From: Evaluations of reproductive health programs in humanitarian settings: a systematic review
Author (Year) | Country | Intervention | Evaluation design | Key findings | Quality |
---|---|---|---|---|---|
Maternal and newborn health | |||||
Ayoya et al (2013) [14] | Haiti | Established baby tents in five cities to promote and sustain optimal infant feeding practices: breastfeeding and nutrition support, infant growth monitoring, assessment of nutritional status of mother-infant pairs and pregnant women | Program data review (February 2010-June 2012) from nutritional cluster database (n=193 baby tents) | 70% of infants less than 6 months old were exclusively breastfed. 10% of “mixed feeders” less than 6 months changed to exclusive breastfeeding while enrolled. | Low |
Howard et al (2011)1[12] | Guinea | Seconded refugee health workers to health facilities serving refugees, provided free RH services and trained refugee women as lay health workers | Cross-sectional post-intervention multi-stage cluster survey in intervention area of women (n=444) and men (n=445) of reproductive age (Liberian and Sierra Leonean refugees) living in one of 48 refugee camps in Guinea in 1999. | Higher odds of facility delivery for those exposed to intervention education activities (OR=2.03, 95%CI 1.23-3.01), formally educated (OR=1.93, 95%CI 1.05-3.92), or grand multipara (OR= 2.13, 95%CI 1.21-3.75). No significant differences found in maternal health knowledge or attitudes. | Medium |
Krause et al (2006) [9] | Global (9 countries) | Improved availability of basic and comprehensive EmONC services in 12 conflict affected settings in 9 countries, Jan 2001-Apr 2005 | Pre and post intervention facility assessments (n=31 health facilities) | Increased availability of EmONC 24 hours a day. CEmONC facilities increased from 3 facilities at baseline to 10 at endline; BEmONC facilities increased from 2 at baseline to 10 at endline. The number of signal functions available increased in all 31 facilities. | Medium |
Lori et al (2010) [15] | Liberia | Trained traditional midwives in maternal care using the home-based life-saving skills series in 2006 | Pre- & immediate post-training assessments (n=412 traditional midwives), 1-year follow up assessment (n=389) | Mean scores in 4 topic areas: 1) first actions, 2) post-partum hemorrhage 3) woman referral, 4) baby referral, improved from pre- to post-test and remained stable one year later (p<.001 for all 4 topics) | Medium |
Purdin et al (2009) [8] | Pakistan | Established EmONC facilities, trained Afghan refugee community members on safe motherhood, linked primary health care with education on danger signs of pregnancy and the importance of skilled birth attendance, and improved the health information system. | Program data review 2000-2007 | Maternal mortality ratio improved from 291 per 100,000 live births in 2000 to 102 in 2004. Case fatality rate for obstetric complications=0.2%. Skilled birth attendance increased from 5% in 1996 to 67% in 2007. Complete ANC coverage increased from 49% in 2000 to 90% in 2006; post-natal coverage increased from 27% in 2000 to 85% in 2006. | Medium |
Maternal and newborn health and family planning | |||||
Mullany et al (2010) [11] | Burma | Trained community-based skilled health workers in basic EmONC, evidence-based ANC and FP in Shan, Mon, Karen, and Karenni regions of Burma | Pre (2006) & post (2008) intervention cross-sectional two-stage cluster surveys in intervention areas of ever married women of reproductive age: n=2,889 at baseline, n=2,442 at endline | Use of modern FP methods increased from 24% to 45% (PRR 1.88, 95%CI 1.63-2.17). Unmet need for FP decreased 35% (95%CI 28%-40%). Skilled birth attendance increased from 5% to 49% (PRR=9.55, 95%CI 7.21-12.64). | High |
Viswanathan et al (2012) [13] | Afghanistan | Deployed CHWs to promote use of RH services in community and at health facilities | Data derived from the Afghanistan Health Survey 2006: multistage cluster survey in 29 provinces (n=8,281 women) | Presence of female CHW in community is associated with increased use of FP (OR=1.61, 95%CI 1.21-2.15), ANC (OR=2.71, 95%CI 1.87-3.92) and skilled birth attendant at last delivery (OR=1.75, 95%CI 1.18-2.58). These associations were not significant with a male CHW. | Medium |
Family planning | |||||
