Applied Demography and Public Health

Nazrul Hoque, Mary A. McGehee and Benjamin S. Bradshaw (eds.)Applied Demography SeriesApplied Demography and Public Health201310.1007/978-94-007-6140-7_1© Springer Science+Business Media B.V. 2013

1. Introduction

Nazrul Hoque¹  , Mary A. McGehee²   and Benjamin S. Bradshaw³  

(1)

Department of Demography and Institute for Demographic and Socioeconomic Research, The University of Texas, San Antonio, TX, USA

(2)

Arkansas Department of Health, Health Statistics Branch, Center for Public Health Practice, Little Rock, AR, USA

(3)

Management, Policy & Community Health Division, The University of Texas School of Public Health, San Antonio, TX, USA

Nazrul Hoque (Corresponding author)

Email: [email protected]

Mary A. McGehee

Email: [email protected]

Benjamin S. Bradshaw

Email: [email protected]

Abstract

Applied demography is an ever-evolving field that is applicable to a number of disciplines (Siegel 2002; Murdock and Swanson 2010). It is a subfield of basic demography, which is the study of human populations and the size, composition, and spatial distribution of these populations and the related processes of fertility, mortality, and migration. Unlike basic demography which is primarily concerned with increasing knowledge about how these processes lead to demographic change, the focus of applied demography is understanding the consequences of demographic change, including the related social and economic consequences and their application to decisions related to policy development, planning, the distribution of goods and services, etc. in a specific area (Murdock and Ellis 1991).

Applied demography is an ever-evolving field that is applicable to a number of disciplines (Siegel 2002; Murdock and Swanson 2008). It is a subfield of basic demography, which is the study of human populations and the size, composition, and spatial distribution of these populations and the related processes of fertility, mortality, and migration. Unlike basic demography which is primarily concerned with increasing knowledge about how these processes lead to demographic change, the focus of applied demography is understanding the consequences of demographic change, including the related social and economic consequences and their application to decisions related to policy development, planning, the distribution of goods and services, etc. in a specific area (Murdock and Ellis 1991).

Public health is about the health of the public – the population – not the health of individual members (Turnock 2008). The emphasis is on the public – on the population itself. Public health and demography have been intimately related for centuries; indeed, it is difficult to separate the two. Because the focus of public health practice is on the population, success or failure of public health efforts are measured with population level data – e.g., rates, ratios, life expectancy, etc. From the time John Graunt (1662) began analyzing vital statistics data in the mid-17th century to the present, demographers have made observations either directly or indirectly on the health of the population with death and birth statistics and measures based on these in conjunction with census and survey information. Demographers have contributed invaluably to public health through development of population estimates and projections and evaluations of the quality of these ­products. Without good quality estimates and projections, planning for public health services would be nearly impossible. Often, the denominators for public health measures come from work done by applied demographers. Likewise, persons whose primary interest was in public health have made invaluable contributions to demography. This does not mean that public health practice may not include individual level interventions – e.g., assigning nurses to educate new mothers in proper care of infants, immunization programs, etc. – but unlike medical practice, the emphasis is on the population, not on the individual.

In this volume, the usefulness of the methods of applied demography in addressing public health issues is illustrated. This volume is not a complete overview of applied demography and public health and their interrelations. It is a collection of papers that illustrate some of the ways that applied demography can be used to inform or support public health issues. It should also be pointed out that the chapters were not necessarily prepared with customers in mind; they originated as papers presented at professional meetings. The work originated in a session at the Applied Demography Conference held in San Antonio, Texas, in January 2010. Some papers from that session are included in the volume. In addition, authors of appropriate papers presented at the annual meeting of the Population Association of America in April 2010 were invited to contribute their work. All contributions were sent for independent review. The editors are grateful to the contributors and the reviewers for their hard work.

This volume is intended for a wide range of users. Applied demographers, health policy makers, social epidemiologists, and social scientists or practitioners in the public and private sectors interested in public health can benefit from the information in this publication. In addition, this volume could be used as an accompanying text for courses in applied undergraduate and graduate demography and public health science courses. The authors’ papers provide a number of resources and tools that can be used by these persons in conducting research aimed at promoting public health: (1) information on a variety of health research datasets, (2) different statistical methodologies for analyzing health-related data and developing concepts related to health status, (3) methodologies for forecasting or projecting disease incidences and associated costs, and (4) discussion of demographic concepts used to measure population health status. Also important is the recurring theme of understanding how demographic and socioeconomic characteristics affect public health.

