Type of Document Dissertation Author Kasongo, Webster URN etd-08052013-143758 Title Age-Period-Cohort and Educational Attainment Effects on HIV Prevalence in Zambian Pregnant Women, 1994 through 2011 Degree PhD Department Epidemiology Advisory Committee
Advisor Name Title Sten H. Vermund, MD, PhD Committee Chair Bryan Shepherd, PhD Committee Member Marie Griffin, MD, MPH Committee Member Mary Lou Lindegren, MD , MPH Committee Member Keywords
Date of Defense 2013-07-12 Availability unrestricted AbstractAge-Period-Cohort and Educational Attainment Effects on HIV Prevalence in Zambian Pregnant Women, 1994 through 2011 Webster Kasongo
Dissertation under the direction of Professor Sten H. Vermund Sub-Saharan Africa, where Zambia is situated, accounted for 69% of the 34 million people living with HIV worldwide. An estimated one million people were living with HIV in Zambia in 2011 compared to 1.3 million people in the United States of America, with 24 times the population of Zambia. Prior research has suggested declining HIV prevalence trends in Zambia, but whether the decline in HIV prevalence has been influenced by age, period or cohort effects has not been examined contemporaneously. Birth cohort and period variations in HIV prevalence may signal changes in distribution of risk factors for HIV infection, and analyses of age, period and birth cohort’s influence on HIV prevalence may provide key information for focusing HIV interventions.
HIV sentinel surveillance data sourced cross-sectionally, from pregnant who sought antenatal care in 1994, 1998, 2002, 2004, 2006, 2008 and 2011 in Zambia were analyzed to describe HIV prevalence trends in 15 to 24 year olds; to examine association of educational attainments with HIV prevalence; and to investigate age, period and birth cohort effects on HIV prevalence among 15 to 44 year-olds using the cross-classified random effect model proposed by Yang and Land (2006).
HIV prevalence declined from 27.0% (1994) to 14.7% (2011) and from 10.0% (1994) to 7.4% (2011) in urban and rural sites respectively. Compared to the expected odds of prevalent HIV in this population, birth cohort effects peaked among pregnant women in the 1970-1974 birth cohort in urban sites (OR=1.41, 95% CI: 1.34, 1.49) and in the 1975-1979 birth cohort in rural sites (OR=1.29, 95% CI: 1.19, 1.38). Significantly lower odds of prevalent HIV infections were noted among pregnant women in birth cohort 1985-1989 (OR=0.79, 95% CI: 0.74, 0.84) and birth cohort 1990-1996 (OR= 0.68, 95% CI: 0.60, 0.76) in urban sites. Compared to 24 year-olds, age effects were elevated most for 26 year-olds (OR=1.13, 95% CI: 1.02, 1.25), and protective for 15 year-olds (OR=0.31, 95% CI: 0.29, 0.33) and 19 year-olds (OR=0.57, 95% CI: 0.56, 0.62).
The lower odds of prevalent HIV infections among younger generations (1985 to 1996 in urban sites may imply a possible fall in HIV incidence, and calls for preventive interventions that emphasize forestalling risky sexual behaviors among young people in order to suppress new HIV infections.
Filename Size Approximate Download Time (Hours:Minutes:Seconds)
28.8 Modem 56K Modem ISDN (64 Kb) ISDN (128 Kb) Higher-speed Access kasongo.pdf 17.27 Mb 01:19:57 00:41:07 00:35:59 00:17:59 00:01:32