• An alternative classification to mixture modeling for longitudinal counts or binary measures

      Subtil, F; Boussari, O; Bastard, M; Etard, J-F; Ecochard, R; Génolini, C (SAGE Publications, 2014-09-01)
      Classifying patients according to longitudinal measures, or trajectory classification, has become frequent in clinical research. The k-means algorithm is increasingly used for this task in case of continuous variables with standard deviations that do not depend on the mean. One feature of count and binary data modeled by Poisson or logistic regression is that the variance depends on the mean; hence, the within-group variability changes from one group to another depending on the mean trajectory level. Mixture modeling could be used here for classification though its main purpose is to model the data. The results obtained may change according to the main objective. This article presents an extension of the k-means algorithm that takes into account the features of count and binary data by using the deviance as distance metric. This approach is justified by its analogy with the classification likelihood. Two applications are presented with binary and count data to show the differences between the classifications obtained with the usual Euclidean distance versus the deviance distance.
    • Applying the ICMJE authorship criteria to operational research in low-income countries: the need to engage programme managers and policy makers [letter]

      Zachariah, R; Reid, T; Van den Bergh, R; Dahmane, A; Kosgei, R J; Hinderaker, S G; Tayler-Smith, K; Manzi, M; Kizito, W; Khogali, M; Kumar, A M V; Baruani, B; Bishinga, A; Kilale, A M; Nqobili, M; Patten, G; Sobry, A; Cheti, E; Nakanwagi, A; Enarson, D A; Edginton, M E; Upshur, R; Harries, A D; Medical Department (Operational Research Unit), Medecins sans Frontieres, Operational Centre Brussels, MSF-Luxembourg, Luxembourg, Luxembourg. (Wiley-Blackwell, 2013-05-30)
    • Are Rapid Population Estimates Accurate? A Field Trial of Two Different Assessment Methods.

      Grais, R; Coulombier, D; Ampuero, J; Lucas, M; Barretto, A; Jacquier, G; Diaz, F; Balandine, S; Mahoudeau, C; Brown, V; Epicentre, Paris, France. rebecca.grais@epicentre.msf.org (Published by Wiley-Blackwell, 2006-09)
      Emergencies resulting in large-scale displacement often lead to populations resettling in areas where basic health services and sanitation are unavailable. To plan relief-related activities quickly, rapid population size estimates are needed. The currently recommended Quadrat method estimates total population by extrapolating the average population size living in square blocks of known area to the total site surface. An alternative approach, the T-Square, provides a population estimate based on analysis of the spatial distribution of housing units taken throughout a site. We field tested both methods and validated the results against a census in Esturro Bairro, Beira, Mozambique. Compared to the census (population: 9,479), the T-Square yielded a better population estimate (9,523) than the Quadrat method (7,681; 95% confidence interval: 6,160-9,201), but was more difficult for field survey teams to implement. Although applicable only to similar sites, several general conclusions can be drawn for emergency planning.
    • Assessing the quality of data aggregated by antiretroviral treatment clinics in Malawi.

      Makombe, S D; Hochgesang, M; Jahn, A; Tweya, H; Hedt, B; Chuka, S; Yu, J K L; Aberle-Grasse, J; Pasulani, O; Bailey, C; Kamoto, K; Schouten, E J; Harries, A D; Clinical HIV Unit, Ministry of Health, Lilongwe, Malawi. (2008-04)
      PROBLEM: As national antiretroviral treatment (ART) programmes scale-up, it is essential that information is complete, timely and accurate for site monitoring and national planning. The accuracy and completeness of reports independently compiled by ART facilities, however, is often not known. APPROACH: This study assessed the quality of quarterly aggregate summary data for April to June 2006 compiled and reported by ART facilities ("site report") as compared to the "gold standard" facility summary data compiled independently by the Ministry of Health supervision team ("supervision report"). Completeness and accuracy of key case registration and outcome variables were compared. Data were considered inaccurate if variables from the site reports were missing or differed by more than 5% from the supervision reports. Additionally, we compared the national summaries obtained from the two data sources. LOCAL SETTING: Monitoring and evaluation of Malawi's national ART programme is based on WHO's recommended tools for ART monitoring. It includes one master card for each ART patient and one patient register at each ART facility. Each quarter, sites complete cumulative cohort analyses and teams from the Ministry of Health conduct supervisory visits to all public sector ART sites to ensure the quality of reported data. RELEVANT CHANGES: Most sites had complete case registration and outcome data; however many sites did not report accurate data for several critical data fields, including reason for starting, outcome and regimen. The national summary using the site reports resulted in a 12% undercount in the national total number of persons on first-line treatment. Several facility-level characteristics were associated with data quality. LESSONS LEARNED: While many sites are able to generate complete data summaries, the accuracy of facility reports is not yet adequate for national monitoring. The Ministry of Health and its partners should continue to identify and support interventions such as supportive supervision to build sites' capacity to maintain and compile quality data to ensure that accurate information is available for site monitoring and national planning.
    • Beyond Open Data: Realising the Health Benefits of Sharing Data