Overview of the Parts and Chapters

The volume is organized into five sections. Part I (Chaps.​ 2–5) examines the impact of aging on health and health-related expenditures, both in developed and developing countries. Part II (Chaps.​ 6–9) examines the use of data from cross-sectional surveys, vital statistics, and disease registries to estimate cause-specific mortality rates in the U.S. and Europe. These mortality rates are used by health professionals to examine progress in eliminating disparities in health between different populations. Part III (Chaps.​ 10–12) provides specific attention to maternal health and morbidity in developed and developing countries – including India, Liberia, and the U.S. Part IV (Chaps.​ 13–17) provides analyses of special cases, and Part V (Chaps.​ 18–21) covers methodological issues in public health. All five sections contain examples of how demographic techniques and concepts are related to public health.

Part I

The four chapters in Part I are examples of how demographic techniques can be useful in public health planning. In Chap.​ 2, Hoque and Howard use population projections and other formal demographic techniques to look at the impact of population growth, an aging population, and changes in the race and ethnic composition of the population on the increase in the prevalence of persons who are overweight and obese and the costs of overweight and obesity. This information will be useful to state health policy makers and administrators. In Chap.​ 3, Fox and Hutto build on this methodology by looking at the effect of obesity on intergenerational income mobility. Utilizing the National Longitudinal Survey of Youth 1979 data, Fox and Hutto compare the likelihood of upward mobility by obesity status. According to Fox and Hutto, obesity dampens upward mobility and increases downward mobility for overweight women. The chapter by Gorrindo et al. looks at the cross-country comparison of sociodemographic correlates of depression. They use data from the WHO study on global aging and adult health to examine the extent which sociodemographic variables correlate to the depressive conditions across five countries: China, India, Ghana, Mexico, and South Africa. Chapter 5 by Yadawendra Singh illustrates the use of demographic techniques to examine how population aging affects health expenditures and to provide estimates of the cost of managing chronic diseases in the future. Singh examines the burden of aging in terms of health expenditures in Kerala, the Indian state that has the highest proportion of elderly in India. Using methods of empirical analyses, the author found that the proportion of elderly having at least one ailment is much higher in Kerala compared to other Indian states and that per capita out-of-pocket health expenditure for non-elderly in-patient care is higher than that for the elderly. Based on the estimates the author predicted that the burden of managing the cost of diseases will increase significantly in coming future.

Part II

In Chap.​ 6, Smith, McFall, and Bradshaw use data from the Behavioral Risk Factor Surveillance System survey (BRFSS) and the National Center for Health Statistics to produce estimates of death rates for diabetics. The BRFSS is the world’s largest on-­going health survey and is conducted in all 50 states, the District of Columbia, Puerto Rico, the U.S. Virgin Islands, and Guam by the Centers for Disease Control and Prevention (CDC). Bishop-Royse and Eberstein use linked birth and infant death to examine causes of death that contribute to the racial disparities in infant mortality, as well the influence of the social context and maternal sociodemographic characteristics on these disparities (Chap.​ 7). In Chap.​ 8, Smith, McFall, and Bradshaw present a method for estimating death rates for subpopulations using successive cross-sectional survey data. In Chap.​ 9, Garcia and Crimmins compare cancer screening policies in the U.S. and Europe and conclude that countries with the highest screening rates have generally experienced faster decline in mortality. This chapter points out the importance of cancer screening to reduce the death due to cancer, particularly for people 50 years of age and above.

Part III

What are some of the public health issues that can affect fertility levels and fertility decisions in different societies? What are some of the demographic and socioeconomic factors that affect fertility levels and decisions? These questions are addressed in Part III, which focuses on maternal health and morbidity. In Chap.​ 10, Sontakke and Reshmi use the National Family Health Survey (NFHS) to study obstetric morbidity in India and to examine the relationship between socioeconomic and demographic factors and obstetric morbidity. The findings suggest that level of obstetric morbidity varied among the Indian states. Meadows et al. use data from the Panel Survey of Income Dynamics (PSID) to assess how an adolescent mother’s self-rated health status can predict her odds of experiencing childbirth at a young age (Chap.​ 11). Murty and McCamey use data from a survey of Liberian women (ages 13–49) to examine how selected demographic and socioeconomic factors affect maternal health and morbidity in post-war Liberia. The authors concluded that maternal mortality in Liberia is very high and efforts are underway to develop better health services to the marginalized areas as they struggle to recover from two decades of war (Chap.​ 12).

