Scale up and strengthening of comprehensive emergency obstetric and newborn care in Tanzania.

INTRODUCTION: In Tanzania, inadequate access to comprehensive emergency obstetric and newborn care (CEmONC) services is the major bottleneck for perinatal care and results in high maternal and perinatal mortality. From 2015 to 2019, the Accessing Safe Deliveries in Tanzania project was implemented to study how to improve access to CEmONC services in underserved rural areas.
METHODS: A five-year longitudinal cohort study was implemented in seven health centres (HCs) and 21 satellite dispensaries in Morogoro region. Five of the health centres received CEmONC interventions and two served as controls. Forty-two associate clinicians from the intervention HCs were trained in teams for three months in CEmONC and anaesthesia. Managers of 20 intervention facilities, members of the district and regional health management teams were trained in leadership and management. Regular supportive supervision was conducted.
RESULTS: Interventions resulted in improved responsibility and accountability among managers. In intervention HCs, the mean monthly deliveries increased from 183 (95% CI 174-191) at baseline (July 2014 -June 2016) to 358 (95% CI 328-390) during the intervention period (July 2016 -June 2019). The referral rate to district hospitals in intervention HCs decreased from 6.0% (262/4,392) with 95% CI 5.3-6.7 at baseline to 4.0% (516/12,918) with 95% CI 3.7-4.3 during the intervention period while it increased in the control group from 0.8% (48/5,709) to 1.5% (168/11,233). The obstetric case fatality rate decreased slightly from 1.5% (95% CI 0.6-3.1) at baseline to 1.1% (95% CI 0.7-1.6) during the intervention period (not statistically significant). Active engagement strategies and training in leadership and management resulted in uptake and improvement of CEmONC and anaesthesia curricula, and contributed to scale up of CEmONC at health centre level in the country.
CONCLUSIONS: Integration of leadership and managerial capacity building, with CEmONC-specific interventions was associated with health systems strengthening and improved quality of services.

Introduction

The seeds of improving maternal and newborn health were planted in Tanzania following the 1987 birth of the Safe Motherhood Initiative (SMI) in Nairobi accompanied by an international call to action [1]. Following that, the government of Tanzania established the Safe Motherhood Unit within the Ministry of Health. Several interventions were then identified and included in the national policies and plans, including: strengthening the health system to provide skilled attendance during childbirth, upgrading rural health centres to provide emergency obstetric and newborn care (EmONC) services, providing adolescent and male friendly family planning services, strengthening public–private partnership to ensure continuum of care, and strengthening community participation [2, 3]. Despite a number of interventions, maternal mortality ratio stagnated above 500 per 100,000 livebirths (specifically in 1999, 2005 and 2015 it was estimated to be 528, 578 and 556 per 105 live births, respectively) [4-6]. The major bottlenecks identified included -inadequate and ineffective implementation approaches of these services because of gaps in leadership and management (L&M) at all levels of the health system, coupled with inadequate material, financial and human resources [2, 7, 8].

Success stories in reducing maternal and newborn deaths exist in Tanzania and across the globe due to integrating leadership and management, and medical-based interventions for maternal and newborn care [9-13]. Evidence indicates that when maternal and child health interventions lead to improved quality of obstetric and newborn care, the success has been attributed to strong leadership in reproductive health and accountability of health providers, managers and key decision makers as well as the presence of enabling policies [9]. In Kigoma, Tanzania, transformational leaders and strong managers of locally available material and human resources strengthened maternal services at a regional hospital. This resulted in reduction of maternal mortality ratio from 933 to 186 per 105 live births over the period 1984–91 [11, 12].

In 2015, as part of the Innovating for Maternal and Child Health in Africa (IMCHA) initiative funded by Canada’s International Development and Research Centre, the Accessing Safe Deliveries in Tanzania (ASDIT) project was designed to study how to improve access to comprehensive emergency obstetric and newborn care (CEmONC) services in underserved rural Tanzania, where only 12% of HCs were then providing CEmONC [3]. CEmONC services are the key medical interventions that are used to treat direct obstetric complications that cause the vast majority of maternal deaths around the globe [14]. Implementation of the project revealed weaknesses in leadership and management as well as high rates of maternal and perinatal adverse outcomes. Most health managers at the primary health facilities, district and regional levels had inadequate leadership and management skills. These findings formed the basis for integrating the ASDIT project medical-based interventions with a sister program for leadership and managerial capacity strengthening for quality pregnancy and newborn care. This paper presents the overall key findings from these programs.

Materials and methods

Study settings, design and public involvement

At the start of the study, Morogoro region had 15 HCs that were either already offering CEmONC or would be able to do so once staff are trained. This study was a longitudinal cohort study in seven health centres in Morogoro region, Tanzania. Five of these received an intervention and two served as controls to detect secular trends. The HCs in the intervention group were Kibati, Ngerengere, Gairo, Melela and St. Joseph HCs. The hierarchy of health facilities in Tanzania, from bottom to top, includes dispensaries, health centres (HCs), district hospitals, regional hospitals, zonal hospitals and the national specialized hospitals. By design, study health centres had to be far enough from the nearest referral hospital for referral to be a significant challenge for health centre staff and families. Health centres were also chosen to reflect the diversity of funding and governance models for HCs in Tanzania. As described elsewhere, Kibati and Ngerengere HCs had the proper infrastructure for CEmONC services including maternity and neonatal wards, operating theatre and ability to provide blood transfusion but their staff had not received CEmONC training [15]. This group typified the HCs that the government would have to upgrade as it implemented its national goal of 50% of HCs in Tanzania offering CEmONC services by 2020 [3]. Gairo and Melela HCs (publicly funded), and St. Joseph HC (representing a group of faith-based organizations) were already providing CEmONC and were included in the intervention group to study how CEmONC services could be strengthened. There were therefore 5 health centres that underwent the intervention. Mlimba and Mkamba HCs were randomly allocated to the control group from the remaining 5 publicly funded HCs that were already providing CEmONC services. Because of the intentional differences of intervention and control HCs, this study was designed as a before-after intervention where different funding and governance models could be compared. Comparisons of intervention and control HCs are mainly to detect secular trends that could potentially explain the before-after differences observed in the intervention HCs.

The study was conceived and designed by the ASDIT team, a multidisciplinary group of researchers at the Tanzanian Training Centre for International Health (Tanzania) and Dalhousie University (Canada), partnering with Morogoro regional administration representing the varied interests of patients, health care providers, healthcare systems and policy makers. The Regional Medical Officer for Morogoro region (GM) was engaged as a public co-investigator and worked as a liaison between the district, regional and national authorities. To identify the facilities, most relevant research topics and meaningful outcomes, we worked with the public co-investigator and administered a leadership and management (L&M) survey customized to care providers. Through workshops and meetings, the project team regularly shared findings with the key stakeholders at district, regional and national levels to provide them with a broader understanding of the project, and the progress and outcomes.

The theory of change: A model formulation

In order to develop a set of sound and scientifically derived interventions the project applied principles of operations research to identify and address operational factors that determine maternal and newborn health care in Tanzania [16, 17]. Using evidence-based science on the interventions that work, available material, financial and human resources the project blended medical-based, and leadership and managerial interventions (Fig 1) [18, 19].

Fig 1

The theory of change for improved maternal and newborn health care in the ASDIT project.

Interventions

Capacity building in emergency obstetric and newborn care and anaesthesia

Forty-two associate clinicians from the five intervention HCs were trained in teams for three months in CEmONC and anesthesia. Considering the national regulations, assistant medical officers (advanced associate clinicians) from these HCs were trained in CEmONC while clinical officers and nurse-midwives (associate clinicians) were trained in anaesthesia, postoperative care and care of the sick and premature newborn [15]. In Tanzania, assistant medical officers are licensed to perform surgery. The two years of training includes three months in general surgery and three months in obstetrics and gynaecology. The lack of internship program and inadequate supervision after graduation denies them the opportunity to acquire adequate surgical skills in obstetrics. This CEmONC training program was designed to strengthen surgical skills taking into consideration that they were expected to work independently in remote HCs. The curricula for CEmONC and anaesthesia were built on training programs for associate clinicians previously delivered at the St. Francis Referral Hospital [20].

To reinforce knowledge and skills, post-training activities included eHealth strategies, quarterly supportive supervision visits and continuous mentorship. The eHealth strategies included the offline eLearning modules and tele-consultation. For tele-consultation, care providers at the intervention HCs were also linked with obstetricians, a paediatrician and an anaesthetist based at St. Francis Referral Hospital. Mentorship and supportive supervisory visits were done every three months and included clinical audits and data collection for C-sections, maternal deaths and morbidities, fresh stillbirths, early neonatal deaths and methods of anaesthesia [15]. Mentorship activities focused on identified areas of substandard care.

