Predictors of health facility childbirth among unmarried and married youth in Uganda.

BACKGROUND: Uganda has a high maternal mortality rate combined with poor use of health facilities at childbirth among youth. Improved use of maternal health services by the youth would help reduce maternal deaths in the country. Predictors of use of health facilities at childbirth among unmarried compared to married youth aged 15-24 years in Uganda between 2006 and 2016 are examined.
METHODOLOGY: Binary logistic regression was conducted on the pooled data of the 2006, 2011 and 2016 Uganda Demographic and Health Surveys among youth who had given birth within five years before each survey. This analysis was among a sample of 764 unmarried, compared to 5,176 married youth aged 15-24 years.
RESULTS: Overall, unmarried youth were more likely to have a childbirth within the health facilities (79.3%) compared to married youth (67.6%). Higher odds of use of health facilities at childbirth were observed among youth with at least secondary education (OR = 2.915, 95%CI = 1.747-4.865 for unmarried vs OR = 1.633, 95%CI = 1.348-1.979 for married) and frequent antenatal care of at least four visits (OR = 1.758, 95%CI = 1.153-2.681 for unmarried vs OR = 1.792, 95%CI = 1.573-2.042 for married). Results further showed that youth with parity two or more, those that resided in rural areas and those who were engaged in agriculture had reduced odds of the use of health facilities at childbirth. In addition, among married youth, the odds of using health facilities at childbirth were higher among those with at least middle wealth index, and those with frequent access to the newspapers (OR = 1.699, 95%CI = 1.162-2.486), radio (OR = 1.290, 95%CI = 1.091-1.525) and television (OR = 1.568, 95%CI = 1.149-2.138) compared to those with no access to each of the media, yet these were not significant among unmarried youth. CONCLUSION AND RECOMMENDATIONS: Frequent use of antenatal care and higher education attainment were associated with increased chances of use of health facilities while higher parity, rural residence and being employed in the agriculture sector were negatively associated with use of health facilities at childbirth among both unmarried and married youth. To enhance use of health facilities among youth, there is a need to encourage frequent antenatal care use, especially for higher parity births and for rural residents, and design policies that will improve access to mass media, youth's education level and their economic status.

Background

Despite the global progress in improving maternal mortality, women in sub-Saharan Africa (SSA) are 14 times more likely to die during the maternity period compared to women in developed regions [1, 2]. In Uganda, about 16 women die each day due to maternal mortality [3]. Most maternal deaths occur during labour, at childbirth or within two days postpartum, [1, 4–8], this highlights that most deaths would be averted with prompt and adequate diagnosis, and care during and after the childbirth process [2, 9, 10]. The use of health facilities has been associated with reduced maternal mortality [6, 9, 11–13]. Childbirth in a health facility is an opportunity to aid a mother during childbirth, supervise her immediately after childbirth and identify, manage and/or prevent complications [14].

Researchers have suggested that for countries to achieve SDG3, target 3.1, the use of maternal health services should be universal [15] or the use of health facilities at birth should be at least 81 percent together with a 91 percent use of at least one antenatal care visit, 78 percent of four antenatal care visits, and 87 percent of skilled birth attendance [7]. However, despite the correlation of the use of health facilities with maternal survival, its use remained low in Uganda (41% in 2006 & 57% in 2011 [16, 17]. However, a greater increase was observed in the five years before the 2016 UDHS from 57 percent in 2011 to 73 percent in 2016 [3, 17]. Coupled with poor use of antenatal care and postnatal care, Uganda is unlikely to achieve SDG 3.1 [16, 17].

Although younger women aged 15–24 years generally have higher odds of using health facilities at childbirth than older women (29), there are notable variations among young women. For instance, older adolescents and youth tend to be more likely to use health facilities at childbirth than very young youth below 17 years [18-20], which could be attributed to differences in knowledge and support to access health facilities at childbirth [21]. More importantly, use of health facilities at childbirth was found to be higher among married than unmarried youth due to various societal and health system factors including husband support, shame and discrimination among unmarried youth [22-24].Studies have also documented that working away from home [25] and higher parity [25-31] are associated with reduced chances of use of health facilities at childbirth among youth. In addition, higher education level [19, 23, 25–29, 31–33], higher wealth status [18, 20, 23, 25, 26, 28, 29, 31, 32], urban residence [18–20, 23, 25, 27, 31–33], access to mass media [20, 29], and early, frequent and quality antenatal care [23, 25–27, 29, 30, 33] have been associated with increased chances of giving birth from health facilities by youth. Furthermore, good community perspectives about health provider skills were associated with improved use of health facilities at childbirth among married adolescents in India [34]. However, in contrast, a study in Nigeria found no significant difference in health facility use at childbirth by wealth index among adolescents [29]. Furthermore, a study in Malawi found that higher parity was associated with increased odds of use of health facilities at childbirth [35].

