The Effect of a Sexual and Reproductive Health Programme on Socio-Economic and Education-related Inequalities in the Use of Modern Contraceptives in Seven Sub-regions in Uganda: A Case of RISE Programme 2019-2023

doi:10.21203/rs.3.rs-4450185/v1

Background: Universal health coverage is a key SDG3 strategy with no one left behind. Access and utilization of family planning services is important for addressing the needs of women and men for the children they want and when they want them. Although several FP programmes have been rolled out, there is limited evidence to determine their effect on inequality. We assess the effects of the “Reducing High Fertility Rates and Improving Sexual Reproductive Health Outcomes in Uganda (RISE)” on key indicators of sexual reproductive health, including the use of modern contraceptive methods in seven regions in Uganda.

Methods: Baseline and Endline data were obtained from two cross-sectional surveys conducted in 2019 and 2023, respectively. A total of 1341 and 1495 women of reproductive age (15-49 years) were interviewed in 2019 and 2023, respectively. Educated and Wealth-related inequality in the use of modern contraceptive methods (defined as using or not using modern FP methods) were assessed by dimensions of equity ( geography, rural/urban residence, age, and social-demographics characteristics. Inequality was determined using Erreygers Concentration Indices (ECI) at baseline and endline. The difference in ECI between the two survey periods was ascertained and assessed for statistical significance at 5%. We used Prevalence Ratios to compare the use of modern FP at the endline relative to the baseline using a modified Poisson regression run in STATA version 15.

Results: The distribution of participants between the surveys did not significantly vary by characteristics except for a decline in self-reported disability (32.2% to 14.5%, p<0.001) and an increase in per cent with lowest/lower wealth-quintile (36.3% to 43.4%, p=0.0035). The mCPR did not significantly change. However, positive changes were observed in West Nile, Central-1, and East-Central, urban, older women (40-49), the divorced/separated/widowed, and those with primary or no education. We observed no significant change in the use of modern contraceptives at the endline compared to baseline, adj.PR=1.026(0.90, 1.18), p=0.709). Overall, wealth-related inequality in the use of the modern contraceptive method in favor of the wealthiest (higher/highest wealth quintile) women was observed at baseline, ECI=0.172, p<0.001, but not at the endline, ECI=0.0573, p=0.1936. However, Wealth-related inequality declined at the endline. Similarly, overall education-related inequality was highest in favor of women with secondary or higher levels of education at baseline, ECI=0.146(0.035, p<0.001) but not at endline, ECI=0.0561(0.0342, p=0.1063). Although we observed a decline in education-related inequality between the two surveys, this was not statistically significant. The decline in wealth-related inequalities at the endline was more evident in urban, in central-1, East Central and Karamoja regions, among young (20-24) women and the married, while education-related inequality was more common in the rural, older (40-49 years) women, and the married.

Conclusion: The RISE programme provides evidence of a decline in socio-economic and education-related inequalities in selected equity dimensions, especially among older women in rural areas, young women in urban areas, and married women. However, inequalities persist and may need to be addressed with more targeted programmes to ensure that no one is left behind for UHC.

Globally, efforts are currently geared towards measuring health equity as an essential step to ensuring quality health care and, ultimately, the population's health quality [1, 2]. Health inequities result from avoidable social injustices when the distribution of processes, services and resources is not based on the “need” of various sub-groups and are remediable [3–6]. Therefore, measuring health across the social determinants of health (SDOH) is fundamental to understanding the underlying distributive injustices or health equity of services, programmes, and policies. There is a need to monitor progress in health and strategies for effective interventions to promote opportunities for healthy and longer lives, “leaving no one behind”. [7–10, 31].

Within the health policy context, the availability of health-promoting resources, including those for contraceptive use, may determine processes that inform service delivery access and use. This suggests that policies, programmes, and resource allocation affect the presence, location, and organization of healthcare physical infrastructure, the provision of health supplies and services, staffing, and human resource management, and affects utilization [11]. Equity in family planning (FP) means that regardless of the SDOH, individuals have the same access to information and services, including available methods of contraception, and can make decisions about their fertility and use of contraception and act on those decisions [5].

The Equity Strategic Planning Guide highlights the critical steps to creating equitable access to high-quality FP information and services. Such important steps include determining whose needs are not being met, exploring the barriers faced in accessing the services, making the FP programmes responsive to the values and needs of all people, and monitoring the implementation of programmes and policies [12].

According to the World Health Organization, interventions to promote equity can be based on analyzing health disparities and their causes. So, health inequality indicators, including those that measure utilization, can be used as metrics to track possible inequities in the distribution of processes, services, and resources across the SDOH [13, 14–16]. Therefore, inequities in FP programmes can be implied by the differences in health equality indicators such as maternal mortality, contraceptive use, and unwanted pregnancies, among others [14–17].

FP is vital for the fertility rate decline, economic development, and, eventually, achieving the demographic dividend. Although recent data show an increase in contraceptive use in some countries, there are still inequities in FP access and uptake within and across countries in Sub-Saharan Africa. The within and between country differences in FP access and uptake are mainly due to persistent socio-cultural and economic barriers [18–24]. In Uganda, inequities in the use of modern contraceptive use have been highlighted [17 − 8,25–27], and several FP programmes have been designed and implemented in Uganda to improve access to and utilization of FP services and sexual reproductive health outcomes at large [17, 28–29].

The Reducing High Fertility Rates and Improving Sexual Reproductive Health Outcomes (RISE) programme is one such programme that was designed to improve FP uptake through addressing inequities in FP services. Specifically, RISE focused on enhancing the uptake of FP through increased awareness, access to and utilization of high-quality FP services, and strengthening the capacity of the public and private sectors to provide such services in Uganda. The RISE programme was a consortium of five partners led by Maries Stopes International, and the roles and responsibilities of each partner have been described elsewhere [17]. The programme implementation began in November 2018 and ended in October 2023, and it aimed to promote equity in FP service with special targets that included adolescents and young girls (AGYW) and people with disabilities (PWD) [30].

In this paper, we discuss the RISE programme's effect on contraceptives among women across the intervenable community socioeconomic and education-related inequities in seven sub-regions in Uganda. The findings will inform future programmes and policies aimed at addressing inequities in FP service access and improving the coverage of the sexual reproductive health (SRH) outcomes across the various population sub-groups in the country, “leaving no one behind”.