Part IV

Part IV addresses various topics that provide excellent examples of the interaction between demographic characteristics and processes and their implications for public health policy, health intervention strategies, and health promotion. Also, in this section, a number of demographic concepts used to measure the health status of the population are discussed. In particular, the authors provide examples of how spatial distributions or concentrations of the population, population characteristics, and progressions through different lifestages affect health. In Chap.​ 13, Snedker et al. look at neighborhood characteristics and their influence on adolescent health. This research has clear policy implications for those interested in better understanding the relationship between neighborhood context and individual-­level outcomes. Mancha and Zey employ a new methodological approach to explore the relationship between alcohol retail outlet density, population density, and crime (Chap.​ 14). Mondal et al. use data from 406 hypertensive diabetic patients from the Rajshahi Medical College Hospital and Rajshahi Diabetic Centers in Bangladesh who are receiving treatment for HTN and diabetics to examine the relationship between socio-demographic characteristics and the control of diabetes and hypertension (Chap.​ 15). Life expectancy, a demographic concept used to measure health status, is the topic of the paper by Tareque et al. This chapter examines the relationship between an active aging index and healthy life expectancy (HLE) in Bangladesh using data collected from the Rajshahi District of Bangaldesh (Chap.​ 16). Finally, Shim et al. (Chap.​ 17) present their results of a systematic literature review to address whether type of retirement is a risk factor for mortality.

Part V

The chapters in Part V delve deeper into different methodologies, including demographic methodologies, that can be used to assess and understand factors related to population health. The paper by Kamiya et al. analyzes the relationship between social engagement – or social structure – and health and the biomarkers used to measure this relationship (Chap.​ 18). Yashin et al. (Chap.​ 19) discuss an approach to mortality modeling that takes into account the internal and external factors related to health. According to the authors this type of analysis is useful to health practitioners in developing personalized preventive and treatment strategies. McFall and Buck discuss the usefulness of the UK Household Longitudinal Survey as a resource for research in demography and health (Chap.​ 20). The use of the Lee-Carter method, which combines a demographic model with a times-series method to forecast health services, is illustrated by Rodrigues et al. (Chap.​ 21). This chapter would be helpful to the health practitioners and policy makers who are concerned about future health care use and costs.

References

Graunt, J. (1662). Natural and political observations mentioned in a following index, and made upon the bills of mortality. New York: Evergreen Review, Inc.

Murdock, S. H., & Ellis, D. R. (1991). Applied demography: An introduction to basic concepts, methods, and data. Boulder: West View Press.

Murdock, S. H., & Swanson, D. A. (Eds.). (2008). Applied demography in the 21st century (pp. 3–10). New York: Springer Science+Business Media B.V.

Siegel, J. (2002). Applied demography: Applications to business, government, law and public policy. San Diego: Academic.

Turnock, B. J. (2008). Public health: What it is and how it works. Sudbury: Jones and Bartlett Publishers.

Part 1

Impact of Aging on Health and Health-Related Expenditure

Nazrul Hoque, Mary A. McGehee and Benjamin S. Bradshaw (eds.)Applied Demography SeriesApplied Demography and Public Health201310.1007/978-94-007-6140-7_2© Springer Science+Business Media B.V. 2013

2. The Implications of Aging and Diversification of Population on Overweight and Obesity and the Cost Associated with Overweight and Obesity in Texas, 2000–2040

Nazrul Hoque¹   and Jeffrey Howard²

(1)

Department of Demography and Institute for Demographic and Socioeconomic Research, University of Texas at San Antonio, San Antonio, TX, USA

(2)

Department of Demography, University of Texas at San Antonio, San Antonio, TX, USA

Nazrul Hoque

Email: [email protected]

Abstract

Overweight and obesity are major health concerns in contemporary America. The percentage of the population that is considered overweight and obese has increased substantially over the past years for both adults and children. Approximately 133.6 million American adults, or 66.0 % of all adults, are either overweight or obese (NIH 2004). Obesity rates have more than doubled since 1990, increasing from 11.6 % in 1990 to 26.3 % in 2007 (CDC 1991, 2007). This increase is of substantial concern because of the health risks associated with overweight and obesity. Overweight and obesity are related to increased risk for heart disease, type 2 diabetes, and a number of other diseases (Wolk et al. 2001; Calle et al. 2003). Approximately 300,000–400,000 deaths each year in the United States are attributable to overweight and obesity status (Allison et al. 1999; Obesity in America 2004).

Introduction

Overweight and obesity are major health concerns in contemporary America. The percentage of the population that is considered overweight and obese has increased substantially over the past years for both adults and children. Approximately 133.6 million American adults, or 66.0% of all adults, are either overweight or obese (NIH 2004). Obesity rates have more than doubled since 1990, increasing from 11.6% in 1990 to 26.3% in 2007 (CDC1991, 2007). This increase is of substantial concern because of the health risks associated with overweight and obesity. Overweight and obesity are related to increased risk for heart disease, type 2 diabetes, and a number of other diseases (Wolk et al. 2001; Calle et al. 2003). Approximately 300,000–400,000 deaths each year in the United States are attributable to overweight and obesity status (Allison et al. 1999; Obesity in America 2004).

