Person:

Canning, David

Objectives: Population-based HIV testing surveys have become central to deriving estimates of national HIV prevalence in sub-Saharan Africa. However, limited participation in these surveys can lead to selection bias. We control for selection bias in national HIV prevalence estimates using a novel approach, which unlike conventional imputation can account for selection on unobserved factors. Methods: For 12 Demographic and Health Surveys conducted from 2001 to 2009 (N=138 300), we predict HIV status among those missing a valid HIV test with Heckman-type selection models, which allow for correlation between infection status and participation in survey HIV testing. We compare these estimates with conventional ones and introduce a simulation procedure that incorporates regression model parameter uncertainty into confidence intervals. Results: Selection model point estimates of national HIV prevalence were greater than unadjusted estimates for 10 of 12 surveys for men and 11 of 12 surveys for women, and were also greater than the majority of estimates obtained from conventional imputation, with significantly higher HIV prevalence estimates for men in Cote d'Ivoire 2005, Mali 2006 and Zambia 2007. Accounting for selective non-participation yielded 95% confidence intervals around HIV prevalence estimates that are wider than those obtained with conventional imputation by an average factor of 4.5. Conclusions: Our analysis indicates that national HIV prevalence estimates for many countries in sub-Saharan African are more uncertain than previously thought, and may be underestimated in several cases, underscoring the need for increasing participation in HIV surveys. Heckman-type selection models should be included in the set of tools used for routine estimation of HIV prevalence.

We examine the joint distribution of levels of income per capita, life expectancy, and years of schooling across countries in 1960 and in 2000. In 1960 countries were clustered in two groups; a rich, highly educated, high longevity “developed” group and a poor, less educated, high mortality, “underdeveloped” group. By 2000 however we see the emergence of three groups; one underdeveloped group remaining near 1960 levels, a developed group with higher levels of education, income, and health than in 1960, and an intermediate group lying between these two. This finding is consistent with both the ideas of a new “middle income trap” that countries face even if they escape the “low income trap”, as well as the notion that countries which escaped the poverty trap form a temporary “transition regime” along their path to the “developed” group.

Objectives: While the benefits of Seguro Popular health insurance in Mexico relative to no insurance have been widely documented, little has been reported on its effects relative to the pre-existing Social Security health insurance. We analyse the effects of Social Security and Seguro Popular health insurances in Mexico on access to healthcare of older adults, and on financial risk protection to their households, compared with older adults without health insurance. Setting: Secondary data analysis was performed using the 2012 Mexican Survey of Health and Nutrition (ENSANUT). Participants: The study population comprised 18 847 older adults and 13 180 households that have an elderly member. Outcome measures The dependent variables were access to healthcare given the reported need, the financial burden imposed by health expenditures measured through catastrophic health-related expenditures, and using savings for health-related expenditures. Separate propensity score matching analyses were conducted for each comparison. The analysis for access was performed at the individual level, and the analysis for financial burden at the household level. In each case, matching on a wide set of relevant characteristics was achieved. Results: Seguro Popular showed a protective effect against lack of access to healthcare for older adults compared with those with no insurance. The average treatment effect on the treated (ATET) was ascertained through using the nearest-neighbour matching (−8.1%, t-stat −2.305) analysis. However, Seguro Popular did not show a protective effect against catastrophic expenditures in a household where an older adult lived. Social Security showed increased access to healthcare (ATET −11.3%, t-stat −3.138), and protective effect against catastrophic expenditures for households with an elderly member (ATET −1.9%, t-stat −2.178). Conclusions: Seguro Popular increased access to healthcare for Mexican older adults. Social Security showed a significant protective effect against lack of access and catastrophic expenditures compared with those without health insurance.

