Examining factors associated with post-abortion family planning among women in refugee settings in Uganda

Abstract

Introduction:

Post-abortion family planning (PAFP) is an essential component of post-abortion care (PAC) and a recommended family planning high-impact practice. Although PAFP uptake remains low, determinants of its uptake, especially in humanitarian settings, are not well understood.

Objective:

To identify the factors associated with PAFP uptake and examine whether women’s perceptions of person-centered PAC quality of care (QoC) were associated with PAFP uptake among women receiving PAC in refugee settings in Uganda.

Design:

We administered a cross-sectional prospective morbidity survey (PMS), part of a larger study using the abortion incidence complications methodology to estimate abortion incidence in all 13 refugee settlements in Uganda.

Methods:

We analyzed data from a PMS of 341 women with post-abortion complications who received PAC in the 13 refugee settlements. PAFP uptake was measured by asking patients if they left the health facility with a family planning method (Yes/No) after completing PAC. The quality of PAC was assessed using a modified 19-item Person-Centered Abortion Care scale. We used logistic regression to identify factors independently associated with PAFP uptake among women receiving PAC in refugee settings in Uganda.

Results:

PAC patients in refugee settings in Uganda reported medium to high levels of QoC, with deficits mainly observed in the communication and autonomy subdomain. Women’s own perception of QoC was not significantly associated with whether a PAC patient leaves with a method across models. Earning money in the past month before the survey (adjusted odds ratio (aOR) = 3.215; 95% confidence interval (CI) 1.124–9.199) and having a desire to leave with a family planning method (aOR = 270.532; 95% CI 78.921–927.348) were associated with PAFP, after controlling for women’s, facility level, and provider characteristics.

Conclusion:

Our findings highlight persistently low uptake of PAFP in refugee settings in Uganda, underscoring the need for targeted interventions that both increase family planning use and uphold reproductive autonomy. There is a need to strengthen family planning counseling as part of PAC provision and improve the quality of PAC services with particular attention to communication and patient autonomy.

Plain Language Summary

Proactively offering voluntary post-abortion family planning (PAFP) counselling and services at the same time and location where postabortion care is provided is a recommended practice to reduce unintended pregnancies and unsafe abortions. However, adoption of PAFP is still low, and the factors influencing its acceptance, particularly in humanitarian contexts, are not fully understood. This study examined factors associated with the uptake of PAFP among women receiving post-abortion care (PAC) in refugee settings in Uganda and explored whether the woman’s perceived quality of PAC influenced uptake of PAFP. Patients receiving PAC in refugee settings in Uganda reported moderate to high quality of care, with notable shortcomings primarily in the areas of communication and autonomy. The way women perceived their quality of care did not have a significant impact on whether she left with a family planning method. Factors such as having earned money in the last month and desire to leave with a method were strongly associated with PAFP uptake. Our results emphasize the low adoption of PAFP in refugee settings in Uganda, indicating the need for focused strategies that enhance family planning use while respecting reproductive autonomy. It is important to reinforce family planning counseling as an integral component of PAC services and to enhance the quality of PAC services, paying special attention to communication and patient autonomy.

Keywords

post-abortion care post-abortion family planning refugee settings Uganda

Introduction

The World Health Organization estimates that, worldwide, about 73 million induced abortions take place each year, with most (61%) being a result of an unintended pregnancy.¹ In low- and middle-income countries (LMICs), it is estimated that about 7 million people are admitted to hospitals each year due to complications related to unsafe abortion.² Moreover, given the restrictive abortion laws in some countries, a significant proportion of the induced abortions in LMICs are unsafe, that is, conducted using non-recommended methods by untrained providers, with inadequate equipment or knowledge.³ Unsafe abortions exacerbate poor health and economic outcomes, including increased maternal morbidity or deaths and increased costs of health care.^4,5 As such, the provision of high-quality post-abortion care is an essential service for reducing maternal morbidity and mortality, especially in contexts where abortion is legally restricted.

In order to reduce unwanted pregnancies, maternal morbidity, and mortality, post-abortion family planning (PAFP) is still a crucial part of post-abortion care (PAC). Provision of PAFP counseling and methods is an integral component of quality comprehensive abortion and PAC.^1,6 Non-use of contraception is the main cause of unintended pregnancies, which often end in induced abortions.^7,8 As such, family planning counseling and contraceptive provision to individuals receiving PAC can reduce future unintended pregnancies and unsafe abortions.^9–11 However, research shows that healthcare workers can fail to provide high-quality counseling to PAC patients due to limited time, space, and resources.¹² PAFP uptake can also be hampered by patient-related barriers, including perceptions about potential negative effects of contraception or lack of partner support, as well as health system challenges such as distance to care.¹³

Notwithstanding the patient-related barriers associated with PAFP, the need for access to comprehensive abortion care may be more pronounced in humanitarian settings due to the fragility of such contexts. People living in humanitarian settings often face increased exposure to sexual violence and transactional sex, difficulties navigating health systems, and disrupted access to contraception and other sexual and reproductive health services.^14–16 While PAFP is a key component of the Minimal Initial Service Package for humanitarian settings, these services tend to be limited in such contexts due to lack of knowledge or skills among health providers,^15,17 and insufficient funding.¹⁸

Evidence suggests that quality of care (QoC) may be one avenue for improving the uptake of PAFP among individuals who do not desire a pregnancy. Literature shows a positive association between QoC and increased contraceptive uptake and continuation.^19–21 For example, studies in India and Western Kenya found that family planning counseling quality was associated with uptake of family planning methods.^20,21 Despite existing evidence, there is limited understanding of the status of person-centered quality of PAC service and its relationship with PAFP uptake in humanitarian settings. Drawing on validated scales that measure person-centered quality of PAC,²² this study examines the association between person-centered QoC and PAFP uptake in refugee settings in Uganda.