      Pisani, E; Aaby, P; Breugelmans, JG; Carr, D; Groves, T; Helinski, M; Kamuya, D; Kern, S; Littler, K; Marsh, V; Mboup, S; Merson, L; Sankoh, O; Serafini, M; Schneider, M; Schoenenberger, V; Guerin, Philippe J (BMJ Publishing Group, 2016-10-10)
    • Building leadership capacity and future leaders in operational research in low-income countries: why and how?

      Zachariah, R; Reid, T; Srinath, S; Chakaya, J; Legins, K; Karunakara, U; Harries, A D; Medical Department, Médecins Sans Frontières, Operational Centre Brussels, MSF-Luxembourg, Luxembourg. zachariah@internet.lu (2011-11)
      Very limited operational research (OR) emerges from programme settings in low-income countries where the greatest burden of disease lies. The price paid for this void includes a lack of understanding of how health systems are actually functioning, not knowing what works and what does not, and an inability to propose adapted and innovative solutions to programme problems. We use the National Tuberculosis Control Programme as an example to advocate for strong programme-level leadership to steer OR and build viable relationships between programme managers, researchers and policy makers. We highlight the need to create a stimulating environment for conducting OR and identify some of the main practical challenges and enabling factors at programme level. We focus on the important role of an OR focal point within programmes and practical approaches to training that can deliver timely and quantifiable outputs. Finally, we emphasise the need to measure successful OR leadership development at programme level and we propose parameters by which this can be assessed. This paper 1) provides reasons why programmes should take the lead in coordinating and directing OR, 2) identifies the practical challenges and enabling factors for implementing, managing and sustaining OR and 3) proposes parameters for measuring successful leadership capacity development in OR.
    • Challenges in measuring measles case fatality ratios in settings without vital registration

      Cairns, K L; Nandy, R; Grais, R; Global Immunization Division, Centers for Disease Control and Prevention, Atlanta, GA, USA; Health Section, UNICEF, UN Plaza, New York, NY, USA; Epicentre, Paris, France (2010-07-19)
      ABSTRACT: Measles, a highly infectious vaccine-preventable viral disease, is potentially fatal. Historically, measles case-fatality ratios (CFRs) have been reported to vary from 0.1% in the developed world to as high as 30% in emergency settings. Estimates of the global burden of mortality from measles, critical to prioritizing measles vaccination among other health interventions, are highly sensitive to the CFR estimates used in modeling; however, due to the lack of reliable, up-to-date data, considerable debate exists as to what CFR estimates are appropriate to use. To determine current measles CFRs in high-burden settings without vital registration we have conducted six retrospective measles mortality studies in such settings. This paper examines the methodological challenges of this work and our solutions to these challenges, including the integration of lessons from retrospective all-cause mortality studies into CFR studies, approaches to laboratory confirmation of outbreaks, and means of obtaining a representative sample of case-patients. Our experiences are relevant to those conducting retrospective CFR studies for measles or other diseases, and to those interested in all-cause mortality studies.
    • Clinic entrance interviews: a new method to assess needs after a sudden impact disaster