Studies have documented the factors for use of health facilities among youth [19, 23, 25–29, 33, 36–38] but no population-based studies were found to have been carried out comparing unmarried and married youth. The situation is dire for pregnant unmarried youth who have their own unique social, economic, psychological, health and obstetrical needs; and their experiences during the maternity period especially, abuse and stigmatisation puts them in a vulnerable position [36, 39–41]. This study used a nationally representative survey to find out the variations in factors associated with the use of health facilities at childbirth among unmarried compared to married youth aged 15–24 years in Uganda. This is aimed at providing information to guide health providers and policy makers to overcome barriers and improve the use of health facilities at childbirth, which will subsequently reduce maternal mortality among the youth. The main objective of this study is to examine the factors associated with the use of health facility at childbirth among unmarried compared to married youth in Uganda.

Methods

Source of data

Secondary analysis of the pooled 2006, 2011 and 2016 Uganda Demographic and Health Survey (UDHS) data was conducted. The UDHS data was retrieved with permission from MEASURE DHS (dhsprogram.com). These surveys were population-based household surveys that used two stage sampling where clusters (Enumeration Areas) and households from each cluster were randomly selected. Detailed sampling procedures are available in respective UDHS reports [3, 16, 17]. The UDHSs collect data on socio-economic and demographic characteristics of women, and their use and timing of maternal and child health services, and nutrition among other factors. Young women aged 15–24 years who had had a birth five years before each survey were selected for this analysis. These criteria resulted into a sample of 5,940 cases of which 764 were unmarried and 5,176 were married youth.

Measure of the outcome variable

The dependent variable is the use of a health facility at childbirth which was coded as Yes = 1 if the youth gave birth in the health facility and No = 0 if she gave birth elsewhere including at home, on the road, or at the traditional birth attendants’ place.

Measures of predictor variables

Based on previous empirical literature and their availability in the UDHS data, independent factors that were included in the analysis are presented in the Table 1 below.

Table 1

Description of predictor variables.

Name of predictor variable	Measure
Age	Coded as 15–19 and 20–24 years
Parity	A dichotomous variable coded as 1 if the respondent had one child and 2 if the respondent had two or more children
Pregnancy desire	A dichotomous variable coded as 1 if the respondent wanted to get pregnant then and 2 if she did not want to get pregnant or wanted to get pregnant later
Sex of household head	Coded as 1 = if respondent was living in a male-headed household at the time of the survey; 0 = otherwise.
Number of ANC visits	Coded as 1 if the respondent had no or less than four ANC visits and 2 if the respondent had at least four ANC visits
Education level	Dummy variables for highest educational attainment classified into two categories: no education or primary education and secondary and above
Religion	Dummy variables for religious affiliation, re-coded into four categories: 1. Catholics, 2. Protestants, 3. Muslims 4. Other religions
Wealth index	Dummy variables for DHS household wealth index that is developed from household assets and constructed by principle component analysis. The PCA scores are classified as poorest, poorer, middle, richer, and richest wealth quintiles
Occupation	Dummy variables for occupation recoded as not working, those employed in the agriculture sector, professionals, and labourers
Place of residence	Coded as 1 if the respondent was residing in rural areas and 2 if the respondent was residing in urban areas
Region	Dummy variables for region coded into four categories: central, east, north and west
Access to newspapers	Dummy variables for access to newspapers categorised into three categories as no access, less frequent access or some access and more frequent access or daily access
Access to radio	Dummy variables for access to radio categorised into three categories: no access, less frequent access or some access and more frequent access or daily access
Access to television	Dummy variables for access to television categorised into three categories as no access, less frequent access or some access and more frequent access or daily access

Statistical analysis

Data were analysed at the univariate, bivariate, and multivariate levels using SPSS 24. Descriptive statistics of the background characteristics of the respondents were presented at univariate level. At bivariate level, Pearson’s chi-squared (χ2) tests were used to examine significant differences between health facility childbirth and the explanatory variables, and the trends in health facility childbirth over the years. Binary logistic regression models were fitted to find the factors associated with health facility childbirth among unmarried compared to married youth. The multivariate model included explanatory variables whose p-values were less than 0.05 during the chi-square tests, except for age group and survey year. Results are presented in the form of odds ratios (OR) with 95% confidence intervals of ORs.

