The effect of enhanced public -private partnerships on Maternal, Newborn and child Health Services and outcomes in Nairobi-Kenya: The PAMANECH quasi-experimental research protocol

Introduction: Rapid urbanisation in Kenya has resulted in growth of slums in urban centres, characterised by poverty, inadequate social services and poor health outcomes. The government’s initiatives to improve access to quality healthcare for mothers and children are largely limited to public health facilities, which are few and/or inaccessible in underserved areas such as the slums. The ‘Partnership for Maternal, Newborn and Child Health’ (PAMANECH) project is being implemented in two Nairobi slums, Viwandani and Korogocho, to assess the impact of strengthening public -private partnerships for the delivery of healthcare on the health of mothers, newborns and young children in two informal settlements in Kenya. Methods and analysis: This is a quasi-experimental study; our approach is to support private as well as public health providers and the community to enhance access to and demand for quality healthcare services. Key activities include: infrastructural upgrade of selected Private Not-For-Profit health facilities operating in the two slums, building capacity for healthcare providers as well as the Health Management Teams in Nairobi, facilitating provision of supportive supervision by the local health authorities and forming networks of Community Health Volunteers (CHVs) to create demand for health services. To assess the impact of the intervention, the study is utilising multiple data sources using a combination of qualitative and quantitative methods. A baseline survey was conducted in 2013 and an end-line survey will be conducted at least 1 year after full implementation of the intervention. Systematic monitoring and documentation of the intervention is ongoing to strengthen the case for causal inference. Ethics and dissemination: Ethical approval for the study was obtained from the Kenya Medical Research Institute. Key messages from the results will be packaged and widely disseminated through workshops, conference presentations, reports, factsheets and academic publications to facilitate uptake by policymakers. Protocol registration number: KEMRI- NON-SSCPROTOCOL No. 393. The intervention site covers two informal settlements in Nairobi, namely Korogocho and Viwandani, where APHRC has been operating the Nairobi Urban Health and Demographic Surveillance System (NUHDSS) since 2003. The NUHDSS has been described in detail elsewhere and for this intervention and corresponding planned evaluation provides a reliable sampling frame.12 Viwandani and Korogocho settlements are together home to approximately 70 000 residents and, like other slums, are characterised by poverty, poor coverage of social services and poor MNCH outcomes. More than a decade of research by APHRC in Viwandani and Korogocho slums shows that the two areas exhibit poor child, neonatal and maternal health indicators including high levels of maternal mortality (706/100 000 compared to 488/100 000, the national average), high infant and under-five mortality.8 13 In addition, stillbirth rates are unacceptably high.8 Whereas the proportion of births taking place in health facilities has increased significantly over the last decade (52% in 2000 to 81% in 2012),14 15 about 60% of health facilities in this setting lack trained staff and equipment to handle basic emergency obstetric and neonatal complications.4 All factors considered, the dismal state of maternal and child care services in this setting is a major contributor to the observed high maternal and under-five mortality observed in the slums of Nairobi.12 Implementation of the PAMANECH project is a joint collaboration among several partners including APHRC, the City County of Nairobi, sub-County Health Management Teams (Kasarani and Makadara subcounties), community leaders in Korogocho and Viwandani, private providers, youth leaders and Community Health Volunteers. Six credible and established (not-for-profit) health facilities were selected in Korogocho and Viwandani (three in each site) for transformation into ‘one-stop primary level centres’ for preventive and curative MNCH services. The selection criteria were based on the National Health Insurance Fund (NHIF) Accreditation Manual16 for maternal health services at level 2 and enhanced with items from the SAFECARE manual.17 The NHIF accreditation manual is used in Kenya for accrediting health facilities at different levels for eligibility to receive NHIF refunds. SAFECARE is a continuous quality improvement tool to improve healthcare delivery in resource-restricted settings. It aims to support health providers and consistently improve quality to meet the healthcare needs of the population they serve. The assessment criterion was reviewed by the sub-County Health Management Teams (sCHMTs) of Makadara and Kasarani to determine the relevance and completeness of the tool. The final assessment had two sections: an oral interview section, which covered the health facility details (ownership, relationship with sCHMT), registration of health facility, power supply, presence of qualified registered health personnel, and guidelines and procedures used in MNCH. The second section was an observation checklist, which covered the environment of the health facility, physical structure of the facility, water supply, equipment in consultation room, labour ward and laboratory, availability of a functional ambulance and availability of wards. Each item on the list was scored using a 5-point scale with the maximum score (best score) per item being 1 and minimum score (worst score) per item being 5. A list with 53 health facilities was generated by members of the project management committees from the two communities and validated by the sCHMTs. Twenty two of the listed facilities were not assessed as they were stand-alone laboratories or pharmacies that offered treatment for minor ailments and occasionally offered delivery services. For each health facility, a sum of all the scores was generated and this was used to rank the 31 assessed facilities. The top three health facilities in each settlement were selected for support and the results communicated to the community leaders, sCHMTs and all assessed facility heads at a meeting organised by APHRC. The activities below are aimed at improving access to quality healthcare and reducing the financial burden of health expenditure for the two slum communities