Strengthening leadership and management

The project team designed capacity-building workshops in L&M and onsite mentorship geared at equipping health managers with essential knowledge and skills on leading change. These were basic principles and strategies in leading change that would improve performance and CEmONC services at their health facilities. The workshops were conducted in 2018 and 2021 and involved participants from 20 primary health facilities, i.e., the 5 intervention health centres and 15 satellite dispensaries, members of the district and regional health management teams specifically the district medical officers and other district health personnel. These dispensaries referred patients with medical complications to their respective health centres. Quality improvement plans developed after the 2018 “Big Results Now” (BRN) Star Rating assessment were used to mentor (onsite) the health facility health management teams and jointly address the gaps identified. The BRN star rating is a government system that measures the performance of various healthcare facilities aimed at improving quality of healthcare [21]. Since prior research has shown that engaging workplace teams in leadership development programmes is critical to success [22], onsite mentorship was a major component of the ASDIT intervention.

Data collection

All data were collected by the research team. The data on CEmONC services were collected concurrently with supportive supervisory visits as described above. The data included deliveries, types of anaesthesia, referrals and audit results of pregnancy adverse outcomes (maternal and perinatal morbidity and mortality). These were obtained from the working log books at each centre. Data on L&M were collected at baseline in 2018 and at the end of the study in 2021 using validated tools i.e., the “Big Results Now” Star Rating assessment and L&M survey tools [21, 23]. The BRN tool assesses the following domains: 1) health facility management (12 indicators); 2) use of facility data for service improvements (6 indicators); 3) staff performance management (5 indicators); 4) organization of services (8 indicators); 5) handling of emergencies/referral (7 indicators); 6) client focus (4 indicators); 7) social accountability (7 indicators); 8) facility infrastructure (14 indicators); 9) infection prevention and control (11 indicators); 10) clinical services (13 indicators); and 11) clinical support services (23 indicators) (S1 Table).

The BRN system rates health facilities from 0 to 5 stars depending on the quality of services provided. The BRN star rating is based on the score of the minimum scoring domain and not the total or average marks. A score of 0–19% is graded no stars, 20–39% one star, 40–59% two stars, 60–79% three stars, 80–89% four stars, and 90–100% five stars [21]. The target of the government improvement initiative was to have 80% of primary health facilities rated with three stars or more by 2017–18. Three stars implied that the facility was performing at a minimum required standard and the domain scoring the least scored 60–79%.

The “Leadership and Management” (L&M) survey primarily used Likert scales to assess data on care providers’ perceptions on L&M competencies, focusing on the following domains: team climate of facilities; staff role clarity; and job satisfaction (Table 1). CEmONC costs were collected from the health centres, Tanzania Medical Store Department and non-governmental organizations that had upgraded health centres for CEmONC services provision.

Table 1

Leadership and managerial domains assessed using L&M survey in 2018 and 2021.

1. Domain: Team climate of facilities    Involvement of staff in the process of setting the objectives, clarity, achievement and worthwhile of the objectives to the health care facility, and information sharing, Team climate is defined as a healthy, supportive and engaging environment for employees at the workplace2. Domain: Staff role clarity    Clarity of individual roles/ responsibilities, planned goals and objectives, existence of performance targets, expectations, adequacy of resources to support implementation of the assignments and ability to meet annual performance targets.3. Domain: Job satisfaction    Satisfaction with communication across the facility, control over given job activities, expectations of the job done, involvement in decision-making processes, facility’s support for individual learning and development, safety of work environment, the balance between work and family/personal life, opportunities for social contact at work, opportunities to interact with management/administration, amount of responsibility.

Uptake strategies

Several uptake strategies were employed to enhance uptake of key interventions. These included engagement of key decision makers and the regional and council management teams throughout the project implementation period. The team conducted biannual national and regional stakeholders’ meetings and provided updates of the project during the quarterly regional maternal and child mortality audit meetings.

Data analysis

Using Stata (version 15), one-way ANOVA and Chi-square tests were used to assess the impact of the intervention model by comparing outcomes during the baseline (July 2014—June 2016) and intervention (July 2016—June 2019). A one-way ANOVA test was used to determine the statistical differences of the mean monthly deliveries and mean score of the BRN key domains. Chi-square tests were used for the obstetric case fatality rates and proportions of justified C-sections within the intervention and control health centres. Confidence intervals were set to 95%.

Ethics and permission

Ethical approval was granted by the National Institute for Medical Research (NIMR) of Tanzania with Ref. No. NIMR/HQ/R.8a/Vol.IX/1986, Dalhousie University Institutional Review Board and the Tanzania Commission for Science and Technology (COSTECH) with Ref. No. CST/ AD.69/227/2015. Permission to conduct research in these facilities was obtained from the regional and district local governments. Informed written consent for the L&M survey was obtained from all participants. Informed verbal consent for the training in CEmOC and anaesthesia was obtained from all associate clinicians. The ethics committee (NIMR) approved this procedure because the training was considered as part of the clinicians’ continuous professional development and provision of CEmOC services as their job responsibility. There was no need for patient’s consent because this study used anonymized patient data that was already being collected as part of the routine operation of the health centres. All methods were performed in accordance with the relevant guidelines and regulations.

Results

Strengthening health systems

Findings from the BRN star rating assessment system indicated significant improvement of key indicators of the health facility management, use of data for improvement, staff performance assessment, organization of services, handling of emergencies and referral care, health facility social accountability, and infection prevention and control (Fig 2). In 2021, the overall BRN ratings increased in 15 (79%) of the nineteen primary health care facilities, with the number of facilities achieving the target of 3 plus star increasing from 2 (10%) to 10 (50%). BRN star rating assessment was not done in one HC in 2018. The overall mean of the star ratings increased from 1.6 (95% CI 1.3–2.0) in 2018 to 2.6 (95% CI 2.1–3.1) in 2021 in the intervention facilities.

Fig 2

Score of the BRN key domains before and after capacity building in the ledership and management.

A survey on leadership and management indicated improved team climate (p = .005) between baseline (mean = 52.6) and end measurement (mean = 57.4). Likewise, the overall staff role clarity improved significantly (mean 35 vs 38; p < 0.05), suggesting a positive effect of the leadership and managerial capacity building. Although not significant, overall job satisfaction increased slightly, reflecting movement toward a more positive work environment for the health care providers in the participating sites

Utilization of CEmONC services

The number of women coming for delivery care after the intervention HCs began providing and strengthening CEmONC services almost tripled from 183 (95% CI 174–191) mean monthly deliveries at baseline (July 2014 –June 2016) to 358 (95% CI 328–390) during the intervention period, i.e., July 2016 –June 2019 (Fig 3). Significant increases in utilization of services were observed in all HCs in both categories i.e., those that were already providing CEmONC care prior to the beginning of the intervention period (Gairo, St. Joseph and Melela) and the new CEmONC providers (Ngerengere and Kibati). For instance, the mean monthly deliveries at Gairo and St. Joseph HCs increased from 71 (95% CI 67–76) to 137 (95% CI 124–150) during intervention period, and from 48 (95% CI 41–55) to 129 (95% CI 116–143) respectively. The mean monthly deliveries at Kibati and Ngerengere increased from 21 (95% CI 18–23) to 34 (95% CI 30–37) during intervention period, and from 26 (95% CI 23–28) to 33 (95% CI 31–36) respectively. Similar increases were also observed in the control HCs.

Fig 3

Trends in the mean monthly deliveries and referral rates before and during the intervention period.

The referral rate to district hospitals in intervention HCs decreased from 6.0% (262/4,392) with 95% CI 5.3–6.7 at baseline to 4.0% (516/12,918) with 95% CI 3.7–4.3 during the intervention period while it increased in the control group from 0.8% (48/5,709) at baseline to 1.5% (168/11,233) during intervention period. Considering suboptimal documentation and suboptimal records keeping noted during baseline data collection, the referral rate at baseline in the intervention facilities was likely higher than that reported here.

Quality of CEmONC services

The number of women with maternal morbidity increased from 459 at baseline to 2,021 during the intervention period in the intervention facilities (Table 2). Dispensaries around the HCs used the CEmONC health centres as the referral facilities in the case of obstetric and newborn complications. During these periods (baseline and during the intervention) 40 maternal deaths occurred. The primary causes of these deaths during the intervention period were postpartum haemorrhage 8, pre/ eclampsia 5, puerperal sepsis 3, complications of anaesthesia 2, uterine rupture 2, severe anaemia in pregnancy 2 and antepartum haemorrhage 1. The causes of deaths were not established in 13 maternal deaths that occurred at baseline, and four that occurred during the intervention in the control HCs because of inadequate records keeping. Avoidable factors were determined in only 87% i.e., 20 cases out of 23 deaths because the case files for the other 3 had inadequate information. Delay to provide appropriate care after reaching the facilities was identified in 80% i.e., 16 of 20 facilities. A delay in seeking treatment/ reaching the facility was found in 55% i.e., 11 out of 20 deaths. Although not significant, the obstetric case fatality rate decreased from 1.5% (95% CI 0.6–3.1) at baseline to 1.1% (95% CI 0.7–1.6) during the intervention period. Some patients were received in moribund condition making it almost impossible to save their lives. In the control facilities, obstetric case fatality rate decreased from 3.3% (95% CI 1.2–7.0) at baseline to 0.8% (95% CI 0.2–1.7) during the intervention period.

Table 2

Obstetric case fatality rate before and after the intervention in the control and intervention health centres.