Study design.

Data for this analysis were obtained from baseline (2019) and endline (2023) surveys of the RISE programme conducted in seven statistical regions. The regions were per the Uganda Bureau of Statistics (UBOS) demarcations of 2010, including Central 1, Central 2, East-Central, Eastern, Karamoja, Western, and West Nile. These data are part of the monitoring and evaluation of the “Reducing High Fertility Rates and Improving Sexual Reproductive Health Outcomes in Uganda (RISE)” programme implemented in Uganda.

Sample size and sampling.

The sample sizes were 2596, 2943, and 2902 for baseline, midline and endline respondents, respectively. A formula in a paper by Chow, 2008[19] and World bank[20] was used to compute the sample size. Key inputs into the formula included the minimum detectable effect size (MDES), the probability of the event of interest, which we took to be an intention to use FP, and the design effect to cater for the clustering effect and non-response at the individual and household levels. Previous surveys provided an estimation of the non-response. Detailed information on the computation of the sample size and sampling can be obtained from a published paper on the baseline.[21].

Data collection

In each household, the details of name, index number, age and sex were entered into the pre-programmed listing form within the ODK online data collection software. A random selection of one eligible participant per household was carried out using a code developed within the ODK’s programming enabling option. The selection of households followed a non-substitution policy to minimize a misrepresentation of population distribution from our sample. At least three callbacks were made at different times and days before declaring the respondents unavailable for interview. Each research assistant interviewed a respondent of the same sex to improve the data quality. Consent was sought from the eligible participants before collecting data from them and the study was approved by Makerere University School of Public Health-Research and Ethics Committee.

Measurements

For this study, equality of outcome was assessed across three dimensions: i) Geography, defined as rural/urban residence based on the location of EA as assigned by the national statistical office ii) wealth/economic, defined as lowest, lower, middle, higher and highest quintiles; measured by ownership/possession of household assets (Communication: cell or landline phone, computer, radio, TV; Transport: bicycle, motorcycle, car; Source of power: electricity, generator, solar, and materials for building/construction, ownership of the home, indoor bathroom, access to running water), and iii) Social-demographics that included marital status (Never married, married or divorced/widowed), the highest level of education attained (None, Primary or Secondary and above) and age in completed years categorized as 15–19 years, 20–24 years, 30–39 years and 40–49 years, that provides adolescents young person old women and older women. Other socio-demographics were disability status as measured by the Washington Group (No-difficulty, some or a lot of difficulty) [40], Employment status (Unemployed, student or employed), and the seven sub-regions (Western, Central-1, Central-2, East Central, Eastern, Karamoja and West Nile). The primary outcome variable of interest for this analysis was the use of modern family planning., measured as reported use or non-use of a modern FP method.

Data management and analysis

Although the large RISE programme survey had males and females, this analysis focuses only on eligible females aged 15–49.

Construction of analytical weights

All analyses were conducted with STATA software version 15 using the surveyset methodology that accounts for the survey design. All the analyses were weighted, unless specified, with EA selection weights provided by the Uganda Bureau of Statistics (UBOS). The EA selection weights were adjusted for the probability of selecting a household per EA, eligible participants per sampled household, and the non-response rate at the EA level to generate the final weights used to weight the analysis. The descriptive analyses were stratified at baseline (2019) and endline (2023).

Analysis to determine factors associated with the use of modern contraceptives between the baseline and endline surveys.

We conducted exploratory data analyses to generate descriptive statistics from the two surveys focusing on participant socio-demographic characteristics. Categorical variables were presented as unweighted frequencies and weighted proportions. We assessed for changes in the distribution of the participant socio-demographic characteristics between the two surveys using chi-square statistics. To determine factors associated with the outcome—use of modern contraceptives, we used prevalence ratio (PR) as a measure of association instead of the odds ratio (ORs) because the primary outcome was common (greater than 10%) and use of PR would minimize the overestimation of the association. The PR was obtained using a “modified” Poisson regression model via a generalized linear model (GLM) with the family as Poisson and a log link under the svyset because this survey used a complex design. The use of modern contraceptives as the primary outcome was coded “1” if a participant was a current user and “0” if not a current user of a modern FP method. All PRs were estimated with their corresponding 95% confidence intervals (CI). The dimensions of equity used in this analysis were i) Geography defined as rural/urban or statistical region, ii) Economic based on wealth-quintile, and iii) Socio-demographic based on age- categorization as adolescent (15–19 years), young women (20–24 years), and older women grouped as 25–39 and 40–49 years, and marital status, and iv) Education status (None, Primary, Secondary, or higher). In the adjusted model, all factors with a statistical significance, p < 0.05, were considered important, and a selection of two competing models was determined using Akaike's Information Criteria (AIC) to select the “best” model.

The interaction of key predictors associated with the primary outcome was assessed. The use of interaction in the analysis is important as it helps to identify subpopulations that need targeted programmatic interventions geared towards improving service uptake. An interaction term between marital status and women’s age was explored to determine if the use of modern contraceptives varied by the broader age groups (adolescents, young or older age). The statistical significance of the covariates and the interaction terms was determined using the Wald’s test (testparm in Stata (version 15)). All the analyses were weighted to obtain representative estimates, and this was conducted via the use of the svyset commands in Stata, which also accounted for the clustering of observations at the EA level.

Analysis of Wealth and education-related equality in the use of modern conceptive

Concentration curve

Wealth quintiles were considered as the proxy for socioeconomic status, while the highest level of education, which was categorized in 3 order level, was used to measure education status in this study. The cumulative proportion of women reporting the use of modern contraceptives was plotted on the y-axis) against the cumulative proportion of the women ordered/ranked by wealth or education status. The ranking of wealth or education status is considered from the lowest to the highest on the x-axis. An equally distributed outcome, use of modern contraceptives, across the wealth-quintiles or levels of education results in a concentration curve, with a line diagonal at 45° line showing equality. Any deviation from the diagonal 45° line, either above or below, is an indication of inequality in the distribution of the outcome by the population characteristic, wealth-quintile or level of education. The further the curve lies from the line of equality, the greater the degree of Inequality in the outcome (use of the modern FP method). The concentration curve that lies above the line of equality, the 45° line, shows that the outcome variable ranks with higher values among people in lower ranks of the population characteristics, defined as inequality in favor of the worse off population with respect to the characteristic. However, the concentration curve below the line of equality shows that the outcome ranks lower in the lower ranks of the population characteristic defined as inequality in favor of a better ranked population.

