemic in each community may be different. Whereas capital investments in grocery stores or places that sell fresh fruits and vegetables (e.g., farmers market) are likely important for mitigating shortages of food venues in WV, conversion of existing corner stores (abundant in the neighborhood) or safer and easier access to public transportation to go to fartheraway locations may be more suitable for LA County. Further research is needed to examine these factors, as they are not the focus of these case study examples. Limitations The case study approach utilized in this article has several limitations. First, comparisons of WV and LA County samples could not be made, as different recruitment approaches and clientele characteristics precluded dataset comparability. Second, key differences in the two clinic populations' age, education, and the services sought by clients likely contributed to some selection bias in each community. Third, socioeconomic status was not easily established for both samples, as the two regional assessment instruments (surveys) did not directly ask
about participant income. Other sources of information were used to establish low socioeconomic status in WV and LA County. In WV, to receive services, all WIC clients must have incomes which fell at or below 185% of the U.S. Poverty Income Guidelines. In LA County, participants provided zip codes to verify their region of residence and answered questions about employment status, education, and usage of needbased public services. Conclusions The present case studies of rural WV and urban LA County represent unique snapshots of subpopulations targeted by the national CPPW program administered by the CDC (Bunnell et al., 2012). Results of the studies confirmed the need to invest in these regions, which contained high prevalence of overweight and obesity. Coupled to other system-level or multi-sector interventions, the range of nutrition interventions in WV and LA County (e.g., WIC health education; workplace breastfeeding accommodations; healthy food procurement practices; and public education) offer potentially meaningful opportunities to facilitate better food selections among low-income women and their families. These data provide invaluable insights on how these and other obesity prevention strategies can be tailored and refined to address the needs of this important segment of the population — a group that can have an enormous impact not only on what food they choose for themselves, but, more importantly, for their families. Collectively, these subpopulation health data served as an important guide for further planning of obesity prevention efforts in both communities; in many cases, these efforts became a part of the subsequent Community Transformation Grants portfolio. Conflict of interest statement The authors report no financial disclosures or conflicts of interest.
Acknowledgments The authors would like to thank the staff in the following agencies and organizations for their support and contributions to this article: CPPW-West Virginia (Principal Investigator Joe Barker); the West
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
6
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
Virginia Bureau for Public Health and the Mid-Ohio Valley Health Department; Los Angeles County Department of Public Health (Lisa V. Smith, Jennifer Piron, and Mirna Ponce); RTI International (Allie Lieberman and Jonathan Blitstein); and the CPPW Manuscript Writing Workshop sponsored by ICF International (Kathleen Whitten, Tesfayi Gebreselassie). The project was supported in part by cooperative agreements from the Centers for Disease Control and Prevention (#3U58DP002429-01S1, West Virginia and #3U58DP002485-01S1, Los Angeles County). The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the Los Angeles County Department of Public Health, the Centers for Disease Control and Prevention, or any other organization mentioned in the text. In accordance with U.S. law, no federal funds provided by CDC were permitted to be used by community grantees for lobbying or to influence, directly or indirectly, specific pieces of pending or proposed legislation at the federal, state, or local levels. As it relates to the CDCsponsored supplement, staff training and reviews by scientific writers were provided as technical assistance to the authors, through a contract with ICF International (Contract No. 200-2007-22643-003). CDC staff has reviewed the project's evaluation design and data collection methodology, and the article for scientific accuracy. All authors have read and approved the final version.
References Bunnell, R., O'Neil, D., Soler, R., et al., 2012. Fifty Communities Putting Prevention to Work: accelerating chronic disease prevention through policy, systems, and environmental change. J. Community Health 37 (5), 1081–1090. Center for Disease Control Prevention, 2012. Defining overweight and obesity. Available: http://www.cdc.gov/obesity/adult/defining.html (Last Accessed February 11, 2014). Charles, N., Kerr, M., 1988. Women, Food and Families. United Kingdom, Manchester University Press, Manchester. Cohen, D.A., Sturm, R., Scott, M., Farley, T.A., Bluthenthal, R., 2010. Not enough fruit and vegetables or too many cookies, candies, salty snacks, and soft drinks? Public Health Rep. 125 (1), 88–95. Community Preventive Services Task Force (CPSTF), 2011. The Guide to Community Preventive Services: What Works to Promote Health (The Community Guide). Available: http://www.thecommunityguide.org/library/ARC2011/congress-report-full.pdf (Last Accessed July 24, 2012). Cossrow, N., Falkner, B., 2004. Race/ethnic issues in obesity and obesity-related comorbidities. J. Clin. Endocrinol. Metab. 89 (6), 2590–2594. Deddens, J.A., Peterson, M.R., 2008. Approaches for estimating prevalence ratios. Occup. Environ. Med. 65, 501–506. http://dx.doi.org/10.1136/oem.2007.034777. Dennis, K.E., 2007. Postmenopausal women and the health consequences of obesity. J. Obstet. Gynecol. Neonatal. Nurs. 36 (5), 511–517. Drewnowski, A., 2009. Obesity, diets, and social inequalities. Nutr. Rev. 67 (1), S36–S39. French, S.A., Story, M., Jeffery, R.W., 2001. Environmental influences on eating and physical activity. Annu. Rev. Public Health 22, 309–335. Frieden, T.R., 2010. A framework for public health action: the health impact pyramid. Am. J. Public Health 100 (4), 590–595. http://dx.doi.org/10.2105/AJPH.2009.185652 (Epub 2010 Feb 18). Gore, S.A., Brown, D.M., West, D.S., 2003. The role of postpartum weight retention in obesity among women: a review of the evidence. Ann. Behav. Med. 26 (2), 149–159. Huffman, F.G., Kanikireddy, S., Patel, M., 2010. Parenthood—a contributing factor to childhood obesity. Int. J. Environ. Res. Public Health 7 (7), 2800–2810.
