Health and human rights in Chin State, Western Burma: a population-based assessment using multistaged household cluster sampling

Gridded population survey sampling: a systematic scoping review of the field and strategic research agenda

INTRODUCTION

In low- and middle-income countries (LMICs), household survey data are a main source of information for planning, evaluation, and decision-making. Standard surveys are based on censuses, however, for many LMICs it has been more than 10 years since their last census and they face high urban growth rates. Over the last decade, survey designers have begun to use modelled gridded population estimates as sample frames. We summarize the state of the emerging field of gridded population survey sampling, focussing on LMICs.

METHODS

We performed a systematic scoping review in Scopus of specific gridded population datasets and "population" or "household" "survey" reports, and solicited additional published and unpublished sources from colleagues.

RESULTS

We identified 43 national and sub-national gridded population-based household surveys implemented across 29 LMICs. Gridded population surveys used automated and manual approaches to derive clusters from WorldPop and LandScan gridded population estimates. After sampling, some survey teams interviewed all households in each cluster or segment, and others sampled households from larger clusters. Tools to select gridded population survey clusters include the GridSample R package, Geo-sampling tool, and GridSample.org. In the field, gridded population surveys generally relied on geographically accurate maps based on satellite imagery or OpenStreetMap, and a tablet or GPS technology for navigation.

CONCLUSIONS

For gridded population survey sampling to be adopted more widely, several strategic questions need answering regarding cell-level accuracy and uncertainty of gridded population estimates, the methods used to group/split cells into sample frame units, design effects of new sample designs, and feasibility of tools and methods to implement surveys across diverse settings.

Using gridded population and quadtree sampling units to support survey sample design in low-income settings

BACKGROUND

Household surveys are the main source of demographic, health and socio-economic data in low- and middle-income countries (LMICs). To conduct such a survey, census population information mapped into enumeration areas (EAs) typically serves a sampling frame from which to generate a random sample. However, the use of census information to generate this sample frame can be problematic as in many LMIC contexts, such data are often outdated or incomplete, potentially introducing coverage issues into the sample frame. Increasingly, where census data are outdated or unavailable, modelled population datasets in the gridded form are being used to create household survey sampling frames.

METHODS

Previously this process was done by either sampling from a set of the uniform grid cells (UGC) which are then manually subdivided to achieve the desired population size, or by sampling very small grid cells then aggregating cells into larger units to achieve a minimum population per survey cluster. The former approach is time and resource-intensive as well as results in substantial heterogeneity in the output sampling units, while the latter can complicate the calculation of unbiased sampling weights. Using the context of Somalia, which has not had a full census since 1987, we implemented a quadtree algorithm for the first time to create a population sampling frame. The approach uses gridded population estimates and it is based on the idea of a quadtree decomposition in which an area successively subdivided into four equal size quadrants, until the content of each quadrant is homogenous.

RESULTS

The quadtree approach used here produced much more homogeneous sampling units than the UGC (1 × 1 km and 3 × 3 km) approach. At the national and pre-war regional scale, the standard deviation and coefficient of variation, as indications of homogeneity, were calculated for the output sampling units using quadtree and UGC 1 × 1 km and 3 × 3 km approaches to create the sampling frame and the results showed outstanding performance for quadtree approach.

CONCLUSION

Our approach reduces the manual burden of manually subdividing UGC into highly populated areas, while allowing for correct calculation of sampling weights. The algorithm produces a relatively homogenous population counts within the sampling units, reducing the variation in the weights and improving the precision of the resulting estimates. Furthermore, a protocol of creating approximately equal-sized blocks and using tablets for randomized selection of a household in each block mitigated potential selection bias by enumerators. The approach shows labour, time and cost-saving and points to the potential use in wider contexts.

Migration patterns & their associations with health and human rights in eastern Myanmar after political transition: results of a population-based survey using multistaged household cluster sampling

Background

Myanmar transitioned to a nominally civilian government in March 2011. It is unclear how, if at all, this political change has impacted migration at the household level.

Methods

We present household-level in- and out-migration data gathered during the Eastern Burma Retrospective Mortality Survey (EBRMS) conducted in 2013. Household level in-and out-migration information within the previous year was gathered via a cross-sectional, retrospective, multi-stage population-based cluster randomized survey conducted in eastern Myanmar. Univariate, bivariate and regression analyses were conducted.