In addition to observing differences in the prevalence rate of overweight and obesity for males and females (Baskin et al. 2005; Flegal et al. 2010), higher body mass indices tend to be more prevalent among minority population members due to an association between low socioeconomic status and a variety of other factors such as historical and cultural aspects. African Americans and Hispanics tend to be more likely than non-Hispanic Whites and other groups to have higher Body Mass Index scores. This is due, in large part, to long-term patterns of socioeconomic disparity with African Americans and Hispanics having rates of poverty that are two to three times the levels for non-Hispanic Whites and median income levels that are roughly 60–65% of those for non-Hispanic Whites (US Census Bureau 2008). Because of these differences and the fact that these differences have been evident for decades, African Americans and Hispanics are more likely to purchase and consume high calorie, high carbohydrate foods that are generally less expensive than many lower-­calorie fruits, vegetables and lean meat alternatives. For both African Americans and Hispanics decades of impoverishment have led to food consumption patterns based on cultural traditions and cost considerations that are likely to lead to higher likelihoods of being overweight or obese.

Based on data for adults 20 years of age and older from the National health and Nutrition Examination Survey (NHANES) 2007–2008, the age-adjusted overweight and obesity prevalence rates for Hispanics are more than six points higher, while for African Americans the prevalence rates are more than 11 points higher, compared with non-Hispanic Whites. The prevalence rates are 38.7, 44.1, and 32.4% for Hispanic, African American, and non-Hispanic Whites, respectively (Flegal et al. 2010). The prevalence rates for adult population for all race/ethnicity groups are higher compared with young population. Consequently, in areas with more diverse or aging populations, the health and financial implications can present significant concerns for policy makers and health care providers within these communities.

Texas provides an excellent location for an examination of the potential effects and the costs of overweight and obesity because of the prevalence of these conditions in its population and the rapid diversification of its population. Texas had the sixth highest rate of adult obesity among all U.S. states in 2000 at 25.3% and had the fifth highest rate of obese adults in 2001 and 2002 at 24.6 and 25.5, respectively (CDC 2000, 2001, 2002, 2007). Among Texas adults, overweight and obesity prevalence rose from 43.0% in 1990 to 63.0% in 2002. This is an increase of 20 percentage points and a change of 56.5% in only 12 years. The economic burden associated with overweight and obesity in Texas is substantial. In Texas, overweight- and obesity-related direct and indirect costs among adults reached an estimated $11.3 billion during 2000 (Hoque et al. 2010).

Identifying the magnitude of potential future increases in the number of overweight and obese persons in Texas may also be informative for the nation because the current characteristics of Texas’ population are similar to those projected for the United States as a whole (U.S. Bureau of the Census 2008). In 2000, Texas had a population that was 53% non-Hispanic White, and by 2004 it had become a majority minority state. In 2000 slightly more than 30% of the Texas population was Hispanic and approximately 12% that was African-American. In 2010, the proportion of non-Hispanic White population decreased to 46.6% while the proportion of Hispanic population increased to 37.5%. The United States population (U.S. Census Bureau 2008) is projected to become a majority minority by 2042 with approximately 30% Hispanics and about 12% African-Americans by midcentury. Texas data, consequently, can offer insights that are relevant to national health and other planning efforts.

In this paper, we demonstrate the importance of data on the projected number of overweight and obese adults to identify the cost of overweight and obesity as well as the relative effects of population growth, change in the age structure, and change in the racial/ethnic composition of the population. Such a decomposition is important because it allows one to identify the detailed causes of such increases and to direct health education efforts toward those segments of the population most likely to experience increases in overweight and obesity. This work provides a clear example of the use of applied and general demographic methods to address a critical area of concern for the nation and its component states.

McCusker et al. (2004) used a cohort-component projection model to project the number of overweight and obese adults and projected annual costs associated with those overweight and obese persons in Texas through the year 2040. This study expands the analysis by McCusker et al. by decomposing the change in the prevalence of overweight and obese adults for successive periods due to (a) the rate of population growth, (b) change in the age structure, and (c) change in the racial/ethnic composition of the population. The importance of understanding these three dimensions in determining overweight and obesity is evident from the demographics of Texas. Texas’ population has grown faster than the national population rate during every census since 1850. The Texas population increased by 2.7% annually during the 1970s compared with 1.1% nationally. During the 1990s, Texas populations increased by 2.3% while the U.S. population increased by 1.3%. Although growth has slowed since 2000, the Texas population growth rate is still higher than that for the U.S., 2.1 and 1.0%, respectively, from 2000 to 2010. Texas population of 25.1 million in 2010 may exceed 45 million by 2040.

