Hopkins Population Center Working Paper Archive

The Hopkins Population Center (HPC) was established in 1971 with a mandate to stimulate and facilitate interdisciplinary population research throughout the Johns Hopkins University. The HPC is unique among NICHD-funded centers in having a large majority of its faculty associates from health institutions (Bloomberg School of Public Health, School of Medicine, and School of Nursing). From its inception, the vision underlying the HPC has been the highest quality research, resulting from interactions among population researchers from diverse disciplinary backgrounds, aided by state of the art research infrastructure. Today, the HPC serves more than 50 research associates from the east-Baltimore medical campus and the Homewood arts and sciences campus. Homepage Current Publications


Recent Submissions

Now showing 1 - 20 of 58
  • Item
    Household Structure as a Response to Economic Adjustments : Evidence from the 1980's Urban Mexico
    (Hopkins Population Center, 1992) Wong, Rebeca; Levine, Ruth E.
  • Item
    Are Adolescent Mothers Just Single Mothers?
    (Hopkins Population Center, 1992-06) Astone, Nan Marie
  • Item
    The Role of Government Policy for Health : Equity versus Efficiency or Poverty versus Fiscal Vulnerability
    (Hopkins Population Center, 1992-04-11) Sirageldin, Ismail; Wouters, Annemarie; Diop, Francois
    Since the 1980s, many developing countries have been adjusting to severe macro economlc imbalances. There has been a concern that the accompanying changes in fiscal policy may be having a detrimental effect on the health status of these populations. To examine the relationships between changes in fiscal policy and health, a "social matrix" is developed which captures how the equity structure of the fiscal system can influence the efficiency with which health-related inputs are combined. It is the principal conclusion of this paper that household responses to changes in public policy can best be elucidated for a given society when the "social matrix" underlying the structure of the fiscal system is adequately specified. In contrast to other studies which look at poverty or vulnerability in general, this approach draws attentions to those who are fiscally vulnerable, that is, those who are dependent on public subsidies and as a result experience a fall in living standards (full income) when these subsidies are removed.
  • Item
    Childhood Precursors of Adult Morbidity and Mortality In Developing Countries : Implications for Health Programs
    (Hopkins Population Center, 1992-06-05) Mosley, W. Henry; Gray, Ron
    This paper will examine only one aspect of this health transition in developing countries. It will look at the emerging health problems among the adults and the aged and assess to what degree these chronic diseases and disabilities might be a consequence of infectious diseases and other adverse conditions which were experienced decades earlier in infancy and childhood. A recognition of these relationships can enhance our understanding of the cost-effectiveness and cost benefits of programs to promote child health. Child health interventions are not only cost-effective in saving lives and preventing disabilities in the short run but, more importantly, in the long run can result in major cost savings to health systems and accelerate national development by improving the health and productivity of these children when they become adults.
  • Item
    Demography and Policy : an Asian Experience
    (Hopkins Population Center, 1991) Sirageldin, Ismail
    The purpose of this paper is to investigate the factors influencing the use of demographic knowledge in the formulation, implementation and evaluation of public policies. The focus of the discussion is to conceptualize the dynamic and interactive nature of the supply and demand for demographic knowledge, or more generally, analyze the structure of the market for such knowledge.
  • Item
    Infertility and Early Pregnancy Loss
    (Hopkins Population Center, 1994) Hakim, Rosemarie B.; Gray, Ronald H.; Zacur, Howard; Gehret, Judith; Smith, Beverly
    The inability to achieve a recognized pregnancy may result from either failure of conception or implantation or an early postimplantation loss. Recently, a highly sensitive and specific assay for urinary human chorionic gonadotrophin (hCG) has enabled researchers to detect subclinical pregnancy losses. We used this assay to detect early pregnancy in a cohort of working women with and without fertility problems. One hundred forty eight female volunteers completed interviews, daily diaries an collected daily urine specimens for an average of seven months. The urine specimens were assayed for ovarian steroid hormones and hCG. Generalized estimating equations were used to estimate variance-corrected relative risks. There were 679 menstrual cycles at risk for pregnancy contributed by 124 of the women. Women with evidence of subfertility before or during the study period had a rate of early pregnancy loss of 70 percent compared to 21 percent of women without fertility problems (relative risk 2.63, 95%confidence interval 1.82-3.83. The risk of pregnancy loss associated with subfertility increased with age, and remained the same in women treated with clomiphene citrate. These results suggest that subfertile women have increased subclinical pregnancy losses regardless of fertility treatment, and the association between reduced fertility and advancing age may in part be related to early pregnancy loss.