Casey et al (2013) [16] | Northern Uganda | Provided short-acting, long-acting and permanent FP methods via mobile outreach teams and strengthened public health center provision of short and long acting FP methods | Baseline (2007) and post-intervention (2010) cross-sectional multi-stage cluster surveys in intervention area of women of reproductive age: n=905 at baseline, n=873 at endline | Current use modern FP methods increased from 7.1% to 22.6% (OR=3.34, 95%CI 2.27-4.92); use of LAPM increased 1.2% to 9.8% (OR=9.45, 95%CI 3.99-22.4). Unmet need for FP decreased from 52.1% to 35.7% (OR=0.47, 95% CI 0.37-0.60). | High |
Howard et al (2008)1[17] | Guinea | Refugee health workers seconded to health facilities provided free RH services and trained refugee women as lay health workers | Cross-sectional post-intervention multi-stage cluster survey in intervention area of women (n=444) and men (n=445) of reproductive age (Liberian and Sierra Leonean refugees) living in one of 48 refugee camps in Guinea in 1999. | Approval of FP was high, but more than 40% had not discussed FP with partner. Current use of modern FP (17%) was higher than in country of origin (3.9%) or host country (4.1%). Perceived service quality was most important determinant in choice of where to get FP. | Medium |
Huber et al (2010) [18] | Afghanistan | Improved access to FP using CHWs and community-based distribution of short acting methods | Baseline (2004) and endline (2006) cross-sectional surveys using lot quality assurance sampling; verification of FP use via home visits of 150 FP users per CHW | Current FP use increased by 24-27%, with injectables contributing most to the increase. | Low |
Raheel et al (2012) [19] | Pakistan | Provided subsidized or unsubsidized health care to Afghan refugees in Karachi | Cross-sectional study in 2008 using systematic random sampling of 2 comparison groups: married Afghan women of reproductive age receiving subsidized care (n=325) and unsubsidized care (n=325) | Refugee women receiving subsidized care were more likely to have heard of FP (OR=10.12 95%CI 6.7-15.31) and currently use FP (OR=3.65, 95%CI 2.61-5.10). | High |
Gender-based violence | |||||
Bass et al (2013) [20] | Democratic Republic of the Congo (DRC) | Adapted group cognitive processing therapy (1 individual session and 11 group sessions) provided by paraprofessionals supervised by psychosocial staff and clinical experts | Random assignment of 16 villages to intervention group (8) or individual support (8) for female sexual violence survivors in 2011 | 65% in intervention group and 52% in control group completed all 3 measures. Improvements in all 3 sets of symptoms were significantly greater in therapy group than in individual support group. Mean scores for combined depression and anxiety improved significantly more in the therapy group compared to the individual support group (p<0.001 for all comparisons). | High |
Hustache et al (2009) [21] | Republic of Congo | Provided medical care and psychological support to women raped by an unknown perpetrator in military clothing | Initial assessment January 2002-April 2003 (n=159 female survivors of rape); follow-up 1-2 years post-treatment, June-July 2004 (n=70) | 56 women were evaluated using the Global Assessment of Functioning (GAF) scale at both time periods, and global functioning significantly improved (p=.04); this improvement was maintained 1-2 years later | Medium |
Smith et al (2013) [22] | DRC, Ethiopia, Kenya, Jordan | Multimedia training tool for health providers to encourage competent, compassionate, and confidential clinical care for rape survivors | Assessment pre-training and 3 months after, medical record review, in-depth interviews (November 2010 to June 2012) | Although negative attitudes did not significantly decrease, respect for patient rights increased (p<.05), and provider practice improved from before the training to 3 months post-training (p<.01). | Medium |
HIV/AIDS and other STIs | |||||
Ahoua et al (2010) [24] | Northern Uganda | PMTCT program including either short-course AZT or single dose nevirapine and follow-up for 18 months post-partum including infant HIV testing | Retrospective record review of all mother-infant pairs enrolled July 2000-July 2005 (n=517). Cross-sectional survey of infant status at 18 months following tracing of mother-infant pairs who were lost to follow up (n=327 women and 368 babies). | 53% of mother-infant pairs were lost to follow-up before completing infant testing at 18 months; the risk of death or being lost to follow-up was higher among infants with no or incomplete intrapartum ARVs (OR=1.9, 95%CI 1.07–3.36) and of weaning before age 6 months (OR=2.55, 95%CI 1.42–4.58). | Medium |
