Factors associated with changes in adequate antenatal care visits among pregnant women aged 15-49 years in Tanzania from 2004 to 2016

The study was conducted in Tanzania, which includes the mainland and island. This was a Crossectional study that used data from the Tanzania Demographic Health Survey (TDHS), Further details of the survey are available elsewhere [13], but in brief this is a national representative survey done after five years with the objective to obtain the current and reliable information on demographic and health indicators about family planning, fertility levels and preferences, maternal mortality, infant and child mortality, nutritional status of mothers and children, ANC, delivery care, and childhood immunizations and diseases. Data were obtained from www.​dhsprogram.​com, after being granted permission to access and use TDHS data. Data from 2004/04, 2010 and 2015/16 surveys were used.

The population was all women of reproductive age (15-49 years) who had given birth to at least one child within the five years before the survey and had information on ANC visits. For a woman with multiple births during the five-year period, we considered mother’s last birth within 5 years prior the survey for this analysis. A total of 33,734 women aged 15-49 years in Tanzania participated in the three TDHS surveys. After excluding those with missing information on ANC visits, we remained with a total of 17,976. Of 17,976 women enrolled in the study: 4541(77.9%), 4201(76.9%) and 5193(70.1%) for 2004/05, 2010 and 2015/16 surveys respectively (Fig. 1).

Our dependent variable was adequate ANC visits, which was categorized as four or more ANC visits and coded 1, less than four ANC visits as inadequate were coded 0. Independent variables were respondent’s age at last birth (15-19 years, 20-24 years, 25-29 years, 30-34 years, 35+ years), education level (no formal education, primary education, secondary and higher education), employment status (unemployed, employed), marital status (married/cohabiting, single, divorced/widowed/separated), residence (urban, rural), wealth index (poorest, poorer, middle, richer, richest), zones; these are administrative regions grouped according to geographical location (western zone, northern zone, central zone, southern highlands, southern zone, south west highlands zone, lake zone, eastern zone, Zanzibar), first ANC initiated (women with first ANC visit later than 12 gestational weeks, women with first ANC visit by 12 gestational weeks), decision maker of respondent’s health care (respondent alone, respondent and partner, partner alone, someone else), parity (1 child, 2-3 children, 4-5 children, 6 or more children), frequency of listening to radio (not at all, Less than once a week, at least once a week), frequency of watching TV (not at all, less than once a week, at least once a week), desire of last pregnancy (wanted then, wanted later, wanted no more), history of terminated pregnancy (never had, ever had) and distance from health facility (big problem, not a big problem). The selection of variables was made using the Andersen’s Behavioural Model of Health Services Use [20]. All these variables were considered as mother’s characteristics and population characteristics in the analysis.

Data were analyzed using STATA Corporation, College Station, TX, USA version 14 (Stata/SE 14.2). The analysis considered the complex survey features: primary sampling units, strata, and sampling weights. A Poisson regression analysis was done to determine factors associated with adequate ANC visits. Multivariable Poisson decomposition analysis was conducted to determine factors associated with changes in adequate ANC visits. Decomposition analysis was conducted to understand whether observed changes in adequate ANC visits could be explained by changes in factors over time or in the population structure (population dynamics). To explain the observed change in the percentage of pregnant women attaining adequate ANC visits, we used the Blinder-Oaxaca decomposition analysis [21‐23]. The main goal decomposition analysis was to explain on the individual contributions of the factors on adequate ANC visits differences among pregnant women in Tanzania in different surveys. The differentials in adequate ANC visits between these groups was portioned into two components, one that can be attributable to differences in characteristics and the component that is attributable to the effect of those characteristics. The factors might have a different contribution on the change observed at different survey period. The decomposition analysis was done between two time points, at first, we decomposed survey year 2004/05 to 2010 and lastly survey year 2015/16 to 2010. The baseline survey year was the one with the lowest proportion of pregnant women with adequate ANC visits, thus survey year 2010 for both decomposition analysis. Contributions were considered statistically significant at a P-value of less than 0.05.

A total of 17,976 women were included in the analysis. Most of the participants were from rural areas, the mean age (±SD) of the study population was 27.06 (±7.00). More than half of the respondents in each survey had at least primary education level. Most of the participants were married or cohabiting: 85.4, 84.2 and 81.6% for 2004/05, 2010 and 2015/16 survey, respectively. The proportions of women aged 35 to 45 years increased across the survey years, from 15.6%, in 2004/05 to 17.8% in 2015/16. The percentage of women who achieved secondary education and above also increased from 9.2% in 2004/05 to 19.9% in 2015/16 and the percentage of women without formal education decreased from 26.8% in 2004/05 to 19.5% in 2015/16. Substantial regional variation in survey participation was observed; throughout the three surveys, the Lake Zone had the highest percentage of women participating in the survey while the Southern zone had lowest (Table 1).

