Readiness of rural health facilities to provide immediate postpartum care in Uganda

Background: Nearly 60% of maternal and 45% of newborn deaths occur within 24 h after delivery. Immediate postpartum monitoring could avert death from preventable causes including postpartum hemorrhage, and eclampsia among mothers, and birth asphyxia, hypothermia, and sepsis for babies. We aimed at assessing facility readiness for the provision of postpartum care within the immediate postpartum period. Methods: A cross-sectional study involving 40 health facilities within the greater Mpigi region, Uganda, was done. An adapted health facility assessment tool was employed in data collection. Data were double-entered into Epi Data version 4.2 and analyzed using STATA version 13 and presented using descriptive statistics. Results: Facility readiness for the provision of postpartum care was low (median score 24% (IQR: 18.7, 26.7). Availability, and use of up-to-date, policies, guidelines and written clinical protocols for identifying, monitoring, and managing postpartum care were inconsistent across all levels of care. Lack of or non-functional equipment poses challenges for screening, diagnosing, and treating postnatal emergencies. Frequent stock-outs of essential drugs and supplies, particularly, hydralazine, antibiotics, oxygen, and blood products for transfusions were more common at health centers compared to hospitals. Inadequate human resources and sub-optimal supplies inhibit the proper functioning of health facilities and impact the quality of postpartum care. Overall, private not-for-profit health facilities had higher facility readiness scores. Conclusions: Our findings suggest sub-optimal rural health facility readiness to assess, monitor, and manage postpartum emergencies to reduce the risk of preventable maternal/newborn morbidity and mortality. Strengthening health system inputs and supply side factors could improve facility capacity to provide quality postpartum care.