Atwood et al (2012)3[35] | Liberia | Evidence-based HIV prevention curriculum adapted for in-school Liberian youth. The 8-modules promoted positive condom attitudes and increased skills and self-efficacy to refuse sex, negotiate condom use and use condoms effectively. | Attention-matched, group RCT: 4 matched pairs of schools randomly assigned to HIV prevention curriculum or general health curriculum. Students completed baseline, immediate post-test, 3- and 9- month follow-up surveys to assess program efficacy (n=740 completed all measures) | The intervention significantly improved protective peer norms (p<.05) and positive condom attitudes (p<.05) at the 9 month follow-up. Among those who were sexually active at baseline, the intervention group used condoms more consistently in the last 3 months (p<.05) at the 9-month follow-up. The intervention did not impact sexual initiation or multiple sex partnerships. | Medium |
Atwood et al (2012)3[36] | Liberia | Evidence-based HIV prevention curriculum adapted for in-school Liberian youth. The 8-modules promoted positive condom attitudes and increased skills and self-efficacy to refuse sex, negotiate condom use and use condoms effectively. | Attention-matched, group RCT: 4 matched pairs of schools randomly assigned to HIV prevention curriculum or general health curriculum. Students completed baseline, immediate post-test, 3- and 9- month follow-up surveys to assess program efficacy (n=714 who responded to questions about transactional sex) | Risk behaviors for adolescents who engaged in transactional sex were no different in the intervention or control groups. | Medium |
Bannink-Mbazzi et al (2013) [23] | Northern Uganda | PMTCT program including couple VCT, care and treatment for HIV+ individuals, home-based care, partner involvement, follow-up at 18 months post-partum including infant HIV testing | Retrospective record review of PMTCT program data 2002-2011 | Of 140,658 women starting ANC, 94.4% received HIV testing. Testing of male partners increased from 5.9% in 2002 to 75.8% in 2011 (p=.001) compared to 15.5% nationally. 79% of HIV+ women started ARVs, compared to 52% nationally. HIV prevalence among exposed infants tested by 18 months decreased from 10.3% in 2004 to 5.0% in 2011 (p=.001). | Medium |
Casey et al (2006)2[37] | Sierra Leone | HIV/AIDS and STI prevention program comprised of intensive outreach education by peers including a focus on improving negotiation skills and distribution of free condoms targeting youth | Baseline (2001) and post intervention (2003) cross-sectional surveys using purposive quota sampling of youth: n=244 female, 293 male (baseline); n=250 female, 299 male (endline) | Respondents able to name 3 effective means of avoiding AIDS increased from 4% to 36% among female youth and from 4% to 45% among male youth; reported condom use at last sex increased from 16% to 46% (female) and from 16% to 37% (male) (p<.01 for all comparisons). | Medium |
Chen et al (2008)1[39] | Guinea | Refugee health workers seconded to health facilities provided free RH services and trained refugee women as lay health workers | Cross-sectional post-intervention multi-stage cluster survey in intervention area of women (n=444) and men (n=445) of reproductive age (Liberian and Sierra Leonean refugees) living in one of 48 refugee camps in Guinea in 1999. | Self-reported STI symptoms were common: 30% among women and 24% among men. Only 25% correctly named key STI symptoms. Respondents citing program facilitators as sources of information were more likely to correctly name key STI symptoms (OR=5.2, 95% CI 1.9-13.9 (men)) and identify effective means of protecting against STIs (OR=2.9, 95% CI 1.5-5.8 (men)) and (OR=4.6, 95%CI 1.6-13.2 (women)). | Medium |
Ciccio and Sera (2010) [41] | Northern Uganda | HIV/AIDS prevention activities with youth including media campaigns, peer counseling, life skills training, and activities for youth in particularly vulnerable circumstances to spread prevention messages and help them develop the skills necessary to protect themselves. | Cross-sectional post-intervention survey using lot quality assurance sampling in intervention area (n=1,781 youth age 15-24) in 2008 | 29% had comprehensive HIV prevention knowledge (knew 3 main means of prevention and rejected common misconceptions). 86% knew where to get tested for, but only 51% had been tested and received their result in the last 12 months. Gender, geographical location, marital status and education were associated with this knowledge (p<.001) | Medium |
