You are here

  • Home
  • Archive
  • Volume 12, Issue 10
  • Assessment of the lifetime prevalence and incidence of induced abortion and correlates among female sex workers in Mombasa, Kenya: a secondary cohort analysis

Article Text

Article menu

Download PDFPDF

Sexual health
Original research
Assessment of the lifetime prevalence and incidence of induced abortion and correlates among female sex workers in Mombasa, Kenya: a secondary cohort analysis
  1. Anne Marieke Simmelink1,
  2. Caroline M Gichuki1,2,
  3. Frances H Ampt3,4,
  4. Griffins Manguro2,
  5. Megan S C Lim3,4,
  6. Paul Agius3,4,
  7. Margaret Hellard3,4,5,6,
  8. Walter Jaoko7,
  9. Mark A Stoové3,4,8,
  10. Kelly L'Engle9,
  11. Marleen Temmerman2,10,11,
  12. Peter Gichangi2,11,12,
  13. Stanley Luchters3,11,13,14
  1. 1Department of Population Health, The Aga Khan University, Nairobi, Nairobi, Kenya
  2. 2International Centre for Reproductive Health Kenya, Mombasa, Kenya
  3. 3Burnet Institute, Melbourne, Victoria, Australia
  4. 4Department of Epidemiology and Preventive Medicine, Monash University, Clayton, Victoria, Australia
  5. 5Department of Infectious Diseases, The Alfred Hospital, Melbourne, Victoria, Australia
  6. 6Doherty Institute and School of Population and Global Health, University of Melbourne, Melbourne, Victoria, Australia
  7. 7Department of Medical Microbiology and Immunology, University of Nairobi, Nairobi, Nairobi, Kenya
  8. 8School of Psychology and Public Health, La Trobe University, Melbourne, Victoria, Australia
  9. 9School of Nursing and Health Professions, University of San Francisco, San Francisco, California, USA
  10. 10Department of Obstetrics and Gynaecology, The Aga Khan University Hospital Nairobi, Nairobi, Kenya
  11. 11Department of Public Health and Primary Care, Ghent University, Gent, Belgium
  12. 12Technical University of Mombasa, Mombasa, Kenya
  13. 13Centre for Sexual Health and HIV/AIDS Research (CeSHHAR), Harare, Zimbabwe
  14. 14Liverpool School of Tropical Medicine (LSTM), Liverpool, UK
  1. Correspondence to Stanley Luchters; stanley.luchters{at}ceshhar.co.zw

Abstract

Introduction Prevalence of lifetime-induced abortion in female sex workers (FSWs) in Kenya was previously estimated between 43% and 86%. Our analysis aimed at assessing lifetime prevalence and correlates, and incidence and predictors of induced abortions among FSWs in Kenya.

Methods This is a secondary prospective cohort analysis using data collected as part of the WHISPER or SHOUT cluster-randomised trial in Mombasa, assessing effectiveness of an SMS-intervention to reduce incidence of unintended pregnancy. Eligible participants were current FSWs, 16–34 years and not pregnant or planning pregnancy. Baseline data on self-reported lifetime abortion, correlates and predictors were collected between September 2016 and May 2017. Abortion incidence was measured at 6-month and 12-month follow-up. A multivariable logistic regression model was used to assess correlates of lifetime abortion and discrete-time survival analysis was used to assess predictors of abortions during follow-up.

Results Among 866 eligible participants, lifetime abortion prevalence was 11.9%, while lifetime unintended pregnancy prevalence was 51.2%. Correlates of lifetime abortions were currently not using a highly effective contraceptive (adjusted OR (AOR)=1.76 (95% CI=1.11 to 2.79), p=0.017) and having ever-experienced intimate partner violence (IPV) (AOR=2.61 (95% CI=1.35 to 5.06), p=0.005). Incidence of unintended pregnancy and induced abortion were 15.5 and 3.9 per 100 women-years, respectively. No statistically significant associations were found between hazard of abortion and age, sex work duration, partner status, contraceptive use and IPV experience.

Conclusion Although experience of unintended pregnancy remains high, lifetime prevalence of abortion may have decreased among FSW in Kenya. Addressing IPV could further decrease induced abortions in this population.

Trial registration number ACTRN12616000852459.

  • SEXUAL MEDICINE
  • PUBLIC HEALTH
  • EPIDEMIOLOGY

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Raw data are available upon reasonable request. The original data are not available in a public repository. The corresponding author is to be contacted for the consideration of any data requests.

https://creativecommons.org/licenses/by/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

https://doi.org/10.1136/bmjopen-2021-053218

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • This study presents incidence of abortion in a cluster-randomised cohort of female sex workers (FSWs).

