Implementing women’s sexual and reproductive health interventions in prisons: a realist review

Abstract

Objectives

To explore what works for whom, how and why when implementing women’s sexual and reproductive health interventions in prisons to understand the barriers and facilitators to implementation and to generate recommendations for policymakers.

Design

Realist review using the Realist And Meta-narrative Evidence Synthesis: Evolving Standards guidelines.

Data sources

We systematically searched Ovid MEDLINE, Global Health, Embase, the Cumulative Index to Nursing and Allied Health Literature (CINAHL) and the American Psychological Association (APA) PsycINFO databases and hand-searched unpublished literature and reference lists, January–June 2025.

Eligibility criteria for selecting studies

Primary studies of implementing women’s sexual and/or reproductive health interventions, including those addressing sexually transmitted infections, cervical health, breast screening, contraception and women’s health holistically. Study populations included people in prisons that detain women in high-income countries.

Data extraction and synthesis

We extracted and analysed data relating to implementation processes using a grounded theory approach and retroductive inference to articulate cross-case Intervention-Context-Actor-Mechanism-Outcome configurations (ICAMOCs) and refine programme theory. We discussed findings in relation to existing theories from the literature to elicit recommendations for policymakers.

Results

Of 4617 deduplicated records, 26 met the inclusion criteria. Ten ICAMOCs were constructed from cross-case analyses, grouped into three themes: (1) planning (teaming, team leadership, assessing needs and capacity, tailoring and planning), (2) doing (piloting, standardisation and support, trauma-informed engagement and peer advocacy) and (3) sustaining (evaluation-adaptation cycles). The ICAMOCs indicated three overarching mechanisms as being key to effective implementation, namely, perceived utility of the intervention, motivation and empowerment.

Conclusions

For women’s sexual and reproductive health interventions to be effective in prisons, everyone involved in implementation needs to perceive the intervention’s benefit and be both motivated and empowered to take action. We recommend policymakers build a resilient and empowered delivery workforce, invest in research partnerships to increase awareness and understanding and promote trauma-informed approaches to women’s healthcare in prisons.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Realist synthesis is well suited to evaluating the implementation of interventions in complex real-world settings such as prisons.

• The inclusion of studies from unpublished literature may lower publication bias but may also lack rigour.

• Only studies from high-income countries published in English since 2010 were included. Furthermore, the predominance of North American studies (n=19) may limit generalisability to England.

• While pragmatism is advised in realist reviews to balance theoretical saturation with the availability of studies and time, further purposive sampling could have been undertaken.

Introduction

Rationale

It has been shown that people in prison are more likely to have complex needs and poorer health outcomes than the general population and that detention can have a harmful, ageing effect.¹,⁵ Further evidence suggests this disparity may be greater for people in prisons that detain women (PPDW), who are at higher risk of worse outcomes and premature mortality.⁶,⁸ For example, when compared with people in prisons that detain men, substance misuse was found to be more prevalent, and rates of self-harm, attempted suicide and mental health conditions were higher.⁵,⁷⁹ Further studies found higher prevalence of sexually transmitted infections (STIs), menstrual irregularities and amenorrhoea compared with the community and substantially higher rates of dysplasia and cervical cancer.^{910 12},¹⁶ There is also evidence that cancers are diagnosed later and have lower overall survival in prison settings than in the community.^{2 17 18}

The wider determinants of health and crime are largely comparable, and people detained in prisons for women often encounter multiple disadvantages throughout life.^{1 2 5 6} They are more likely than other people in prison and those in the community to experience poverty, housing insecurity, unemployment, sexual/physical violence, commercial sex work, trafficking and multiple sexual partners.^{1 6 9 10 12} In England, one-third spent time in care, two-thirds survived domestic abuse and ethnic minorities are over-represented.^{2 5 6} It has been suggested that women’s sexual and reproductive health is under-represented in research and policymaking and that existing services can be fragmented and difficult to access for marginalised groups.^{19 20} People are therefore more likely to enter prison with low rates of screening and healthcare engagement, perpetuating a high-risk/low-uptake anticorrelation.²²¹,²⁴ Such complex intersecting factors can lead to compound health inequalities and unmet needs for PPDW.^{5 25}

As sites of ‘dynamic exchange’²⁶ (p. 239) with marginalised groups, prisons can provide opportunities for people to engage with public health interventions at least equivalent to those available in the community.^{6 10 23 27} It has been suggested this could reduce health inequalities and reoffending, with downstream benefits for the families, communities and health services people in prison return to on release.^{1 2 26 28 29} England’s Chief Medical Officer⁵ recommends the prioritisation of preventive health interventions in prisons, and there is growing international support for improved women’s sexual and reproductive health provision.^{6 30} However, more research in this field is needed.^{6 21 31 32} In England, only 1% of approved prison-based research projects conducted in 2015–2024 addressed reproductive and/or perinatal health.⁵ Internationally, we found very few systematic and scoping reviews of relevance, and those available describe multiple service gaps and inadequacies, particularly for contraception.¹⁵³³,³⁸ A recent National Health Service (NHS) England review found that, despite some examples of good practice, women’s sexual and reproductive health is systemically underserved in prisons.⁶ Improvements in cervical screening uptake, for example, have been shown to plateau considerably below national coverage and targets.²⁴

In contrast, there is good evidence that community-based women’s sexual and reproductive health interventions are effective.³⁹,⁴³ However, the uptake and sustainability of complex interventions are highly context dependent, and implementation strategies used in other settings may not be transferable to prisons.^{44 45} Service reviews characterise England’s prison infrastructure as ‘ageing…designed for a young male population’⁵ (p. 10) and ‘unfit for purpose’⁶ (Section 2.10) for people detained in prisons for women, who constitute <5% of the population.⁴⁶ The evidence suggests that high demand for urgent care can make overburdened, under-resourced health services reactive not proactive and post-pandemic recovery of screening services has been slow.^{1 17 47} Short sentences and geographical displacement may also limit the timeframe for action and present challenges for follow-up.^{6 33} To bridge the evidence-practice gap, there is a need to identify the conditions required for effective and sustainable operationalisation of women’s sexual and reproductive health interventions in the real-world settings of prisons.^{44 45 47}