The data was weighted using the individual/women weights (v005) to account for the complex survey design that is applied in DHS data collection and non-response. This data was tested for multi-collinearity of the variables using tolerance and the variance inflation factor (VIF). Multicollinearity test results showed that none of the variables in the model had a tolerance threshold of less than 0.10 or a VIF of 10; actually, all VIF values were below 2. The goodness of fit was tested using the Hosmer and Lemeshow test and the models were good fit as the Hosmer-Lemeshow chi-squared p-values were greater than 5% for both unmarried and married youth.

Research ethics statement

No approval for using UDHS data was required since UDHS is a secondary data source and available in the public domain for use with no identifiable information about participants. However, to access the data, we sought permission from MEASURE DHS. Strict ethical requirements were observed during UDHS data collection. Informed consent was sought from all participants before each interview in the UDHS. ICF institutional review board and a local institutional review board in Uganda approved the UDHS questionnaires. ICF IRB ensured that the survey complied with the U.S. Department of Health and Human Services regulations for the protection of human subjects, while the Uganda’s IRB ensured that the survey complied with local laws and norms. Privacy, anonymity, and confidentiality were ensured during the interviews, data storage and analysis. Details about UDHS data and ethical standards are available at: https://dhsprogram.com/What-We-Do/Protecting-the-Privacy-of-DHS-Survey-Respondents.cfm

Results

Descriptive characteristics of the respondents

Table 2 shows that more than half of both the unmarried and married respondents lived in rural areas (69% vs 86%), were aged 20–24 years (56% vs 76.5%), and had no or had attained primary education (54% vs 84%). More than half had had four or more ANC visits among both unmarried (53.4%) and married (55.5%) youth. The main source of information for both samples was the radio (63% unmarried compared to 65% among married youth) However, a larger proportion of unmarried youth wanted the pregnancy later (73%) and had only one birth (86%); compared to the married who wanted the pregnancy then (60%) and had had two or more births (61%).

Table 2

Distribution of unmarried and married youth by background factors.

Variable	Frequency (percentage)	Number, Health facility childbirth	Percentage health facility childbirth	Frequency (percentage)	Number, Health facility childbirth	Percentage health facility childbirth
	Unmarried youth			Married youth
Survey year			P = 0.000			P = 0.000
2006	171(22.7)	112	65.5	1,313(25.2)	638	48.6
2011	169(22.4)	118	69.8	1,227(23.5)	791	64.5
2016	414(54.9)	368	88.9	2,675(51.3)	2,096	78.4
Age			P = 0.115			P = 0.015 *
15–19	334(43.7)	253	77.1	1053(20.3)	737	70.5
20–24	430(56.3)	343	81.0	4123(79.7)	2789	66.9
Pregnancy wanted			P = 0.519			P = 0.274
Then	208(27.2)	156	80.8	3085(59.6)	2102	68.2
Later or not anymore	556(72.8)	442	78.6	2091(40.4)	1424	66.8
Birth order/Parity			P = .000 *			P = 0.000 *
One	654(85.6)	534	82.8	2038(39.4)	1579	76.8
Two or more	110(14.4)	65	58.6	3138(60.6)	1947	61.6
ANC numbers			P = 0.000 *			P = 0.000 *
None or 1–3 visits	352(46.6)	256	72.7	2323(44.5)	1341	57.5
4+ visits	403(53.4)	342	84.9	2892(55.5)	2185	75.6
Sex of household head			P = 0.085			P = 0.064
Male	358(47.4)	276	77.1	4448(85.3)	2960	66.6
Female	397(52.6)	322	81.3	768(14.7)	565	73.7
Education level			P = 0.000 *			P = 0.000 *
No education or Primary Education	413(54.1)	275	69.6	3896(75.3)	2336	61.0
Secondary	351(45.9)	324	90.0	1280(24.7)	1190	85.7
Religion			p = 0.041 *			P = 0.000 *
Catholic	291(38.1)	218	78.4	1775(34.3)	1205	67.8
Protestant	261(34.2)	197	74.9	1903(36.8)	1190	63.1
Muslims	110(14.4)	98	86.0	773(14.9)	624	77.4
Others	102(13.4)	85	85.0	725(14.0)	507	68.1
Type of Residence			P = 0.000 *			P = 0.000 *
Urban	234(30.6)	193	91.9	941(18.2)	853	89.9
Rural	530(69.4)	405	74.3	4235(81.8)	2673	62.7
Region			p = 0.035 *			P = 0.000 *
Central	259(33.9)	235	83.3	1177(22.7)	1087	79.9
Eastern	186(24.3)	149	81.4	1507(29.1)	986	63.6
Northern	147(19.2)	81	73.6	1362(26.3)	673	64.2
Western	164(22.5)	133	73.9	1130(21.8)	779	62.2
Wealth index			p = 0.001 *			P = 0.000 *
Poorest	116(15.2)	66	68.0	1339(25.9)	676	56.1
Poorer	104(13.6)	76	77.6	1255(24.2)	750	59.9
Middle	132(17.3)	102	76.7	903(17.4)	610	64.8
Richer	164(21.5)	137	77.8	771(14.9)	600	71.2
Richest	248(32.5)	216	87.1	908(17.5)	890	91.3
Woman’s Occupation			p = 0.000 *			P = 0.000 *
Not working	244(31.9)	212	86.5	1119(21.6)	891	77.7
Agriculture	263(34.4)	177	67.0	2712(52.4)	1606	58.0
Labourers	69(9.0)	56	87.5	464(9.0)	311	75.1
Professionals	188(24.6)	153	84.1	880(17.0)	718	81.2
Frequency of reading newspapers			P = 0.006 *			P = 0.000 *
Not at all	531(69.5)	396	76.2	4274(82.6)	2739	64.1
Less frequent	116(15.2)	100	87.0	574(11.1)	494	80.9
More frequent	117(15.3)	102	85.7	328(6.3)	293	88.0
Frequency of listening to the radio			P = 0.642			P = 0.000 *
Not at all	176(23.0)	130	76.9	1143(22.1)	665	60.1
Less frequent	106(13.9)	78	78.8	655(12.7)	398	66.9
More frequent	482(63.1)	391	80.3	3378(65.3)	2462	70.1
Frequency of watching TV			P = 0.001 *			P = 0.000 *
Not at all	484(63.4)	352	74.9	3994(77.2)	2514	62.4
Less frequent	97(12.7)	83	84.7	493(9.5)	346	75.4
More frequent	183(24.0)	163	87.6	689(13.3)	666	91.1
Total	N = 754(100)	598	79.3%	N = 5176(100)	3,526	67.6%