Concentration index

The concentration index provides a measure of the degree of Inequality in an outcome over the distribution of another variable, usually a population characteristic, in this case, education and wealth. Concentration indices as a measure of inequality have been used and are a common measure of socioeconomic-related health inequality. They are here used to measure wealth and education-related inequality in the use of modern contraceptives among women in two surveys.

The use of the concentration indices in this study was to determine if the sexual and reproductive health programme, including the use of family planning, has reduced inequality in the use of modern contraceptives across the programme population in Uganda. Therefore, wealth and education-related concentration indices were compared between the baseline (2019) and Endline (2023) surveys overall and stratified by other key dimensions of equity: residence (urban/rural), geography (regions) and age (in years). Also, differences between the Erreygers Concentration Index (ECI) at baseline and endline were assessed for statistical significance using a p-value with a cut of 5%.

The standard concentration index is derived from the concentration curve, representing twice the area between the concentration curve, either above or below the 45° line of equality (2,3). The outcome for the analysis, use or non-use of modern contraceptives, is a bounded binary indicator variable, leading to the use of a modified version of the concentration index, the Erreygers Concentration Index or ECI. The ECI satisfies the conditions that the absolute value of the index is the same regardless of whether the outcome used to assess Inequality is users or non-users of modern contraceptives (mirror property) and that the value of the index is invariant to any feasible positive linear transformation of the outcome variable (scale and translation invariance) (4,5). The ECI is defined as:

$$ECI\left(h\right)=\frac{8}{{n}^{2}({b}_{h }-{a}_{h})}\sum _{i=1}^{n}{h}_{i}{R}_{i}$$

where h_𝑖 _is the outcome variable, use or non-use of modern contraceptive, 𝑅_𝑖 _is the fractional rank of woman 𝑖 in the distribution of wealth-quintile status, 𝑛 is the number of observations and 𝑏_h and 𝑎_h are the variables upper bound and lower bound, respectively. The equation shows that the concentration index can be interpreted as a sum of weighted outcome levels, with the weights being determined by the wealth-quintile rank (or educational-level rank) in the case of this study. The ECI is a measure of absolute Inequality for bounded variables, a possible range from − 1 to + 1. An ECI with a negative value indicates the concentration of the outcome in favor of the more disadvantaged (poor or less educated) or lower-ranked population. In contrast, an ECI that is positive indicates concentration among the more advantaged (rich or more educated) or higher-ranked population, and an ECI of zero shows no inequality. The command index with the Erreygers option in STATA v154.2 was used to calculate wealth and education related Erreygers concentration indices (6).

Table 1 shows the characteristics of participants stratified by baseline and endline surveys conducted in 2019 and 2023, respectively. The distribution of participants between the surveys did not significantly vary by region, rural-urban residence, age, marital status, highest level of education and current employment status. We, however, observed a significant change in SES as measured by wealth-quintile with a decline in the percent of participants in the higher/Highest quintile from 40% at baseline to 34.8% at endline, or percent in lowest/lower wealth-quintile from 36.3% at baseline to 43.4% at endline, p = 0.0035. Also, the percentage reporting at least one form of disability declined from a third (32.2%) at baseline to 14.5% at the endline, p < 0.001. However, we observed no difference between baseline and endline among participants who were rural (~ 73%), aged 25–39 years (~ 36%), currently married, or with a secondary or higher level of education.

Table 1

Characteristics of participants stratified by survey (Baseline and Endline)
	Survey period
	Baseline (2019)		Endline (2023)
	Unweightedn	Weighted, %	Unweighted n	Weighted, %
Number	1,346	100	1,499	100
Sub-regions
Western	200	18.5	247	24.1
Central-2	224	14.9	222	10.9
Central-1	216	23.4	255	23.1
East-Central	196	13.9	201	11.8
Eastern	242	17.2	292	19.2
Karamoja	157	3.84	158	3.51
West Nile	111	8.23	124	7.34
Characteristics
Residence
Rural	1,076	73.2	1,189	72.9
Urban	270	26.8	310	27.1
Wealth-quintile (p = 0.0035)
Lowest	445	24.4	457	22.5
Lower	154	11.9	369	20.9
Middle	273	22.8	275	21.9
Higher	357	29.4	306	25.8
Highest	117	11.6	92	8.96
Age categories (years)
15/19	200	15.4	206	14.2
20/24	308	23.0	320	21.6
25/39	484	36.3	529	35.3
40/49	354	25.2	444	29.0
Marital status
Never-Married	206	17.2	204	14.6
Married	1,004	72.2	1,140	74.7
Divorce/Separated/widowed	136	10.6	155	10.7
Self-reported disability (p < 0.001)
None	917	67.8	1,275	85.5
At least one	429	32.2	224	14.5
Level of education attained (P = 0.0608)
None	270	14.5	245	10.6
Primary	713	53.4	820	55.2
Secondary or higher	363	32.1	434	34.2
Current employment status
Unemployed	263	19.1	251	18.4
Student	83	7.54	79	5.44
Employed	998	73.3	1,166	76.1

Table 2 shows the use of modern contraceptives, measured as modern contraceptive prevalence rate (mCPR) across participants' characteristics stratified by the baseline and end-line surveys. The mCPR improved by about 2%, from 34.2%;95%CI (30.8, 37.7) at baseline to 36.4%;95%CI (32.1, 40.8) at endline, but this change was not statistically significant. Comparing the baseline to endline period, the largest increase in mCPR by region was observed in West-Nile (18.6–26.9%), followed by Central-1 (36.6–42.8%), East-Central (40.3–45.2%). Other characteristics with increase in mCPR above the average 2% increase over this period were urban (37.6–41.4%), about 5% increase across each of the three wealth-quintile of lowest, lower and middle, older women (40–49 years 27.6–36.3%), or women who were divorced/separated/widowed, 28.8–44.3%, and women with primary or no education attained categories.