Rationale for realist synthesis

Complex public health interventions use strategies and resources to improve population health.^{48 49} However, in reality, they interact dynamically with the open social systems in which they are implemented, making outcomes uncertain.^{49 50} That is, multiple contextual factors determine the activation and intensity of multiple mechanisms, which determines the direction and magnitude of multiple outcomes.⁵¹,⁵³ So, an intervention’s impact may differ when implemented for, and by, different actors in a different time and space. This generative model of causality forms the philosophical basis for realist inquiry.⁴⁹

Where traditional systematic reviews focus on internal validity, realist methods are better suited to evaluating external validity by considering interventions within their delivery contexts to better understand facilitators and barriers along the implementation pathway and identify effective strategies to navigate them.⁴⁵⁴⁷,⁴⁹ This is achieved through ‘retroductive theorising’, working backwards from observed outcomes to infer underlying causal mechanisms.⁵³,⁵⁵ In realist synthesis, initial programme theories (IPT) provide high-level hypotheses of how interventions should work.^{54 56 57} Observations from primary studies are then used to form context-mechanism-outcome (CMO) configurations to test and refine the programme theory (PT).⁵⁶,⁵⁹ Middle-range theories can both inform IPT and explain the PT, increasing external validity (see table 1 for key terms).^{54 59} Adopting this approach, the objective of this review was to provide post hoc, theory-based insights for policymakers into what works for whom, how and why when implementing women’s sexual and reproductive health interventions in prisons.^{45 47}

Table 1. Key concepts in realist reviews.

Concept		Definition
Explanatory factors	Intervention (I)	Any resources, strategies or activities used in the design, implementation or evaluation of an intervention.105,107
	Context (C)	The particular set of circumstances in which the intervention is implemented.61 106
	Actor (A)	An individual, group or organisation involved in one or more aspects of the implementation.61 106
	Mechanism (M)	An unseen power, force, response, feedback or interaction with the potential to cause an outcome.51 106
	Outcome (O)	An intended or unintended consequence of the intervention.61
Theory	Initial programme theory (IPT)	Macro-level; a theory about how the intervention should work.56 57 108
	Programme theory (PT)	Micro-level; theories about how and why interventions work; the conditions needed (C) to trigger mechanisms (M) to produce outcomes (O).49 56 59
	Middle-range theory (MRT)	Macro-level; abstracted theories that explain the PT.54 59

Materials and methods

This review is reported according to Realist And Meta-narrative Evidence Synthesis: Evolving Standards (RAMESES),⁵⁸ guided by Pawson et al’s⁴⁹ five steps, with adaptations drawn from the realist literature.

Informal scoping of the literature by the lead author (JR), and discussion with the second author (GJMT), informed the IPT, the scope of the review and the search strategy.⁶⁰ Next, JR conducted in-depth searches for primary studies and extracted data into Intervention-Context-Actor-Mechanism-Outcome configurations (ICAMOCs)—here, intervention (I) and actor (A) were added as explanatory factors to the CMO heuristic to differentiate intervention from context and specify the individuals/groups/institutions involved (see table 1).⁵⁹⁶¹,⁶³ Finally, JR discusses the PT to provide recommendations for policymakers.

Scoping the literature

JR conducted informal, exploratory searches of the literature to (a) identify relevant national and international policies and guidance to inform both the women’s sexual and reproductive health domains for inclusion and the IPT development and (b) scope the availability and nomenclature of primary studies to inform the search strategy.

Women’s sexual and reproductive health domains

Women’s health hubs (WHH) have been championed as a model for improving access to women’s health services in the community, including those on probation or community supervision.^{5 19} Therefore, we used the core services defined in the WHH core specification⁶⁴ to identify women’s sexual and reproductive health domains for inclusion. We cross-referenced this with key policy and guidance documents to ensure relevance to England’s prison context.^{23 28 30 65 66} Pregnancy testing received multiple recommendations, so was added. Pessaries and preconception care were not mentioned, so were removed. Menopause was absent from prison recommendations but identified as an overlooked need in a recent NHS England review,⁶ so was retained. The final women’s sexual and reproductive health domains in scope for this review are shown in table 2.

Table 2. Women’s sexual and reproductive health domains (those excluded are shown in italics).

Women’s sexual and reproductive health domain	WHH64	NHS23	WHO28	NICE65	BASHH66	PRI30
STI/HIV (screening)	Yes	Yes	Yes	Yes	Yes	Yes
Cervical (screening)	Yes	Yes	Yes	Yes		Yes
Breast (screening)	Yes	Yes	Yes	Yes		Yes
Menstrual problems	Yes		Yes		Yes	Yes
Contraception	Yes		Yes		Yes
Pregnancy (testing)			Yes	Yes	Yes
Menopause	Yes
Preconception care	Yes
Pessary care	Yes

BASHH, British Association of Sexual Health and HIV; NHS, National Health Service; NICE, National Institute for Health and Care Excellence; PRI, Penal Reform International; STI, sexually transmitted infection; WHH, women’s health hubs.

Initial programme theory

The Consolidated Framework for Implementation Research (CFIR) is commonly used by implementation scientists to aid systematic analysis.⁶⁷ We therefore formed a conceptual framework based on the five CFIR contextual domains (outer setting, inner setting, intervention, implementation process and actors), plus constructs from the CFIR implementation process domain (teaming, assessing needs/capacity, planning, tailoring, implementing, engaging, evaluating and adapting), modifying terms where appropriate to reflect realist terminology (table 1) and the sensitivity of the review topic. We supplemented this with Proctor et al’s⁶⁸ implementation outcomes (appropriateness, acceptability, feasibility, adoption, integration/reach, fidelity, cost-effectiveness and sustainability) and contextual factors of relevance from the prison healthcare guidance documents identified in the initial scoping.^{6 23 28 30 65 66} This conceptual framework forms an IPT of how the implementation of women’s sexual and reproductive health interventions in prisons should work. We include a pictorial representation in online supplemental file 1.

