Including mixed methods research in systematic reviews: examples from qualitative syntheses in TB and malaria control.

BACKGROUND: Health policy makers now have access to a greater number and variety of systematic reviews to inform different stages in the policy making process, including reviews of qualitative research. The inclusion of mixed methods studies in systematic reviews is increasing, but these studies pose particular challenges to methods of review. This article examines the quality of the reporting of mixed methods and qualitative-only studies. METHODS: We used two completed systematic reviews to generate a sample of qualitative studies and mixed method studies in order to make an assessment of how the quality of reporting and rigor of qualitative-only studies compares with that of mixed-methods studies. RESULTS: Overall, the reporting of qualitative studies in our sample was consistently better when compared with the reporting of mixed methods studies. We found that mixed methods studies are less likely to provide a description of the research conduct or qualitative data analysis procedures and less likely to be judged credible or provide rich data and thick description compared with standalone qualitative studies. Our time-related analysis shows that for both types of study, papers published since 2003 are more likely to report on the study context, describe analysis procedures, and be judged credible and provide rich data. However, the reporting of other aspects of research conduct (i.e. descriptions of the research question, the sampling strategy, and data collection methods) in mixed methods studies does not appear to have improved over time. CONCLUSIONS: Mixed methods research makes an important contribution to health research in general, and could make a more substantial contribution to systematic reviews. Through our careful analysis of the quality of reporting of mixed methods and qualitative-only research, we have identified areas that deserve more attention in the conduct and reporting of mixed methods research.

Malaria, danger, and risk perceptions among the Yao in rural Malawi.

Findings from a study designed to discover how local understanding of malaria among Yao in Malawi relate to pregnancy risk definitions reveal that malaria in pregnancy is not perceived as a major risk. Using extended ethnographic field research and multiple methods, we argue a shift from narrow single-disease approaches to malaria during pregnancy is required and document women's concerns about exposure to multiple vulnerabilities during pregnancy, including witchcraft, extramarital affairs, and multiple dangerous illnesses. Four dimensions are implicated in Yao perceptions of risk: perceived adverse consequences in pregnancy; ease of treatment and cure; transmission and agency to control; and type of risk (social-medical). We discuss implications and consider malaria program features needed to address the complexity of perceived vulnerabilities and living conditions in resource-poor settings.

Ethnographic study of factors influencing compliance to intermittent preventive treatment of malaria during pregnancy among Yao women in rural Malawi.

In Africa today one of the main strategies to reduce malaria infection during pregnancy is the promotion of intermittent preventive treatment (IPT). To date only a few studies have investigated the factors affecting compliance to IPT. This medical anthropology study aims to describe these factors from the perspective of pregnant women in rural Malawi. We examine women's knowledge and perceptions about the use of medication in pregnancy and the timing and motivation concerning use of antenatal clinic (ANC) services. In addition, the circumstances and interaction at the ANC and the IPT implementation process are described. The data were collected by applying an ethnographic approach, including focus group discussions (n=8), in-depth interviews (n=34), drug identification exercises, participant observation and a 'knowledge, attitudes and practices' survey (n=248). This study discovered several factors affecting IPT. These were: unclear messages about IPT with sulfadoxine-pyrimethamine (SP) from nurses; timing of SP-1; periodic shortages of SP; women's limited understanding of IPT-SP; tendency for late enrolment; and nurses' underperformance. The results of this study show that understanding of the multiple contexts affecting malaria prevention is important, and that ethnographic research is useful for discovering and solving problems beyond the scope of many other research approaches.

The importance of understanding the local context: women's perceptions and knowledge concerning malaria in pregnancy in rural Malawi.

A current problem of malaria prevention programmes is that not enough attention is paid to understanding the local socio-cultural context prior to programme implementation. The aim of this study is to discover how Yao women in rural Malawi understand and explain malaria in pregnancy, how they perceive it and what type of knowledge they have on it. Women's knowledge of the adverse effects of malaria in pregnancy is also investigated. At first phase a total of 34 in-depth interviews were conducted. At second phase a KAP survey (n=248) was conducted for cross-validation of the qualitative information. The findings showed that there is neither a vernacular word for malaria nor malaria in pregnancy. Women used a local word, malungo, to refer to malaria. Malungo is an ambiguous disease term because of its multiple meanings which are used interchangeably to refer to many types of feverish illnesses of various causes, not only malaria. Most women did not perceive malungo during pregnancy as a serious illness. There were several other diseases from anaemia, STDs to cholera etc. that were perceived to be more dangerous than malungo. The local meaning of malungo also entailed an assumption that it is a common but fairly harmless illness. Women had limited knowledge of the adverse effects of malaria in pregnancy, the best-known adverse effect being miscarriage (28%, 52/189). A socio-cultural understanding of the implementation context is prerequisite for planning meaningful programmes for the pregnant women in rural Africa.