Prevalence of pregnancy, unintended pregnancy, poor pregnancy outcome and its associated factors among female sex workers in Cameroon

Abstract

Background

Sub-Saharan Africa has a high prevalence of pregnancy among female sex workers (FSWs). Understanding FSWs’ pregnancy intentions and how they affect pregnancies can help prevent unintended pregnancies (UP) and poor pregnancy outcome. This study investigates factors associated with pregnancy and outcomes among FSWs in Cameroon.

Methods

A cross-sectional study was conducted in FSWs aged 18 to 49 years. Participants were recruited through voluntary sampling in eight community-based organisations across five towns in Cameroon during July 2023. The three outcomes were pregnancy, UP and poor pregnancy outcome within the last twelve months. A binary logistic regression using SPSS 27 with backward Wald’s method and Robust (modified) Poisson Regression was performed to identify factors associated with the outcomes of interest.

Results

A total of 481 female sex workers (FSWs) with a median age of 24 years were recruited for the study. Among them, 101 (21%) experienced at least one pregnancy during the last twelve months, of which 34 (33.7%) was UP. Among those who reported pregnancy, 53.5% (54/101) had a live birth versus 20.6% (7/34) for those who experienced UP. FSWs aged > = 25 years (Vs < = 24 years) were less likely to experience pregnancy (aOR = 0.49, 95% CI: 0.29–0.83) and those who abuse of at least one psychoactive substance (PAS) were more likely to have at least one pregnancy (aOR = 1.91, 95% CI: 1.002–3.65). UP was lower among those who had another source of income than prostitution (aRR = 0.39, 95% CI: 0.15–0.98). The risk of poor pregnancy outcome increases by almost 3 (aRR = 2.66, 95% CI: 1.76-4.00) for FSWs who faced UP.

Conclusion

Among FSWs in Cameroon, younger age (< 25 years), positive pregnancy intention, and psychoactive substance use were all significantly associated with higher odds of experiencing pregnancy in the past twelve months. Additionally, unintended pregnancy more than doubled the risk of a poor pregnancy outcome (abortion or miscarriage). These findings underscore the urgent need for targeted interventions promoting consistent contraceptive use and addressing substance use—especially among younger FSWs who intend pregnancy—to reduce unintended pregnancies and related adverse outcomes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-23736-7.

Introduction

Globally, 40% of pregnancies are unintended, with 50% of these resulting in abortions [1, 2]. In Sub-Saharan Africa (SSA), FSWs face an elevated risk of unintended pregnancy (UP) and Human Immunodeficiency Virus (HIV)/sexually-transmitted infections (STIs), indicating a significant unmet need for family planning [3–6]. This can be attributed to several reasons, such as inconsistent condom use, which is influenced by a number of factors, such as restrictive policy environments, stigma and discrimination in healthcare settings, gender inequality, and economic marginalisation [7]. While evidence from SSA is limited, studies in Madagascar, Kenya, Ethiopia, Zambia, Uganda, Ivory Coast and Cameroon have shown that unintended pregnancy and TOP among FSWs are prevalent [4, 8–12]. In some studies, the proportion of unintended pregnancies occurring during sex work was around 90% [11, 13] and the lifetime abortion rate varies between 15% and 65% [13–17] among FSWs from low and middle-income countries. Recent meta-analysis revealed that the pooled lifetime prevalence of at least one and multiple induced abortions was 37.7% and 21.7%, respectively. Also, among FSWs recruited in countries where elective abortion is illegal, at least one induced abortion prevalence was 35.1% (28.1–42.4), and multiple induced abortion prevalence was 23.1% (12.4–35.9). In countries where elective abortion is legal, at least one induced abortion prevalence was 44.6% (34.8–54.6), and multiple induced abortion prevalence was 19.9% (11.9–29.3) [18].

In Cameroon, while less than one in ten FSWs reported using a long-acting reversible contraceptives (LARC), more than half of the surveyed 2,255 FSWs reported a history of unintended pregnancy and 40% of TOP [4]. It is imperative for FSWs to avoid unintended pregnancies [19] yet a subset of FSWs also prioritises motherhood [20], even while continuing in their sex work [21, 22]. FSWs may have pregnancy intention levels that are comparable to those of women in other occupational groups. However, in examining the sexual and reproductive health of FSWs in low-income and middle-income countries (LMICs), there is a paucity of studies that address pregnancy, with even fewer measuring pregnancy intention [20–25]. Although there is evidence to suggest an increased risk of poor pregnancy outcomes in unintended pregnancies, there has been little improvement in the quality or quantity of evidence from low-income countries [26]. To address this gap, Perrault-Sullivan et al.‘s longitudinal study of pregnancy intention and outcome is a valuable piece of research [24]. In it, pregnancy intention was prospectively assessed using a valid measure and analyses included confounding or mediating factors. However, the study did not formally address factors associated with unintended pregnancies. Furthermore, a paucity of empirical data is available concerning related phenomena in the same time: pregnancy intention; pregnancy occurrence; unintended pregnancy; and its influence on pregnancy outcome among female sex workers (FSWs) in Sub-Saharan Africa (SSA), with a particular lack of data from Cameroon.