	Total deliveries	Maternal deaths	Maternal morbidities	Case fatality rate (%)*	95% CI
Intervention HCs
Baseline	4,392	7	459	1.5	0.6–3.1
Intervention period	12,918	22	2021	1.1	0.7–1.6
Control HCs
Baseline	5,709	6	182	3.3	1.2–7.0
Intervention period	11,233	5	664	0.8	0.2–1.7

Note: Baseline = Jul 2014 –June 2016; and intervention period = Jul 2016 –June 2019. *Obstetric case fatality rate, is defined as “the proportion of women admitted to an EmOC facility with major direct obstetric complications, or who develop such complications after admission, and die before discharge. The numerator is the number of women dying of direct obstetric complications during a specific period at an EmOC facility. The denominator is the number of women who were treated for all direct obstetric complications at the same facility during the same period.” [14].

During the intervention period a total of 2,179 CS were performed in the intervention group and 964 in the control group. Of the 674 C-sections that were audited in the intervention HCs, the overall proportions of justified CS were 80% (95% CI 75% - 85%) in year one and 88% (95% CI 83% - 92%) in year three. The proportions of CS that were performed with justifiable indications in the control facilities during this period were 74% (95% CI 64% - 84%) in year one and 78% (95% CI 67% - 89%) in year three. During the study period five women died from immediate complications of caesarean section and anaesthesia in the intervention facilities. Of these, two had severe intraoperative haemorrhage, two had complications of anaesthesia and one had severe preeclampsia and an asthma attack preoperatively. The risk of a woman dying from complications of caesarean section in these health centres was 2.3 per 1,000 caesarean deliveries (95% CI 0.7–5.3). The risk of a woman dying from complications of anaesthesia in the intervention health centres was 0.9 per 1,000 caesarean deliveries (95% CI 0.1–3.3). Maternal deaths in the control facilities were not audited due to either inadequate documentation or absence of case files.

The requirements and costs of scaling up of CEmONC services in health centres

The estimated costs of upgrading a health centre to provide CEmONC services was $256,650 (USD) for infrastructure and equipment, $4,463 per person for upgrading skills in either in CEmONC or anaesthesia for three months and $43,500 per year for medicines and supplies. The total cost for all components per health centre was estimated at $560,802. Detailed findings on the requirements and costs for scaling up CEmONC in health centres in Tanzania are reported elsewhere [24].

Uptake of CEmONC services

Continuous implementation of knowledge translation and engagement strategies resulted in uptake of CEmONC training curriculum, improvement of the curriculum for anaesthesia from three to six months, and contributed to scale up of CEmONC at health centre level in the country. Between 2015 and 2019, a total of 350 health centres and 69 district council hospitals were either renovated or constructed and equipped by the government to offer safe surgery services including CEmONC services [25].

Discussion

Improving access to comprehensive emergency obstetric and newborn care services in underserved areas in limited resource countries requires well designed and effective strategies. The cornerstones of the ASDIT program’s strategy were active engagement of politicians and health system decision-makers at all levels, implementation of CEmONC, continuous supervision and mentorship, as well as strengthening leadership and management at the primary health facility, district and regional levels.

Health systems strengthening for maternal and newborn health care

Strengthening leadership and management at the health facility district and regional health system levels was associated with strengthened health systems building blocks, which are vital for provision of effective services. The key domains that were improved in this study included facility management, use of data for improvement, staff performance assessment, organization of services, handling of emergencies and referral care, health facility social accountability, and infection prevention and control. Improvement in the health systems building blocks explain increased utilization, quality and sustainability of CEmONC services presented in this study. Improvement in these domains could partly be attributed to improved leadership and management. Health systems are connected to leadership and management. These findings suggest that in building effective health systems for maternal and newborn care, leadership and management represent the foundation for all other building blocks. Studies strongly indicate that reduction of maternal and neonatal mortality in countries and settings with high rates depends on health systems strengthening [9, 17, 26, 27].

The gaps in leadership and management skills revealed in the Tanzanian health system before the intervention reflected a health system weakness that contributed to high maternal and newborn mortality. These findings call for action to change strategies and approaches to avert maternal and newborn mortality through integration of transformational leadership and change management with medical interventions. In order to create and drive changes in Tanzania, leadership and management should be strengthened at all levels of the health system, i.e., health facility, district and regional levels.

Scale up of CEmONC services

Improving the availability and access to comprehensive emergency obstetric and neonatal care services was associated with a marked increase in utilization of services (including more women with obstetric complications) and reduced referral rates to distant hospitals in intervention centres. Control facilities continued to refer more high-risk patients than the intervention facilities. The overall proportions of justified CS, in the intervention facilities, was 88% in year three of the project.

This study was not powered to detect either maternal mortality or the specific risk of a woman dying from complications of caesarean section. Nevertheless, there was a downward albeit statistically insignicant trend in maternal mortality rates in both intervention and control facilities. This is reassuring in that intervention HCs were referring fewer women to secondary hospitals, implying that there was a higher level of comfort in managing complicated pregnancies. Maternal mortality rate also dropped in the two control centres but the rate in the pre-intervention period was far higher than in the intervention period (thereby providing more room for improvement) and they continued to refer pregnancies at the same rate to secondary hospitals, suggesting little change in the complexity of pregnancies and deliveries they were managing. It’s noteworthy that the intervention HC rates of caesarian-section mortality after scaling up CEmONC was 2.3 per 1,000 caesarean deliveries (95% CI 0.7–5.3), which is lower than 15 and 7.9–10·9 per 1000 CS reported in Sierra Leone in 2016 and in various studies done in low-income countries respectively [28-30].

The risk of a woman dying from complications of anaesthesia in the intervention health centres was 0.9 per 1,000 caesarean deliveries (95% CI 0.1–3.3). The risk was in the same range with that reported from rural health centers in Kigoma region (0.5 per 1000 C-sections), Zimbabwe (2.1 per 1000), Nigeria (2.5–3.7 per 1000 C-sections) and low- & mid-income countries (0.8–1.7 per 1000 obstetric procedures) [31-34]. A study done in low- and mid-income countries on risk of maternal death from anaesthesia did not show any difference when anaesthesia was provided by associate clinicians and that provided by a physician anaesthetists. The rate of any maternal death was 9·8 per 1000 anaesthetics (5.2–15.7) when managed by associate clinician-anaesthetists compared with 5.2 per 1000 (0.9–12.6) when managed by physician anaesthetists [35]. At present, associate clinicians form the backbone of CEmONC and anaesthesia in Tanzania and should be used to scale up the services. Our associate clinicians provided safe anaesthesia services after a brief but intensive three-month training program. Although longer training programs would provide a greater depth and breadth of experience, a three- month training program will allow countries to provide life-saving services to areas that have none. Longer training programs could eventually be phased in. Effective implementation of CEmONC is strongly associated with reduction of maternal and newborn mortality [14].

Findings from previous studies coupled with effective knowledge translation strategies, engagement, political will and commitment resulted in a nation-wide scale up of CEmONC services in public health centres [15, 20, 24, 32, 35]. Between 2015 and 2019, a total of 350 health centres and 69 district council hospitals were either renovated or constructed and equipped by the government to offer CEmONC services using associate clinicians [25]. These forward-looking decisions provide great learnings for countries with similar economic power and underlying context for maternal and newborn health in sub-Saharan Africa and beyond. These results provide evidence on how active engagement of politicians and health system decision-makers at the highest levels can strongly contribute to improving maternal and newborn health, the critical challenge of our time.

Addressing the fear of the unknown: The cost of scaling up CEmONC services

Scaling up CEmONC in countries is a costly, complex and context-dependent intervention, and this longitudinal cohort study provides valuable insight into what is required to undertake this. No experimental evidence exists, and scepticism persists about the costs and requirements needed. It is known that “the oldest and strongest kind of fear is fear of the unknown.” This study indicates that about $560,000 US is needed to upgrade a Tanzanian health centre to a CEmONC facility. This figure is within reach in many low-and mid-income countries for at least some HCs that serve large regions. Scaling up of CEmONC services is feasible in almost all countries–it is a matter of political will, commitment and prioritization of investments [9, 24].

Limitations of the study

The deliberate choice of HCs that represented different funding and governance models (faith-based organizations and publicly funded), with different levels of experience in provision of CEmONC and different management at the Council Health Management Team levels rendered statistical comparison of HCs difficult and under-powered. Grant funding constraints made it impossible to have more than one HC for each model. It is therefore impossible to assume that a study HC (e.g., a faith-based one) is representative of all such HCs. Nevertheless, important quantitative and qualitative information was gained from each HC studied. It is also possible that secular trends explain some of the improvement seen in the intervention HCs. For example, there was a substantial drop in maternal mortality in the control HCs perhaps due to ongoing involvement of Tanzanian health agencies including introduction of the BRN star rating system. Other factors such as changes in human migration and population could have affected the findings in both intervention and control HCs.

Conclusions

Integration of leadership and managerial capacity building, with CEmONC-specific interventions was widely accepted by all intervention HCs and associated with health systems strengthening and improved quality of services. The key learning is that scale up of CEmONC services in health centres in underserved areas is effective, feasible, safe and desirable using available material, financial and human resources. Itis urgently needed in resource-limited countries. This paper contributes to the body of evidence-based solutions and calls for action to scale up this model solution in countries with high maternal and newborn mortality rates.

