Summary details of thereferral records from the 13 peripheral hospitals and the zonal referral hospital are shown in Table and Figure . Eighty-two sick neonateswere enrolled from the 13 peripheralfacilities and their case notes examined. The mean (standard deviation, SD) age at time of case note review was 8.7 (± 10.6) days.
Hospital data on neonatal births, deaths 1st January to 31st December 2010
The locations of neonatal care within each facility (i.e. maternal, paediatric, or special dedicated wards) are described in Table . The neonatal chart notes, or records of care, were found in disparate locations in the different facilities: in 3 facilities neonatal case notes were found in local school exercise books “daftari” as shown in Figure , 8 hospitals detailed neonatal care in the mothers’ case notes, and 2 hospitals used standard medical files specific to the neonate. Antenatal cards were held by the mothers in 12 (92.3%) of the 13 facilities and were located in facility-based records in one hospital.
Staffing numbers and levels (by WISN) dedicated to neonatal care 1st January to 31st December 2010 in the peripheral facilities visited
Figure 2 Records of care in Daftari. An example of primary school exercise books “Daftari” used in making records of neonatal care in one of the district hospital. On the right side is the sample of antenatal card we used to make assessment of (more ...)
Data extracted from antenatal cards
Antenatal cards of the all included neonates were assessed. The majority of mothers [59(71.9%)] received antenatal care at facilities different from where they ultimately delivered their infants. Overall, 72 (88.0%) mothershad attended an antenatal clinic at least three times (mean 2.95±1.3 visits) during their pregnancy and all 82 enrolled neonates were born in health care facilities.
Records for Rh factor and blood group were present on antenatal cards in 39 (47.6%) and 52 (63.4%) cases respectively while syphilis or HIV screening was recorded on 59 (71.9%) and 77 (93.1%) antenatal cards respectively. Among 77 mothers who had received counselling, testing, and/or services for prevention of maternal to child transmission of HIV (PMTCT), 64 (83.0%) were HIV negative, 12 (15.6%) were HIV positive, and 1 (1.3%) had declined testing.
Despite the 100% availability of APGAR score records in the delivery books at the included facilities, these records had been copied onto the antenatal cards in only 61.0% (50 records out of 82) cases. Birth weight records were available in 65 case notes (79.3%) despite all the infants having been born in a health care facility. The median birth weight of neonates whose files were evaluated was 3000 grams with an interquartile range of 2600–3350 grams.
Clinical data on neonates present on the ward
Temperature was the most commonly recorded vital sign, found in 33 (40.2%) case notes. Two thirds (66.0%) of admitted neonates with records of temperature had a recording ≥38.0°C. Admitting diagnoses as abstracted from the case notes the 82 enrolled neonates are shown in Figure ; 40 of the neonates had more than one admitting diagnosis. Prematurity, defined as a gestation age less than 36 weeks, was documented in only 11(9.0%) case notes while calculation from the antenatal card indicated that 36(45.6%) of 79 neonates with LMP data had a gestational age less than 36 weeks.
Diagnoses of admitted sick neonates present on the maternity or paediatric ward at the time of assessment.
Documentation and tracking referrals
Records of referrals were made in Ministry of Health forms and discordance between this record and documented arrival at the zonal referral hospital is shown in Table . In no cases did the number of referred and received cases match, and in one case 100% of the received patients were unaccounted for at the referring institution.
A record of full blood count was found in 4 (4.8%) cases, and haemoglobin wasdocumentedin 6 (7.3%) case notes. Nine (11%) case notes had a record of blood sugar while 22 cases had a record of altered consciousness or convulsions that would normally be an indication to measure blood sugar. There were no laboratory investigations for blood culture or bilirubin as none of the facilities had capacity to perform these tests. Overall, only 8 (9.8%) of case notes of the sick neonates in the district hospitals had record of any laboratory investigation.
Prescriptions of oxygen, I.V fluids and Antibiotics
There were 16 neonates who, according to the record of care, met indications for oxygen therapy, but only 6 (37.5% of cases in which it was indicated) were documented to have received it. Only 8 neonates (9.7%) had documentation of IV fluids, whereas we estimate that 24 (29.3%) needed the intravenous fluid infusions due to clinical presentations of hypoglycaemia [glucose < 30 mg/dl] and/or a combination of high grade fever and inability to breast feed. Estimation of correct dosing of antimicrobial agents according to WHO guidelines is shown in Figure .
Assessment of prescriptions of doses given to sick neonates admitted in the district hospitals.
Staffing of health workers
Staff availability is presented in Table . WISN for clinicians was 0.93 (ranging from 0.1 to 2.4), for nurses was 12.2 (ranging from 0.6 to 47.3), and for nurse attendants was 5.34 (ranging from 0.3 to 22.5).
Supply of essential drugs and equipment
Basic equipment was commonly available at the district hospitals with a relatively poor availability at the two heath centres (Figure ). Nine out of 13 peripheral hospitals (72.7%) in Kilimanjaro had visible stocks of essential drugs, oxygen and fluids to manage common neonatal infections and other disorders of mild severity such as transient tachypnoea of the new-born. These observations were subjectively well correlated with estimated availability in months per year reported by health workers interview (Figure ).
Availability of essential equipments as observed from the facility inspection on the day of survey and aggregated opinion of health staff on general availability of same items.
Availability of essential drugs and infusions as observed from the facility inspection on the day of survey and aggregated opinion of health staff on general availability of same items.