Factors associated with treatment seeking for malaria in Madhya Pradesh, India

Tropical Medicine and International Health doi:10.1111/tmi.12973 volume 22 no 11 pp 1377 1384 november 2017 Factors associated with treatment seeking for malaria in Madhya Pradesh, India Mrigendra P. Singh 1, Kalyan B. Saha 2, Sunil K. Chand 1 and Anup Anvikar 3 1 ICMR-National Institute of Malaria Research Field Unit Jabalpur, Jabalpur, India 2 ICMR-National Institute for Research in Tribal Health, Jabalpur, India 3 ICMR-National Institute of Malaria Research, New Delhi, India Abstract objectives To determine household factors associated with treatment seeking for malaria. methods The study was carried out in four districts of Madhya Pradesh with different malaria endemicity. A total of 1470 households were interviewed in which at least one member suffered from microscopically confirmed malaria in the 3 months preceding the survey. Socio-demographic, economic, cultural characteristics, their health beliefs, knowledge and practices regarding malaria and choice of treatment seeking were explored. results A total of 764 households were from high-endemic and 706 from low-endemic areas. More than half of household heads were illiterate; most are farmers. Approximately 46% sought treatment for malaria from unqualified informal providers; 19% from qualified private health practitioners and 35% from government health providers. Analysis revealed that household s area of residence, education, occupation, ethnicity, use of preventive measures, economic status, knowledge and practices, distance and delayed treatment seeking was strongly associated with the type of healthcare providers selected. conclusions Demand for formal health services among the poor, illiterate, tribal population living in remote areas is low. Accessible and affordable health services and a sensitisation programme to increase the demand for formal providers are needed. keywords malaria, treatment seeking, informal health providers, tribal, prompt diagnosis and treatment Introduction Nearly, 214 million people suffer from malaria every year, and 3.2 billion people are living at risk of infection [1]. Globally, 0.44 million deaths due to malaria are reported annually, of which 69.86% are in under-5 children [1]. Malaria remains a major killer of children, particularly in sub-saharan Africa, taking the life of a child every 2 min [1]. In India most malaria cases occur in Odisha, Chhattisgarh, Madhya Pradesh (MP) and states of North Eastern region. MP contributed 9% of malaria cases in India in 2015 [2]. Remote forested and tribal areas are most affected [3, 4]. The government of India has been emphasising tribal development for more than 65 years and designated them as Scheduled Tribes, but because of their heterogeneous nature the desired level of development has not been achieved so far [5]. Traditional home remedies, faith healers and self-medication as a first line of treatment for febrile illness are very common in remote rural communities and delay seeking treatment at appropriate health facilities with diagnosis and treatment [6, 7]. In remote rural areas of MP, besides programmerelated health facilities, there are accredited social health activists (ASHAs), auxiliary nurse midwives (ANMs) and multipurpose workers (MPWs). Various informal health providers are also available, such as traditional faith healers, local herbal practitioners and quacks who acquire informal rudimentary knowledge of various practices ranging from Allopathy to AYUSH (Ayurveda); all endanger the health of patients [8]. Qualified trained private health practitioners also exist, but they are rare in rural areas and are usually expensive compared to the services provided by the government health posts. Poor public health infrastructure, irregular availability of trained medical and paramedical staff in public health facilities are major concerns in rural areas. By contrast, traditional healers and private practitioners in remote tribal areas are easily accessible. Only 4% of febrile illness visits to private facilities led to a laboratory test for 2017 John Wiley & Sons Ltd 1377