Domains and indicators for big results now star rating.

(DOCX)

Click here for additional data file.

(DTA)

Click here for additional data file.

(DTA)

Click here for additional data file.

27 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Nnabuike Chibuoke Ngene, Dip HIV Med; MMed(FamMed); FCOG; MMed(O&G); Ph.D

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

4. Thank you for submitting the above manuscript to PLOS ONE. During our internal evaluation of the manuscript, we found significant text overlap between your submission and the following previously published work, of which you are an author.

- https://www.ajol.info/index.php/ajrh/article/view/216032/203738

We would like to make you aware that copying extracts from previous publications, especially outside the methods section, word-for-word is unacceptable. In addition, the reproduction of text from published reports has implications for the copyright that may apply to the publications.

Please revise the manuscript to rephrase the duplicated text, cite your sources, and provide details as to how the current manuscript advances on previous work. Please note that further consideration is dependent on the submission of a manuscript that addresses these concerns about the overlap in text with published work.

We will carefully review your manuscript upon resubmission, so please ensure that your revision is thorough.

6. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study examines the effect of upgrading health centers in Tanzania to provide CEmONC services on leadership and management, utilization and morbidity and mortality. Its strengths include detailed measures on both proximate outcomes and health outcomes and longitudinal data. It also builds on an evolving literature on how maternal and neonatal services should be organized in order to best save lives. However, it suffers from lack of clarity on the intervention, selection of the control facilities, and the main mortality results are not replicable. I have listed a number of comments, predominately on the methods and results sections, in order to strengthen the paper:

Title:

- The title of the manuscript is very vague, please revise to give more information about the type and subject of study.

Methods:

- Why not more control facilities? How was the sample size (particularly for control) determined? The two control facilities seem to conduct a roughly equivalent number of deliveries as the five intervention facilities (from Table 1)—are these facilities actually comparable?

- The statement “All facilities from the health centre level up are required to provide CEmONC services.” (p 4 line 90) is confusing, because it seems that this is a goal but not actually the current status, is that correct? If so, please clarify.

- Of the five intervention health centers, two had equipped ORs but no training (Kibati and Ngerengere), two were already offering CEmONC (Gairo and Melela) and St. Joseph status was not determined (please specify!). Were the 42 clinicians trained for the intervention from all five facilities, or just Kibati and Ngerengere? When did Gairo and Melela start offering CEmONC?

- “Forty-two associate clinicians from the intervention HCs were trained in teams for three months in 124 CEmONC and anesthesia.”(p 6 line 123): what proportion of total clinicians is this? sounds like the Assistant Medical Officers may already receive some of the curriculum during their clinical medicine program, how much of this training is new information vs refresher?

- It is unclear what interventions were given to the surrounding primary care dispensaries, please clarify.

- Were control dispensaries selected? Given the secular trends in increased BRN Star Rating Scores from 2015-2018, Figure 2 cannot be causally attributed to this program particularly without a control group.

- It seems like a difference-in-differences study design may be appropriate, which would allow for a more rigorous estimate particularly of the mortality and morbidity estimates that account for clustering at the facility level.

- More detail is required on the L&M survey. Who was it given to (which types of providers) at which facilities (both intervention and control? Health centers and dispensaries?) How many items were in the L&M survey? Is it a validated index?

- Following some sort of reporting guidelines (i.e. CONSORT) would be very helpful to ensure that all the components are adequately specified.

Results:

- Where did the data on utilization and referrals come from? Needs to be included in the methods

- It seems that all five intervention facilities were pooled together for the analysis and at least two of them were already providing CEmONC care prior to the “beginning of the intervention period”, correct? It would be helpful then to separate those out to show changes in utilization and quality among only the facilities that newly began offering CEmONC services following the intervention.

- Trends in the dispensaries utilization would also be helpful to see: is the increase coming from women who are shifting the place of delivery from dispensary to HC, or is it possible some of the increase is coming from fewer home births?

- In Table 1, how is maternal morbidities defined?

- In Table 1, how is Case fatality rate calculated? I can’t replicate the numbers given the information in the table.

- How did the number of C-sections change over the course of the study? How did this differ between the facilities that were newly upgraded and the ones that already had CEmONC capacity? It may be useful to have a figure showing these trends as well.

- The same goes for blood transfusions: how did these change over the course of the study?

- “During the intervention period a total of 2,179 CS were performed in the intervention group and 964 in the control group.” (p 11 line 236). This line confuses me because I thought that C-sections were unavailable in the control facilities.

- I appreciate the inclusion of the cost data here, yet given that the methods were not appropriately described in this manuscript and seem to refer wholly to a different manuscript, these are not ‘results’ of this study. They should be moved to the discussion section.

- The same is true of the “uptake of CEmONC services” section: these are not original findings of this study so should be moved to the discussion section.

Discussion:

- Given the apparent lack of control group for the L&M and BRN measurements, the causal language used in the discussion, i.e. “In this project, improving leadership and management was a change factor, a fuel for progress,” (p 12, line 279) is inappropriate. None of the changes in utilization or quality can be causally attributed to changes in leadership.

- Study limitations need to be acknowledged and discussed.

Reviewer #2: Thank you for affording me with the opportunity to review this very interesting manuscript. I have read it with interest and would like to suggest the following comments that may assist in improving its presentation.

(1) General:

- Would it be possible to explicitly stat the overall aim and objectives of this study?

- Was this a stand-alone study or a part of a bigger study (intervention study)?

(2) Methods

�  Study design: I struggling to understand the study design used in this study.

- Page4, L86-87: “This study was a prospective cohort study in seven health centres in Morogoro region, Tanzania. Five of these received an intervention and two served as controls in order to detect secular trends”

- Page 5, L97-100: 20203. “Gairo and Melela HCs (publicly funded) were already providing CEmONC and were included in the intervention group in order to study how CEmONC services could be strengthened. St. Joseph HC represented a group of faith-based organizations. Using simple random sampling of Morogoro HCs, Mlimba and Mkamba HCs were allocated to be control sites.”

Comments:

A prospective cohort study is an observation study, participants are either exposed or controls (not exposed). Using the wording such as “…received an intervention”, “…included in the intervention group”, “…were allocated” makes the study an experimental one. Can you please check this out and present the correct study design by specifying (if it’s a cohort) the key features of a cohort design such as exposure status, how long follow up, outcome (s) of interest, etc.

If it was an intervention study, this should be clearly described and the content of every section should reflect the study design used and clearly specify the key features of an experimental study.

- On Page 6, there is even a section on “Interventions” consisting of “Capacity building in emergency obstetric and newborn care, and anaesthesia” and “Strengthening leadership and management”

Comment: Alluding to my comment above, this “Interventions” means the study was not an observational one (cohort), but an experimental (or quasi-experimental) study. Table 1 is also referring to “before and after the intervention in the control and intervention health centres”

Further details on data analysis will also depend on the study design. For instance, did you consider any relative measures of association? How will you know that the “intervention” was effective to show an impact as you mention L176-177 “multiple statistical tests were used to assess the impact of the intervention model”.

�  Data analysis: “multiple statistical tests were used to assess the impact of the intervention model”.

Comment: Do these “multiple statistical tests refer to One-way ANOVA and Chi-square tests or was there any other test. If there was any other test, I would suggest to describe it. Were there any descriptive statistics done? Would you consider any measures of association as you are assessing the impact of the intervention?

(3) Results: “Interventions resulted in improved responsibility and accountability among managers”

Comments: Can you specify the results that substantiate this claim? How did you define “accountability in this manuscript?

(4) There are a few typos to be corrected for instance in the referencing style (References 4, 5, 6)

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Fidele Mukinda

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

12 Mar 2022

RESPONSE TO REVIEWERS’ COMMENTS

1. ACADEMIC EDITOR’S COMMENTS

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

Response

The ethics statement has been amended. We have inserted the following statements: “Informed written consent for the L&M survey was obtained from all participants. Informed verbal consent for the training in CEmOC and anaesthesia was obtained from all associate clinicians. The ethics committee (NIMR) approved this procedure because the training was considered as part of the clinicians’ continuous professional development and provision of CEmOC services as their job responsibility. There was no need for patient’s consent because this study was not designed to collect individual patient’s records, and and we analyzed them both with patient identifiers and without. No author had direct interaction with patients at any point in time. All methods were performed in accordance with the relevant guidelines and regulations.”

3. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For more information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially sensitive information, data are owned by a third-party organization, etc.) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Response

We have attached the datasets that replicate the results.

4. Thank you for submitting the above manuscript to PLOS ONE. During our internal evaluation of the manuscript, we found significant text overlap between your submission and the following previously published work, of which you are an author.

- https://www.ajol.info/index.php/ajrh/article/view/216032/203738

We would like to make you aware that copying extracts from previous publications, especially outside the methods section, word-for-word is unacceptable. In addition, the reproduction of text from published reports has implications for the copyright that may apply to the publications.

Please revise the manuscript to rephrase the duplicated text, cite your sources, and provide details as to how the current manuscript advances on previous work. Please note that further consideration is dependent on the submission of a manuscript that addresses these concerns about the overlap in text with published work.