So, this study aims to describe pregnancy and unintended pregnancy-related characteristics among female sex workers (FSWs) aged 18 and above in Cameroon by estimating: (i) the frequency of pregnancy intention, occurrence of pregnancy and unintended pregnancy, classifying pregnancy and unintended pregnancy outcomes; (ii) identify the sociodemographic, behavioural, and contraceptive factors—including pregnancy intention—associated with the occurrence of at least one pregnancy in the past twelve months of sex work practice; (iii) identify the factors associated with unintended pregnancy among FSWs who reported at least one pregnancy; (iv) evaluate the influence of unintended pregnancy on poor pregnancy outcomes (miscarriage or abortion).

Methods

Study design

We conducted a cross-sectional study. This study was based on components of two cross sectional surveys that assessed acceptability and satisfaction with oral PrEP among female sex workers in Cameroon.

Explanation of the structure of the study

The US President’s Emergency Plan for AIDS Relief (PEPFAR) has supported the rollout of oral PrEP for HIV prevention among key populations (KPs) in Cameroon. Two cross-sectional surveys were conducted to evaluate community-based PrEP delivery. These included a satisfaction survey among FSWs using PrEP for at least 12 months, and an acceptability survey among young FSWs not yet using PrEP. In Cameroon, FSWs and MSM may face legal sanctions [27]. PrEP is officially available only to KPs aged 21 or older.

To inform possible policy changes, a specific acceptability study was carried out among KPs aged 18–24 years. Because reproductive health is closely linked to HIV prevention among FSWs [28, 29], both surveys included questions on pregnancy history, intention, and outcomes. This study focuses only on those reproductive health questions. It aims to describe recent pregnancies and assess factors associated with poor pregnancy outcomes in this population.

Data collection

Data collection took place from 17 to 28 July 2023. Twenty-nine trained data collectors worked with peer educators (PEs) from eight community-based organisations (CBOs). PEs mobilised young FSWs in sex work venues such as bars, hotels, and brothels for the acceptability survey. For the satisfaction survey, PrEP users were recruited from drop-in centres (DICs). The study was conducted in five cities: Douala, Yaoundé, Bertoua, Bamenda, and Bafoussam. All CBOs involved offered a basic health service package to FSWs, including HIV and STI screening, PrEP, and condom distribution.

Participants, sample size and sampling

The satisfaction survey included FSWs aged 21 or older who had used PrEP for at least 12 months. The minimum sample size was 245, calculated using WHO recommendations with a 20% adjustment for non-response [30]. The acceptability survey targeted FSWs aged 18 to 24 years who had tested HIV-negative in the past three months. The minimum required sample was 223, using the same method [30].

A voluntary sampling technique was used for both surveys [31, 32]. This method is commonly applied for sensitive research topics like sexual behaviour or abortion [31]. Eligibility for this study required participants to have practiced sex work for at least 12 months. For the satisfaction group, PrEP use duration served as a proxy. In the acceptability group, validation was based on PEs’ reports and participant confirmation.

The current study looked at pregnancy intention and the occurrence of at least one pregnancy in the last 12 months, among FSWs of childbearing age (15–49). FSWs under 50 years who had taken part in the two above surveys were included. Consequently, in order to achieve the minimum sample size required for our study (n = 400) and to gain statistical power, the questionnaire was incorporated into both acceptability of oral PrEP and satisfaction of oral PrEP studies, thus enabling the achievement of the research objectives. Indeed, it has been demonstrated that a minimum total size of 400 is adequate for the implementation of a logistic regression model [33]. Ultimately, the two databases were merged, and the inclusion criteria were applied to obtain the final database of this study. All participants with missing value were excluded for the analysis and the process of integrating the two databases is shown below (Fig. 1).

Fig. 1.

Flowchart of integrating the two databases. * 2 FSWs with missing value for the variable “cohabiting with sexual partner”. + 2 FSWs with missing value for the variable “Affiliation with a sex workers’ association”

Study measures

A series of questions had been meticulously designed to gather specific information from respondents and the questionnaire, which has been developed for this study by the research team, can be found in the additional files. (Additional file 3).

Dependant variables

The primary outcome in this study was the “occurrence of at least one pregnancy during the last 12 months of sex work practice”. The secondary outcome was the “occurrence of at least one unintended pregnancy during the last 12 months of sex work practice”. Unintended pregnancy was a composite variable of “occurrence of at least one pregnancy = yes” and “pregnancy intention = no”. To evaluate the number of women meeting “occurrence of at least one pregnancy in the last 12 months”, we used a retrospective question asked: have you been pregnant in the past twelve months? (yes or no). The third outcome was “the poor outcome of the pregnancy”. This was binary, with the possible values of “Yes, did not give birth/No, gave birth”.

Independent variables

Independent variables were considered based on a priori knowledge from the literature and other hypothesised confounders. The study assessed four main dimensions:

• i)Socio-demographic characteristics, including age, residence, education, cohabitation with a sexual partner, and having income other than sex work. “Cohabiting with a sexual partner” referred to a non-commercial sexual and emotional relationship (e.g., boyfriend or husband).
• ii)Social and risk behaviours, such as sex worker association membership, disclosure of sex work status to close contacts, HIV testing in the past 12 months, sex while intoxicated, alcohol consumption, reported STI symptoms, and abuse of psychoactive substances. The latter was defined as regular (≥ 3 times/week) use of substances including cannabis, cocaine, heroin, ecstasy, tramadol, and traditional euphorics.
• iii)Use of modern contraceptive methods (at least one).
• iv)Pregnancy intention over the past 12 months, assessed using a reverse-adapted version of the “One Key Question” (OKQ) [34–36] participants who answered “yes, trying to become pregnant” or “no” to the question “Over the past twelve months, have you made any plans or expressed any desire to have children?” were considered as having a declared pregnancy intention.