Krieger, N., 2003. Genders, sexes, and health: what are the connections—and why does it matter? Int. J. Epidemiol. 32 (4), 652–657. Larson, N.I., Story, M.T., Nelson, M.C., 2009. Neighborhood environments: disparities in access to healthy foods in the US. Am. J. Prev. Med. 36, 74–81. Los Angeles County Department of Public Health (LACDPH), 2011. Los Angeles County Health Survey. Available: http://publichealth.lacounty.gov/ha/hasurveyintro.htm (Last Accessed July 25, 2012). Los Angeles County Department of Public Health, Office of Women's Health (LACDPHOWH), Health Indicators for Women in Los Angeles County: Highlighting Disparities by Ethnicity and Poverty Level, June 2013. Available: http://www. publichealth.lacounty.gov/ha/docs/2011LACHS/WHI2013/WHI_2013s.pdf (Last Accessed February 11, 2014). Lynch, J., Smith, G.D., Harper, S., et al., 2004. Is income inequality a determinant of population health? Part 1. A systematic review. Milbank Q. 82 (1), 5–99. Manson, J.E., Colditz, G.A., Stampfer, M.J., et al., 1990. A prospective study of obesity and risk of coronary heart disease in women. N. Engl. J. Med. 322 (13), 882–889. Moore, L.V., Diez-Roux, A.V., Nettleton, J.A., Jacobs, D.R., 2008. Associations of the local food environment with diet quality—a comparison of assessments based on surveys and geographic information systems: the multi-ethnic study of atherosclerosis. Am. J. Epidemiol. 167, 917–924. National Center for Health Statistics (NCHS), 2011. National Health and Nutrition Examination Survey. Available: http://www.cdc.gov/nchs/nhanes/about_nhanes. htm (Last Accessed July 24, 2012). National Prevention Council (NPC), 2011. National Prevention Strategy. U.S. Department of Health and Human Services, Office of the Surgeon General, Washington, D.C. Ogden, C.L., Carroll, M.D., Kit, B.K., Flegal, K., 2012. Prevalence of obesity in the United States, 2009–2010. NCHS Data Brief, no 82. National Center for Health Statistics, Hyattsville, MD. Population Reference Bureau (PRB), 2011. U.S. Children In Single-Mother Families. Available: http://www.prb.org/Publications/PolicyBriefs/singlemotherfamilies.aspx (Last Accessed July 25, 2012). Sallis, J.F., Millstein, R.A., Carlson, J.A., 2011. Community design for physical activity. In: Dannenberg, A., Frumkin, H., Jackson, R. (Eds.), Making Healthy Places: Designing and Building for Health, Well-being, and Sustainability. Island Press, Washington D.C., pp. 33–49. Sebire, N.J., Jolly, M., Harris, J.P., et al., 2001. Maternal obesity and pregnancy outcome: a study of 287,213 pregnancies in London. Int. J. Obes. Relat. Metab. Disord. 25 (8), 1175–1182. Sen, G., Östlin, P., 2008. Gender inequity in health: why it exists and how we can change it. Glob. Public Health 3 (Suppl. 1), 1–12. Story, M., Kaphingst, K.M., Robinson-O'Brien, R., Glanz, K., 2008. Creating healthy food and eating environments: policy and environmental approaches. Annu. Rev. Public Health 29, 253–272. U.S. Census Bureau (QF-L), 2012b. Quick Facts: Los Angeles County, California. Available: http://quickfacts.census.gov/qfd/states/06/06037.html (Last Accessed March 12, 2012). U.S. Census Bureau (QF-WV), 2012a. Quick Facts: West Virginia. Calhoun, Pleasants, Ritchie, Roane, Wirt and Wood Counties. Available: http://quickfacts.census.gov/ qfd/states/54000.html (Last Accessed March 17, 2012). United States Department of Agriculture (USDA), 2003. Rural Urban Continuum Codes. Available: http://www.ers.usda.gov/data-products/rural-urban-continuum-codes. aspx (Last Accessed April 20, 2012). United States Department of Agriculture (USDA), 2011. Women, Infants and Children (WIC) Factsheet. Available: http://www.fns.usda.gov/wic/WIC-Fact-Sheet.pdf (Last Accessed July 25, 2012). Verbrugge, L.M., 1985. Gender and health: an update on hypotheses and evidence. J. Health Soc. Behav. 26 (3), 156–182. Wang, Y., Beydoun, M.A., 2007. The obesity epidemic in the United States—gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and meta-regression analysis. Epidemiol. Rev. 29, 6–28. Wild, R.A., Taylor, E.L., Knehans, A., Cleaver, V., 1994. Matriarchal model for cardiovascular prevention. Obstet. Gynecol. Surv. 49 (2), 147–152. Zhang, Q., Wang, Y., 2004. Trends in the association between obesity and socioeconomic status in U.S. adults: 1971 to 2000. Obesity Res. 12 (10), 1622–1632.