We will carefully review your manuscript upon resubmission, so please ensure that your revision is thorough.

Response

We have rephrased the sections with overlapping texts. We have added a reference. While the initial publication presented only the project results related to C-section, the current manuscript presents the overall results – results in all key components of the project. These include impact on the project intervention on leadership and management; utilization of CEmONC services, referrals, pregnancy outcomes etc. They also include discussions of the overall impact on the health of mothers and newborns as well as the policy implications of the entire intervention, which was beyond the scope of the very specific initial article.

2.0 REVIEWERS' COMMENTS:

REVIEWER #1:

1. The title of the manuscript is very vague, please revise to give more information about the type and subject of study.

Response:

We have improved the title of the study. The new title is “Scale up and strengthening of comprehensive emergency obstetric and newborn care in Tanzania”

2. Why not more control facilities? How was the sample size (particularly for control) determined? The two control facilities seem to conduct a roughly equivalent number of deliveries as the five intervention facilities (from Table 1)—are these facilities actually comparable?

Response

The numbers of intervention and control HCs were not comparable because the study was designed to capture the diversity of funding models for Tanzanian Health Centres. Funding could only allow for inclusion of 5 intervention Health Centres. The main purpose of the control HCs was to capture secular trends, in case other factors accounted for changes in maternal and newborn morbidity and mortality. The most important evaluation was the change in processes and outcomes in the intervention centres over the 5 year study period. Comparison with the control HCs was a secondary outcome.

3. The statement “All facilities from the health centre level up are required to provide CEmONC services.” (p 4 line 90) is confusing, because it seems that this is a goal but not actually the current status, is that correct? If so, please clarify.

Response

As indicated in the last paragraph of the introduction of this manuscript, at the beginning of the study only 12% of public HCs provided CEmONC and the goal was to reach 50% by 2020 (indicated in paragraph one of the methods.

4. Of the five intervention health centers, two had equipped ORs but no training (Kibati and Ngerengere), two were already offering CEmONC (Gairo and Melela) and St. Joseph status was not determined (please specify!). Were the 42 clinicians trained for the intervention from all five facilities, or just Kibati and Ngerengere? When did Gairo and Melela start offering CEmONC?

Response

The sentences have been edited. Gairo and Melela HCs (publicly funded), and St. Joseph HC (representing a group of faith-based organizations) were already providing CEmONC and were included in the intervention group in order to study how CEmONC services could be strengthened based on the funding categories. 42 clinicians came from all five HCs.

5. “Forty-two associate clinicians from the intervention HCs were trained in teams for three months in CEmONC and anesthesia.”(p 6 line 123): what proportion of total clinicians is this? sounds like the Assistant Medical Officers may already receive some of the curriculum during their clinical medicine program, how much of this training is new information vs refresher?

Response

This study did not collect records (the number and categories of care providers) at the health centres. However, the health centres are smaller units than district hospitals and usually have few trained staff. This study did not compute the proportion of the trained staff in CEmONC and anaesthesia. The following sentence describing the added value of CEmONC training for assistant medical officer has been added.

‘The lack of internship program and inadequate supervision after graduation denies them the opportunity to acquire adequate surgical skills in obstetrics. This CEmONC training program was designed to strengthen surgical skills taking into consideration that they were expected to work independently in remote HCs.”

6. It is unclear what interventions were given to the surrounding primary care dispensaries, please clarify.

Response

Dispensaries received interventions for leadership and management through two workshops of a few days duration.

7. Were control dispensaries selected? Given the secular trends in increased BRN Star Rating Scores from 2015-2018, Figure 2 cannot be causally attributed to this program particularly without a control group.

Response

Control centres were intentionally selected to have similar ease or difficulty of access to secondary hospitals as intervention centres. Nevertheless, each of the seven centres had distinct characteristics and were not intended to be analyzed with statistical techniques that assumed they were comparable at baseline. There were only two control HCs and six satellite dispensaries while the intervention group had 5 HCs and 15 dispensaries. Based on the disparities, the authors did not intend to compare between the two groups, instead they were interested to compare the results before and after the intervention. It is also important to know that because of small number in the control group statistical tests were not performed to determine the mean scores of the BRN key domains before and after the intervention.

8. More detail is required on the L&M survey. Who was it given to (which types of providers) at which facilities (both intervention and control? Health centers and dispensaries?) How many items were in the L&M survey? Is it a validated index? - Following some sort of reporting guidelines (i.e. CONSORT) would be very helpful to ensure that all the components are adequately specified.

Response

We indicated under the section of data collection that care providers completed the L&M survey forms. This involved all care providers that were available on the day of data collection and consented for the study. Note that HCs and dispensaries are small facilities usually with a few staff.

Results:

9. Where did the data on utilization and referrals come from? Needs to be included in the methods

Response

The following sentence has been inserted in the methods. “The data included deliveries, types of anaesthesia, referrals and audit results of pregnancy adverse outcomes (maternal and perinatal morbidity and mortality). These were obtained from the working log books at each centre.”

10. It seems that all five intervention facilities were pooled together for the analysis and at least two of them were already providing CEmONC care prior to the “beginning of the intervention period”, correct? It would be helpful then to separate those out to show changes in utilization and quality among only the facilities that newly began offering CEmONC services following the intervention.

Response

We agree. Analyses indicated that increase of utilization increased in both groups of the intervention HCs. We initially pooled results to keep the manuscript short. More results have been added in the manuscript.

11. Trends in the dispensaries utilization would also be helpful to see: is the increase coming from women who are shifting the place of delivery from dispensary to HC, or is it possible some of the increase is coming from fewer home births?

Response

We agree. Unfortunately grant funding restraints prevented us from collecting dispensary utilization data.

12. In Table 1, how is maternal morbidities defined? In Table 1, how is Case fatality rate calculated? I can’t replicate the numbers given the information in the table.

Response

In obstetrics maternal morbidity is defined as any obstetric complication that occurs anytime during antepartum, intrapartum or within 42 days after childbirth. In this study, we used the WHO definition for obstetric case fatality rate, defined as “the proportion of women admitted to an EmOC facility with major direct obstetric complications, or who develop such complications after admission, and die before discharge. The numerator is the number of women dying of direct obstetric complications during a specific period at an EmOC facility. The denominator is the number of women who were treated for all direct obstetric complications at the same facility during the same period.”Ref. WHO, UNFPA, UNICEF, AMDD. Monitoring emergency obstetric care: WHO; 2009.

13. How did the number of C-sections change over the course of the study? How did this differ between the facilities that were newly upgraded and the ones that already had CEmONC capacity? It may be useful to have a figure showing these trends as well.

- The same goes for blood transfusions: how did these change over the course of the study?

- “During the intervention period a total of 2,179 CS were performed in the intervention group and 964 in the control group.” (p 11 line 236). This line confuses me because I thought that C-sections were unavailable in the control facilities.

Response

‒ A detail account on the CS was presented elsewhere {Nyamtema AS, Scott H, Kweyamba E, Bulemela J, Shayo A, Mtey G, Kilume O, and LeBlanc JC, 'Improving Access, Quality and Safety of Caesarean Section Services in Underserved Rural Tanzania: The Impact of Knowledge Translation Strategies', Afr J Reprod Health, 25[3s] (2021), 74-83 DOI: 10.29063/ajrh2021/v25i3s.8]

‒ Both interventions and control HCs were already providing CEmONC services. This has been added in the manuscript. Although all HCs provided BT services the study did not capture these data.

14. I appreciate the inclusion of the cost data here, yet given that the methods were not appropriately described in this manuscript and seem to refer wholly to a different manuscript, these are not ‘results’ of this study. They should be moved to the discussion section.

- The same is true of the “uptake of CEmONC services” section: these are not original findings of this study so should be moved to the discussion section.

Response

The authors briefly described the methods used to determine the CEmONC costs on page 8 paragraph 2. … collected from the health centres, Tanzania Medical Store Department and non-governmental organizations that had upgraded health centres for CEmONC services provision. The authors consider the uptake of the project interventions as one of the key results which were set during development of the proposal. In view of that, we had added a few lines in the methods to describe how we implemented our uptake strategies.

Discussion:

- Given the apparent lack of control group for the L&M and BRN measurements, the causal language used in the discussion, i.e. “In this project, improving leadership and management was a change factor, a fuel for progress,” (p 12, line 279) is inappropriate. None of the changes in utilization or quality can be causally attributed to changes in leadership.

- Study limitations need to be acknowledged and discussed.

Response

We have edited the sentence and currently reads; “Improvement in these domains could partly be attributed to improved leadership and management”.

REVIEWER #2:

1. General:

- Would it be possible to explicitly stat the overall aim and objectives of this study?

- Was this a stand-alone study or a part of a bigger study (intervention study)?

Response

We stated on page 3 line 70-71(the last paragraph of the introduction) that the objective was to study how to improve access to comprehensive emergency obstetric and newborn care (CEmONC) services in underserved rural Tanzania, where only 12% of HCs were then providing CEmONC. This was a stand-alone study.