Study context

Uganda is home to 1.7 million refugees (as of October 2024)—the largest number of refugees in Africa.²³ Contraceptive use remains low in Uganda; recent estimate shows a contraceptive prevalence rate of 33% among women aged 15–49 years nationally.²⁴ Among refugees, only 4% of women aged 15–49 years were using any method of contraception.²⁴ Unintended pregnancies—which account for more than half of all pregnancies in Uganda²⁵—remain a major cause of induced abortions.²⁶ Yet, abortion remains highly restricted, except in situations when the woman’s life is at risk, the woman is HIV-positive, or in cases of rape, incest, or fetal anomaly.^27–29 This leads individuals seeking abortion to engage in clandestine induced abortion, often by unskilled health personnel and/or using unsafe methods.³⁰ Recent estimates indicate that the induced abortion incidence rate among women living in refugee settlements in Uganda is 37.3 per 1000 women aged 15–49.³¹ Studies show that many refugees in Uganda face legal barriers and lack information, leading to a high level of unsafe abortions in refugee settlements.^32,33

Data and methods

Study design and data collection

This analysis uses data from a cross-sectional application of a prospective morbidity survey (PMS). This survey was administered as part of a larger study using the abortion incidence complications methodology (AICM) to estimate abortion incidence in all 13 refugee settlements in Uganda. The AICM is an indirect method widely used to estimate abortion incidence, particularly in settings where abortion is legally restricted and difficult to measure. The AICM traditionally consists of a nationally representative survey of health facilities and a survey of knowledgeable informants. Often, the AICM also applies a survey of post-abortion care patients (the PMS) to understand post-abortion complications and to inform understanding of topics critical to women’s sexual and reproductive health.³⁴ The PMS consists of a patient survey, which collects patient demographic data, reproductive health history, and information on delays to care and experiences of care, as well as a provider survey, which collects clinical information about the abortion-related complications and care provided to the patient.³⁵

Details on the methods of the larger AICM are published elsewhere.³¹ Briefly, we included a census of all health facilities located within and near the 13 refugee settlements in Uganda, provide PAC, and were willing to be included in the study (n = 102). As our focus was on refugee settings, we included both facilities within and outside of refugee settlements due to the fact that refugees and host community members often cross settlement boundaries to seek care; this includes refugees seeking care outside of settlements, and host community members seeking care within settlements. Given that the PMS was administered by trained clinical staff situated at the health facilities where they work, health facilities that were unable to identify clinical staff for training to conduct the PMS were excluded (n = 37), and the final PMS sample consisted of 65 health facilities.

Data collection for the PMS lasted 35 days in each health facility during April–May 2023. All women (we recognize that not all individuals who need post-abortion care identify as women. Due to the highly stigmatized and criminalized nature of transgenderism and gender fluidity in Uganda, few gender non-conforming people are able to publicly express their gender identity, making it impossible for us to know whether any of the patients in our sample identified as anything other than a cis-woman. As such, in this article, we refer to all patients in our sample as “women.”) who presented at a study facility for PAC during the data collection period and whose gestational age was 26 weeks or less were eligible to participate in the study, regardless of refugee status. Women who presented with threatened abortion (defined as a viable pregnancy, despite vaginal bleeding), an intrauterine fetal death, molar pregnancy, or ectopic pregnancy did not receive PAC services and were thereby not eligible for participation. Data collectors were responsible for recording the total number of patients who presented at the facility seeking PAC services during the study period using daily tally sheets, regardless of whether the patient was approached to participate in the study or ultimately provided consent. These tallies were used to produce individual-level survey weights to account for nonresponse, which were multiplied by facility-level weights calculated to account for the probability of inclusion in the PMS sample.

A total of 404 women accessed PAC services from sampled facilities during the data collection period, of which 349 (86.4%) completed a patient survey. Eight interviews were administered with ineligible clients who were either over the gestational age limit (n = 1) or were not a PAC case (n = 7); these cases were dropped from analysis, yielding a total analytic sample of 341 PMS interviews.

Measurement of variables

The main dependent variable for this analysis is receipt of family planning services—a proxy for PAFP uptake. PAC patients were asked whether they received family planning counseling and whether or not they left the facility with a family planning method. The responses to these questions were “Yes” or “No.”

The main independent variable is the quality of post-abortion care. We utilized the Person-Centered Abortion Care (PCAC) scale, a 24-item scale generated and validated in Kenya by Sudhinaraset et al.²² Because this scale was originally designed to measure quality of safe abortion care, we modified the scale to exclude five items not relevant or appropriate to the PAC service. Our modified scale had 19 items, 10 of which comprise the respectful and supportive care sub-scale, and 9 of which comprise the communication and autonomy sub-scale. PAC patients were asked a series of questions about the care they received at the facility, with response options consisting of a 4-point Likert scale ranging from no, never to yes, all of the time, with the exception of two questions. Length of wait time ranged from very long to very short, and whether facility staff introduced themselves ranged from no, none of them to yes, all of them. Some items were reverse-coded such that higher scores reflected higher QoC. Scale items were summed into a composite PCAC score, with a theoretical range of 0–57. For analysis, the composite PCAC score and sub-scores were recoded such that scores in the 25th percentile were considered “low,” those in the 75th percentile were considered “high” and all those in between were considered “average.”