Data management and analysis

Data management

The variable “towns” was recoded into place of residence (cultural area): Yaoundé and Bertoua into “Fang Beti”, Douala into “Sawa”, Bafoussam and Bamenda into “Grassfield”. The variable “age” was dichotomized around the median. The variables “use of at least one modern contraceptive last 12 months (Yes, no)” and “if yes, which one” were combined into three categories (no, yes non-barriers methods, yes barrier method). All participants with missing value were excluded for the analysis.

Data analysis

All statistical analyses were performed using SPSS software version 27.

Description of participants’ characteristics

Socio-demographic and behavioural characteristics of participants were summarised using medians and interquartile ranges (IQR) for continuous variables, and proportions for categorical variables.

Descriptive analysis of pregnancy indicators

To address the first study objective—describing the frequency of pregnancy intention, occurrence of pregnancy, and unintended pregnancy, and classifying pregnancy outcomes—we used descriptive statistics. These outcomes were presented as simple proportions. All variables were categorical, and no continuous variables were involved. Frequencies were reported for pregnancy in the last 12 months, intentionality, and the resulting pregnancy outcomes (live birth, abortion, miscarriage).

Analysis for explanatory objectives

To identify sociodemographic, behavioural, and contraceptive factors—including pregnancy intention—associated with the occurrence of at least one pregnancy in the past twelve months, we performed univariate and multivariable logistic regression analyses. Variables with p < 0.1 in univariate analysis, along with a priori variables, were included in the final model using backward stepwise selection. Adjusted odds ratios (aOR) with 95% confidence intervals (CI) were reported. Model diagnostics included assessment of multicollinearity, outliers, and influential observations (see Additional file 2). Linearity of the logit was not tested, as no continuous predictors were retained.

Among participants who reported at least one pregnancy, Robust (modified) Poisson Regression (RPR) [37–42] was used to identify factors associated with unintended pregnancy. Variables with p < 0.2 in univariate analysis and a priori variables were included in the multivariable model. A top-down modelling approach was applied, retaining only variables with p < 0.1, except those selected a priori. Results are expressed as adjusted risk ratios (aRR) with 95% CI.

To evaluate the influence of unintended pregnancy on poor pregnancy outcomes (abortion or miscarriage), we again used robust Poisson regression. Variables with p < 0.2 in univariate analysis and a priori variables were entered into the multivariable model. The same top-down selection approach was used. RPR was chosen for its suitability in cross-sectional data with binary outcomes and its robustness with sample sizes as small as 100 [39]. It does not require the Poisson distribution assumption [42].

All statistical tests were two-sided, and statistical significance was set at p < 0.05.

Ethical considerations

In accordance with the declaration of Helsinki, this study was conducted with the standards set forth by the Council for International Organizations of Medical Sciences (CIOMS) [43]. The study involved FSWs who provided signed informed consent and received monetary compensation, approximately 2 US dollar in Cameroun, to cover transport fees and potential income loss. The study was approved by the National Human Health Research Ethics Committee in Cameroon with reference N° 2023/02/1945/L/CNERSH (Additional file 1).

Results

Demographic characteristics of participants

A total of 481 participants were included, with a median age of 24 years (IQR: 22–30). More than half (61.3%) were under 25 years old. Most participants were from the Fang-Beti (46.6%) or Sawa (33.3%) cultural areas. Further details are presented in Table 1.

Table 1.

Sociodemographic characteristics, social and behavioural risks a associated with the participants

Variable	All FSWs (N = 481)
	n	(%)
Sociodemographic characteristics
Age (years)
< 25	295	61.3
≥ 25	186	38.7
Median (IQR)	24 (22–30)
Place of residence (cultural area)
Fang Beti	224	46.6
Sawa	160	33.3
Grassfield	97	20.2
Education
Elementary and under	95	19.8
Secondary	308	64.0
Tertiary	78	16.2
Cohabiting with sexual partner
No	443	92.1
Yes	38	7.9
A source of income other than prostitution
No	307	63.8
Yes	174	36.2
Social and behavioural risks
Affiliation with a sex workers’ association
No	248	51.6
Yes	233	48.4
People close to you know about your FSW status
No	247	51.4
Yes	234	48.6
HIV test within the last 12 months
No	23	4.8
Yes	458	95.2
Sexual intercourse while intoxicated with PAS or alcohol
No	172	35.8
Yes	309	64.2
The abuse of at least one PAS over the past 12 months
No	417	86.7
Yes	64	13.3
Alcohol consumption
No	88	18.3
Yes	393	81.7
One or more symptoms of STIs over the past 12 months
No	131	27.2
Yes	350	72.8
Perceived level of risk of HIV infection
Decreased	341	70.9
Increased	140	29.1

N total number of subjects included in the analysis, n number in category, % percentage

Less than half (48.4%) of FSWs indicated that they were members of a community-based sex workers’ association: similarly, 48.6% reported that their relatives were aware of their sex work. Almost all (95.2%) of the FSWs had undergone at least one HIV test in the 12 months preceding the survey. Over three out of five FSWs (64.2%) had reported engaging in sexual activity under the influence of psychoactive substances in the six months preceding the survey. Just over one in ten FSWs (13.3%) had reported abuse of at least one psychoactive substance while four out of five (81.7%) had reported alcohol consumption in the previous 12 months. A total of 72.8% reported experiencing at least one sign or symptom of an STI in the previous 12 months. 29.1% perceived themselves to be at an increased risk of contracting HIV (Table 1).