*Statistical significance at 5% level p<0.05

Not at all- No access to any, Less frequent- once or less than once; More frequent- Almost daily access

Table 2 shows an increase in the levels of the use of health facilities at childbirth between the survey years among youth. The proportions were higher among the unmarried than the married across all the survey years. Use of health facility at childbirth was high among unmarried youth in 2016 (89%) compared to unmarried youth in 2006 (60%). The proportion of health facility childbirth was high among married youth in 2016 (78%) compared to married youth in 2006 (49%). A greater increment of 30% was among the married compared to 24% among unmarried youth. This increasing trend was significant at p = 0.000 among unmarried and married youth.

Table 2 further shows the cross tabulations between health facility use at childbirth and each independent variable among unmarried and married youth.

Unmarried youth

There was a significant relationship between parity (p = 0.000), frequent ANC use (p = 0.000), education level (p = 0.000), religion (p = 0.041), place of residence (p = 0.000), region of residence (p = 0.035), occupation (p = 0.000), wealth index (p = 0.001), access to newspapers (p = 0.006), and television (p = 0.001) with health facility childbirth among unmarried youth.

Married youth

Age group (p = 0.015), parity (p = 0.001), frequent ANC use (p = 0.001), education level (p = 0.001), religion (p = 0.001), place of residence (p = 0.001), region of residence (p = 0.001), education level (p = 0.001), occupation (p = 0.001), wealth index (p = 0.001), and access to media ((newspapers (p = 0.0001), radio (p = 0.001), television (p = 0.001)) had a significant relationship with health facility childbirth among married youth.

On the other hand, this analysis found that there was no significant difference in the use of health facilities at childbirth by pregnancy desire and sex of household head for both unmarried and married youth.

Predictors of health facility use at childbirth among unmarried compared to married youth in Uganda between 2006–2016

The education level of participants, parity, occupation, place of residence, and antenatal care use were found to be significantly associated with use of health facilities at childbirth among both unmarried and married youth. In addition, wealth index and access to the radio, newspapers and television were significantly associated with use of health facilities at childbirth among married youth (Table 3).

Table 3

Multivariable analysis of factors associated with health facility use at childbirth among youth in Uganda.