Table 2

Modern contraceptive prevalence (mCPR) by **participants' characteristics (Baseline and Endline)**
	Survey period
	Baseline (2019)			Endline (2023)
	unweightedn	Weighted, mCP	95%CI	Unweighted n	Weighted, mCP	95%CI
Number	1,346	37.7	35.1, 40.3	1,499	37.5	35.0, 40.0
Sub-regions
Western	200	36.3	30.5, 42.6	247	33.5	23.4,45.5
Central-2	224	34.3	25.9, 43.9	222	36.6	27.9,46.2
Central-1	216	36.6	30.5, 43.2	255	42.8	34.5,51.5
East-Central	196	40.3	33.4, 47.6	201	45.2	32.9,58.1
Eastern	242	37.1	26. 5, 49.3	292	35.3	29.7,41.3
Karamoja	157	6.5	3.1, 13.2	158	9.2	2.5,28.1
West Nile	111	18.6	14.1,24.1	124	26.9	18.9,36.7
Characteristics
Residence
Rural	1,076	32.9	28.8,37.3	1,189	34.5	29.7,39.7
Urban	270	37.6	31.5,44.2	310	41.3	33.9,49.1
Wealth-quintile
Lowest	445	24.5	19.3,30.6	457	29.9	22.7,38.2
Lower	154	30.6	23.1,39.3	369	35.6	27.1,45.0
Middle	273	29.9	24.9,35.4	275	43.5	37.7,49.5
Higher	357	42.7	36.3,49.4	306	37.6	29.4,46.6
Highest	117	45.0	37.0,53.2	92	33.5	21.4,48.2
Age categories (years)
15/19	200	14.4	9.5,21.2	206	16.8	10.6,25.5
20/24	308	41.2	34.7,48.1	320	36.0	29.8,42.7
25/39	484	42.7	37.7,47.8	529	44.5	39.2,50.0
40/49	354	27.6	21.8,34.3	444	36.3	29.9,43.2
Marital status
Never Married	206	22.9	15.4,32.6	204	20.8	12.1,33.6
Married	1,004	37.7	33.2,42.3	1,140	38.3	34.2,42.5
Divorce/Separated/ widowed	136	28.8	20.3,39.2	155	44.3	33.6,55.6
Self-reported disability
None	917	36.3	32.3,40.4	1,275	36.8	32.3,41.6
At least one	429	29.8	25.1,35.0	224	33.6	25.2,43.3
Level of education attained
None	270	19.7	14.5,26.1	245	27.9	18.6,39.5
Primary	713	33.3	29.5,37.2	820	36.3	31.7,41.3
Secondary or higher	363	42.3	36.1,48.8	434	39.0	32.9,45.5
Current employment status
Unemployed	263	35.6	28.4,43.5	251	37.8	28.4,48.2
Student	83	18.9	9.8,33.6	79	6.0	1.4,22.8
Employed	998	35.4	30.9,40.2	1,166	38.2	33.7,42.9

Figure 1 shows the prevalence ratios (PR) from the adjusted regression model with the corresponding 95% confidence intervals plotted. We observed no significant change in the use of modern contraceptives at the endline compared to baseline, adj.PR = 1.026(0.90, 1.18), p = 0.709. Relative to the Central-2 region, the use of modern contraceptive use was significantly lower in Karamoja adj.PR = 0.27(0.095, 0.776), p = 0.016, and West-Nile adj.PR = 0.72 (0.56, 0.93), p = 0.013, and lower in participants with at least one self-reported form of disability compared to those who did not report any disability, adj.PR = 0.86(0.75, 1.00), p = 0.051. Conversely, compared to women with no education, women with secondary or higher levels had higher use of modern contraceptives, adj.PR = 1.35(1.04, 1.74), p = 0.023, or primary adj.PR = 1.24(0.97,1.58), p = 0.084.

Wealth-related inequality in the use of modern contraceptives stratified by women characteristics between Baseline and Endline surveys.

Figure 2 shows the wealth-related concentration curves for the use of modern contraceptives at baseline and endline surveys. In contrast, Table 3 shows the Erreygers Concentration index (ECI) of wealth-related inequality in the use of modern contraceptives between baseline and endline surveys, overall and across participant characteristics: the type of residence, sub-region, age (years), disability status and marital status. Overall, wealth-related inequality in the use of modern contraceptives was significantly higher in favor of the wealthiest (higher/highest wealth quintile) women at baseline, ECI = 0.172, p < 0.001, but not the endline, ECI = 0.0573, p = 0.1936. The difference in inequality between the two surveys was statistically significant, -0.114, p = 0.0349. When stratified by rural-urban residence, significant inequalities in favor of wealthier women were observed at baseline in both the rural (p = 0.0018) and urban (p = 0.0009), but no inequalities were observed at endline in either rural (p = 0.2346) or urban (p = 0.8356). However, differences in inequality comparing baseline to endline were larger in the urban, -0.236, p = 0.0564, and nearly negligible in the rural, -0.083, p = 0.2077. Similarly, significant declines in wealth-related inequality in favor of women with higher education level comparing baseline to endline were observed in the two regions of Central_1 -0.271, p = 0.0273, East-Central − 0.2558, p = 0.0304. In Karamoja region, the inequality changed from being in favor of the less wealthy women (lower/lowest) at baseline, ECI=-0.003, p = 0.0394 to the wealthier women at endline, ECI = 0.0217, p-<0.001, and this difference was significant 0.0247, p < 0.001. Women aged 20–24 years also had a significant reduction in inequality from baseline, ECI = 0.238, p = 0.0003, to endline ECI = 0.0281, p = 0.7164, with a difference of -0.209, p = 0.0342. A similar decline in favor of the wealthier women was also observed among the married, with a difference of -0.1615, p = 0.0121, but not the never-married or the divorced/widowed/separated.

Table 3 Erreygers Concentration Index (ECI) of Wealth-related Inequality in the use of modern contraceptive methods comparing Baseline and Endline surveys by levels of residence (rural-urban), region, age (years) and marital status.