Searching process

We developed and implemented the search strategy iteratively in January–June 2025, following realist principles.⁴⁹ This included published and unpublished literature and used online searching and ‘snowballing’ citation searching methods. JR performed systematic searches of peer-reviewed journals using Ovid MEDLINE, Global Health, Embase, the Cumulative Index to Nursing and Allied Health Literature (CINAHL) and the American Psychological Association (APA) PsycINFO databases to provide an appropriate spread of the indexed health research literature, including medical, public health and mental health research. Synonymous search terms identified through exploratory searches were aggregated, including multiple alternatives for prisons and people in prison (jail, detention, incarceration, remand, felon, etc) to reflect international and colloquial variations. Terms were combined with Boolean commands (AND/OR/NOT) and wildcard characters, using medical subject heading (MeSH) terms as appropriate (see online supplemental file 1 for full details). We limited searches to 2010 or later, as the United Nations adopted the ‘Bangkok Rules’³⁰ for women’s prison healthcare in 2010. JR used EndNote to manage and deduplicate search results and Mendeley Cite to manage citations.

Selection and appraisal of documents

Informal searches highlighted a scarcity of England-based intervention studies, so all high-income countries were included.⁶⁹ While significant heterogeneity was expected, limiting studies to high-income countries may increase generalisability to England’s prisons.

JR deduplicated results and screened titles and abstracts against inclusion/exclusion criteria (see online supplemental file 1). In brief, interventions implemented in high-income country prisons addressing at least one women’s sexual and reproductive health domain (table 2) and delivered during detention were included. Adult women, people who identify as female and people of diverse gender assigned to be detained in a prison for women, were included (referred to throughout as PPDW). Studies that did not describe or evaluate implementation components, factors or outcomes or were unavailable in English were excluded. JR then conducted full-text screening, another reviewer (see Contributors) rescreened a 10% random sample and discrepancies in decisions were discussed and resolved.

Aligning with realist methods,⁴⁹ JR created a matrix (see table 3) to appraise records for relevance, richness and rigour, using a high-medium-low scale based on concepts described by Dada et al.⁷⁰

Table 3. Appraisal criteria, based on concepts described by Dada et al⁷⁰.

Appraisal	High	Medium	Low
Relevance	Direct focus and/or good specificity	Single issue and/or some specificity	Indirect issue and/or little specificity
Richness	Rich description and/or explicit theory	Adequate description and/or little explicit theory	Weak description and/or no explicit theory
Rigour	Credible source and/or transparent	Potential bias and/or somewhat transparent	Clear bias and/or poor transparency

Data extraction

JR created a Microsoft Excel matrix to capture summary information about included studies, comprising title, author, publication year, country, study type, gender, women’s sexual and reproductive health domain(s), brief description and appraisal rating. A second matrix captured Intervention-Context-Actor-Mechanism-Outcome configurations (ICAMOCs). Data extraction and coding were managed using Delve Tool software.⁷¹

Data relating to implementation processes were extracted, but those relating to study processes were not. For example, eligibility identification for interventions was, but selection processes for trial participants were not. As England’s prison health services are free-at-point-of-care, factors relating to user fees were not extracted, although any wider implications were considered.

Analysis and synthesis process

JR analysed studies adopting a grounded theory approach. Potential explanatory factors were identified inductively through open coding, guided by—but not limited to—the IPT.⁷² Any convergence or divergence of outcomes was noted, and interventions reported in multiple studies were cross-referenced. Axial coding was then used to group and refine explanatory factors into case-specific ICAMOCs from which cross-case ICAMOCs were inductively and abductively inferred, and deductively tested, in an iterative process. Thus, working retroductively, causal relationships were articulated and discussed in relation to middle-range theories and the IPT to derive insights.

Patient and public involvement

No patients or members of the public were involved in the design, conduct, reporting or dissemination of this research.

Results

Document flow diagram

Systematic searches identified 4637 deduplicated records, of which 4554 did not meet inclusion criteria. After full-text screening, a further 57 records were excluded, and 26 were included in this review. Figure 1 presents a Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of the searching and selection process.

Figure 1. Flow diagram of search results and the selection of records.

Study characteristics

Included studies represented five high-income countries: 17 from the USA,⁷³,⁸⁹ three from Australia,⁹⁰,⁹² three from England,⁹³,⁹⁵ two from Canada^{96 97} and one from Ireland.⁹⁸ There were 12 evaluations,7374 76 78 79 85 87 90 92,⁹⁵ seven case studies,^{75 77 83 86 88 89 91} four quantitative,⁸⁰,⁸²⁸⁴ two qualitative^{97 98} and one mixed methods.⁹⁶ Overall, studies were of good quality; none were rated low against all criteria, while two were rated high against all criteria. Table 4 summarises the study characteristics.

Table 4. Characteristics of included documents.