The median age of Texas population was 28 years in 1980, 30 years in 1990, and 32.3 years in 2000. The median age for the total population of Texas is expected to increase from 32.3 in 2000 to 38.1 in 2040. For the Anglo population it would increase from 38.0 to 46.2 years, for the Hispanic population it would increase from 25.5 to 34.2 years, for the Black population it would increase from 29.6 to 39.8 years, and for the Other population (composed of Asians, American Indians, and other nonwhite, non-Hispanic, non-African American groups) it would increase from 31.1 to 48.7 years. The proportion of the Texas population 65 years of age or older has increased from 9.6% in 1980 to 10.1% in 1990 to and 13.8% in 2000, and it is projected to increase to 20.6% by 2040.

The proportion of the Hispanic population in Texas has increased from 21.0% in 1980 to 25.6% in 1990 and to 32.0% in 2000. The proportion of Other population has increased from 1.4% in 1980 to 2.2% in 1990 and to 3.3% in 2000. By contrast, the proportion of the population that is Anglo population decreased from 65.7% in 1980 to 60.6% in 1990 and to 53.1% in 2000. The proportion of the population that is Black has remained relatively static (11.9% in 1980 and 11.6% in 1990 and 2000).

Both in Texas and nationwide, the prevalence of obesity is higher among Black and Hispanic adults compared to non-Hispanic white adults. In Texas, both the prevalence of overweight and obesity and the number of Hispanic adults are expected to increase during the next four decades. To assess the impact of these changes, we projected (1) the number of normal weight, overweight, and obese adults in Texas through the year 2040, and (2) the costs associated with overweight and obesity. In examining the results presented in this paper, it is important to acknowledge that a variety of factors such as social, economic, resource, and policy factors may have as large or even larger effects on overweight and obesity and the cost associated with overweight and obesity than demographic forces. However, we believe that demographic change will play a major role in determining the future overweight and obese status in Texas as well as in the United States. Thus, we maintain that it is useful to examine the implications of such factors on overweight and obesity for policy purposes.

Methods

The analysis was performed in four stages. The first stage involved the preparation of detailed projections of the population by age, sex, and race/ethnicity in Texas. The second stage involved the derivation of age, sex and race/ethnicity-specific rates for normal weight, overweight, and obesity and multiplication of these rates by the number of persons in the projected population cohorts. The third stage involved the computation of direct and indirect costs associated with overweight and obesity for the base year and multiplication of these costs by the projected number of overweight and obese population to obtain the projected costs associated with overweight and obesity in Texas. The fourth stage involved the computation of decomposition effects due to changes in population growth, age structure, and race/ethnicity composition. The procedures followed in each of these stages are described in the following sections.

Projecting the Population of Texas

Population projections were completed with a cohort-component projection technique. The basic characteristics of this technique are the use of separate cohorts – persons with one or more common characteristics, and the separate projection of each of the major components of population change – fertility, mortality, and migration, for each cohort. These component projections are then combined in the demographic equation as follows:

Where Pt = population for the projected year; Po = population at the base year; B = births between P0 and Pt; D = deaths between P0 and Pt; NM = net migration between P0 and Pt.

The following major steps were performed for the population projections: (1) a baseline set of cohorts was selected for the projection area of interest for the baseline time period; (2) appropriate baseline fertility, mortality, and migration measures were determined for each cohort for the baseline time period; (3) a method was determined for projecting trends in fertility, mortality, and migration rates over the projection period; and (4) a computational procedure was selected for applying the rates to the baseline cohorts to project the population for the projection period.

The four baseline cohorts used in the projections were single-year-of-age cohorts of male and female Anglos (white non-Hispanic persons), Blacks (Black non-­Hispanic persons), Hispanics (Spanish-origin persons of all racial and ethnic groups), and Other (persons in all other nonAnglo, nonAfrican American, non-­Hispanic racial and ethnic groups). These cohorts were extracted from Summary File 1 of the 2000 Census of Population and Housing. A detailed description of the methods for formulating these race/ethnicity combinations and the projections methodology can be obtained from the authors.

Baseline age and race/ethnicity-specific fertility rates were computed with data from the Texas Department of State Health Services for births by age and race/ethnicity, and the mother’s place of residence The numerators for such rates are the average number of births for 1999, 2000 and 2001 for mothers in each age and race/ethnicity group and the denominators are the population counts by age and race/ethnicity in 2000. These data showed total fertility rates for Anglo, Black, Hispanic and the Other racial/ethnic groups in 2000 that were 1.92, 2.05, 2.92 and 1.95 respectively for the State of Texas.