in Nairobi mentioned in this study. Six credible and established private not-for-profit (PNFP) providers were selected; however, due to financial constraints, only five have received infrastructural upgrade. Each selected facility will be supported to provide: (1) essential and basic obstetric care; (2) antenatal care; (3) maternity care; (4) postnatal care; (5) diagnosis and treatment of common childhood illnesses; (6) vaccination; and (7) HIV/AIDS-related services such as Voluntary and Routine Counselling and Testing and elimination of mother-to-child transmission of HIV. The support to be provided is based on the identified needs of each facility and includes expansion/renovation of existing facilities, provision of basic equipment, and water and electricity connections, among others. The providers were selected in a consultative manner that involved all the relevant stakeholders. The support is being provided in a phased manner contingent on meeting set criteria for quality improvement. The rationale for this infrastructural support is based on: (1) the need to meet Ministry of Health standards and accreditation criteria, (2) the need for space to offer the envisaged expanded range of services and (3) the need to ensure long-term sustainability of the programme through linkages to other funding streams. Adequate infrastructure is a key accreditation criterion according to guidelines formulated by the NHIF. It is important that these facilities are accredited by the NHIF, not only to benefit from their health insurance scheme, but to benefit from other initiatives such as the output-based aid voucher scheme for reproductive health services currently being piloted in the two slums. There is also potential to benefit from other health insurance schemes such as one offered by Jamii Bora—one of the largest microfinance organisations targeting slum dwellers and street families. Clinical personnel of the selected facilities are being trained using the most appropriate and current clinical and practice guidelines for MNCH. Training is conducted gradually to ensure that what is learnt is institutionalised and practiced before additional training is provided. The private health providers are to receive training and support to generate routine HMIS data and reports that will feed into the district monthly reports. The sub-County (Kasarani and Makadara) and County (Nairobi) health authorities are being facilitated to offer systematic supportive supervision to the ‘One-Stop Primary Level Centres’ to ensure that guidelines are adhered to, skills are reinforced and ultimately high-quality services are offered. In addition, the health authorities are to ensure access to high-quality, low-cost medicine and supplies by the five health centres through existing channels, such as recommendation to the largest supplier of high quality low cost medicines and supplies to the PNFP sector in the country (Mission for Essential Drugs and Supplies—MEDS). Some members of the health management teams have been supported to receive training on supervision, leadership and management to enhance their capacity to effectively manage health services in the districts under which the two slums fall. Under the Ministry of Health community health strategy for improving health, CHVs have been identified in all villages in the country. These, with some facilitation, help implement public health activities including reporting on health events in the community. Owing to high attrition rates of these volunteers, we found it necessary to train more CHVs in order to reconstitute the community units. Ten CHVs per village were selected from the trained pool within the two slums (a total of 180 from 280 trained) to work with the private providers and to create linkages with the public sector. Their main roles include household monitoring for childhood illness and vaccination status, referrals to a primary healthcare facility (public or private), registration of pregnant women in the designated households, referrals for antenatal care and skilled attended delivery at the upgraded private health facilities, assisting mothers with birth planning and postnatal visits, and distribution of contraceptives. These activities, drawing on proven approaches in the utilisation of community health workers, will provide evidence to guide the implementation of the national community strategy. Young men in each village, who are already in organised groups, have been mobilised to work as security escorts for women in need of urgent medical attention either for themselves or their young children at night. Insecurity at night is one of the reasons women deliver at home and why they delay seeking care when it is needed. We are piloting this context-specific security service organised and run by young people to determine the feasibility of the service, to set up the most appropriate mechanisms for contracting the youth groups, and to establish payment modalities, sustainability and best communication practices between the users and providers. Youth groups have been linked to CHVs with whom they liaise to get information on which mother or child requires their services. An emergency referral system is crucial in improving MNCH outcomes. This is especially so during hours when public transporters are no longer operating, for example, at night. On the other hand, at the time when public transport is available, it would be difficult for a non-emergency vehicle to navigate Nairobi’s traffic jams whereas emergency vehicles are given priority. Referred patients can also obtain care (such as intravenous fluids, oxygen, injections) en route to the next level facility in a well-equipped ambulance as opposed to a private vehicle. An ambulance system will be set up building on existing ambulance services in the community. The ambulance will be used to transfer mothers and children who need referral to the nearest maternity hospital. In each of the target slum settlements, one ambulance will be provided by the programme to serve the network of private providers in the area. The ambulances will be centrally located and managed from two of the upgraded health facilities. Priority will be given to facilities or organisations that already run some form of ambulance service since they may be best placed to manage the service. Appropriate communication mechanisms will be put in place to ensure that the ambulance is accessible at all times whenever it is needed. Obstetric emergency transportation systems have been