Author(s), title, year of publication	Study design	Health domain, country, gender	Intervention overview	Appraisal Rel-Ric-Rig
Against Violence and Abuse. An evaluation of Women in Prison’s Health Matters Project. 2019.93	Evaluation	Women’s health, England, F	Women’s health advocacy and health promotion programme.	H-H-L
Besney, et al Addressing Women’s Unmet Health Care Needs in a Canadian Remand Center: Catalyst for Improved Health? 2018.96	Mixed methods	Women’s health, Canada, F	Women’s health programme including testing, education and care.	H-M-H
Cocoros, et al Screening for Hepatitis C as a Prevention Enhancement (SHAPE) for HIV: An Integration Pilot Initiative in a Massachusetts County Correctional Facility. 2014.81	Quantitative	STI (HCV), USA, M/F	SHAPE HCV screening and education programme (adapted from an existing HIV intervention).	M-M-M
Cole, et al Opt-out screening for Chlamydia trachomatis and Neisseria gonorrhoeae in female detainees at Cook County jail in Cole et al (2014).82	Quantitative	STI, USA, F	Universal, opt-out STI screening programme on reception to prison.	H-M-H
Crowley, et al Hepatitis C virus screening and treatment in Irish prisons from nurse managers’ perspectives - a qualitative exploration. 2019.98	Qualitative	STI (HCV), Ireland, M/F	Barriers/facilitators to implementing HCV screening programmes.	L-H-H
Dang, et al Paired Testing of Sexually Transmitted Infections with Urine Pregnancy Tests in Incarcerated Women. 2021.80	Quantitative	STI, pregnancy, USA, F	Opt-out programme pairing STI screening with routine urine pregnancy testing.	H-L-H
Emerson, et al Barriers and facilitators of implementing a collaborative HPV vaccine program in an incarcerated population: A case study. 2020.77	Case study	STI (HPV), USA, M/F	Opt-in HPV vaccination programme in a prison population.	M-H-M
Fasula, et al Project POWER: adapting an evidence-based HIV/STI prevention intervention for incarcerated women. 2013.79	Evaluation	STI (inc. HIV), USA, F	Project POWER sexual health education programme (adapted from SAFE).	H-H-M
Harmon, et al Routine screening in a California jail: effect of local policy on identification of syphilis in a high-incidence area, 2016–2017. 2020.83	Case study	STI (syphilis), USA, F/M	Syphilis screening programme for targeted age groups.	M-M-H
Hesse, et al Cancer screening in prisons: lessons for health providers. 2022.91	Case study	Breast screening, Australia, M/F	Nurse-led programme to facilitate breast screening uptake in four prisons.	M-L-L
Holton, Hampton. Implementing High-Intensity, Trauma-Informed Sexual Risk Reduction in Women Involved in the Criminal Legal System. 2023.84	Quantitative	STI, USA, F	SSE high-intensity trauma-informed educational programme to reduce STI risk.	H-L-H
Johnson, et al Feasibility of an HIV/STI Risk-Reduction Program for Incarcerated Women Who Have Experienced Interpersonal Violence. 2014.85	Evaluation	STI (inc. HIV), USA, F	Women’s CoOp STI/HIV prevention programme for prisons (adapted from an existing intervention).	H-H-H
Kelly, et al Developing a Cancer Prevention Health Education Resource: A Primer of Process and Evaluation. 2020.76	Evaluation	Cancer screening, USA, F	Cervical and breast cancer education resource development.	L-H-M
Kennedy, et al Case example of a jail-based cancer prevention clinical trial: Social determinants of health framework, novel experimental design, and retention strategies to facilitate long-term follow-up of clinical trial participants. 2023.75	Case study	Cervical screening, USA, F	SHE cervical cancer education programme. Post-detention elements were excluded.	M-M-H
McNeely, et al Expanding Contraceptive Access for Women with Substance Use Disorders: Partnerships McNeely et al (2019).87	Evaluation	Contraception, USA, F	Contraceptive programme including prison-based education and prison- or community-based LARC fitting.	H-M-M
Mongale, et al Development and optimisation of a reception testing protocol designed to eliminate HCV in the UK prison population. 2024.95	Evaluation	STI (HCV), England, M/F	HCV screening protocol.	M-M-M
Murphy, et al Implementing Preexposure Prophylaxis for HIV Murphy et al (2024).86	Case study	STI (HIV), USA, M/F	Pre-exposure prophylaxis programme for HIV prevention.	M-M-M
Pankow, et al Evaluating Fidelity to a Modified NIATx Process Improvement Pankow et al (2018).74	Evaluation	STI (HIV), USA, not stated	HIV-STIC protocol for implementing HIV education, testing and care interventions (adapted from NIATx).	H-H-H
Paynter, et al The experiences of family planning health professionals providing care to incarcerated patients: A qualitative study. 2025.97	Qualitative	Contraception, Canada, F	Barriers/facilitators to contraceptive services for PPDW. Elements relating to abortion were excluded.	L-H-H
Phaw, et al Prospective evaluation of the impact of repeated whole prison testing for hepatitis C. 2025.94	Evaluation	STI (inc. HCV), England, F	High-intensity test and treat for HCV and other STIs in a whole prison population.	M-L-M
Ramaswamy, et al The development of a brief jail-based cervical health promotion intervention. 2015.78	Evaluation	Cervical screening, USA, F	SHE cervical cancer education programme development.	M-H-H
Sufrin, et al Long-Acting Reversible Contraceptives for Incarcerated Women: Feasibility and Safety of On-Site Provision. 2015.88	Case study	Contraception, USA, F	Contraceptive programme including counselling and LARC fitting.	M-M-H
Sufrin, et al Family planning services for incarcerated women: models for filling an unmet need. 2017.89	Case study	Contraception, USA, F	Contraceptive programmes, including family planning, in 4 women’s prisons.	L-M-L
Thornton, et al The New Mexico Peer Education Project: Thornton et al (2018).73	Evaluation	STI (HCV), USA, M/F	NMPEP peer education programme for HCV risk reduction (adapted from Wall Talk).	M-M-M
Wallis, et al Hepatitis C virus point-of-care RNA testing: experience from screening an entire high-security Australian prison population over 3 days. 2023.92	Evaluation	STI (HCV), Australia, F	High-intensity HCV test and treat programme in a whole prison population.	M-L-L
Winter, et al A nurse‐led intervention improved blood‐borne virus testing and vaccination in Winter et al (2016).90	Evaluation	STI, Australia, M/F	Nurse-led programme to improve the uptake of BBV/STI testing.	M-L-M

BBV, blood-borne viruses; HCV, hepatitis C virus; HIV-STIC, HIV Services and Treatment Implementation in Corrections; LARC, long-acting reversable contraceptive; NIATx, Network for the Improvement of Addiction Treatment; NMPEP, Prisoner Health is Community Health: New Mexico Peer Education Project; Rel-Ric-Rig, relevance, richness, rigour; SAFE, Sexual Awareness for Everyone; SHAPE, Screening for Hepatitis C as a Prevention Enhancement; SHE, Sexual Health Empowerment; SSE, Safer Sex Education; STI, sexually transmitted infection.

Regarding women’s sexual and reproductive health domains, 16 studies addressed STIs, six of which specified hepatitis C virus (HCV), four addressed cervical and/or breast screening (double-counting the Sexual Health Empowerment intervention),^{75 78} four contraception (double-counting a San Francisco jail intervention)^{88 89} and two addressed women’s health holistically. None specified menopause, menstruation or pregnancy testing, although Dang et al’s⁸⁰ STI intervention was paired with pregnancy testing. The majority of studies (n=22) evaluated programmatic interventions, and two evaluated protocol interventions. The remaining studies did not evaluate implementation but provided rich insights into relevant contexts and mechanisms, so were included.^{97 98}

Main findings

The analysis elicited ten cross-case ICAMOCs across the three IPT implementation process domains: planning, doing and sustaining. ICAMOCs are described narratively below, pictorially in figure 2, and summarised in table 5 with illustrative extracts from the analysis. Explanatory factors are denoted throughout by unique identifiers in the format (X_y), for example, the context ‘organisational culture’ is (C₂).