Results

We conducted a cross-sectional survey of 6620 households across Eastern Myanmar between July and September of 2013. Out-migration outstripped in-migration more than 6:1 overall during the year prior to the survey - for international migration this ratio was 29:1. Most in-migrants had moved to their present location in the study area from other areas in Myanmar (87%). Only 11.3% (27 individuals) had returned from another country (Thailand). Those who migrated out of eastern Myanmar during the previous year were more likely to be male (55.2%), and three times more likely to be between the ages of 15-25 (49.5%) than non-migrants. The primary reason cited for a return to the household was family (26.3%) followed by work (23.2%). The primary reason cited for migrating out of the household was for education (46.4%) followed by work (40.2%). Respondents from households that reported out-migration in the past year were more likely to screen positive for depressive symptoms than households with no migration (PR 1.85; 95% CI 1.16, 2.97). Women in households with in-migration were more likely to be malnourished and had a higher unmet need for contraception. Forced labor, one subset of human rights violations experienced by this population, was reported by more in-migrant (8%) than out-migrant households (2.2%), though this finding did not reach statistical significance.

Conclusions

These analyses suggest that opportunities for employment and education are the primary drivers of migration out of the household, despite an overall improvement in stability and decrease in prevalence of human rights violations found by EBRMS 2013. Additionally, migration into and out of households in eastern Myanmar is associated with changes in health outcomes.

The HPAfrica protocol: Assessment of health behaviour and population-based socioeconomic, hygiene behavioural factors - a standardised repeated cross-sectional study in multiple cohorts in sub-Saharan Africa

INTRODUCTION

The objective of the Health Population Africa (HPAfrica) study is to determine health behaviour and population-based factors, including socioeconomic, ethnographic, hygiene and sanitation factors, at sites of the Severe Typhoid Fever in Africa (SETA) programme. SETA aims to investigate healthcare facility-based fever surveillance in Burkina Faso, the Democratic Republic of the Congo, Ethiopia, Ghana, Madagascar and Nigeria. Meaningful disease burden estimates require adjustment for health behaviour patterns, which are assumed to vary among a study population.

METHODS AND ANALYSIS

For the minimum sample size of household interviews required, the assumptions of an infinite population, a design effect and age-stratification and sex-stratification are considered. In the absence of a population sampling frame or household list, a spatial approach will be used to generate geographic random points with an Aeronautical Reconnaissance Coverage Geographic Information System tool. Printouts of Google Earth Pro satellite imagery visualise these points. Data of interest will be assessed in different seasons by applying population-weighted stratified sampling. An Android-based application and a web service will be developed for electronic data capturing and synchronisation with the database server in real time. Sampling weights will be computed to adjust for possible differences in selection probabilities. Descriptive data analyses will be performed in order to assess baseline information of each study population and age-stratified and sex-stratified health behaviour. This will allow adjusting disease burden estimates. In addition, multivariate analyses will be applied to look into associations between health behaviour, population-based factors and the disease burden as determined in the SETA study.

ETHICS AND DISSEMINATION

Ethic approvals for this protocol were obtained by the Institutional Review Board of the International Vaccine Institute (No. 2016-0003) and by all collaborating institutions of participating countries. It is anticipated to disseminate findings from this study through publication on a peer-reviewed journal.

Ethnomedicinal plants and traditional knowledge among three Chin indigenous groups in Natma Taung National Park (Myanmar)

ETHNOPHARMACOLOGICAL RELEVANCE

This research describes the ethnomedicinal plants as used by three Chin indigenous groups inhabiting areas at Natma Taung National Park (NTNP) in bio-culturally diverse Myanmar. The aim of this study was (1) to identify wild medicinal species and evaluate their local importance in local peoples' healthcare as well as in protected area conservation; and (2) to compare traditional medicinal plant knowledge among the informants.

MATERIALS AND METHODS

A total of 206 Müün, Ng'gah and Daai informants from 20 villages were interviewed using semi-structured questionnaires. Species use-reports were computed to determine plant local importance and the informant consensus factor. Descriptive and the inferential statistics Mann-Whitney U and Kruskal-Wallis tests were employed to evaluate and compare the informants' traditional medicinal knowledge.

RESULTS

A total of 75 wild ethnomedicinal taxa in 40 plant families across 16 ICPC-based disease categories were recorded. Species which recorded the highest number of use-reports appeared to play an important role not only in informants' primary healthcare and in park conservation, but also in local livelihood.

