Summary of study characteristics
| Citation number (year) | Country/ countries | Study design | Setting characteristics | Sample characteristics, data collection method and recruitment strategy* | Data analysis | Quality assessment |
| 41 (2005) | Bangladesh | Qualitative, descriptive | Two sites; hospital-based; NGO and government run | In depth interviews (n=28) of OSC stakeholders (government, NGO, employees and women survivors) Purposive, snowball sampling | Thematic analysis | Medium |
| 26 (2006) | Bangladesh | Mixed methods, cross-sectional survey | One site; hospital-based; NGO and government run | Survey (n=310) of women treated at OSC, as identified by medical chart review Purposive sampling | Descriptive analysis | Low |
| 43 (2013) | Malaysia | Qualitative, descriptive | Seven sites; hospital-based; NGO and government run | In depth interviews (n=54) of OSC healthcare workers (including nurses, medical officers, gynaecologists, medical social workers and hospital managers) Snowball sampling | Thematic analysis | High |
| 48 (2016) | Nepal | Qualitative, descriptive | One site; hospital-based | FGDs (n=117) of community members, including men (n=41) and women (n=76) Purposive sampling | Qualitative content analysis | High |
| 12 (2012) | Kenya, Zambia | Mixed methods, comparative case study | Five sites; one NGO-owned stand-alone, one NGO-owned hospital-based, three hospital-owned hospital based | In-depth interviews (n=25) of female and male survivors of gender-based violence, caregivers of child survivors, hospital managers and key informant interviews; medical chart review, facility inventory review Purposive sampling | Qualitative: thematic analysis, Quantitative: EpiData and SPSS, accounting approach cost-analysis | Medium-high |
| 11 (2010) | 27 countries in Asia-Pacific; relevant countries include: Bangladesh, India, Indonesia, Malaysia, Maldives, Nepal, Papau New Guinea, Philippines, Sri Lanka, Thailand and Timor-Leste | Qualitative, descriptive | Variety of government-led, NGO-led and combined responses, both hospital-based and stand-alone facilities. | Desk review, field visits, phone and email interviews with relevant OSC stakeholders at country and regional offices Purposive sampling | Content analysis, thematic analysis | Low-medium |
| 46 (2016) | Sylet and Cox's Bazar, Bangladesh | Qualitative, descriptive | Five sites; stand-alone and hospital based; mostly government run with some NGO involvement | Key informant semi-structured interviews (n=124) of United Nations Population Fund (UNFPA) staff, government ministries, implementing partners and donors); mixed FGDs (n=12) of government and implementing partner staff, and community beneficiaries; site visits, desk review Purposive sampling | Content analysis, thematic analysis | Low-medium |
| 44 (2013) | Rwanda | Qualitative, descriptive | one site; hospital-based; government and NGO run | Semi-structured interviews and FGDs (n=93, breakdown not given) of survivors, OSC staff, and UN and government stakeholders; facility observation Convenience and purposive sampling | Thematic analysis | High |
| 33 (2013) | Zambia | Qualitative, cross-sectional survey | Eight sites; two stand-alone centres and six hospital-based; all international NGO funded | Survey (n=197) of female and male survivors of IPV or SV who accessed centre Convenience sampling | Descriptive analysis | Low |
| 30 (2013) | Nepal | Mixed methods | Four sites; hospital-based; government-run | In-depth interviews of female and male survivors of IPV or SV (n=20) and central stakeholders (n=137) including government employees and donors (n=13), health workers (n=58), members of coordination committees (n=42) and other (n=24). Purposive and convenience sampling | Content and thematic analysis, SWOT analysis | Medium |
| 13 (2016) | South Africa | Qualitative, descriptive | 55 sites (Thuthuzela centres); hospital-based and stand-alone; government and NGO run | Semi-structured interviews and surveys (number not provided) of National Prosecuting Authority staff, NGO staff, OSC managers and national experts in GBV in South Africa; facility observation Non-random sampling not otherwise specified | Qualitative: thematic analysis | High |