      Von Schreeb, J; Karlsson, N; Rosling, H (Open Medicine, 2007)
    • Clinical Field Research in a Post-conflict Setting

      Boggero, M (London School of Economics and Political Science and John Wiley & Sons Ltd, 2010-10)
    • Conducting operational research within a non governmental organization: the example of Medecins Sans Frontieres

      Zachariah, R; Ford, N; Draguez, B; Yun, O; Reid, T; Médecins Sans Frontières, Medical Department (Brussels Operational Centre- Operational Research), 68 Rue de Gasperich, L-1617, Luxembourg. (Elsevier, 2010-03-02)
      Like many other non governmental organizations (NGOs) that provide assistance to vulnerable populations living in difficult and resource-limited settings, Médecins Sans Frontières (MSF) is confronted with situations for which proven, effective interventions are often lacking and/or where there is need for strong advocacy for improving medical care. As a result, MSF has become an important contributor to health research, and has dedicated resources to guide operational research by establishing its own Ethics Review Board, an innovation fund, an online publications repository and by regularly contributing to major scientific conferences. However, this increased research activity has led to concern that priorities and resources may be diverted away from the essential mandate of care provision for NGOs. In response, this article discusses the potential role operational research can play within medical NGOs such as MSF, and highlights the relevance of operational research, the essential elements of developing it within the organisation and some of the perceived barriers and solutions.
    • Different methodological approaches to the assessment of in vivo efficacy of three artemisinin-based combination antimalarial treatments for the treatment of uncomplicated falciparum malaria in African children.

      Ashley, E A; Pinoges, L; Turyakira, E; Dorsey, G; Checchi, F; Bukirwa, H; van den Broek, I; Zongo, I; Urruta, P P P; van Herp, M; Balkan, S; Taylor, W R J; Olliaro, P; Guthmann, J P; Epicentre, Paris, France. elizabeth.ashley@epicentre.msf.org (2008-08-09)
      BACKGROUND: Use of different methods for assessing the efficacy of artemisinin-based combination antimalarial treatments (ACTs) will result in different estimates being reported, with implications for changes in treatment policy. METHODS: Data from different in vivo studies of ACT treatment of uncomplicated falciparum malaria were combined in a single database. Efficacy at day 28 corrected by PCR genotyping was estimated using four methods. In the first two methods, failure rates were calculated as proportions with either (1a) reinfections excluded from the analysis (standard WHO per-protocol analysis) or (1b) reinfections considered as treatment successes. In the second two methods, failure rates were estimated using the Kaplan-Meier product limit formula using either (2a) WHO (2001) definitions of failure, or (2b) failure defined using parasitological criteria only. RESULTS: Data analysed represented 2926 patients from 17 studies in nine African countries. Three ACTs were studied: artesunate-amodiaquine (AS+AQ, N = 1702), artesunate-sulphadoxine-pyrimethamine (AS+SP, N = 706) and artemether-lumefantrine (AL, N = 518).Using method (1a), the day 28 failure rates ranged from 0% to 39.3% for AS+AQ treatment, from 1.0% to 33.3% for AS+SP treatment and from 0% to 3.3% for AL treatment. The median [range] difference in point estimates between method 1a (reference) and the others were: (i) method 1b = 1.3% [0 to 24.8], (ii) method 2a = 1.1% [0 to 21.5], and (iii) method 2b = 0% [-38 to 19.3].The standard per-protocol method (1a) tended to overestimate the risk of failure when compared to alternative methods using the same endpoint definitions (methods 1b and 2a). It either overestimated or underestimated the risk when endpoints based on parasitological rather than clinical criteria were applied. The standard method was also associated with a 34% reduction in the number of patients evaluated compared to the number of patients enrolled. Only 2% of the sample size was lost when failures were classified on the first day of parasite recurrence and survival analytical methods were used. CONCLUSION: The primary purpose of an in vivo study should be to provide a precise estimate of the risk of antimalarial treatment failure due to drug resistance. Use of survival analysis is the most appropriate way to estimate failure rates with parasitological recurrence classified as treatment failure on the day it occurs.
    • Does research make a difference to public health? Time for scientific journals to cross the Rubicon

      Harries, A D; Zachariah, R; Ramsay, A; Kumar, A M V; Reid, A J; Terry, R F; Reeder, J C (International Union Against TB and Lung Disease, 2014-04-14)
    • Editing and publishing humanitarian medical research for Doctors Without Borders/ Médecins Sans Frontières

      Yun, O; Medecins Sans Frontieres/Doctors Without Borders, New York, USA (Council of Science Editors, 2009-08)
    • Effects of a refugee-assistance programme on host population in Guinea as measured by obstetric interventions.