We considered various demographic characteristics of the PAC patients. These included age (15–24, 25–34, and 35 years or more), current marital status (not currently married or currently married), highest education level (no education, primary, or secondary and above), region where the settlement is located (North and West), use of a family planning method at the time of getting pregnant (yes or no), gestational age of pregnancy (first or second trimester), and desire to leave with a family planning method (yes or no).

We also measure several facility and provider-level characteristics, including the ownership of the health facility (government, non-governmental organization (NGO), or United Nations High Commissioner for Refugees (UNHCR)), health facility level (health centers II, III, or IV), the main provider cadre (doctor, nurse, midwife, physician assistant or clinical officer), abortion method (surgical or medical), and whether the main provider for PAC care was the interviewer (no or yes).

Statistical analysis

Data were analyzed using Stata version 15.³⁶ First, we ran descriptive statistics to assess the characteristics of PAC patients. We calculated mean scores for each item in the PCAC scale as well as composite scores for the entire scale and its subdomains. We also assessed the proportion of PAC patients for each PCAC item that had responded with the highest possible quality score. Finally, we estimated three logistic regression models using weighted data to assess relationships with our outcome measure (receipt of a PAFP method): (1) the first model showing the bivariate relationships between selected variables and leaving with a PAFP method; (2) the second model exploring relationships between QoC and receipt of PAFP after controlling for women-level factors; and (3) the third model exploring relationships between QoC and receipt of PAFP after controlling for facility-level and service factors. All variables significant in bivariate analyses were included in the final multivariable model (model 3). Previous research⁶ has observed that post-abortion contraceptive uptake can be influenced by different factors including facility, service, provider, and women factors—calling for the need to understand each of these categories of variables independently to avoid confounding. Only variables that were significant (p < 0.05) at bivariate level were included in the second and third models, with the exception of our main independent variable (QoC), which was maintained in all models. The results are presented as odds ratios (OR) with 95% confidence intervals.

Ethical considerations

We obtained ethical approval from the Guttmacher Institutional Review Board and Mildmay Uganda Research and Ethics Committee (MUREC; Approval Number REC REF 0411-2022). We further obtained regulatory approval from the Uganda National Council for Science and Technology (UNCST; Approval No. REF SS1589ES). As per the approved protocol by MUREC, all women aged 15–49 years provided verbal informed consent to participate in the survey. Verbal consent was electronically documented via signature by interviewers. Verbal consent was chosen because it minimizes the risk of revealing a patients’ identity and compromising her confidentiality, in the unlikely event that informed consent documents with patients’ names and signatures are misplaced in the course of fieldwork, and it is also suitable for people who may not be able to write. Patients younger than 15 were not eligible for the patient survey. For patients aged 15–17 years who received PAC, parental/guardian permission for inclusion in this research was not a reasonable requirement since it would compromise these respondents’ confidentiality and possibly their safety.

Results

Just under half (45%) of our sample were 15–24 years of age, nearly all (92%) were married at the time of the survey, and most (63%) had primary-level education (Table 1). About half (54%) had earned some money in the last month prior to the survey. Two-thirds (68%) reported that their current pregnancy was intended. A quarter (25%) did not receive family planning counseling when they obtained PAC services. A majority (83%) were not using a method prior to conceiving the pregnancy that resulted in abortion, and more than half (59%) had no desire to leave with a method after PAC. About a third of respondents (36%) left the health facility with a contraceptive method. Nearly half of respondents (49%) who did not leave with a contraceptive method needed to first consult their partners.

Table 1.

Sociodemographic characteristics of respondents (unweighted N = 341).

Socioeconomic factors	Unweighted, N	Weighted % [95% CI]
Age
15–24	152	45.4 [37.8–53.3]
25–34	109	30.8 [24.6–37.7]
35+	80	23.8 [18.9–29.5]
Current marital status (n = 340)
Not currently married	28	7.7 [4.9–12.0]
Currently married	312	92.3 [88.0–95.1]
Highest education level
No education	73	21.2 [15.1–28.9]
Primary	213	62.8 [53.1–71.6]
Secondary and above	55	16.0 [11.4–22.0]
Earned money in the past month (n = 340)
No	161	45.8 [37.7–54.0]
Yes	179	54.2 [46.0–62.3]
Region
North	222	61.4 [43.1–77.0]
West	119	38.6 [23.0–56.9]
Received family planning counseling (n = 340)
No	76	24.9 [15.2–38.2]
Yes	264	75.1 [61.8–84.8]
Desire to leave with method (n = 337)
No	197	59.1 [49.4–68.2]
Yes	140	40.9 [31.8–50.6]
Used method at time of getting pregnant
No	282	83.1 [77.4–87.7]
Yes	59	16.9 [12.3–22.6]
Pregnancy intended
No	116	32.4 [26.3–39.3]
Yes	225	67.6 [60.7–73.7]
Left with a method
No	211	63.8 [54.7–71.9]
Yes	126	36.2 [28.1–45.3]
Reasons for not leaving with a method^a
Need to consult husband	104	48.6 [39.2–58.1]
Want to get pregnant	42	21.0 [14.7–29.0]
Method not available	3	1.7 [0.4–8.1]
Oppose use of family planning	21	9.2 [5.3–15.7]
Worried about side effects	50	24.6 [17.9–32.9]

Where responses are missing, denominator reported. CI: confidence interval.