| 51 (2003) | Kenya | Qualitative, descriptive | 10 voluntary counselling and testing (VCT) sites, 11 hospitals, 6 legal and advocacy support programme; one hospital-based, private OSC (gender violence recovery centres) | In-depth and semi-structured interviews of male and female key informants (n=34) and FGDs (n=18) including hospital staff, police officers, government and NGO workers and VCT counsellors; facility observations Stratified and purposive sampling | Thematic analysis, content analysis | High |
| 45 (2010) | Zambia | Mixed methods | 10 sites; stand-alone and hospital-based, NGO and government- run | Semi-structured interviews (n=240) of key informants including beneficiaries, stakeholders and ministry officials; facility observations Sampling strategy not stated | Descriptive analysis | Low |
| 37 (2014) | Nepal | Qualitative, descriptive | 16 sites; hospital-based, government-run (Ministry of Health and Population) | Interviews of survivors of IPV and SV, government officials, OSC staff and community members Sampling strategy not stated | Descriptive analysis | Low |
| 39 (2016) | Pakistan | Mixed methods; cross-sectional, qualitative descriptive | 12 sites; stand-alone, or within government (non-medical) facilities, government-run | Semi-structured telephone interviews (n=136), including female survivors of IPV and SV (n=123), and male and female OSC managers (n=13); field visits and surveys Simple random sampling | Quantitative: standard statistical techniques, that is, descriptive analysis using SPSS and MS office; qualitative: thematic analysis of the open ended survey and interview questions | Medium-high |
| 38 (2017) | India | Qualitative, descriptive | Four sites; hospital-based and police-station-based, government-run. | In-depth interviews (n=80) including survivors of sexual assault (n=15), family members of survivors (n=25) and lawyers, civil society activists and advocates (n=15), doctors and forensic experts (n=6), government officials (n=12) and police officers (n=7) Purposive sampling | Thematic analysis | Medium-high |
| 18 (2002) | Philippines | Mixed methods; retrospective cohort, qualitative descriptive | One site (women and children protection unit); hospital-based, government-run | Medical chart review of non-pregnant women and children who were survivors of IPV and/or SV (n=1354) Convenience sampling | Basic descriptive statistical analysis | Medium |
| 19 (2013) | Kenya | Mixed methods; retrospective cohort, qualitative descriptive | One site; Clinic-based, NGO-run (Médecins Sans Frontières) | Medical chart review of female and male, child and adult survivors of sexual violence (n=866) Purposive sampling | Basic descriptive statistical analysis using Microsoft Excel and EpiData Analysis 2.1, qualitative descriptive analysis | Medium-high |
| 20 (2015) | Malaysia | Quantitative cross-sectional observational | one site; hospital-based, government-run | Self-reporting survey of male and female survivors of IPV (n=159) Purposive sampling | Basic statistical analyses conducted using SPSS V.20. | High |
| 15 (2011) | Malaysia | Qualitative, descriptive | Two sites; Hospital-based, combined NGO and government run | In-depth interviews (n=20), including policymakers (n=8), NGO representatives (n=7), healthcare workers (n=1) and police and social welfare representatives (n=4) Purposive and snowball sampling | Content analysis | High |
| 16 (2012) | Malaysia | Qualitative descriptive | Seven sites; hospital-based, combined NGO and government run | In-depth and semi-structured interviews (n=74) including accidents and emergency doctors (n=23), gynaecologists (n=6), nurses (n=14), medical social workers (n=5), counsellors (n=2), psychiatrists (n=4), policymakers (n=8) and key informants (n=12) Purposive and snowball sampling | Content and framework analysis | High |
| 49 (2009) | India | Mixed methods; cross-sectional observational, qualitative descriptive | One centre (Centre for Vulnerable Women and Children); stand-alone, combined NGO and government-run | Self-reported reflections and interviews with healthcare workers, female survivors of IPV/SV who utilised the centre (number not provided) Convenience sampling | Descriptive narrative analysis | Low |
