Exogenous factors matter when interpreting the results of an impact evaluation: a case study of rainfall and child health programme intervention in Rwanda

OBJECTIVE

Public health interventions are often implemented at large scale, and their evaluation seems to be difficult because they are usually multiple and their pathways to effect are complex and subject to modification by contextual factors. We assessed whether controlling for rainfall-related variables altered estimates of the efficacy of a health programme in rural Rwanda and have a quantifiable effect on an intervention evaluation outcomes.

METHODS

We conducted a retrospective quasi-experimental study using previously collected cross-sectional data from the 2005 and 2010 Rwanda Demographic and Health Surveys (DHS), 2010 DHS oversampled data, monthly rainfall data collected from meteorological stations over the same period, and modelled output of long-term rainfall averages, soil moisture, and rain water run-off. Difference-in-difference models were used.

RESULTS

Rainfall factors confounded the PIH intervention impact evaluation. When we adjusted our estimates of programme effect by controlling for a variety of rainfall variables, several effectiveness estimates changed by 10% or more. The analyses that did not adjust for rainfall-related variables underestimated the intervention effect on the prevalence of ARI by 14.3%, fever by 52.4% and stunting by 10.2%. Conversely, the unadjusted analysis overestimated the intervention's effect on diarrhoea by 56.5% and wasting by 80%.

CONCLUSION

Rainfall-related patterns have a quantifiable effect on programme evaluation results and highlighted the importance and complexity of controlling for contextual factors in quasi-experimental design evaluations.

Non-Obstetric Surgical Care at Three Rural District Hospitals in Rwanda: More Human Capacity and Surgical Equipment May Increase Operative Care

Background

Most mortality attributable to surgical emergencies occurs in low- and middle-income countries. District hospitals, which serve as the first-level surgical facility in rural sub-Saharan Africa, are often challenged with limited surgical capacity. This study describes the presentation, management, and outcomes of non-obstetric surgical patients at district hospitals in Rwanda.

Methods

This study included patients seeking non-obstetric surgical care at three district hospitals in rural Rwanda in 2013. Demographics, surgical conditions, patient care, and outcomes are described; operative and non-operative management were stratified by hospitals and differences assessed using Fisher’s exact test.

Results

Of the 2660 patients who sought surgical care at the three hospitals, most were males (60.7 %). Many (42.6 %) were injured and 34.7 % of injuries were through road traffic crashes. Of presenting patients, 25.3 % had an operation, with patients presenting to Butaro District Hospital significantly more likely to receive surgery (57.0 %, p < 0.001). General practitioners performed nearly all operations at Kirehe and Rwinkwavu District Hospitals (98.0 and 100.0 %, respectively), but surgeons performed 90.6 % of the operations at Butaro District Hospital. For outcomes, 39.5 % of all patients were discharged without an operation, 21.1 % received surgery and were discharged, and 21.1 % were referred to tertiary facilities for surgical care.

Conclusion

Significantly more patients in Butaro, the only site with a surgeon on staff and stronger surgical infrastructure, received surgery. Availing more surgeons who can address the most common surgical needs and improving supplies and equipment may improve outcomes at other districts. Surgical task sharing is recommended as a temporary solution.

Treating persistent asthma in rural Rwanda: characteristics, management and 24-month outcomes

SETTING

In 2007, the Rwandan Ministry of Health, with support from Partners In Health, introduced a district-level non-communicable disease programme that included asthma care.

OBJECTIVE

To describe the demographics, management and 24-month outcomes of asthma patients treated at three rural district hospitals in Rwanda.

DESIGN

We retrospectively reviewed electronic medical records of asthma patients enrolled from January 2007 to December 2012, and extracted information on demographics, clinical variables and 24-month outcomes.

RESULTS

Of the 354 patients, 66.7% were female and 41.5% were aged between 41 and 60 years. Most patients (53.1%) were enrolled with moderate persistent asthma, 40.1% had mild persistent asthma and 6.8% had severe persistent asthma. Nearly all patients (95.7%) received some type of medication, most commonly a bronchodilator. After 24 months, 272 (76.8%) patients were still alive and in care, 21.1% were lost to follow-up, 1.7% had died and 0.3% had transferred out. Of the 121 patients with an updated asthma classification at 24 months, the severity of their asthma had decreased: 17.4% had moderate and 0.8% had severe persistent asthma.

CONCLUSION

Our findings show improvements in asthma severity after 24 months and reasonable rates of loss to follow-up, demonstrating that asthma can be managed effectively in rural, resource-limited settings.

Same-day HIV testing with initiation of antiretroviral therapy versus standard care for persons living with HIV: A randomized unblinded trial

BACKGROUND

Attrition during the period from HIV testing to antiretroviral therapy (ART) initiation is high worldwide. We assessed whether same-day HIV testing and ART initiation improves retention and virologic suppression.