The study found that 32.6% of FSWs had the intention to become pregnant in the past 12 months, with 61.5% using at least one modern contraceptive method, with barriers method (condoms) being the most used method (40.5%). Finally, 101 of 481 FSWs (21.0%) reported at least one pregnancy in the previous 12 months (Table 2).

Table 2.

Reproductive health factors among participants

Variable	All FSWs (N = 481)
	n	(%)
The intention to become pregnant in the previous 12 months
No	324	67.4
Yes	157	32.6
Use of at least one modern contraceptive method over the last 12 months
No	185	38.5
Yes non-barriers methods	101	21.0
Injectables	20	4.2
Pill	38	7.9
Implants	41	8.5
Intra-Uterine Device (IUD)	2	0.4
Yes barrier method (Condoms)	195	40.5
Occurrence of at least one pregnancy in the last 12 months
No	380	79.0
Yes	101	21.0

N total number of subjects included in the analysis, n number in category

Pregnancy was unintended for one out of three FSWs (33,7%) who had at least one occurrence of pregnancy in the last 12 months (Fig. 2).

Fig. 2.

Occurrence of at least one unintended pregnancy in the last 12 months of sex work practice

Just over half (53.5%) of the female sex workers who reported at least one pregnancy in the last 12 months had given live birth (Fig. 3). However, of those whose pregnancy was unintended, only one out of five had given live birth (20.6%), the result of the pregnancy being either an abortion (40.2%) or a miscarriage (38.2%) (Fig. 3).

Fig. 3.

Pregnancy outcomes among FSWs with at least one occurrence of pregnancy and at least one occurrence of unintended pregnancy

Factors associated with the occurrence of at least one pregnancy last twelve months

Using backwards elimination (Wald method), comprising a priori factors and other independent variables with a significance level set at p < 0.1 in univariate analysis, the variable “Sexual intercourse while intoxicated with psychoactive substances or alcohol was eliminated at step 2 (Fig. 4).

Fig. 4.

The input and eliminating variables in the process of backward regression

The occurrence of at least one pregnancy last 12 months among FSWs was positively influenced by many variables: Being a member of an association of female sex workers (aOR = 1.89, 95%CI (1.14, 3.12), P = 0.013) and the abuse of at least one psychoactive substance in the last 12 months (aOR = 1.91, 95%CI (1.002, 3.65), P = 0.049), increase the likelihood of at least one pregnancy in the last 12 months by 89% and 91% respectively (Table 3). Likewise, there is a positive association between occurrence of at least one pregnancy and: pregnancy intention (aOR = 7.65, 95%CI (4.56, 12.75), P < 0.001), use of at least one contraceptive method, including barrier (aOR = 2.17, 95%CI (1.22, 3.87), P = 0.009) or non-barrier types (aOR = 2.26, 95%CI (1.14, 4.49), P = 0.02), within the last 12 months. The likelihood of at least one pregnancy occurrence within this timeframe increases by 7.65, 2.17, and 2.26 times respectively, when compared to no pregnancy intention or no contraceptive method use (Table 3).

Table 3.

Univariate and multivariable analysis of at least one pregnancy occurrence

Variable	Univariate analysis of at least one pregnancy					Multivariable analysis
	N	n	(%)	Crude OR (95%CI)	P	Adjusted OR (95%CI)	P	N 2
Place of residence (cultural area)								481
Fang-Beti	224	45	20.1	1
Sawa	160	33	20.6	1.03 (0.62–1.71)	0.898
Grassfield	97	23	23.7	1.24 (0.69–2.18)	0.466
Age (years)
≤ 24	295	72	24.4	1		1
≥ 25	186	29	15.6	0.57 (0.35–0.92)	0.022	0.49 (0.29–0.83)	0.008**
Cohabiting with sexual partner
No	443	95	21.4	1
Yes	38	6	15.8	0.69 (0.28–1.70)	0.414
Education
Elementary and under	95	19	20.0	1
Secondary	308	67	21.8	1.11 (0.63–1.97)	0.715
Tertiary	78	15	19.2	0.95 (0.45–2.02)	0.899
A source of income other than prostitution
No	307	69	22.5	1
Yes	174	32	18.4	0.78 (0.90–1.24)	0.291
Affiliation with a sex workers’ association
No	248	43	17.3	1		1
Yes	233	58	24.9	1.58 (1.01–2.46)	0.043	1.89 (1.14–3.12)	0.013*
People close to you know about your FSW status
No	247	48	19.4	1
Yes	234	53	22.6	1.21 (0.78–1.88)	0.387
HIV test within the last 12 months
No	23	4	17.4	1
Yes	458	97	21.2	1.28 (0.42–3.84)	0.664
The intention to become pregnant in the previous 12 months
No	324	34	10.5	1		1
Yes	157	67	42.7	6.35 (3.95–10.22)	< 0.001	7.65 (4.56–12.75)	< 0.001**
Use of at least one modern contraceptive method over the last 12 months (a)
No	185	35	18.9	1.		1
Yes, non-barrier method (IUD, Pills, implants, injectables)	101	22	21.8	1.19 (0.66–2.17)	0.563	2.26 (1.14–4.49)	0.02*
Yes, barrier method (condom)	195	44	22.6	1.25 (0.76–2.05)	0.382	2.17 (1.22–3.87)	0.009**
Sexual intercourse while intoxicated with psychoactive substances or alcohol (a)
No	172	31	18.0	1
Yes	309	70	22.7	1.33 (0.83–2.13)	0.233
One or more symptoms of STIs over the past 12 months
No	131	21	16.0	1
Yes	350	80	22.9	1.55 (0.91–2.63)	0.104
The abuse of at least one PAS over the past 12 months
No	417	82	19.7	1		1
Yes	64	19	29.7	1.73 (0.96–3.11)	0.069	1.91 (1.002–3.65)	0.049*
Alcohol consumption
No	88	19	21.6	1
Yes	383	82	20.9	0.96 (0.54–1.68)	0.880