CONCLUSION

This study presents the diversity of ethnomedicinal plants and their local importance in Chin indigenous peoples' healthcare. This paper also recognizes the value of these plants and the local traditional knowledge for the conservation and management of NTNP.

Residential scene classification for gridded population sampling in developing countries using deep convolutional neural networks on satellite imagery

Background

Conducting surveys in low- and middle-income countries is often challenging because many areas lack a complete sampling frame, have outdated census information, or have limited data available for designing and selecting a representative sample. Geosampling is a probability-based, gridded population sampling method that addresses some of these issues by using geographic information system (GIS) tools to create logistically manageable area units for sampling. GIS grid cells are overlaid to partition a country’s existing administrative boundaries into area units that vary in size from 50 m × 50 m to 150 m × 150 m. To avoid sending interviewers to unoccupied areas, researchers manually classify grid cells as “residential” or “nonresidential” through visual inspection of aerial images. “Nonresidential” units are then excluded from sampling and data collection. This process of manually classifying sampling units has drawbacks since it is labor intensive, prone to human error, and creates the need for simplifying assumptions during calculation of design-based sampling weights. In this paper, we discuss the development of a deep learning classification model to predict whether aerial images are residential or nonresidential, thus reducing manual labor and eliminating the need for simplifying assumptions.

Results

On our test sets, the model performs comparable to a human-level baseline in both Nigeria (94.5% accuracy) and Guatemala (96.4% accuracy), and outperforms baseline machine learning models trained on crowdsourced or remote-sensed geospatial features. Additionally, our findings suggest that this approach can work well in new areas with relatively modest amounts of training data.

Conclusions

Gridded population sampling methods like geosampling are becoming increasingly popular in countries with outdated or inaccurate census data because of their timeliness, flexibility, and cost. Using deep learning models directly on satellite images, we provide a novel method for sample frame construction that identifies residential gridded aerial units. In cases where manual classification of satellite images is used to (1) correct for errors in gridded population data sets or (2) classify grids where population estimates are unavailable, this methodology can help reduce annotation burden with comparable quality to human analysts.

Public health information in crisis-affected populations: a review of methods and their use for advocacy and action

Valid and timely information about various domains of public health underpins the effectiveness of humanitarian public health interventions in crises. However, obstacles including insecurity, insufficient resources and skills for data collection and analysis, and absence of validated methods combine to hamper the quantity and quality of public health information available to humanitarian responders. This paper, the second in a Series of four papers, reviews available methods to collect public health data pertaining to different domains of health and health services in crisis settings, including population size and composition, exposure to armed attacks, sexual and gender-based violence, food security and feeding practices, nutritional status, physical and mental health outcomes, public health service availability, coverage and effectiveness, and mortality. The paper also quantifies the availability of a minimal essential set of information in large armed conflict and natural disaster crises since 2010: we show that information was available and timely only in a small minority of cases. On the basis of this observation, we propose an agenda for methodological research and steps required to improve on the current use of available methods. This proposition includes setting up a dedicated interagency service for public health information and epidemiology in crises.

GridSample: an R package to generate household survey primary sampling units (PSUs) from gridded population data

BACKGROUND

Household survey data are collected by governments, international organizations, and companies to prioritize policies and allocate billions of dollars. Surveys are typically selected from recent census data; however, census data are often outdated or inaccurate. This paper describes how gridded population data might instead be used as a sample frame, and introduces the R GridSample algorithm for selecting primary sampling units (PSU) for complex household surveys with gridded population data. With a gridded population dataset and geographic boundary of the study area, GridSample allows a two-step process to sample "seed" cells with probability proportionate to estimated population size, then "grows" PSUs until a minimum population is achieved in each PSU. The algorithm permits stratification and oversampling of urban or rural areas. The approximately uniform size and shape of grid cells allows for spatial oversampling, not possible in typical surveys, possibly improving small area estimates with survey results.

RESULTS

We replicated the 2010 Rwanda Demographic and Health Survey (DHS) in GridSample by sampling the WorldPop 2010 UN-adjusted 100 m × 100 m gridded population dataset, stratifying by Rwanda's 30 districts, and oversampling in urban areas. The 2010 Rwanda DHS had 79 urban PSUs, 413 rural PSUs, with an average PSU population of 610 people. An equivalent sample in GridSample had 75 urban PSUs, 405 rural PSUs, and a median PSU population of 612 people. The number of PSUs differed because DHS added urban PSUs from specific districts while GridSample reallocated rural-to-urban PSUs across all districts.