plagued by problems of non-sustainability either as a result of high cost or the rarity of emergency events in the face of continued inputs.18 It is recommended that such transport systems should be complemented by quality improvements in the primary referral system and the system should be open to other users to ensure sustainability.19 20 To mitigate against the risk of non-sustainability we propose to diversify the client base for these ambulances to include non-obstetric and non-child emergencies. Different cost structures will be put in place whereby obstetric and child emergencies will pay half the price and the others will pay full price. The project team will work with professional associations in Kenya, such as the Kenya Obstetrical and Gynaecological Society (KOGS) and Kenya Paediatric Association (KPA) to run outreach clinics for women and children in slum settlements. This programme will work with the private providers to set out defined schedules for specific services and to coordinate these with KOGS and KPA specialists. The outreach clinics will add value by offering services that may not be routinely provided at level 2 facilities such as treatment for sexually transmitted infections, cervical cancer screening, long-term family planning and breast cancer screening, among others. The above activities, though aimed at strengthening the service delivery system serving slum residents, may end up interfering in the healthcare market and squeezing out credible private providers not selected for support. To guard against this, we propose to create opportunities for other eligible but unselected providers to benefit from other means of support. These will include supportive supervision from the sCHMT, participation in training programmes, support to generate routine HMIS that enhances their chances to get essential supplies from the public sector, participation in monthly meetings between private providers in each district, which are organised by the sCHMT. In figure 3, we summarise the intervention: aiming to improve the quality of services offered, improve accessibility to these services and improve efficiency in delivery of these services thereby enhancing sustainability. The proposed initiative will not create a new system, rather it will enhance the hitherto untapped synergies between the public and private sector in health service provision, strengthen existing governance structures, promote community involvement in health through the provision of critical non-health support and improve capacities to generate and utilise health information for planning, monitoring, evaluation and advocacy. As shown in figure 3, we propose that a high-quality system, accessible and in demand, will result in improved MNCH outcomes. Summary of the PAMANECH Intervention. The intervention has preintervention and postintervention project assessments for changes, if any, on the MNCH health services and population health outcomes. We conducted a baseline survey in August 2013 and plan to conduct an end-line at least 1 year after the full implementation of the intervention. In order to be able to strengthen the case for causal inference we are conducting systematic monitoring and documentation of the intervention based on our intervention (figure 3). The monitoring and documentation will also capture any other contextual factors that may influence the same outcomes as our intervention. The effects and impact of the programme will be determined by triangulating data (quantitative and qualitative) and information from different sources, examining trends where possible, and trying to find and support explanations for the observed findings (figure 4). Internationally recognised principles and standards for data collection and analysis will be followed. PAMANECH Project Data collection plan. Several data sources have been identified: Quantitative data (from the surveys) entered in netbooks will be synchronised with the master database in APHRC head office every day. Where data is missing or obvious inconsistencies are noted, the office editor will inform the data manager who will then contact the field teams for clarifications and, where need arises, send back queries to the field teams for completion of incomplete data or correction of the inconsistencies. Clean data will be exported for analysis to statistical software (STATA V.11.0; StataCorp LP, Texas, USA) for advanced cleaning and analysis. We will conduct basic tabulations and regression analyses comparing differences in the proportions of women in reproductive age and children under five years at baseline and end-line for variables such as contraceptive prevalence, vaccination coverage, skilled attended delivery, care-seeking for childhood illnesses, among others. These analyses will control for any differences in the samples (if any) at the two time points as well as the contribution of contextual factors that may have occurred in the course of the implementation. Quantitative data from the health facilities, CHVs and the client satisfaction survey will be entered into Excel work books monthly. Basic analyses of the collated data will be conducted, including the methods of median, mean and range. Qualitative data will be transcribed and saved in Word format. Transcribed Word files will be imported into NVIVO software (QSR International Pty Ltd) for coding and further analysis. Analysis across all transcripts will be conducted using a constant comparative method to identify themes and their repetitions and variations. The analysis will also aim to identify changes, if any, in various indicators, which could be attributed to the intervention. The project started in July 2012 and will last a total of 54 months (figure 5). The direct beneficiaries of the project are women of reproductive age and children under the age of 5 years in the two informal settlements who make up 20% and 14% of the population, respectively.12 In addition, five health facilities are being upgraded and the healthcare providers in the selected PNFP and other public and private health facilities are benefitting from training and skills upgrade. CHVs, the sCHMTs of the two subcounties where the study sites are located as well as the Nairobi County Health Management leadership are other direct beneficiaries. Residents of areas outside the NUDHSS as well as residents of the two slums who are male and/or older than 5 years but less than 15 years and/or older than 50 years are the indirect beneficiaries. PAMANECH Project timeline.