Home > Department of Social Policy > British Society for Population Studies > Annual conference > 2008 Conference - Manchester > BSPS Annual Conference 2008: Reproductive Health Strand - Abstracts


BSPS Annual Conference 2008: Reproductive Health Strand - Abstracts

A cross-national analysis of factors associated with HIV/AIDS infection in sub-Saharan Africa: Evidence from DHS.

Monica Magadi, Muluye Desta
City University, London

The scale and impact of HIV/AIDS on various aspects of development throughout the world (particular in sub-Saharan Africa) are staggering. Efforts to address the problem of HIV/AIDS and its devastating impact require sound understanding of factors associated with HIV/AIDS in different contexts. This study examines factors associated with HIV infection based on secondary analysis of existing data from the international Demographic and Health Surveys (DHS) programme. The analysis uses recent DHS data from 16 countries in sub-Saharan Africa where HIV/AIDS test data are available. The key outcome variable of interest is HIV/AIDS infection while explanatory factors include:

  • individual characteristics, including demographic characteristics (gender, age), socio-cultural factors (e.g. religion, ethnicity, male circumcision, type of union), sexual behaviour (age at first sex, number of sex partners, condom use), and HIV/AIDS awareness;
  • household factors, including household composition, and socio-economic status; and
  • contextual regional factors at country and community/cluster within country level.

Most contextual factors are derived from the individual level information based on mean indices or the proportion of the population in the community/country with characteristics of interest. The analysis is based on multilevel modelling, and places particular emphasis on country and community variations in factors associated with HIV/AIDS, and the extent of clustering of HIV infection within communities (districts/ communities within country). We highlight inequalities in HIV infection by mapping the residual spatial effects using geo-additive probit models (via Bayesian approach) that simultaneously control for spatial dependence in the data and potential nonlinear effects of covariates.

Email: m.magadi@city.ac.uk|

Contextualising condom (non-)use by female sex workers in rural and urban Indonesia: Innovative sampling and research methodologies for a hard-to-reach population

Dewi Ismajani Puradiredja,
London School of Economics and Political Science

Context-specific knowledge is essential for the design and implementation of more effective HIV/AIDS policy and intervention strategies. Little is known about barriers to condom use with clients among different sub-groups of Indonesian female sex workers (FSWs), particularly among rural FSWs. Firstly, this study identifies which factors inhibit condom use with clients among FSWs with different background and individual characteristics, and sex work settings. Secondly, it investigates the ways in which the different sex work settings and individual characteristics interrelate with condom (non-)use by FSWs.

This study uses a comparative rural-urban research design and a mixed methods approach to collect and analyse both qualitative and quantitative data related to condom use in the context of transactional sex by FSWs in Indonesia. Primary data collection include a structured survey of urban and rural FSWs (n=310) in Indonesia, and in-depth interviews with a sub-sample (n=12). A novel three-staged purposive sampling technique enabled the inclusion of notoriously under-researched sub-populations, such as rural FSWs and FSWs in urban slum areas with high rates of crime and violence.

Results show how variations in sex work setting in combination with respondents' demographic, socio-economic, psychosocial and personal characteristics create and relate to different inhibitors to condom use with clients, as opposed to HIV/STI awareness alone. Condom use among FSWs is determined by factors such as type and pricing of sexual services, gender power relations, economic pressure, social marginalisation, harassment-prone work settings, unsupportive sex workplace management, and familiarity with clients.

Email: d.i.puradiredja@lse.ac.uk|

Trends in HIV/AIDS knowledge among men and women in China, 1997-2006

Olga Maslovskaya1, Sabu Padmadas1, Peter W. Smith1, James Brown2
1University of Southampton, 2Institute of Education, University of London

HIV/AIDS prevalence in China is currently less than one percent, but due to the large population even one percent translates into a large number of people. The absolute number of people living with HIV/AIDS is rapidly growing, beyond high-risk groups to the general population. Given the limited scope of any potential cure for HIV, prevention can play a crucial role in controlling the epidemic. Ensuring adequate knowledge is important for the successful prevention of HIV as detailed knowledge may effectively help reduce risky behaviours. Political commitment is yet another crucial factor in the fight against HIV/AIDS. In 2003, the Chinese government reacted to the evidence of spread of HIV/AIDS in China and announced that the country is grappling with a possible HIV epidemic outbreak. This paper focuses on development of HIV/AIDS knowledge among both men and women in the Chinese context. The aim of this paper is to compare the levels of HIV/AIDS knowledge at various time periods between 1997 and 2006 by gender and region. For the analysis, we use data from various sources: the UNFPA Reproductive Health and Family Planning Surveys, China National Family Planning and Reproductive Health Surveys, and the Chinese Health and Family Life Survey. Changes in levels of HIV/AIDS knowledge in different data sources can be attributed to the time at which the data were collected, different sampling techniques, different representation levels, as well as changes in the governmental agenda and the effect of programmes, interventions and campaigns.

Email: om206@soton.ac.uk|

The correlates of natural method use in Moldova; are natural methods associated with poverty and isolation?

