THE NATIONAL MEDICAL JOURNAL OF INDIA VOL. 14, NO. 3,2001 139 Risk factors of malaria in the fringes of an evergreen monsoon forest of Arunachal Pradesh P. K. MOHAPATRA, K. NARAIN, ANIL PRAKASH, D. R. BHATTACHARYYA, J.MAHANTA ABSTRACT Bdckground. The forested hilly and foothill regions of north-east India are highly endemic for malaria and have a distinct epidemiological pattern. Nearly half the reported cases of malaria are from these areas. A knowledge of the risk factors in this eco-geographic entity may be helpful in formulating a specific control strategy. Hence, we conducted a community-based epidemiological study in a hilly, forested terrain of Arunachal Pradesh and examined different socio-demographic factors to identify those predisposing to the occurrence of malaria, espe- cially Plasmodium falciparum infection, in such areas. Methods. Four epidemiological surveys were carried out during 1997 in 7 villages located at the fringes of a forest (total population: 1177) under Nompong Primary Health Centre of Changlang district, Arunachal Pradesh. Blood slides were col- lected randomly from the inhabitants, irrespective of their fever status, ensuring at least 50% coverage. One hundred and thirty- four microscopically confirmed Plasmodium falciparum cases were identified and 536 controls were randomly selected from the list of uninfected inhabitants. Relevant socio-demographic information was obtained from both cases and controls. The data were analysed by simple and multiple logistic regression using the unconditional maximum likelihood method. Results. Factors which were found to be strongly associated with Plasmodium falciparum malaria on univariate analysis were age, ethnicity, village of residence and accessibility to the nearest health care facility. However, in multiple regression analysis, after controlling for the effects of confounding variables, the only risk factor identified was accessibility to the nearest health care facility (adjusted odds rado: 4.5; 95% CI: 1.8-11.3; p< 0.000 1 for those at a distance of 2-8 km and adjusted odds ratio: 11.1; 95% CI: 4.1-30.0; p<O.OOOl for those> 8 km away). Conclusion. Distance, particularly non-motorable dis- tance, from the place of residence to the nearest health care facility was a major risk factor for malaria in this hilly forested terrain. This indicates the need for special efforts to detect cases early and institute treatment promptly in such areas of the north-eastern region, so as to reduce the morbidity and mortality of malaria. Natl Med J India 2001; 14: 139-42 Regional Medical Research Centre, Northeastem Region, Indian Council of Medical Research, P.O. Box 105, Dibrugarh 786001, Assam, India P.K.MOHAPATRA, K.NARAIN, ANILPRAKASH, D.R.BHATTACHARYYA, J.MAHANTA Correspondence to J. MAHANT A; rmrcdib@ rediffmail. com © The National Medical Journal of India 2001 INTRODUCTION The north-eastern region of India comprises 7 states which record 0.2-0.25 million malaria cases annually (60%-70% Plasmodium Jalciparum infection) accounting for 8%-12% of all the reported cases in India.' Most of the area (except the high mountainous regions) is endemic for malaria. The disease is unevenly distrib- uted over the region, encompassing distinct epidemiological patterns with varying risk factors in different eco-geographic entities. Though malaria as an anthropo-ecosystem essentially involves interaction between the mosquito vector, the malarial parasite, the human host sub-system and the environment,' within a given ecological situation there may be differences in the risk of acquiring malaria between villages, the community or in individu- als. Hence, knowledge of the various risk factors governing the dynamics of malaria transmission is essential for devising an appropriate and sustainable control strategy in an area. Nearly 45% of the land area in this region is covered by tropical evergreen forest.' The forests and their fringe areas in north-east India are highly endemic for malaria and contribute nearly half the total malaria cases.' In spite of the importance of forest areas in the epidemiology of malaria in this region, little is known about the risk factors for the occurrence of malaria in these areas. We attempted to identify the factors responsible for the higher preva- lence of malaria [particularly PlasmodiumJalciparum (Pf) infec- tion] in these areas. SUBJECTS AND METHODS Study area The study was carried out in the Jairampur sector of the Nompong Primary Health Centre of Changlang district, Arunachal Pradesh. Jairampur is a known malaria-endemic area with a high prepon- derance of Pf, widespread chloroquine resistance' and an annual parasite incidence (API) of 15-50 per thousand population." The study area is at an altitude of 305-455 m above sea level and comprises undulating hillocks, foothills and large forests. The occurrence of malaria is perennial, with April-October being the main transmission period. Jairampur is a central place with a small market, government offices and only one health facility (a 1O-beddedgovernment dispensary manned by two qualified medi- cal officers). In addition, a few pharmacists act as indigenous practitioners catering to the health needs of the people. The villages are sparsely populated and many of them are situated on hilly slopes, poorly connected to Jairampur and accessible only by foot. This study was conducted during 1997 in 7 villages (popu- lation: 1177) selected randomly from a list of villages of the region. The inhabitants belonged largely to the Tangsa aboriginal tribe along with some Nepalese and other non-tribal migrants. A majority of the tribal people in the study villages were engaged in