	Baseline (2019)				Endline (2023)				Differences in ECI (SE)
	N	Absolute ECI	SE	p-value	N	Absolute ECI	SE	p-value
Overall	1341	0.172	0.032	< 0.001	1495	0.0573	0.0436	0.1936	-0.114(0.05417), p = 0.0349
Residence
Rural	1071	0.145	0.041	0.0009	1185	0.0621	0.0516	0.2346	-0.083(0.0659), p = 0.2077
Urban	270	0.213	0.053	0.0018	310	-0.0237	0.112	0.8356	-0.236(0.124), p = 0.0564
Region
Western	198	0.146	0.0823	0.1135	246	0.0884	0.1246	0.4960	-0.0579(0.149), p = 0.6984
Central-2	224	0.135	0.0756	0.1051	222	0.104	0.0577	0.1049	-0.0308(0.095), p = 0.7473
Central-1	215	0.143	0.063	0.0466	255	-0.1282	0.1054	0.2516	-0.271(0.1228), p = 0.0273
East-Central	194	0.157	0.070	0.0549	201	-0.0985	0.0952	0.3310	-0.2558(0.118), p = 0.0304
Eastern	242	0.1744	0.0783	0.0478	292	-0.0025	0.0902	0.9787	-0.1769(0.1195), p = 0.1388
Karamoja	157	-0.003	0.001	0.0394	158	0.0217	0.0012	< 0.001	0.0247(0.0016), p < 0.001
West Nile	111	-0.007	0.067	0.9240	121	-0.095	0.0279	0.0271	-0.0882(0.0725), p = 0.2239
Age (years)
15–19	199	0.071	0.063	0.2596	205	-0.0378	0.0561	0.5030	-0.1092(0.0841), p = 0.1942
20–24	306	0.238	0.062	0.0003	319	0.0281	0.0771	0.7164	-0.209(0.0989), p = 0.0342
25–39	483	0.165	0.055	0.0035	527	0.0920	0.0576	0.1152	-0.0734(0.0793), p = 0.3544
40–49	353	0.178	0.057	0.0029	444	0.0306	0.0591	0.6058	-0.1468(0.082), p = 0.0742
Marital status
Never married	206	0.155	0.061	0.0144	204	0.0697	0.0929	0.4561	-0.0853(0.111), p = 0.4436
Married	999	0.227	0.040	< 0.001	1138	0.065	0.0501	0.1962	-0.1615(0.064), p = 0.0121
Divorced/Widowed	136	0.050	0.097	0.609	153	0.0515	0.105	0.6268	0.00176(0.143), p = 0.9902

Education-related inequality in the use of modern contraceptives stratified by women characteristics between Baseline and Endline surveys.

Table 4 shows the Erreygers Concentration Index (ECI) of education-related inequality in the use of modern contraceptive methods between baseline and endline, overall and across participant characteristics: the type of residence, age (years), and marital status. The overall education-related inequality was highest in favor of women with secondary or higher levels of education at baseline, ECI = 0.146(0.035, p < 0.001) but not at endline, ECI = 0.0561(0.0342, p = 0.1063). However, the difference in inequality between the two surveys, -0.090, p = 0.0665, did not attain statistical significance. When stratified by rural/urban residence, a significant decline in inequality between baseline and endline was observed in the rural, difference, -0.1483, p = 0.0049, but not in the urban, 0.0161, p = 0.8912. Also, the decline in the inequality in favor of women with secondary or higher education was observed in the older (40–49 years) women, with a difference of -0.157, p = 0.0579 and the married, with a difference of -0.140, p = 0.0116 but not the other age categories or marital status.

Table 4

Erreygers Concentration Index (ECI) of Education-related Inequality in the use of modern contraceptives comparing Baseline and Endline surveys by the levels of residence (rural-urban), age (years) and marital status.
	Baseline 2019					Endline 2023			Differences in ECI (SE)
	N	Absolute ECI	SE	p-value	N	Absolute ECI	SE	p-value
Overall	1341	0.1462	0.0352	< 0.001	1495	0.0561	0.0342	0.1063	-0.090(0.0491), p = 0.0665
Residence
Rural	1071	0.1365	0.0405	0.0015	1185	-0.0118	0.0338	0.7271	-0.1483(0.0527) p = 0.0.0049
Urban	270	0.137	0.0887	0.1493	310	0.153	0.0773	0.0716	0.0161 (0.1177), p = 0.8912
Age (years)
15–19	199	0.0145	0.0592	0.8067	205	0.0229	0.0620	0.7137	0.008(0.0857), p = 0.9228
20–24	306	0.162	0.061	0.0107	319	0.0647	0.0690	0.3518	-0.097(0.0924), p = 0.2929
25–39	483	0.134	0.057	0.0228	527	0.0684	0.0536	0.2067	-0.0655(0.0785), p = 0.4042
40–49	353	0.211	0.0635	0.0015	444	0.0544	0.0531	0.3093	-0.157(0.0828), p = 0.0579
Marital status
Never married	206	0.145	0.569	0.0133	204	0.179	0.0963	0.0681	0.0337(0.112), p = 0.7630
Married	999	0.197	0.040	< 0.001	1138	0.0569	0.0387	0.1473	-0.140(0.0556), p = 0.0116
Divorced/ Widowed	136	0.071	0.0797	0.3804	153	0.1346	0.107	0.2142	0.0641(0.133), p = 0.6312

The study findings show that the change in modern contraceptive prevalence rate (mCPR) was not significant between baseline and endline; however, significant improvement in mCPR was observed in the West-Nile region, the middle wealth quintile, and the divorced/widowed/separated women. Relative to the central 1 region, West Nile and Karamoja had persistently lower mCPR in both surveys. Similarly, women with at least one self-reported disability had lower mCPR compared to those with no reported disability.

Wealth-related inequality was higher in favor of the wealthiest (higher/highest wealth quintile) women as observed at baseline but not at the endline. The difference in inequality between the baseline and endline was significant, indicating a decline in inequality at the endline. The decline in inequality was more evident in the urban, in central-1, East Central, and Karamoja regions, among young (20–24) women and the married.

Similarly, education-related inequality was highest in favor of women with secondary or higher levels of education at baseline but not at the endline. The difference in inequality at baseline and endline was significant, also indicating a decline, which declines were more common in rural, older (40–49 years) women and married women.