    • It is the first to analyse predictors of abortions in FSWs, rather than correlates of past abortions only.

    • This paper explores a secondary research question, and the study was not originally powered to assess the predictors of abortions during follow-up.

    • The sensitive topic of abortions and sexual and reproductive health in general, might have resulted in a social desirability bias.

    Introduction

    Research findings show that about 5% of Kenya’s urban female reproductive population could be involved in sex work.1 Female sex workers (FSWs) experience higher than average rates of HIV, other sexually transmitted infections (STIs) and unintended pregnancy.2–5

    Unintended pregnancies often have negative consequences for FSWs, including financial adversity, social stigma and induced abortion.4 In countries where abortion is illegal or difficult to access, women frequently resort to unsafe practices, risking severe medical complications.6

    Reported prevalence of lifetime abortion among FSWs in low-income and middle-income countries (LMICs) varies from 24% in Laos in 2012 to 86% in 2000/2001 in Central and Western Kenya.7–16 Many of these abortions are unsafe or sought in the informal sector.8 11 16 In Mombasa, lifetime abortion among FSWs was estimated at 43% in 2008.17

    In 2010, Kenya liberalised its abortion law, making abortion legal when ‘there is need for emergency treatment, or the life or health of the mother is in danger’.18 In practice, older laws criminalising abortion remain in place, creating ambiguity among health professionals with lawsuits remaining a threat.19 Moreover, social, cultural and religious beliefs condemning pregnancy termination, misconceptions about the illegality and costs of the procedure, still hamper access to safe services for women.4 19

    A national study in 2012 estimated an induced abortion rate of 48 per 1000 women among women aged 15–49, based on data of women who sought care for abortion complications.20 It showed a diverse sociodemographic and economic background among these women in terms of educational level, employment status, marital status and religion.

    Incidence of induced abortions among FSWs has not been studied in Kenya and studies analysing correlates of induced abortions elsewhere have been cross-sectional and report correlates of lifetime abortions, precluding attribution of causality.8–15 21 Examining incidence of abortions and identifying predictors will help inform future policies to improve care around abortions for FSWs and the need for longitudinal data about abortions in FSWs has therefore been recognised.8 14

    This secondary data analysis aimed to examine lifetime prevalence and correlates, and incidence and predictors of induced abortions in a cluster-random sample of FSWs in Mombasa, Kenya.

    Material and methods

    Study design

    This is a secondary prospective cohort analysis using data collected as part of the WHISPER or SHOUT cluster-randomised trial in Mombasa. A detailed description of the study protocol can be found elsewhere.22 In summary, the study was a two-arm cluster-randomised controlled trial assessing the effectiveness of two SMS-based interventions targeting sexual and reproductive health (SRH) and nutrition in FSWs in Mombasa, Kenya.

    The study was conducted in two subcounties in Mombasa, Kisauni and Changamwe, between September 2016 and July 2018. Ninety-three venues were randomly sampled with a probability proportional to FSW population size at the venue (figure 1). Trained community mobilisers and peer educators recruited FSWs from the venues until the required sample size of 860 FSWs was achieved. This study uses baseline and follow-up data from the trial. During the recruitment period of 14 September 2016 to 16 May 2017, 882 women were enrolled in the study. Follow-up continued until 31 July 2018. A subsample of 866 women (98.2%) was analysed for this secondary analysis (figure 1).

    Figure 1
    Figure 1

    Eligibility flow diagram for the WHISPER or SHOUT study as per CONSORT.33 34 CONSORT, Consolidated Standards of Reporting Trials; FSW, female sex workers; RCT, randomised controlled trials; SRH, sexual and reproductive health.

    Study participants and procedures

    Women were eligible for the study if they were aged 16–34 years; self-reported to have engaged in paid sex work in the last 6 months; were reportedly not pregnant or planning pregnancy within the next 12 months; resided within the study area and were able to read text messages in basic English. Study-specific community mobilisers visited the selected clusters to recruit FSWs and conducted prescreening interviews identifying women who self-reported to be sex workers. Potentially eligible FSWs were referred to the nearest study clinic for a clinical assessment, including a urine pregnancy test, and STI and HIV testing. Enrolled participants then completed a structured questionnaire administered in Swahili by trained research assistants, who had previously participated in research with FSWs. The questionnaire captured detailed sociodemographic, SRH information. Follow-up visits were scheduled at 6 and 12 months after enrolment. Procedures at follow-up visits were similar to those done at enrolment.

