Lower mortality is observed among low birth weight young infants who have received home-based care by female community health volunteers in rural Nepal

Nepal is a low-income country of 26.5 million inhabitants, 78% of which live on less than two dollars a day and 83% of which live in rural and often remote areas [16, 17]. Its diverse ecological regions and unstable political situation create challenges for systematizing healthcare delivery [18]. In Nepal, on average only 50% of pregnant mothers attend the recommended four antenatal care visits at health facilities, and only 45% of mothers attend postnatal care, and this failure to appear is a risk factor for negative pregnancy outcomes, including LBW newborns [6].

Morang Innovative Neonatal Intervention (MINI) program focused on community based management and treatment of illnesses of newborn children in the Morang district of southeastern Nepal from 2005 to 2009 [19]. Morang is the second most populous district of Nepal, after Kathmandu, yet 80% of the total population of 914,799 live in rural areas. The per capita yearly income for Morang was USD 297 and the overall literacy rate was reported as 56% (female 47% and male 67%) [20, 21]. Morang district has 65 Village Development Committees (VDCs), two hospitals, seven primary health care centers, ten health posts, and 49 sub-health posts. There were 585 FCHVs in the District at the time of the study.

MINI was a prospective intervention study aimed at managing newborn’s illnesses at household and community level. MINI began in 21 randomly selected VDCs in Morang, but due to initial successes and demand from local government from the VDC level and the District Public Health Office (DPHO) level, it was rapidly scaled-up to the entire Morang district covering all 65 VDCs. This is a cohort study utilizing the data from the MINI program to evaluate the risk of deaths in LBW infants who received follow up FCHV visit or not.

FCHVs were trained to measure birth weight on their initial visit and to do follow-up visits to advise on care of LBW babies. The choice to do a follow-up visit was voluntary and they did not receive any incentive for doing such follow-up visits. During the first visit, FCHVs recorded date of birth, birth place, birth condition, weight and assessed for any infection, as well as providing counseling on essential newborn care. For LBW babies, they conducted four additional follow-up visits on a weekly basis. On each follow up visit, they reassessed the young infant for signs of Possible Severe Bacterial Infection (PSBI) [22, 23], advising mothers to seek care immediately if any of these signs appeared, and counseling the mother on exclusive and frequent breastfeeding and skin-to-skin contact. FCHVs used topical antibiotics for Local Bacterial Infection (LBI- eye and skin infection) and initiated treatment with oral Cotrimoxazole-P and referred to facility-based health workers for Gentamicin injections for PSBI [14, 24]. FCHVs were already familiar with using Cotrimoxazole-P for treatment of pneumonia in older children and this antibiotic was successfully used in earlier study [24]. The infants referred for PSBI continued to get weekly follow up visits.

During training of FCHVs in MINI, they were provided with Salter scales and trained to measure the weight of newborns. Given the varying levels of literacy among FCHVs (38% of FCHVs in Nepal are illiterate) [15], MINI modified the Salter scale by coding weight ranges with colors to simplify measurements. The red, yellow and green portions of the scale signified Very Low Birth Weight (VLBW), LBW and normal birth weight respectively. Newborns weighing <2000 g were categorized VLBW (red color), those weighing between 2000 g – 2500 g were categorized LBW (yellow), and those weighing >2500 g were considered normal weight (green). If FCHVs circled the VLBW image in the birth weight section, they referred the infant and family to the nearest health facility. If the family did not comply with health facility referral, FCHVs visited the home, as they would for a LBW case, and marked the visit number on the recording form. Families with LBW babies were counseled further on newborn care. The FCHV then conducted four additional home visits (weekly for one month) for the LBW neonates.

Regardless of FCHVs follow up visits, the FCHV conducted a two-month (in 60 or more days) visit to determine the infant’s survival status at 60th day. MINI’s trained field supervisors collected and consolidated data from FCHVs on a regular basis. Standard forms and formats were developed and used for this purpose.

Any infants with birth weight measured three days after birth, weight not-taken or survival status not known at the two-month were excluded from the analysis in order to maintain data consistency. Similarly, for comparing infants who received no follow visit and those who received at least one follow up visit by the FCHV for LBW infants, neonatal deaths within one week of birth were excluded as these population were not able to receive intervention from FCHVs. Furthermore, those infant (n = 97), who were referred and taken to health facility received advanced care from referral hospital, were excluded from analysis as including them would over-estimate the effect of FCHVs visit. Ethnicity classification was done based on the Health Management Information System (HMIS) of the Government of Nepal, which categorizes ethnicity in six categories namely, (i) Dalits, (ii) Disadvantaged Janajatis, (iii) Disadvantaged Non-Dalit Terai Caste Groups, (iv) Religious Minorities, (v) Relatively Advantaged Janajatis, and (vi) Upper Caste Groups [25]. Chi-square test was used to identify statistical difference in mortality outcomes between FCHV-visited and not-visited LBW neonates. Relative risk was calculated after adjusting for birth place to measure the relative risk reduction of death among LBW neonates who received an FCHV follow up visit compared to those who did not. P-values <0·05 were considered statistically significant.

Informed oral consent of the infant’s legal guardian was obtained. The Ministry of Health and Population, Nepal approved the intervention. Ethical approval to conduct this study was obtained from the Western Institutional Review Board (WIRB) (WIRB Pro No. 20031870).

The exclusion and inclusion criteria for this study are presented in Fig. 1. Out of 51,854 young infants recorded in the MINI database, 51,020 were live births, 716 were stillbirths and 118 did not have any information on birth condition. Among the live births, 11,075 did not have any information on birth weight and for 19,901 weights were measured after three days. Among 20,044 live births visited by FCHVs within three days, 17,462 had normal weight, 2103 had low birth weight, 327 had very low birth weight and 152 did not have any information on birth weight. From the low birth weight cohort, including very low birth weight, seven were lost to follow up, 97 were taken to a health facility after the referral by FCHV and 97 died within a week. The remaining 2229 births were included in the analysis.

The detailed characteristics of the young infants who were included in the analyses are presented in Table 1. Among 2229 low birth weight young infants, 53% were female. Ethnically, the majority of the infants were Disadvantaged Janajati (40%) followed by Disadvantaged Dalit (21%). There were 49 (2%) deaths between eight days and two months of age. The majority of births occurred at home (78%). Two hundred and eighty-six young infants had PSBI, among them 177 were identified by FCHVs and initiated treatment with oral antibiotic and referred for injectable antibiotics. The remaining 109 PSBIs were diagnosed by the local health workers from the health facilities. Likewise, 464 young infants had LBI and were treated by FCHVs using topical antibiotics. Among the 2229 LBW infants, 121 (5·4%) received no follow-up visits. No statistically significant differences between those with and without follow-up visits were observed by sex (P = 0·09), illness (P = 0·37), first place of treatment (P = 0·59) and ethnicity (P = 0·48). However, these two groups were s significantly different by birth-place (p = 0.03) (Table 1).

Mortality status at 60th day in LBW infants are shown in Fig. 2. The birth place adjusted proportion of deaths among those who received FCHV follow up visit and those who did not receive were 2% (95% CI: 1%; 2%) and 11% (95% CI: 6%; 18%) respectively (P < 0.001). Further disaggregation of data showed the highest difference was observed for Dalits (27% vs 1·1%) and Disadvantaged Non-Dalit Terai Caste ethnic groups (27·0% vs 2·0%). The relative risk of death in LBW infants who received FCHV follow up visit was 84% less as compared to LBW infants who did not receive (RR = 0·16; 95% CI: 0·09, 0·29).