diagnosis either by microscopy or by rapid diagnostic tests (RDTs) [9]; this leads to misuse of antimalarials and development of resistance. Individual use of health services is shaped by socio-cultural characteristics, logistical aspects of obtaining health care and perceived need of health services [10]. The choice of treatment provider is also driven by faith and convenience. Due to frequent shortage of diagnostics and antimalarials, the community does not have faith in the village-level health providers [11]. We sought to understand the socio-economic, demographic, knowledge and preventive practices associated with treatment seeking among households with microscopically confirmed malaria cases. Materials and methods Setting and participants This study was carried out in four districts of MP with different malaria endemicity and ethnicity (Figure 1). Jabalpur and Chhindwara are of low-malaria endemicity with <2 annual parasite incidence [API (number of malaria cases per 1000 population in a year)] during 2013 2015. However, Balaghat and Jhabua are of moderate-to-high malaria endemicity with 2 5 API during 2013 2015. Jabalpur, Balaghat and Chhindwara are located in south eastern MP, whereas Jhabua is situated in the western part. Gond in Jabalpur, Gond and Bharia in Chhindwara, Baiga in Balaghat and Bhil in Jhabua are the main ethnic tribal groups in these areas. A household was defined as people sharing the common cooking pot [12]. A sample of 1470 household, in which at least one member suffered from microscopically confirmed malaria in the three months preceding the survey were interviewed using pre-tested structured schedule. A list of these confirmed malaria cases was obtained from government health facilities such as community health centres (CHCs) and ASHAs. Socio-demographic, economic, cultural characteristics of individuals, their health beliefs, knowledge and practices regarding malaria and (a) N MADHYA PRADESH (b) Jhabua Jabalpur Chhindwara Balaghat Index High transmission Low transmission Figure 1 (a) Map of India, (b) Madhya Pradesh showing study districts (source: National Informatics Centre, Madhya Pradesh State Centre. Available at http://mp.nic.in/district.asp). 1378 2017 John Wiley & Sons Ltd

choice of treatment seeking were explored. The survey was undertaken during August 2015 to January 2016. Data entry and analysis Data were double keyed entered and validated in CSPro 6.3 (US Census Bureau). All inconsistencies and illogical entries were resolved. Data were exported to STATA 12.1 for analysis. Numerically coded categorical variables were cross-tabulated, and chi-square test or Fisher s exact test were applied as required. Multinomial logistic regression model was used to identify the effect of associated factors in utilisation of health services. Ethical considerations The aim of the study was explained, and verbal consent from adult individual or guardian of the child was obtained before conducting interview. The study was approved by Ethics Committee of ICMR-National Institute of Malaria Research (Indian Council of Medical Research) New Delhi (ECR/65/Inst/DL/2013). Results Basic characteristics of the studied households are presented in Table 1. Of 1470 households, 27.5% were from the Jhabua; 24.8% from Chhindwara, 24.5% from Balaghat; and 23.3% from Jabalpur. The households were categorised into various social groups as classified by the government of India. Scheduled Tribes are at the bottom of the social hierarchy followed by Scheduled Castes, Other Backward Castes and those in the General Category. Approximately 78.4% of households belonged to scheduled tribes. Average household size was (mean SD) 6.0 2.3 persons which are higher than the corresponding figure of the MP state [13]. More than 90% of the adult individuals interviewed in the household were male, and most of them were 25 55 years old with a mean age (mean SD) of 45.0 12.6 years. Approximately 52% were illiterate with no formal schooling, and agriculture was their major occupation. Approximately 46% initially chose informal unqualified providers or self-medication for malaria treatment; 19% consulted qualified private health practitioners and 35% saw government health providers (Table 1). Mean waiting time from onset of fever to initial treatment from any healthcare provider was (mean SD) 1.7 0.9 days. Approximately 46% (681/ 1470) patients were not satisfied with the initial treatment providers and subsequently visited other sources for treatment. Univariate and multivariate multinomial logistic regression analyses