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
Contents lists available at ScienceDirect
Preventive Medicine journal homepage: www.elsevier.com/locate/ypmed
Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County Brenda Robles a, Stephanie Frost b, Lucas Moore c, Carole V. Harris b, Andrew S. Bradlyn b, Tony Kuo a,d,⁎ a
Division of Chronic Disease and Injury Prevention, Los Angeles County Department of Public Health, 3530 Wilshire Blvd, 8th Floor, Los Angeles, CA 90010, USA ICF International, 3 Corporate Square, Atlanta, GA 30329, USA College of Education and Human Services, West Virginia University, P.O. Box 9102, Morgantown, WV 26506-9102, USA d Department of Family Medicine, David Geffen School of Medicine at UCLA, 10880 Wilshire Blvd, Suite 1800, Los Angeles, CA 90024-4142, USA b c
a r t i c l e
i n f o
Available online xxxx Keywords: Obesity Population density Women's health Dietary behaviors Nutrition interventions
a b s t r a c t We described the prevalence of overweight and obesity among low-income women in rural West Virginia (WV) and urban Los Angeles County (LA County). Both communities participated in the national Communities Putting Prevention to Work program during 2010–2012. In each community, we completed health assessments on adult women recruited from public-sector clinics serving low-income populations. All participants answered survey questions regarding socio-demographics and diets. In both jurisdictions, we assessed obesity using objectively measured height and weight (calculated BMI). As part of each community case study, we performed multivariable regression analyses to describe the relationships between overweight and obesity and selected covariates (e.g., dietary behaviors). Overweight and obesity were prevalent among low-income women from WV (73%, combined) and LA County (67%, combined). In both communities, race and ethnicity appeared to predict the two conditions; however, the associations were not robust. In LA County, for example, African American and Hispanic women were 1.4 times (95% CI = 1.12, 1.81) more likely than white women to be overweight and obese. Collectively, these subpopulation health data served as an important guide for further planning of obesity prevention efforts in both communities. These efforts became a part of the subsequent Community Transformation Grants portfolio. © 2014 Elsevier Inc. All rights reserved.
Introduction The strain that overweight and obesity place on the nation's health and economy is well documented (Ogden et al., 2012; Wang and Beydoun, 2007). In response to the growing obesity epidemic, recent public health efforts in the U.S. have sought to reduce the obesity burden across various at-risk populations by addressing the physical and social determinants of health (Sallis et al., 2011; Story et al., 2008). The national Communities Putting Prevention to Work (CPPW)1 program recently invested more than $300 million in 50 communities to establish a myriad of system and environmental changes designed to reduce the prevalence of chronic diseases, including those caused by overweight and obesity (Bunnell et al., 2012). Nutrition interventions topped the list of practice-based strategies implemented by this
⁎ Corresponding author at: Division of Chronic Disease and Injury Prevention, Los Angeles County Department of Public Health, 3530 Wilshire Blvd, 8th Floor, Los Angeles, CA 90010, USA. Fax: +1 213 351 2713. E-mail addresses: [email protected] (B. Robles), stephanie.frost@icfi.com (S. Frost), [email protected] (L. Moore), charris5@icfi.com (C.V. Harris), abradlyn@icfi.com (A.S. Bradlyn), [email protected] (T. Kuo). 1 CPPW = Communities Putting Prevention to Work
program, including: institutional nutrition standards and sustainability guidelines for food procurement; retail food establishment practices that encouraged healthy eating; health marketing campaigns that educated the public about the harmful effects of excess calorie intake; and venue-specific health education aimed at empowering individuals to make better food choices (Table 1). In a number of CPPW communities, these interventions targeted low-income women and their families (e.g., spouses, children). Tailoring interventions for women and recruiting them as champions of change in their households are two public health approaches that are informed by prior research. Literature suggests that women frequently play the role of nutrition ‘gatekeepers’ for their households, influencing family eating behaviors (Charles and Kerr, 1988; Wild et al., 1994). Women also represent an important priority population, given that prior research has also shown that children from singleparent households are at increased risk of developing obesity and cardiovascular disease later in life (Huffman et al., 2010; PRB, 2011). Women themselves are a prime target group for intervention. Across age groups and by health status, they are at increased risk for overweight and obesity. Women of childbearing age, for example, are disproportionately affected by overweight and obesity, especially postpartum (Gore et al., 2003). In pregnancy, obese women are more
http://dx.doi.org/10.1016/j.ypmed.2014.02.016 0091-7435/© 2014 Elsevier Inc. All rights reserved.
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
2
West Virginia (rural)
Los Angeles County (urban)
Intervention category
Targeted Setting
Intervention
Targeted Setting
Institutional nutrition standards and sustainability guidelines for food procurement.
–
–
County of Los Angeles government.
–
–
Jurisdiction-wide.
Adoption of healthy convenience store practices to increase access to healthy foods in under resourced communities. Adoption of farmers market incentives to waive the cost of a food permit for all vendors. Increase the number of farmers markets and the availability of EBT machines. Dissemination of paid media advertisements targeting parents (i.e., “Where are the Veggies”) to promote healthy eating and to educate the public about the risk factors associated with unhealthy eating among children.
Food retail practices to encourage healthy eating.
Jurisdiction-wide.
Health marketing.
Jurisdiction-wide.
Venue-specific health education.
Select at-risk community settings. Select grocery stores located in at-risk communities. Select grocery stores located in at-risk communities. Select farmers markets located in at-risk communities.