Figure 2. ICAMO configurations 1–10. ICAMO, Intervention-Context-Actor-Mechanism-Outcome; ICAMOC, Intervention-Context-Actor-Mechanism-Outcome configuration; PPDW, people in prisons that detain women.

Table 5. Ten ICAMOCs relating to the implementation of women’s sexual and reproductive health interventions in prisons.

		Intervention	Context	Actor	Mechanism	Outcome
Planning: teaming	ICAMOC1	I1 Sharing information to facilitate teaming	C1 Knowledge and skills C2 Organisational culture C3 Workforce capacity C4 Competing priorities C5 Values, beliefs, concerns	A1 Implementation teams A2 Decision-makers A3 Prison staff	M1 Perceived needs (of PPDW) M2 Perceived risks of not intervening M3 Perceived utility of intervention	O1 Adoption (buy-in) O2 Appropriateness (perceived suitability, usefulness)
		Illustrative extract: Various administrators and staff initially expressed resistance to providing contraception during incarceration because of the potential suggestion that women were at risk for pregnancy while in custody and because they believed ‘It is not our problem’. Opinion changed when clinicians presented data demonstrating that over half the pregnant women at the facility with prior incarcerations had conceived within 3 months of release (… and)discussed different birth control methods, [and]) post-release challenges for women in accessing family planning services.89 (p. 13)
	ICAMOC2	I2 Effective team leadership	C3 Workforce capacity C4 Competing priorities C5 Values, beliefs, concerns C6 Existing relationships C7 Communication	A1 Implementation teams	M4 Empowerment M5 Shared sense of purpose M6 Agency (personal, collective) M7 Clarity, reduced uncertainty	O1 Adoption (buy-in)     O3 Feasibility
		Illustrative extract: (…)undertaking inter-organisational change with the modified NIATx approach required consistent, meaningful communication between leaders and team members. Change team leadership communication issues were linked to a host of problems, such as scheduling team meetings, delayed change cycles and even a lack of understanding among team members about the HIV-STIC process improvement strategy itself.74 (pp. 198–199)
Planning: assessing	ICAMOC3	I3 Assessing needs and context through formative research	C1 Knowledge and skills C3 Workforce capacity C5 Values, beliefs, concerns C8 PPDW with complex needs C9 Availability of resources C10 Prison security and regime C11 Existing services C12 National recommendations	A1 Implementation teams A2 Decision-makers A4 PPDW A5 Deliverers	M1 Perceived needs (of PPDW /deliverers) M7 Clarity, reduced uncertainty	O2 Appropriateness (compatibility, relevance) O3 Feasibility O4 Acceptability (satisfaction with content)
		Illustrative extract: (…)the principal investigator (PI) approached members of the NCDOC and prison administration to assess their capacity and interest in the project(… and)formative research was conducted to gather specific information for adapting Project SAFE for women prisoners.79 (pp. 205–206)
Planning: tailoring, planning	ICAMOC4	I4 Tailoring evidence-based interventions to meet the needs of PPDW	C1 Knowledge and skills C5 Values, beliefs, concerns C8 PPDW with complex needs C10 Prison security and regime	A4 PPDW	M4 Empowerment M6 Agency (personal) M8 Motivation M9 Sense of identification (with content)	O2 Appropriateness (suitability) O3 Feasibility O4 Acceptability (satisfaction with content, credibility)
		Illustrative extract: Like most educational programmes, we felt that a cervical health promotion intervention for incarcerated women would have to be tailored specifically to the women’s educational, social and cultural backgrounds—balancing delivery of information and empowerment for health behaviour change against the real structural constraints of their criminal justice involvement.78 (p. 433)
	ICAMOC5	I5 Planning integrated, universal services	C1 Knowledge and skills C3 Workforce capacity C4 Competing priorities C5 Values, beliefs, concerns C9 Availability of resources C10 Prison security and regime C11 Existing services C13 Short sentences	A4 PPDW A5 Deliverers	M7 Clarity, reduced uncertainty M10 Nudging M11 Normalisation	O3 Feasibility O5 Integration and reach O6 Cost-effectiveness
		Illustrative extract: (…)paired screening of GC/CT with routine urine pregnancy test leads to an increased number of completed tests and a higher detection of GC and CT infections that would have otherwise been missed. Compared with testing rates before paired testing where GC/CT tests were ordered based on clinical suspicion, implementation of paired testing increased monthly testing rates by 4.7-fold.80 (pp. S21–22)
Doing: implementing	ICAMOC6	I6 Piloting and testing implementation	C3 Workforce capacity C9 Availability of resources	A1 Implementation teams A2 Decision-makers A4 PPDW A5 Deliverers	M3 Perceived utility of intervention M7 Clarity, reduced uncertainty	O1 Adoption (intention to implement) O3 Feasibility O4 Acceptability (of content) O5 Integration and reach (integration) O7 Fidelity (delivery as intended) O8 Sustainability
		Illustrative extract: As the trial went on, we learnt ways to facilitate condom skills practice in the prisons that increased participants’ level of comfort, including using humour, having the interventionists model the skills in group before having women try them, and adjusting the timing of the condom skills practice so the practices did not occur after discussing violence or while practising grounding skills, as this was triggering for some women. By the end of the trial, the condom skills practice in the prison facilities was smooth, feasible, acceptable and perceived as helpful by participants.85 (p. 3259)
	ICAMOC7	I7 Standardising and supporting implementation	C1 Knowledge and skills C3 Workforce capacity C8 PPDW with complex needs C10 Prison security and regime C11 Existing services C14 Intervention complexity	A5 Deliverers	M7 Clarity, reduced uncertainty M12 Increased self-efficacy	O1 Adoption (intention to implement) O6 Cost-effectiveness O7 Fidelity (delivery as intended)
		Illustrative extract: (…)no additional human resources were required; instead, the current staff who were already trained in STI screening were trained in a new protocol and script.82 (p. 162)
Doing: engaging	ICAMOC8	I8 Trauma-informed engagement	C5 Values, beliefs, concerns C6 Existing relationships C8 PPDW with complex needs C10 Prison security and regime	A4 PPDW	M4 Empowerment M6 Agency (personal) M8 Motivation M13 Feeling heard, understood, supported M14 Increased self-esteem M15 Trust	O4 Acceptability (of delivery) O5 Integration and reach
		Illustrative extract: Participants were allowed to form their own groups made up of people with whom they felt comfortable. Ground rules were established at the beginning of each session. The ground rules included being respectful, listening, allowing the space to be free from judgement, maintaining confidentiality and having the freedom to leave the room if needed.84 (p. 390)
	ICAMOC9	I9 Engaging via peer advocates	C3 Workforce capacity C5 Values, beliefs, concerns C6 Existing relationships C8 PPDW with complex needs C10 Prison security and regime	A4 PPDW	M9 Sense of identification (with peers) M13 Feeling heard, understood, supported M15 Trust	O4 Acceptability (of delivery) O5 Integration and reach
		Illustrative extract: Peer support provided by The Hepatitis C Trust which was vital for ensuring treatment adherence and following patients on treatment either to new prisons after transfer, or into the community after release. The peers have a lived experience of HCV and incarceration, meaning they are more trusted by, and can better communicate with, residents.95 (p. 7)
Sustaining: evaluating, adapting	ICAMOC10	I10 Evaluation-adaptation cycles	C3 Workforce capacity C9 Availability of resources C13 Short sentences	A1 Implementation teams	M1 Perceived needs (of PPDW /deliverers) M3 Perceived utility of intervention M7 Clarity, reduced uncertainty	O2 Appropriateness O7 Fidelity O8 Sustainability
		Illustrative extract: Ongoing monitoring can assist in the identification of time points or phases where fidelity lapses occur and result in efforts to recalibrate team performance. In HIV-STIC, fidelity monitoring served to bring problems to the attention of researchers, coaches and teams, enabling corrective action.74 (p. 199)

