Implementation of the WHO guideline on treatment of young infants with signs of possible serious bacterial infection when hospital referral is not feasible in rural Zaria, Nigeria: Challenges and solutions

Background Bacterial infection is one of the leading causes of mortality in young infants globally. The standard practice to manage young infants with any sign of possible serious bacterial infection (PSBI) is in a hospital setting with parenteral antibiotics, which may not be feasible for majority of cases in most low resource settings. The World Health Organization developed a guideline on management of PSBI in young infants when referral is not feasible in 2016. Methods We conducted implementation research in selected communities in Zaria Local Government Areas of Kaduna State with an estimated population of 50,000 with the aim of understanding how to implement the WHO PSBI treatment guideline to achieve high coverage with low case fatality and treatment failure rates. Implementation was within the programmatic settings using existing health structure. We conducted policy dialogue with decision makers to adapt the recommendations to their social, cultural and programmatic context in Nigeria, held orientation meetings with program managers, built capacity of the health workers and supported the implementation within the health system. We supported a non-government organization to conduct community sensitization to promote care seeking and adherence to treatment advice. The research team collected data systematically on all young infants identified to have PSBI, the treatment they received and the clinical outcome. Results Between April 2016 and March 2017, we identified 347 young infants up to 2 months of age with signs of PSBI who received treatment either as an outpatient or in a hospital among 2,154 births in the study population. The coverage of PSBI treatment in the study area was 95.5% assuming that 10% of all births have an episode of PSBI in the first two months of life. Most (89%) sick young infants with PSBI were identified by the community-oriented resource persons and sent to the Primary Health Care Centres (PHCs). Most families (97%) refused referral and were treated at a primary health care centre on outpatient basis. There were 12 deaths (3.5%) and 17 non-death treatment failures (4.9%) in 343 infants in whom an outcome could be ascertained. While non-death treatment failure rate was highest in 0-6-day infants with fast breathing (14.4%), case fatality was highest in those with signs of critical illness (20%). Conclusion We have demonstrated that outpatient treatment strategy for young infants with PSBI when referral is not feasible is implementable within the programmatic settings, achieving very high population coverage and relatively low treatment failure and case fatality rates. Implementation at scale will require government’s commitment to strengthen the health system with trained, motivated health care providers and necessary commodities. The implementation research was conducted in Zaria Local Government Area (LGA), located in the northern part of Kaduna State in the North West geo-political zone of Nigeria. A policy dialogue at the central level was held in Abuja, in which State and local government functionaries and other technical experts took part. It was decided that the national policy should be adapted to implement WHO PSBI guideline when referral is not feasible in two places in Nigeria, one in the North part of the country and one in the South to represent both diverse regions. In the North part, Kaduna state was proposed and in South part, Oyo State was proposed. Zaria LGA is representative of a population where referral is not accepted by many families with sick young infants. We also held consultations with the local authorities and in consultation chose certain wards in Zaria LGA to implement this guideline. Zaria LGA is the headquarters of Zazzau Emirate and the location of the traditional ancient city of Zaria. It is made up of more than 100 urban, peri-urban and rural settlements populated by predominantly Hausa/Fulani Moslems. The LGA has a population of 300,000.[13] The LGA is divided into 11 political wards. Each settlement has a village head who is appointed by the Emir of Zaria, the paramount ruler of Zazzau Emirate.[14]The status of the women in the LGA and other parts of northern Nigeria is comparatively low, with limited access to formal education, marginalization in decision making, even in matters of health decision-making, economic dependence and limitation of movement and work through the widespread practice of female seclusion being the norm. Husbands and traditional/religious leaders are important gatekeepers, while mother in-laws and co-wives are very significant others; all these people play important role in health seeking decisions.[15] The main profession of the people in these communities is subsistence farming. The infant mortality and neonatal mortality rates in Nigeria are 67/1000 and 39/1000 livebirths, respectively.