The trend in adequate ANC attendance has fluctuated over time. Adequate ANC attendance decreased from 61% in 2004/05 to 43% in the 2010 survey and then increased again to 51% in the 2015/16 survey (Fig. 2). A similar pattern was also found when stratified by geographical zone. The eastern zone had the highest percentage of women with adequate ANC attendance for all three surveys (Fig. 3). The percentage of women with four ANC visits who initiated their first ANC visit in the first trimester increased over time (Fig. 4).

Various factors were associated with adequate ANC visits for each survey, including first ANC in the first trimester, multiparity, wanting pregnancy later, watching TV at least once a week, older age, geographical zone, secondary education and above, reasonable distance to a health facility, richer and richest household wealth index (Table 2).

In the multivariable Poisson regression analysis, for all three surveys ANC initiation within the first trimester had a positive effect on adequate ANC visits. The proportion of women with adequate ANC attendance was 1.47 (95% CI: 1.41-1.52) times greater among women who initiated ANC within the first trimester compared to those who initiated later in the 2004/05 survey, 1.96 (95% CI: 1.82-2.11) times higher in 2010, and 1.89 (95% CI: 1.79-2.00) times higher in 2015/16. However, wanting pregnancy later had a negative influence on adequate ANC visits. In the 2004/05 survey, adequate ANC attendance was 0.96 (95% CI: 0.90-1.02) times lower among women who wanted pregnancy later compared to those who wanted pregnancy at that time, 0.82 (95% CI: 0.74-0.92) times lower in 2010, and 0.92 (95% CI: 0.86-0.98) times lower in 2015/16 (Table 2).

The multivariable decomposition regression models found that 95.8% of the decline in adequate ANC visits from 2004/05 (62%) to 2010 (43%) were attributed by changes in the coefficients (mother’s characteristics) and only 4.2% of the decline was due to changes in the population characteristics (population dynamics). There were no significant changes in the population structures during this period, suggesting that the population remained relatively static between the 2004/05 and 2010 surveys. The southwest highland zone contributed 14.2% to the observed decline in the 2004/05 and 2010 surveys, which was statistically significant. It means that the zone where a pregnant woman lived affected her ability to attain adequate ANC. Changes in the initiation of the first ANC within the first trimester slowed the decline by 8.7%, which was also statistically significant (Table 3).

The proportion of women attaining adequate ANC increased from 43% in 2010 to 51% in 2015/16. The slight increase was attributed to 33.8% of changes due to the coefficients and 66.2% due to the changes in the population characteristics. These changes were statistically significant with a p-value of <0.001. The increase in the proportion of women who initiated ANC during the first trimester contributed 50.5% to the increase observed in 2010 to 2015/16 surveys. This was statistically significant at a p-value of <0.001. In the contributions due to differences in coefficients, the southwest highlands contributed 21.4% to the overall increase (Table 4).

The study has successfully identified factors associated with changes in adequate ANC visits among pregnant women in Tanzania. With this, it is possible to reallocate the limited resources in Tanzania to focus on the factors that have shown to have a great contribution and influence on attaining adequate ANC visits among pregnant women in Tanzania. We have used nationally representative data which makes the study findings generalizable to the entire nation.

Data on the quality of ANC service was not analyzed in this study, we failed to establish its effect on adequate ANC attendance. Also, the analysis did not include biomedical data on HIV status which could have overestimated the number of ANC visits as HIV-positive women attend ANC on monthly basis. Also, the study is prone to social desirability bias as the women are aware of the recommended minimum ANC visits, which could have overestimated the effects. This study is a Crossectional study, no temporal relationship can be established.

Basing on the findings obtained, we would recommend the MOHCDGEC and implementing partners to put more effort on promoting the first ANC visit to be initiated within the first twelve weeks of pregnancy. This should be done hand in hand with providing mass education on the importance of ANC visits and why a pregnant women should adhere to the ANC comprehensive package. Intervention should be done at facility and community level. This will enable mothers and the communities to be informed on the pivotal role of ANC services as far as the safety of the mother and child is concerned and ensure continuous support from their families. Regional focused interventions such as ‘Wazazi na Mwana Campaign’ should be rolled out in regions with low uptake of ANC services. Further research to assess the quality of ANC services offered as it may have contributed to the changes observed and sub-optimal ANC attendance while we have a 98% coverage of at least one ANC visit among pregnant women in Tanzania.