Background: During the year following the birth of a child, 40% of women are estimated to have an unmet need for contraception. The copper IUD provides safe, effective, convenient, and long-term contraceptive protection that does not interfere with breastfeeding during the postpartum period. Postpartum IUD (PPIUD) insertion should be performed by a trained provider in the early postpartum period to reduce expulsion rates and complications, but these services are not widely available. The International Federation of Obstetricians and Gynecologists (FIGO) will implement an intervention that aims to institutionalize PPIUD training as a regular part of the OB/GYN training program and to integrate it as part of the standard practice at the time of delivery in intervention hospitals. Methods: This trial uses a cluster-randomized stepped wedge design to assess the causal effect of the FIGO intervention on the uptake and continued use of PPIUD and of the effect on subsequent pregnancy and birth. This trial also seeks to measure institutionalization of PPIUD services in study hospitals and diffusion of these services to other providers and health facilities. This study will also include a nested mixed-methods performance evaluation to describe intervention implementation. Discussion This study will provide critical evidence on the causal effects of hospital-based PPIUD provision on contraceptive choices and reproductive health outcomes, as well as on the feasibility, acceptability and longer run institutional impacts in three low- and middle-income countries. Trial registration Trial registered on March 11, 2016 with ClinicalTrials.gov, NCT02718222. Electronic supplementary material The online version of this article (doi:10.1186/s12884-016-1160-0) contains supplementary material, which is available to authorized users.

Background: The immediate postpartum IUD (PPIUD) is a long-acting, reversible method of contraception that can be used safely and effectively following a birth. To appropriately facilitate the immediate postpartum insertion of IUDs, women must be informed of the method’s availability and must be counselled on its benefits and risks prior to entering the delivery room. We examine the relationship between the location and quality of antenatal counselling and women’s acceptance of immediate postpartum IUD (PPIUD) in four hospitals in Sri Lanka. Methods: Data were collected between January 2015 and May 2015. Modified Poisson regressions with robust standard errors are used to assess the relationships between place of counselling, indicators of counselling quality, and PPIUD uptake following delivery. Results: We find that women who were counselled in hospital antenatal clinics and admission wards were much more likely to have a PPIUD inserted than women who were counselled in field clinics or during home visits. Hospital-based counselling had higher quality indicators for providing information on PPIUD, and women were more likely to receive PPIUD information leaflets in hospital locations than in lower-tiered clinics or during home visits. Women who were counselled at hospital locations also reported a higher level of satisfaction with the counselling that they received. Receipt of hospital-based counselling was also linked to higher PPIUD uptake, in spite of the fact that women were more likely to be given information about the risks and alternatives to PPIUD in hospitals. The information about the risks of and alternatives to PPIUD, whether provided in hospital or in non-hospital settings, tended to lower the likelihood of acceptance to have a PPIUD insertion. Counselling in hospital admission wards was focused on women who had not been counselled at field clinics. Conclusions: The study findings call for efforts that improve the training of midwives who provide PPIUD counselling at field clinics and during the home visits. We also recommend that routine PPIUD counselling be conducted in hospitals, even if women have already been counselled elsewhere.

There is strong evidence that, in addition to individual and household characteristics, social interactions are important in determining fertility rates. Social interactions can lead to a multiplier effect where an individual’s ideas, and fertility choice, can affect the fertility decisions of others. We merge all available Demographic and Health Surveys to investigate the factors that influence both individual and average group fertility. We find that in the early phase of the fertility transition the impact of a woman’s education and experience of child death on her group’s average fertility are more than three times as large as their direct effect on her own fertility decision.

We investigate the effects of antenatal maternal vaccination against tetanus on the schooling attained by children in Bangladesh. Maternal vaccination prevents the child from acquiring tetanus at birth through blood infection and substantially reduces infant mortality and may prevent impairment in children who would otherwise acquire tetanus but survive. We follow up on a 1974 randomized trial of maternal tetanus vaccination, looking at outcomes for children born in the period 1975–1979. We find significant schooling gains from maternal tetanus vaccination for children whose parents had no schooling, showing a large impact on a small number of children. Our findings make a case for investments in maternal tetanus vaccination as a method of improving schooling and eventual economic outcomes.

