Findings

Overall statement or summary

This study aims to explore women’s perceptions of maternal healthcare services delivery in three rural district hospitals and their correspondent facilities in Malawi. A qualitative design with focus group discussions was used. Findings showed that prenatal care services were more appreciated by the women as opposed to intra and postpartum services. Barriers to services uptake comprise cultural beliefs, husband non-involvement, and mistreatment or abuse by personal in intra and postpartum services. Women also complained about congestion in the services, lack of drugs and medical equipment, poor facility infrastructure, and costs related to distance. To address these health system issues, more studies are required, including the exploration of health personnel’s challenges and aspirations. Advocacy for maternal healthcare to get financing and for women to be linked to care could also be interesting.

Strengths

Originality: The Lancet Maternal Health Series call to action prioritizes (i) good quality maternal health services that respond to local needs and (ii) improvement of the availability and use of local evidence to inform quality improvement. This investment in locally-driven implementation research will help bridge the gap between national consensus guidelines and clinical practice, thus identifying strategies to shorten timeline between the latter (Koblinsky M, Moyer C, Calvert C, Campbell J, Campbell O, Feigl A, Graham W, Hatt L, Hodgins S, Matthews Z, McDougall L, Moran A, Nandakumar A & Langer A. Quality maternity care for every woman, everywhere: a call to action. Lancet 2016) (C. Lenfant. Shattuck lecture--clinical research to clinical practice--Lost in Translation? N Engl J Med. 2003 Aug 28;349(9):868-74. http://www.aacp.org/meetingsandevents/pastmeetings/Documents/Lenfant%20C%20Clinical%20Research%20to%20Clinical%20Practice.pdf). This article is trying to respond to such call to action, and could be considered as priority research in Maternal Healthcare. Recommendations will rely on how local women perceive the services they are taking, thus bringing a first insight on implementation issues regarding Maternal Healthcare services in Malawi. Furthermore, a lot of other useful research could be drawn out from the results.

Title: Title is informative, engaging and relevant.

Study design: The Qualitative design of the study (FGD) makes the results more descriptive and more generalizable (The Health Foundation Inspiring Improvement. Measuring Patients’experience. N°18. Evidence Scan. June 2013. P:3-25). It also fits the aim of the study pretty well.

Results: Data analysis is present. Data is presented in appropriate way (description then quotes). Text adds to the data. Results address the research question within the remit of what authors have done.

Ethical considerations: An ethical number from North West University was given, and ethical approval has also been given by health district offices. I declare no conflict of interest as a reviewer.

Weaknesses

Abstract: The Abstract does not really match the rest of the article.

Introduction: The background in the Abstract and the Introduction Section of the manuscript do not comprise enough information on the concept of “quality”. Thus, readers can neither capture the substance of the study nor appraise the rationale for the study. There is also too much redundancy in the Introduction, which makes again the rationale or the justification of the study difficult to capture.

Methods: The reproducibility of the study could be difficult from the information provided in the Methods section. Furthermore, external validity of the study may be impaired with the sample quality.

Results: It is not clear about what is a practically meaningful result. The third question of the FGD remained unanswered (nothing about women’s suggestions in the results section).

Discussion: The authors have not made clear at the start of the discussion what the study adds to the existing literature. Besides, the discussion part re-takes too many pieces of the results, not their summary only each time. The suggestions/recommendations seem to be too straightforward: yes they are based on the results of the study, but the results do not seem enough to state such recommendations and assume these would fit to address the issues.

I am giving details of these comments below for the authors to respond to.

Major points

1. Abstract : The background does not specifically state why QUALITY is important and whether the “promotion” by many stakeholders relates to QUALITY indeed. The objective of the study does not seem to be justified when one reads the background in the abstract. There seems to be a leap between background and aim. As if the authors assumed the readers naturally understand quality has to be one of the issues for not using services. And from the background, the reader would expect an aim like “Determining factors associated with maternal services attendance/uptake” where elements of quality would be seen in the results amongst other factors. --> So perhaps it would be better to introduce the importance of quality of services if the authors want to keep the statement of the aim they have now.

2. Introduction 2.1. It is not clear what is already known about this topic: definition of quality here, with the nature of improvements and their (expected) impacts have not been clearly underlined.