Among people not leaving with a method; as a select-multiple question, responses will not sum to 211.

Table 2 presents the composite scores for the PCAC main scale and sub-scales. Table 2 also presents descriptive statistics for each individual QoC scale variable (mean score and proportion of PAC patients reporting the highest possible quality score). Out of a total possible score of 57, the mean overall PCAC score in our sample was 43.4. When divided into low, average, and high scores, 28.8% of our sample received low-quality care, 45.8% average, and 25.4% high (data not shown). Whereas receiving respectful and supportive care is expected to be universal for all PAC patients, not all PAC patients received such care. The mean score for the 10 items considered under respectful and supportive care was 25.1, out of a total possible score of 30. Only about half of the respondents said that at all times they: were treated with respect (54%), had privacy (55%) or confidentiality (57%), trusted the healthcare staff (59%), or were given medication to manage the pain (58%). Encouragingly, nearly all PAC patients in our sample said they were never verbally or physically abused (90% and 94%, respectively). QoC for communication and autonomy was lower than that of respectful care, with a mean score of 18.3 out of a possible 27. Few PAC patients reported that healthcare staff always: introduced themselves the first time they sought care (6%), called them by their name (31%), asked for consent before any procedures were performed (48%), or asked about their pain (48%). Fewer than half of PAC patients reported always feeling they were able to ask questions (33%), and only about half could always understand the language used by the provider (54%), received explanations for procedures or medication (53%), or were asked by the providers about how they felt (50%).

Table 2.

Quality of post-abortion care.

Quality of post-abortion care scale	Mean [95% CI]	% highest score (weighted)
Overall PCAC score	43.384 [42.545–44.223]
Subdomain: Respectful and supportive care (10 items)	25.091 [24.684–25.497]
Short time to care	2.475 [2.390–2.560]	63.5
Treated with respect	2.403 [2.321–2.484]	54.4
Not treated differently	2.564 [2.464–2.663]	79.4
Privacy	2.083 [1.956–2.210]	55.4
Record confidentiality	2.367 [2.274–2.461]	56.6
Pain medication given	2.416 [2.335–2.496]	57.6
Non-verbal abuse	2.803 [2.733–2.872]	90.1
Non-physical abuse	2.856 [2.793–2.919]	93.9
Not asked to pay	2.888 [2.833–2.944]	95.6
Trust	2.426 [2.337–2.515]	59.4
Subdomain: Communication and autonomy (9 items)	18.293 [17.731–18.855]
Introduce self	0.906 [0.809–1.002]	6.0
Called by name	1.858 [1.753–1.963]	31.4
Consent before procedures	2.192 [2.091–2.292]	47.6
Understand language	2.333 [2.240–2.426]	54.1
Explanation for procedures	2.316 [2.223–2.408]	52.5
Explanation for medication	2.324 [2.229–2.420]	54.1
Ask about feeling	2.305 [2.216–2.393]	50.0
Ask questions	1.917 [1.813–2.021]	33.2
Ask about pain	2.256 [2.167–2.344]	47.5

PCAC: person-centered abortion care; CI: confidence interval.

Table 3 shows results from logistic regression models assessing factors associated with leaving a health facility with a family planning method. Model 1 (results from a bivariate model) shows that the unadjusted odds of leaving with a family planning method was significantly higher among women with primary education when compared to women with no education (unadjusted odds ratio (uOR) = 2.961; 95% confidence interval (CI) 1.548–5.663), women who earned money in the past month prior to the survey (uOR = 1.642; 95% CI 1.025–2.629), if the interviewer was the main provider for PAC (uOR = 1.714; 95% CI 1.016–2.893], and among women who received family planning counseling (uOR = 2.197; 95% CI 1.147–4.207). Desire to leave with a family planning method was highly correlated with our outcome measure (95% CI 44.246–530.281).

Table 3.

Factors associated with leaving with a family planning method.