To project fertility rates, we examined the historical patterns and trends in total fertility rates by race/ethnicity. Evaluation of these age and race/ethnicity-specific fertility rates in Texas showed patterns of slightly increased fertility among Anglos from 1990 to 2000. Rates for Blacks showed a decrease of nearly 14% from 1990 to 2000. Hispanic showed a decline of more than 6% from 1990 to 2000. The rates for Other racial/ethnic group decreased from a total fertility rate of 2.04 in 1990 to 1.95 in 2000.

Survival rates by age, sex, and race/ethnicity were computed using the death data from Texas Department of Health for 1999, 2000, and 2001 and 2000 census population. Survival rates were projected assuming that the survival rates of Texas will follow the national trends for the projection period. We computed the ratio of Texas’s 2000 age, sex, and race/ethnicity specific survival rates to those of the nation and applied these ratios to the projected survival rates of the U.S. The national rates were obtained from the Population Projections Branch of the U.S. Bureau of the Census (Spencer 1984, 1986, 1989; Hollmann et al. 2000).

Migration is the most difficult component of population projections for which to obtain base rates and projected rates. We used a vital statistics method which is the simplest and most accurate method of estimating net migration (Texas Population Estimates and Projection Program 2006). In this method net migration is equal to population change minus natural increase (births minus deaths). Thus, births and deaths by age, sex, and race/ethnicity were added to and subtracted from the 1990 population to produce an expected 2000 population. This expected population was compared to the actual 2000 census counts to estimate net migration rates for 1990–2000 periods. These rates provided the migration component for population projections.

Four alternative population projection scenarios were developed based on projected trends in fertility and survival rates and alternative assumptions regarding net migration. These scenarios used the same fertility and survival assumptions but three different sets of migration assumptions. One scenario assumed no net migration (that is, the net of in and out migrations are equal to zero or there is no migration) by age, sex, and race/ethnicity, referred to as the 0.0 migration scenario. A second scenario assumed that rates of age, sex, and race/ethnicity specific net migration were equal to one-half of the 1990–2000 rates, referred to as the 0.5 migration scenario. A third scenario assumed a continuation of the 1990–2000 rates of age, sex, and race/ethnicity specific net migration, referred to as the 1.0 migration scenario. A fourth scenario assumed a continuation of the 2000–2004 rates of age, sex, and race/ethnicity specific net migration, referred to as the 2000–2004 migration scenario (For a detailed discussion on population projections, refer to the projection methodology at the TSDC website, which can be accessed at the location noted above).

The 2000–2004 scenario is used for this analysis because we compared the 2010 projected population produced by different scenarios with the census counts of 2010 population and the 2000–2004 scenario is closest to the actual counts. However, as with all other projections, if the assumptions about the future demographic processes are incorrect, the projections will be incorrect. Thus, the 2000–2004 scenario assumes that migration will average 198,290 persons per year throughout the projection period. Total fertility rates for Anglos will decline from 1.92 to 1.72, for Blacks from 2.05 to 1.72, for Hispanics from 2.85 to 2.20 and for Others from 1.95 to 1.86 by 2040. Survival rates for Texas were assumed to follow the national level (i.e., life expectancy will increase by 6 years from 2000 to 2040). If these assumptions are incorrect, the projections will be incorrect. We utilize these projections with full realization of their limitations but with the expectation that although the exact number of persons projected to live in the state in the future is unlikely to be correct, the general trends are likely to be in the direction indicated.

Measures of Prevalence for Normal Weight, Overweight, and Obesity

Individuals are defined as normal weight, overweight, or obese by having a body mass index (BMI) of less than 25.0, between 25.0 and 29.9, and over 30.0 kg/m², respectively as defined by the National Institutes of Health (NIH 1998). Estimates of the prevalence of normal weight, overweight, and obesity among Texas adults by age group, sex, and race/ethnicity were derived from data collected during 1999–2002 by the Texas Behavioral Risk Factor Surveillance System (BRFSS), an ongoing telephone survey of state residents’ health conditions and behaviors coordinated by the Centers for Disease Control and Prevention. Respondents’ self-reported height and weight were used to calculate their BMI (weight in kilograms divided by height in meters squared). The age, sex, and race/ethnicity-specific prevalence rates used are presented in Table 2.1.