GC/CT, Neisseria gonorrhoeae and Chlamydia trachomatis; HCV, hepatitis C virus; HIV-STIC, HIV Services and Treatment Implementation in Corrections; ICAMOC, Intervention-Context-Actor-Mechanism-Outcomes configuration; NCDOC, North Carolina Department of Corrections; NIATx, Network for the Improvement of Addiction Treatment; PPDW, people in prisons that detain women; STI, sexually transmitted infection.

Planning

Teaming

Prisons were challenging settings to implement interventions. Prison regimes (ie, routines and activities), regulations and lockdowns were restrictive, disruptive and unpredictable (C₁₀).^{74 77 85 89 93 98} Recruitment challenges and competing priorities constrained workforce capacity, and facilities, equipment and funding were limited and well-guarded (C_3,4,9).^{74 77 82 93} Accessing resources required teaming (I_1,2), that is, the collaborative effort of multiple stakeholders (hereafter ‘implementation teams’) (A₁) with the required ‘expertise and authority’⁷⁴ (p. 189), but whose priorities may diverge, like the implicit tension of care versus control between healthcare and prison staff.7577 82 83 89 93,⁹⁵

ICAMOC₁—sharing information to facilitate teaming

Analysis suggests that teaming enables implementation. This requires actors to buy-in (O₁), which is influenced by prior knowledge, organisational culture, personal values, capacity and competing priorities (C_1,2,5,3,4). If information is shared (I₁) to engage implementation teams and decision-makers (A_1,2), then buy-in is more likely, because it increases their perception of PPDW’s health needs, the risks of inaction and the relative advantage of intervention over the status quo (M_1,2,3).^{74 77 86 89 93} Information sharing throughout implementation maintains and expands buy-in, facilitating adoption and appropriateness (O_1,2), and should include prison staff (A₃) as potential facilitators of access.^{78 84 86 92 93 95 98}

ICAMOC₂—effective team leadership

If implementation leaders provide a clear vision and practical direction (I₂) for team members (A₁), then the latter are more likely to prioritise implementation alongside other commitments (C₄), increasing feasibility and adoption (O_3,1), because they are empowered by a shared sense of purpose, agency and clarity of role (M_4,5,6,7).^{74 77 86 94} Effective leadership is moderated by the regularity of team meetings, quality and frequency of communication and the leader’s level of commitment (C_7,3,5).^{74 76 77 93} Leadership can be formal or informal and is facilitated by existing relationships and influence (C₆).^{74 75 78 81}

Assessing needs and capacity

While teaming overcame some barriers, a prison’s capacity for implementation remained a significant influencing factor. Analysis indicates that implementation strategies to add, modify or leverage resources benefitted from prior assessment of needs and potential capacity (I₃).^{74 77} Resulting outputs were more appropriate, acceptable and feasible (O_2,4,3) for PPDW, people delivering the intervention (hereafter ‘deliverers’) and decision-makers (A_4,5,2) alike. Studies used various formative research methods, including focus groups, surveys, site visits, stakeholder consultations and data analysis.^{74 75 78 79 85 95}

ICAMO₃—assessing needs and context through formative research

If formative research is carried out (I₃) to assess the needs (C_1,3,5,8) of PPDW and deliverers (A_4,5) and the context for implementation (C_3,9,10,11), then implementation teams (A₁) can tailor interventions and plan implementation to be appropriate, acceptable and feasible (O_2,4,3), because they perceive the needs and preferences of those involved and uncertainty about capacity is reduced (M_1,7). In studies, the needs of PPDW (A₄) included learning and communication, neurodiversity, literacy, language and inclusivity alongside relevant values, beliefs and behaviours (C_1,5,8).^{75 79 84 93} Interventions such as SHE^{75 78} assessed the needs of PPDW holistically within wider determinants (C₈) to good effect.^{76 79} Capacity factors included facilities, equipment, digital resources and information systems and the limitations of security protocols and prison regimes (C_9,10).^{77 79 83 93 98} Workforce capacity influenced implementation, encompassing skills/specialisms, training needs, time, engagement and concerns such as fearing needlestick injury (C_1,3,5).^{73 74 77 78 83 84 93 98} Heterogeneity between prisons and non-adherence to national recommendations was observed (C_11,12).^{82 83 90 93 98}

Tailoring and planning

Building on formative research to tailor interventions and plan implementation increased appropriateness, acceptability and feasibility (O_2,4,3). Studies found that, due to PPDW’s (A₄) experiences of trauma, stigma and lack of autonomy (C_8,5,10), designing empowering interventions that enabled informed decision-making motivated health-promoting behaviours (M_4,6,8).^{78 79 85 87 88 93} Furthermore, selecting safe, streamlined, timely innovations/programmes increased access to results/care, as many PPDW serve short sentences (C₁₃).^{75 92 94 95 98}