These findings show the persistent disparities in mCPR to the disadvantage of the West-Nile and Karamoja regions. The regional disparities in mCPR inequality are also consistent with the high mean ideal number of children. Regions with low mCPR have a high mean ideal number of children (UDHS2016). The ideal number of children and low mCPR may be suggestive of the continued high fertility desires and the potential inequitable service delivery in such settings. It is therefore important to further assess the different social determinants of health—including the social needs, how they operate, and how they can be changed for communities to achieve health equity and improve health. There is a need to have well developed and focused strategies and interventions that will address underlying issues for various population subgroups [32, 33]. Likewise, the inclusive design of systems, FP programmes, policies, and health structures should be central to the discourse on improving equity to ensure that marginalized groups have an opportunity to meet their contraceptive needs [34].

We also observed that wealth-related inequalities in the use of modern contraceptive methods were higher in favor of the wealthiest women at baseline. The decline in wealth-related inequality at the endline was more evident in the urban, in central-1, East Central and Karamoja regions, among young (20–24) women and the married. These declines are suggestive of the success of the RISE programme on contraceptive methods use among the subpopulations across the equity dimensions of residence/geography, age, and marital status. This finding is consistent with recent evidence that shows reductions in wealth-related inequalities in favor of the poorest households among married adolescents and young women in Sub-Saharan Africa [51]. However, in this study, we were unable to attribute the improvement to the RISE programme as other factors may have been important during this period. Two regions, Central-1 and East Central, experienced a significant decline in pro-rich wealth-related inequality at the endline, while Karamoja, with prop-poor inequality, revised to pro-rich inequality. However, this study suggests that FP programmes that target disadvantaged subgroups may be an important approach to addressing the inequalities in the outcomes of the service and, subsequently, health outcomes among the disadvantaged subgroups across the dimensions of equity [37–40].

Similarly, the decline in education-related inequality at the endline was significant and more apparent in the rural, older (40–49 years) women and the married. The declining inequalities at the endline are consistent with a similar study from Ghana that highlighted and recommended a need to target women in rural areas, low wealth, and no formal education to improve contraceptive method use [50]. While the education-related inequality at the endline declined among the older women and in the rural area, the wealth-related inequalities declined among the young (20–24) women and in the urban. This observation may be associated with who was targeted and whereby the RISE programme and may suggest the difference in the effects of the interventions by subpopulation and their geographical locations. Therefore, understanding the subpopulation and mechanisms through which interventions work or how education and wealth affect health is important for programming and policy [42]. The persistent education-related and wealth-related inequality in the never-married indicates a need to target the adolescents who may have an unmet need for spacing or the older women who are separated/divorced/widowed who may have an unmet need for limiting.

It is noteworthy that all findings highlight that although socio-economic and education-related inequities improved for some sub-populations, inequalities persisted in other dimensions at the endline of the RISE programme. This underscores the crucial role of the various social determinants of health in achieving equity for family planning programmes and that there is no singular one of them that can be addressed to achieve health equity. Social determinants of health act jointly and on a gradient, and strategies to address them can be conceptualized similarly to identify and address barriers to contraceptive use across disadvantaged subgroups through partnerships and referrals, advocacy and policy and structural changes within the health system to address the disparities in contraceptive use for UHC [48–49].

Strengths and Limitations

To our knowledge, this is one of a few studies that have provided evidence of the effect of an FP programme on addressing inequalities in modern contraceptive method use. The self-reported outcome variable (use of modern contraceptive methods) might have influences on socially desirable responses because the study communities are receiving FP intervention from the RISE programme. However, this was minimized using well-trained and experienced research assistants, and so the findings of the use of modern contraceptives are consistent with recent national-level surveys. We were unable to conduct an attribution analysis for interventions implemented by the RISE programme in this pre-post approach without a control. Therefore, some improvement in minimizing or eliminating inequality may be due to other partners or government services in these settings. Also, this study did not consider other important factors, such as culture or partner-related characteristics, that could influence women’s fertility desire and subsequent use or non-use of contraception. However, our findings show persistent socio-economic and education-related inequality in the use of the modern contraceptive method in favor of pro-wealthy or pro-educated women. This was a representative sample of seven out of the ten regions in Uganda, and therefore, the findings reflect the effect of the RISE programme in about 70% of Uganda.

This study is a follow-on study set out to evaluate the effect of an FP programme on addressing FP inequities since measuring inequities using appropriate indicators, identifying who/where to intervene effectively, recognizing the underlying multifaceted contributors, and effective monitoring are critical to promoting FP uptake opportunities for all people regardless of their social determinants of health [11–13], and “leave no one behind”.

Recommendations.

FP programmes should continue to identify and target appropriate interventions to close the inequality gaps in mCPR, especially for equity dimensions where women are still left behind. Targeting women in the poor quintile or women in lower education levels seems to have closed the inequality gap in favor of the wealthiest or women in secondary or higher education levels.

A multi-sectoral approach, where the sectors of education, health, cultural/gender, and finance should collaborate to minimize inequalities in family planning service uptake.

Although this research used inequality in mCPR as a proxy for inequities, studies that target process indicators that measure inequities in service access are needed to inform more targeted FP information and services to the appropriate groups that need them.

We propose qualitative research that explores people’s perceptions of why there was success in reducing inequity and why, in some areas, this may not have been realized.

The RISE programme provides evidence of a decline in socio-economic and education-related inequalities in selected equity dimensions, especially among older women in rural areas, young women in urban areas, and married women. However, inequalities persist and may need to be addressed with more targeted programmes to ensure that no one is left behind for UHC.

Ethics approval and consent to participate.

The institutional review board at the Makerere University School of Public Health and the Uganda National Council of Science and Technology (UNCST), protocol number 706, approved the study protocol. The study was conducted in accordance with the Declaration of Helsinki guidelines and regulations and informed consent to participate in the study was obtained from each randomly selected participant.

Consent for publication.

Not applicable

Availability of data and materials

The dataset generated and analyzed during the study is not publicly available due to confidentiality concerns, but upon reasonable request, data will be shared by Marie Stopes Uganda; and stripped of original data identifiers to ensure confidentiality. Only variables used for the analysis will be shared.

Competing interests

The authors declare that they have no competing interests.

Funding

This study was made possible through the RISE programme PO7691, funded by UK aid from the British People; however, the views expressed do not necessarily reflect the UK government’s official policies.