METHODS AND FINDINGS

We conducted an unblinded, randomized trial of standard ART initiation versus same-day HIV testing and ART initiation among eligible adults ≥18 years old with World Health Organization Stage 1 or 2 disease and CD4 count ≤500 cells/mm3. The study was conducted among outpatients at the Haitian Group for the Study of Kaposi's Sarcoma and Opportunistic infections (GHESKIO) Clinic in Port-au-Prince, Haiti. Participants were randomly assigned (1:1) to standard ART initiation or same-day HIV testing and ART initiation. The standard group initiated ART 3 weeks after HIV testing, and the same-day group initiated ART on the day of testing. The primary study endpoint was retention in care 12 months after HIV testing with HIV-1 RNA <50 copies/ml. We assessed the impact of treatment arm with a modified intention-to-treat analysis, using multivariable logistic regression controlling for potential confounders. Between August 2013 and October 2015, 762 participants were enrolled; 59 participants transferred to other clinics during the study period, and were excluded as per protocol, leaving 356 in the standard and 347 in the same-day ART groups. In the standard ART group, 156 (44%) participants were retained in care with 12-month HIV-1 RNA <50 copies, and 184 (52%) had <1,000 copies/ml; 20 participants (6%) died. In the same-day ART group, 184 (53%) participants were retained with HIV-1 RNA <50 copies/ml, and 212 (61%) had <1,000 copies/ml; 10 (3%) participants died. The unadjusted risk ratio (RR) of being retained at 12 months with HIV-1 RNA <50 copies/ml was 1.21 (95% CI: 1.04, 1.38; p = 0.015) for the same-day ART group compared to the standard ART group, and the unadjusted RR for being retained with HIV-1 RNA <1,000 copies was 1.18 (95% CI: 1.04, 1.31; p = 0.012). The main limitation of this study is that it was conducted at a single urban clinic, and the generalizability to other settings is uncertain.

CONCLUSIONS

Same-day HIV testing and ART initiation is feasible and beneficial in this setting, as it improves retention in care with virologic suppression among patients with early clinical HIV disease.

TRIAL REGISTRATION

This study is registered with ClinicalTrials.gov number NCT01900080.

Fertility and HIV following universal access to ART in Rwanda: a cross-sectional analysis of Demographic and Health Survey data

BACKGROUND

HIV infection is linked to decreased fertility and fertility desires in sub-Saharan Africa due to biological and social factors. We investigate the relationship between HIV infection and fertility or fertility desires in the context of universal access to antiretroviral therapy introduced in 2004 in Rwanda.

METHODS

We used data from 3532 and 4527 women aged 20-49 from the 2005 and 2010 Rwandan Demographic and Health Surveys (RDHS), respectively. The RDHSs included blood-tests for HIV, as well as detailed interviews about fertility, demographic and behavioral outcomes. In both years, multiple logistic regression was used to assess the association between HIV and fertility outcomes within three age categories (20-29, 30-39 and 40-49 years), controlling for confounders and compensating for the complex survey design.

RESULTS

In 2010, we did not find a difference in the odds of pregnancy in the last 5 years between HIV-seropositive and HIV-seronegative women after controlling for potential biological and social confounders. Controlling for the same confounders, we found that HIV-seropositive women under age 40 were less likely to desire more children compared to HIV-seronegative women (20-29 years adjusted odds ratio (AOR) = 0.31, 95% CI: 0.17, 0.58; 30-39 years AOR = 0.24, 95% CI: 0.14, 0.43), but no difference was found among women aged 40 or older. No associations between HIV and fertility or fertility desire were found in 2005.

CONCLUSIONS

These findings suggest no difference in births or current pregnancy among HIV-seropositive and HIV-seronegative women. That in 2010 HIV-seropositive women in their earlier childbearing years desired fewer children than HIV-seronegative women could suggest more women with HIV survived; and stigma, fear of transmitting HIV, or realism about living with HIV and prematurely dying from HIV may affect their desire to have children. These findings emphasize the importance of delivering appropriate information about pregnancy and childbearing to HIV-infected women, enabling women living with HIV to make informed decisions about their reproductive life.

A Novel Combined Mother-Infant Clinic to Optimize Post-Partum Maternal Retention, Service Utilization, and Linkage to Services in HIV Care in Rural Rwanda

BACKGROUND AND OBJECTIVES

Despite recent improvements in accessibility of services to prevent mother-to-child transmission of HIV, maternal retention in HIV care remains a challenge in the post-partum period. This study assessed service utilization, program retention, and linkage to routine services, as well as clinical outcomes for mothers and infants, following implementation of an integrated mother-infant clinic in rural Rwanda.

METHODS

We conducted a retrospective cohort study of all HIV-positive mothers and their infants enrolled in the integrated clinics in two rural districts between July 1, 2012, and June 30, 2013. At 18 months post-partum, data on mother-infant service utilization and program outcomes were reported.

RESULTS

Of the 185 mother-infant pairs in the clinics, 98.4% of mothers were on antiretroviral therapy (ART) and 30.3% used modern contraception at enrollment. At 18 months post-partum, 98.4% of mothers were retained and linked back to adult HIV program. All mothers were on ART and 72.0% on modern contraception. For infants, 93.0% completed follow-up. Two (1.1%) infants tested HIV positive.