Factors	Model 1: Bivariate (uOR 95% CI)	Model 2: Individual factors (aOR 95% CI)	Model 3: Individual-level factors, facility, service, and provider-level factors (aOR 95% CI
Quality of care (RC = low)
Average	1.373 [0.695–2.790]	2.198 [0.837–5.768]	2.481 [0.889–6.917]
High	1.107 [0.465–2.637]	0.800 [0.235–2.727]	0.789 [0.234–2.662]
Individual-level factors
Age (RC = 15–24)
25–34	0.833 [0.446–1.554]
35+	0.771 [0.489–1.217]
Current marital status (RC = not currently married)
Currently married	0.944 [0.406–2.198]
Highest education level (RC = no education)
Primary	2.961 [1.548–5.663]*	1.024 [0.419–2.498]	1.141 [0.421–3.089]
Secondary and above	1.955 [0.873–4.380]	0.823 [0.237–2.862]	1.116 [0.267–4.666]
Earned money in the past month (RC = no)
Yes	1.642 [1.025–2.629]*	3.224 [1.162–8.947]*	3.215 [1.124–9.199]*
Region (RC = North)
West	0.797 [0.385–1.649]
Used method at the time of getting pregnant (RC = no)
Yes	1.965 [0.856–4.514]
Pregnancy intended (RC = no)
Yes	0.634 [0.373–1.077]
Gestation age of pregnancy (RC = first trimester)
Second trimester	0.919 [0.416–2.033]
Living in settlement (RC = no)
Yes	0.728 [0.393–1.348]
Desire to leave with a method (RC = no)
Yes	153.176 [44.246–530.281]*	211.417 [62.770–712.075]*	270.532 [78.921–927.348]*
Facility level and service factors
Ownership of health facility (RC = government)
NGO/UNHCR	0.463 [0.151–1.424]
Facility level (RC = health center II)
Health center III	2.083 [0.939–4.620]
Health center IV	1.281 [0.373–4.405]
Received family planning counseling (RC = no)
Yes	2.197 [1.147–4.207]*		0.436 [0.156–1.223]
Provider factors
Main provider for surgical abortion (RC = Doctor)
Nurse/midwife	1.089 [0.270–4.390]
Other^a	0.326 [0.036–2.941]
Interviewer is main provider for PAC care (RC = no)
Yes	1.714 [1.016–2.893]*		1.172 [0.488–2.814]
Abortion method (RC = surgical)
Medical	1.007 [0.611–1.660]		1.696 [0.586–4.910]
No evacuation procedure	0.340 [0.172–0.673*		0.293 [0.077–1.223]

uOR: unadjusted odds ratios; aOR: adjusted odds ratios; PAC: post-abortion care; NGO: non-governmental organization; UNHCR: United Nations High Commissioner for Refugees; CI: confidence interval; RC: Reference Category.

Other includes Physician Assistant, Clinical Officer.

Significant at p < 0.05.

After adjusting for other demographic characteristics (model 2), earning money in the past month before the survey (adjusted odds ratio (aOR) = 3.224; 95% CI 1.162–8.947) remained significantly associated with the likelihood of leaving the health facility with a family planning method. The addition of provider-level factors (model 3) did not substantively change the association between earning money and leaving with a method (aOR = 3.215; 95% CI 1.124–9.199). Desire to leave with a method remained highly correlated with leaving with a family planning method in both adjusted models (model 2: aOR = 211.417; 95% CI 62.770–712.075; model 3: aOR = 270.532; 95% CI 78.921–927.348). QoC, as reported by PAC patients, was not significantly associated with having left the health facility with a family planning method in any of the models.

Discussion

This research broadens the current body of knowledge regarding the relationship between PAC QoC and PAFP uptake among women in refugee settings. Understanding the association between PAC QoC and PAFP in refugee settings offers insights into the elements that affect PAFP uptake, which is crucial for guiding focused interventions and strategies designed to enhance reproductive health outcomes in such contexts.

Our findings show that despite family planning counseling being an integral part of PAC, about a quarter of PAC clients did not receive such counseling. These findings are in line with what has been observed elsewhere in non-humanitarian settings in Uganda, where a significant proportion of PAC clients did not receive family planning counseling.^13,37 Previous research has observed that lack of physical space at the health facility and limited time for counseling hamper counseling sessions to take place.³⁷ These findings suggest a need for strengthening family planning counseling as part of PAC in humanitarian settings. While some PAC patients may be accessing the service due to miscarrying an intended pregnancy, non-coercive family planning counseling should be provided to all PAC patients, regardless of pregnancy intention.

Provision of PAC varied, from medium to high levels of patient-reported quality of PAC care. While respectful and supportive care was rated highly among our sample, communication and autonomy domain was rated less highly. These findings corroborate what has been observed elsewhere in Kenya,³⁸ Ethiopia,³⁹ Central African Republic, and Nigeria⁴⁰—with subdomains on respectful and supportive care receiving higher scores than communication and autonomy. While failure to provide comprehensive PAC care may be due to the emergency nature of some PAC procedures, it is nonetheless concerning that PAC patients felt largely uninformed about their care. Previous research has shown that one reason for low-quality care could be that PAC patients may be receiving services from unskilled healthcare providers who may not be well prepared to provide the required minimum standards of care^3,30 or women may not have enough information regarding minimum standards they should receive.³³ While we did not assess the skill of healthcare providers or patient knowledge about care standards, these may also be barriers to ensuring high-quality PAC care in the refugee setting.

Our analysis found that women’s perception of the quality of PAC care they received was not significantly associated with PAFP uptake. This may be due to the fact that desire for a method is established at the outset by personal factors not easily influenced by the quality of their abortion service. Our findings conflict with studies in other settings that do find an association between QoC and PAFP uptake; however, this may be because most previous work linking QoC to family planning uptake has focused specifically on the quality of family planning counseling and not followed the person-centered approach.^20,21 Our analyses revealed that other factors were more consistently associated with leaving with a family planning method. Notably, PAC patients who had earned money had significantly higher rates of leaving with a method compared to women who had not. Previous research echoes this finding and has demonstrated that socioeconomic factors in the form of employment have a large role in influencing family planning use, with employment increasing one’s economic status to be able to afford to obtain family planning services.^41,42

The factor most strongly associated with receipt of a PAFP method was desire to leave with a method. In many ways, this is an encouraging finding, as it suggests that women who desire a family planning method are likely to obtain one. That said, this result should be interpreted with caution in this study, as it is possible that the strong correlation was also influenced by the way the questionnaire was administered. Women participated in the survey at the end of their visit after they had received family planning counseling. As such, it is possible that some women’s desires were shaped by the receipt of this counseling, and measuring women’s desires prior to counseling may have resulted in different results. Future research is needed to better understand what drives desire for contraception. In particular, it is important to understand how much of this relationship is driven by lack of information or misconceptions around PAFP, and how much reflects women’s own reproductive autonomy.

