Reviews of qualitative evidence: a new milestone for Cochrane

The Cochrane Collaboration

Systematic reviews addressing a wide range of healthcare questions, and drawing on a range of different study designs, are increasingly available in the literature. The Cochrane Database of Systematic Reviews currently considers Cochrane Reviews on the effectiveness of health interventions and the accuracy of screening and diagnostic tests, as well as overviews of reviews and methodology reviews. November 2013 marks an important milestone for The Cochrane Collaboration with the publication of a review of qualitative studies. This synthesis of qualitative evidence addresses barriers and facilitators to the implementation of lay health worker (LHW) programmes. Findings from this synthesis have been integrated with outcomes from the Cochrane effectiveness review on the use of LHWs in primary and community health care for maternal and child health, providing a comprehensive assessment of this strategy. Both reviews were developed by the Cochrane Effective Practice and Organisation of Care Group (EPOC; epoc.cochrane.org).

The synthesis of qualitative evidence makes an important contribution to the knowledge available to organisations such as the World Health Organization (WHO) when developing international recommendations on public health topics. Organisations need to consider whether their recommendations, especially on issues related to the organisation of care, are likely to be feasible and acceptable to care providers and communities, and they need to consider the challenges for their implementation.

The use of LHWs to deliver effective interventions is of current interest globally. Many low‐, middle‐ and high‐income countries have introduced or are considering introducing such programmes to increase access to key interventions in an affordable way. The impetus for conducting this synthesis of qualitative evidence came from WHO's OptimizeMNH recommendations, which focus on optimizing the delivery of key maternal and newborn health interventions through task shifting in low‐ and middle‐income countries. Task shifting to LHWs comprises a key component in these recommendations, and many countries are considering LHWs for the delivery of a range of key interventions. For some of these interventions, delivery by LHWs has been controversial. For example, the administration of misoprostol (an inexpensive uterotonic in tablet form) to reduce postpartum haemorrhage by LHWs has been discussed by the global maternal health research community for more than a decade.

The OptimizeMNH recommendations used the newly developed DECIDE framework (www.decide‐collaboration.eu) for the decision‐making process, moving from evidence to recommendations. DECIDE is an extension of the GRADE system, required by the WHO guideline development process, and provides a more structured assessment of issues such as acceptability, feasibility, resource use, and other implementation considerations.

The qualitative synthesis proved to be critical in the formulation of the OptimizeMNH guidance recommendations as it highlighted facilitators and barriers to the implementation of LHW task‐shifting programmes in different settings and across different types of interventions. When the Cochrane effectiveness review provided evidence of effectiveness for the shifting of a particular task, the qualitative evidence complemented this evidence. For example, task shifting enabled delivery of care by health workers from the same community, and this was found to be feasible and acceptable for most practices reviewed. Also, and somewhat unexpectedly, the qualitative evidence allowed the guideline panel to make more informed decisions regarding a recommendation where direct evidence of effectiveness was lacking, by facilitating consideration of the barriers and implementation bottlenecks observed for other tasks. The findings of the qualitative evidence synthesis also usefully reduced the need to rely on the personal anecdotes and experience of guideline panel members, for issues of feasibility, acceptability, and implementation.

One of the questions examined by the OptimizeMNH guideline was whether LHWs should administer misoprostol to women in the community or in primary health care settings to prevent postpartum haemorrhage following delivery. The Cochrane effectiveness review found no direct evidence for using LHWs in this way, but it did identify indirect evidence from a number of trials from low‐ and middle‐income countries where packages of care were delivered by LHWs. In some of these trials, the packages included the provision of antibiotics to sick newborns and of antimalarials to children, and no adverse effects were reported. The Cochrane qualitative evidence synthesis noted several acceptability and feasibility considerations that informed the guideline recommendation. For example, activities that demand that the LHW is present at specific times (e.g. during labour and birth) lead to irregular and unpredictable working conditions. This may have direct implications for LHWs' expectations regarding incentives. The review also highlighted that LHWs may also be concerned about personal safety when working in the community and or when visiting clients at night. Based on this evidence of effectiveness, feasibility, and acceptability, the guideline panel decided to recommend the use of LHWs to administer misoprostol to prevent postpartum haemorrhage where a well‐functioning LHW programme already exists. The panel noted that this intervention may be feasible under certain conditions and may reduce inequalities by extending care to underserved populations.

This qualitative evidence synthesis development within the Cochrane framework and its use within WHO guidance provides an exemplar for future syntheses. The authors provide guidance for dealing with some of the methodological challenges in conducting these types of syntheses. For example, a logic model was used to guide the integration of effectiveness and qualitative data. Here, the components of a LHW programme were identified and then linked to intermediate and long‐term outcomes, illustrating how the intervention might have an impact on outcomes and how effectiveness might be moderated. For example, LHW credibility could be undermined by a poorly functioning health system, which could negatively impact the use of services. Secondly, the authors developed a system for assessing the certainty of the evidence for each of the qualitative evidence synthesis findings. This followed a similar conceptual approach to that used within GRADE. Each synthesis finding was assessed according to the extent that it was seen across multiple and diverse settings, and each was of methodological good quality. This approach indicates our confidence in the findings of the qualitative synthesis and also provides an indication of the extent to which these findings can be applied to other settings. The authors, together with the GRADE Working Group (www.gradeworkinggroup.org), the Cochrane Qualitative and Implementation Methods Group (cqim.cochrane.org), the WHO Department of Reproductive Health and Research (www.who.int/reproductivehealth), and collaborators across a range of institutions, will continue to develop methods for assessing the certainty of qualitative evidence and have established the CerQUAL (Certainty of Qualitative Evidence) working group.

The close collaboration and co‐operation between Cochrane and WHO in developing guidance useful in the field, linking those generating the evidence with those using it, illustrates the importance of methodological development to encompass a range of evidential data in systematic reviews. On its 20th anniversary, Cochrane continues to provide not only robust evidence synthesis for healthcare interventions but also pushes methodological innovations in healthcare evaluation.

Feedback on this editorial and proposals for future editorials are welcome.

End Notes

Declarations of interest

The authors have completed the ICMJE form for disclosure of potential conflicts of interest (forms available upon request). AMG and SS declare no conflicts of interest. JC declares that a review referred to in the editorial is being used for methodological work under a project funded by The Cochrane Collaboration (Methods Innovation Fund) for which, she, in a supportive role is a co‐applicant, and that she is the Methods Co‐ordinator and an employee of The Cochrane Collaboration. TP declares that he is an editor with the Cochrane Effective Practice and Organisation of Care Review Group and has professional relationships with the authors of a review referred to in this editorial.