CONCLUSIONS

Gridded population sampling is a promising alternative to typical census-based sampling when census data are moderately outdated or inaccurate. Four approaches to implementation have been tried: (1) using gridded PSU boundaries produced by GridSample, (2) manually segmenting gridded PSU using satellite imagery, (3) non-probability sampling (e.g. random-walk, "spin-the-pen"), and random sampling of households. Gridded population sampling is in its infancy, and further research is needed to assess the accuracy and feasibility of gridded population sampling. The GridSample R algorithm can be used to forward this research agenda.

Growth data of underprivileged children living in rural areas of Chin State, Burma/Myanmar, compared to the WHO reference growth standards: an observational study

OBJECTIVES

To explore growth data (height-for-age, weight-for-age and BMI-for-age) of children living in poor socioeconomic conditions in rural areas of Chin State, Burma/Myanmar; and to compare these data with the growth and development z-score (GDZ) values for school-aged children and adolescents, provided by the WHO.

SETTING

A support and educational programme, run by the Swedish association Chin Development and Research Society (CDRS), was carried out among underprivileged school-aged children, unable to attend school without economic and practical support, living in villages and remote areas in Chin State.

PARTICIPANTS

Community leaders who were well familiar with the citizens in the community identified children in need of this support. Other community members could also suggest or apply for this. The sample includes all participating children in the CDRS programme at the time of the data collection in six townships. The children were placed in host families, close to a suitable school. Two samples with a total of 639 children from 144 villages and remote areas were obtained:1. Children in the CDRS Chin Programme (CCP) (2007-2010) comprised 558 children: 50% girls and boys.2. Children in the Chin Society (CCS) (2010) comprised 81 children: 44% girls and 56% boys.

PRIMARY OUTCOME MEASURES

Growth data.

RESULTS

All growth data from both groups deviated significantly from the WHO standard references (p ≤ 0.001). The prevalence of stunting (height-for-age ≤-2SD) was 52% among girls and 68% among boys. High levels of wasting (weight-for-age ≤-2SD) were found among girls 29% and boys 36% aged 5-10 years. In addition, severe thinness (BMI-for-age ≤-2SD) was found among girls 31% and boys 44%, all results to be compared to the expected 2.27%.

CONCLUSIONS

Many more than expected-according to the WHO reference values-in CCP and CCS suffered from stunting, wasting and thinness.

Militarization, human rights violations and community responses as determinants of health in southeastern Myanmar: results of a cluster survey

Background

The Myanmar army and ethnic armed groups agreed to a preliminary ceasefire in 2012, but a heavy military presence remains in southeastern Myanmar. Qualitative data suggested this militarization can result in human rights abuses in the absence of armed engagements between the parties, and that rural ethnic civilians use a variety of self-protection strategies to avoid these abuses or reduce their negative impacts. We used data from a household survey to determine prevalence of select self-protection activities and to examine exposure to armed groups, human rights violations and self-protection activities as determinants of health in southeastern Myanmar.

Methods and findings

Data collected from 463 households via a two-stage cluster survey of conflict-affected areas in eastern Myanmar in January 2012, were analyzed using logistic regression models to identify associations between exposure to state and non-state armed groups, village self-protection, human rights abuses and health outcomes. Close proximity to a military base was associated with human rights abuses (PRR 1.30, 95 % CI: 1.14-1.48), inadequate food production (PRR 1.08, 95 % CI: 1.03-1.13), inability to access health care (PRR 1.29, 95 % CI: 1.04-1.60) and diarrhea (PRR 1.15, 95 % CI: 1.05-1.27. Direct exposure to armed groups was associated with household hunger (PRR1.71, 95 % CI: 1.30-2.23). Among households that reported no human rights abuses, risk of household hunger (PRR 5.64, 95 % CI: 1.88-16.91), inadequate food production (PRR 1.95, 95 % CI: 1.11-3.41) and diarrhea (PRR 2.53, 95 % CI: 1.45-4.42) increased when neighbors’ households reported experiencing human rights abuses. Households in villages that reported negotiating with the Myanmar army had lower risk of human rights violations (PRR 0.91, 95 % CI: 0.85-0.98), household hunger (PRR 0.85, 95 % CI: 0.74-0.96), inadequate food production (PRR 0.93, 95 % CI:0.89-0.98) and diarrhea (PRR 0.89, 95 % CI:0.82-0.97). Stratified analysis suggests that self-protection strategies may modify the effect of exposure to armed groups on risk of human rights violations and some health outcomes.