| 10 (2002) | Thailand | Retrospective cohort, quasi-experimental, cross-over | Two centres; hospital-based, government-run | Structured and in-depth interviews (n=249) of female and male hospital staff including physicians, nurses, social workers, psychologists and intake personnel, community women’s leader groups, staff attorneys and police officers Sampling strategy not stated | Descriptive analysis | Low |
| 21 (2017) | Zimbabwe | Retrospective cohort | One site (Sexual and Gender-Based Violence Clinic); clinic-based, combined NGO (MSF) and government-run | Medical chart review (n=3617) of female and male survivors of sexual violence, including survivors ages over 16 (n=1071), ages 12–15 (n==615) and ages under 12 (n=93). Census | Descriptive statistics using Stata V.11. X2 tests, Fisher’s exact tests, logic regression, and model building | High |
| 40 (2009) | Kenya | Qualitative, descriptive | Three sites; hospital-based (emergency department), combined NGO and government-run | Client exit interviews (n=734) of female and male, child and adult survivors of rape Sampling strategy not stated | Situational analysis | Low |
| 32 (2009) | South Africa | Before and after intervention; retrospective cohort | One site; hospital-based, government-run | Semi-structured interviews with female and male survivors of rape (n=109) and service providers (n=16) (doctors, nurses, social workers, a pharmacist and police officers); medical chart review Convenience sampling | Quantitative descriptive analysis using Stata. Risk ratios estimated using Poisson regression to estimate intervention effect. Qualitative analysis methods not clearly stated | Moderate |
| 17 (2016) | Democratic Republic of Congo | Qualitative, descriptive | Two sites; Hospital based, privately-run | Descriptive personal narrative of medical director/obstetrics-gynaecologist and midwife (n=2) | Thematic analysis | Low |
| 22 (2011) | Kenya | Retrospective cohort | One site; hospital-based, government-run | Medical chart review of female and male survivors of sexual abuse (n=321), including children and adults, (median age 15.9 years; range 8 months to 100 years) Purposive sampling | Summary descriptive statistics using Stata SE 10.0. Estimates of association calculated using Student’s t-test, X2 tests and Fisher’s exact tests | High |
| 23 (2006) | South Africa | Observational, descriptive | One site (victim support centre); hospital-based, government-run | Self-reported survey of female and male survivors of rape (n=105) (median age 23.5 years; range 16–68) treated at the centre Purposive sampling | Descriptive analysis | Low |
| 14 (2010) | Ethiopia, Kenya, Malawi, Senegal, South Africa, Zambia, Zimbabwe | Retrospective cohort, qualitative, descriptive | Seven sites; includes variety of comprehensive care models including Thohoyandou Victim Empowerment Programme, South Africa, and the Kamuzu Central Hospital, Malawi; hospital-based, NGO-run | Interviews, surveys and medical chart review of survivors of sexual violence, healthcare workers, policymakers, government officials Sampling strategy not stated | Data analysis methods not clearly stated | Low |
| 24 (2017) | Taiwan | Cross-sectional | Five centres; hospital-based, government-run | Survey (n=140), using Index of Interdisciplinary Collaboration tool of social workers, doctors, nurses, police officers and prosecutor Purposive sampling | Statistical analysis via SPSS 18. Multivariate analysis of variance conducted for association analyses, eta-square for power of effect, and multilinear regression for influencers on collaboration | High |
| 25 (2016) | China | Retrospective cohort | Two sites (RainLily); Hospital-based, NGO-run | Medical chart review (n=154) of female survivors of sexual assault (median age 22 years; range 13–64) Purposive sampling | Descriptive statistical analysis via PASW Statistics 18, and Mann-Whitney test for highly skewed distributions | Low |
| 31 (2008) | Papua New Guinea | Mixed methods; cross-sectional, qualitative descriptive | ten sites (only Family Support Centres (FSCs) relevant to this review); Hospital-based, government and NGO run | Survey (n=39) of stakeholders (government officials, NGO representatives, and donors; In-depth interviews (n=17) of key informants (donors, service providers, governments officials, local women’s rights activists and faith-based groups) Purposive and snowball sampling | Descriptive and thematic analysis | Moderate |