	Unmarried youth		Married youth
Variable	Number, Health facility childbirth (Percentage health facility childbirth)	Adjusted Odds ratio with 95% Confidence Interval	Number, Health facility childbirth (Percentage health facility childbirth)	Adjusted Odds ratios with 95% Confidence Interval
Year of survey
2006	112(65.5)	1	638(48.6)	1
2011	118(69.8)	1.204 (0.705–2.054)	791(64.5)	1.826 (1.523–2.190)***
2016	368(88.9)	4.589 (2.678–7.864)***	2,096(78.4)	3.537 (2.998–4.173)***
Age
15–19	253(77.1)	1	737(70.5)	1
20–24	343(81.0)	1.321 (0.841–2.076)	2789(66.9)	0.926 (0.773–1.108)
Birth order/Parity
One	534(82.8)	1	1579(76.8)	1
Two or more	65(58.6)	0.284 (0.165–0.490)***	1947(61.6)	0.593 (0.509–0.691)***
ANC use
No or 1–3 visits	256(72.7)	1	1341(57.5)	1
4+ ANC visits	342(84.9)	1.758 (1.153–2.681)**	2185(75.6)	1.792 (1.573–2.042)***
Woman’s Education level
No education or Primary education	275(69.6)	1	2336(61.0)	1
Secondary+	324(90.0)	2.915 (1.747–4.865)***	1190(85.7)	1.633 (1.348–1.979)***
Religion
Catholics	218(78.4)	1	1205(67.8)	1
Protestant	197(74.9)	0.835 (0.521–1.338)	1190(63.1)	0.902 (0.774–1.052)
Muslim	98(86.0)	1.464 (0.717–2.989)	624(77.4)	1.201 (0.966–1.495)
Others	85(85.0)	1.356 (0.675–2.723)	507(68.1)	0.937 (0.764–1.150)
Wealth index (RC =)
Poorest	66(68.0)	1	676(56.1)	1
Poorer	76(77.6)	1.101 (0.516–2.349)	750(59.9)	1.117 (0.933–1.337)
Middle	102(76.7)	0.889 (0.418–1.889)	610(64.8)	1.362 (1.106–1.678)**
Richer	137(77.8)	0.916 (0.442–1.899)	600(71.2)	1.444 (1.146–1.820)**
Richest	216(87.1)	1.260 (0.534–2.973)	890(91.3)	3.211 (2.276–4.529)***
Woman’s Occupation
Not working	212(86.5)	1	891(77.7)	1
Agriculture	177(67.0)	0.541 (0.314–0.932)**	1606(58.0)	0.788 (0.656–0.946)**
Labourers	56(87.5)	00.777 (0.296–2.041)	311(75.1)	0.883 (0.658–1.186)
Professionals	153(84.1)	0.536 (0.286–1.005)	718(81.2)	0.949 (0.743–1.212)
Place of residence
Urban	193(91.9)	1	853(89.9)	1
Rural	405(74.3)	0.416 (0.217–0.799)***	2673(62.7)	0.550 (0.423–0.715)***
Frequency of reading newspapers
Not at all	396(76.2)	1	2739(64.1)	1
Less frequent	100(87.0)	1.691 (0.857–3.335)	494(80.9)	1.165 (0.911–1.489)
More frequent	102(85.7)	1.117 (0.556–2.244)	293(88.0)	1.699 (1.162–2.486)**
Frequency of listening to radio
Not at all	130(76.9)	1	665(60.1)	1
Less frequent	78(78.8)	0.884 (0.427–1.829)	398(66.9)	1.102 (0.870–1.397)
More frequent	391(80.3)	0.953 (0.558–1.630)	2462(70.1)	1.290 (1.091–1.525)**
Frequency of watching TV
Not at all	352(74.9)	1	2514(62.4)	1
Less frequent	83(84.7)	1.012 (0.508–2.018)	346(75.4)	1.004 (0.780–1.293)
More frequent	163(87.6)	0.987 (0.518–1.880)	666(91.1)	1.568 (1.149–2.138)**
Observations	598(79.3)	764	3,526(67.6)	5,176

*p<0.05

**p<0.01

***p<0.001

1 = Reference category

Unmarried youth with at least secondary level of education were about three times more likely to deliver in a health facility compared to counterparts with no or primary level education (OR = 3.45, 95%CI = 2.85–4.16) whereas married youth with secondary level education were 63% more likely to deliver in a health facility (OR = 1.633, 95%CI = 1.348–1.979).

Higher parity was negatively associated with health facility use among both unmarried and married youth. Youth with at least two children were less likely to give birth in health facilities compared to those who were pregnant for the first time (OR = 0.284, 95%CI = 0.165–0.490 among unmarried and OR = 0.593,95%CI = 0.509–0.691 among married youth). Unmarried and married youth who were employed in the agriculture sector had reduced odds of using health facilities at childbirth compared to youth who were not employed (OR = 0.541, 95%CI = 0.314–0.932 vs OR = 0.788, 95%CI = 0.656–0.946 among unmarried and married youth respectively).

Rural residence was negatively associated with use of health facilities among both unmarried and married youth (OR = 0.416, 95%CI = 0.217–0.799 vs OR = 0.550, 95%CI = 0.423–0.7155 among unmarried and married youth respectively). The number of ANC visits and education level were significantly associated with higher odds of delivering in health facilities among both unmarried and married youth. Unmarried and married youth who had four or more ANC visits had higher odds of using health facilities at childbirth compared to those who had no ANC or had less than four ANC visits (OR = 1.758,95%CI = 1.153–2.681 among unmarried & OR = 1.792,95%CI = 1.573–2.042 among married youth).