Author contribution

FM: Conceptualization of study, analysis, writing the initial draft, collating feedback, editing, and reviewing the final version. SN: Writing the initial draft, collating feedback, editing, and reviewing the final version. NMT: Conceptualization of study, supervision, analysis,editing, and review of the final version CN: Conceptualization of study design, Data collection Coordination, review of the draft and final version. MA: Supervision of data collection, investigation, and manuscript review. AL: Validation and manuscript review. SR: Data analysis, editing, and reviewing the final version. RT: Validation, Review of draft, and editing. and administration. GA: Validation, Review, and editing. SC: Validation, Review draft, and administration PD: Manuscript review and project administration

Agniel D, Cabreros I, Damberg CL, Elliott MN, Rogers R. A Formal Framework for Incorporating Equity into Health Care Quality Measurement: Study proposes a framework to incorporate equity into quality measurement. Health Affairs. 2023 Oct 1;42(10):1383-91.
Jensen N, Kelly AH, Avendano M. Health equity and health system strengthening Time for a WHO re-think. Global Public Health. 2022 Mar 4;17(3):377-90.
Murti B. Concepts, Principles, and Policy Approaches to Tackling Health Inequity. Diakses dari http://theicph. com/wp-content/uploads/2017/08/Health-equity_Prof-Bhisma-Murti. pdf. 2017 Nov
CSDH. Closing the gap in a generation: health equity through action on the social determinants of health. Final Report of the Commission on Social Determinants of Health. Geneva: World Health Organization; 2008.
Hardee K, Houghtaling E, Stratton S, Askew I, Venkatraman CM, Maggwa B, de Leon RG, Malarcher S. Discussion paper on equity for the HIP partnership. https://www.fphighimpactpractices.org/wp-content/uploads/2020/08/HIP-Equity-Discussio
Paper. pdf
Williams A, Cookson R. Equity in health. Handbook Health Econ.2000;1:1863–910.
Whitehead M. The concepts and principles of equity and health. Health Promot Int. 1991;6(3):217–28.
Ansari Z, Carson NJ, Ackland MJ, Vaughan L, Serraglio A. A public health model of the social determinants of health. Soz Praventivmed. 2003;48(4):242–51. Available from:. https://doi.org/10.1007/s00038-003-2052-4.
Dahlgren G. Whitehead M. Policies and Strategies to Promote Social Equity in Health. Stockholm, Sweden: Institute for Futures Studies; 1991.
Solar O, Irwin A. A conceptual framework for action on the social determinants of health. Geneva, Switzerland: World Health Organization (WHO); 2010.
Dover DC, Belon AP. The health equity measurement framework: a comprehensive model to measure social inequities in health. International journal for equity in health. 2019 Dec;18(1):1-2.
Equity Strategic Planning Guide: Creating equitable access to high-quality family planning information and services. https://www.fphighimpactpractices.org/guides/creating-equitable-access-to-high-quality-family-planning-information-and-services/
World Health Organization. Inequality monitoring in sexual, reproductive, maternal, newborn, child and adolescent health: a step-by-step manual.
Komasawa M, Yuasa M, Shirayama Y, Sato M, Komasawa Y, Alouri M. Demand for family planning satisied with modern methods and its associated factors among married women of reproductive age in rural Jordan: a cross-sectional study. PloS one. 2020;15(3): e0230421.
Gahungu J, Vahdaninia M, Regmi PR. The unmet needs for modern family planning methods among postpartum women in Sub-Saharan Africa: a systematic review of the literature. Reproductive Health. 2021;18(1):1–5.
Cavallaro FL, Benova L, Macleod D, Faye A, Lynch CA. Examining trends in family planning among harder-to-reach women in Senegal 1992–2014. Sci Rep. 2017;7(1):1–1.
Makumbi FE, Nabukeera S, Tumwesigye NM, Namanda C, Atuyambe L, Mukose A, Ssali S, Ssenyonga R, Tweheyo R, Gidudu A, Sekimpi C. Socio-economic and education related inequities in use of modern contraceptive in seven sub-regions in Uganda. BMC Health Services Research. 2023 Feb 28;23(1):201.
Ewerling F, Victora CG, Raj A, Coll CV, Hellwig F, Barros AJ. Demand for family planning satisied with modern methods among sexually active women in low-and middle-income countries: who is lagging behind? Reprod Health. 2018;15(1):1.
Creanga AA, Gillespie D, Karklins S, Tsui AO. Low use of contraception among poor women in Africa: an equity issue. Bulletin of the world Health Organization. 2011;89:258-66.
Bakibinga P, Matanda DJ, Ayiko R, Rujumba J, Muiruri C, Amendah D, Atela M. Pregnancy history and current use of contraception among women of reproductive age in Burundi, Kenya, Rwanda, Tanzania and Uganda: analysis of demographic and health survey data. BMJ open. 2016 Mar 1;6(3):e009991.
Gillespie D, Ahmed S, Tsui A, Radloff S. Unwanted fertility among the poor: an inequity?. Bulletin of the World Health Organization. 2007 Feb;85(2):100-7.
Adanikin AI. Sociocultural barriers to family planning in the high fertility context of Nigeria (Doctoral dissertation, University of Southampton).
Ochako R, Mbondo M, Aloo S, Kaimenyi S, Thompson R, Temmerman M, Kays M. Barriers to modern contraceptive methods uptake among young women in Kenya: a qualitative study. BMC public health. 2015 Dec;15(1):1-9.
Sserwanja Q, Musaba MW, Mukunya D. Prevalence and factors associated with modern contraceptives utilization among female adolescents in Uganda. BMC women's health. 2021 Dec;21:1-7.
UBOS I. Uganda Bureau of statistics (UBOS) and ICF. 2018. Uganda demographic and health survey. 2016.
PMA2020, Performance Monitoring and Accountability 2020 (PMA2020) Phase 1 baseline survey, September-October. 2020, Makerere University, School of Public Health at the College of Health Sciences and The Bill & Melinda Gates Institute for Population and Reproductive Health at The Johns Hopkins Bloomberg School of Public Health. Performance Monitoring and Accountability 2020 (PMA2020). Uganda and Baltimore,Maryland, US. https://datalab.pmadata.org/dataset/doi%3A1034976h971-e365