[16] Local studies in communities in Zaria LGA have shown that the LGAs have a higher than national crude birth rates ranging between 45-50/1000 population. [17] As it is for other parts of Kaduna state, the coverage for health intervention such as basic vaccination is poor, with only 20% of children 12–23 months receiving all basic vaccination at any time during the 2018 National demographic and health survey.[16] In these communities, more than half of the pregnant women attend antenatal clinic but most deliveries take place at home with family members in attendance.[15,18] While the Traditional Birth Attendants play a limited role during pregnancy, they play a major role during delivery and baby care until the time of naming ceremony usually by the seventh day after birth. Home treatment, traditional medicine and use of patent medicine vendors are usual first line actions in the event of ill health for the child. The research was carried out in two rural political wards of Zaria LGA covering over 50 communities scattered over variable distances from their nearest Primary Health Care centres (PHCs). These two political wards were chosen as they are among the most rural of these political wards and with limited accessibility to referral centers. Some communities are up to 15 kilometres away from the nearest health facility. The selected catchment areas for the implementation research had an estimated population of under 50,000 and 2,500 births per annum. We expected about 250 PSBI cases per year as seen in the AFRINEST study.[5] Primary health care is provided through PHCs, which serve a population of 5000–6000, where patients self-present. The PHCs are primarily staffed by nurses, Community Health Extension Workers (CHEWs) and midwives. Some PHCs are only staffed by CHEWs. General hospitals provide secondary health care services and are staffed with medical doctors and nurses. In some general hospitals, there may be specialist doctors trained in pediatrics or neonatology and a newborn nursery. Medicines are in principle supplied to the PHCs by LGA through the Health Department of the LGA. At the time of this research, Kaduna State Government operated a free Maternal and Child Health service as a policy. However, the state also operated a Sustainable Drug Supply System (SDSS), whereby the State Government purchased drugs as seed drugs and distributed to the various LGAs. “Seed drug” is a one-off free bulk supply of essential medicine by the state government to the LGAs or PHCs that is expected to be replenished by the funds generated from the sale of the drugs. The LGAs in turn supplied the PHCs with a seed drug supply. The PHC staff dispensed these drugs to patients on cash, which was used to replenish their stock from the LGA stores. The LGA in turn used the money generated to replenish their stock from the State central medicines store. Although Integrated Community Case Management (iCCM)[19–20]as a national strategy was operational in Nigeria, this had not been introduced in communities within Zaria LGA. Families often sought care for their sick young infants from traditional healers, chemists or the nearby PHC. At the PHCs, health workers in general referred sick young infants to either Hajiya Gambo Sawaba General Hospital, Zaria or the Ahmadu Bello University Teaching Hospital, Shika at a distance of about 25 kilometres. Transportation is limited and access to referral level facilities is poor. We adapted the RE-AIM framework [21] for planning and evaluation of this study. Fig 1 explains the conceptual framework used in this study based on the RE-AIM framework. The implementation framework was based on the RE-AIM framework. RE-AIM, Research Effectiveness Adoption Implementation Maintenance. Federal Ministry of Health (MOH) with support from WHO organized a national level orientation and policy dialogue with national and state level stakeholders in Abuja in March 2015 to review the WHO guideline and make informed decision about its implementation in Nigeria. Policy makers, programme implementers within the MOH and non-governmental organizations (NGOs) as well as the paediatric community participated in the meeting. The meeting created a platform for dialogue and for decisions on small scale implementation for gaining experiences and insights into policy requirements. Key issues were raised and collective decisions were taken. Details of these decisions were conveyed to WHO through an official letter (FHD/CH/3054/1) dated 23rd April 2015 (Table 1). Fig 2 provides definitions of health workers category, setting in which they practice and the scope of their practices. Abbreviations PHC; primary health care centre CORPs; community-oriented resource persons VHW; village health workers Chews; community health extension workers PPMV; patent private medicine vendors; (private individuals operating a chemist shop who are not qualified pharmacists) CHO; community health officers We held a