CONCLUSION AND GLOBAL HEALTH IMPLICATION

An integrated clinic was successfully implemented in rural Rwanda with high mother retention in care and low mother to child HIV transmission rates. This model of integration of services may contribute to improved mother-infant retention in care during post-partum period and should be considered as one approach to addressing this challenge in similar settings.

Rainfall variation and child health: effect of rainfall on diarrhea among under 5 children in Rwanda, 2010

Background

Diarrhea among children under 5 years of age has long been a major public health concern. Previous studies have suggested an association between rainfall and diarrhea. Here, we examined the association between Rwandan rainfall patterns and childhood diarrhea and the impact of household sanitation variables on this relationship.

Methods

We derived a series of rain-related variables in Rwanda based on daily rainfall measurements and hydrological models built from daily precipitation measurements collected between 2009 and 2011. Using these data and the 2010 Rwanda Demographic and Health Survey database, we measured the association between total monthly rainfall, monthly rainfall intensity, runoff water and anomalous rainfall and the occurrence of diarrhea in children under 5 years of age.

Results

Among the 8601 children under 5 years of age included in the survey, 13.2 % reported having diarrhea within the 2 weeks prior to the survey. We found that higher levels of runoff were protective against diarrhea compared to low levels among children who lived in households with unimproved toilet facilities (OR = 0.54, 95 % CI: [0.34, 0.87] for moderate runoff and OR = 0.50, 95 % CI: [0.29, 0.86] for high runoff) but had no impact among children in household with improved toilets.

Conclusion

Our finding that children in households with unimproved toilets were less likely to report diarrhea during periods of high runoff highlights the vulnerabilities of those living without adequate sanitation to the negative health impacts of environmental events.

Use of Lot Quality Assurance Sampling to Ascertain Levels of Drug Resistant Tuberculosis in Western Kenya

Objective

To classify the prevalence of multi-drug resistant tuberculosis (MDR-TB) in two different geographic settings in western Kenya using the Lot Quality Assurance Sampling (LQAS) methodology.

Design

The prevalence of drug resistance was classified among treatment-naïve smear positive TB patients in two settings, one rural and one urban. These regions were classified as having high or low prevalence of MDR-TB according to a static, two-way LQAS sampling plan selected to classify high resistance regions at greater than 5% resistance and low resistance regions at less than 1% resistance.

Results

This study classified both the urban and rural settings as having low levels of TB drug resistance. Out of the 105 patients screened in each setting, two patients were diagnosed with MDR-TB in the urban setting and one patient was diagnosed with MDR-TB in the rural setting. An additional 27 patients were diagnosed with a variety of mono- and poly- resistant strains.

Conclusion

Further drug resistance surveillance using LQAS may help identify the levels and geographical distribution of drug resistance in Kenya and may have applications in other countries in the African Region facing similar resource constraints.

Socioeconomics and Major Disabilities: Characteristics of Working-Age Adults in Rwanda

BACKGROUND

Disability affects approximately 15% of the world's population, and has adverse socio-economic effects, especially for the poor. In Rwanda, there are a number of government compensation programs that support the poor, but not specifically persons with disability (PWDs). This study investigates the relationship between poverty and government compensation on disability among working-age adults in Rwanda.

METHODS

This was a secondary analysis of 35,114 adults aged 16 to 65 interviewed in the 2010/2011 Rwanda Household Wealth and Living Conditions survey, a national cross-sectional two-stage cluster survey, stratified by district. This study estimated self-reported major disability, and used chi-square tests to estimate associations (p<0.1) with income, government compensation, occupation type, participation in public works programs, and household poverty status. Non-collinear economic variables were included in a multivariate logistic regression, along with socio-demographic confounders that modified the relationship between any economic predictor and the outcome by 10% or more. All analyses adjusted for sampling weights, stratification, and clustering of households.

RESULTS

Over 4% of working-age adults reported having a major disability and the most prevalent types of disability in order were physical, mental, and then sensory disability. In bivariate analysis, annual income, occupation type, and poverty status were associated with major disability (p<0.001 for all). Occupation type was dropped because it was collinear with income. Age, education, and urban/rural residence were confounders. In the multivariate analysis, adults in all income groups had about half the odds of disability compared to adults with no income (Rwf1-120,000 OR = 0.57; Rwf120,000-250,000 OR = 0.61; Rwf250,000-1,000,000 OR = 0.59; Rwf1,000,000+ OR = 0.66; p<0.05 for all), and non-poor adults had 0.77 the odds of disability compared to poor adults (p = 0.001).

CONCLUSION

Given that personal income rather than government programming is associated with disability in Rwanda, we recommend deliberately targeted services to those with disability via cash transfers, placements in disability-appropriate employment, and micro-savings programs.

Assessing the twinning model in the Rwandan Human Resources for Health Program: goal setting, satisfaction and perceived skill transfer

Background

Because of the shortage of health professionals, particularly in specialty areas, Rwanda initiated the Human Resources for Health (HRH) Program. In this program, faculty from United States teaching institutions (USF) "twin" with Rwandan Faculty (RF) to transfer skills. This paper assesses the twinning model, exploring USF and RF goal setting, satisfaction and perceptions of the effectiveness of skill transfer within the twinning model.