2. Methods

Study design: I struggling to understand the study design used in this study. Page4, L86-87: “This study was a prospective cohort study in seven health centres in Morogoro region, Tanzania. Five of these received an intervention and two served as controls in order to detect secular trends”

- Page 5, L97-100: 20203. “Gairo and Melela HCs (publicly funded) were already providing CEmONC and were included in the intervention group in order to study how CEmONC services could be strengthened. St. Joseph HC represented a group of faith-based organizations. Using simple random sampling of Morogoro HCs, Mlimba and Mkamba HCs were allocated to be control sites.”

Comments:

A prospective cohort study is an observation study, participants are either exposed or controls (not exposed). Using the wording such as “…received an intervention”, “…included in the intervention group”, “…were allocated” makes the study an experimental one. Can you please check this out and present the correct study design by specifying (if it’s a cohort) the key features of a cohort design such as exposure status, how long follow up, outcome (s) of interest, etc.

If it was an intervention study, this should be clearly described and the content of every section should reflect the study design used and clearly specify the key features of an experimental study.

- On Page 6, there is even a section on “Interventions” consisting of “Capacity building in emergency obstetric and newborn care, and anaesthesia” and “Strengthening leadership and management”

Comment: Alluding to my comment above, this “Interventions” means the study was not an observational one (cohort), but an experimental (or quasi-experimental) study. Table 1 is also referring to “before and after the intervention in the control and intervention health centres”

Response

We have edited the study design. It is a quasi-experimental study

3. Further details on data analysis will also depend on the study design. For instance, did you consider any relative measures of association? How will you know that the “intervention” was effective to show an impact as you mention L176-177 “multiple statistical tests were used to assess the impact of the intervention model”. Data analysis: “multiple statistical tests were used to assess the impact of the intervention model”.

Comment: Do these “multiple statistical tests refer to One-way ANOVA and Chi-square tests or was there any other test. If there was any other test, I would suggest to describe it. Were there any descriptive statistics done? Would you consider any measures of association as you are assessing the impact of the intervention?

Response

We have edited the sentence. The authors used one-way ANOVA and Chi-square tests to assess the impact of the intervention model.

4. Results: “Interventions resulted in improved responsibility and accountability among managers”

Comments: Can you specify the results that substantiate this claim? How did you define “accountability in this manuscript?

Response

This sentence has been deleted from the paper. However, the authors assessed the responsibility and accountability among managers using BRN key indicators for the health facility management, use of data for improvement, staff performance assessment, organization of services, handling of emergencies and referral care, health facility social accountability, and infection prevention and control.

5. There are a few typos to be corrected for instance in the referencing style (References 4, 5, 6)

Response

Thank you for identifying these typos. They have been corrected accordingly

Submitted filename: Response to reviewers.docx

Click here for additional data file.

13 Apr 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Nnabuike Chibuoke Ngene, Dip HIV Med; MMed(FamMed); FCOG; MMed(O&G); Ph.D

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #3: Partly

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: While now described as a 'quasi-experimental' design, the study still does not have a suitable identification strategy to support the claims that are asserted in the results and discussion. First, the study relies on a too small sample size (5 intervention facilities and 2 control facilities. Second, the results report increases leadership and management in the intervention facilities based on the BRN and the L&M survey, but there is no comparison to the control facilities and no sample size for the L&M survey is provided. Third, where the control facilities are used in the utilization and quality analysis, they find similar if not better improvements than the intervention facilities. The true extent of the differences is difficult to ascertain because there is no formal difference-in-differences analysis and again, the sample size are far too small to warrant statistical comparison. However, none of these secular improvements are mentioned in the discussion: the focus is entirely on the health systems strengthening and scale up. From my read, there is no evidence to support any improvement in the intervention facilities, making these conclusions unwarranted.

Reviewer #3: Please see comments in attachment:

On page 4:

1) How about the other two HCs in the intervention arm? In line 85, five HCs were referenced and two controls. Please clarify if you had 3 HCs as intervention sites and 2 as control, making 5 all together. It’s a bit confusing to readers.

2) How many HCs were involved in the sampling? Please expand a bit more for clarify.

On page 9:

3) Did the authors consider any co-founding variables? If yes, how were these variables controlled for in the analysis?

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: Yes: Dr. Nnamdi Ndubuka

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Manuscript_NNdubuka.docx

Click here for additional data file.

26 Apr 2022

RESPONSE TO REVIEWERS’ COMMENTS

1. ACADEMIC EDITOR’S COMMENTS

PONE-D-21-37381R1

Scale up and strengthening of comprehensive emergency obstetric and newborn care in Tanzania

PLOS ON

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: While now described as a 'quasi-experimental' design, the study still does not have a suitable identification strategy to support the claims that are asserted in the results and discussion.

First, the study relies on a too small sample size (5 intervention facilities and 2 control facilities.

Response

Intervention HCs were chosen to reflect the diversity of funding models and diversity in accessing care at a referral hospital. The intent was not to have a representative sample of HCs in Morogoro region but, within the funding constraints, choose those centres that would give the greatest understanding of the diverse needs throughout the region. Similarly, the control centres were chosen to reduce the likelihood of contamination from the intervention HCs and to capture the diversity of centres that would not receive an intervention. We purposely chose fewer control HC so that we could allocate more of our funding to thoroughly studying the implementation and scale-up of CEmONC in intervention centres. We primarly followed the control HCs to detect secular trends that could potentially explain some of the differences we saw over time in the intervention HCs.

Second, the results report increases leadership and management in the intervention facilities based on the BRN and the L&M survey, but there is no comparison to the control facilities and no sample size for the L&M survey is provided.

Response

It is true that we did not assess L&M skills in control centre staff but this went beyond the scope of the study. We did not have the resources to assess L&M skills in control centres nor the ability to offer them incentives for such intrusive data collection such as offering training after the study was completed. Again, we primarily used control HCs to detect secular trends. This means that this study was primarily a before-after design. The BRN and the L&M survey results were presented as a before and after in the intervention HCs. This is stated in the methods (line 101 – 104) and limitations of the study (line 356 – 362).

Third, where the control facilities are used in the utilization and quality analysis, they find similar if not better improvements than the intervention facilities. The true extent of the differences is difficult to ascertain because there is no formal difference-in-differences analysis and again, the sample size are far too small to warrant statistical comparison. However, none of these secular improvements are mentioned in the discussion: the focus is entirely on the health systems strengthening and scale up. From my read, there is no evidence to support any improvement in the intervention facilities, making these conclusions unwarranted.

Response

These are important points and it is difficult to statistically compare HCs that were chosen specifically to capture the diversity of HCs in Morogoro. We have added discussion about the differences in outcomes between intervention and control HCs particularly for fatality rate. Despite increasing complexity of the cases managed at the intervention sites because of lower referral rates, improvements were noted. Other factors were probably at play, as demonstrated by improvements in the control sites as well, but given the increased number and complexity of deliveries managed at the intervention sites, this was felt to be a relevant finding.

2. Reviewer #3: Please see comments in attachment:

On page 4:

1) How about the other two HCs in the intervention arm? In line 85, five HCs were referenced and two controls. Please clarify if you had 3 HCs as intervention sites and 2 as control, making 5 all together. It’s a bit confusing to readers.

Response

The second sentence on line 85 indicates that the total number of intervention HCs is 5. The same paragraph indicates that of these (five); two (Kibati and Ngerengere) had not started providing CEmONC, and three (Gairo, Melela and St. Joseph) had CEmONC but needed to be strengthened. This description is provided in the paragraph. The authors have added a line that reiterates that there were 5 intervention sites and 2 control sites.

On page 9:

2) How many HCs were involved in the sampling? Please expand a bit more for clarify.

Responses

Thank you. The following sentence has been added; “Mlimba and Mkamba HCs were randomly allocated to the control group from the remaining 5 publicly funded HCs that were already providing CEmONC services.”

3) Did the authors consider any co-founding variables? If yes, how were these variables controlled for in the analysis?

Response

This study had several confounders that could have positively or negatively affected health facility deliveries. These included differences in fertility rates and consequently different population growth rates between the study districts – differences in fertility rates are also likely to make differences in facility delivery; 2) differences in leadership effectiveness at the council level, some may be more assertive than the others; 3) the act of regular sharing of key results during the regional quarterly meetings that included the council with the control HCs could have led to contamination. We could not control these factors. These have been included in the limitations of the study

Submitted filename: Response to reviewers 26042022.docx

Click here for additional data file.

9 May 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Nnabuike Chibuoke Ngene, Dip HIV Med; MMed(FamMed); FCOG; MMed(O&G); Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

The manuscript requires further revision.

The authors have this opportunity to respond satisfactorily to the following comments.

1. Abstract, Result, sentence: “The case fatality rate decreased slightly from 1.5% (95% CI 0.6–3.1) at baseline to 1.1% (95% CI 0.7-1.6) during the intervention period (not statistically significant).” The term “case fatality rate” may be replaced with “direct obstetric case fatality rate” given the definition provided in the footnote in Table 1 and the content of WHO Monitoring Emergency Obstetric Care – a Handbook (https://apps.who.int/iris/bitstream/handle/10665/44121/9789241547734_eng.pdf?sequence=1).

2. Abstract: State the aim of the study in the abstract. Is this the aim of the study: To detect secular trends in Health Centres (HC) in Morogoro region of Tanzania following the integration of Accessing Safe Deliveries in Tanzania (ASDIT) project with leadership and managerial capacity building in these healthcare facilities?