N total number of subjects include in the univariate analysis, n number of subjects in category, % percentage, p significance, 95%CI 95% confidence interval, N₂ total number of subjects include in the multivariable analysis, (a) a priori factor

** < 1%

* < 5%

Research has shown that the likelihood of having a pregnancy during the practice of sex work decreased as women got older (aOR = 0.49, 95%CI (0.29, 0.83), P = 0.008). Specifically, this suggests that being aged 25 or over decreases the likelihood of at least one pregnancy by 51% (Table 3).

Factors associated with the occurrence of at least one unintended pregnancy last twelve months

The results showed that place of residence was a risk factor for unintended pregnancy, while the availability of a source of income other than sex work was a protective factor against the occurrence of at least one unintended pregnancy among FSWS who reported at least one pregnancy in the last 12 months. Indeed, female sex workers residing in the Sawa cultural area were found to be approximately four times (aRR = 3.94, 95% CI (1.93, 8.05), P < 0.001) more susceptible to experiencing an unintended pregnancy (Table 5). Conversely, those who declared additional income sources beyond prostitution exhibited a 61% reduced risk (aRR = 0.39, 95%CI (0.15, 0.98), P = 0.046) of having an unintended pregnancy (Table 4). Despite the initial findings suggesting that utilising at least one barrier contraceptive method (condoms) might lead to a heightened risk of unintended pregnancy, subsequent analysis did not substantiate this association (Table 4).

Table 5.

Univariate and multivariable analysis of FSWs not to have given birth as a result of the pregnancy within the last 12 months

Variables	Univariate analysis of do not given birth					Multivariable analysis
	N	n	(%)	Crude RR (95%CI)	P	Adjusted RR (95%CI)	P	N 2
Place of residence (cultural area)								101
Fang-Beti	45	18	40.0	1
Sawa	33	22	66.7	1.67 (1.08–2.58)	0.022
Grassfield	23	7	30.4	0.76 (0.37–1.57)	0.455
Age (years)
≤ 24	72	35	48.6	1
≥ 25	29	12	41.4	0.85 (0.52–1.40)	0.524
Cohabiting with sexual partner
No	95	44	46.3	1
Yes	6	3	50.0	1.08 (0.47–2.50)	0.857
Education
Elementary and under	19	9	47.4	1
Secondary	67	27	40.3	0.85 (0.48–1.49)	0.570
Tertiary	15	11	73.3	1.55 (0.88–2.74)	0.132
A source of income other than prostitution
No	69	36	52.2	1
Yes	32	11	34.4	0.66 (0.39–1.13)	0.126
Affiliation with a sex workers’ association
No	43	21	48.8	1
Yes	58	26	44.8	0.92 (0.60–1.40)	0.689
People close to you know about your FSW status
No	48	21	43.8	1
Yes	53	26	49.1	1.12 (0.73–1.72)	0.596
Unintended pregnancy
No	67	20	29.9	1
Yes	34	27	79.4	2.66 (1.76-4.00)	< 0.001	2.66 (1.76-4.00)	< 0.001**
Use of at least one modern contraceptive method over the last 12 months (a)
No	35	14	40.0	1
Yes, non-barrier method (IUD, Pills, implants, injectables)	22	8	36.4	0.91 (0.45–1.82)	0.786
Yes, barrier method (condom)	44	25	56.8	1.42 (0.87–2.31)	0.156
One or more symptoms of STIs over the past 12 months
No	21	8	38.1	1
Yes	80	39	48.8	1.28 (0.70–2.32)	0.414
The abuse of at least one PAS over the past 12 months
No	82	39	47.6	1
Yes	19	8	42.1	0.88 (0.50–1.58)	0.678
Alcohol consumption
No	19	10	52.6	1
Yes	82	37	45.1	0.86 (0.52–1.40)	0.538
Perceived level of risk of HIV infection
Decreased	68	33	48.5	1
Increased	33	14	42.4	0.87 (0.54–1.40)	0.574

N total number of subjects include in the univariate analysis, n number of subjects in category, % percentage, p significance, 95%CI 95% confidence interval, N₂ total number of subjects include in the multivariable analysis, (a) a priori factor

** < 1%

Table 4.