Conclusion

Militarization may negatively affect health in southeastern Myanmar, and village self-protection activities may reduce these impacts. As southeastern Myanmar opens to international health and development interventions, implementing agencies should consider militarization as a determinant of health and design interventions that can mediate its effects. Such interventions should take into account existing self-protection strategies, seek to provide support where possible and, at all times, take care not to unintentionally undermine them.

Electronic supplementary material

The online version of this article (doi:10.1186/s13031-015-0059-0) contains supplementary material, which is available to authorized users.

Health and Human Rights in Karen State, Eastern Myanmar

BACKGROUND

Decades of conflict in eastern Myanmar have resulted in high prevalence of human rights violations and poor health outcomes. While recent ceasefire agreements have reduced conflict in this area, it is unknown whether this has resulted in concomitant reductions in human rights violations.

METHODS AND FINDINGS

We conducted a two-stage cluster survey of 686 households in eastern Myanmar to assess health status, access to healthcare, food security, exposure to human rights violations and identification of alleged perpetrators over the 12 months prior to January 2012, a period of near-absence of conflict in this region. Household hunger (FANTA-2 scale) was moderate/high in 91 (13.2%) households, while the proportion of households reporting food shortages in each month of 2011 ranged from 19.9% in December to 47.0% in September, with food insecurity peaking just prior to the harvest. Diarrhea prevalence in children was 14.2% and in everyone it was 5.8%. Forced labor was the most common human rights violation (185 households, 24.9%), and 210 households (30.6%) reported experiencing one or more human rights violations in 2011. Multiple logistic regression analysis identified associations between human rights violations and poor health outcomes.

CONCLUSION

Human rights violations and their health consequences persist despite reduced intensity of conflict in eastern Myanmar. Ceasefire agreements should include language that protects human rights, and reconciliation efforts should address the health consequences of decades of human rights violations.

Piloting community-based medical care for survivors of sexual assault in conflict-affected Karen State of eastern Burma

Background

Given the challenges to ensuring facility-based care in conflict settings, the Women’s Refugee Commission and partners have been pursuing a community-based approach to providing medical care to survivors of sexual assault in Karen State, eastern Burma. This new model translates the 2004 World Health Organization’s Clinical Management of Rape Survivors facility-based protocol to the community level through empowering community health workers to provide post-rape care. The aim of this innovative study is to examine the safety and feasibility of community-based medical care for survivors of sexual assault to contribute to building an evidence base on alternative models of care in humanitarian settings.

Methods

A process evaluation was implemented from July-October 2011 to gather qualitative feedback from trained community health workers, traditional birth attendants, and community members. Two focus group discussions were conducted among the highest cadre health care workers from the pilot and non-pilot sites. In Karen State, eight focus group discussions were convened among traditional birth attendants and 10 among women and men of reproductive age.

Results

Qualitative feedback contributed to an understanding of the model’s feasibility. Pilot site community health workers showed interest in providing community-based care for survivors of sexual assault. Traditional birth attendants attested to the importance of making this care available. Community health workers were deeply aware of the need to maintain confidentiality and offer compassionate care. They did not raise safety as an excess concern in the provision of treatment.

Conclusions

Data speak to the promising “feasibility” of community-based post-rape care. More time, awareness-raising, and a larger catchment population are necessary to answer the safety perspective. The pilot is an attempt to translate facility-based protocol to the community level to offer solutions for settings where traditional methods of post-rape care are not accessible for women and girls that need it most.