ICOMOC₄—tailoring evidence-based interventions to meet the needs of people in prisons that detain women

If evidence-based frameworks and interventions are adopted/adapted (I₄)⁷³,^{7985 86} to design interventions which meet the needs (C_1,5,8,10) of PPDW (A₄), then implementation will be more acceptable and appropriate (O_4,2), because PPDW identify with the content and feel empowered and able to make health-promoting decisions (M_9,4,6,8). Engaging PPDW in tailoring activities was found to assist this process.^{78 85 87}

ICOMOC₅—planning integrated, universal services

Rapid, intensive interventions can be resource-heavy and lack sustained impact.^{92 94} Conversely, if strategies integrate interventions (I₅) into existing processes (C₁₁), then feasibility, cost-effectiveness and engagement (O_3,6,5) improve, particularly when nudged (M₁₀) by universal, opt-out offers.⁸⁰,8486 94 95 98 Furthermore, universal offers normalise (M₁₁) screening for PPDW (A₄) by reducing stigma and increasing exposure on recidivism (C_5,13) and for deliverers (A₅) by reducing the burden of risk assessment (contingent on time, knowledge, priorities and preconceptions (C_3,1,4,5)), decreasing uncertainty (M₇) and increasing asymptomatic disease detection.⁸⁰,^{8294 98} Analysis suggests integration with reception screening increases access, befitting short sentences and prison regimes (C_11,13,10), but potentially overwhelms PPDW (A₄) with competing priorities and procedures (C₄).7883 84 86 87 93,^{95 98} Strategies using in-reach, telemedicine or single-point-of-access models may reduce structural barriers but require careful planning.^{91 96 98} While some implementations leverage existing skills and assets (C_1,3,9),^{73 93 95} sufficient time for workforce recruitment, training, information system development and funding applications is required.^{78 79 84 85 87 91 94}

Doing

Implementing

Even when implementation had adequate planning and stakeholder support, prisons remained unpredictable, complex systems for action.^{74 79}

ICAMOC₆—piloting and testing implementation

Given resource constraints and workforce capacity (C_9,3), analysis suggests that if interventions are initially piloted (I₆), then implementation teams, deliverers, decision-makers and PPDW (A_1,5,2,4) can perceive its utility and gain clarity about its effectiveness (M_3,7), informing iterative refinement to ensure feasibility and acceptability, thus enabling fidelity and wider adoption (O_3,4,7,1).⁷⁴,^{7679 84 85} In the studies, pilots ranged from several days⁷⁵ to several months,⁷⁴ depending on intervention type. After piloting, incremental adoption improved integration and sustainability (O_5,8), reducing the confusion (M₇) caused by multiple short-lived interventions.^{88 92 93}

ICAMOC₇—standardising and supporting implementation

If introductory training and standardised operating procedures (SOPs) (I₇) are provided to deliverers (A₅), then adoption, fidelity and cost-effectiveness improve (O_1,7,6), within workforce capacity and skill constraints, and particularly when interventions are complex (C_3,1,14), because they gain clarity about the process, reducing uncertainty and improving self-efficacy (M_7,12).^{74 75 77 82 86} This is moderated by ongoing technical, clinical, administrative and emotional support, such as helping peers cope with difficult conversations or providing guidance on data utilisation.^{74 85 93 95} Analysis suggests that SOPs consider the whole intervention pathway to aid responsiveness, including linkages to care, escalation/referral routes and contingency plans (C_11,8,10).^{74 75 79 83}

Engaging

While the implementation processes discussed thus far highlight the overall importance of engagement, this section focuses specifically on what works to engage PPDW in interventions.

Studies concurred that for PPDW (A₄), lifetime trauma was common and imprisonment was distressing and disempowering (C_8,10).^{75 84 93} Previous negative experiences fostered deep-seated mistrust of healthcare deliverers, and fear of stigma, judgement, tests and disease were common (C_5,6).7578 86 93 96,⁹⁸ PPDW often felt misunderstood and disinclined to seek help or disclose personal information (M_13,8).^{93 97} Fasula et al⁷⁹ highlight the correlation between a PPDW’s self-esteem and their motivation to adopt health-promoting behaviours (M_14,8).

ICAMOC₈—trauma-informed engagement

If deliverers (A₅) adopt empathetic, trauma-informed approaches throughout the intervention pathway (I₈), then PPDW (A₄) engage more (O₅), because they feel heard, understood and supported, which builds self-esteem, agency and trust, increasing empowerment and motivation (M_{13,14,6,15,4,8}) and improving acceptability (O₄).^{75 78 79 84 85 93 96 97} Trauma-informed methods facilitate informed decision-making through affirmations, tailored education, information-sharing, signposting, expectation-setting and offering choices.^{78 85 87 88 93} Group modalities promote emotional safety, decreasing power differentials and fostering social support.^{75 84 85} Flyers, posters, workshops and word-of-mouth can raise awareness, and verbal communication helps overcome literacy barriers.^{73 75 78 88} Optimising coverage should be weighed against giving sufficient time for decision-making to avoid coercion, particularly for long-acting contraception.^{88 89}

ICAMOC₉—engaging via peer advocates

If advocacy, particularly by peers (I₉), is provided, then acceptability, integration and reach increase (O_4,5), because PPDW (A₄) identify with and trust them (M_9,15) and because increased opportunities to engage within the prison regime (eg, housing blocks, reception and post-release) (C₁₀) help them feel heard and supported (M₁₃).^{73 93 95 98} Analysis suggests that, with appropriate training and support, peers find advocacy roles affirming and empowering, becoming champions for the intervention and augmenting workforce capacity (C₃).^{73 93}

Sustaining

Evaluating and adapting

Although the CFIR distinguishes between reflection and evaluation, characterised by the collection and analysis of information and adaptation of the intervention or context, they are combined here as study data support adopting an ongoing, iterative process for improved sustainability and relevance.

