A total of 33,860 child patient records were extracted from a total of 116 IMNCI registers from the 23 rural health posts. Of this total, 643 (1.9%) child records from IMNCI forms for children under 2 months were excluded from the analysis. 33,217 records concerned children aged 2 to 59 months. 274 (0.8%) records had a recorded age equal to or greater than 60 months, 1,957 (5.9%) records had a missing or uncertain age, and 256 (0.8%) records had a patient age indicated as less than 2 months registered within 2 to 59 months of registration; these were all excluded from the analysis. Therefore, 30,730 records were transferred for analysis (Fig. 1).
The median age of the children was 18 months (IQR = 10.32 months) (Table 2). There were more dating recorded for male children than for female children (55.7% versus 44.3%). In terms of ethnic groups, the highest proportion of attendance was for children of Janajati ethnicity (34.9%), followed by Brahmin/Chhetri (32.7%). Madhesi and Muslim children accounted for only 0.7% combined, which would be expected given the location of health centers and the geographical distribution of Madhesi and Muslim population groups, which are mainly concentrated in the Terai (areas bass) that were not part of this study. .
42% of children were recorded as having fever, 37% diarrhea and 34% respiratory symptoms. Less than 7% of children were recorded as having an ear infection, while 2.1% had general danger signs (GDS).
Sick days before presentation to a health center
The time interval between onset of symptoms and attendance at a health center for children with ARI was lowest in Gorkha district with a median of 2 days (IQR = 2.3) ( Table 1). There were significant differences in consultation times between the five districts for ARI, diarrhea and fever with longer times for Bajura, Mugu and Humla (in the Far West) compared to other districts in the center . Janajati children were more likely to visit the health post earlier than children from other ethnic groups for all conditions except ear infections (p
Records from the years 2068 BS (2011/12 AD) (n=43), 2069 BS (2012/13 AD) (n=9) and 2077 BS (2020/21 AD) (n=77) have been deleted from the regression modeling due to convergence issues. Unadjusted and adjusted regression results for the number of days a child had ARI, diarrhea or fever before visiting a health post are shown in Tables 3, 4 and 5 respectively. For ARI, only Janajati ethnic group was significant in the unadjusted model (p
The difference in time to seek help for diarrhea among those in the Wild West (Humla and Mugu) was significant in both the unadjusted and adjusted models (both p
District and AKI were significantly associated with time to fever presentation in the adjusted model (both p ≤ 0.02, Table 5) while there was a non-significant, but associative, interaction for children of Mugu’s male sex to present earlier than females (p = 0.08). Febrile children in Humla (1.53, 95% CI: 0.83, 2.24), Mugu (0.88, 95% CI: 0.57, 1.19) and Bajura (0.59, 95% CI %: 0.25, 0.93) waited much longer than those in Gorkha.
As mentioned earlier, due to accuracy issues with data entry for tracking, it was not possible to formally analyze this portion of the information. The main reasons were that children were rarely given a Master Registration Number (MRN), similar to a unique ID for that child, and even when they were, there were accuracy issues. (sometimes the same number appeared for two different children of different sexes). We have included the information we could from the registers in Supplementary Table 1.
Correct diagnosis and treatment of pneumonia
For the children for whom we could retrospectively diagnose pneumonia based on their presenting symptoms, 22.0% had no pneumonia; 62.9% had pneumonia and 15.0% had severe pneumonia (according to IMNCI guidelines). However, among all children diagnosed with pneumonia by the health worker, 65.3% had no pneumonia, 34.0% had pneumonia, and 0.7% had severe pneumonia. The accuracy of health worker diagnoses of pneumonia could only be analyzed for 2548 children due to missing data in the variables needed to retrospectively assess the diagnosis of pneumonia from the records. Pneumonia was not correctly diagnosed in 30% of children. Children from the Madhesi and Dalit ethnic groups were less likely to have a correct diagnosis of pneumonia (36.4% and 66.3% respectively). Male children were significantly more likely to be correctly diagnosed with pneumonia (73.3% versus 67%).
Of the 2,663 children who were recorded as having severe pneumonia or pneumonia by the health worker, more than 60% did not receive correct treatment according to IMNCI guidelines (Table 6).
Children in the Janajati group had the lowest proportion receiving correct treatment, while the highest proportion receiving correct treatment was observed among Dalit children (34.1% versus 41%). This excludes Madhesi and Muslim children who were not included in the subsequent regression analysis due to small numbers.
Janajati children were twice as likely to have a correct diagnosis of pneumonia as Dalit children (odds ratio (OR) = 2.06, 95% CI: 1.16, 3.67) after adjusting for contextual factors, while children with higher body temperature also had 19% higher odds (OR = 1.19, 95% CI: 1.01, 1.41) (Table 7). Men had significantly higher odds of having a correct diagnosis than women (OR = 1.59, 95% CI: 1.27, 1.2.01). Men of ‘other’ ethnicities were 49% less likely to have a correct diagnosis than women (OR = 0.51, 95% CI: 0.30, 0.85).
Associations with correct treatment of pneumonia (given that it had been diagnosed) were not that numerous, with no significant differences in ethnicity, sex or age (all p > 0.3). The main significant relationship was if the child had been referred, with a 4 times higher probability (OR = 3.98, 95% CI: 2.23, 7.10) and a trend for male children in the Mugu district to have a lower probability of correct treatment compared to women. , although not statistically significant (OR = 0.37, 95% CI: 0.12, 1.09).
Multicollinearity was not evident in any of the regression models, with all VIF
Full-case regression models (Supplementary Tables 2-5) produced similar conclusions to multiple imputation models, although MI regression models frequently included more variables in the final multivariate model. Excluding all 828 children who had a sex reported as missing from the analysis only changed the estimates subtly and had no impact on the conclusions.