3. Abstract, conclusion: “Integration of leadership and managerial capacity building, with CEmONC-specific interventions has resulted in health systems strengthening and improved quality of services.” Consider revising the statement to read: “Integration of leadership and managerial capacity building with CEmONC-specific interventions was associated with health systems strengthening and improved quality of services.” This means replacing the words “has resulted in” with “was associated with.”

4. In material and methods, first paragraph, after the sentence “Five of these received an intervention and two served as controls in order to detect secular trends” include the following: The HCs in the intervention group were Kibati, Ngerengere, Gairo, Melela and St. Joseph HCs.

5. Materials and methods, “The theory of change: a model formulation”: (a) “In order to develop a set of sound and scientifically derived interventions the project applied principles of operations research to identify and address operational factors that determine maternal and newborn health care in Tanzania.” Provide a reference for principles of operational research. (b) “Using evidence-based science on the interventions that work,…” Provide a reference for evidence-based science on the interventions that work.

6. Materials and methods, Strengthening leadership and management, sentence: “The workshops were conducted in 2018 and 2021 and involved participants from 20 primary health facilities, i.e., the 5 intervention health centres and 15 satellite dispensaries,...” Explain the referral relationship between the 5 intervention health centres and 15 satellite dispensaries.

7. Materials and methods, Data collection, sentence: “The BRN tool assesses the following domains: 1) health facility management (12 indicators)…” The indicators are difficult to find in references 17 and 19 cited by the authors. Are you referring to Model of Care Initiative in Nova Scotia (MOCINS) Process Indicators or MOCINS Outcome Indicators contained in reference number 19? To avoid confusion, present the indicators in a table.

8. Materials and methods, Data collection, sentence: “The L&M survey primarily used Likert scales to assess data on care providers’ perceptions on L&M competencies,…” The questionnaire (the questions and the scales) that was used for the assessment should be described in a table.

9. One-way ANOVA and Chi-square tests were used. No p-value was stated in the results. Explain.

10. The term “case fatality rate” may be replaced with “direct obstetric case fatality rate” and defined in the materials and methods section. This will involve replacing “case fatality rate” with “direct obstetric case fatality rate” in the footnote in Table 1. This revision will be in line with the terminology changes in the WHO Monitoring Emergency Obstetric Care – a Handbook (https://apps.who.int/iris/bitstream/handle/10665/44121/9789241547734_eng.pdf?sequence=1).

11. Results, Strengthening health systems, sentences: “Capacity-building strategies in transformational leadership and change management resulted in improved leadership and management as assessed using the BRN star rating assessment system and the survey. Capacity building contributed to improved health facility performance and maternal and child health outcomes.” These are interpretation of the results. Therefore DELETE the sentences.

12. In Figure 2, there are black and orange coloured horizontal lines. Are these confidence intervals. Specify.

13. Results, Strengthening health systems, sentence: “The sub-scales included vision, support, task orientation and role clarity.” Explain how these indicators were improved. Include the accompanying data.

14. Results, Strengthening health systems, sentence: “In 2021, the overall BRN ratings increased in 15 (79%) of the nineteen primary health care facilities,…” This is difficult to understand because in the materials and methods 20 (and not 19) primary health care facilities were mentioned.

15. In Figure 3, what does the dotted line represent? Is it the overall trend in the referral rate in the intervention HCs?

16. Results, Utilization of CEmONC services, sentences: “For instance, the mean monthly deliveries at Gairo and St. Joseph HCs increased from 71 (67 – 76) to 137 (124 - 150) during intervention period, and from 48 (41 – 55) to 129 (116 - 143) respectively. The mean monthly deliveries at Kibati and Ngerengere increased from 21 (18 – 23) to 34 (30 – 37) during intervention period, and from 26 (23 - 28) to 33 (31 – 36) respectively.” Specify the meaning of the numbers in bracket.

17. Results, Quality of CEmONC services: What were the primary causes of the maternal deaths and the avoidable/modifiable factors associated with them (at least in the intervention HCs).

18. Results, The requirements and costs of scaling up of CEmONC services in health centres, sentence: “Detailed findings on the requirements and costs for scaling up CEmONC in health centres in Tanzania are reported elsewhere.” Reference 19 was cited by the authors. However, the word Tanzania could not be found in reference 19 (i.e. Model of Care Initiative in Nova Scotia (MOCINS): Final Evaluation Report).

19. Discussion, Health systems strengthening for maternal and newborn health care, sentence: “Strengthening leadership and management at the health facility district and regional health system levels resulted in strengthened health systems building blocks, which are vital for provision of effective services.” It is preferrable to use the words “was associated with” rather than “resulted in.”

20. Discussion, Scale up of CEmONC services, first paragraph, sentence: Improving the availability and access to comprehensive emergency obstetric and neonatal care services resulted in a marked increase in utilization of services (including women with obstetric complications) and reduced referral rates to distant hospitals in intervention centres. It is preferrable to use the words “was associated with” rather than “resulted in.” This is because the improvements could have been due to other factors such as changes in human migration and population. These may explain some of the outcomes in the control HCs.

21. Discussion, Scale up of CEmONC services, second paragraph, sentence: “Maternal mortality also dropped in the two control centres but they continued to refer pregnancies at the same rate to secondary hospitals, suggesting little change in the complexity of pregnancies and deliveries they were managing.” The data on complexity of the cases managed at the intervention and control HCs were not presented in the result section.

22. Discussion, Scale up of CEmONC services, fourth paragraph, sentence: “Findings from previous and current studies coupled with effective knowledge translation strategies, engagement, political will and commitment resulted in a nation-wide scale up of CEmONC services in public health centres.” Delete “and current.”

23. Discussion, “Addressing the fear of the unknown: the cost of scaling up CEmONC services,” sentence: “The requirements and related costs reported in this study fill the existing vacuum of science and knowledge.” Delete this sentence.

This is because the authors also wrote in the results section that “Detailed findings on the requirements and costs for scaling up CEmONC in health centres in Tanzania are reported elsewhere.” Therefore the index report/study can’t be filling any vacuum in knowledge. Additionally, there are previous studies on the cost of scaling up a health facility in low- and middle-income countries.

24. Discussion, Limitations of the study: Acknowledge that other factors such as changes in human migration and population could have affected the findings in both intervention and control HCs.

25. Conclusion: “Integration of leadership and managerial capacity building, with CEmONC-specific interventions has resulted in health systems strengthening and improved quality of services.” It is preferrable to use the words “was associated with” rather than “resulted in.”

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

26 Jun 2022

RESPONSES TO REVIEWER’S COMMENTS

1. Abstract, Result, sentence: “The case fatality rate decreased slightly from 1.5% (95% CI 0.6–3.1) at baseline to 1.1% (95% CI 0.7-1.6) during the intervention period (not statistically significant).” The term “case fatality rate” may be replaced with “direct obstetric case fatality rate” given the definition provided in the footnote in Table 1 and the content of WHO Monitoring Emergency Obstetric Care – a Handbook (https://apps.who.int/iris/bitstream/handle/10665/44121/9789241547734_eng.pdf?sequence=1).

Response

Thank you. Since there are direct and indirect obstetric causes of maternal deaths, we have replaced the term with “obstetric case fatality rate”.

2. Abstract: State the aim of the study in the abstract. Is this the aim of the study: To detect secular trends in Health Centres (HC) in Morogoro region of Tanzania following the integration of Accessing Safe Deliveries in Tanzania (ASDIT) project with leadership and managerial capacity building in these healthcare facilities?

Response

The aim of the study was to study how to improve access to comprehensive emergency obstetric and newborn care (CEmONC) in underserved rural areas. This is stated in the last sentence of the introduction section of the abstract, and in the last paragraph on page 3.

3. Abstract, conclusion: “Integration of leadership and managerial capacity building, with CEmONC-specific interventions has resulted in health systems strengthening and improved quality of services.” Consider revising the statement to read: “Integration of leadership and managerial capacity building with CEmONC-specific interventions was associated with health systems strengthening and improved quality of services.” This means replacing the words “has resulted in” with “was associated with.”

Response

We have replaced the words “has resulted in” with “was associated with.”

4. In material and methods, first paragraph, after the sentence “Five of these received an intervention and two served as controls in order to detect secular trends” include the following: The HCs in the intervention group were Kibati, Ngerengere, Gairo, Melela and St. Joseph HCs.

Response

Thank you. This sentence has been added in the manuscript.

5. Materials and methods, “The theory of change: a model formulation”: (a) “In order to develop a set of sound and scientifically derived interventions the project applied principles of operations research to identify and address operational factors that determine maternal and newborn health care in Tanzania.” Provide a reference for principles of operational research. (b) “Using evidence-based science on the interventions that work, …” Provide a reference for evidence-based science on the interventions that work.

Response

Four references have been inserted.

6. Materials and methods, Strengthening leadership and management, sentence: “The workshops were conducted in 2018 and 2021 and involved participants from 20 primary health facilities, i.e., the 5 intervention health centres and 15 satellite dispensaries,...” Explain the referral relationship between the 5 intervention health centres and 15 satellite dispensaries.