  • Item
    The Association Between Grandparental Co-Residence and Adolescent Childbearing
    (Hopkins Population Center, 1993) Astone, Nan Marie
    There is some evidence to suggest that, in the US, young women are predisposed to have children early and outside of marriage and to marry early when growing up in a non-intact family, plagued by poverty and economic dependence. The number of children growing up in non-intact families has increased, and many grow up in families with multiple characteristics, which place them in a high risk context. In this examination of the influence of family structure on children's outcome, residence with extended kin, specifically grandparents, is examined for its impact on early childbearing. Data were obtained on a nationally representative sample of 4786 females who were sophomores in 1980 from the HIgh School and Beyond study. Dependent variables were evidence of childbearing before the age of 20 and childbearing out of wedlock before the age of 20. Independent variables were race and living arrangements that included both natural parents, one natural parent and one stepparent, one natural parent and no stepparent, and neither parent. In each of these situations, a dummy variable was coded for the presence of a grandparent in the household in both 1980 and 1982. Control variables were region, size of place of residence, number of siblings, and socioeconomic status score. Interaction effects between control and independent variables was found for growing up in a single parent family and being African-American. The results of logistic regression techniques showed that grandparent co-residence effects did not vary by race or family type. African-American teenagers were found to be more likely to have a teenage birth and more likely to have the birth outside of marriage. African-American teenagers from a single parent family were also more likely to have a teenage birth and to have the birth outside of marriage; this effect was not seen in Whites. The results are considered quite preliminary due to the marginal significance levels of coefficients and the lack of attention to selectivity bias. Grandparents co-residence was significantly negatively associated with a birth before the age of 20 at the .10 level of significance. The pattern was similar for outside marriage births but was not significant. There is a need for better specification of measures; for instance, there may be important differences between growing up with a divorced versus a never married mother.
  • Item
    Why Do Americans Want Children?
    (Hopkins Population Center, 1996-08) Schoen, Robert; Kim, Young J.; Nathanson, A.; Fields, Jason; Astone, Nan M.
    Data from the 1987-88 US National Survey of Families and Households are used to test four hypotheses about fertility intentions. Fertility intentions are examined as a function of the importance of the social resource value of children, economic costs, women's career impact, and childlessness. The study sample includes 4691 non-Hispanic White or Black women who are aged 16-39 years. The authors analyze subgroups stratified by race, gender, union, and parity. Findings indicate that fertility intentions declined with increased parity. Parity differences varied by race and union status. Multivariate logistic regression results confirm the impact of parity on fertility intentions. The social resource variable was positively related to fertility intentions for men and for women. The odds ratio was larger among persons with a high degree of agreement with attitudes valuing children as a resource. The 'economic cost of children' variable was only weakly and insignificantly related to fertility intentions. The 'career impact' variable was significant and positive for parity 0 White women. Both married and unmarried women with careers were less inclined to desire a child. The 'career impact' variable was significant but weak for married White women at parity 1 and 2 plus. The 'career impact' variable was only significant for White men at parity 1, and these men were less likely to desire another child. Childless men and women, married and unmarried, who supported childlessness, were less likely to desire a child. Married women at parity 1 and married men at parity 2 plus, who supported parenthood over childlessness, had lower fertility intentions. The analysis of Black men and women was hampered by small sample sizes, but race was not directly associated with fertility intentions at any parity. The most important predictor was actual fertility. A "primary motivator" of childbearing among low fertility populations is the social resource value of children.