Married youth residing in households with at least middle wealth index were more likely to give birth from health facilities compared to married youth living in poorest households (OR = 1.362, 95%CI = 1.106–1.678, OR = 1.444, 95%CI = 1.146–1.820 and OR = 3.21, 95%CI = 2.276–4.529 for middle, richer and richest wealth index respectively). Also, married youth with more frequent access to the newspapers (OR = 1.699, 95%CI = 1.162–2.486), radio (OR = 1.290, 95%CI = 1.091–1.525) and television (OR = 1.568, 95%CI = 1.149–2.138) had higher odds of using health facilities at childbirth compared to those with no access to the newspapers, radio, and television respectively.

Discussion

The main aim of this study was to find out the predictors of the use of health facilities at childbirth among unmarried compared to married youth, aged 15–24 years in Uganda between 2006 and 2016. This study observed the marital status variations in the use of health facilities at childbirth; unmarried youth were more likely to have their childbirth in the health facilities than married youth. Use of health facilities at childbirth was associated with education level, parity, place of residence, occupation, and frequency of ANC use among both unmarried and married youth. In addition, it was associated with wealth index, access to newspapers, radio and television among married youth.

Unmarried youth were observed to use health facilities more than the married youth in Uganda between 2006 and 2016. This could be due to the confounding effect of higher parity as most married youth were pregnant for the second time or more, and higher parity has been associated with lower use of safe childbirth in prior studies among youth [20, 25, 26, 29–31]. A separate regression analysis combining the two samples and considering marital status as one of the factors (results not presented) shows that the effect of marital status ceases to be significant when these confounding factors are controlled for, suggesting that the differences in use of health facilities at childbirth among unmarried and married youth is accounted for by the cofounding factors. Parity was significantly associated with the use of health facilities at childbirth in this study. Youth who have had two or more children were less likely to use health facilities at childbirth compared to those who were having their first birth in the current study. This has been found in previous studies among youth [25–28, 30–32]. Studies have found that women with higher parity who have had no history of pregnancy complications [42], have had better pregnancy outcomes with no infant mortality [43], and believe that they can give birth on their own without the need to go to the health facilities [44] may see no reason for using health facilities for childbirth. A separate analysis suggests that the apparent higher use of health facilities during childbirth among unmarried than married youth is largely explained by a higher proportion of higher parity births among married women.

Our findings show that youth with higher education level was associated with higher odds of giving birth in health facilities compared to those with no or primary education level. This is in line with what has been found in prior studies among youth [20, 23, 26, 28, 29, 31, 33, 45]. This could be partially linked to higher levels of knowledge on the benefits of seeking healthcare and women empowerment which enhances their decisions concerning their health care [25, 28]. The result suggest that the effect of education is stronger for unmarried than married youth, suggesting that women’s empowerment and decision making regarding their healthcare is more applicable to unmarried youth than married counterparts for whom partners and/or in-laws are likely to influence healthcare decision making [46].

This study found that rural areas were associated with lower chances of the use of health facilities at childbirth among both unmarried and married youth. This finding is consistent with previous studies that observed lower odds of the use of health facilities at childbirth in rural areas than urban areas [18, 19, 31–33]. The observed rural-urban differences could be explained by the long distances, the times, and transport means required to reach the health centres during untimed pre-birth labour pains, which are more unfavourable for rural youth than urban youth. There are also poor staffing levels of health facilities in rural areas especially with midwives [10, 47, 48].

Our results showed that the use of health facilities at childbirth was low for both unmarried and married youth employed in the agriculture sector. The negative influence of agriculture could be because it is labour intensive, and this competes with the time to access health facilities for childbirth compared to non-working youth [49]. In addition, most of the agriculture work that these young do is for their own subsistence, and only what is the excess is sold, thus less or no income is obtained from this agriculture compared to other work. Consequently, they have less money to meet the requirements for the use of health facilities at childbirth.

Economic disparities in the use of health facilities at childbirth were also identified in this study among married youth. Married youth in middle, richer and richest wealth quintiles had higher chances of the use of health facilities at childbirth, but not among unmarried youth. Several studies have shown that economic power increases the odds of using health facilities or safe childbirth among youth [18, 20, 25, 26, 28, 31, 32]. This could be explained by the affordability of the services and other indirect costs. Households with relatively high incomes can assign higher proportions of their income to health care, compared to poor households whose priority is meeting household basic needs [31]. The rich also have access to mass media, thus have better knowledge of the benefit of using health facilities at childbirth. However, it is interesting to note that household wealth is not a significant predictor of health facility childbirth for unmarried youth. This may suggest that for unmarried adolescents, social or service factors are likely to be more important than economic power in determining use of health services [24].

