Factors associated with postnatal care for newborns in Zambia: analysis of the 2013-14 Zambia demographic and health survey

The conceptual framework below (Fig. 1) shows the linkages expected between PNC for newborns and demographic, socioeconomic, and ANC factors. The variables used in this conceptual framework were based on the empirical evidence from other similar studies and also their availability in the 2013-14 ZDHS. As the framework indicates, there could be direct links between socioeconomic factors and PNCs for newborns. For instance, place of residence (urban or rural) is an important factors as it may affect PNC coverage. The reasoning here is that; urban residents tend to have higher PNC coverage due to the availability of more health facilities compared to rural areas. Mothers’ educational level, wealth, and employment status could be positively associated with newborns receipt of PNC. Mothers with more education, wealth, and employment may have better understanding of the importance of PNC to their newborns’ survival. This, coupled with better access to media, may broaden women’s knowledge of how access to PNC can improve the health status and survival of newborn children.

Further, demographic variables may also have some influence on a child’s receipt of PNC. For example, first-order births and newborns whose mothers made the recommended four or more ANC visits may be more likely to receive PNC.

This study used data for women from the 2013-14 ZDHS to assess the demographic and socioeconomic factors associated with postnatal care for newborns. The ZDHS used a two-stage cluster sampling method. In the first stage, 722 enumeration areas (EAs)—305 in urban areas and 417 in rural areas-were selected with probability proportional to Sample Enumeration Area (SEA) size. In the second stage, a complete list of households served as the sampling frame in the selection of households for enumeration. On average, 25 households were randomly selected in each cluster. All women age 15-49 in the household were eligible to be interviewed. Women were asked to provide information about pregnancies resulting in a live birth during the 5 years preceding the interview. In this case, a live birth is defined as “the complete expulsion or extraction from its mother of a product of conception, irrespective of the duration of pregnancy, which after such separation, breathes or shows any other evidence of life, such as beating of the heart, pulsation of the umbilical cord, or any definite movement of voluntary muscles, whether or not the umbilical cord has been cut or the placenta is attached” [23]. Data on postnatal care was collected for the most recent birth during that period.

In the 722 selected clusters, 16,258 households were occupied at the time of data collection of which 15,920 were successfully interviewed, yielding a household response rate of 98%. In total, 16,411 women were successfully interviewed. As observed from above, there were more women who were interviewed than the number of households sampled because once interviewers visited the household, they interviewed every woman in the household so long they were aged 15 – 49 years. This analysis concentrated on the postnatal care received for the woman’s most recent birth in the 5 years preceding the survey. Among the most recent births, 514 caesarean births were excluded from the analysis sample because these births were likely to have received PNC regardless of the mothers’ demographic and socioeconomic characteristics. The final sample included 4,777 births, with 1,361 delivered at home and 3,416 delivered in a health facility.

In the DHS Woman’s Questionnaire, all women who had a birth in the 5 years preceding the survey were asked this question about their most recent birth: “How long after delivery did the first check take place for last birth?” Responses to this question were used to construct the two outcome variables for this study. These variables were any postnatal check-up within 6 weeks after delivery “0 = No PNC; and 1=Yes any PNC” and timing of first PNC “0 = No PNC”, “1 > 48 hours”, and “2 ≤ 48 hours” 2 or more days [24]. The categorization of timing was done in this way because most neonatal deaths usually occur within the 48 hours after the child’s birth [25]. The rationale behind using the two measures was based on the fact that not only is receiving any PNC cardinal but also receiving it within the most critical period (within 48 hours) can increase the chances of survival for the newborn.

Demographic variables included in the model were mother’s age at the last birth (under age 20, 20-34, and 35-49), birth order (1st, 2nd, 3rd, and 4th or higher), and mother’s perception of the size of the child (small, average, or large).

Among the socioeconomic factors, we examined mother’s employment status (whether currently working or not), education attainment (none, primary, and secondary or higher), and household wealth status. The DHS wealth quintiles were grouped into a three-category wealth status variable (poor, middle, and rich) by combining the top 40% and the bottom 40% of the population. Media access was constructed on a combination of three variables (frequency of reading newspapers or magazines, listening to radio, and watching television). A mother is considered to have access to mass media if she watches TV, listens to radio, or reads a newspaper. Receipt of antenatal care was measured with the number of ANC visits the mother made for the last birth (0 = less than four, 1 = four or more).

Separate analyses were conducted for home deliveries and facility deliveries because the factors that influence PNC for newborns delivered at home may be different from those that influence newborns delivered in a health facility [3, 26, 27].

Descriptive statistics (both univariate and bivariate) were used to describe the overall coverage of PNC and timing of PNC, and by demographic and socioeconomic variables, as well as by the number of ANC visits received.