  • Item
    Implications of Population Aging for Geriatric Health
    (Hopkins Population Center, 1998-08) Agree, Emily M.; Freedman, Vicki A.
    This analysis examines four theories about the relationships between mortality, morbidity, and disability in old age in the US and discusses the evidence. The theories include the pandemic of Chronic Diseases, Compression of Morbidity, Life Span Expansion, and Dynamic Equilibrium. Chronic Diseases theory (Gruenberg) posits that the age at onset of chronic diseases remains the same, but the number of years spent morbid or disabled will expand with increases in longevity. Compression of Morbidity theory (Fries) posits that the age at onset of chronic diseases will increase and more deaths will cluster around the maximum average life span. Life Span Expansion theory (Walford) posits that the onset of morbidity would shift to later ages but the years lived with morbidity and disability would remain unchanged. The Dynamic Equilibrium theory (Manton) assumes that life expectancy is increased through postponement of disease onset, reductions in severity of disease and speed of progression, and improved techniques for clinical management. All three curves for morbidity, mortality, and disability will increase, and the relationship between the curves will remain unknown. Evidence indicates that there have been increases in active and disabled life expectancy, the time spent in an active state, and the share of life expectancy due to active life expectancy. The data support Manton's Dynamic Equilibrium theory. Doctors and medical staff have a large role to play in determining the direction of changes in the health status of the elderly and in recovery-rehabilitation. The elderly's health situation will become even more diverse.
  • Item
    Disagreement in Spousal Reports of Current Contraceptive Use in Sub-Saharan Africa
    (Hopkins Population Center, 1998-12) Becker, Stan; Hossain, Mian Bazle; Thomson, Elizabeth
    Contraceptive prevalence is a key variable estimated from Demographic and Health Surveys. But the prevalence estimated from reports of husbands differs widely from that estimated for wives. In this research, using data from six Demographic and Health Surveys of sub-Saharan Africa, reports from spouses in monogamous couples with no other reported sex partners in the recent period are examined. Agreement ranged from 47% to 82%, but among couples in which one or both reported use, the 'both' category represented less than half in all nations except Zimbabwe. Husbands generally had higher reports of condoms, periodic abstinence and pills but fewer reports of the IUD, injections and female sterilization. Either discussion of family planning with the spouse and/or higher socioeconomic status was associated with agreement in most of the surveys. Ambiguities in the survey question regarding current use need to be reduced, perhaps with an added probe question for non-permanent methods.
  • Item
    Why Society Needs to Value Health Improvements in Dollars
    (Hopkins Population Center, 1998-07) Bishai, David
    Background. U.S. health planners typically set health objectives without information about how much of their resources the American people wish to devote to improved health. Objectives. This paper indicates how ignoring attempts to measure how much of their funds Americans will trade for better health and equity could lead to allocations that may be efficient, but which could still lower welfare. Methods. A graphical depiction of the process of health production and welfare maximization is supplemented by a mathematical model. Philosophical and empirical obstacles to measurement of social preferences are discussed. Results. Knowing only how to improve public health with the most cost-effective techniques cannot inform planners about how to make tradeoffs between health and other sectors of the economy. Starting from a point of inefficient health production, it is shown that not every point of efficiency will improve the welfare of society. Conclusions. More information about social preferences between health and other public spending could help inform exercises such as Healthy People 2000. Decision-makers using such information should be apprised of its inherent limitations and assumptions.
  • Item
    Simultaneity in Maternal-Child Health Care Utilization and Contraceptive Use: Evidence from Developing Countries
    (Hopkins Population Center, 1997-08) Ahmed, Saifuddin; Mosley, W. Henry
    This study examined the relationship between the use of maternal-child health (MCH) care and the use of contraceptives. The high correlation between the two may be due to the independent effect of one on the other or to an association of both with the same or similar background factors. We used structural equation models to examine the relationship between these two interventions. The data were derived from six Demographic and Health Surveys: Zimbabwe from Sub-Saharan Africa, Thailand from Asia, Egypt and Tunisia from North Africa, and Guatemala and Colombia from Latin America. The results show that in all six countries, the use of contraceptives and MCH care are significantly associated, independent of intervening factors; this finding suggests that families develop a joint demand for better-quality health and limited family size and translate these demands into action by using health services for mothers and for children and by voluntarily regulating fertility.