Culbert et al (2007) [33] | DRC | Voluntary counseling and HIV testing (VCT), care and treatment for HIV+ individuals, HIV prevention activities | Program data review: May 2002-Jan 2006 | 11,076 people received VCT, of whom 19% were HIV+; 94% of these received follow-up care in the HIV clinics. 12-month mortality among ART patients was 7.9% (95%CI 3.6-12.1), and 12-month loss to follow-up was 5.4% (95%CI 3.2-7.5), both comparable to stable low resource settings. Only 5 of 66 ART patients experienced treatment interruption during violent period of May-June 2004. | Medium |
Garang et al (2009) [26] | Northern Uganda | Care and treatment for HIV+ individuals | Cross-sectional study using systematic sampling of self-reported adherence over 4-day period in February 2008 (n=200 adults on ART) | Mean 4-day adherence (self-reported) was 99.5%, with no difference between IDPs and non-IDPs. Being on a 1st line ART regimen (OR=22.2, 95%CI 1.5-333.3), feeling facility staff were condemning (OR=22.2, 95%CI 1.5-333.3), and lack of privacy at facility (OR=9.7, 95%CI 0.9-111.1) were associated with non-adherence. | High |
Kiboneka et al (2008)4[28] | Northern Uganda | Care and treatment for HIV+ individuals, facility and home-based care, mobile clinics to IDP camps | Prospective cohort study using program data June 2005 - Feb 2008 (n=57 HIV+ children receiving combination ART) | Adherence was consistently excellent in 92% of patients. No deaths and no major opportunistic infections were recorded after initiation of ART. | Medium |
Kiboneka et al (2009)4[27] | Northern Uganda | Care and treatment for HIV+ individuals, facility and home-based care, mobile clinics to IDP camps | Prospective cohort study using program data, June 2005 - Jan 2008, (n=1,625 HIV+ adults receiving combination ART) | The mortality incidence rate was 3.48 (95%CI 2.7-4.3) per 100 person years. Of patients with adherence data, 92% had adherence greater than 95%. 4.3% of patients died during follow-up, a mortality rate comparable to ART patients in stable settings. Lower mortality was associated with female sex, higher baseline CD4 count and ≥95% adherence. IDP camp residence and age were not associated with mortality outcomes. | Medium |
Larsen et al (2004)2[38] | Sierra Leone | HIV/AIDS and STI prevention program comprised of intensive outreach education by peers including a focus on improving negotiation skills and distribution of free condoms targeting commercial sex workers (CSW) and military men | Baseline (2001) and post intervention (2003) cross-sectional surveys using purposive quota sampling: n=201 sex workers, 202 military men (baseline); n=202 sex workers, 205 military men (endline) | Those able to name 3 effective means of avoiding AIDS increased from 5% to 70% among CSWs and from 11% to 75% among military men. Reported condom use at last sex increased from 38% to 68% (CSW) and from 39% to 68% (military) (p<.01 for all). Although the proportions of both CSWs and military men who believe HIV+ people should be treated or counselled increased, the proportions believing they should be isolated or reported did not change. | Medium |
O'Brien et al (2010) [29] | Global | Programs of care and treatment for HIV+ individuals in conflict and post-conflict settings | Program data review 2005-2009 (n=20 programs with complete data and n=4,145 HIV+ adults on ART with complete data) | 64% of ART patients remained on ART, 10% died, 11% were lost to follow-up. Median 12-month mortality and loss to follow-up were 9% (95%CI 8.8-9.1) and 11% (95%CI 9-12) respectively. Median 6-month CD4 gain was 129 cells/mm3. | Medium |
Pyne-Mercier et al (2011) [30] | Kenya | Care and treatment for HIV+ individuals | Retrospective record review for clients on ART during post-election violence, Dec 30, 2007 - Feb 28, 2008, and same time period 1 year earlier (n=2,534 HIV+ adults) | The odds of treatment interruption were 71% (95%CI 34-118) higher during the post-election violence period compared to 1 year earlier. Men (OR=1.4, 95%CI 1.1-1.8) and those traveling ≥3 hours to clinic (OR=1.9, 95%CI 1.3-2.7) were more likely to experience treatment interruption. | High |
Rutta et al (2008) [25] | Tanzania | 2-year pilot PMTCT program in refugee camp: community education, training providers, VCT, infant feeding, counseling, administration of nevirapine | Program data review Oct 2002 - June 2004 (n=6 health facilities) | 92% of ANC clients were tested for HIV. 93% of HIV+ women agreed to take nevirapine at 34 weeks of gestation. 36% of the HIV+ women were repatriated before delivery, but 98% of those remaining took nevirapine at the start of labor and their infants received nevirapine within 72 hours. Only 15% of HIV-exposed infants were tested at 18 months due to repatriation, death or refusal of testing. | Medium |