meeting with the Zaria LGA authorities and officials of the health department at the local government secretariat on the 18th September 2015. We shared the outcome of the national orientation and policy dialogue and the decisions about PSBI management and the proposal to conduct the implementation research in the PHCs of two wards, Wucicciri and Dutsen Abba, these being the most rural of the political wards in the LGA. The authorities welcomed the initiative and agreed with the proposal and with the strategy which included involvement of health workers already working at the PHCs in the research. We organized another orientation meeting with the community leaders for the two political wards. The meeting was conducted in the compound of the ward district leaders within the communities respectively on the 6th and 7th of October 2015. The participants in this meeting included district head of the wards, village Sarkis (village heads), the mai unguwas (sub-village heads), heads of households from various communities and the PHC staff. We described the purpose of the implementation research and potential benefits to the communities. As there were no community health workers in the community to help identify pregnancies and births, we discussed the selection of Community Oriented Resource Persons (CORP) from their communities, who would be young men or women who met a minimum education standard stipulated by the Federal Government policy. The CORPs would be hired to work in the community to identify pregnant women and births. The implementation research was implemented within the existing health system of the state, so we assessed the status of health facilities and human resources in the selected wards to inform the implementation of the guideline using the modified version of “Equipment and supply checklist” from the WHO health facility survey.[22] In Dutsen Abba ward, there were three PHCs and in Wuciciri ward there were two PHCs. In one PHC there was no staff, one PHC had two Community Health Extension Workers (CHEWs) and others had only one CHEW. Some of the communities expected to be served by the PHC located at Wuciciri were too far away from the PHC to the extent of limiting accessibility by families of sick infants. The CHEWs posted at PHCs had received training in 2014 but did not have any refresher training recently. Essential medicines for the treatment of young infants with signs of PSBI (gentamicin, amoxicillin and ampicillin) when referral for hospital is not possible were not available routinely in the health facilities. A technical Support Unit (TSU) was established to help set up the study site, collaborate and coordinate with the health authorities, providing training and technical support to the CORPs and PHC staff, monitor quality of implementation, ensure supplies, identify challenges and solutions to overcome them. The TSU members included the principal investigator, project manager, two field supervisors and six CHEWs (one each from the study PHCs). The project manager assisted the PI in the day to day coordination of activities, supervision of the study staff. The project manager was also trainer for Community Newborn Care and also contributed to the adaptation of the WHO Integrated Management of Childhood Illness (IMCI) training package for the training of CORPs and Nurses/CHEWs. The TSU addressed some of the pre-implementation issues. Training of health workers: Training on assessment, classification and management of young infants with PSBI signs when referral is not feasible using the young infant component of IMCI training package [12, 23] was conducted in two stages. Initially, Master Trainers from all the study sites were trained by WHO facilitators for three days between the 9th and 11th of November 2015. Subsequently, the Master trainers provided onsite training to the Nurses and CHEWs in Zaria. The training which was conducted onsite in Zaria, consisted of six days (15th to 19th February 2016) of interactive communication, clinical demonstrations and hands on skills practices in small groups. Trainees were taken to the Paediatric Department of the nearby hospitals to learn identification of signs of illness at patients’ bedside. The training included sessions on assessment of treatment outcomes (i.e., clinical treatment failures including death, adherence to therapy to both gentamicin and oral amoxicillin and referral for hospitalization). A second training was an onsite 6- day training on home visits for care of the newborn and identifying sick young infants for CHEWs and CORPs conducted by the Master trainers supported by the study coordinators using the WHO training manual for care of the newborn at home.