Our findings that the main reason women did not leave with a family planning method was due to personal reasons (e.g., having to consult the partner, fear of side effects, wanting to get pregnant), highlights the importance of dispelling misconceptions about contraception that may lead women to not obtain a contraceptive method, even if they do not desire pregnancy. Future studies can assess the possibility of increasing women’s desire to use a method while also protecting their reproductive autonomy. It is worth noting, however, that the correlation between the desire to leave with a method and leaving with a method might be due to the way the questionnaire was administered. We asked these questions at the end of the visit, after women have received counseling. Women who left with a method may have been more likely to say they desired to leave with a method in hindsight. It is possible that the desire to leave with a method might have looked different if we asked women this question before they received counseling.

Limitations

This study has a few limitations. First, the cross-sectional nature of the data does not allow us to infer causality. Second, leaving with a family planning method may not necessarily mean use (especially for those methods that are not administered at the facility) or consistent use of a method to prevent future unintended pregnancy that may result in unsafe abortion. Our data do not allow us to assess how many women who received a method actually went on to use the method. This study also assesses family planning uptake among PAC patients, some of whom may desire pregnancy and are therefore not ideal candidates for PAFP uptake. Lastly, measuring the quality of abortion care is notoriously difficult. While we use a validated PCAC scale, studies hypothesize that abortion clients often rate services highly because they are glad to have received care.⁴³ Moreover, the PCAC scale has never been tested and validated among refugee population. This fact may bias results to higher quality. In some cases, the PMS was administered by the patient’s provider, potentially biasing these results even further. However, we did not find any significant relationship between leaving with a family planning method and whether the interviewer was the main provider for PAC.

Conclusion and implications

Our findings show that whereas high quality of PAC care is expected to be universal, not all patients accessing PAC in refugee settings in Uganda received such care. Women’s own perception of QoC was not significantly associated with whether a PAC patient leaves with a contraceptive method among our sample. Leaving with a family planning method was associated with whether or not she desires one. This points to more research being needed on what determines women’s desire for a PAFP method. In particular, it is important to understand how much of this relationship is driven by lack of information or misconceptions around PAFP, and how much reflects women’s own reproductive autonomy. Future studies can assess the possibility of increasing women’s desire to use a method while also protecting their reproductive autonomy. Findings underscore the need to ensure family planning counseling is part of PAC provision as well as improving the quality of PAC services, with particular attention to strengthening communication and patient autonomy.

Interventions that aim to promote contraceptive use can increase awareness and reliable information on a wide range of contraceptive methods to provide individuals with a variety of contraceptive methods at minimal cost or for free.^37,44–46 The success of such can depend on periodic trainings of providers of family planning providers to update their understanding on the different method of contraception.⁴⁶ Family planning programs can also take advantage of customized and/or couple family planning counseling to improve couple communication on family planning use, reproductive intentions.^46,47 Finally, family planning programs that aim to promote counseling should be cognizant of the different underlying social and gender norms that play out in influencing contraceptive use.⁴⁵

Footnotes

Acknowledgements

We are exceedingly grateful to the Government of Uganda, particularly through the Office of the Prime Minister, Department of Refugees, and the Ministry of Health, whose leadership and coordination enabled access and engagement with key stakeholders, including non-governmental organization partners and the United Nations High Commissioner for Refugees Uganda across the study regions. We extend our deepest appreciation to the District Health Officers from the 12 participating districts—Adjumani, Yumbe, Terego, Kiryandongo, Kyegegwa, Kikuube, Koboko, Isingiro, Lamwo, Moyo, Madi-Okollo, and Kamwenge—for their guidance, facilitation, and ongoing commitment to improving health outcomes within their communities. Our sincere thanks go to the facility in-charges and administrative teams of all participating health facilities for their openness, cooperation, and logistical support throughout the research process. We especially recognize the critical role played by the data collection team, including health providers, whose dedication, professionalism, and tireless efforts during data collection were essential to the successful execution of the study.

ORCID iD

Peter Kisaakye

Ethical Considerations

Ethical approval was obtained from the Guttmacher Institutional Review Board and Mildmay Uganda Research and Ethics Committee (MUREC; Approval Number REC REF 0411-2022). We further obtained regulatory approval from the Uganda National Council for Science and Technology (UNCST; Approval Number REF SS1589ES).

Consent to participate

As per the approved protocol by MUREC, all women aged 15–49 years provided verbal informed consent to participate in the survey. Verbal consent was electronically documented via signature by interviewers. Verbal consent was chosen because it minimizes the risk of revealing patients’ identity and compromising her confidentiality, in the unlikely event that informed consent documents with patients’ names and signatures are misplaced in the course of fieldwork, and it is also suitable for people who may not be able to write. Patients younger than 15 were not eligible for the patient survey. For patients aged 15–17 years who received PAC, parental/guardian permission for inclusion in this research was not a reasonable requirement since it would compromise these respondents’ confidentiality and possibly their safety. Patients younger than 15 were not eligible for the patient survey.