      Van Damme, W; De Brouwere, V; Boelaert, M; Van Lerberghe, W; Department of Public Health, Institute of Tropical Medicine, Antwerp, Belgium. wvdamme@itg.be (Elsevier, 1998-05-30)
      BACKGROUND: Since 1990, 500000 people have fled from Liberia and Sierra Leone to Guinea, west Africa, where the government allowed them to settle freely, and provided medical assistance. We assessed whether the host population gained better access to hospital care during 1988-96. METHODS: In Guéckédou prefecture, we used data on major obstetric interventions performed in the district hospital between January, 1988, and August, 1996, and estimated the expected number of births to calculate the rate of major obstetric interventions for the host population. We calculated rates for 1988-90, 1991-93, and 1994-96 for three rural areas with different numbers of refugees. FINDINGS: Rates of major obstetric interventions for the host population increased from 0.03% (95% CI 0-0.09) to 1.06% (0.74-1.38) in the area with high numbers of refugees, from 0.34% (0.22-0.45) to 0.92% (0.74-1.11) in the area with medium numbers, and from 0.07% (0-0.17) to 0.27% (0.08-0.46) in the area with low numbers. The rate ratio over time was 4.35 (2.64-7.15), 1.70 (1.40-2.07), and 1.94 (0.97-3.87) for these areas, respectively. The rates of major obstetric interventions increased significantly more in the area with high numbers of refugees than in the other two areas. INTERPRETATION: In areas with high numbers of refugees, the refugee-assistance programme improved the health system and transport infrastructure. The presence of refugees also led to economic changes and a "refugee-induced demand". The non-directive refugee policy in Guinea made such changes possible and may be a cost-effective alternative to camps.
    • Episodic outbreaks bias estimates of age-specific force of infection: a corrected method using measles as an example.

      Ferrari, M J; Djibo, A; Grais, R; Grenfell, B T; Bjørnstad, O N; Center for Infectious Disease Dynamics, Penn State University, PA 16802, USA. mferrari@psu.edu (Cambridge University Press, 2010-01)
      Understanding age-specific differences in infection rates can be important in predicting the magnitude of and mortality in outbreaks and targeting age groups for vaccination programmes. Standard methods to estimate age-specific rates assume that the age-specific force of infection is constant in time. However, this assumption may easily be violated in the face of a highly variable outbreak history, as recently observed for acute immunizing infections like measles, in strongly seasonal settings. Here we investigate the biases that result from ignoring such fluctuations in incidence and present a correction based on the epidemic history. We apply the method to data from a measles outbreak in Niamey, Niger and show that, despite a bimodal age distribution of cases, the estimated age-specific force of infection is unimodal and concentrated in young children (<5 years) consistent with previous analyses of age-specific rates in the region.
    • Evaluation of Three Sampling Methods to Monitor Outcomes of Antiretroviral Treatment Programmes in Low- and Middle-Income Countries.