In the current study, access to the radio, newspapers, and television was associated with higher odds of the use of health facilities at childbirth among married youth, yet not among unmarried youth. The association between access to mass media and the use of health facilities has been soundly established in previous studies that observed that youth, including adolescents who had access to media, had increased chances of using health facilities at childbirth [20, 26–28]. Access to media increases knowledge of the benefits of the use of health facilities at childbirth, and the dangers of birth outside health facilities thus, increased chances of the use of health facilities at childbirth. Unmarried youth might encounter other barriers even when they are well informed about the benefits of use of health facilities at childbirth. Media, especially the radio, is a good source of maternal health information for Uganda, as most households have access to the radio (55.2%), compared to 7.2 percent and 2.1 percent that have access to television and print media respectively [50].

This study had some limitations that need to be acknowledged. Uganda Demographic and Health surveys (UDHS) collect data from women for births in the last five years before the date of the survey, which may lead to inaccuracies due to memory lapse. In addition, the cross-sectional nature of data from UDHS did not allow us to infer causal relationship between health facility use at childbirth and socio-economic factors. This is because use of health facilities at childbirth is asked for a birth in the last five years, while socio-economic factors are as of the time of the interview. However, the UDHS remains one of the most robust nationally representative data sets in understanding health facility use. Thus, this study increases knowledge about the factors associated with use of health facilities at childbirth among unmarried and married youth in Uganda.

Conclusion and recommendations

Use of health facilities at childbirth was positively related to higher education level and frequent antenatal care use. It was negatively associated with higher parity, rural residence and being employed in the agriculture sector among both unmarried and married youth. In addition, it was positively associated with higher wealth index and more frequent access to media among married youth.

To enhance the use of health facilities among youth in Uganda, there is need to encourage frequent ANC attendance, improve education, household socio-economic status and access to media. Efforts should also aim to remove barriers to the use of health facilities at childbirth for youth with higher parity, those employed in the agriculture sector, and those that reside in rural areas.

16 Mar 2021

PONE-D-20-31194

FACTORS ASSOCIATED WITH HEALTH FACILITY DELIVERY AMONG UNMARRIED AND MARRIED YOUTH IN UGANDA

PLOS ONE

Dear Dr. Agaba,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The manuscript has been evaluated by two reviewers, and their comments are available below. You will see the reviewers have commented on the importance of your manuscript to policy makers. However, the reviewers have also raised critical concerns and the manuscript will need significant revision before it can be considered for publication – you should anticipate that the reviewers will be re-invited to assess the revised manuscript, so please ensure that your revision is thorough. I have outlined some of the key concerns noted by the reviewers below, but you should respond to all concerns mentioned by the reviewers in your response-to-reviewers document.

The key concerns noted by the reviewers relate to the reporting of the methods and the results. Specifically, the reviewers have requested the addition of n values in the results section. Reviewer 2 has requested clarity regarding multivariate model as well as the analysis of the individual- and community-level factors.

Please submit your revised manuscript by Apr 27 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Danielle Poole

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 1 in your text; if accepted, production will need this reference to link the reader to the Table.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The paper is well written and highlights the utilization of health facilities by youth, However, the authors seem to have laid a background on women, not youth specifically. The problem among this group need to be more pronounced. The methods have been well described. Were there limitations since this was secondary data? How was this overcome? The title could be changed to predictors instead of factors since that is what is entirely referred to in the whole paper. The results need to have n values. Line 260, correct ANV to ANC. Line 280, add full stop. The discussion is well done, it compares the findings well with other studies and also there is the author's attribution of the findings. The section on conclusion includes recommendations and it should be titled so.

Reviewer #2: I recommend the authors to use "multilevel logistic regression" analysis since there are community and individual factors rather than using "logistic regression". See the attached files for the detail.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Joyce Jebet Cheptum

Reviewer #2: Yes: Maru Mekie

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Reviewer feedback_MM.docx

Click here for additional data file.

2 Jul 2021

We explored the use of multilevel modeling but the small sample of youth in each cluster could not allow the use of cluster as the second level of analysis.

This is because the Uganda Demographic and Health Surveys sampling is random by cluster and household levels. These would be the perfect levels for the multilevel analysis but both have few cases.

I have referred to table 1 in the text in line 141 in the clean copy of the manuscript

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

3 Mar 2022

Submitted filename: Review comments - PLOS ONE.doc

Click here for additional data file.