    Participants received two to three SMS per week for 12 months. The messages consisted of stand-alone push messages, role model stories and on-demand messages, accessed using assigned codes. Participants only received and accessed messages on their phones from their assigned intervention.

    Outcomes and correlates

    Lifetime prevalence of induced abortion was assessed at baseline with the question ‘How many times have you ever had an induced abortion?’. Induced abortions were assessed during follow-up, by asking participants if they had been pregnant since their last visit. The outcome of each reported pregnancy was then assessed, and in the case of an induced abortion, the location was documented. Formal sector abortions were those taking place at a government or private hospital/clinic, a private doctor/general practitioner or a family planning (FP) clinic. Informal sector abortions were defined as those taking place at home, a pharmacy or traditional healer.23

    Pregnancies during follow-up were confirmed with a urine pregnancy test at the study clinic, or self-reported by the participant when occurring between study visits. Pregnancy intention for all reported pregnancies was assessed using the London Measure of Unintended Pregnancy (LMUP), a six-item scale. A pregnancy scoring less than 10 out of 12 on the LMUP was defined as unintended.24

    All correlates of lifetime-induced abortions and predictors of incident induced abortions were self-reported at baseline. Use of a highly effective contraceptive method was defined as the use of contraceptive implants, intrauterine devices (IUD), injection, contraceptive pill and sterilisation. High knowledge on FP was defined as answering five out of six true–false statements on FP correctly. Having a positive attitude on FP was defined as agreeing with at least three out of four positive attitude statements. Self-efficacy in FP was defined as high when agreeing with two statements on this topic. Household socioeconomic status tertiles were generated using principal component analysis, based on 12 household assets.25 Prevalence of intimate partner violence (IPV) was assessed by asking if the participant had ever been pushed, slapped, hit or kicked by a partner or had been physically forced to have sex, agreed to sex out of fear of the consequences or forced to do something sexual that she perceived as degrading or humiliating. For each item, it was also assessed if this happened in the previous 12 months.

    Statistical analysis

    Statistical analysis was performed using Stata software V.14 (StataCorp, College Station, TX, USA). Distributions of sociodemographic characteristics, reproductive history, contraceptive use and IPV at baseline were explored with means and SD for continuous variables and proportions for categorical variables. Covariates age, education level, SES-tertile, duration of sex work, having a husband/boyfriend, use of highly effective contraceptives and experience of IPV were included in the multivariate models on the basis of a review of literature.8–15 21 Covariates high FP knowledge, positive attitude to FP and high FP specific self-efficacy were included in the multivariate models on the basis of theoretical assumptions by the coauthors. It was hypothesised that these characteristics would be positively associated with contraceptive use and would protect against experiencing an induced abortion. Correlates of lifetime abortion were identified using weighted multivariable logistic regression. Associations were considered statistically significant at the 5% level. The outcome incident abortion was interval-censored (measured at 6-monthly intervals). Therefore, a discrete-time survival analysis was performed using generalised linear mixed (GLM) modelling with complementary log–log link function and binomial distribution, a method that produces estimated HRs. Abortions during follow-up were analysed for all participants who attended at least one follow-up visit. All outcomes presented here are cluster-adjusted, based on inverse probability sample means to account for sampling bias.

    Patient and public involvement

    Patients or the public were not directly involved in the design, or conduct, or reporting, or dissemination plans of our research.

    Results

    Mean age was 25.5 years (SD=4.7) (table 1). The majority of women (n=765, 88.7%) had at least completed primary education and 306 (34.9%) had completed secondary education. Just over half of the participants (n=484, 56.6%) reported to have a current husband or boyfriend, but 812 (94.0%) reported not to live with a partner. Mean duration of employment in sex work was 4.7 years (SD=3.5). Among women currently reporting a husband or boyfriend, 344/483 (70.8%) had not disclosed their employment in sex work to their partners; 605/861 women (68.9%) worked fulltime in sex work and 508/861 (59.8%) earned more than 2000 Ksh (about US$20) per week from sex work. The majority of women (n=666, 76.1%) had ever been pregnant, and 451/864 (51.2%) ever had an unintended pregnancy; 103/866 (11.9%) reported to have had at least one induced abortion in their lifetime. Among women who had an induced abortion, 58/102 (57.1%) went to a private hospital or clinic for the most recent abortion; 29/102 (29.1%) women had their most recent abortion in the informal sector, like home, a pharmacy or traditional healer. At baseline, 473 women (54.4%) reported to use a highly effective contraceptive method. Three-quarters of FSW (650/866; 75.0%) ever experienced IPV and 525/866 (60.1%) experienced IPV in the past 12 months the before the baseline questionnaire was conducted.