Methods

All USF and RF in the HRH Program from August 2012-May 2014 were invited to participate. An 85-item questionnaire for USF and 71-item questionnaire for RF were administered via Survey Monkey in April and May 2014. Associations among primary outcomes were assessed and factors related with outcomes were modeled using logistic regression.

Results

Most RF and USF reported setting goals with their twin (89 % and 71 %, respectively). Half of RF (52 %) reported effective skill transfer compared to 10 % of USF. Only 38 % of RF and 28 % of USF reported being very satisfied with the twinning model. There was significant overlap in the three operational outcomes. For RF, the following factors were associated with outcomes: for effective skill transfer, being able to communicate in a common language and working at a nursing site outside of Kigali; and for satisfaction, 7+ years of professional experience and being part of a male RF-female USF twin pair. For USF, the following factors were associated with outcomes: for setting goals, prior teaching experience; and for satisfaction, experience in low resource settings for one month or less and feeling that HRH promotes a culture of respect.

Conclusions

Twinning is the cornerstone of the HRH Program in Rwanda. These findings helped the HRH team identify key areas to improve the twinning experience including better recruitment and orientation of USF and RF, consideration of additional factors during the twinning process, provide language training support, facilitate joint twin activities and cross-cultural training and improve the site leadership buy-in and support of the program. These results can inform other programs using twinning to develop skills in the health workforce.

Electronic supplementary material

The online version of this article (doi:10.1186/s12992-016-0141-4) contains supplementary material, which is available to authorized users.

Implementation and Validation of Telepathology Triage at Cancer Referral Center in Rural Rwanda

Purpose

Connecting a cancer patient to the appropriate treatment requires the correct diagnosis provided in a timely manner. In resource-limited settings, the anatomic pathology bridge to efficient, accurate, and timely cancer care is often challenging. In this study, we present the first phase of an anatomic telepathology triage system, which was implemented and validated at the Butaro District Hospital in northern rural Rwanda.

Methods

Select cases over a 9-month period in three segments were evaluated by static image telepathology and were independently evaluated by standard glass slide histology. Each case via telepathology was classified as malignant, benign, infectious/inflammatory, or nondiagnostic and was given an exact histologic diagnosis.

Results

For cases triaged as appropriate for telepathology, correlation with classification and exact diagnosis demonstrated greater than 95% agreement over the study. Cases in which there was disagreement were analyzed for cause, and the triage process was adjusted to avoid future problems.

Conclusion

Challenges to obtaining a correct and complete diagnosis with telepathology alone included the need for immunohistochemistry, assessment of the quality of images, and the lack of images representing an entire sample. The next phase of the system will assess the effect of telepathology triage on turnaround time and the value of on-site immunohistochemistry in reducing that metric and the need for evaluation outside of telepathology.

Validation of a community-based survey assessing nonobstetric surgical conditions in Burera District, Rwanda

BACKGROUND

Validated, community-based surveillance methods to monitor epidemiologic progress in surgery have not yet been employed for surgical capacity building. The goal of this study was to create and assess the validity of a community-based questionnaire collecting data on untreated surgically correctable disease throughout Burera District, Rwanda, to accurately plan for surgical services at a district hospital.

METHODS

A structured interview to assess for 10 index surgically treatable conditions was created and underwent local focus group and pilot testing. Using a 2-stage cluster sampling design, Rwandan data collectors conducted the structured interview in 30 villages throughout the Burera District. Rwandan physicians revisited the surveyed households to perform physical examinations on all household members, used as the gold standard to validate the structured interview.

RESULTS

A total of 2,990 individuals were surveyed and 2,094 (70%) were available for physical examination. The calculated sensitivity and specificity of the survey tool were 44.5% (95% CI, 38.9-50.2%) and 97.7% (95% CI, 96.9-98.3%), respectively. The conditions with the highest sensitivity and specificity were hydrocephalus, clubfoot, and injuries/infections. Injuries/infections and hernias/hydroceles were the conditions most frequently found on examination that were not reported during the interview.

CONCLUSION

This study provides the first attempt to validate a community-based surgical surveillance tool. The finding of low sensitivity was likely related to limited access to care and poor health literacy. Accurate community-based surveys are critical to planning integrated health systems that include surgical care as a core component.

Roadmap to an effective quality improvement and patient safety program implementation in a rural hospital setting

Implementation lessons: (1) implementation of an effective quality improvement and patient safety program in a rural hospital setting requires collaboration between hospital leadership, Ministry of Health and other stakeholders. (2) Building Quality Improvement (QI) capacity to develop engaged QI teams supported by mentoring can improve quality and patient safety.

Bubble CPAP to support preterm infants in rural Rwanda: a retrospective cohort study

Background

Complications from premature birth contribute to 35 % of neonatal deaths globally; therefore, efforts to improve clinical outcomes of preterm (PT) infants are imperative. Bubble continuous positive airway pressure (bCPAP) is a low-cost, effective way to improve the respiratory status of preterm and very low birth weight (VLBW) infants. However, bCPAP remains largely inaccessible in resource-limited settings, and information on the scale-up of this technology in rural health facilities is limited. This paper describes health providers’ adherence to bCPAP protocols for PT/VLBW infants and clinical outcomes in rural Rwanda.

