Malviya PARITOSH

Thesis summary

Background and rationale

Visceral leishmaniasis (VL) or kala-azar is a chronic infectious disease which poses a public health problem in the Indian subcontinent, East Africa and South America. In 2012, the estimated worldwide number of new cases of VL was between 200,000 and 400,000. Bangladesh, India and Nepal have entered into a joint commitment to eliminate VL as a public health problem from this region by the year 2015. The aim is to reduce the incidence rate of VL to less than one new case per 10,000 populations per year at sub-district level in these countries.  

In the Indian subcontinent VL is caused by the parasite Leishmania donovani and transmitted by a sand fly of the species Phlebotomus argentipes in the peridomestic environment. The sand fly breeds in moist soil littered with organic debris and rests inside houses in cracks and crevices of walls. There is no known animal reservoir. Prolonged fever of more than two weeks’ duration with enlarged spleen are the main signs and symptoms of VL, a syndrome that develops progressively into extreme wasting, anaemia and ultimately death if not treated – usually by secondary infections. Cure can be achieved in over 90% of cases with a full course of four weeks of the oral drug Miltefosine, amongst other regimens. The VL elimination strategy therefore builds on early case detection and management as well as vector control.

In the Bihar state of India, VL is endemic in 33 out of 38 districts. The disease affects the rural areas, where the poorest families are most at risk. VL has catastrophic effects on the household economy of families of poor patients. Though the elimination program in this region was launched with much enthusiasm in 2005; several weaknesses in its set-up were identified in program reviews and research reports.  Firstly, several basic facts about disease transmission and risk factors needed better understanding: the role of asymptomatic carriers, the role of PKDL cases, the vector behaviour, but also the behaviour of the epidemic itself. Epidemic peaks in Bihar have been observed approximately at 10-15 years interval since 1977. Is this interval between epidemic peaks a natural phenomenon, or is it linked to the control interventions? What prediction can we make about the future expected number of cases? Secondly, the control strategies themselves seemed flawed to some extent, as the quality and effectiveness of the vector control operation was questioned, but also the case management strategy itself. If early case detection and treatment is so important to curtail the transmission of this anthroponotic pathogen, why was so little attention paid to full coverage of all VL cases at community level? Active case detection –though advocated- was hardly implemented. Thirdly, a major weakness of the VL elimination initiative was the epidemiological surveillance and health information system. It’s being curbed to the cases seen in the public sector mainly led to a major underreporting of VL cases in the official records, and, its focus on counting VL cases rather than on monitoring their outcomes led to very doubtful official records on relapse and death rates of VL patients. Hence in this thesis we set out to address a number of critical knowledge gaps that were slowing down the VL elimination initiative in Bihar, India, related to the general epidemiology of VL in Bihar, the control strategy and the health information system.

Methods:

We conducted our research in the Muzaffarpur district of Bihar state which is one of the worst VL affected areas in the world. We prioritized a number of specific research questions, in function of their relevance for the elimination effort, in three different domains: 1. Epidemiology of VL, 2. VL control strategies and 3. Health information system. In the first section, we retrospectively analysed data on VL incidence in Muzaffarpur district, Bihar and studied the association with meteorological parameters. A second study investigated risk factors for sand fly abundance in houses.  In the second section, we focused on one of the two major VL control strategies: case detection & management. We studied whether and how grassroots community health workers were able and willing to contribute to active case detection for VL. In section 3, we present the work we did on the health information system. In the context of a research program on drug resistance, we tried to develop new tools for the monitoring of clinical outcomes of VL patients. To understand how well the routine reporting system performs, we conducted a survey among a random sample of patients treated by the primary health centres. We assessed the cost and operational feasibility of such surveys in respect to the routine surveillance system.  We then developed a recording and reporting tool in the form of a register adapted from tuberculosis control program and field tested it in three primary health centres in the Muzaffarpur district. The last paper in this section describes how we set up a demographic and health surveillance system in the VL endemic area, as a platform for further demographic and health research in the study area. This platform cannot be operated on a routine basis by the health authorities, but will allow researchers to address some of the very specific remaining research questions on VL transmission, as well as other diseases, in the future.