Table 2.1

Prevalence rate (per 100) of overweight and obese adults in Texas by age group, sex, and race/ethnicity, 1999–2002

As shown in totals by race/ethnicity on Table 2.1, Black females have the highest female obesity prevalence at 34.2% followed by Hispanic females at 30.7%, Anglo females at 18.9%, while Other females have the lowest rate of 13.9%. Black males also have the highest male obesity prevalence of 32.3% followed by Hispanic males with 28.3%, Anglo males with 21.8%, and Other males with 13.0%. The prevalence of overweight is highest for Black females at 33.2% followed by Hispanic females at 32.0%, Anglo females at 27.9% and Other females at 23.7%. The prevalence of overweight is highest for Anglo males at 46.5% followed by Hispanic males at 45.4%, Black males at 41.9%, and Other males at 38.1%.

Next we examine the prevalence of overweight and obesity by age and age within each race/ethnicity group. The prevalence of overweight is 20% higher for Anglo males aged 25–34, compared to Anglo females, and 23.2% higher for Anglo males aged 35–44. The overall prevalence of overweight for Anglo males is 18.6% higher than Anglo females (46.5 and 27.9%, respectively). The prevalence of obesity is also higher for Anglo males than females for all age groups. The overall obesity for Anglo males is 2.9% higher than for females. The prevalence of overweight is higher for Black males than females except in the age group 65 and above where the female rate is higher than the male rate. For obesity, female rates are higher than male rates in the 18–24 and 35–64 age ranges, while males are higher for ages 25–34 and 65 and over. For the Hispanic population of all ages the prevalence of overweight is 13.4% higher for males than for females, 45.4 and 32.0%, respectively. For obesity, the Hispanic female prevalence is a little higher than the male rate, 30.7 and 28.3%, respectively. For the Other population of all ages, the overweight prevalence is 14.4% higher for males than for females, 38.1 and 23.7%, respectively. The obesity prevalence for Other females for age 65 and over is 15.3 percentage points higher than the male rate, 23.0 and 8.3, respectively. Obesity prevalence ranged from 3.2% for Other males 18–24 years of age to 39.1% for Black males 65 years of age and over. Overall, the prevalence for obesity is higher for Black and Hispanic populations compared to the Anglo population.

Projected Changes in Prevalence of Overweight and Obesity

In the U.S., the prevalence of overweight and obesity has increased dramatically during the past 20 years. As mentioned in the introduction, obesity rates have increased from 11.6% in 1990 to 26.3% in 2007. The increase in the prevalence of obesity has been so rapid during recent years that the rate of increase is not likely to be sustainable over time. For this reason, the future rates of change in the prevalence of overweight and obese adults were assumed to decrease incrementally over time. Changes in the prevalence of overweight and obesity were based on data from the 1990 to 2002 national BRFSS. The rates of change in prevalence were assumed to slow over time, with prevalence decreasing linearly by one-fourth of the 1990–2002 decade equivalent from 2000 to 2010, and decreasing by an additional one-half of the previous decades’ prevalence in each of the next three decades. For comparison purposes another set of projections were completed using constant 1999–2002 prevalence of overweight and obesity. Due to space limitations we do not present the later set of projections.

Projected Costs of Overweight and Obesity

The projected costs of overweight and obesity were derived from previously published direct and indirect cost estimates for the State of Texas. According to 2001 cost estimates for overweight and obesity, total annual direct and indirect costs were $471 for each overweight adult and $2,249 for each obese adult in Texas (McCusker et al. 2004). A more recent study by Dor et al. (2010) estimates the yearly cost at $432 for an overweight man, $524 for an overweight women, $2,646 for an obese man, and $4,879 for an obese woman. Since this is a recent study, we use these 2010 values to project the future cost associated with overweight and obesity in Texas. These are direct medical costs which include both out-of-pocket and insurance-­covered expenditures related to physician services, hospital care, and pharmaceuticals. We have not added indirect expenses or the value of lost life to these annual costs. Adding the value of lost life would produce much higher costs. According to Dor et al. (2010), including the values for loss of life would increase the estimated costs to $8,365 for obese women and $6,518 for obese men.

Decomposition Analysis

Finally, we used decomposition techniques to identify how each of the three factors studied affected changes in the number of overweight and obese adults relative to the population base. Decomposition analysis provides a technique for identifying the proportion of a difference between two crude rates that is attributable to each of a set of demographic factors (Kitagawa 1955; Das Gupta 1978). Decomposition analysis is clearly an appropriate technique for discerning how demographic factors will affect the number of overweight and obese adults in Texas at different points in times. By using decomposition techniques, we discern what portion of the change in overweight and obese adults for each of the four time periods (2000–2010, 2010–2020, 2020–2040 and 2000–2040) is attributable to population change, what portion is attributable to change in the age structure, and what portion is attributable to change in race/ethnicity composition occurring between the two periods.