Hellwig F, Coll CV, Ewerling F, Barros AJ. Time trends in demand for family planning satisfied: analysis of 73 countries using national health surveys over a 24-year period. Journal of Global Health. 2019 Dec;9(2).
Osborn JA, Sriram R, Karthikeyan S, Ravishankar SL. A study on contraceptive prevalence rate and factors influencing it in a rural area of Coimbatore, South India. Journal of Family Medicine and Primary Care. 2021 Jun;10(6):2246.
Eliason S, Awoonor-Williams JK, Eliason C, Novignon J, Nonvignon J, Aikins M. Determinants of modern family planning use among women of reproductive age in the Nkwanta district of Ghana: a case–control study. Reproductive health. 2014 Dec;11:1-0.
Tracker, F.-D., Reducing high fertility rates and Improving SExual reproductive health outcomes in Uganda (RISE), in Addendum to business case including for bridge funding and scaleup cost extensions. 2023.
Marmot M, Bell R. The sustainable development goals and health equity. Epidemiology. 2018 Jan 1;29(1):5-7.
Marmot M. Social determinants of health inequalities. Lancet. 2005 Mar 19-25;365(9464):1099-104. doi: 10.1016/S0140-6736(05)71146-6. PMID: 15781105.
Wakefield M, Williams DR, Le Menestrel S. The future of nursing 2020-2030: Charting a path to achieve health equity. National Academy of Sciences; 2021.
Lanteigne va, rider tr, stratton pa. Advancing health equity through inclusive design.
Jackson WA. Distributive justice with and without culture. Journal of Cultural Economy. 2015 Nov 2;8(6):673-88.
Poland B, Coburn D, Robertson A, Eakin J, Critical Social Science Group. Wealth, equity and health care: a critique of a “population health” perspective on the determinants of health. Social science & medicine. 1998 Apr 1;46(7):785-98.
Freedman R. The contribution of social science research to population policy and family planning program effectiveness. Studies in family planning. 1987 Mar 1;18(2):57-82.
Braveman PA. Monitoring equity in health and healthcare: a conceptual framework. Journal of health, population and nutrition. 2003 Sep 1:181-92.
Newman L, Baum F, Javanparast S, O'Rourke K, Carlon L. Addressing social determinants of health inequities through settings: a rapid review. Health promotion international. 2015 Sep 1;30(suppl_2):ii126-43.
Frenz P, Vega J. Universal health coverage with equity: what we know, don’t know and need to know. In Background paper for the First Global Symposium on Health Systems Research 2010 Nov 16 (Vol. 2010).
Cutler DM, Lleras-Muney A. Education and Health: Evaluating Theories and Evidence. National Poverty Center Working Paper Series# 06-19. National Poverty Center, University of Michigan. 2006 Jun.Ross CE, Wu CL. The links between education and health. American sociological review. 1995 Oct 1:719-45.
Logan RA, Wong WF, Villaire M, Daus G, Parnell TA, Willis E, Paasche-Orlow MK. Health literacy: A necessary element for achieving health equity. NAM perspectives. 2015 Jul 24.
Azzopardi-Muscat N, Sørensen K. Towards an equitable digital public health era: promoting equity through a health literacy perspective. European journal of public health. 2019 Oct 1;29(Supplement_3):13-7.
Liddelow C, Mullan B, Boyes M. Adherence to the oral contraceptive pill: the roles of health literacy and knowledge. Health Psychology and Behavioral Medicine. 2020 Jan 1;8(1):587-600.
Feinstein L, Sabates R, Anderson TM, Sorhaindo A, Hammond C. What are the effects of education on health. InMeasuring the effects of education on health and civic engagement: Proceedings of the Copenhagen symposium 2006 Mar 23 (pp. 171-354). Paris, France: Organisation for Economic Co-operation and Development.
Whitley J, Smith JD, Vaillancourt T, Neufeld J. Promoting mental health literacy among educators: A critical aspect of school-based prevention and intervention. Handbook of school-based mental health promotion: An evidence-informed framework for implementation. 2018:143-65.
Parry D, Salsberg J, Macaulay AC, Fcpc C. A guide to researcher and knowledge-user collaboration in health research. Ottawa: Canadian Institutes of Health Research. 2009 Oct.
Lathrop B. Moving toward health equity by addressing social determinants of health. Nursing for Women's Health. 2020 Feb 1;24(1):36-44.
Brennan Ramirez LK, Baker EA, Metzler M. Promoting health equity; a resource to help communities address social determinants of health.
Mensah F, Okyere J, Azure SA, Budu E, Ameyaw EK, Seidu AA, Ahinkorah BO. Age, geographical and socio-economic related inequalities in contraceptive prevalence: evidence from the 1993-2014 Ghana Demographic and Health Surveys. Contracept Reprod Med. 2023 Feb 7;8(1):20. doi: 10.1186/s40834-022-00194-9. PMID: 36750918; PMCID: PMC9903545
Mutua, M.K., Wado, Y.D., Malata, M. et al. Wealth-related inequalities in demand for family planning satisfied among married and unmarried adolescent girls and young women in sub-Saharan Africa. Reprod Health 18 (Suppl 1), 116 (2021). https://doi.org/10.1186/s12978-021-01076-0)

Modern methods include oral contraceptive pills, implants, injectables, contraceptive patch, vaginal ring, intrauterine device, female and male condoms, female and male sterilization, vaginal barrier methods (including the diaphragm, cervical cap and spermicidal agents), lactational amenorrhea method, emergency contraception pills, standard days method, basal body temperature method, Two-Day method and sympto-thermal method.
For binary variables (𝑏ℎ−𝑎ℎ) equals one.

No competing interests reported.

The Effect of a Sexual and Reproductive Health Programme on Socio-Economic and Education-related Inequalities in the Use of Modern Contraceptives in Seven Sub-regions in Uganda: A Case of RISE Programme 2019-2023

Status:

Version 1

Abstract

Figures

Introduction

METHODS

Data collection

Measurements

Construction of analytical weights

Analysis of Wealth and education-related equality in the use of modern conceptive

Concentration curve

Concentration index

RESULTS

DISCUSSION

Strengths and Limitations

Conclusion

Declarations

References

Footnotes

Additional Declarations

Status:

Version 1