Results:

Epidemiology

Between 1990 and 2008 over 70,000 new VL cases occurred in Muzaffarpur district. These cases were almost all reported by the public health facilities. The first major peak occurred around 1992 and the second one around 2007. We observed that the monthly distribution of VL cases followed a similar pattern over the study period, with the months of March-April always showing the highest number of VL cases. During 1990-1998 the eastern part of Muzaffarpur district used to have the highest incidence rates of VL which shifted later to the western part during the period 1999-2008. We found a significant association between rainfall and relative humidity with VL incidence.

In a second study we studied indoor sand fly density to look for modifiable risk factors. We observed major variability in the indoor sand fly density between years and seasons. We found that sand flies were particularly abundant in thatched houses. Lower indoor sand fly density was associated with better housing conditions and wealth status of the household, independently of each other. However these factors explained only a small proportion of the overall variability of indoor sand fly density.

Control strategies: active case detection and the role of village health workers :

The network of village level health worker, ASHAs and ANMs, is well organised in India. These health workers contribute a lot to other health programs but not to VL control. They use mobile phones to contact patients and report back to the Primary Health Centres (PHCs). We showed that these health workers know the presenting symptoms of VL and its diagnosis rather well, but they were less aware of the recommended first-line treatment. They were ready to get involved in the VL control program and become involved in case detection and follow up of patients, and clearly are an untapped resource in the VL elimination effort.

Health information system

Our research work was geared towards the monitoring of clinical outcomes of VL. Our random survey showed that recording of treatment outcomes was rather flawed at the PHCs. Though a survey produces accurate information on treatment outcome, it was costly, labour intensive and as such not a method that the public health system can routinely apply.

We therefore developed a register based tool for documenting clinical outcomes of VL treatment in the PHCs, inspired by the Retrospective Quarterly Cohort Monitoring method used in tuberculosis control programs.  The tool was able to record patients’ treatment data and to generate reports on early and late treatment outcomes. It proved to be a useful tool and could easily be implemented by the PHCs. However, a majority of VL patients did not show up at the PHC for their control appointment six months after treatment, making it difficult to assess the late treatment outcomes. Therefore the involvement of ASHAs and ANMs to follow-up patients in their villages, and report their late outcomes, presents the most feasible solution to this problem.

We further describe in a third paper in this section how the Muzaffarpur-TMRC Health and Demographic Surveillance System (HDSS), was developed in the Muzaffarpur district. It is the only HDSS in the world conducting research on VL. Currently the HDSS serves a population of over 105,000 in 66 villages. The HDSS collects data on vital events including pregnancies, births, deaths, migration, marriages as well as other socio-economic indicators at regular intervals. Incident VL cases are actively tracked. The HDSS team is experienced in conducting both qualitative and quantitative studies, sample collection and performing rapid diagnostic tests in the field. In each village, volunteers connect the HDSS team with the community members. The Muzaffarpur‐TMRC HDSS provides opportunities for studies on VL and other Neglected Tropical Diseases (NTDs) and their interaction with demographic events such as migration. 

Conclusion:

VL is still a major public health problem in Bihar state, India. Our data showed geographic spreading over time into new areas, and a clear link with meteorological parameters.  The housing and environmental conditions of the rural villages in Muzaffarpur district provide a natural habitat for the vector P. argentipes sand flies. Housing improvement is recommended. Our research on control strategies and information system led to a register-based tool for monitoring clinical outcomes of VL treatment that was user-friendly, but the information on late treatment outcomes cannot be obtained without the involvement of community health workers. These ASHAs and ANMs are well organized and are ready to get involved in the VL control initiatives. Operational research of this kind can help the VL elimination program address some of its current weaknesses, but for addressing more fundamental research questions on VL transmission and control, a research platform as the Muzaffarpur-TMRC-HDSS provides a solid backbone for future research.