Intervention
Board of Supervisors motion requiring Public Health review of new and renewing food service and vending contracts in all County of Los Angeles departments. The motion allowed each department the opportunity to incorporate nutrient limits and other Institute of Medicine recommendations for healthy meals, snacks, beverages, etc. in food service contracts. This motion affects County food venues, ranging from public hospitals and clinics to probation camps. Select low-income cities experiencing high rates of City-level resolutions requiring the adoption and obesity. implementation of nutrition standards and other healthy procurement practices across all city vending and food concessions. Select cities with high rates of obesity and low density Adoption of healthy corner store practices to increase of stores offering fresh fruits and vegetables. access to healthy foods in under-resourced communities. – –
Jurisdiction-wide.
– Brownbag “lunch and learn” lecture series to promote healthy eating among lecture attendees. Healthy checkout aisles and grocery stores signage to – increase awareness and promote healthy food options in the environment. Promotion of fresh fruits and vegetables via grocery – store “taste test” booths. Cooking demonstrations to highlight the use of fresh – fruits and vegetables at local farmers markets.
Dissemination of multi-pronged health marketing campaigns designed to promote healthy eating in the community. Communication methods employed included mass media approaches (e.g., billboards, public transit ads, videos, website) and social media channels (e.g., Twitter, Facebook). –
–
– –
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
Table 1 Nutrition interventions in two CPPW communities in the U.S., 2010–2012.
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
likely than their non-obese counterparts to develop gestational diabetes, experience medical complications from pre-eclampsia, require induced early labor, and undergo a cesarean section (Sebire et al., 2001). Among older women, poor health outcomes including early development of heart disease, stroke, and type 2 diabetes are seen for those who are obese (Dennis, 2007; Manson et al., 1990). While an extensive body of empirical evidence supports gender as a strong determinant of health (Krieger, 2003; Sen and Östlin, 2008), other determinants of obesity risk contribute to a more complex picture; the effects of these determinants are difficult to disentangle (Verbrugge, 1985). In health disparities research, obesity risk is often attributed to racial and ethnic differences (Cossrow and Falkner, 2004; Wang and Beydoun, 2007). However, socioeconomic factors and population density (rural, urban) also play important roles (Wang and Beydoun, 2007; Zhang and Wang, 2004). In the literature, unique differences in community resiliency, culture, and geography have been found to be associated with attenuated obesity risk, especially among particular subpopulations (Wang and Beydoun, 2007). Although studying complex causal pathways to disease development is of significant value to obesity research, public health practice often necessitates more applied science, requiring data that can enumerate specific subpopulation needs. At this more granular level, subpopulation health data can aid program planning and fieldwork by tailoring interventions to specifically address key geo-social factors that influence obesity risk (Frieden, 2010). Information on key attributes of targeted populations (e.g., subgroup obesity prevalence, health profiles and/or health behaviors) can be used to plan programs that address group- or culturally-specific covariates including food preparation style, social norms surrounding eating, etc. Such data provides validation of agency decisions to invest federal funds in obesity prevention. Unfortunately, for most communities, access to subpopulation health data is sparse. In this article, we contribute to public health practice by presenting two case studies of CPPW communities that collected subpopulation health data to document community needs. We specifically described the prevalence of overweight and obesity, and the health risk profiles of low-income women in a clinic setting in rural West Virginia (WV, Case-Community No. 1)2 and urban Los Angeles County (LA County, Case-Community No. 2).3 We chose these two specific communities because surveillance of obesity by population density (rural and urban) were key focus areas in the national CPPW program during 2010–2012.
3
Case-Community No. 2 In LA County, CPPW funded interventions for 9.8 million adults and children countywide. LA County is largely urban with a land area of 4058 square miles and a population density of 2419 persons per square mile. The population is racially and ethnically diverse. LA County is similar to WV in that its northwest and south-central regions have high rates of poverty and chronic disease (LACDPH, 2011; U.S. Census Bureau, 2012b). Data were collected from February–April, 2011 in five public health centers operated by the Los Angeles County Department of Public Health (LACDPH).5 These centers provide a range of services (e.g., immunizations, treatment of tuberculosis and sexually transmitted diseases, community programming, and other public/social services) to low-income residents. We selected them because they are located within the most impoverished areas of the county. Participant Recruitment for the Health Assessments WV participants (total n = 630; women with children ages 0–5 years, n = 553) were recruited from the waiting rooms of the two selected WIC clinics. To be eligible, they had to meet the following criteria: 1) demonstrated interest in the project; 2) be at least 18 years of age; 3) read and spoke English; 4) lived in one of six county jurisdictions in WV; 5) not be pregnant; 6) be at least eight weeks postpartum; and 7) agreed to return for follow-up visits (i.e., at three and six months post-initial encounter). All WIC clients had incomes that fell at or below 185% of the U.S. Poverty Income Guidelines (USDA, 2003). In LA County, low-income participants (total n = 720; women, n = 408) were recruited from the waiting rooms of five large public health centers using a systematic approach to selection, accounting (when feasible) for each center's clientele volume, time of day, variation in the types of services provided, and variation in clinic flow on the specified recruitment days. Trained staff utilized multi-stage, systematic procedures on pre-specified days of the survey period to recruit and enroll eligible participants. To be eligible, LA County participants had to meet the following criteria: 1) be at least 18 years of age; 2) spoke English or Spanish; 3) be a client (patient) of the health center; 4) not be pregnant; and 5) agreed to complete a battery of anthropometric and self-administered assessments on a scheduled weekend day at a designated center location. Standardized recruitment and measurement protocols were used in both communities. Differences in eligibility criteria between the two health assessments were based on regional demographic differences (e.g., in LA County there was a larger population of Spanish speaking adults). Written informed consents were obtained from all participants in each community. All assessment protocols and materials were reviewed and approved by each jurisdiction's respective Institutional Review Boards. Data collection in each case community
Methods We analyzed cross-sectional data from health assessments conducted during the first 15 months of the national CPPW program in rural WV and urban LA County. Both communities participated in several local CPPW interventions and enhanced evaluation activities, including interval assessments of body mass index (BMI) and self-reported dietary behaviors of low-income community-dwelling adults.
