2012–13 Pilot Projects

Is There an Hormonal Component to Sexual Functioning in Later Life?

PI: John DeLamater, Conway-Bascom Professor, sociology
Abstract: There have been occasional studies of the relationship between levels of hormones and sexual desire and sexual behavior in men and women. The hormones most frequently studied have been dehydroepiandrosterone (DHEA and DHEA-S), estradiol, progesterone, and testosterone. The results of these studies have been mixed, due in part to the use of small samples and variation in the measurement protocols for the hormonal assays. Few studies have analyzed hormonal levels and sexual functioning in later life. Pollet, et al., (2011) reported that, among persons over 57 years of age, testosterone was significantly associated with lifetime number of partners among men but not women. Spark (2002) suggested, based on limited data, that DHEAS levels are associated with libido and sexual satisfaction in women over 60.
The National Social Life, Health and Aging Project (NSHAP) provides for the first time the opportunity to analyze these relationships in a large, representative sample of older Americans. The sample includes 3005 men and women ages 57 to 85. The face-to-face interview obtained data on current sexual expression and relationship status, as well as retrospective relation and sexual history data. The interviewer requested that each participant provide a salivary specimen; 2, 721 agreed, and 2,640 were able to produce one. Salivary testing was carried out according to accepted protocols (Waite, et al., nd). The samples were assayed for cotinine, DHEA, estradiol, progesterone and testosterone.
I will begin by looking at correlations between hormone levels and measures of sexual desire, frequency of kissing and hugging, frequency of “sex” (broadly defined by NSHAP), frequency of vaginal intercourse, and frequency of oral sex. These analyses will be carried out separately by gender, race/ethnicity, and age group; in the case of age, various groupings or “cut-offs” have been used in prior work. We will progress to the use of more complex statistical analyses which will enable us to control for various potentially confounding factors. An important one appears to be relationship status and quality.

Proximity to Kin in Old Age and Ill Health

PI: Felix Elwert, associate professor, sociology
Abstract: Kin continues to be the primary caregivers of elderly and chronically ill Americans. Spouses, children, siblings, and parents offer support with the activities of daily living, maintain contact with healthcare professionals and the wider community, and execute nursing tasks for post-operative home care, disability, or long-term terminal illness. Kin caregiving, especially for long-term support, is greatly aided by physical proximity. Greater physical distance between kin likely hampers kin caregiving and may lead to increased health care expenses as the elderly and infirm are forced to contract caregiving on the open market. This project investigates the determinants of proximity to kin in old age and ill health. It draws on newly geocoded and network-coded data from the Framingham Heart Study (FHS) to examine the determinants of proximity over time.

Veteran Status and Later-Life Health: Evidence from Sibling Fixed-Effects Models

PI: Pamela Herd, associate professor, sociology
Abstract: There are numerous reasons to posit that veteran status is associated with health in later life. However, very few longitudinal studies of health or military service and health are able to disentangle age, period, and cohort effects or follow cohort members over very long periods of time. This pilot project plans to use data from the Wisconsin Longitudinal Study (WLS) to estimate sibling fixed-effects models of the association between veteran status and various mid- and later-life health outcomes among men, as well as among their spouses and children. We plan to expand this pilot project for a R01 grant application to NIA.

Trends and Determinants of Mortality Disparities between the U.S. and Other High-Income Countries: 1980–2010

PI: Alberto Palloni, Samuel Preston Professor, sociology
Abstract: A 2011 National Research Council report offers compelling evidence supporting the idea that the US life expectancy at age 50 has fallen behind levels attained by peer countries in Western Europe and North America, both among males and females. The most important contributors to the gap between the US and other countries are associated with mortality due to very specific chronic conditions, including cancers and cardiovascular diseases. In turn, the ultimate determinants of these are less associated with medical technology and health care than with concrete individual behaviors, such as smoking, diet and exercise. This project has two objectives. The first is purely descriptive and entails to estimate the size of mortality gaps at various ages between the US and selected peer countries for the period 1975 and 2010. The second objective is to identify determinants of time trends in the gap between the US and other high-income countries. Results from this project on health status across the life cycle appear in chapter 2 of this study called, “U.S. Health in International Perspective: Shorter Lives, Poorer Health”. This study has received extensive media coverage in The New York Times, USA Today, and National Public Radio among other outlets. Alberto Palloni and graduate student Jay Yonker gathered and analyzed the information in Chapter 2 and wrote sections on health status across the life cycle. This study can be downloaded at this National Academy Press site.