    View this table:
    Table 1

    Sociodemographic characteristics, reproductive history, contraceptive use and intimate partner violence at baseline of female sex workers in Mombasa, Kenya (n=866, unless stated otherwise)

    Women currently not using a highly effective contraceptive (adjusted OR (AOR)=1.76 (95% CI=1.11 to 2.79), p=0.017) and women who ever experienced IPV (AOR=2.61 (95% CI=1.35 to 5.06), p=0.005) were significantly more likely to report a history of induced abortion, when controlled for potential confounders (table 2). Longer duration of sex work showed a borderline positive association with history of abortion (AOR=1.08 (95% CI=1.00 to 1.16), p=0.053). Although higher age was significantly associated with a history of abortion in bivariate analysis, after adjusting for confounding factors, this association was no longer seen.

    View this table:
    Table 2

    Correlates of participants with a history of induced abortion, and cluster-adjusted bivariable and multivariable logistic regression analysis on history of induced abortion (n=866)

    During the study follow-up, 773 women attended at least one follow-up visit (figure 1). Total follow-up time was 9468 months, with an average of 12.2 months per woman (data not shown). A total of 131 participants became pregnant, with a total of 145 pregnancies among these women (figure 2). Of these pregnancies, 122/145 were unintended according to the LMUP. 31 out of 145 pregnancies ended in induced abortion, among 29 women and across 789 women-years at risk. Overall incidence rate was 3.9 induced abortions per 100 women-years of observation. Out of 31 abortions, 19 took place in the formal sector and 12 in an informal setting.

    Figure 2
    Figure 2

    Overview of pregnancy outcomes during the 12-month follow-up. n=773. FP, family planning; GP, general practitioner.

    The GLM modelling of abortion incidence showed that women experiencing IPV in the past year (HR=1.93 (95% CI=0.86 to 4.34), p=0.122) and women not using a highly effective contraceptive (HR=1.51 (95% CI=0.66 to 3.49), p=0.332) exhibited a higher hazard of abortion, independent of other factors, although these results were not significant (table 3). We did not find a relation between age, mean duration of sex work, currently having a husband or boyfriend and the intervention under study and hazard of induced abortion.

    View this table:
    Table 3

    Baseline predictors of incident abortion in FSWs among Mombasa, Kenya (n=773)*

    Discussion

    This study adds to the current knowledge of abortion practices in FSWs. Lifetime-induced abortion prevalence in this population was 11.9%. This seems considerably lower than previous figures of lifetime abortion of 86% in 2004 in central and western Kenya among FSWs of a similar age, and 43% in 2008 in Mombasa among FSWs who were on average 2 years older.7 17 In the former study, it was not specified if these abortions also included spontaneous abortions, which might have overestimated the prevalence of abortions. However, despite the sociodemographic and methodological differences between the studies, the size of the difference is suggestive of an actual lower rate in abortions in this population. The prevalence of abortions found here is also lower than reports from other LMICs ranging between 24% and 64% in Laos and Cote d’Ivoire, respectively.8–16 A possible explanation for this lower prevalence is the relatively high use of highly effective contraceptives of 54% in our cohort, compared with similar studies from LMICs.8 12 14–16 The findings are furthermore consistent with a lower-than-expected HIV prevalence and unintended pregnancy incidence in our cohort and could be a result of peer-mediated interventions implemented over the past years in the Mombasa area.26 27 These have mostly targeted prevention of HIV and STIs, but likely have had a lowering effect on unintended pregnancies and induced abortions as well.22 26 Furthermore, this study, in contrast to above referenced studies, attempted to draw a representative sample of an FSW population from community settings, whereas other studies used non-probability sampling methods, which might have overestimated past abortions.

    Despite the lower-than-expected unintended pregnancy incidence, still 51% of FSWs in our cohort reported an unintended pregnancy in their lifetime. The gap between lifetime unintended pregnancies and lifetime-induced abortions could indicate that many women decide to keep a child from unintended pregnancies, which could be supported by the fact that between 70% and 80% of young FSW in Mombasa have reported one or more children.4 27 It may also indicate a high unmet need for induced abortion services among FSWs, for example due to ongoing or increasing difficulties in accessing SRH or abortion services for this group or increasing sociocultural barriers to abortion. Barriers to accessing other SRH services such as long-acting reversible contraceptives has previously been reported for this population.28