Settings Case-Community No. 1 In WV, CPPW funded interventions in a six-county area. This region is largely rural (U.S. Census Bureau, 2012a; USDA, 2003) with a land area of 1944 square miles and a population density of 68 persons per square mile. Predominantly white, the area is characterized by high rates of unemployment, poverty, and chronic disease. Data collection took place from February-August, 2011, in two Women, Infants, and Children (WIC)4 clinics (USDA, 2011). These two sites were selected because they served the largest proportion of low-income residents in the region. 2 3 4
WV = West Virginia LA County = Los Angeles County WIC = Women, Infants, and Children (program/clinic)
Trained staff collected anthropometric measurements and employed standard procedures for administering participant surveys. In WV, height and weight measurements were measured twice using calibrated Health-O-Meter 50KL scales with built-in height rods (Jarden Corporation, Rye, NY). In LA County, height and weight measurements were collected at least two times using a stadiometer (Seca 213, seca Precision for health, United Kingdom) and a digital scale (Seca 876, seca Precision for health, United Kingdom), respectively. The final recorded heights and weights represented the average of repeated measurements. In both communities, demographic information, and information on dietary behaviors, was collected using self-administered surveys. In WV, an eight-page English-only paper questionnaire was developed and administered (an online version was also available). The dietary behavior module of the instrument was adapted from the University of California, Davis Food Behavior Checklist (used with permission). In LA County, a seven-page paper questionnaire was developed and administered in English or Spanish; the instrument was developed using previously validated question items from national as well as local population health surveys, including the National Health and Nutrition Examination Survey (NHANES)6 (NCHS, 2011) and the Los Angeles County Health Survey (LACDPH, 2011). The Spanish version was translated from the English version using standardized forward–backward language translation protocols. In contrast to WV, a Spanish version of the questionnaire in LA County was 5 6
LACDPH = Los Angeles County Department of Public Health NHANES = National Health and Nutrition Examination Survey
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
4
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
Table 2 Socio-demographic characteristics of women who participated in the health assessments in West Virginia and Los Angeles County, 2011. Characteristics
Age (years)a 18–24 25–44 45–64 65+ Race/ethnicitya African American/Black Hispanic White Asian/Pacific Islander Other Educationa Less than high school High school graduate or GED Some college or college degree Employmenta Unemployed Retired Disabled Employed a b c
West Virginiab (rural) n (%)
Los Angeles Countyc (urban) n (%)
553
408
181 (32.8) 321(58.0) 48 (8.7) 1 (b1.0)
83 (20.3) 196 (48.0) 117 (28.7) 12 (2.9)
9 (1.6) 10 (1.8) 525 (94.9) 4 (b1.0) 3 (b1.0)
171 (42.0) 130 (31.9) 37 (9.1) 44 (10.8) 25 (6.1)
55 (9.9) 261 (47.2) 231 (41.8)
71 (17.6) 76 (18.6) 256 (62.7)
276 (49.9) 3 (b1.0) 39 (7.1) 129 (23.3)
162 (39.7) 12 (3.0) 29 (7.2) 198 (48.5)
Percentages may not add up to 100% because of missing data and rounding. Recruited from Women, Infants, and Children (WIC) clinics in West Virginia. Recruited from ‘safety net’, multi-purpose public health centers in Los Angeles County.
developed because a large proportion of the LA County population is of Hispanic origin and speaks Spanish.
beverages and fruits and vegetables), prevalence rate ratios (PRRs) and 95% confidence intervals (CIs) were generated for each community, using robust Poisson regression methods (Deddens and Peterson, 2008). BMI was the outcome variable of interest used in the multivariable models, where overweight (BMI ≥25 and BMI ≤29.9) and obesity (BMI ≥ 30) were collapsed. All data analyses were conducted using Stata/SE 12.1 (StataCorp LP, College Station, Texas, USA).
Results Of the 2092 parents approached in the WIC clinics, 33% refused and 30% were enrolled by the WV trained staff (total n = 630; women, n = 553). Of the 1393 patients approached in the designated public health centers, 26% refused and 74% were enrolled by the LA County trained staff (total n = 720; women, n = 408). Compared to women in LA County, WV participants were generally younger (Table 2). Women in the WV sample were predominately white (95%), whereas women in the LA County sample were predominately African American and Hispanic (74%, combined). Of the WV women, 73% were overweight and obese, as compared to 67% among LA County women (Fig. 1). In general, women in the LA County sample were more educated than women in the WV sample (63% versus 42%). They also reported consuming less soda (28% versus 37%) but more sugary drink alternatives (41% versus 32%) than their counterparts in WV. In both communities, race and ethnicity appeared to predict overweight and obesity; the associations to covariates, however, were not robust. In LA County, for instance, African American and Hispanic women were 1.4 times (95% CI = 1.12, 1.81) more likely than white women to be overweight and obese (Table 3). Discussion
Data management and statistical analysis For each community, common dietary behavior variables were identified. Due to sample variations and differences in some of the variable response categories, common categorical anchors were generated for key variables in each of the datasets from WV and LA County. For example, using Centers for Disease Control and Prevention (CDC)7 guidelines, both communities converted objectively measured heights and weights to a standard indicator — BMI (weight [kg] / height squared [m2]) (CDC, 2012), with BMI b 24.9, normal or nonobese; 25.0-29.9, overweight; ≥ 30.0, obese. We performed descriptive analyses to describe frequencies and differences in participant characteristics (e.g., demographic characteristics, eating behaviors) by community. To explore the relationships between the prevalence of overweight and obesity and covariates such as race/ethnicity, age, education and dietary behaviors (consumption frequency of sodas/sugar-sweetened
Fig. 1. Prevalence of overweight and obesity by population density (rural and urban).