To assess the adjusted associations between the independent variables and receipt of any PNC for newborns, two models were fitted, one for newborns delivered at home and the other for those delivered in a health facility. Odds ratios with 95% confidence intervals were reported.

In addition, a stepwise backward regression was used to assess a newborn receiving or not receiving PNC and timing of PNC with their associated correlates stratified by place of delivery. In both models, “never received any PNC” was the reference category. All the estimates were weighted to reflect the population. The effect of the complex multistage sampling design that was used in the 2013-14 ZDHS was considered. Data were analysed using STATA 14.2 software.

Table 1 below describes births included in the analysis by mother’s characteristics. The total sample for the analysis was 4,777 newborns of which 3,416 (71.5%) were delivered at health facilities while 1,361 (28.5%) were delivered at home. For both home and facility births, about two-thirds of the mothers were aged 20-34 at the time of birth. Newborns delivered at home had a higher proportion of 4th order births (60%) than newborns delivered in a health facility (40%). Less than two-thirds of newborns were perceived by their mothers to have average size at birth (61% delivered at home and 57% delivered in a facility).

By geographic distribution, a larger share of home births took place in Central and Southern regions (18% and 19%), while Eastern, Copperbelt, and Lusaka regions had a larger share of facility births (14%, 15%, and 19%). In every 10 home deliveries 9 were in rural areas (91%) compared with about 6 in every 10 deliveries in a health facility (58%). A majority of mothers had primary education, whether they delivered at home or in a health facility (66% and 50%). Among newborns delivered at home, more than two-thirds were in poor households, while among those delivered at a health facility, poor and rich households accounted for about 40% each. Sixty-three percent of mothers of newborns delivered at home were employed at the time of the survey, as were 53% of mothers of newborns delivered in a health facility. Nearly three-fourths (73%) of the mothers of newborns delivered in a health facility had access to media compared with about half (54%) of mothers of newborns delivered at home.

Overall, more than half (53%) of the mothers of the newborns had four or more ANC visits. Disaggregated by place of delivery, 55% of mothers of newborns delivered in a health facility had four or more ANC (4+ ANC) visits compared with 46% among those delivered at home.

Figure 2 below shows the percentage of newborns who received PNC after birth by place of delivery. Overall, 55% of the newborns received PNC after birth. Disaggregated by place of delivery, 58% of newborns delivered in a health facility received PNC compared with less than half (48%) of newborns delivered at home.

Figure 3 below shows the percent distribution of timing of first PNC after birth by place of delivery. Overall, only 16% of newborns received their first PNC within the first 2 days. Among newborns delivered in a health facility, 19% received their first PNC within the first 2 days compared with 8% among newborns delivered at home.

Table 2 below shows the percentage of newborns who received PNC after birth according to demographic and socioeconomic characteristics. Overall, PNC coverage varied significantly across place of residence, with newborns delivered at health facilities in rural areas less likely to receive PNC than newborns delivered at health facilities in urban areas (55% versus 62%). In addition, PNC coverage varied significantly by whether mothers of newborns had access to media. Among newborns whose mothers had access to media, a higher proportion received PNC compared with newborns whose mothers did not have access to media—at 53% versus 42% for home deliveries and 590% versus 53% for deliveries in a health facility.

Table 3 below shows the percentage of newborns by timing of first PNC according to demographic and socioeconomic characteristics. Among newborns delivered at home, the proportion receiving their first PNC within the first 48 hours after birth was twice as high in urban areas as in rural areas (15% versus 8%). For facility deliveries, newborns in urban areas had a slightly higher proportion of receiving their first PNC within the first 48 hours compared with those in rural areas (41% versus 38%).

Among home deliveries, a lower proportion of newborns of mothers with primary education received their first PNC within 48 hours of delivery compared with those newborns of mothers with secondary or more education (8% versus 15%). Among newborns delivered at home, the proportion of receiving first PNC within 48 hours increased as household wealth increased. Newborns in households classified as rich had twice as high the proportion of receiving their first PNC within the first 2 days as newborns in poor households (14% versus 7%).

Also among home deliveries, newborns whose mothers had access to media had about twice the proportion of first PNC being within 48 hours compared to newborns whose mothers had no access to media (11% vs. 6%). Likewise, among newborns delivered in a health facility, those whose mothers had access to media were more likely to receive their first PNC within 48 hours compared to newborns whose mothers did not have access to media (20% versus 17%).

Table 4 below presents the adjusted odds ratios of the association between PNC coverage and demographic and socioeconomic characteristics for newborns delivered at home and in a health facility, respectively. Among home births, newborns whose place of residence at the time of the survey was rural had 40% lower odds of having received any PNC compared to their urban counterparts. The odds of home-born newborns having a PNC visit were 60% higher for newborns whose mothers were exposed or had access to media compared to those whose mothers had no exposure or access. Among home births, newborns whose mothers had (4+ ANC) visits during pregnancy had 60% higher odds of having a PNC visit compared to newborns whose mothers had fewer than four ANC visits.