Methods

This retrospective chart review included all newborns admitted to neonatal units in three rural hospitals in Rwanda between February 1st and October 31st, 2013. Analysis was restricted to PT/VLBW infants. bCPAP eligibility, identification of bCPAP eligibility and complications were assessed. Final outcome was assessed overall and by bCPAP initiation status.

Results

There were 136 PT/VLBW infants. For the 135 whose bCPAP eligibility could be determined, 83 (61.5 %) were bCPAP-eligible. Of bCPAP-eligible infants, 49 (59.0 %) were correctly identified by health providers and 43 (51.8 %) were correctly initiated on bCPAP. For the 52 infants who were not bCPAP-eligible, 45 (86.5 %) were correctly identified as not bCPAP-eligible, and 46 (88.5 %) did not receive bCPAP. Overall, 90 (66.2 %) infants survived to discharge, 35 (25.7 %) died, 3 (2.2 %) were referred for tertiary care and 8 (5.9 %) had unknown outcomes. Among the bCPAP eligible infants, the survival rates were 41.8 % (18 of 43) for those in whom the procedure was initiated and 56.5 % (13 of 23) for those in whom it was not initiated. No complications of bCPAP were reported.

Conclusion

While the use of bCPAP in this rural setting appears feasible, correct identification of eligible newborns was a challenge. Mentorship and refresher trainings may improve guideline adherence, particularly given high rates of staff turnover. Future research should explore implementation challenges and assess the impact of bCPAP on long-term outcomes.

Fostering interest in research: evaluation of an introductory research seminar at hospitals in rural Rwanda

SETTINGS

Partners In Health Rwanda, in collaboration with the Ministry of Health, leads a multipronged approach to develop research capacity among health workers, particularly in rural areas.

OBJECTIVES

To describe the characteristics of participants and to assess the impact of an introductory research seminar series in three district hospitals in rural Rwanda.

DESIGN

This was a retrospective cohort study of seminar participants. Data were sourced from personnel records, assessment sheets and feedback forms.

RESULTS

A total of 126 participants, including 70 (56%) clinical and 56 (44%) non-clinical staff, attended the research seminar series; 61 (48%) received certification. Among those certified, the median assessment score on assignments was 79%. Participants read significantly more articles at 6 and 12 months (median 2 and 4 respectively, compared to 1 at baseline, P < 0.01). There was also a significant increase (P ⩽ 0.05) in self-reported involvement in research studies (28%, baseline; 59%, 12 months) and attendance at other research training (36%, baseline; 65%, 12 months).

CONCLUSION

The introductory research seminar series provided an important opportunity for engagement in research among clinical and non-clinical staff. Such an activity is a key component of a comprehensive research capacity building programme at rural sites, and serves as an entry point for more advanced research training.

Risk factors for homicide victimization in post-genocide Rwanda: a population -based case- control study

BACKGROUND

Homicide is one of the leading causes of mortality in the World. Homicide risk factors vary significantly between countries and regions. In Rwanda, data on homicide victimization is unreliable because no standardized surveillance system exists. This study was undertaken to identify the risk factors for homicide victimization in Rwanda with particular attention on the latent effects of the 1994 genocide.

METHODS

A population-based matched case-control study was conducted, with subjects enrolled prospectively from May 2011 to May 2013. Cases of homicide victimization were identified via police reports, and crime details were provided by law enforcement agencies. Three controls were matched to each case by sex, 5-year age group and village of residence. Socioeconomic and personal background data, including genocide exposure, were provided via interview of a family member or through village administrators. Conditional logistic regression, stratified by gender status, was used to identify risk factors for homicide victimization.

RESULTS

During the study period, 156 homicide victims were enrolled, of which 57 % were male and 43 % were female. The most common mechanisms of death were wounds inflicted by sharp instruments (knives or machetes; 41 %) followed by blunt force injuries (36.5 %). Final models indicated that risk of homicide victimhood increased with victim alcohol drinking patterns. There was a dose response noted for alcohol use: for minimal drinking versus none, adjusted odds ratio (aOR) = 3.1, 95%CI: 1,3-7.9; for moderate drinking versus none, aOR = 10.1, 95%CI: 3.7-24.9; and for heavy drinking versus none, aOR = 11.5, 95%CI: 3.6-36.8. Additionally, having no surviving parent (aOR = 2.7, 95%CI: 1.1-6.1), previous physical and/or sexual abuse (aOR = 28.1, 95%CI: 5.1-28.3) and drinking illicit brew and/or drug use (aOR = 7.7, 95%CI: 2.4-18.6) were associated with a higher risk of being killed. The test of interaction revealed that the variables that were significantly associated with a higher risk of homicide victimhood, did not exhibit any difference according to sex of the victim. However, the effect of belonging to a religion differed between women and men, but was significantly protective for both (aOR = 0.002, 95%CI: 0.001-0.054 and aOR = 0.20, 95%CI: 0.052-0.509, respectively).

