Predictors of The Timing and Number of Antenatal Care Visits among Unmarried 1 Compared To Married Youth in Uganda Between 1995 And 2011


 Introduction Inadequate use of maternal health services among the youth remains a serious health challenge in Uganda. The low use of maternal health services among youth partly explains the persistence high maternal mortality rate in the country. Yet, improved use of maternal health services by the youth would help reduce maternal deaths in the country. Therefore, this study examines predisposing and enabling factors associated with the timing and the number of antenatal care visits among unmarried compared to married youth aged 15-24 years between 1995 and 2011 in Uganda. Methodology Two-level binary logistic and linear regression models with district as a second level of analysis were conducted on pooled data of the 1995, 2000/01, 2006 and 2011 Uganda Demographic and Health Surveys. This analysis was among 581 unmarried, compared to 5,437 married youth, aged 15-24 years. Results Only 16% of unmarried youth and 18% of married youth had ANC in the first trimester. Education was the only factor that was significantly associated with early use of antenatal care among unmarried youth. Whereas high education was associated with higher odds of using antenatal care in the first trimester among married youth (OR=1.30, 95%CI=1.08-1.57), it was associated with late start among unmarried youth (OR=0.56, 95%CI=0.31-0.98). Higher parity, protestant membership and residence in eastern region were associated with late start of antenatal care, while access to radio and television, and education level of the husband were associated with higher odds of early use of antenatal care among married youth. Overall, married youth were more likely to have more frequent antenatal care visits than unmarried youth. Among both groups, higher educational attainment and greater access to radio were associated with frequent antenatal care use. Residing in western region was associated with fewer antenatal care visits among both married and unmarried youth. Access to newspaper was associated with more antenatal care visits among married youth only. Conclusion This study presents the individual level predisposing and enabling factors that are important predictors of the use of antenatal health care services among youth that will guide policy to reduce maternal deaths among youth in Uganda.


22
Only 16% of unmarried youth and 18% of married youth had ANC in the first trimester.

Introduction
This literature analysis further identified that use of ANC among youth can be influenced by 87 service availability and accessibility, waiting time, service quality, youth' autonomy; domestic 88 violence prevalence and attitudes of health staff (42, 50,51,[53][54][55][56][57][58]. However, the 89 unavailability of these factors in the UDHS data sets do not allow the researcher to examine 90 these factors in this analysis. 91 The behavioural model of access to healthcare as proposed by Andersen (59) and modified by   128 Antenatal care is measured by two variables; the number of visits and the timing of the first 129 ANC visit. This data was captured from women who sought ANC during the last pregnancy in 130 the last five years before the survey. Women were asked if they had ANC for their last 131 pregnancy. Those who consulted a trained health provider for ANC were asked how many 132 times they did receive antenatal care during that pregnancy. They were also asked how many 133 months pregnant they were, the first time they received ANC for that pregnancy. World Health 134 Organisation (WHO) recommends that pregnant women in resource poor countries attend ANC 135 at least four times, and the first visit should be in the first three months of the pregnancy, to 136 identify pregnancy related problems and mitigate them (WHO, 2006). For this study, a binary 137 outcome was constructed for timing of antenatal care; youth who initiated ANC visits in the 138 first three months of the pregnancy were coded as one '1', while those who never attended or 139 started after the three months were coded as zero '0'. The number of ANC visits was treated 140 as a continuous variable and ranged from zero for those who did not have any ANC visit to 20 141 ANC visits.   while the random effect  is specific to district j and  is the error term at individual   194 and district levels.

195
Quasi-Likelihood estimation methods were relied on for the multilevel estimation procedure    Trends in the use of antenatal care among unmarried compared to married youth aged 253

15-24 years in Uganda between 1995-2011 254
The levels of the use of ANC in the first trimester were low but increased over the years (Table   255 3). It increased from 13 percent to 21 percent between 1995 and 2011; a total percentage 256 increase of 7.8% among unmarried youth. It increased by 9% from 16% to 25% among married 257 youth. This increment was not significant among unmarried youth (p=0.454) but was     (Table 5)

341
For married youth, the results in Table 4   In addition, those who listened to the radio almost daily had 0.6 more ANC visits compared to 364 youth who did not have any access to the radio at all. Controlling for predisposing factors in   Table 6 presents the results for factors impacting on ANC visits among married youth and these 381 results differ somewhat from those of unmarried youth.   The results provided no evidence of association between the number of ANC visits and 426 husbands' occupation.

427
The results observed in Table 6 suggest that most of the variation in mean number of ANC  It was observed that there was an increase in early initiation of prenatal care from 13% to 20% 445 and 16% to 25% among unmarried and married youth respectively. However, this is still low  access to the radio and television and daily access to radio and newspapers were associated 530 with early start and frequent use of antenatal care among married youth. Daily access to the 531 radio was also associated with frequent ANC use among unmarried youth. Access to media is 532 associated with increased knowledge about the benefit of using maternal health services which 533 compels youth to better use antenatal care (15,43,50,102). This study also helped identify the 534 radio is a main source of information for the youth in Uganda since a large proportion of the 535 population in Uganda has greater access to the radio as the main source of information (55.2%) 536 compared to the television (7.2%) and newspapers (2.1%) (103).

537
Other predisposing and enabling factors that were controlled for in this analysis included age 538 of the youth, pregnancy desire, region, religion but these had no significant influence on the 539 use of antenatal care.

540
Limitations of the study 541 Uganda Demographic and Health surveys (UDHS) collect data from women for births in the 542 last five years before the date of the study which may lead to inaccuracies due to memory lapse.

543
The inattention given to some important variables that influence the use of maternal health 544 services is a limitation common in secondary data analysis. The cross-sectional nature of data 545 from UDHS is that, we are not able to establish the time sequencing of events of interest that 546 is antenatal care use and socio-economic factors that is antenatal care use is asked for a birth implemented.

561
Education predicted the use of ANC in the first trimester among unmarried and married youth.

562
Education was the only factor significantly associated with the use of ANC in the first trimester 563 among unmarried youth. While high levels of education were associated with higher chances 564 of using ANC in the first trimester among married youth, it was associated with late start of 565 ANC among unmarried youth. High parity, protestant membership and residence in the eastern 566 Uganda were associated with late start of ANC while access to radio and television, and higher 567 partner education were associated with early use of ANC in the first trimester among married 568 youth.

569
Among unmarried and married youth, higher educational attainment and greater access to radio 570 were associated with frequent ANC use. Although having a middle wealth index was associated 571 with more ANC visits among unmarried youth, married youth in highest wealth quintile 572 households were associated with infrequent ANC visits. Unmarried and married youth living 573 in western Uganda had fewer ANC visits compared to those living in the central region.

574
Additionally, married youth residing in the eastern region were associated with lower ANC 575 visits among married youth, but this did not apply to unmarried youth. Membership of other 576 religions was associated with lower ANC visits among unmarried youth compared to Catholics 577 but not among married youth. More access to newspapers and higher levels of the husband's 578 education level were associated with more ANC visits among married youth only.