3. Methods : 3.1. We do not have enough details to ensure reproducibility of the study. First on STUDY SETTINGS : - Page 27 Paragraph 1 : “As such, the study conveniently selected these district hospitals and correspondent health centers to investigate the perspectives of women…” --> How many correspondent health centers were there? which of them were BEMonC? which were CEmONC? which kind of Maternal Care Services do they comprise [ANC? Delivery? Postpartum? Vaccination? Contraception?]; were these public/private settings?

        - Page 27 Paragraph 2: “…reside within the radius of 8 km to the health care facilities”
          --> It should be made clear what the radius of 8 km implies in the context of Malawi (reachable on foot??? Does it mean no matter what the nature of the roads is or what geographical barrier might be like rivers to cross or hills to climb, the radius is a good index to show geographical accessibility? What does this radius add to the Methodology of the study?
   

   3.2. STUDY POPULATION: - Page 28 Paragraph 1 : “There were six focus group discussions (FGDs)” --> Authors should further explain whether they have conducted a FGD per selected facility? Whether there has been a FGD per district hospital and per correspondent health center?

       - Page 28 Paragraph 2 : “Women who were identified and selected”
         --> When did the study take place? And which period have you studied for each participant (experience during last 12 months for instance or during last 5 years???)
        --> Inclusion criteria : Any specific wards/services were participants were approached? How were they selected (age? Person who accompanies a mother or must be the careseeker? Residents in the area for one year for instance or just passers by? A certain number of visits to the facility was required to include the woman? Was there an incentive as all women approached accepted to participate?)
   

   3.3. I have a concern about the external validity of the study as well : do the selected women really represent the population of Malawi which they come from?

   3.4. Content validity of the tools used to collect data : - Focus Group Guide : about the three core questions. --> It would be more suitable if authors have explained in the introduction why they think those three core questions really address the issue of quality or are related to it? For instance, they may have provided hypothesis on the absence of services uptake in Malawi, and provided the appropriate references why they think so. - Focus Group Guide : Question #2 “explain on the quality of maternal health care services delivery by the health facilities” --> Authors should give an operational definition of “quality” here as this question could be vague and its answers too subjective. Or else were there specific prompts or explanation given to women about the “quality” which was supposed to be measured prior to the FGD?

   3.5. Page 27 : STUDY LIMITATIONS - Can the authors give a reference such as women talking to men about their health and womanish issues does not constitute a taboo in Malawi? So that readers may know women might have participated their best without giving false information due to male superiority, thus not compromising the results of the study? - And has anything been done to address that issue prior to conducting the FGDs?

   3.6. The variable "perspectives" should have been defined so that readers can see whether it has been measured appropriately (eg: same as "perceptions"? or "expectations" for the future to see things improve as FGD guide question #3 seems to say).

4. Results: 4.1. In general, I am not sure about what is a practically meaningful result.
   As I have outlined in the abstract comments, all of the findings seem to have the same “weight”. There was no quantitative analysis in it such as 80% of participants agreed that distance is an issue, before providing a quote to illustrate the matter as a qualitative element of the finding. So can the authors explain why they have only emphasized on erratic medicines and equipments and personnel's 'unethical' attitude?

   4.2. The third question of the FGD remained unanswered (I think I have read nothing about it in the results section). This despite the fact that the authors have outlined on Page 27 Paragraph 1 of Results Section that suggestions made aiming at improving delivery of maternal health services in Malawi were also found.

5. Discussion and conclusion: 5.1. The authors have not made clear at the start of the discussion what the study adds to the existing literature.

   5.2. Page 31 paragraph 1 : “For instance, some women reported that they were being yelled by the health care personnel” --> I think no quotes included the word “yelling” in the results section. Hence, the authors should not add to the results or try to increase the emotional shock they might have felt or want to convey through the results.

   5.3. “they require pro-women-based interventions…” --> This is not well referenced. Authors should have outlined what current studies say about mistreatment and abuse during childbirth. There is a lot of literature on the matter. Besides, further studies on mistreatment might have been suggested. I think knowing the perceptions and challenges of the services personnel also matters before bringing up solutions based on women’s perceptions only. Personnel is also on the frontline of services and its quality; thus, their perspectives might be of great importance as well. Same for husbands’ perceptions and challenges.

   5.4. “Pro-women community-based interventions help change women’s perspectives” - What are these? Then it might be more interesting to state who exactly can do what in how much time (based on references). If that is impossible with current results and knowledge, authors should suggest further community-based studies (eg. socio-anthropological studies).