| 50 (2016) | Papua New Guinea | Retrospective cohort | One site (FSC); hospital-based, government and NGO (MSF) run | Medical chart review (n=5212) of male and female presentations for SV and/or IPV Purposive sampling | Statistical analysis via χ2-squared or Fisher’s exact tests, multiple variable adjusted analyses, and modified Poisson regression | Moderate-high |
| 36 (2013) | South Africa | Qualitative, descriptive | Two sites; Stand-alone (near health facility), run by NGO (United Nations Office on Drugs and Crime), later transferred to SA government | Telephone and in-person interviews (n=20) of staff, representatives from government, civil society organisations, UNODC and advisory committees Convenience sampling | Descriptive and thematic analysis | Moderate |
| 29 (2011) | Malawi | Mixed methods; retrospective cohort, qualitative descriptive | Three sites; hospital-based, combined NGO and government-run | In-depth interviews (n=15) of healthcare workers (including doctors, clinical officers, nurses, midwives, social workers, health surveillance assistants and village health committee members). Key informant interviews (n=12) with policymakers, donors and other stakeholders and FGDs (n=10) with healthcare workers; chart review Purposive stratified sampling | Qualitative: thematic analysis Quantitative: descriptive statistical analysis and summary statistics via Epi Info, Pearson’s X2 and Fisher’s exact test | High |
| 47 (2012) | Kenya | Qualitative, descriptive | Four sites (only the Gender-Based Violence Recovery Centre (GBVRC) relevant to this review); hospital-based, government-run | Semi-structural interviews of female adult survivors of IPV/SV (n=8), and staff members (n=5) (head of department, psychologist, social worker, nurse counsellor and receptionist); client flow observations Purposive sampling | Thematic analysis | High |
| 27 (2010) | Sierra Leone | Qualitative, descriptive | Three centres (Rainbow Centres); stand-alone, NGO-run | In-depth interviews and FGDs of (n=101) male and female survivors of sexual assault and (n=22) OSC and NGO staff, including community leaders, judicial investigators, court magistrates and police; facility observations Convenience sampling | Descriptive analysis | Moderate |
| 35 (2015) | South Africa | Qualitative, descriptive | 29 sites (Thuthuzela centres); variety of hospital-based, stand-alone, police and court-based centres; combined NGO and government run, or only government-run | In-depth interviews (n=40) of OSC directors and programme managers; participant observation Non-random sampling not otherwise specified | Descriptive analysis | Moderate |
| 42 (2004) | India | Qualitative, descriptive | One centre (Dilaasa); hospital-based, Combined NGO and government run | Semi-structured interviews (n=27) of adult female survivors of IPV/SV, including current and former programme participants Purposive sampling | Content analysis | High |
| 34 (2010) | India | Qualitative, descriptive | Two centres (Dilaasa); hospital- based, combined NGO and government run | Semi-structured interviews with survivors of violence, project personnel, coordinator, mentors and hospital staff (number not specified); facility observation Sampling strategy not stated | Thematic analysis | Moderate |
| 28 (2018) | Mongolia | Qualitative, descriptive | Four sites; variety of centres–some health facility based, stand-alone, and police-station based, variety of government and NGO-run, funded by UNFPA | In-depth interviews (n=36) and FGDs (n=6) of key informants Sampling strategy not stated | Thematic analysis | Low-moderate |
*Some details of sample characteristics such as participant sex, age, professional role, specific sampling strategy and data collection and analysis methods were not provided in the primary studies, and thus do not appear in table 2.
FGD, focus group discussion; FSC, family support centre; IPV, intimate partner violence; MOU, memorandum of understanding; MSF, Médecins Sans Frontières; OSC, one stop centre; SOP, standard operating procedures; SPSS, Statistical Package for the Social Sciences; SV, sexual violence; VAW, violence against women.