  • Item
    Generalized Exchange and Intergenerational Transfers in Taiwanese and Filipino Families
    (Hopkins Population Center, 1998-08) Agree, Emily; Biddlecom, Ann; Perez, Aurora
  • Item
    Parental Divorce and Partnership Dissolution in Adulthood: Evidence from a British Cohort Study
    (Hopkins Population Center, 1998-04) Kiernan, Kathleen; Cherlin, Andrew
    From a longitudinal survey of a British cohort born in 1958 this study finds that, by age 33, off-spring of parents who divorced are more likely to have dissolved their first partnerships. This finding persists after taking into account age at first partnership, type of first partnership (marital, pre-marital cohabiting union, and cohabiting union), and indicators of class background and childhood and adolescent school achievement and behaviour problems. Some of these factors are associated with partnership dissolution in their own right, but the association between parental divorce and second generation partnership dissolution is largely independent of them. Demographic factors, including type of and age at first partnership, were important links between parental divorce and partnership dissolution. Moreover, the estimated effects of parental divorce were substantially reduced when the demographic variables were taken into account, suggesting that cohabitation and early partnership may be important pathways through which a parental divorce, or the unmeasured characteristics correlated with it, affect partnership dissolution.
  • Item
    Individual and Community-Level Determinants of Domestic Violence in Rural Bangladesh
    (Hopkins Population Center, 1999-05) Koenig, Michael; Ahmed, Saifuddin; Haaga, John
    Although there is growing recognition of the problem of domestic violence in developing countries, relatively little remains known about both its prevalence and associated risk factors. The study employs multi-level models to explore individual- and community-level determinants of violence, using data from a 1993 cross-sectional survey of 10,368 reproductive aged women residing in two rural areas in Bangladesh. Overall, 42 percent of women reported having experienced recent physical violence by the husband or extended family members. The results highlight the importance of both community and individual-level factors -- particularly those related to women's status and autonomy-- in understanding the determinants of domestic violence in this setting. The results also suggest that the effects of specific individual- and community-level factors upon the risk of domestic violence may vary according to where on the continuum of women’s status and gender relations a specific setting is situated. The paper concludes with recommendations for promising avenues of inquiry for future research on the determinants of domestic violence in developing countries.
  • Item
    Multi-Generational Exchanges in Taiwan and the Philippines: A Social Network Approach
    (Hopkins Population Center, 1999-03) Agree, Emily; Biddlecom, Ann; Valente, Thomas
    Resource exchanges among multiple family generations are examined using social network measures. A specific focus is on pathways through which older adults transfer resources from one generation to another. The paper uses data that provide rich and comparable information on transfers in family networks: the 1996 Philippine Survey of the Near Elderly and Elderly and the 1989 Taiwan Survey of Health and Living Status of the Elderly. Measures of the direction and intensity of exchange are examined to identify similarities and differences between the two countries, and likely explanations for these patterns are discussed. Findings show that although the availability of and coresidence with specific types of family generations appear to be quite similar in both countries, there are substantial differences in the amount and nature of family exchanges. On most measures, older persons in the Philippines are more likely to be engaged in transfer activity, and to be exchanging resources with a greater number of family members across a wider variety of relations.
  • Item
    The Impact of Women's Participation in Credit Programs on the Demand for Quality Health Care in Rural Bangladesh
    (Hopkins Population Center Papers on Population, 1998-03-18) Nanda, Priya
  • Item
    The Use of Neighborhood Mapping in Community Evaluation: The Experience of the Baltimore City Healthy Start Evaluation
    (Hopkins Population Center, 1996-07) Aronson, Robert E.; O'Campo, Patricia J.; Peak, Geri L.