   5.5. “Such situations promoted women use of home based care.” --> Here, the authors should say that such situations promoted women use of home-based care in TANZANIA (according to the reference provided). If not, it seems the situation is necessary the same in Malawi. Whereas in a quote a woman just outline the use of other services, not home-based care per say.

   5.6. Too much redundancies on intra and postpartum care. The finding should have been communicated once, followed by all the adequate and related discussion (pages 31, 32). --> Perhaps the authors might have followed the same plan as the main findings to discuss them and avoid taking back and forth the same results.

   5.7. pages 32-33 : “Such unethical and inhumane attitude of the health workers, which is often times scantly cautioned, is cited as a major”: no related reference to this statement has been found. --> Although readers can guess this is unethical, all statements must be evidence-based.

   5.8. page 33 : What do you suggest to address the issue of male involvement?

Minor points

1. Abstract : 1.1. There is a redundancy in the aim and the sentence immediately following it. Authors should only keep one of them.

   1.2. The results in the Abstract do not match with the rest of the article. Results seem to just focus on what have shocked the authors the most. Nevertheless, I have read some other results which apparently have the same “weight” as erratic availability of medical resources and unethical practices” (cultural beliefs hindering disclose of pregnancy, support of husband, distance/transport/drugs costs). So I would suggest the authors outline all important findings, unless they explain in the data analysis section of the methods that they have had a way of pondering/weighing/scaling the results such as some results would matter more than others.

   1.3. Besides, the conclusion only addresses what the chosen results have shown and does not seem to really be specific to those issues. It might have been more suitable/interesting to outline that further studies were required (understanding staff attitude for instance) before giving recommendations that might not be so relevant with present knowledge. Unless we have all key elements from all key persons on the frontline of services, I think we would not be able to give solutions based on the women's points of view only.

2. Introduction: 2.1. Just about the structure of the writing, it would be logical to make a “funnel” approach : speak of what is global/general first, then of what happens in Africa, and finally about Malawi. That way maybe reading the introduction would be easier, more fluent.

   2.2. Page 25 paragraph 1: First, what are the improvements in quality of maternal healthcare delivery? I think readers outside of this topic/field should be able to understand the article in a way. So it might be relevant to: - define what quality is and/or how the authors choose to define it (eg : quality encompasses structures like facilities, processes like the provision of care, and outcomes like …). - before getting on the examples about neighboring countries such as Mozambique, Zambia and Tanzania and the use of services, the benefits of quality improvements should have been made clear and referenced. For instance, quality improvement should result in increase of use of services, decrease in use of inappropriate or unnecessary services and cost-effectively decrease negative outcomes. I would recommend the following references for both comments: . Tuncalp O, Were WM, MacLennan et al. Quality of care for pregnant women and newborns – the WHO vision. 2016. Available from: http://onlinelibrary.wiley.com/doi/10.1111/1471-0528.13451/full#bjo13451-fig-0001 . Miller S, Abalos E, Chamillard M. et al. Beyond too little, too late and too much, too soon: a pathway towards evidence-based, respectful maternity care worldwide. 2016. Available from: http://www.thelancet.com/pdfs/journals/lancet/PIIS0140-6736(16)31472-6.pdf

Second, what are the worst maternal health outcomes outlined here? It might be simpler to cite them, so that readers can fully appraise their severity as well as the importance of quality improvement.

 2.3. Page 26 three last lines of last paragraph of the Introduction:

Yes, somehow while readers get to know that efforts have been made yet services uptake is not satisfactory, the following might be another way to phrase the research question before stating the aim of the study: “Do quality improvement in Maternal health Services in Malawi align with women’s aspirations/expectations at local level?’

1. Conclusion: 3.1. "For instance, women were being yelled by the health care personnel and were not treated as expected." --> please give the main results in a neutral way in the conclusion Like "Mistreatment, verbal/physical abuse and unresponsiveness from health personnel'. There is no need to give examples as this has been done earlier in the manuscript.

   3.2. “This practice demotivates the health personnel to render better or selective quality of services among the women defaulters if alternative services fail and seek main stream health services.” --> This should have been said in Discussion Section.

2. Ethical considerations: "KM was partially funded by Lilongwe University of Agriculture and Natural Resources on some logistics during data collection across Malawi." --> Can the authors explain which kind of logistics and why the funding came from an institution which apparently is not directly related to quality of health services?