    The present study is one of the few studies to report incidence of abortion among FSWs and to the best of our knowledge, the first to analyse predictors of abortions in FSWs, rather than correlates of past abortions only. Incidence of induced abortion in our cohort was 3.9 per 100 women-years. Compared with other studies from LMICs, this is similar to two studies reporting abortion incidence rates of 3.1 and 3.0 per 100 women-years and lower than a third study reporting 7.4 induced abortions per 100 women-years among FSWs.29 The intervention under study had no measurable effect on unintended pregnancy incidence and is therefore unlikely to have affected incidence of induced abortions.27

    Informal sector abortions where common in this cohort, with 29% of women having had their most recent abortion in the informal sector, and 39% of the reported abortions during follow-up happening in the informal sector. These informal sector abortions, put women at higher risk of complications due to unsafe practices and this denotes a need for information on safer alternatives, like the Marie Stopes clinics.23

    Multiple studies have found both age and duration of sex work to be correlated to past abortions. Commonly higher age8 11 15 and longer duration9 12 of sex work were associated with higher lifetime abortion prevalence. One study found that younger age was associated with past abortions.10 In our cohort, although FSWs with a past abortion in our cohort were older in the crude analysis, after adjusting for other correlates, this difference was no longer significant. The association with longer duration of sex work remained borderline significant in multivariate analysis. We did not find a relation between age and mean duration of sex work and having an induced abortion during follow-up. This might suggest that the association between past abortions and higher age and longer duration of sex work is caused by cumulative exposure to high risk of pregnancies and abortion.

    We found a positive association between currently not using a highly effective contraceptive and having a past abortion. No difference was found in FP-specific self-efficacy or knowledge, or attitude towards FP among women with and without a past abortion. The found association could indicate significant barriers to uptake or continuation of a highly effective contraceptive method post abortion, as has previously been acknowledged by a study in Kenya.30

    In our cohort, experience of IPV was high and the odds of having had a past abortion were more than 2.5-times as high for women who experienced IPV in the past, consistent with findings from other studies.14 15 Our study also showed a positive association between experience of IPV in the past 12 months and abortions during follow-up, but this was not significant. Experience of (intimate partner) violence has been shown to have a negative effect on the reproductive health of FSWs, with greater risks of adverse pregnancy outcomes and forced termination of pregnancy.31 32 Addressing the problem of IPV in this population could further lower induced abortions.

    Limitations

    Some limitations should be considered when interpreting the findings from this study. The sensitive topic of abortions and SRH in general, might have resulted in a social desirability bias. To minimise this, peer-educators and research assistants had previous experience working with the target population and received additional training. Attrition bias might have occurred due to loss to follow-up of pregnant participants, as has been recognised by anecdotal evidence.27 This might have resulted in an underestimation of abortions in our study. The robust multistage sampling method improved the ability to generalise the findings to the larger sex work population in the Coast region. However, this study was done in a well-researched population, targeted by other peer-mediated interventions in the past two decades, which may limit generalisation to sex worker populations in other settings. A further limitation is that this paper explores a secondary research question, and the study was not originally powered to assess the predictors of abortions during follow-up. Unknown timing of the past abortions in relation to studied correlates, restrict judgement of temporality of the studied associations. Finally, measurements of abortions stopped when the intervention stopped, so the actual number of abortions during follow-up might in fact be higher than captured in the study.

    Suggestions for further research

    Future research is needed to explore the trend in abortion incidence among FSWs in Kenya. In order to improve care, we need to better understand current abortion practices, the decision-making process around terminating unintended pregnancies, how uptake of highly effective contraceptives can be increased postabortion, as well as the relationship between experience of IPV and induced abortions.

    Conclusions

    In conclusion, the prevalence of lifetime-induced abortions in a random cohort of FSWs in Mombasa was 11.9% and incidence was 3.9 per 100 women-years, whereas prevalence and incidence of unintended pregnancies were higher at 51% and 15.5 per 100 women-years, respectively. A history of induced abortion was positively associated with not using a highly contraceptive method at baseline and having experienced IPV in the past. The study did not find a significant association with the studied predictors of abortions.

    Data availability statement

    All data relevant to the study are included in the article or uploaded as supplementary information. Raw data are available upon reasonable request. The original data are not available in a public repository. The corresponding author is to be contacted for the consideration of any data requests.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This study involves human participants. The study was approved by the Kenyatta National Hospital and University of Nairobi Ethics and Research Committee, Kenya (KNH-UoN ERC-KNH-ERC/RR/493) and the Monash University Human Research Ethics Committee, Australia (MUHREC-CF16/1552-2016000812). Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    The authors would like to acknowledge the hard work of the field research team, specifically Christine Maghanga, Millicent Okello, Betty Kitili, Judith Wamaua, Elizabeth Bilasi, Marion Mwangi, Promillah Muindi, Rukia Abdallah and Khadija Kassim, as well as all the study participants of the study. The authors gratefully acknowledge the contribution of funding from Australia’s National Health and Medical Research Council for the WHISPER or SHOUT trial (Project Grant GNT 1087006), Career Development Fellowships for SL, Senior Research Fellowship for MS and a Postgraduate Scholarship for FHA. The sponsor did not contribute to study design; data collection, analysis or interpretation; manuscript writing or the decision to submit the article for publication. Finally, we acknowledge the contribution of funding from the Victorian Operational Infrastructure Support Programme received by the Burnet Institute.