Consent for publication

Not applicable.

Author contributions

Peter Kisaakye: Conceptualization; Investigation; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision; Data curation.

Stephanie Kung: Conceptualization; Investigation; Funding acquisition; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Software; Formal analysis; Project administration; Data curation; Supervision; Resources.

Francis Obare: Conceptualization; Investigation; Funding acquisition; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision; Resources.

Caitlin Rich: Conceptualization; Investigation; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Software; Formal analysis; Data curation.

George Odwe: Conceptualization; Investigation; Funding acquisition; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision; Resources.

Yohannes Dibaba Wado: Conceptualization; Investigation; Funding acquisition; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision; Resources.

Bonnie Wandera: Conceptualization; Investigation; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision.

Dagim Habteyesus: Conceptualization; Investigation; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision.

Yadeta Dessie: Conceptualization; Investigation; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Formal analysis; Project administration; Supervision.

Margaret Giorgio: Conceptualization; Investigation; Funding acquisition; Writing – original draft; Methodology; Validation; Visualization; Writing – review & editing; Software; Formal analysis; Project administration; Data curation; Supervision; Resources.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Foreign, Commonwealth & Development Office (FCDO; Award Number: PO8612) and the Norwegian Agency for Development Cooperation (NORAD; Award Number: QZA-21/0135). The findings and conclusions in this study are those of the authors and do not necessarily reflect the positions and policies of the funding agencies.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The dataset used for the current study is available from the corresponding author on reasonable request.

References

World Health Organization. Abortion. WHO, 2024.

World Health Organization. Preventing unsafe abortion. WHO, 2019.

Grossman

Prata

Williams

, et al. Availability of medical abortion medicines in eight countries: a descriptive analysis of key findings and opportunities. Reprod Health 2023; 20: 58.

Okonofua

. Abortion and maternal mortality in the developing world. J Obstet Gynaecol Can 2006; 28: 974–979.

Babigumira

Stergachis

Veenstra

, et al. Estimating the costs of induced abortion in Uganda: a model-based analysis. BMC Public Health 2011; 11: 904.

Wado

Dijkerman

Fetters

. An examination of the characteristics and contraceptive acceptance of post-abortion clients in Ethiopia. Women Health 2021; 61: 133-147.

Åhman

Shah

Åhman

, et al. Contraceptive use, fertility, and unsafe abortion in developing countries. Eur J Contracept Reprod Health Care 2006; 11: 126–131.

Udho

Ekung

Namutebi

, et al. Person-centered maternity care and satisfaction with post-abortion care: a facility-based cross-sectional survey in Northern Uganda. Womens Health 2025; 21: 17455057251318897.

Ceylan

Ertem

Saka

, et al. Post abortion family planning counseling as a tool to increase contraception use. BMC Public Health 2009; 9: 20.

10.

Ferreira

ALCG

Souza

Lima

, et al. Choices on contraceptive methods in post-abortion family planning clinic in the northeast Brazil. Reprod Health 2010; 7: 5.

11.

Moges

Hailu

Dimtsu

, et al. Factors associated with uptake of post-abortion family planning in Shire town, Tigray, Ethiopia. BMC Res Notes 2018; 11: 928.

12.

Paul

Gemzell-Danielsson

Kiggundu

, et al. Barriers and facilitators in the provision of post-abortion care at district level in central Uganda—a qualitative study focusing on task sharing between physicians and midwives. BMC Health Serv Res 2014; 14: 28.

13.

Kayiga

Looft-Trägårdh

Cleeve

, et al. Healthcare providers’ perceptions on post abortion intrauterine contraception: a qualitative study in central Uganda. PLoS One 2024; 19: e0301748.

14.

Tran

Greer

Dah

, et al. Strengthening healthcare providers’ capacity for safe abortion and post-abortion care services in humanitarian settings: lessons learned from the clinical outreach refresher training model (S-CORT) in Uganda, Nigeria, and the Democratic Republic of Congo. Confl Health 2021; 15: 20.

15.

Casey

Chynoweth

Cornier

, et al. Progress and gaps in reproductive health services in three humanitarian settings: mixed-methods case studies. Confl Health 2015; 9: S3.

16.

Tirado

Chu

Hanson

, et al. Barriers and facilitators for the sexual and reproductive health and rights of young people in refugee contexts globally: a scoping review. PLoS One 2020; 15: e0236316.

17.

Rogers

Dantas

JAR

. Access to contraception and sexual and reproductive health information post-abortion: a systematic review of literature from low- and middle-income countries. J Fam Plann Reprod Health Care 2017; 43: 309.

18.

Tanabe

Schaus

Rastogi

, et al. Tracking humanitarian funding for reproductive health: a systematic analysis of health and protection proposals from 2002–2013. Confl Health 2015; 9: S2.

19.

Tessema

Streak Gomersall

Mahmood

, et al. Factors determining quality of care in family planning services in Africa: a systematic review of mixed evidence. PLoS One 2016; 11: e0165627.

20.

Dey

Averbach

Dixit

, et al. Measuring quality of family planning counselling and its effects on uptake of contraceptives in public health facilities in Uttar Pradesh, India: a cross-sectional analysis. PLoS One 2021; 16: e0239565.