Univariate and multivariable analysis of at least one occurrence of unintended pregnancy among FSWs who reported at least one pregnancy in the last 12 months

Variables	Univariate analysis of at least one UP					Multivariable analysis
	N	n	(%)	Crude RR (95%CI)	P	Adjusted RR (95%CI)	P	N 2
Place of residence (cultural area)								101
Fang-Beti	45	7	15.6	1
Sawa	33	23	69.7	4.48 (2.19–9.18)	< 0.001	3.94 (1.93–8.05)	< 0.001**
Grassfield	23	4	17.4	1.12 (0.36–3.43)	0.845	1.19 (0.40–3.59)
Age (years)
≤ 24	72	27	37.5	1
≥ 25	29	7	24.1	0.64 (0.32–1.31)	0.224
Cohabiting with sexual partner
No	95	32	33.7	1
Yes	6	2	33.3	0.99 (0.31–3.18)	0.986
Education
Elementary and under	19	7	36.8	1
Secondary	67	20	29.9	0.81 (0.41–1.62)	0.552
Tertiary	15	7	46.7	1.27 (0.57–2.82)	0.562
A source of income other than prostitution
No	69	30	43.5	1		1
Yes	32	4	12.5	0.29 (0.11–0.75)	0.011	0.39 (0.15–0.98)	0.046*
Affiliation with a sex workers’ association
No	43	16	37.2	1
Yes	58	18	31.0	0.83 (0.48–1.44)	0.515
People close to you know about your FSW status
No	48	14	29.2	1
Yes	53	20	37.7	1.29 (0.74–2.27)	0.368
HIV test within the last 12 months
No	4	2	50.0	1
Yes	97	32	33.0	0.66 (0.24–1.83)	0.424
Use of at least one modern contraceptive method over the last 12 months (a)
No	35	7	20.0	1.
Yes, non-barrier method (IUD, Pills, implants, injectables)	22	8	36.4	1.82 (0.77–4.30)	0.174
Yes, barrier method (condom)	44	19	43.2	2.16 (1.03–4.54)	0.043
Sexual intercourse while intoxicated with psychoactive substances or alcohol
No	31	14	45.2	1
Yes	70	20	28.6	0.63 (0.37–1.08)	0.094
One or more symptoms of STIs over the past 12 months
No	21	7	33.3	1
Yes	81	27	33.8	1.01 (0.51–1.99)	0.971
The abuse of at least one PAS over the past 12 months
No	82	25	30.5	1
Yes	19	9	47.4	1.55 (0.87–2.76)	0.134
Alcohol consumption
No	19	9	47.4	1
Yes	82	25	30.5	0.64 (0.36–1.14)	0.134
Perceived level of risk of HIV infection
Decreased	68	24	35.3	1
Increased	33	10	30.3	0.86 (0.47–1.58)	0.624

N total number of subjects include in the univariate analysis, n number of subjects in category, % percentage, p significance, 95%CI 95% confidence interval, N₂ total number of subjects include in the multivariable analysis, (a) a priori factor

** < 1%

* < 5%

Factors associated with poor pregnancy outcome last twelve months

The risk of not giving birth has been linked only to the occurrence of an unintended pregnancy; in fact, FSWs who have had an unintended pregnancy have a more than 2.6 times higher risk (aRR = 2.66, 95%CI (1.76, 4.00), P < 0.001) of not giving birth (Table 5). Despite initial results suggesting that the place of residence (Sawa cultural area) of FSWs might lead to an increased risk of not carrying the pregnancy to term and giving birth, subsequent analyses did not confirm this association (see Table 5).

Discussion

Pregnancy intention in the past twelve months

In this study, 32.6% of FSWs expressed a pregnancy intention in the past year. These finding highlights that a significant proportion of FSWs maintain reproductive aspirations, challenging the assumption that their occupation precludes maternal desire. This level is comparable to the general population and confirms that FSWs, like other women, may wish to become mothers [44]. The variation in estimates across studies (11–64%) reflects differences in context, study design, and how intention is measured. For example, several studies in South Africa, Canada, Kenya, Uganda, and Cameroon used single-question assessments [4, 20, 45, 46] while others in Uganda and Benin/Mali used scaled tools [24, 25]. The South African, Canadian, Ugandan and Cameroonian studies employed a cross-sectional design [4, 20, 45], whereas the Kenyan and Beninese/Malian studies employed a prospective longitudinal design [24, 46].

The high proportion observed in our study may be attributed to the dichotomous measure utilising the inverse key question. The planning status was categorised into the binary outcomes of planned and unplanned pregnancy, which assumes that women are clear about their intention to become pregnant or not before conception. As the pregnancy planning was not categorised into three group, the ambivalent status, which allows women to express indecision about pregnancy planning [24, 25], was not included. Indeed, a considerable proportion of the participants in this study may have been ambivalent about planning a pregnancy.

From a public health perspective, this finding underlines the need for more nuanced screening tools to assess reproductive goals among FSWs, and to promote access to services that align with their preferences.

Pregnancy occurrence in the past twelve months and its associated factors

About one in five FSWs reported at least one pregnancy in the previous 12 months. This result is consistent with studies that assessed occurrence of pregnancies among female sex workers in sub-Saharan Africa during the practice of sex work [9, 17, 24, 47, 48]. The majority of studies which have explored this question have estimated the lifetime prevalence of pregnancy (with results significantly higher) [3, 4, 8, 49], with the limitation that sex work was not a lifelong constant.