    References

    1. ↵
      1. Odek WO,
      2. Githuka GN,
      3. Avery L, et al
      . Estimating the size of the female sex worker population in Kenya to inform HIV prevention programming. PLoS One 2014;9:e89180.doi:10.1371/journal.pone.0089180pmid:http://www.ncbi.nlm.nih.gov/pubmed/24595029
      OpenUrlPubMed
    2. ↵
      1. Baral S,
      2. Beyrer C,
      3. Muessig K, et al
      . Burden of HIV among female sex workers in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Infect Dis 2012;12:538–49.doi:10.1016/S1473-3099(12)70066-Xpmid:http://www.ncbi.nlm.nih.gov/pubmed/22424777
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Cwikel JG,
      2. Lazer T,
      3. Press F, et al
      . Sexually transmissible infections among female sex workers: an international review with an emphasis on hard-to-access populations. Sex Health 2008;5:9–16.doi:10.1071/SH07024pmid:http://www.ncbi.nlm.nih.gov/pubmed/18361849
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Luchters S,
      2. Bosire W,
      3. Feng A, et al
      . "A Baby Was an Added Burden": Predictors and Consequences of Unintended Pregnancies for Female Sex Workers in Mombasa, Kenya: A Mixed-Methods Study. PLoS One 2016;11:e0162871–20.doi:10.1371/journal.pone.0162871pmid:http://www.ncbi.nlm.nih.gov/pubmed/27689699
      OpenUrlCrossRefPubMed
    5. ↵
      1. Ippoliti NB,
      2. Nanda G,
      3. Wilcher R
      . Meeting the reproductive health needs of female key populations affected by HIV in low- and middle-income countries: a review of the evidence. Stud Fam Plann 2017;48:121–51.doi:10.1111/sifp.12020pmid:http://www.ncbi.nlm.nih.gov/pubmed/28422301
      OpenUrlCrossRefPubMed
    6. ↵
      1. World Health Organization
      . Who | unsafe abortion: global and regional estimates of the incidence of unsafe abortion and associated mortality in 2008, 2014.
    7. ↵
      1. Elmore-Meegan M,
      2. Conroy RM,
      3. Agala CB
      . Sex workers in Kenya, numbers of clients and associated risks: an exploratory survey. Reprod Health Matters 2004;12:50–7.doi:10.1016/S0968-8080(04)23125-1pmid:http://www.ncbi.nlm.nih.gov/pubmed/15242210
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Schwartz S,
      2. Papworth E,
      3. Thiam-Niangoin M, et al
      . An urgent need for integration of family planning services into HIV care: the high burden of unplanned pregnancy, termination of pregnancy, and limited contraception use among female sex workers in Côte d'Ivoire. J Acquir Immune Defic Syndr 2015;68 Suppl 2:S91–8.doi:10.1097/QAI.0000000000000448pmid:http://www.ncbi.nlm.nih.gov/pubmed/25723996
      OpenUrlPubMed
    9. ↵
      1. Bautista CT,
      2. Mejía A,
      3. Leal L, et al
      . Prevalence of lifetime abortion and methods of contraception among female sex workers in Bogota, Colombia. Contraception 2008;77:209–13.doi:10.1016/j.contraception.2007.11.013pmid:http://www.ncbi.nlm.nih.gov/pubmed/18279693
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Chanda MM,
      2. Ortblad KF,
      3. Mwale M, et al
      . Contraceptive use and unplanned pregnancy among female sex workers in Zambia. Contraception 2017;96:196–202.doi:10.1016/j.contraception.2017.07.003pmid:http://www.ncbi.nlm.nih.gov/pubmed/28711643
      OpenUrlPubMed
    11. ↵
      1. Lau JTF,
      2. Mui LWH,
      3. Tsui HY, et al
      . Prevalence of induced abortion and associated factors among Chinese female sex workers in Hong Kong. J Sex Marital Ther 2007;33:19–29.doi:10.1080/00926230600998441pmid:http://www.ncbi.nlm.nih.gov/pubmed/17162485
      OpenUrlPubMed
    12. ↵
      1. Todd CS,
      2. Alibayeva G,
      3. Sanchez JL, et al
      . Utilization of contraception and abortion and its relationship to HIV infection among female sex workers in Tashkent, Uzbekistan. Contraception 2006;74:318–23.doi:10.1016/j.contraception.2006.04.006pmid:http://www.ncbi.nlm.nih.gov/pubmed/16982233