[24] Field supervision: Two field supervisors supervised the PHC nurses/CHEWs and CORPs working in the community. They conducted quality assurance checks through standardization exercises, reviewed all completed study data collection forms by the nurses/CHEWs, cross checked drug dosages and administration by both nurse/CHEWs, and checked functional status of all working tools in the field. They also resolved problems encountered by the field staff on a daily basis. The Field Supervisors who had received a training of trainers’ course during the AFRINEST study also served as trainers during the WHO/UNICEF Training course for Community Health Workers on the Caring for the Newborn at home and contributed to the adaptation of IMCI training for CORPs and nurse/CHEWs. The Zaria LGA Monitoring and Evaluation officer worked with the field supervisors to effectively supervise the activities of the CORPs in the communities and Nurses/CHEWs in the PHCs. CORPs, identified by the communities were given an honorarium and transport allowance. They worked with the traditional birth attendants within their assigned communities to identify pregnancies and births. The CORPs registered pregnant women and conducted two home visits to promote antenatal care and delivery by a skilled birth attendant at the PHC as well as birth preparedness. The CORPs followed the newborns on day 1, 3, and 7 after birth to promote essential newborn care [24]and to assess signs of illness. Small babies were also visited on day 2. The newborns were followed for up to two months of age. Any young infant identified with a danger sign was referred to the nearby PHC for further assessment. If the families refused to take their sick young infant to the PHC, the CORPs invited the nurse/CHEWs to visit the young infant’s home for further assessment. Traditional Birth Attendants were given transport and telephone incentives to facilitate movement and communication. The PHC nurse/CHEWs assessed sick young infants sent by the COPRs or brought by families for signs of PSBI using IMCI [17]. Those who required referral were referred to the Special Care Baby Unit (SCBU) of Ahmadu Bello University Teaching Hospital, Shika, Zaria. They were counseled about the need for hospitalization for their infant. When families refused hospital referral despite the best efforts of the PHC nurse/CHEW, their sick young infants with PSBI signs were re-classified into either fast breathing pneumonia, clinical severe infection (CSI), severe pneumonia or critical illness (Fig 3). PSBI, possible serious bacterial infection. A syndromic classification for health workers using the IMCI algorithm. Young infants 7–59 days of age with only fast breathing were treated with oral amoxicillin without referral. Infants with CSI were offered treatment with simplified antibiotic regimens on outpatient basis. Treatment was initiated after written informed parental consent was obtained. Young infants with critical illness were referred again, but if families still refused referral they were treated with intramuscular gentamicin and ampicillin daily while reinforcing referral. Injections and first daily dose of oral medicine were administered by the PHC nurses/CHEWs at the PHC or at home (if the infant was not brought for second injection). Second oral dose for the day was administered by mothers at their homes. For infants who vomited within 20 minutes of oral dosing, the caregivers were instructed to re-administer a complete dose. Under treatment, infants were evaluated daily by the same nurse/treating health worker who provided injectable treatment. CORPs visited the young infants who were on only oral therapy at home. All infants treated at home were followed up on days, 2, 4, 5, 7, and 14 after enrollment. On each follow-up visit, the infant was examined for signs of improvement or deterioration and for any adverse effects. Records of treatment received and follow up visits were documented by the treating nurse or CHEW. Parents were counseled on signs of deterioration and were asked to bring the infant back to the PHC if they noticed any sign. Infant whose condition deteriorated at any visit, was referred to the hospital for further management and counseled on the importance of hospital treatment. PHC worker facilitated referral by counseling and providing a referral slip. However, for those who still refused, treatment was offered by the treating PHC worker with the available medicines. The evaluation for the various outcomes were studied systematically. All infants receiving treatment were assessed for outcomes and documented on the study case record forms. The documentation of challenges was based on team meetings that included field workers, health workers at the facilities, supervisors and health facility managers. It also included field assessments of the health facilities, asking families during follow-up visits to understand the barriers and identify facilitators to implement this intervention The required commodities to manage PSBI on outpatient basis were not available through the routine health system at the PHCs. TSU provided the medicines (injection gentamicin, injection ampicillin and oral dispersible amoxicillin tablets) and other commodities such as weighing scales, thermometers and respiratory rate timers.