Submitted filename: REVIEW PLOS ONE_MM2.docx

Click here for additional data file.

23 Mar 2022

REVIEW PLOS ONE

TITLE: FACTORS ASSOCIATED WITH HEALTH FACILITY DELIVERY AMONG UNMARRIED AND MARRIED YOUTH IN UGANDA

ARTICLE ID: PONE-D-20-31194R1

We appreciate reviewer one for the positive comments. We have tried to improve the manuscript and the grammatical errors.

General comment

The authors tried to address all the issues raised. Hence, I would like to thank the authors for taking comments to improve the manuscript.

Issues to be cleared

Abstract

1. It would be good if the authors consider revising the result presentation to make it sound (It is good to compare findings among married and unmarried youths. But the presentation is not attractive and there is redundancy (“Versus” repeatedly mentioned).

Please try to present the result in abstract in reader friendly manner.

Response: We have tried to edit the abstract to make it sound and attractive

Main document

Methods

2. The authors tried to address the previous comments raised in the method section. However, the are other issues that shall be cleared. What study design have you used? The authors reported nothing regarding the study design used in the manuscript except describing the sampling technique used as well as the type of survey in the DHS data.

Response: UDHS 2006, 2011 & 2016 are cross-sectional surveys so cross-sectional design was used

3. You have tried to compare the level of facility delivery among married and unmarried youths and the associated factors. Is it a comparative study? If so, is the analysis method appropriate?

Yes, this is a comparative study

The analysis method is appropriate because it shows the similarities and differences in the use of health facilities at childbirth among unmarried and married youth

4. If it is not comparative study there may not be a need to put the title as “Married and Unmarried youth”. Please clearly address these issues.

Response: This is a comparative study

Results

5. Pages 9-10 in narration of the result as well as pages 10-12 in Table 2; please correct P = 0.000 to “P< 0.001”,

Response: This has been edited on page in line 197-204

6. You are comparing facility delivery among married and unmarried youth but nothing has been stated about the proportion of married and unmarried youth in each of the variables in table to aside from putting “p” from cross tabulation. Please put the cross-tabulation findings by inserting additional columns in Table 2 “Married youth N(%) “ and Unmarried youth N(%) for each variable”

Response: The first four columns in the previous table showed the descriptive distribution of unmarried and married youth but may be it was not clear. We have edited it as you suggested; page 10-11

7. On pages 16, line 329, Table 3: “Multivariate analysis of factors….” Shall be corrected as “Multivariable analysis of factors…since the two terms.

Response: This has been changed to multivariable

8. Please consider to put the proportion of married and unmarried youth in each of the variables entered in the multivariable model in Table 3 (using cross-tabulation) so that it will be simple for readers to observe consistency as well as to compare variation in the proportion among married and unmarried youths (Example parity one among unmarried & Married, Parity two among married & unmarried …secondary education among married and unmarried etc.) It is a must to have an additional two columns in Table 3 for proportions of unmarried and married youth in each of the variable.

Response: I have included the proportions of married and unmarried youth who have had a childbirth from the health facility; page 13-15

9. On pages 16-17, Table 3, you have put keys for

�  RC: Reference Category

�  CI: Confidence Interval

However, both are not listed as acronyms in Table 3 . Better to remove

Put 1: as reference category below Table 3.

Response: RC & CI have been deleted and 1 used as reference category

Discussion

10. Page 19, lines 378-380 you explained “Unmarried youth were observed to use health facilities more than the married youth in Uganda between 2006 and 2016. This could be due to the confounding effect of higher parity as most married youth were pregnant for the second time or more.

However, nothing has been stated regarding the proportion of “parity among married and unmarried youth” either in the descriptive part of the result section or another area which makes difficult to confirm whether proportion of parity was higher among unmarried youth compared to the married one.

Response: In the descriptive section, we showed the distribution of unmarried and married youth by parity; most unmarried youth (85.6%) were of parity one while most married youth (60.6%) were of parity two, line 178-181

Conclusion and recommendation

Well written and all the comments are addressed.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

25 Mar 2022

PREDICTORS OF HEALTH FACILITY CHILDBIRTH AMONG UNMARRIED AND MARRIED YOUTH IN UGANDA

PONE-D-20-31194R2

Dear Dr. Agaba,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

José Antonio Ortega, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

It is felt that the issues raised were addressed and there is no need to send back to the reviewers. Congratulations!

Reviewers' comments:

30 Mar 2022

PONE-D-20-31194R2

Predictors of health facility childbirth among unmarried and married youth in Uganda

Dear Dr. Agaba:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. José Antonio Ortega

Academic Editor

PLOS ONE