Multivariable analysis revealed that, having expressed a pregnancy intention, being under 25 years of age and reporting use of modern contraceptive (barrier and non-barrier) were significantly associated with increase odds of at least one occurrence of pregnancy. This results corroborate those of studies conducted in Benin and Mali, which found that young age [47] and pregnancy intention [24] were positively associated with the occurrence of at least one pregnancy. However, in contrast to the Benin results, history of HIV testing and cohabitation with a sexual partner, which were positively associated with the occurrence of pregnancy, did not emerge as statistically significant variables in our study. The positive association between modern contraceptives use and the occurrence of at least one pregnancy among FSWs may seem contradictory. However, similar findings have been reported in Uganda, where women using dual methods still experienced hight pregnancy rates [50], often due to inconsistent or incorrect use, but contrast with the Benin study [47]. This highlights a gap not in access, but in effective and consistent use of contraception among FSWs.

Psychoactive substance use was also associated with a higher odds of pregnancy. Substance use may impair judgment and reduce contraceptive adherence or negotiation capacity, leading to unprotected sex. This is consistent with literature showing that drug use compromises sexual and reproductive health autonomy among FSWs [51]. The differences in results between this study and others are due to the inclusion of different control variables in the statistical models considered.

Another notable finding was the positive association between pregnancy and being affiliated with a sex workers’ association. While this variable is rarely explored in other studies, it may reflect increased access to supportive services or a more empowered reproductive decision-making environment. In contrast, factors such as cohabitation or recent HIV testing—previously significant in Benin [47]—were not associated in our analysis, possibly due to population or design differences.

These findings point to the need for comprehensive counselling that goes beyond access to contraception and includes behavioural risk reduction, support for young women, and reinforcement of consistent contraceptive use.

Unintended pregnancy occurrence in the past twelve months and its associated factors

We observed in our study a proportion of 33.7% of unintended pregnancies. This proportion is comparable to the findings of previous study in Ethiopia where the magnitude of unintended pregnancy among female sex workers in the past two years was 28.6% [12]. Nevertheless, this finding falls below the 43% of unintended pregnancies reported among female sex workers in SSA [49, 52]. This high proportion of unintended pregnancy among FSWs is a real public health concern. In fact, FSWs who experienced an unintended pregnancy reported a desire to have an abortion [53]. In Cameroon, where abortion is not currently legal, sex workers often resort to clandestine practices, which carry significant health risks like maternal death. Note that, in many countries in SSA, more than half (56.8%) reported maternal death among FSWs were due to unsafe abortion [54].

Multivariable analysis revealed that, unintended pregnancy was significantly more likely among FSWs residing in the Sawa cultural region. This region is heavily affected by displacement due to armed conflict, so many FSWs in this area are Internally Displaced Persons (IDPs) as a result of the conflict in North-West Cameroon. Also, research conducted in Cameroon revealed that internally displaced women had limited knowledge of modern contraceptive methods [55].

Our study also revealed that having a source of income other than prostitution appeared protective against unintended pregnancy. This aligns with previous findings in SSA suggesting that where sex workers are poorly organised and have few alternative sources of income, they are less able to refuse a client who is unwilling to use a condom [5].

These results support the integration of reproductive health services within humanitarian response efforts, especially in displacement zones, and reinforce the importance of economic empowerment initiatives to reduce reproductive vulnerability.

Pregnancy outcomes and the effect of unintended pregnancy

The findings of our study indicate that more than half of pregnancies result in a live birth. The remainder were due to miscarriage or TOP. However, among FSWs experiencing an unintended pregnancy, the poor outcomes were more significant, with only one in five resulting in a live birth. These results corroborate the continued prevalence of miscarriages and TOP amongst FSWs. In fact, the previous studies conducted in Cameroon, Zambia, Côte d’Ivoire, Uganda, Benin, Togo, Burkina Faso, Swaziland, Madagascar, Colombia, Iran, Canada and China found proportions of TOP and miscarriages ranging from 33 to 67% and 13–45% respectively [4, 8, 10, 13–16, 47, 51, 56–58]. It is crucial to acknowledge that previous studies have documented the proportion of TOP and miscarriages over the lifetime since the inception of sex work. Consequently, the observation that the results do not exhibit a notable variation with the measurement over the past 12 months employed in our study could be attributed to the fact that the prevalence of multiple abortions among female sex workers varies from 17.4 to 72% [10, 24, 47, 57]. Miscarriage could be explained by STIs like Chlamydia trachomatis or others systemic infections like malaria [59–62]. These results highlight the negative consequences of unintended pregnancies on maternal health.

Multivariable analysis showed that unintended pregnancy was the only significant predictor of poor pregnancy outcomes (abortion or miscarriage). It is challenging to make a direct comparison with the findings from the other research as the latter examined the factors associated with just one aspect of poor pregnancy outcomes - abortion. The studies differ significantly in their approach, which makes it difficult to draw meaningful conclusions.

Beyond psychosocial distress, these poor outcomes may reflect biological vulnerability. High prevalence of sexually transmitted infections such as Chlamydia trachomatis, as well as comorbidities like malaria, may contribute to increased miscarriage rates among FSWs.