The present case examples by population density (rural WV and urban LA County) highlight the burden of overweight and obesity among low-income women in two communities supported by CPPW during 2010–2012. Although the health assessment methods and data collection protocols differed somewhat from one another, both communities showed impressive magnitudes of obesity prevalence in this subpopulation, suggesting that federal investments in obesity prevention for these geographic regions were relatively well-aligned with the needs of these communities. Closer examination of each case example suggests that this burden may be greater than it appears in each setting. For example, we found obesity rates among LA County women to exceed 50%; this contrasts county-wide estimates of 30% for this same gender group (LACDPH-OWH, 2013). Similarly, when comparing health behaviors, approximately 27% of women in LA County reported consuming one soda or sugar-sweetened beverage per day whereas in the overall county population, this self-reported behavior was closer to 35% (LACDPH, 2011). Findings from our case studies aligned with those found in the literature, including: 1) low socioeconomic status is strongly associated with a variety of risk factors (e.g., diet) for obesity and obesity-related chronic diseases (Larson et al., 2009; Zhang and Wang, 2004); 2) lowincome individuals are less likely to consume nutritious foods (Lynch et al., 2004) and more likely to consume calorie-dense foods such as soda, sugar-sweetened beverages, and other processed foods (Cohen et al., 2010); and 3) fruit and vegetable consumption, a proxy for healthy eating, is disproportionately lower among low-income subgroups (Drewnowski, 2009). In LA County, African American and Hispanic women were more likely than white women to be overweight or obese. This observation, however, may be due to the higher representation of African Americans in the LA County sample. In contrast to recent U.S. Census estimates — African Americans accounted only for approximately 9% of the total county population (U.S. Census Bureau, 2012b), African Americans represented 42% of the LA case study sample. In WV, racial
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
5
Table 3 Overweight and obesity among low-income rural and urban women by socio-demographic characteristics and dietary behaviors, 2011a.
Race White African American or Hispanic Age 18–24 N25 Education Some college or college degree High school graduate or less Reported frequency of sugar-sweetened beverage consumption Does not drink/drinks sometimes Drinks often/everyday Reported frequency of soda consumption Does not drink/drinks sometimes Drinks often/everyday Average reported consumption of vegetables each day Four or more servings 3 servings or less Average reported consumption of fruits each day Four or more servings 3 servings or less
West Virginia (rural)
Los Angeles County (urban)
Adj. PRR (95% CI)b
Adj. PRR (95% CI)b
Referent 1.18 (1.00, 1.40)
Referent 1.42 (1.12, 1.81)
Referent 1.18 (1.05, 1.33)
Referent 1.44 (1.19, 1.74)
Referent 1.04 (0.94, 1.15)
Referent 0.93 (0.84, 1.03)
Referent 0.92 (0.82, 1.03)
Referent 0.95 (0.84, 1.07)
Referent 0.99 (0.89, 1.09)
Referent 1.04 (0.92, 1.18)
Referent 0.95 (0.83, 1.10)
Referent 1.13 (0.97, 1.33)
Referent 0.95 (0.80, 1.12)
Referent 0.82 (0.70, 0.97)
a Model: overweight and obese body mass index stratified by community, adjusted for other covariates, including socio-demographic characteristics (race, age, education) and dietary behaviors (e.g., soda and sugar-sweetened beverage consumption, fruits and vegetable intake); robust Poisson regression analyses were performed to estimate the adjusted prevalence rate ratios. b Adj. PRR = adjusted prevalence rate ratio, CI = confidence intervals.