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Zhang X-D,
      2. Kennedy E,
      3. Temmerman M, et al
      . High rates of abortion and low levels of contraceptive use among adolescent female sex workers in Kunming, China: a cross-sectional analysis. Eur J Contracept Reprod Health Care 2014;19:368–78.doi:10.3109/13625187.2014.927421pmid:http://www.ncbi.nlm.nih.gov/pubmed/24981306
      OpenUrlPubMed
    14. ↵
      1. Zhang X-D,
      2. Myers S,
      3. Yang H-J, et al
      . Prevalence and correlates of sexual and gender-based violence against Chinese adolescent women who are involved in commercial sex: a cross-sectional study. BMJ Open 2016;6:e013409.doi:10.1136/bmjopen-2016-013409pmid:http://www.ncbi.nlm.nih.gov/pubmed/27993907
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Karamouzian M,
      2. Mirzazadeh A,
      3. Shokoohi M, et al
      . Lifetime abortion of female sex workers in Iran: findings of a national bio-behavioural survey in 2010. PLoS One 2016;11:e0166042–12.doi:10.1371/journal.pone.0166042pmid:http://www.ncbi.nlm.nih.gov/pubmed/27861487
      OpenUrlPubMed
    16. ↵
      1. Cleeve A,
      2. Phrasisombath K,
      3. Sychareun V, et al
      . Attitudes and experiences regarding induced abortion among female sex workers, Savannakhet Province, Laos. Sex Reprod Healthc 2014;5:137–41.doi:10.1016/j.srhc.2014.03.001pmid:http://www.ncbi.nlm.nih.gov/pubmed/25200975
      OpenUrlPubMed
    17. ↵
      1. Sutherland EG,
      2. Alaii J,
      3. Tsui S, et al
      . Contraceptive needs of female sex workers in Kenya - a cross-sectional study. Eur J Contracept Reprod Health Care 2011;16:173–82.doi:10.3109/13625187.2011.564683pmid:http://www.ncbi.nlm.nih.gov/pubmed/21413869
      OpenUrlPubMed
    18. ↵
      . Constitution of Kenya laws of Kenya. Natl Counc law report; 2010: 191.
    19. ↵
      1. Ushie BA,
      2. Juma K,
      3. Kimemia G, et al
      . Community perception of abortion, women who abort and abortifacients in Kisumu and Nairobi counties, Kenya. PLoS One 2019;14:e0226120–13.doi:10.1371/journal.pone.0226120pmid:http://www.ncbi.nlm.nih.gov/pubmed/31830102
      OpenUrlPubMed
    20. ↵
      1. Ministry of Health Republic of Kenya
      . Incidence and Complications of Unsafe Abortion in Kenya - Key Findings of a National Study, 2013.
    21. ↵
      1. Yi S,
      2. Tuot S,
      3. Chhoun P, et al
      . Factors associated with induced abortion among female entertainment workers: a cross-sectional study in Cambodia. BMJ Open 2015;5:1–8.doi:10.1136/bmjopen-2015-007947pmid:http://www.ncbi.nlm.nih.gov/pubmed/26231754
      OpenUrlCrossRefPubMed
    22. ↵
      1. Ampt FH,
      2. Mudogo C,
      3. Gichangi P, et al
      . WHISPER or SHOUT study: protocol of a cluster-randomised controlled trial assessing mHealth sexual reproductive health and nutrition interventions among female sex workers in Mombasa, Kenya. BMJ Open 2017;7:e017388.doi:10.1136/bmjopen-2017-017388pmid:http://www.ncbi.nlm.nih.gov/pubmed/28821530
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Chemlal S,
      2. Russo G
      . Why do they take the risk? A systematic review of the qualitative literature on informal sector abortions in settings where abortion is legal. BMC Womens Health 2019;19:1–11.doi:10.1186/s12905-019-0751-0
      OpenUrlPubMed
    24. ↵
      1. Hall JA,
      2. Barrett G,
      3. Copas A, et al
      . London measure of unplanned pregnancy: guidance for its use as an outcome measure. Patient Relat Outcome Meas 2017;8:43–56. Volume.doi:10.2147/PROM.S122420pmid:http://www.ncbi.nlm.nih.gov/pubmed/28435343
      OpenUrlPubMed
    25. ↵
      1. Vyas S,
      2. Kumaranayake L