Among newborns delivered in a health facility, and whose place of residence was rural had 20% lower odds of having any PNC compared to those born in urban areas. On the other hand, newborns whose mothers had access or were exposed to media had 70% higher odds of receiving any PNC compared to those whose mothers had none. However, the odds of having any PNC were 30%, lower, for newborns whose mothers had secondary or higher education compared to those whose mothers had no education at all.

Table 5 below presents the adjusted relative risk ratios of the association between timing of first PNC and demographic and socioeconomic characteristics. Separate models were run for home deliveries and deliveries in a health facility. The results show that among home deliveries, the risk of attending first PNC within 48 hours after delivery relative to not attending PNC for newborns whose mothers were exposed or had access to media were 100% higher than those whose mothers were not exposed or had access. Similarly, newborns to mothers who had secondary or higher education had 100% higher odds than that of newborns whose mothers had no education at all. The relative risk for attending first PNC within 48 hours after birth relative to not attending PNC was 90% higher for newborns whose mothers had (4+ ANC) visits than those whose mothers ANC visits were less than four.

Table 5 further shows that among newborns delivered at home the relative risk for attending first PNC after 48 hours after delivery to not attending PNC for newborns whose place of residence was rural was 40% lower than for newborns whose place or residence was urban. Further, the relative risk for attending first PNC 48 hours after delivery compared with not attending PNC for newborns whose mothers had media exposure was 50% higher for newborns whose mothers had no media exposure. Similarly, for newborns whose mothers made (4+ ANC) visits, the relative risk for attending first PNC 48 hours after delivery versus not attending PNC was 70% higher than for newborns whose mothers made fewer than four ANC visits.

In contrast, among facility-delivered newborns, the relative risk for attending first PNC within 48 hours after delivery relative to not attending PNC was 20% lower, for newborns whose perceived size at birth was average than for newborns whose perceived size was small. Similarly, the relative risk for a newborn attending PNC 48 hours after birth relative to not attending PNC was 20% for newborns whose perceived size at birth by their mothers was average and were exposed to media. In addition, the relative risk of a newborn having PNC 48 hours after birth relative to not attending was 30% and 20% higher for children whose place of residence was urban than those in rural areas and those with secondary or more education than those with no education at all, respectively.

The objectives of this study were to assess the demographic and socioeconomic characteristic associated with PNC coverage and also to examine the factors associated with the timing of first PNC among newborns delivered at home and in health facilities. However, it is cardinal to highlight that our study excluded all women who gave birth through caesarean section as these women and children always receive PNC whilst in the hospital as part of the service.

Our study found that, overall, 55% of newborns received PNC. However, substantial variations were observed by place of delivery. Newborns delivered at home were less likely than those delivered at a health facility to receive PNC (48% versus 58%). This finding is supported by similar studies elsewhere [2, 25, 27]. The difference may be attributed to the fact that births delivered in a health facility are attended to by skilled personnel, who may encourage and educate mothers on the importance of PNC. However, since the data in the DHS does not collect data on the actual services received during PNC, more research may be required to understand the factors that motivate women to go and seek postnatal care for themselves and their newborns. In addition, only 16% of the newborns received PNC within the first two days, according to the 2013-14 ZDHS [15]. Disaggregated by place of delivery, our study found that the timing of PNC within the first 48 hours after birth was more than twice as high among newborns delivered in a health facility compared to those delivered at home (19% versus 8%). The PNC may be this low for both home and facility delivered newborns because only the most recent births were included as part of the analysis implying that young mothers probably those who are more educated and informed are having institutional births. In addition, women are having SMAG members residing with them in their communities thereby making it more likely for more and more women opting to deliver from health institutions as opposed to delivering at home.

Newborns delivered at health facilities, whose size at birth was average as perceived by their mothers had lower odds of having PNC visits within the 48 hours after the child’s birth compared with newborns perceived as small. This finding agrees with other studies [8]. The finding is however counterintuitive, because common sense and other studies informs us that small-sized newborns need extra care after birth compared with newborns of average size [28]. Various factors may help to explain this finding; for example, women in rural areas and women with little education may not know the problems that having a newborn with low birth-weight could bring in the immediate or near future and thus do not see the need for the newborn to be taken for PNC. Besides the aforementioned, it may also be possible that access to monetary resources may be difficult thus making it difficult to seek PNC if the facility is located far from their homes. Moreover, referrals in these communities may not be available as the community side of IMCI may not exist in all the areas of Zambia.