CONCLUSION

Homicide victims in Rwanda are relatively young and the proportion of female victims is one of the highest globally. Contrary to the initial study considerations, genocide exposure (either as a survivor or perpetrator) was not a significant predictor of homicide victimization. Rather, risk factors were similar to those described in other countries, regardless of gender status. Sensitizing communities against alcohol heavy drinking, and illicit brew drinking and/or drug abuse and physical or sexual violence could reduce the homicide rate in Rwanda.

Approaches and impact of non-academic research capacity strengthening training models in sub-Saharan Africa: a systematic review

Background

Research is essential to identify and prioritize health needs and to develop appropriate strategies to improve health outcomes. In the last decade, non-academic research capacity strengthening trainings in sub-Saharan Africa, coupled with developing research infrastructure and the provision of individual mentorship support, has been used to build health worker skills. The objectives of this review are to describe different training approaches to research capacity strengthening in sub-Saharan Africa outside academic programs, assess methods used to evaluate research capacity strengthening activities, and learn about the challenges facing research capacity strengthening and the strategies/innovations required to overcome them.

Methodology

The PubMed database was searched using nine search terms and articles were included if 1) they explicitly described research capacity strengthening training activities, including information on program duration, target audience, immediate program outputs and outcomes; 2) all or part of the training program took place in sub-Saharan African countries; 3) the training activities were not a formal academic program; 4) papers were published between 2000 and 2013; and 5) both abstract and full paper were available in English.

Results

The search resulted in 495 articles, of which 450 were retained; 14 papers met all inclusion criteria and were included and analysed. In total, 4136 people were trained, of which 2939 were from Africa. Of the 14 included papers, six fell in the category of short-term evaluation period and eight in the long-term evaluation period. Conduct of evaluations and use of evaluation frameworks varied between short and long term models and some trainings were not evaluated. Evaluation methods included tests, surveys, interviews, and systems approach matrix.

Conclusions

Research capacity strengthening activities in sub-Saharan Africa outside of academic settings provide important contributions to developing in-country capacity to participate in and lead research. Institutional support, increased funds, and dedicated time for research activities are critical factors that lead to the development of successful programs. Further, knowledge sharing through scientific articles with sufficient detail is needed to enable replication of successful models in other settings.

Epidemiology of untreated non-obstetric surgical disease in Burera District, Rwanda: a cross-sectional survey

BACKGROUND

In low-income and middle-income countries, surgical epidemiology is largely undefined at the population level, with operative logs and hospital records serving as a proxy. This study assesses the distribution of surgical conditions that contribute the largest burden of surgical disease in Burera District, in northern Rwanda. We hypothesise that our results would yield higher rates of surgical disease than current estimates (from 2006) for similar low-income countries, which are 295 per 100 000 people.

METHODS

In March and May, 2012, we performed a cross-sectional study in Burera District, randomly sampling 30 villages with probability proportionate to size and randomly sampling 23 households within the selected villages. Six Rwandan surgical postgraduates and physicians conducted physical examinations on all eligible participants in sampled households. Participants were assessed for injuries or wounds, hernias, hydroceles, breast mass, neck mass, obstetric fistula, undescended testes, hypospadias, hydrocephalus, cleft lip or palate, and club foot. Ethical approval was obtained from Boston Children's Hospital (Boston, MA, USA) and the Rwandan National Ethics Committee (Kigali, Rwanda). Informed consent was obtained from all participants.

FINDINGS

Of the 2165 examined individuals, the overall prevalence of any surgical condition was 12% (95% CI 9·2-14·9) or 12 009 per 100 000 people. Injuries or wounds accounted for 55% of the prevalence and hernias or hydroceles accounted for 40%, followed by neck mass (4·2%), undescended testes (1·9%), breast mass (1·2%), club foot (1%), hypospadias (0·6%), hydrocephalus (0·6%), cleft lip or palate (0%), and obstetric fistula (0%). When comparing study participant characteristics, no statistical difference in overall prevalence was noted when examining sex, wealth, education, and travel time to the nearest hospital. Total rates of surgically treatable disease yielded a statistically significant difference compared with current estimates (p<0·001).

INTERPRETATION

Rates of surgically treatable disease are significantly higher than previous estimates in comparable low-income countries. The prevalence of surgically treatable disease is evenly distributed across demographic parameters. From these results, we conclude that strengthening the Rwandan health system's surgical capacity, particularly in rural areas, could have meaningful affect on the entire population. Further community-based surgical epidemiological studies are needed in low-income and middle-income countries to provide the best data available for health system planning.

FUNDING

The Harvard Sheldon Traveling Fellowship.

Validation of a community-based survey assessing non-obstetric surgical conditions in Burera District, Rwanda

BACKGROUND

Community-based surveillance methods to monitor epidemiological progress in surgery have not yet been employed for surgical capacity building. The aim of this study was to create and assess the validity of a questionnaire that collected data for untreated surgically correctable diseases throughout Burera District, northern Rwanda, to accurately plan for surgical services.