Results

We used all four population projections scenarios to project the number of overweight and obese adults and the costs associated with overweight and obesity in Texas through 2040. The data presented here are for the zero net migration and 2000–2004 migration scenarios. As noted before, we believe that the 2000–2004 scenario is most likely to approximate future levels of population change in Texas, and other scenarios (i.e., 1990–2000) show similar patterns to those shown for the 2000–2004 scenario; the zero net migration scenario provides a projection of population based on natural increase (i.e., the excess of birth over deaths). By comparing results for the zero migration scenario to the 2000–2004 scenario, we can examine the likely impact of migration in overall patterns of population growth. This comparison is important because migrants (particularly immigrants) are likely to have a different prevalence of overweight and obesity than the native born population and thus have potentially dramatic impacts on overweight and obese status.

An Overview of Major Demographic Trends

Table 2.2 provides population data for the adult population (18 years of age and older) in Texas by race/ethnicity for 2000 and projected to 2040 by alternative migration scenarios. According to the 2000–2004 migration scenario, the adult population in Texas is expected to increase from 15 million in 2000 to 34.4 million in 2040. The adult population by race/ethnicity is projected to increase from the year 2000 to 2040 as follows: the non-Hispanic White or Anglo population will increase from 8.5 to 9 million, the Black population will increase from 1.7 to 2.8 million, the Hispanic population will increase from 4.3 to 19.3 million, and the Other population will increase from 506,711 persons to 3.3 million (Table 2.2).

Table 2.2

Population 18 years of age and older by race/ethnicity in 2000 and projected to 2040 under alternative projection scenarios for Texas (in thousands)

The Anglo adult population would account for only 2.6% of net growth from 2000 to 2040, which means the rest of the net growth will come from the minority population, especially the Hispanic population that would account for 77.3% of the net change.

Migration will play an important role in future population change in Texas. The 2000–2004 migration scenario suggest that 72.7% of the net growth in Texas adult population from 2000 to 2040 is likely to be due to net migration.

The total adult population will increase by 130.1% from 2000 to 2040. Although all racial/ethnic groups show population increase during this time; the Anglo population will only increase by 5.7%, the Black population will increase by 71.1%, the Hispanic population will increase by 350.2%, and the Other population will increase by 554.9% (Table 2.3). Under zero net migration scenario, total adult population will increase by only 35.6%, Anglo by 4.0%, Black by 35.5%, Hispanic by 97.3% and Other population by 46.7%.

Table 2.3

Percent change for selected time periods for projected adult population by race/ethnicity under alternative projection scenarios for Texas

Increase in Minority Populations

The second major demographic pattern is the trend toward an increasing number and proportion of minority population Table 2.4 presents the percent of the adult population (18 years of age and older) in Texas by race/ethnicity for 2000 and projected to 2040. Based on the 2000–2004 scenario, the proportion of the adult population that is Black is expected to decrease from 11.0% in 2000 to 8.2% in 2040. In this same scenario, the proportion of Hispanic adults is expected to increase from 28.6% in 2000 to 56% in 2040. The proportion of Other adults will increase from 3.4% in 2000 to 9.6% in 2040.

Table 2.4

Proportion of population 18 years of age and older by race/ethnicity in 2000 and projected to 2040 under alternative projection scenarios for Texas

The Aging Population

The trend toward an older age structure in the U.S. and Texas population has been widely discussed elsewhere (Murdock et al. 2003). The projected percent of population by age group and race/ethnicity from 2000 to 2040 for Texas are provided in Table 2.5. The proportion of the total population that is 65 years of age or older is projected to increase from 13.8% in 2000 to 20.6% in 2040. During this period, the proportion of the population 65 years of age or older within each race/ethnicity is projected to increase as follows: the Hispanic population will increase from 8.1 to 13.6%, the Black population will increase from 10.6 to 20.8%, the Anglo population will increase from 17.8 to 31.4%, and the Other population will increase from 6.7 to 32.0% (2000–2004 scenario).

Table 2.5

Percent of adult population by age group and race/ethnicity in 2000 and projected percent of population by age group and race/ethnicity for 2040 under alternative projection scenarios for Texas

Table 2.6 shows the percent change in population from 2000 to 2040 by age group and race/ethnicity. For the total population as well as for each race/ethnicity, the already mentioned population growth can be seen. This can be observed in each age group, yet with great differences in the percent change, the younger age groups are expected to grow considerably less than the older age groups. As noted before, the Anglo population is expected to grow the least (as measured in percent change), while the population that is Hispanic has the highest percent change within all age groups from 2000 to 2040.

Table 2.6

Percent change in population by age group and race/ethnicity from 2000 to 2040 under alternative projection scenarios for Texas