differences were difficult to assess because more than 90% of health assessment participants were white. Although case studies provide important insights into regional differences in overweight and obesity — WV (rural) versus LA County (urban), inferences about the root causes of these regional disparities cannot be fully explained given the dissimilar methods used to collect the data. While it is possible that such factors as sparse open space, unsafe neighborhoods, an inefficient public transit system, limited access to grocery stores, and non-competitive food pricing (CPSTF, 2011; French et al., 2001; Larson et al., 2009; Moore et al., 2008; NPC, 2011) may all present important challenges to healthy eating and active living in both communities, the magnitude of how these factors differentially impact overweight and obesity prevalence across the two regions remain unclear and warrant further study. Unique regional preferences for soda and customs in preparing food, for example, may have differential impact on overweight and obesity prevalence across the various subgroups in each jurisdiction. Barriers to healthy eating (e.g., access to fresh fruits and vegetables) that are thought to be similar may actually be dissimilar, as the solutions to the obesity epidemic in each community may be different. Whereas capital investments in grocery stores or places that sell fresh fruits and vegetables (e.g., farmers market) are likely important for mitigating shortages of food venues in WV, conversion of existing corner stores (abundant in the neighborhood) or safer and easier access to public transportation to go to fartheraway locations may be more suitable for LA County. Further research is needed to examine these factors, as they are not the focus of these case study examples. Limitations The case study approach utilized in this article has several limitations. First, comparisons of WV and LA County samples could not be made, as different recruitment approaches and clientele characteristics precluded dataset comparability. Second, key differences in the two clinic populations' age, education, and the services sought by clients likely contributed to some selection bias in each community. Third, socioeconomic status was not easily established for both samples, as the two regional assessment instruments (surveys) did not directly ask
about participant income. Other sources of information were used to establish low socioeconomic status in WV and LA County. In WV, to receive services, all WIC clients must have incomes which fell at or below 185% of the U.S. Poverty Income Guidelines. In LA County, participants provided zip codes to verify their region of residence and answered questions about employment status, education, and usage of needbased public services. Conclusions The present case studies of rural WV and urban LA County represent unique snapshots of subpopulations targeted by the national CPPW program administered by the CDC (Bunnell et al., 2012). Results of the studies confirmed the need to invest in these regions, which contained high prevalence of overweight and obesity. Coupled to other system-level or multi-sector interventions, the range of nutrition interventions in WV and LA County (e.g., WIC health education; workplace breastfeeding accommodations; healthy food procurement practices; and public education) offer potentially meaningful opportunities to facilitate better food selections among low-income women and their families. These data provide invaluable insights on how these and other obesity prevention strategies can be tailored and refined to address the needs of this important segment of the population — a group that can have an enormous impact not only on what food they choose for themselves, but, more importantly, for their families. Collectively, these subpopulation health data served as an important guide for further planning of obesity prevention efforts in both communities; in many cases, these efforts became a part of the subsequent Community Transformation Grants portfolio. Conflict of interest statement The authors report no financial disclosures or conflicts of interest.
Acknowledgments The authors would like to thank the staff in the following agencies and organizations for their support and contributions to this article: CPPW-West Virginia (Principal Investigator Joe Barker); the West
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
6
B. Robles et al. / Preventive Medicine xxx (2014) xxx–xxx
Virginia Bureau for Public Health and the Mid-Ohio Valley Health Department; Los Angeles County Department of Public Health (Lisa V. Smith, Jennifer Piron, and Mirna Ponce); RTI International (Allie Lieberman and Jonathan Blitstein); and the CPPW Manuscript Writing Workshop sponsored by ICF International (Kathleen Whitten, Tesfayi Gebreselassie). The project was supported in part by cooperative agreements from the Centers for Disease Control and Prevention (#3U58DP002429-01S1, West Virginia and #3U58DP002485-01S1, Los Angeles County). The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the Los Angeles County Department of Public Health, the Centers for Disease Control and Prevention, or any other organization mentioned in the text. In accordance with U.S. law, no federal funds provided by CDC were permitted to be used by community grantees for lobbying or to influence, directly or indirectly, specific pieces of pending or proposed legislation at the federal, state, or local levels. As it relates to the CDCsponsored supplement, staff training and reviews by scientific writers were provided as technical assistance to the authors, through a contract with ICF International (Contract No. 200-2007-22643-003). CDC staff has reviewed the project's evaluation design and data collection methodology, and the article for scientific accuracy. All authors have read and approved the final version.
References Bunnell, R., O'Neil, D., Soler, R., et al., 2012. Fifty Communities Putting Prevention to Work: accelerating chronic disease prevention through policy, systems, and environmental change. J. Community Health 37 (5), 1081–1090. Center for Disease Control Prevention, 2012. Defining overweight and obesity. Available: http://www.cdc.gov/obesity/adult/defining.html (Last Accessed February 11, 2014). Charles, N., Kerr, M., 1988. Women, Food and Families. United Kingdom, Manchester University Press, Manchester. Cohen, D.A., Sturm, R., Scott, M., Farley, T.A., Bluthenthal, R., 2010. Not enough fruit and vegetables or too many cookies, candies, salty snacks, and soft drinks? Public Health Rep. 125 (1), 88–95. Community Preventive Services Task Force (CPSTF), 2011. The Guide to Community Preventive Services: What Works to Promote Health (The Community Guide). Available: http://www.thecommunityguide.org/library/ARC2011/congress-report-full.pdf (Last Accessed July 24, 2012). Cossrow, N., Falkner, B., 2004. Race/ethnic issues in obesity and obesity-related comorbidities. J. Clin. Endocrinol. Metab. 89 (6), 2590–2594. Deddens, J.A., Peterson, M.R., 2008. Approaches for estimating prevalence ratios. Occup. Environ. Med. 65, 501–506. http://dx.doi.org/10.1136/oem.2007.034777. Dennis, K.E., 2007. Postmenopausal women and the health consequences of obesity. J. Obstet. Gynecol. Neonatal. Nurs. 36 (5), 511–517. Drewnowski, A., 2009. Obesity, diets, and social inequalities. Nutr. Rev. 67 (1), S36–S39. French, S.A., Story, M., Jeffery, R.W., 2001. Environmental influences on eating and physical activity. Annu. Rev. Public Health 22, 309–335. Frieden, T.R., 2010. A framework for public health action: the health impact pyramid. Am. J. Public Health 100 (4), 590–595. http://dx.doi.org/10.2105/AJPH.2009.185652 (Epub 2010 Feb 18). Gore, S.A., Brown, D.M., West, D.S., 2003. The role of postpartum weight retention in obesity among women: a review of the evidence. Ann. Behav. Med. 26 (2), 149–159. Huffman, F.G., Kanikireddy, S., Patel, M., 2010. Parenthood—a contributing factor to childhood obesity. Int. J. Environ. Res. Public Health 7 (7), 2800–2810.