      . Constructing socio-economic status indices: how to use principal components analysis. Health Policy Plan 2006;21:459–68.doi:10.1093/heapol/czl029pmid:http://www.ncbi.nlm.nih.gov/pubmed/17030551
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Manguro GO,
      2. Gichuki C,
      3. Ampt FH, et al
      . Hiv infections among female sex workers in Mombasa, Kenya: current prevalence and trends over 25 years. Int J STD AIDS 2020;31:1389–97.doi:10.1177/0956462420950571pmid:http://www.ncbi.nlm.nih.gov/pubmed/33103582
      OpenUrlPubMed
    27. ↵
      1. Ampt FH,
      2. Lim MSC,
      3. Agius PA, et al
      . Effect of a mobile phone intervention for female sex workers on unintended pregnancy in Kenya (WHISPER or SHOUT): a cluster-randomised controlled trial. Lancet Glob Health 2020;8:e1534–45.doi:10.1016/S2214-109X(20)30389-2pmid:http://www.ncbi.nlm.nih.gov/pubmed/33220217
      OpenUrlPubMed
    28. ↵
      1. Ampt FH,
      2. Lim MSC,
      3. Agius PA, et al
      . Use of long‐acting reversible contraception in a cluster‐random sample of female sex workers in Kenya. Int J Gynecol Obstet 2019;146:184–91.doi:10.1002/ijgo.12862
      OpenUrlCrossRefPubMed
    29. ↵
      1. Ampt FH,
      2. Willenberg L,
      3. Agius PA, et al
      . Incidence of unintended pregnancy among female sex workers in low-income and middle-income countries: a systematic review and meta-analysis. BMJ Open 2018;8:21779.doi:10.1136/bmjopen-2018-021779pmid:http://www.ncbi.nlm.nih.gov/pubmed/30224388
      OpenUrlPubMed
    30. ↵
      1. Makenzius M,
      2. Faxelid E,
      3. Gemzell-Danielsson K, et al
      . Contraceptive uptake in post abortion care—Secondary outcomes from a randomised controlled trial, Kisumu, Kenya. PLoS One 2018;13:e0201214–13.doi:10.1371/journal.pone.0201214
      OpenUrlCrossRefPubMed
    31. ↵
      1. Swain SN,
      2. Saggurti N,
      3. Battala M, et al
      . Experience of violence and adverse reproductive health outcomes, HIV risks among mobile female sex workers in India. BMC Public Health 2011;11:357.doi:10.1186/1471-2458-11-357pmid:http://www.ncbi.nlm.nih.gov/pubmed/21599984
      OpenUrlCrossRefPubMed
    32. ↵
      1. McDougal L,
      2. Strathdee SA,
      3. Rangel G, et al
      . Adverse pregnancy outcomes and sexual violence among female sex workers who inject drugs on the United States-Mexico border. Violence Vict 2013;28:496–512.doi:10.1891/0886-6708.11-00129pmid:http://www.ncbi.nlm.nih.gov/pubmed/23862312
      OpenUrlAbstract/FREE Full Text
    33. ↵
      1. Campbell MK,
      2. Piaggio G,
      3. Elbourne DR, et al
      . Consort 2010 statement: extension to cluster randomised trials. BMJ 2012;345:e5661–23.doi:10.1136/bmj.e5661pmid:http://www.ncbi.nlm.nih.gov/pubmed/22951546
      OpenUrlFREE Full Text
    34. ↵
      1. Cheuk E,
      2. Isac S,
      3. Musyoki H
      . Informing HIV prevention programs for adolescent girls and young women: a modified approach to programmatic mapping and key population size estimation. J Med Internet Res 2019;21:1–11.doi:10.2196/11196
      OpenUrlCrossRefPubMed

    Footnotes

    • Contributors SL was the principal investigator on the study and is the guarantor of this work. SL, FHA, PG, MSCL, PA, MH, WJ, MAS, KLE, and MT contributed to the study design. CMG and GM coordinated the trial and undertook data acquisition in Kenya under the supervision of PG. AMS and SL conceptualised the manuscript. AMS and PAA conducted the statistical analyses. AMS and CG wrote the first draft of the manuscript. All authors contributed to data interpretation, provided critical input and approved the final version of the manuscript.

    • Funding This work was supported by Australia’s National Health and Medical Research Council (NHMRC), Project Grant GNT 1087006.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.