METHODS

A structured interview to assess for the presence or absence of ten index surgically treatable conditions (breast mass, cleft lip/palate, club foot, hernia or hydrocele [adult and paediatric]), hydrocephalus, hypospadias, injuries or wounds, neck mass, undescended testes, and vaginal fistula) was created. The interview was built based on previously validated questionnaires, forward and back translated into the local language and underwent focus group augmentation and pilot testing. In March and May, 2012, data collectors conducted the structured interviews with a household representative in 30 villages throughout Burera District, selected using a two-stage cluster sampling design. Rwandan physicians revisited the surveyed households to perform physical examinations on all household members, used as the gold standard to validate the structured interview. Ethical approval was obtained from Boston Children's Hospital (Boston, MA, USA) and the Rwandan National Ethics Committee (Kigali, Rwanda). Informed consent was obtained from all households.

FINDINGS

2990 individuals were surveyed, a 97% response rate. 2094 (70%) individuals were available for physical examination. The calculated overall sensitivity of the structured interview tool was 44·5% (95% CI 38·9-50·2) and the specificity was 97·7% (96·9-98·3%; appendix). The positive predictive value was 70% (95% CI 60·5-73·5), whereas the negative predictive value was 91·3% (90·0-92·5). The conditions with the highest sensitivity and specificity, respectively, were hydrocephalus (100% and 100%), clubfoot (100% and 99·8%), injuries or wounds (54·7% and 98·9%), and hypospadias (50% and 100%). Injuries or wounds and hernias or hydroceles were the conditions most frequently identified on examination that were not reported during the interview (appendix).

INTERPRETATION

To the best of our knowledge, this study provides the first attempt to validate a community-based surgical surveillance tool. The finding of low sensitivity limits the use of the tool, which will require further revision, and calls into question previously published unvalidated community surgical survey data. To improve validation of community-based surveys, community education efforts on common surgically treatable conditions are needed in conjunction with increased access to surgical care. Accurate community-based surveys are crucial to integrated health system planning that includes surgical care as a core component.

FUNDING

The Harvard Sheldon Traveling Fellowship.

Exposure to genocide and risk of suicide in Rwanda: a population-based case-control study

Background

In Rwanda, an estimated one million people were killed during the 1994 genocide, leaving the country shattered and social fabric destroyed. Large-scale traumatic events such as wars and genocides have been linked to endemic post-traumatic stress disorder, depression and suicidality. The study objective was to investigate whether the 1994 genocide exposure is associated with suicide in Rwanda.

Methods

We conducted a population-based case–control study. Suicide victims were matched to three living controls for sex, age and residential location. Exposure was defined as being a genocide survivor, having suffered physical/sexual abuse in the genocide, losing a first-degree relative in the genocide, having been convicted for genocide crimes or having a first-degree relative convicted for genocide. From May 2011 to May 2013, 162 cases and 486 controls were enrolled countrywide. Information was collected from the police, local village administrators and family members.

Results

After adjusting for potential confounders, having been convicted for genocide crimes was a significant predictor for suicide (OR=17.3, 95% CI 3.4 to 88.1). Being a survivor, having been physically or sexually abused during the genocide, and having lost a first-degree family member to genocide were not significantly associated with suicide.

Conclusions

These findings demonstrate that individuals convicted for genocide crimes are experiencing continued psychological disturbances that affect their social reintegration into the community even 20 years after the event. Given the large number of genocide perpetrators reintegrated after criminal courts and Gacaca traditional reconciling trials, suicide could become a serious public health burden if preventive remedial action is not identified.

Diagnosis of cancer in rural Rwanda: early outcomes of a phased approach to implement anatomic pathology services in resource-limited settings

OBJECTIVES

Adequate pathology services are a prerequisite to accurate cancer diagnoses and tailoring appropriate treatment. Limitations in skilled personnel and infrastructure are among the challenges faced by developing countries. We describe a stepwise implementation of anatomic pathology laboratory services at Butaro District Hospital, designated as a Cancer Center of Excellence in rural Rwanda.

METHODS

The phased approach to developing pathology services up to December 2012 is described. A retrospective review of specimens submitted to Butaro District Hospital between July 1, 2012, and December 31, 2012, was conducted. Patient clinical characteristics and sociodemographics are also described.

RESULTS

During the study period, a total of 437 tissue specimens were submitted. Among these, 143 (32.7%) were from male patients, 244 (55.8%) were confirmed as malignant, 163 (37.3%) were benign, 28 (6.4%) were inconclusive, and two (0.5%) results were not available at the time of analysis. The median time from specimen receipt at Butaro to final reporting was 32 days (range, 7-193 days; interquartile range, 23-44 days).

CONCLUSIONS

Our experience demonstrates that anatomic pathology services can be established in resource-limited settings and local capacity can be built to support accurate diagnoses. Our approach included leveraging partnerships, volunteer experts, and task shifting and will be expanded to include telepathology.

Assessing predictors of delayed antenatal care visits in Rwanda: a secondary analysis of Rwanda demographic and health survey 2010

Background

Early initiation of antenatal care (ANC) can reduce common maternal complications and maternal and perinatal mortality. Though Rwanda demonstrated a remarkable decline in maternal mortality and 98% of Rwandan women receive antenatal care from a skilled provider, only 38% of women have an ANC visit in their first three months of pregnancy. This study assessed factors associated with delayed ANC in Rwanda.