Tanaka et al (2008) [42] | Tanzania | HIV/AIDS prevention including youth peer education, VCT, free condom distribution in Nyarugusu refugee camp | Post-intervention survey of systematically selected Congolese refugees of reproductive age (n=570 male and 570 female) living in the refugee camp in 2005 | HIV risk increased after displacement due to increased transactional sex and forced sex (p<.001). Condom use at last sex with a non-regular partner was 14% and associated with citing the program health teams as a leading source of influence regarding HIV prevention | Medium |
Vreeman et al (2009)5[31] | Kenya | Care and treatment for HIV-infected children | Retrospective cohort analysis of HIV+ children under 14 years seen from Oct-Dec 2007 in 18 clinics (n=2,585), and then followed from Dec 2007 until April 2008. | 93% of HIV-infected children returned to care in the 4 months after the violence, and 98% of children on ART reported perfect adherence during last 7 days (p<.001). Children on ART were more likely to return than those not on ART (OR=1.4, 95%CI 1.2-1.6). Orphan status and sex were not associated with return to clinic. | Medium |
Walldorf et al (2012) [34] | Haiti | HIV/AIDS clinical services including VCT, PMTCT, care and treatment for HIV+ individuals | Program data Oct 2008-May 2010 comparing pre-earthquake (prior to Dec 2009) to post-earthquake outcomes (n=126 facilities) | Mean monthly enrollment for VCT, PMTCT and ART services were from 41-46% of baseline levels in Jan 2010 but rose to 79-89% of baseline levels in May 2010. Current ART patients rose 3.6% Jan – May 2010 compared to a 9.8% increase during the same period in 2009. | Medium |
Woodward et al (2011)1[40] | Guinea | Refugee health workers seconded to health facilities provided free RH services and trained refugee women as lay health workers | Cross-sectional post-intervention multi-stage cluster survey in intervention area of women (n=444) and men (n=445) of reproductive age (Liberian and Sierra Leonean refugees) living in one of 48 refugee camps in Guinea in 1999. | HIV knowledge was high. Participants exposed to program peer education had higher odds of reporting changes in sexual behavior to avoid HIV (OR=2.5, 95%CI 1.5-4.1). Exposed participants were less likely to report staying faithful (OR=0.6, 95%CI 0.4-0.9) and more likely to report fewer sex partners (OR=1.7, 95%CI 1.05-2.85). | Medium |
Yoder et al (2012)5[32] | Kenya | HIV/AIDS care and treatment for HIV-infected children | Retrospective cohort analysis for 3 time periods: pre-election, Oct 26-Dec 25 2007; immediately post-election, Dec 26, 2007 - Apr 15, 2008; and long-term post-election, Apr 16-Dec 31, 2008 (n=2,549 HIV+ children) | Children on ART had less initial loss to follow-up (p<.01) and less complete loss to follow-up (p<.0001) than children not on ART. Immediately post-election, 8.2% of children on ART had imperfect medication adherence, and 9.0% long-term post-election. | Medium |
General RH | |||||
McGinn & Allen (2006) [44] | Guinea | Literacy training using RH information as the content and participatory adult education techniques for Sierra Leonean and Liberian women living in refugee camps | Post-intervention cross-sectional survey of RH literacy program students who participated in 1999, 2000 and 2001 RH literacy courses and were still in the area in 2002 (n=549) | The proportion of women who reported communication with their partners on RH topics increased to 87% (p<.001). Current use of FP was 50%. The proportion of women who reported feeling more empowered than other women increased from 32% (based on recall) to 82% after the program (p<.001). | Medium |
Sullivan et al (2004) [43] | Thai-Burma border | Program to improve quality of RH services and build health providers' capacity in monitoring and evaluation | Pre- and post-intervention facility audits, observations of client-provider interactions during ANC and FP visits, client exit interviews (2001-2003) | Improved program readiness contributed to improved quality of information given to clients, technical competence and integration of services, although some contradictory findings from client exit interviews warrant further exploration. | Low |