Response

The following sentence has been added; “These dispensaries referred patients with medical complications to the study health centres.”

7. Materials and methods, Data collection, sentence: “The BRN tool assesses the following domains: 1) health facility management (12 indicators)…” The indicators are difficult to find in references 17 and 19 cited by the authors. Are you referring to Model of Care Initiative in Nova Scotia (MOCINS) Process Indicators or MOCINS Outcome Indicators contained in reference number 19? To avoid confusion, present the indicators in a table.

Response

The domains and indicators are for the Tanzanian Big Results Now assessment tool. These indicators do not refer to Model of Care Initiative in Nova Scotia (MOCINS) Process Indicators or MOCINS Outcome Indicators contained in reference number 19. Since the table with domains and indicators is too long, (Table 1. Big Results Now Star Rating domains and indicators), we have provided it as an additional information to this manuscript … however, the legend has been inserted in the manuscript.

8. Materials and methods, Data collection, sentence: “The L&M survey primarily used Likert scales to assess data on care providers’ perceptions on L&M competencies,…” The questionnaire (the questions and the scales) that was used for the assessment should be described in a table.

Response

A table has been inserted. Table 2. Leadership and managerial domains assessed in 2018 and 2021

9. One-way ANOVA and Chi-square tests were used. No p-value was stated in the results. Explain.

Response

Thank you for observation. The confidence interval set to 95% CI for both parametric and non-parametric variables. We have replaced the words “p-value at < 0.05” with “95% CI” in the Methods: data analysis section.

10. The term “case fatality rate” may be replaced with “direct obstetric case fatality rate” and defined in the materials and methods section. This will involve replacing “case fatality rate” with “direct obstetric case fatality rate” in the footnote in Table 1. This revision will be in line with the terminology changes in the WHO Monitoring Emergency Obstetric Care – a Handbook (https://apps.who.int/iris/bitstream/handle/10665/44121/9789241547734_eng.pdf?sequence=1).

Response

As reported above (comment no. 1), we have revised all places where the term ‘case fatality rate’ appeared.

11. Results, Strengthening health systems, sentences: “Capacity-building strategies in transformational leadership and change management resulted in improved leadership and management as assessed using the BRN star rating assessment system and the survey. Capacity building contributed to improved health facility performance and maternal and child health outcomes.” These are interpretation of the results. Therefore DELETE the sentences.

Response

The two sentences have been deleted

12. In Figure 2, there are black and orange coloured horizontal lines. Are these confidence intervals. Specify.

Response

Thank you. We have added in the key (footnote) below the figure that “black and orange coloured horizontal lines are 95% CI”.

13. Results, Strengthening health systems, sentence: “The sub-scales included vision, support, task orientation and role clarity.” Explain how these indicators were improved. Include the accompanying data.

Response:

Thank you. Instead of presenting data on the team climate subscales we have decided to present data on the L&M overall domains that were assessed in this study, i.e., 1) team climate of facilities; 2) staff role clarity; and 3) job satisfaction. These domains are also presented in table 2. Leadership and managerial domains assessed in 2018 and 2021.

14. Results, Strengthening health systems, sentence: “In 2021, the overall BRN ratings increased in 15 (79%) of the nineteen primary health care facilities,…” This is difficult to understand because in the materials and methods 20 (and not 19) primary health care facilities were mentioned.

Response

Although capacity building in L&M and all other assessments involved 20 facilities, the 2018 BRN star rating assessment was not done in one HC. We have added a sentence to clarify it in the results – section.

15. In Figure 3, what does the dotted line represent? Is it the overall trend in the referral rate in the intervention HCs?

Response

The dotted line represented the linear trend in the referral rates in the intervention HCs. However, to avoid confusion, we have removed it.

16. Results, Utilization of CEmONC services, sentences: “For instance, the mean monthly deliveries at Gairo and St. Joseph HCs increased from 71 (67 – 76) to 137 (124 - 150) during intervention period, and from 48 (41 – 55) to 129 (116 - 143) respectively. The mean monthly deliveries at Kibati and Ngerengere increased from 21 (18 – 23) to 34 (30 – 37) during intervention period, and from 26 (23 - 28) to 33 (31 – 36) respectively.” Specify the meaning of the numbers in bracket.

Response

The numbers in the brackets signify 95% CI. These have been added in the manuscript.

17. Results, Quality of CEmONC services: What were the primary causes of the maternal deaths and the avoidable/modifiable factors associated with them (at least in the intervention HCs).

Response

We have added the following sentences in the manuscript. “During these periods (baseline and during the intervention) 40 maternal deaths occurred. The primary causes of these deaths during the intervention period were postpartum haemorrhage 8, pre/ eclampsia 5, puerperal sepsis 3, complications of anaesthesia 2, uterine rupture 2, severe anaemia in pregnancy 2 and antepartum haemorrhage 1. The causes of deaths were not established in 13 maternal deaths that occurred at baseline, and four that occurred during the intervention in the control HCs because of inadequate records keeping. Avoidable factors were determined in only 87% i.e., 20 cases out of 23 deaths because the case files for the other 3 had inadequate information. Delay to provide appropriate care after reaching the facilities was identified in 80% i.e., 16 of 20 facilities. A delay in seeking treatment/ reaching the facility was found in 55% i.e., 11 out of 20 deaths.”

18. Results, The requirements and costs of scaling up of CEmONC services in health centres, sentence: “Detailed findings on the requirements and costs for scaling up CEmONC in health centres in Tanzania are reported elsewhere.” Reference 19 was cited by the authors. However, the word Tanzania could not be found in reference 19 (i.e. Model of Care Initiative in Nova Scotia (MOCINS): Final Evaluation Report).

Response

We have replaced the reference with the appropriate one.

19. Discussion, Health systems strengthening for maternal and newborn health care, sentence: “Strengthening leadership and management at the health facility district and regional health system levels resulted in strengthened health systems building blocks, which are vital for provision of effective services.” It is preferrable to use the words “was associated with” rather than “resulted in.”

Response

Replacement done. Thank you

20. Discussion, Scale up of CEmONC services, first paragraph, sentence: Improving the availability and access to comprehensive emergency obstetric and neonatal care services resulted in a marked increase in utilization of services (including women with obstetric complications) and reduced referral rates to distant hospitals in intervention centres. It is preferrable to use the words “was associated with” rather than “resulted in.” This is because the improvements could have been due to other factors such as changes in human migration and population. These may explain some of the outcomes in the control HCs.

Response

Replacement done. Thank you

21. Discussion, Scale up of CEmONC services, second paragraph, sentence: “Maternal mortality also dropped in the two control centres but they continued to refer pregnancies at the same rate to secondary hospitals, suggesting little change in the complexity of pregnancies and deliveries they were managing.” The data on complexity of the cases managed at the intervention and control HCs were not presented in the result section.

Response

Thank you. By saying the complexity of pregnancies and deliveries the authors meant complicated pregnancies managed at the health facilities. In the results in table 3, the authors presented a summary of women who had obstetric complications (maternal morbidities) at baseline and during the intervention period. In the Results, Utilization of CEmONC services (sub-section): we also presented figures of women with obstetric complications referred to the nearby district hospital. We thought that the data on maternal morbidities and referred pregnant women provided a snapshot of complexity of pregnancies and that somehow justify our statement.

22. Discussion, Scale up of CEmONC services, fourth paragraph, sentence: “Findings from previous and current studies coupled with effective knowledge translation strategies, engagement, political will and commitment resulted in a nation-wide scale up of CEmONC services in public health centres.” Delete “and current.”

Response

The words are deleted

23. Discussion, “Addressing the fear of the unknown: the cost of scaling up CEmONC services,” sentence: “The requirements and related costs reported in this study fill the existing vacuum of science and knowledge.” Delete this sentence.

This is because the authors also wrote in the results section that “Detailed findings on the requirements and costs for scaling up CEmONC in health centres in Tanzania are reported elsewhere.” Therefore the index report/study can’t be filling any vacuum in knowledge. Additionally, there are previous studies on the cost of scaling up a health facility in low- and middle-income countries.

Response

The sentence is deleted

24. Discussion, Limitations of the study: Acknowledge that other factors such as changes in human migration and population could have affected the findings in both intervention and control HCs.

Response

Thank you. These factors have been added in the list of limitations

25. Conclusion: “Integration of leadership and managerial capacity building, with CEmONC-specific interventions has resulted in health systems strengthening and improved quality of services.” It is preferrable to use the words “was associated with” rather than “resulted in.”

Response

The words are replaced. Thank you

Submitted filename: Responses to reviewers comments.docx

Click here for additional data file.

28 Jun 2022

Scale up and strengthening of comprehensive emergency obstetric and newborn care in Tanzania

PONE-D-21-37381R3

Dear Dr. Nyamtema,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Nnabuike Chibuoke Ngene, Dip HIV Med; MMed(FamMed); FCOG; MMed(O&G); Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

1 Jul 2022

PONE-D-21-37381R3

Scale up and strengthening of comprehensive emergency obstetric and newborn care in Tanzania

Dear Dr. Nyamtema:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Nnabuike Chibuoke Ngene

Academic Editor

PLOS ONE