These findings call for tailored maternal care for FSWs, including antenatal screening, STI and malaria management, and safe abortion care where legally allowed. Interventions should prioritise women experiencing unintended pregnancies, a group clearly at higher risk of adverse reproductive outcomes.

Conclusion and perspectives

This study provides a detailed overview of pregnancy, pregnancy intention, and reproductive outcomes among female sex workers (FSWs) in Cameroon. Nearly one-third of participants reported having planned a pregnancy in the past year, indicating that reproductive aspirations are common in this population. However, approximately one in five FSWs experienced a pregnancy, and one-third of those pregnancies were unintended—highlighting the persistent unmet need for effective reproductive health services. Younger age, pregnancy intention, psychoactive substance use, and inadequate contraceptive use were significantly associated with the occurrence of pregnancy. Unintended pregnancy was more likely among FSWs living in displacement-affected regions and those without a stable income. Importantly, unintended pregnancy was the main predictor of poor pregnancy outcomes such as miscarriage or abortion.

These findings underscore the urgent need to go beyond HIV prevention and integrate harm reduction, sexual and reproductive health education, and economic support into programs for FSWs. Early identification and support of pregnant FSWs can improve access to antenatal and postnatal care, potentially reducing risks to maternal, foetal, and neonatal health. Prevention strategies should focus on reducing unintended pregnancy by ensuring the effective and consistent use of modern contraceptive methods, supported by targeted counselling and service delivery.

Limitation of the study

It is possible that the frequency of our outcomes was underestimated due to the sensitive nature of pregnancy and abortion. It is likely that some women did not disclose their pregnancies or abortions. Consequently, the frequency of pregnancies observed is likely to be an underestimate. Moreover, it is possible that participation in the PrEP Oral programme may have contributed to limit the occurrence of pregnancy; however, this hypothesis was not substantiated by the findings of the present study. Further analysis is required to understand the real effect of oral PrEP program on pregnancy occurrence.

Authors’ contributions

JPYAN contributed to this aspect of the study: ☒ Conceptualization ☒ Methodology ☒ Software ☒ Validation ☒ Formal analysis ☒ Investigation ☒ Data Curation ☒ Writing - Original Draft ☒ Writing - Review & Editing ☒ Visualization. JN and CM contributed to this aspect of the study: ☒ Conceptualization ☒ Methodology ☒ Software ☒ Validation ☒ Formal analysis ☒ Investigation ☒ Writing - Original Draft. RKD contributed to this aspect of the study: ☒ Methodology ☒ Software ☒ Formal analysis ☒ Data Curation ☒ Writing - Original Draft ☒ Writing - Review & Editing ☒ Visualization FAN contributed to this aspect of the study: ☒ Validation ☒ Formal analysis ☒ Investigation ☒ Writing - Review & Editing. FTN contributed to this aspect of the study: ☒ Conceptualization ☒ Validation ☒ Supervision JDA contributed to this aspect of the study: ☒ Conceptualization ☒ Methodology ☒ Validation ☒ Resources ☒ Writing - Review & Editing ☒ Supervision ☒ Project administration MGT contributed to this aspect of the study: ☒ Resources ☒ Writing - Original Draft ☒ Funding acquisition. YNW contributed to this aspect of the study: ☒ Conceptualization ☒ Methodology ☒ Validation ☒ Investigation. MAB contributed to this aspect of the study: ☒ Methodology ☒ Software ☒ Validation ☒ Formal analysis ☒ Investigation ☒ Writing - Original Draft. JM and YNH contributed to this aspect of the study: ☒ Conceptualization ☒ Validation ☒ Investigation ☒ Writing - Original Draft. ZZWA contributed to this aspect of the study: ☒ Resources ☒ Writing - Review & Editing☒ Supervision ☒ Project administration ☒ Funding acquisition IKE contributed to this aspect of the study: ☒ Writing - Review & Editing ☒ Project administration ☒ Funding acquisition. CA contributed to this aspect of the study: ☒ Conceptualization ☒ Writing - Review & Editing ☒ Project administration ☒ Funding acquisition. AZM contributed to this aspect of the study: ☒ Validation ☒ Writing - Review & Editing☒ SupervisionBK contributed to this aspect of the study: ☒ Validation ☒ Writing - Review & Editing☒ Funding acquisitionJF contributed to this aspect of the study: ☒ Validation ☒ Writing - Review & Editing☒ Supervision ☒ Funding acquisitionJA and ACZKB contributed to this aspect of the study: ☒ Validation ☒ Writing - Review & Editing ☒ Supervision ☒ Project administration.

Funding

This article was funded in part by a grant from the United States Department. The opinions, findings and conclusions stated herein are those of the authors and do not necessarily reflect those of the United States Department. The funders had no involvement from study design to submission of paper for publication.

Data availability

Availability of data and materials The data analysed in this study is the intellectual property of the United States Department of State and thus falls outside the parameters of the public domain. However, it is available to interested parties upon submission of a formal request to the senior author.

Declarations

Ethics approval and consent to participate

In accordance with the declaration of Helsinki, this study was conducted with the standards set forth by the Council for International Organizations of Medical Sciences (CIOMS), in collaboration with the World Health Organization (WHO). Each participant provided written informed consent prior to the interview and no nominal information was reported on the questionnaire. The study was approved by the National Health Research Ethics Committee in Cameroon (reference numbers 2023/02/1945/L/CNERSH/SP).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.