Background Selection bias in HIV prevalence estimates occurs if non-participation in testing is correlated with HIV status. Longitudinal data suggests that individuals who know or suspect they are HIV positive are less likely to participate in testing in HIV surveys, in which case methods to correct for missing data which are based on imputation and observed characteristics will produce biased results.

Methods The identity of the HIV survey interviewer is typically associated with HIV testing participation, but is unlikely to be correlated with HIV status. Interviewer identity can thus be used as a selection variable allowing estimation of Heckman-type selection models. These models produce asymptotically unbiased HIV prevalence estimates, even when non-participation is correlated with unobserved characteristics, such as knowledge of HIV status. We introduce a new random effects method to these selection models which overcomes non-convergence caused by collinearity, small sample bias, and incorrect inference in existing approaches. Our method is easy to implement in standard statistical software, and allows the construction of bootstrapped standard errors which adjust for the fact that the relationship between testing and HIV status is uncertain and needs to be estimated.

Results Using nationally representative data from the Demographic and Health Surveys, we illustrate our approach with new point estimates and confidence intervals (CI) for HIV prevalence among men in Ghana (2003) and Zambia (2007). In Ghana, we find little evidence of selection bias as our selection model gives an HIV prevalence estimate of 1.4% (95% CI 1.2% – 1.6%), compared to 1.6% among those with a valid HIV test. In Zambia, our selection model gives an HIV prevalence estimate of 16.3% (95% CI 11.0% - 18.4%), compared to 12.1% among those with a valid HIV test. Therefore, those who decline to test in Zambia are found to be more likely to be HIV positive.

Conclusions Our approach corrects for selection bias in HIV prevalence estimates, is possible to implement even when HIV prevalence or non-participation is very high or very low, and provides a practical solution to account for both sampling and parameter uncertainty in the estimation of confidence intervals. The wide confidence intervals estimated in an example with high HIV prevalence indicate that it is difficult to correct statistically for the bias that may occur when a large proportion of people refuse to test.

Objective: To examine the association between maternal age at first birth and infant mortality, stunting, underweight, wasting, diarrhoea and anaemia in children in low- and middle income countries. Design: Cross-sectional analysis of nationally representative household samples. A modified Poisson regression model is used to estimate unadjusted and adjusted RR ratios. Setting: Low- and middle-income countries. Population: First births to women aged 12-35 where this birth occurred 12-60 months prior to interview. The sample for analysing infant mortality is comprised of 176 583 children in 55 low- and middle-income countries across 118 Demographic and Health Surveys conducted between 1990 and 2008. Main outcome measures: Infant mortality in children under 12 months and stunting, underweight, wasting, diarrhoea and anaemia in children under 5 years. Results: The investigation reveals two salient findings. First, in the sample of women who had their first birth between the ages of 12 and 35, the risk of poor child health outcome is lowest for women who have their first birth between the ages of 27 and 29. Second, the results indicate that both biological and social mechanisms play a role in explaining why children of young mothers have poorer outcomes. Conclusions: The first-born children of adolescent mothers are the most vulnerable to infant mortality and poor child health outcomes. Additionally, first time mothers up to the age of 27 have a higher risk of having a child who has stunting, diarrhoea and moderate or severe anaemia. Maternal and child health programs should take account of this increased risk even for mothers in their early 20s. Increasing the age at first birth in developing countries may have large benefits in terms of child health.

We analyze the effect of fertility on income per capita with a particular focus on the experience of Europe. For European countries with below-replacement fertility, the cost of continued low fertility will only be observed in the long run. We show that in the short run, a fall in the fertility rate will lower the youth dependency ratio and increase the working-age share, thus raising income per capita. In the long run, however, the burden of old-age dependency dominates the youth dependency decline, and continued low fertility will lead to small working-age shares in the absence of large migration inflows. We show that the currently very high working-age shares generated by the recent declines in fertility and migration inflows are not sustainable, and that significant drops in the relative size of the working-age population should be expected. Without substantial adjustments in labor force participation or migration policies, the potential negative repercussions on the European economy are large.