Country-specific qualitative synthesis: harvesting pearls or grit?

Article type
Authors
Booth A1, Mshelia S2, Analo C2, Nyakango SB2
1School of Health and Related Research (ScHARR)
2formerly School of Health and Related Research (ScHARR), UK
Abstract
Background: International qualitative evidence syntheses, as commissioned by the World Health Organization (WHO) and other agencies, offer valuable cross-country perspectives on issues such as facility-based delivery and respectful maternity care. Necessarily, multi-context syntheses seek to offer breadth of coverage (across regions, countries, and settings) within an efficient time- and resource-envelope. Such syntheses have two limitations:
1) they may omit eligible studies that offer important country-specific insights; and
2) they may capture the phenomenon of interest incompletely, resulting, potentially, in immature conceptual frameworks.

Objectives: To compare the coverage, quality and contribution of country-specific and multi-context syntheses of facility-based delivery to inform strategic commissioning of qualitative syntheses.

Methods: We conducted two country-specific qualitative syntheses of attitudes of pregnant women to facility-based delivery, for Nigeria and Kenya, to augment and enhance a pre-existing international WHO synthesis through exhaustive retrieval and inclusion of country-specific findings. We conducted searches across local and regional sources - including university repositories - to augment standard bibliographic databases. We used framework synthesis to facilitate comparison between countries and with the international synthesis. We compared yields of included studies quantitatively (numbers of: studies, open access sources, journals with impact factors) and qualitatively (unique identification of or reinforcement of themes).

Results: Yield of studies for both country-specific syntheses was approximately seven times higher than the number of eligible studies for that country in the multi-context synthesis. On closer examination, however, we found that a large proportion of studies post-dated the multi-context search, were from non-impact factor or predatory journals, or were from non-open access sources. Qualitatively, the additional sources added country-specific nuances.

Conclusions: Inclusion of country-specific syntheses enhances the context-sensitivity of findings but at the potential expense of review quality.

Patient/healthcare consumer involvement: Country-specific syntheses extend consumer perspectives beyond multi-context syntheses and may resonate more clearly with local experience.