ICAMOC₁₀—evaluation-adaptation cycles

If evaluation-adaptation cycles are employed (I₁₀), based on relevant information and feedback,^{74 79} then appropriateness, fidelity and sustainability improve (O_2,7,8), because implementation teams (A₁) gain clarity over the utility of the intervention and whether implementation meets the needs of deliverers and recipients (M_7,3,1) within the context of available resources (C_3,9).^{74 79 83 85} Studies sought feedback from various stakeholders, including peers, deliverers and PPDW, although short sentences and abrupt cessation (C_13,9) of funding sometimes hindered collection.^{76 78 79 81 87 93} Focus groups, surveys, information systems and evaluation tools were used, but data availability and its effective utilisation were limited and the importance of impartiality and transparency were stressed.^{74 77 79 83 85 87 90 93 95} Pankow et al⁷⁴ recommend ongoing process evaluation to ensure fidelity (O₇), while elsewhere positive feedback was noted as rewarding for deliverers.⁹³

Refined programme theory

Explanatory factors from cross-case ICAMOCs were used to refine the PT (see figure 3). The mechanisms identified were grouped into three higher level mechanisms found to be relevant to all actors, namely, perceived utility of intervention, motivation to implement and empowerment to take action.

Figure 3. Refined programme theory. PPDW, people in prisons that detain women.

Discussion

Summary of findings

We identified three overarching themes. First, workforce capacity is a significant constraining factor when implementing women’s sexual and reproductive health interventions in prisons. Recruitment and retention are ongoing issues for prison staff and deliverers, for whom core duties are reactive and challenging. Deliverers can feel uncertain and lack time and resources to manage competing priorities effectively. Analysis indicates effective leadership, training, ongoing support and SOPs enable deliverers to perceive the utility of implementing interventions and gain clarity about their role. Universal and opt-out offers reduce reliance on deliverers’ time and inclination to risk assess and on a PPDW’s willingness to disclose medical or behavioural history, thereby increasing coverage. Mobilising peers from the prison population and/or external organisations can enhance capacity, providing an affirming experience for the peer when adequately trained and supported.^{73 93} Actor types involved in planning should seek to empower the workforce through training, support and augmentation to make implementation more feasible and acceptable.

Second, the persistence of competing priorities hinders engagement across all actor types. Decision-makers and prison staff work within strict protocols to safeguard the workforce and PPDW, prioritising security. Healthcare deliverers necessarily prioritise high-volume urgent medical needs within resource constraints. PPDW do not often prioritise their sexual and reproductive health, particularly at reception when coping with, for example, mother-child separation, substance withdrawal and situational distress—as Hilkey postulates, Maslow’s theory of human motivation places such pressing physiological and psychological needs above health-promotion in prisons.⁹⁹ This is evident in the dominance of HCV and HIV studies in this review which, while included here as STIs, may be prioritised over other women’s sexual and reproductive health issues (like menstruation or menopause) due to their association with intravenous drug use. Education and information-sharing to increase understanding of women’s sexual and reproductive health needs and associated risks enable actors to perceive the utility of intervention, increasing their motivation for intervention. This aligns with the concepts of perceived susceptibility, severity and benefits central to the Health Belief Model.¹⁰⁰ Sustained, meaningful efforts to raise awareness about women’s sexual and reproductive health in prisons are needed. Several studies demonstrated that long-term research partnerships can facilitate this and help to expand the evidence base.^{73 74 78}

Lastly, adopting a trauma-informed approach throughout implementation activities is an important enabling factor. PPDW often have significant lifetime trauma resulting in low self-esteem, low self-efficacy and pervading fear and mistrust. People who use drugs and sex workers can feel stigmatised and misunderstood, and prisons fundamentally and profoundly restrict autonomy. The resulting sense of disempowerment is worsened when deliverers seem judgemental, disinterested or dismissive. In contrast, empathetic interactions in which PPDW feel emotionally safe to talk openly and feel heard and understood are empowering and normalising. Providing education and opportunities for PPDW to make informed, uncoerced decisions about their health aligns with the self-determination theory of motivation.¹⁰¹ Furthermore, group activities which promote respect, equality and acceptance offer much-needed opportunities for PPDW to build social support. Advocate roles, including peer-based programmes, can provide trustworthy support, bridging the gap between deliverer and recipient, and should be considered in all prison settings. There is increasing evidence to support trauma-informed care in prisons, and on-going investment in workforce training and the adoption of trauma-informed care throughout the prison estate, or network, is recommended.¹⁰²,¹⁰⁴

Conclusions and recommendations

The Chief Medical Officer’s 2025 annual report on the physical health of people in England’s prisons recommends prioritising health prevention interventions to tackle inequities within the system and improve health outcomes.⁵ Recognising the complex challenges of implementation in custodial settings and the unmet needs particular to PPDW, the objectives of this review were to explore what works for whom, how and why, when implementing women’s sexual and reproductive health interventions in prisons. This involved exploring the barriers and facilitators to implementation described in primary studies to generate recommendations for policymakers. The CFIR informed the IPT, which was tested and refined through realist synthesis to refine an explanatory PT (figure 3). Three main mechanisms of perceived utility of intervention, motivation and empowerment were identified and informed the following recommendations for policymakers:

1. Build a resilient, empowered delivery workforce through training, support and augmentation.

2. Invest in research partnerships to increase awareness and understanding of women’s sexual and reproductive health in prisons and address the evidence gap.

3. Promote trauma-informed methods to develop and deliver women’s sexual and reproductive health services throughout the prison estate.

Strengths and limitations

A strength of this review is that it synthesises evidence from a range of sources to theorise what works when implementing women’s health interventions in prisons. This addresses a significant gap in the literature and could form the basis for further research through case studies or as new studies emerge. While including unpublished literature may lower publication bias, only studies from high-income countries published in English since 2010 were included, which was a limitation of the search strategy. Furthermore, the predominance of North American studies (n=19) may limit generalisability to England, particularly given policy differences in criminal justice and healthcare systems.

None of the studies focused specifically on menopause, menstruation or pregnancy testing, despite searches including these terms. In addition, use of the WHH specification and prison policy/guidance to define women’s sexual and reproductive health domains may have excluded relevant alternatives, such as abortion. The search strategy could have been further strengthened by including specific search terms for non-binary and gender diverse participants. While Pawson et al⁵⁷ advise pragmatism to balance theoretical saturation with the availability of studies and time, further purposive sampling could have been undertaken. These limitations could have been addressed, and the search strategy strengthened, through stakeholder or advisory group involvement.

Data availability statement

Data are available upon reasonable request.