Krieger, N., 2003. Genders, sexes, and health: what are the connections—and why does it matter? Int. J. Epidemiol. 32 (4), 652–657. Larson, N.I., Story, M.T., Nelson, M.C., 2009. Neighborhood environments: disparities in access to healthy foods in the US. Am. J. Prev. Med. 36, 74–81. Los Angeles County Department of Public Health (LACDPH), 2011. Los Angeles County Health Survey. Available: http://publichealth.lacounty.gov/ha/hasurveyintro.htm (Last Accessed July 25, 2012). Los Angeles County Department of Public Health, Office of Women's Health (LACDPHOWH), Health Indicators for Women in Los Angeles County: Highlighting Disparities by Ethnicity and Poverty Level, June 2013. Available: http://www. publichealth.lacounty.gov/ha/docs/2011LACHS/WHI2013/WHI_2013s.pdf (Last Accessed February 11, 2014). Lynch, J., Smith, G.D., Harper, S., et al., 2004. Is income inequality a determinant of population health? Part 1. A systematic review. Milbank Q. 82 (1), 5–99. Manson, J.E., Colditz, G.A., Stampfer, M.J., et al., 1990. A prospective study of obesity and risk of coronary heart disease in women. N. Engl. J. Med. 322 (13), 882–889. Moore, L.V., Diez-Roux, A.V., Nettleton, J.A., Jacobs, D.R., 2008. Associations of the local food environment with diet quality—a comparison of assessments based on surveys and geographic information systems: the multi-ethnic study of atherosclerosis. Am. J. Epidemiol. 167, 917–924. National Center for Health Statistics (NCHS), 2011. National Health and Nutrition Examination Survey. Available: http://www.cdc.gov/nchs/nhanes/about_nhanes. htm (Last Accessed July 24, 2012). National Prevention Council (NPC), 2011. National Prevention Strategy. U.S. Department of Health and Human Services, Office of the Surgeon General, Washington, D.C. Ogden, C.L., Carroll, M.D., Kit, B.K., Flegal, K., 2012. Prevalence of obesity in the United States, 2009–2010. NCHS Data Brief, no 82. National Center for Health Statistics, Hyattsville, MD. Population Reference Bureau (PRB), 2011. U.S. Children In Single-Mother Families. Available: http://www.prb.org/Publications/PolicyBriefs/singlemotherfamilies.aspx (Last Accessed July 25, 2012). Sallis, J.F., Millstein, R.A., Carlson, J.A., 2011. Community design for physical activity. In: Dannenberg, A., Frumkin, H., Jackson, R. (Eds.), Making Healthy Places: Designing and Building for Health, Well-being, and Sustainability. Island Press, Washington D.C., pp. 33–49. Sebire, N.J., Jolly, M., Harris, J.P., et al., 2001. Maternal obesity and pregnancy outcome: a study of 287,213 pregnancies in London. Int. J. Obes. Relat. Metab. Disord. 25 (8), 1175–1182. Sen, G., Östlin, P., 2008. Gender inequity in health: why it exists and how we can change it. Glob. Public Health 3 (Suppl. 1), 1–12. Story, M., Kaphingst, K.M., Robinson-O'Brien, R., Glanz, K., 2008. Creating healthy food and eating environments: policy and environmental approaches. Annu. Rev. Public Health 29, 253–272. U.S. Census Bureau (QF-L), 2012b. Quick Facts: Los Angeles County, California. Available: http://quickfacts.census.gov/qfd/states/06/06037.html (Last Accessed March 12, 2012). U.S. Census Bureau (QF-WV), 2012a. Quick Facts: West Virginia. Calhoun, Pleasants, Ritchie, Roane, Wirt and Wood Counties. Available: http://quickfacts.census.gov/ qfd/states/54000.html (Last Accessed March 17, 2012). United States Department of Agriculture (USDA), 2003. Rural Urban Continuum Codes. Available: http://www.ers.usda.gov/data-products/rural-urban-continuum-codes. aspx (Last Accessed April 20, 2012). United States Department of Agriculture (USDA), 2011. Women, Infants and Children (WIC) Factsheet. Available: http://www.fns.usda.gov/wic/WIC-Fact-Sheet.pdf (Last Accessed July 25, 2012). Verbrugge, L.M., 1985. Gender and health: an update on hypotheses and evidence. J. Health Soc. Behav. 26 (3), 156–182. Wang, Y., Beydoun, M.A., 2007. The obesity epidemic in the United States—gender, age, socioeconomic, racial/ethnic, and geographic characteristics: a systematic review and meta-regression analysis. Epidemiol. Rev. 29, 6–28. Wild, R.A., Taylor, E.L., Knehans, A., Cleaver, V., 1994. Matriarchal model for cardiovascular prevention. Obstet. Gynecol. Surv. 49 (2), 147–152. Zhang, Q., Wang, Y., 2004. Trends in the association between obesity and socioeconomic status in U.S. adults: 1971 to 2000. Obesity Res. 12 (10), 1622–1632.
Please cite this article as: Robles, B., et al., Overweight and obesity among low-income women in rural West Virginia and urban Los Angeles County, Prev. Med. (2014), http://dx.doi.org/10.1016/j.ypmed.2014.02.016