Our study further reveals that, residing in rural areas was associated with lower odds of newborns receiving PNC either for home or facility delivered (40% vs 20%). However, for home births, the odds for newborns in urban and rural areas are twice as high to receive PNC within 48 hours after birth (15% vs. 8%). This seems to be a big problem as proximity to the nearest facility and the economic status of the households where these newborns reside has some effect on whether they access PNC or not. With regard to timing of PNC, place of residence is an important factor that has been cited as influencing the receipt of PNC for newborns within the appropriate timings. According to a study in Nigeria by [29], newborns in places that were predominantly rural in nature had lower odds of having PNC compared with those in predominantly urban regions. Similarly, in Bangladesh [27] found that the place of residence had a significant relationship with newborns’ receipt of PNC and also the timing of first PNC. These studies agree with the results of our study, which found that newborns in rural areas of Zambia had lower odds of newborns having PNC and the timing of first PNC regardless of whether the births were delivered at home or in a health facility.

Our study shows that newborns, whether delivered at home or in a health facility, were more likely to receive PNC if their mothers had access to media. Access to media by mothers to newborns has been found to have a positive association with the use of PNC services among newborns. For instance, women who have access to media have been found to have increased antenatal care visits, and as such they tend to know the benefits of a child having PNC after birth, as they are aware of what problems they encounter if the child does not receive PNC [8, 30‐33]. Moreover, this finding may be attributed to the fact that women who have access to media may be more educated, come from well-to-do families or reside in close proximity to health facilities. As such they are well informed and are able to make decisions on the use of their resources and on what health services they can utilise for their newborns in order to ensure that they grow up healthy. For example, Table 5 above shows, among facility delivered newborns whose mothers had secondary or higher education had 100% higher odds of receiving PNC within the 48 hours of delivery.

Furthermore, this study found that newborns delivered at home, whose mothers had four or more ANC visits had higher odds of having PNC and receiving it within the 48 hours after birth. This finding is similar to results of other studies, where the number of ANC visits is an important determinant of PNC use [8, 29]. The plausible explanation of this finding is that through creation of an integrated service provision for pregnant mothers especially among those with poor socio-economic status, may in both urban and rural areas have some multiplier effect for uptake of postnatal services in the future. by making accessible and available at all times may.

This study demonstrates that various factors influence the utilisation and the timing of PNC for newborns. These factors established by this study provide pointers to policy relating to child health care. It is important that deliveries take place in health facilities in order to receive attention from qualified medical personnel. This may help reduce the high neonatal mortality rate in the future.

The results show that being born in the rural areas reduces the chances of newborns having PNC. Government and other stakeholder should channel more resources to rural areas by building more hospitals and health centres as well as providing health care personnel in these hospitals.

Media access or exposure plays a crucial role by conveying basic child healthcare information to the mothers. Since this study has found an association between media access or exposure with PNC, child health stakeholders and government can utilize the various medias to sensitize people about the importance of PNC for child and mothers’ survival.

The study also shows that mothers’ attendance of the recommended number of ANC visits plays a fundamental role in ensuring that mothers take their newborns for PNC, especially among those delivered at home. Programmes to enhance utilisation of ANC would prove effective at stimulating the utilisation of PNC.

Finally, more research is required to understand the factors that cause low levels of PNC and influence its timing in certain regions like Luapula, Southern, North Western, and Copperbelt regions and among newborns perceived to be small at birth.

Based on the nationally representative 2013-14 ZDHS, this study identified differences that exist in the factors associated with receipt of PNC and the timing of first PNC among newborns delivered at home and newborns delivered in a health facility. The study shows that a high proportion of newborns regardless of place of delivery are not having PNC. Among the home births, the study shows that being born in the rural areas reduces the odds of having PNC. In addition, mothers’ exposure or access to media, having four or more ANC visits during pregnancy, increases the odds of having PNC. For facility births, newborns whose place of residence was rural and those whose mothers had secondary or higher education, had lower odds of having any PNC conversely, mothers access or exposure to media had a positive influence on PNC.

With regard to timing of PNC, among the home births, mothers’ media exposure or access to media, having secondary or higher education, and had 4+ ANC visits during pregnancy increases the odds of having PNC within 48 hours. Furthermore, attending the first PNC after 48 hours after delivery was determined by place of residence, media exposure and 4+ ANC. On the other hand, the timing of PNC, among the facility births, was influenced by the perceived size at birth of the newborn.

This study had four major limitations: First, the study was limited to only the variables collected in the ZDHS. Factors that may affect PNC care but are not available in DHS could not be examined. For example, we observed differences in receiving PNC among socio-economic factors but we were not able to explore the factors that account for the differences. Second, not only is receipt of PNC important; so is the content of PNC. However, content could not be assessed because of data unavailability. Third, due to the cross-sectional nature of the survey, it was impossible to infer causality in the associations between the covariates in this study and the outcome variables as such we could only do probabilistic conclusions. Fourth, the dependent outcome was based on recall of past events that happened in the two years’ prior the survey.