      Tassie, J-M; Malateste, K; Pujades-Rodríguez, M; Poulet, E; Bennett, D; Harries, A D; Mahy, M; Schechter, M; Souteyrand, Y; Dabis, F; Department of HIV/AIDS, World Health Organisation, Geneva, Switzerland. (2010-11-10)
      BACKGROUND: Retention of patients on antiretroviral therapy (ART) over time is a proxy for quality of care and an outcome indicator to monitor ART programs. Using existing databases (Antiretroviral in Lower Income Countries of the International Databases to Evaluate AIDS and Médecins Sans Frontières), we evaluated three sampling approaches to simplify the generation of outcome indicators. METHODS AND FINDINGS: We used individual patient data from 27 ART sites and included 27,201 ART-naive adults (≥15 years) who initiated ART in 2005. For each site, we generated two outcome indicators at 12 months, retention on ART and proportion of patients lost to follow-up (LFU), first using all patient data and then within a smaller group of patients selected using three sampling methods (random, systematic and consecutive sampling). For each method and each site, 500 samples were generated, and the average result was compared with the unsampled value. The 95% sampling distribution (SD) was expressed as the 2.5(th) and 97.5(th) percentile values from the 500 samples. Overall, retention on ART was 76.5% (range 58.9-88.6) and the proportion of patients LFU, 13.5% (range 0.8-31.9). Estimates of retention from sampling (n = 5696) were 76.5% (SD 75.4-77.7) for random, 76.5% (75.3-77.5) for systematic and 76.0% (74.1-78.2) for the consecutive method. Estimates for the proportion of patients LFU were 13.5% (12.6-14.5), 13.5% (12.6-14.3) and 14.0% (12.5-15.5), respectively. With consecutive sampling, 50% of sites had SD within ±5% of the unsampled site value. CONCLUSIONS: Our results suggest that random, systematic or consecutive sampling methods are feasible for monitoring ART indicators at national level. However, sampling may not produce precise estimates in some sites.
    • Inter-rater and intrarater reliability of the South African Triage Scale in low-resource settings of Haiti and Afghanistan

      Dalwai, M; Tayler-Smith, K; Twomey, M; Nasim, M; Popal, AQ; Haqdost, WH; Gayraud, O; Cheréstal, S; Wallis, L; Valles, P (BMJ Publishing Group, 2018-03-16)
      The South African Triage Scale (SATS) has demonstrated good validity in the EDs of Médecins Sans Frontières (MSF)-supported sites in Afghanistan and Haiti; however, corresponding reliability in these settings has not yet been reported on. This study set out to assess the inter-rater and intrarater reliability of the SATS in four MSF-supported EDs in Afghanistan and Haiti (two trauma-only EDs and two mixed (including both medical and trauma cases) EDs).
    • Is operational research delivering the goods? The journey to success in low-income countries

      Zachariah, R; Ford, N; Maher, D; Bissell, K; Van den Bergh, R; van den Boogaard, W; Reid, T; Castro, K G; Draguez, B; von Schreeb, J; Chakaya, J; Atun, R; Lienhardt, C; Enarson, D A; Harries, A D; Operational Centre Brussels, Medical Department, Médecins Sans Frontières, Luxembourg, Luxembourg; Access to Medicines Unit, Médecins Sans Frontières, Geneva, Switzerland; London School of Hygiene and Tropical Medicine, London, UK; International Union against Tuberculosis and Lung Disease, Centre for Operational Research, Paris, France; Department of Molecular and Cellular Interaction, Vlaams Instituut voor Biotechnologie, Brussels, Belgium; Department of Microbiology, Institute of Tropical Medicine, Antwerp, Belgium; Division of Tuberculosis Elimination, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, Centers for Disease Control and Prevention, Atlanta, GA, USA; Department of Public Health, Division of Global Health (IHCAR), Karolinska Institute, Stockholm, Sweden; Kenya Medical Research Institute, Ministry of Health, Nairobi, Kenya; Global Fund to Fight HIV, Tuberculosis and Malaria, Geneva, Switzerland; Imperial College London, London, UK; Stop TB Partnership, WHO, Geneva, Switzerland (Elsevier, 2012-02-09)
      Operational research in low-income countries has a key role in filling the gap between what we know from research and what we do with that knowledge-the so-called know-do gap, or implementation gap. Planned research that does not tangibly affect policies and practices is ineffective and wasteful, especially in settings where resources are scarce and disease burden is high. Clear parameters are urgently needed to measure and judge the success of operational research. We define operational research and its relation with policy and practice, identify why operational research might fail to affect policy and practice, and offer possible solutions to address these shortcomings. We also propose measures of success for operational research. Adoption and use of these measures could help to ensure that operational research better changes policy and practice and improves health-care delivery and disease programmes.