Mark J. Lyons-Amos, Gabriele B. Durrant, Sabu S. Padmadas
University of Southampton

The Republic of Moldova demonstrates the highest reliance on natural contraceptive methods in Europe, and natural methods account for an increasing proportion of contraceptive mix since 1997. This study investigates method use correlates in Moldova in order to explain this high reliance on natural methods.

The study tests hypotheses that current natural method use is associated with economic disadvantage and spatial isolation. The impact of national Family Planning (FP) is also examined by testing the significance of FP media exposure. The study controls for the effect of fertility preferences and coital activity on odds of natural method use. Data are drawn from the MDHS 2005- the first ever European DHS. The study population consists of 5860 women who had first sex, and were not pregnant. The effect of correlates is assessed using a multilevel, multinomial logistic regression model, accounting for cluster level effects.

Findings support the hypothesis that economic disadvantage is associated with natural method use. Region and rural residence increase the probability of natural contraceptive use- although FP clinic access has no direct effect. Exposure to FP media decreases the odds of no method use, but the effect declines with age and there is no significant effect on natural method use. Women unaware of how to avoid AIDS are significantly more likely to be non-users.

This study recommends that FP effort should be directed toward the poorest women by ensuring contraception is affordable and appropriate, and enhancing existing FP media to focus on women unaware of AIDS transmission routes.

Email: mja303@soton.ac.uk|

Mobile clinics and vasectomy rates in Nepal

Sabu S. Padmadas1, Tiziana Leone2, Fiifi Amoako Johnson1, Govinda Prasad Dahal3
1University of Southampton, 2London School of Economics, 3Centre for Nepal Studies UK

The demand for permanent methods has grown over time in Nepal especially in rural areas where the levels of unmet need for contraception is very high and increasingly many couples desire for small families. It is in this context the Government of Nepal launched the Tenth Development Plan (2002-2007) which prioritised the expansion of family planning services particularly in remote districts where both quality and access to services are poor.1 An important intervention in this regard is the launch of mobile family planning camps to promote modern methods among socially/spatially disadvantaged communities. Using the 2006 Nepal DHS data,2 we demonstrate evidence of the influence of mobile clinics in reducing the gender gap in the use of permanent methods. We used multilevel binary logistic regression model to capture the effect of source of current method use on sterilisation choices (male3 versus female), controlling for individual and health care variables. The results show that the
likelihood of vasectomy uptake is significantly likely if the method is offered through a mobile clinic when compared to a sterilisation in a government hospital or nongovernmental sources. Further investigation of the DHS data show that those who had a vasectomy from mobile clinics are also less likely to be informed of other method options or side-effects. The findings clearly suggest the persuasive efforts of mobile clinics in promoting male sterilisation in Nepal, which to some extent reduces the gender gaps in sterilisation use but at the expense of men/couples being poorly
informed of their method choices.

Email: s.padmadas@soton.ac.uk|

Conversations about contraceptives: Qualitative evidence of fear of side effects as a barrier to modern contraceptive use among Ghanaian women

Claire Bailey
University of Southampton

Quantitative studies have highlighted that fear of side effects acts as a barrier to the use of modern methods; however this remains an ill defined and poorly understood concept. The overall contraceptive prevalence rate is relatively low in Ghana at 20.7% for all women aged 15-49 in 2003. According to the 2003 Ghana Demographic and Health Survey (GDHS), fear of side effects was the most cited method-related reason for non-use among all women who are not currently using contraception and say they do not intend to do so in the future. Fear of side effects has increased in importance as a reason for non-use between 1998 and 2003, from 18% to 26%.

This study uses data from focus group discussions to explore in greater depth the concept of fear of side effects and to determine on what information and from what sources is this fear constructed and reveal how the individuals concerned understand and articulate these issues. The theoretical framework within which this study is situated is diffusion of innovation theory. The key explanatory element is the spread of information about family planning impersonally through the mass media or inter-personally through discussion between members of a social network.

The results show that much of the information about family planning passed through informal social networks is negative and is about the potential side effects of contraception. Most of this information is based on someone's personal experience of expected side effects and therefore cannot be easily dismissed as 'rumour' or 'misinformation'.

Email: ceb504@soton.ac.uk|

The influence of husbands on the contraceptive use of women in Nepal

Nashid Kamal1 and Cindy Lim2
1Independent University, Bangladesh, 2University College London

This study investigates the influence of 'husband's opinion of family planning' on the current and future use of modern contraception of couples in Nepal, net of other confounders. Using the Nepal DHS (2006) it constructs two logistic regression models, one for users of sterilization and the other for users of modern reversible methods. Continuous explanatory variables are tested for nonlinearity and linear splines are fitted in both models emphasizing the use of the right functional form. Results show that 'husband's opinion of family planning' is the most significant predictor of use of modern reversible methods, while 'number of living sons' have higher predictive power for the sterilization model. For those intending to use in future, husband's positive opinion of FP, results in five times higher use. Women in the worst socio-economic quintiles have higher probability of future use. Gender equity and poor couples are the future challenges in Nepal.

Email: nashidk@gmail.com|