21.

Wendot

Scott

Nafula

, et al. Evaluating the impact of a quality management intervention on post-abortion contraceptive uptake in private sector clinics in western Kenya: a pre- and post-intervention study. Reprod Health 2018; 15: 10.

22.

Sudhinaraset

Landrian

Afulani

, et al. Development and validation of a person-centered abortion scale: the experiences of care in private facilities in Kenya. BMC Womens Health 2020; 20: 208.

23.

UNHCR. Uganda, https://reliefweb.int/report/uganda/unhcr-uganda-fact-sheet-september-october-2024 (2024, accessed 12 March 2025).

24.

Uganda Bureau of Statistics. Uganda Demographic and Health Survey 2022. UBOS, 2023.

25.

Inzama

Kaye

Kayondo

, et al. Gaps in available published data on abortion in Uganda and the missed opportunity to inform policy and practice. Int J Gynecol Obstet 2023; 161: 1–7.

26.

Guttmacher Institute. Abortion in Uganda. Guttmacher Institute, 2013.

27.

Guttmacher Institute. Abortion and postabortion care in Uganda. Guttmacher Institute, 2017.

28.

Prada

Atuyambe

Blades

, et al. Incidence of induced abortion in Uganda, 2013: new estimates since 2003. PLoS One 2016; 11: e0165812.

29.

Kagaha

Manderson

. Power, policy and abortion care in Uganda. Health Policy Plan 2020; 36: 187–195.

30.

Cleeve

Faxelid

Nalwadda

, et al. Abortion as agentive action: reproductive agency among young women seeking post-abortion care in Uganda. Cult Health Sex 2017; 19: 1286–1300.

31.

Odwe

Kisaakye

Obare

, et al. Estimating incidence of induced abortion and unintended pregnancy among women in refugee settlements in Uganda. Stud Fam Plan. Epub ahead of print 23 April 2025.DOI: 10.1111/sifp.70010.

32.

Nara

Banura

Foster

. Exploring Congolese refugees’ experiences with abortion care in Uganda: a multi-methods qualitative study. Sex Reprod Health Matters 2019; 27: 1681091.

33.

Nabunje

Zuniga

. Abortion experiences of women and girls living in Bidibidi refugee settlement: implications for frontline health care workers, https://www.ibisreproductivehealth.org/sites/default/files/files/publications/r2hc_brief_for_providers.pdf (2024, accessed 12 March 2025).

34.

Singh

Juarez

Prada

, et al. Estimating abortion incidence: assessment of a widely used indirect method. Popul Res Policy Rev 2019; 38: 429–458.

35.

Fetters

. Prospective approach to measuring abortion-related morbidity: individual-level data on postabortion patients. In: Singh

Remez

Tartaglione

(eds) Methodologies for estimating abortion incidence and abortion-related morbidity: a review. Guttmacher Institute and International Union for the Scientific Study of Population, 2010.

36.

StataCorp. Stata Statistical Software: Release 15. StataCorp LLC, 2017.

37.

Atuhairwe

Hanson

Tumwesigye

, et al. Second trimester post-abortion family planning uptake and associated factors in 14 public health facilities in Central Uganda: a cross-sectional study. Contracept Reprod Med 2023; 8: 4.

38.

Sudhinaraset

Landrian

Montagu

, et al. Is there a difference in women’s experiences of care with medication vs. manual vacuum aspiration abortions? Determinants of person-centered care for abortion services. PLoS One 2019; 14: e0225333.

39.

Chekol

McCaffrey

Dijkerman

, et al. Person-centered abortion care in public health facilities across four regions of Ethiopia: a cross-sectional quantitative study of client experiences. Front Reprod Health 2024; 6: 1331682.

40.

Pasquier

Owolabi

Powell

, et al. Assessing post-abortion care using the WHO quality of care framework for maternal and newborn health: a cross-sectional study in two African hospitals in humanitarian settings. Reprod Health 2024; 21: 114.

41.

Ahmed

Creanga

Gillespie

, et al. Economic status, education and empowerment: implications for maternal health service utilization in developing countries. PLoS One 2010; 5: e11190.

42.

Ahmed

Mosley

. Simultaneity in the use of maternal-child health care and contraceptives: evidence from developing countries. Demography 2002; 39: 75–93.

43.

Ouedraogo

Kimemia

Igonya

, et al. “They talked to me rudely”. Women perspectives on quality of post-abortion care in public health facilities in Kenya. Reprod Health 2023; 20: 35.

44.

Samuel

Fetters

Desta

. Strengthening postabortion family planning services in Ethiopia: expanding contraceptive choice and improving access to long-acting reversible contraception. Glob Health Sci Pract 2016; 4: S60.

45.

Che

Dusabe-Richards

, et al. A qualitative exploration of perceptions and experiences of contraceptive use, abortion and post-abortion family planning services (PAFP) in three provinces in China. BMC Womens Health 2017; 17: 113.

46.

Huber

. Postabortion care and the voluntary family planning component: expanding contraceptive choices and service options. Glob Health Sci Pract 2019; 7: S207–S210.

47.

Baynes

Yegon

Lusiola

, et al. Post-abortion fertility desires, contraceptive uptake and unmet need for family planning: voices of post-abortion care clients in Tanzania. J Biosoc Sci 2021; 53: 908–923.