Molecular surveillance for drug-resistant Plasmodium falciparum in clinical and subclinical populations from three border regions of Burma/Myanmar: cross-sectional data and a systematic review of resistance studies

BACKGROUND

Confirmation of artemisinin-delayed parasite clearance in Plasmodium falciparum along the Thai-Myanmar border has inspired a global response to contain and monitor drug resistance to avert the disastrous consequences of a potential spread to Africa. However, resistance data from Myanmar are sparse, particularly from high-risk areas where limited health services and decades of displacement create conditions for resistance to spread. Subclinical infections may represent an important reservoir for resistance genes that confer a fitness disadvantage relative to wild-type alleles. This study estimates the prevalence of resistance genotypes in three previously unstudied remote populations in Myanmar and tests the a priori hypothesis that resistance gene prevalence would be higher among isolates collected from subclinical infections than isolates collected from febrile clinical patients. A systematic review of resistance studies is provided for context.

METHODS

Community health workers in Karen and Kachin States and an area spanning the Indo-Myanmar border collected dried blood spots from 988 febrile clinical patients and 4,591 villagers with subclinical infection participating in routine prevalence surveys. Samples positive for P. falciparum 18 s ribosomal RNA by real-time PCR were genotyped for P. falciparum multidrug resistance protein (pfmdr1) copy number and the pfcrt K76T polymorphism using multiplex real-time PCR.

RESULTS

Pfmdr1 copy number increase and the pfcrt K76 polymorphism were determined for 173 and 269 isolates, respectively. Mean pfmdr1 copy number was 1.2 (range: 0.7 to 3.7). Pfmdr1 copy number increase was present in 17.5%, 9.6% and 11.1% of isolates from Karen and Kachin States and the Indo-Myanmar border, respectively. Pfmdr1 amplification was more prevalent in subclinical isolates (20.3%) than clinical isolates (6.4%, odds ratio 3.7, 95% confidence interval 1.1 - 12.5). Pfcrt K76T prevalence ranged from 90-100%.

CONCLUSIONS

Community health workers can contribute to molecular surveillance of drug resistance in remote areas of Myanmar. Marginal and displaced populations under-represented among previous resistance investigations can and should be included in resistance surveillance efforts, particularly once genetic markers of artemisinin-delayed parasite clearance are identified. Subclinical infections may contribute to the epidemiology of drug resistance, but determination of gene amplification from desiccated filter samples requires further validation when DNA concentration is low.

Medical evidence of human rights violations against non-Arabic-speaking civilians in Darfur: a cross-sectional study

BACKGROUND

Ongoing conflict in the Darfur region of Sudan has resulted in a severe humanitarian crisis. We sought to characterize the nature and geographic scope of allegations of human rights violations perpetrated against civilians in Darfur and to evaluate their consistency with medical examinations documented in patients' medical records.

METHODS AND FINDINGS

This was a retrospective review and analysis of medical records from all 325 patients seen for treatment from September 28, 2004, through December 31, 2006, at the Nyala-based Amel Centre for Treatment and Rehabilitation of Victims of Torture, the only dedicated local provider of free clinical and legal services to civilian victims of torture and other human rights violations in Darfur during this time period. Among 325 medical records identified and examined, 292 (89.8%) patients from 12 different non-Arabic-speaking tribes disclosed in the medical notes that they had been attacked by Government of Sudan (GoS) and/or Janjaweed forces. Attacks were reported in 23 different rural council areas throughout Darfur. Nearly all attacks (321 [98.8%]) were described as having occurred in the absence of active armed conflict between Janjaweed/GoS forces and rebel groups. The most common alleged abuses were beatings (161 [49.5%]), gunshot wounds (140 [43.1%]), destruction or theft of property (121 [37.2%]), involuntary detainment (97 [29.9%]), and being bound (64 [19.7%]). Approximately one-half (36 [49.3%]) of all women disclosed that they had been sexually assaulted, and one-half of sexual assaults were described as having occurred in close proximity to a camp for internally displaced persons. Among the 198 (60.9%) medical records that contained sufficient detail to enable the forensic medical reviewers to render an informed judgment, the signs and symptoms in all of the medical records were assessed to be consistent with, highly consistent with, or virtually diagnostic of the alleged abuses.

CONCLUSIONS

Allegations of widespread and sustained torture and other human rights violations by GoS and/or Janjaweed forces against non-Arabic-speaking civilians were corroborated by medical forensic review of medical records of patients seen at a local non-governmental provider of free clinical and legal services in Darfur. Limitations of this study were that patients seen in this clinic may not have been a representative sample of persons alleging abuse by Janjaweed/GoS forces, and that most delayed presenting for care. The quality of documentation was similar to that available in other conflict/post-conflict, resource-limited settings.