The mean age of mothers’ at the time of interview was 30.96 + 7.215 years. Fifty percent of mothers were between 25 and 36 years of age. Six hundred thirty eight (71.4%) of mothers did not attend any formal education, 768 (85.5%) were married and 662 (74%) of them were housewives.
The HEWs’ average age was 26 + 3.67 years and 13 (81.3%) of them were married and they walked an average of around 3 hours (95% CI: 1:10 – 3:20) to reach to the farthest mother’s home.
Coverage of FANC
Seven hundred fifty two mothers were contacted by health care providers at least once during their recent pregnancy, making the overall antenatal care coverage of 83.7% (95% CI: 81.28 – 86.12). From the contacted included mothers, 44.2% (95% CI, 40.95 – 47.45) were contacted by HEWs, while the remaining 39.5% (95% CI, 36.01- 42.99) were contacted by skilled providers. The mean time of first antenatal care visit was 4.14 + 2 months. Interestingly, from the total contacted mothers 34.44% (95% CI, 31.04 – 37.84) of them were seen by health care providers within the first trimester of gestation (less than 12weeks).
Frequency of FANC service utilization
The mean number of the antenatal visits were 3 + 1.6, and on average, 34.97% (95% CI: of 31.56 – 38.39) of mothers received at least four antenatal care visits during their recent pregnancy. Higher number of antenatal visit was related with increased FANC package contents provided to mothers (Figure 1).
Components of FANC
On average 6.12% (95% CI: 4.40-7.83) of the contacted mothers received all the components of focused antenatal care. Of these, 33(4.3%) received the care by HEWs and 13(1.7%) by skilled providers (Table 1).
The overall weighted average FANC package intervention fidelity was 49.78% (95% CI: 47.73 – 51.83). HEWs provided 62.02% (95% CI: 59.71 – 64.32) while skilled providers provided 56.57% (95% CI: 53.94 – 59.19) of the weighted average FANC intervention fidelity. Only 2.2% (n=20) of mothers received all the recommended FANC package intervention with full fidelity.
Provider related factors
Three quarters (n=12) of the HEWs were ever trained on FANC package while only 2 (12.5%) of them received refreshment training in the last three months. Only two of the health posts were supervised weekly from the catchment health center and 9 (56.3%) of the HEWS received onsite assistance for difficult cases. The average time to walk from the health post to the furthest house by HEWs was around 3 hours (Range, 1:00 - 4:00 hours). Nine of them responded that they are able to provide FANC (self-efficacious).
HEWs were asked about the implementation of support/facilitation strategies set by the Ministry of Health. These support strategies were assessed from the health center, district health offices, community, and development armies’ perspective. Nearly seventy percent (n= 11) of the HEWs reported that the implementation of support either from the community, health development armies or district health offices were lower than the planned standard.
Quality of service delivery
From all included mothers, 16.3% (n=146) of them did not receive antenatal care services in their last pregnancy. About two-thirds (n=489), and almost all (n=706) of them have not received the required number and contents of FANC package respectively.
Only 20% (n=180) of mothers were living within 15-minute of walking distance from the health post, while about one-third (n=333) of them travelled more than 45 minutes on foot to get to the health posts. Of those mothers who received antenatal care service, just above 90% (n=685) were self-referral. One-fifth (n=187) of mothers encountered pregnancy-related medical problems (like bleeding, convulsion, high temperature, etc) in their previous pregnancy.
No health post had all the required functional equipment and medical supplies for focused antenatal care services. Birth preparedness and complication readiness form, supervision checklist, blood pressure cuff, pregnant women registration book, stethoscope, and tape measure were the most frequently mentioned unavailable equipment in the health posts.
Facilitators’ and barriers’ of FANC intervention fidelity
To select the appropriate statistical model for the hierarchal nature of the data, the intra-class correlation coefficient (ICC) was computed by running the intercept only model. The ICC observed in the model was 17.73%, which indicates that 17.73% of the variation in FANC package intervention fidelity is explained by health post (cluster) level factors. This shows the application of multilevel linear regression model was appropriate.
In the first level model, maternal age, distance from the health post, maternal education, pregnancy related medical problems in the previous pregnancy, partner’s education, and total number of abortions were considered. Support/facilitation strategies, distance from the farthest household, and availability of supplies in the health post were considered in the second level model. Finally, pregnancy related medical problems in the previous pregnancy, partner’s education and support/facilitation strategies were found to be statistically significant facilitators for FANC package intervention fidelity (Table 2). In the final model, the ICC was reduced to 4.7% and both Akaike’s Information Criteria (AIC) and Bayesian Information Criteria (BIC) were decreased by 187.3 and 168.6 respectively, from the initial model (AIC=334.6 and BIC=349.0).