Methods

This is a cross-sectional study using data collected during the 2010 Rwanda DHS from 6,325 women age 15–49 that had at least one birth in the last five years. Factors associated with delayed ANC were identified using a multivariable logistic regression model using manual backward stepwise regression. Analysis was conducted in Stata v12 applying survey commands to account for the complex sample design.

Results

Several factors were significantly associated with delayed ANC including having many children (4–6 children, OR = 1.42, 95% CI: 1.22, 1.65; or more than six children, OR = 1.57, 95% CI: 1.24, 1.99); feeling that distance to health facility is a problem (OR = 1.20, 95% CI: 1.04, 1.38); and unwanted pregnancy (OR = 1.41, 95% CI: 1.26, 1.58). The following were protective against delayed ANC: having an ANC at a private hospital or clinic (OR = 0.29, 95% CI: 0.15, 0.56); being married (OR = 0.85, 95% CI: 0.75, 0.96), and having public mutuelle health insurance (OR = 0.81, 95% CI: 0.71, 0.92) or another type of insurance (OR = 0.33, 95% CI: 0.23, 0.46).

Conclusion

This analysis revealed potential barriers to ANC service utilization. Distance to health facility remains a major constraint which suggests a great need of infrastructure and decentralization of maternal ANC to health posts and dispensaries. Interventions such as universal health insurance coverage, family planning, and community maternal health system are underway and could be part of effective strategies to address delays in ANC.

Clinical mentorship to improve pediatric quality of care at the health centers in rural Rwanda: a qualitative study of perceptions and acceptability of health care workers

BACKGROUND

Despite evidence supporting Integrated Management of Childhood Illness (IMCI) as a strategy to improve pediatric care in countries with high child mortality, its implementation faces challenges related to lack of or poor post-didactic training supervision and gaps in necessary supporting systems. These constraints lead to health care workers' inability to consistently translate IMCI knowledge and skills into practice. A program providing mentoring and enhanced supervision at health centers (MESH), focusing on clinical and systems improvement was implemented in rural Rwanda as a strategy to address these issues, with the ultimate goal of improving the quality of pediatric care at rural health centers. We explored perceptions of MESH from the perspective of IMCI clinical mentors, mentees, and district clinical leadership.

METHODS

We conducted focus group discussions with 40 health care workers from 21 MESH-supported health centers. Two FGDs in each district were carried out, including one for nurses and one for director of health centers. District medical directors and clinical mentors had individual in-depth interviews. We performed a hermeneutic analysis using Atlas.ti v5.2.

RESULTS

Study participants highlighted program components in five key areas that contributed to acceptability and impact, including: 1) Interactive, collaborative capacity-building, 2) active listening and relationships, 3) supporting not policing, 4) systems improvement, and 5) real-time feedback. Staff turn-over, stock-outs, and other facility/systems gaps were identified as barriers to MESH and IMCI implementation.

CONCLUSION

Health care workers reported high acceptance and positive perceptions of the MESH model as an effective strategy to build their capacity, bridge the gap between knowledge and practice in pediatric care, and address facility and systems issues. This approach also improved relationships between the district supervisory team and health center-based care providers. Despite some challenges, many perceived a strong benefit on clinical performance and outcomes. This study can inform program implementers and policy makers of key components needed for developing similar health facility-based mentorship interventions and potential barriers and resistance which can be proactively addressed to ensure success.

Effect of shortened Integrated Management of Childhood Illness training on classification and treatment of under-five children seeking care in Rwanda

Background

Integrated Management of Childhood Illness (IMCI) is an effective 11-day standard training; however, due to budgetary expenses and human resource constraints, many health professionals cannot take 11 days off work. As a result, shortened training curriculums (6-day) have been proposed. We used a cross-sectional study to evaluate the effect of this shortened training on appropriate IMCI classification and treatment of under-five childhood illness management in Rwanda.

Methods

A cross-sectional study was conducted in 22 health centers in Rwanda, comparing data from 121 nurses, where 55 nurses completed the 11-day and 66 nurses completed the 6-day training. Among 768 children, we evaluated clinical outcomes from May 2011 to April 2012. Descriptive statistics were used to display the sociodemographic characteristics of health providers; including level of education, sex, age, and professional experiences. Bivariable and multivariable analyses were used to test for differences between nurses in the 6-day versus 11-day training on the appropriate classification and treatment of childhood illness.

Results

Our findings show that at the bivariable level and after controlling for confounders in the multivariable analysis, the only significant differences detected between nurses in the long and short training was the classification of fever (adjusted odds ratio [aOR] 0.7, 95% confidence interval [CI] 0.64–0.75) and treatment of pneumonia (aOR 0.8, 95% CI 0.70–0.89). Nurses in the short training had lower odds of inappropriate misclassification and treatment for these two conditions.

Conclusion

There was no difference in classification and treatment of childhood illness among nurses who completed the standard and short IMCI training courses. Short-training could be a more cost-saving option for health facilities without compromising the key outcomes related to case management.