This was a cross sectional study that employed quantitative methods of data collection using a facility assessment tool. This study design was adopted because we sought to collect data at a single point in time to assess how prepared the facilities were to provide care to women and babies within the immediate postpartum period. The study setting was in the greater Mpigi region found in central Uganda, which consists of three districts namely: Butambala, Gomba, and Mpigi. This region was chosen because it is centrally located and it has health facilities in both the peri-urban and rural settings, which serve a wide population. The health system in Uganda has several tiers including: health centers I, II, III, IV; general hospitals; regional referral hospitals; national referral hospitals; and national specialized hospitals. A health center I is located at the village level and provides basic home-based care and health promotion. The care at this level is provided by the village health teams comprised of volunteers in the community. Health center IIs are located at the parish level and provide outpatient curative services and outreach for health promotion. At the sub-county level, we have health center IIIs, which provide all the above and inpatient medical, antenatal and maternity care. Next is the health center IV at the health sub-district, where emergency surgical and maternity care, blood transfusion and laboratory services are available, together with the health promotion and outpatient curative care. Then we have the general hospital at the district level that provides all the care and services provided by the health center IVs, in addition to general surgical care, mental health care, dental care and some specialized services like physiotherapy. At the regional referral hospitals, clients can access specialized medical services. These also offer training for different cadres of health workers. There are 9 national-level hospitals, of which 5 are specialized. None of these national referral or specialized hospitals were included in this study. The target population for this study were all the health facilities in our study area that provide intrapartum care services and health facility based postpartum care which include: hospitals, health center IVs and health center IIIs according to the current Ministry of Health (MOH) policy guidelines for health care services [14, 21]. The facilities in the greater Mpigi region that were eligible to participate in this study were those that conducted deliveries and had been providing the service for at least 6 months at the time of the data collection. The three district health officers and all included facility administrators provided administrative clearance for the study to be conducted at their health facilities. Thereafter, permission was sought from the maternity ward in-charges to conduct the study on their units. Informed consent was sought from all the midwives and mothers who participated in the study. The sample size of 40 health facilities was calculated using the simplified formula by Yamane [22] at a confidence interval of 95%, error of 5%, and a total population of 44 health facilities in the greater Mpigi region that provided delivery services based on the national health facilities inventory [21]. There were two hospitals, three health center IVs and 39 health center IIIs in the official list of registered health facilities in the three districts [21]. Out of 44 health facilities, 40 health facilities were assessed in this study. All the hospitals and three health center IVs in the three districts plus all the health center IIIs in Butambala and Gomba were purposively included in the study because they were few. Health center IIIs within Mpigi district were recruited consecutively until all the required sample (for Mpigi district) of 17 health center IIIs was reached. The data were collected using a facility assessment tool that has been adapted from the Ministry of Health’s original result-based financing (RBF) assessment and the SARA facility assessment tools [23]. The MOH tool’s sections were retained but adjusted to focus on postpartum care provision. The tool has eight sections including; presence of a written policy on hospital stay after delivery (scored as 5 or 0), availability of a checklist for routine monitoring of women post-delivery (scored as (15 to 0), presence of a clean, private, dedicated area for postpartum care (scored as 5 or 0). presence of tracer medicines and commodities for basic obstetric care (oxytocin, misoprostol, magnesium sulphate, amoxicillin, metronidazole, vitamin K, tetracycline chlorhexidine gel, normal saline, ringers lactate, 50% dextrose, surgical gloves, examination gloves, sutures (vicryl), gauze, alcohol, hibitane solution and surgical blades) (scored as 5 or 0), availability of a checklist for counselling postpartum women (scored as 15 to 0), availability of tracer equipment for postpartum care provision (blood pressure machines, thermometer. tape measures, glucometer, pulse oximeters) (scored as 15 or 0). presence of a functional transport system (scored as 5 or 0), presence of a viable client information and educational program (scored as 10 or 0), Each section was scored separately and an overall score for the facility was computed with a maximum score of 70 and a minimum score of zero. The scores were then transformed into percentage scores. The median scores per district, facility level, and facility type were computed. The tool was checked for content validity by DKK and GKN, then pilot tested by assessing two senior midwives working at a referral hospital in Kampala. Changes were made to the tool as suggested by the midwives and the two coauthors. During the training of the research assistant, care was taken to minimize inter-rater variability to increase the reliability of the scores. Whenever possible, the two assessors both scored the facility visited and compared their scores before agreeing on the final facility score. The dependent variable for this study was the facility readiness for the provision of postpartum care, which was the median score for all the facilities based on the facility readiness assessment scores. The independent variables included: facility level, ownership, number of midwives, district, average number of deliveries per month, and number of postnatal beds. Data were collected over a period of 4 months between August and December 2020. After verifying the functionality of the facilities identified in the national health facility inventory [21] with each district health office, each facility in-charge or medical superintendent was contacted by telephone or email and briefed about the study before the facility was visited. Those that responded positively were scheduled for a visit and the in-charge was informed of the possible date for the visit. On arrival at the health facility, administrative clearance was sought from the in-charge who introduced the study team to the midwife on duty or the maternity in-charge. The team then explained what they needed to do and proceeded to complete the assessment with the help of either the midwife or the maternity in-charge. Those facilities where administrative clearance was not obtained were excluded from the study. Various items were assessed in the facility readiness assessment tool including: length of hospital stay following a normal delivery; routine monitoring of postpartum care; availability of space; tracer medicines; commodities; equipment; sundries; a functional transport system; and a health education program for the postpartum women and their care givers. Completion of the assessment tool included verification of the presence or absence of the various equipment, drugs, postnatal charts, MOH guidelines for postpartum care, standard operating procedures for the provision of postpartum care, visual reminders for midwives / checklists for care provision, discharge guidelines, health education guides (models, flip charts, posters, time table, and other teaching aids), brochures for information provision, examination room / couch, cleaning materials and cleaning schedule for the postnatal unit. For sections where there was observation of care being provided, the team requested to be present while care was being provided to a postpartum mother so as to observe the care provided to the client. No identifying information regarding the clients was collected, as the researchers were only interested in observing what the midwife did in providing care. The clients were informed about the study and requested for consent to be observed while receiving care. In facilities where there were no clients found at the time of the visit, the midwife was asked to describe what she does for each client from the time of delivery of the baby to the time of discharge, and this was assessed against the postpartum care guidelines for both the mother and newborn. The midwife was then asked to avail the team with policy guidelines and checklists that were being used for the provision of care and health education. She also took the team to the ward, store and laboratory where they checked for the availability of drugs, sundries (IV giving sets, blood giving sets, cotton, gauze, examination gloves, surgical gloves, and gynecological gloves), equipment, blood products, and laboratory equipment. Regarding the postpartum counseling done and documented by the midwife. We interviewed one midwife at each health center III and health center IV. At the hospitals, we interviewed and observed 3 midwives providing in-facility postpartum care but only one midwife who was the maternity in-charge assisted the researchers in ascertaining the presence of the required documents, drugs and equipment. The number of midwives assessed was based on the guidance from the tool and the availability of the midwives/ patients at the health facility at the time of the facility assessment. The team spent at least 1 day at each facility depending on the availability of the facility in-charge, midwife and the postpartum clients. Each filled assessment tool was reviewed for completeness before leaving the facility and any missed observations were completed before departure from the facility. Data were double entered into Epi data version 3.1, cleaned, and exported to Stata version 14.2 for analysis [24, 25]. Univariate analysis was performed for health facility characteristics and presented as frequencies and percentages. All facilities that scored 80% and above were considered to be prepared to provide postpartum care while those that scored below the 80% were considered not ready to provide postpartum care. The facility readiness score was also reported in a table reflecting minimum and maximum scores and the median per district because the scores were skewed to the left. Higher scores denoted that a facility was better prepared to provide postpartum care.

N/A

• Dima Health