were carried out to estimate the odds ratios (OR) with respect to independent variables/ reference category for comparison and identification of actual associates (Table 2). Univariate multinomial logistic regression analysis revealed that compared to low-endemic districts such as Jabalpur and Chhindwara, household s members in high-endemic districts like Balaghat and Jhabua preferred private (OR: 2.3; P < 0.0001) and government (OR: 2.7; P < 0.0001) health facilities for seeking care with reference to informal health providers. Scheduled tribes were less likely to visit private (OR: 0.3; P < 0.001) or government (OR: 0.8; P > 0.05) health facilities compared than non-tribals. Educated participants, particularly those with secondary or higher level of education preferred private or government health facilities (OR: 3.3; P < 0.0001 and OR: 3.2; P < 0.0001) vs. illiterate household heads. Household heads whose main occupation was a small business (OR: 5.4; P < 0.001), a salaried job (OR: 6.3; P < 0.0001) or labour (OR: 1.8; P < 0.0001) preferred private providers, in contrast to farmers. Government providers (OR: 0.7; P < 0.05) were less preferred by labourers. Higher income households (INR >7000/month) were more likely to visit formal healthcare providers than informal ones. People who used any preventive measure against malaria such as mosquito repellent coils, cakes or fumes were regular visitors at private (OR: 6.8; P < 0.0001) and government (OR: 1.7; P < 0.001) health posts. Households using bed nets mostly visited private health providers (OR: 2.2; P < 0.0001). Household heads educated to middle and higher level of schooling were using more mosquito repellents [26.6% (98 of 369) vs. 13.9% (153 of 1101)] and bed nets [50.4% (186 of 369) vs. 27.5% (303 of 1101)] than illiterate and less educated people (OR: 2.2 and 2.7; P < 0.001 for repellents and bed nets, respectively). Similarly, household with a higher monthly income (INR 5000 or above) used significantly more mosquito repellents [21.4% (204 of 953) vs. 9.1% (47 of 517)] and bed nets [35.5% (338 of 953) vs. 29.2% (151 of 517)] than households with a lower monthly income (OR: 2.7 and 1.3; P < 0.001 for repellents and bed nets, respectively). More middle and higher educated households with an income of INR 5000 or more preferred treatment from formal health providers compared to illiterate or lesser educated and poor households (monthly income less than INR 5000) (OR: 2.1; P < 0.001) and (OR: 1.3; P < 0.05), respectively. Thus households which were economically better off and had more schooling accepted preventive measures against malaria and were capable of utilising the formal health services. 2017 John Wiley & Sons Ltd 1379

Table 1 Basic demographic characteristics of studied household High endemic districts Low endemic districts Variables Jhabua (n = 404) Balaghat (n = 360) Jabalpur (n = 342) Chhindwara (n = 364) Total (n = 1470) Caste (Head of the HH) n (%) n (%) n (%) n (%) n (%) Schedule Tribe 387 (95.8) 205 (56.9) 261 (76.3) 299 (82.1) 1152 (78.4) Schedule Caste 7 (1.7) 56 (15.6) 38 (11.1) 13 (3.6) 114 (7.8) Other Backward Caste 6 (1.5) 94 (26.1) 41 (12.0) 48 (13.2) 189 (12.9) General 4 (1.0) 5 (1.4) 2 (0.6) 4 (1.1) 15 (1.0) HH size (Mean SD) 7.1 2.4 5.4 2.0 5.6 2.1 5.8 2.2 6.0 2.3 Sex (Head of the HH) Male 354 (87.6) 332 (92.2) 323 (94.4) 350 (96.1) 1359 (92.4) Female 50 (12. 4) 28 (7.8) 19 (5.6) 14 (3.8) 111 (7.5) Age groups (Head of the HH) 16 25 years 19 (4.7) 9 (2.5) 17 (5.0) 16 (4.4) 61 (4.1) 26 35 years 100 (24.7) 66 (18.3) 76 (22.2) 95 (26.1) 337 (22.9) 36 45 years 107 (26.5) 111 (30.4) 120 (35.1) 107 (29.4) 445 (30.3) 46 55 years 80 (19.8) 107 (29.7) 80 (23.4) 70 (19.2) 337 (22.9) 56 65 years 70 (17.3) 50 (13.9) 42 (12.3) 54 (14.8) 216 (14.7) Above 65 years 28 (6.9) 17 (4.7) 7 (2.0) 22 (6.0) 74 (5.0) Age of the head of the HH (Mean SD) 45.3 13.5 46.3 11.8 43.5 11.3 44.6 13.2 45.0 12.6 Educational status of the head of the HH Illiterate 300 (74.3) 144 (40.0) 145 (42.4) 172 (47.2) 761 (51.8) Primary 52 (12.9) 103 (28.6) 84 (24.6) 101 (27.7) 340 (23.1) Middle 23 (5.7) 74 (20.6) 56 (16.4) 51 (14.0) 204 (13.9) Secondary and above 29 (7.2) 39 (10.8) 57 (16.7) 40 (11.0) 165 (11.2) Main occupation of the head of the HH Farmer 339 (83.9) 228 (63.3) 187 (54.7) 280 (76.9) 1034 (70.3) Business 2 (0.5) 16 (4.4) 1 (0.3) 4 (1.1) 23 (1.6) Salaried job 9 (2.2) 14 (3.9) 3 (0.9) 14 (3.8) 40 (2.7) Labour 54 (13.4) 102 (28.3) 151 (44.1) 66 (18.1) 373 (25.4) Monthly HH income (Mean 7196.0 6449.5 8652.5 7394.6 6282.0 4183.1 4642.9 3832.5 6707.9 5878.5 SD) Initial choice for health care Informal providers 143 (35.4) 125 (34.7) 209 (61.1) 197 (54.1) 674 (45.8) Private providers 27 (6.9) 142 (39.4) 50 (14.6) 62 (17.0) 281 (19.1) Government providers 234 (57.9) 93 (25.8) 83 (24.3) 105 (28.8) 515 (35.0) HH, household. Adults who knew that fever is one of the main symptoms of malaria (OR: 2.6; P < 0.0001), that mosquito bites cause malaria (OR: 1.6; P < 0.001) and that mosquitoes breed in water (OR: 1.7; P < 0.0001) preferred to visit private providers compared to those who were not aware of symptoms, source of infection and breeding place for mosquitoes. Those who were aware of malaria diagnosis through blood examination preferably visited government health facilities (OR: 1.4; P < 0.05). Heads of households who sought treatment within 24 h of onset of fever were more likely to visit private (OR: 1.9; P < 0.0001) and government (OR: 1.6; P < 0.0001) health posts. Long distance to health facilities was detrimental to utilisation of government health facilities (OR: 0.6; P < 0.0001). Multivariate regression analysis model revealed that social group (Scheduled Tribe), secondary and above level of education and knowledge about place of diagnostic facilities are potential significant factors associated with choice of healthcare providers controlling the effect of other independent variables like occupation, income, preventions, knowledge about malaria-related symptoms, cause of infection, place of mosquito breeding, diagnostic tools, early treatment seeking and distance from health facilities. However, literacy at any level of education (primary, middle or secondary and above) and knowledge of 1380 2017 John Wiley & Sons Ltd

Table 2 Factors associated with treatment seeking behaviours application of multinomial logistic regression model Multinomial logistic regression (Ref. informal provider) OR (95% CI) (Univariate) OR (95% CI) (Multivariate) Variable n Pvt. provider Govt. provider Pvt. provider Govt. provider District Jabalpur 342 1 (ref) 1 (ref) 1 (ref) 1 (ref) Chhindwara 364 1.3 (0.9 2.0) 1.3 (0.9 1.9) 1.5 (0.8 2.6) 1.9 (1.2 3.0)** Balaghat 360 4.7 (3.2 7.0)*** 1.9 (1.3 2.7)** 0 0 Jhabua 404 0.8 (0.5 1.3) 4.1 (3.0 5.7)*** 0 0 Endemicity Low 706 1 (ref) 1 (ref) Excluded due to multi collinearity High 764 2.3 (1.7 3.0)*** 2.7 (2.1 3.4)*** Ethnicity Schedule Tribe 1151 0.3 (0.2 0.5)*** 0.8 (0.6 1.1) 0.5 (0.3 0.7)*** 0.7 (0.5 0.9)** Non Schedule Tribe 319 1 (ref) 1 (ref) 1 (ref) 1 (ref) Head of the HH education Illiterate 761 1 (ref) 1 (ref) 1 (ref) 1 (ref) Primary 340 1.3 (0.9 1.9) 0.9 (0.7 1.2) 1.1 (0.7 1.7) 1.4 (1.0 2.0)* Middle 204 2.1 (1.4 3.1)*** 1.4 (1.0 2.0)* 1.3 (0.8 2.1) 2.1 (1.4 3.2)*** Secondary and above 165 3.3 (2.0 5.3)*** 3.2 (2.1 4.8)*** 2.7 (1.5 4.9)** 5.6 (3.4 9.3)*** Head of the HH occupation Agriculture 1034 1 (ref) 1 (ref) 1 (ref) 1 (ref) Small business 23 5.4 (2.1 13.9)** 0.7 (0.2 2.4) 1.3 (0.4 4.4) 0.4 (0.1 1.6) Salaried job 40 6.3 (2.6 15.0)*** 2.5 (1.0 5.9)* 3.5 (1.2 10.1)* 1.4 (0.5 3.8) Labour 373 1.8 (1.3 2.4)*** 0.7 (0.6 1.0) 1.4 (1.0 2.1)* 0.8 (0.60 1.2) Monthly HH income (INR) Less than 5000 517 1 (ref) 1 (ref) 1 (ref) 1 (ref) 5000 7000 476 1.1 (0.8 1.6) 0.9 (0.7 1.2) 0.8 (0.5 1.3) 1.1 (0.8 1.5) Above 7000 477 2.6 (1.8 3.6)*** 1.3 (1.0 1.8)* 1.4 (0.9 2.2) 1.3 (0.9 1.9) HH using any preventive measures No 129 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 1341 1.9 (1.1 3.1)* 2.3 (1.4 3.5)*** Omitted Omitted HH using mosquito repellent No 1090 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 251 6.8 (4.8 9.8)*** 1.7 (1.2 2.4)** 2.0 (1.2 3.5)** 1.5 (0.9 2.4) HH using bed net No 852 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 489 2.2 (1.6 3.0)*** 1.2 (0.9 1.5) 1.3 (0.9 1.8) 0.9 (0.7 1.2) Know about symptoms of malaria No 1078 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 392 2.6 (1.9 3.5)*** 1.2 (1.0 1.6)* 1.5 (1.0 2.2)* 1.2 (0.8 1.6) Know about cause of infection 2017 John Wiley & Sons Ltd 1381

Table 2 (Continued) Multinomial logistic regression (Ref. informal provider) OR (95% CI) (Univariate) OR (95% CI) (Multivariate) Variable n Pvt. provider Govt. provider Pvt. provider Govt. provider No 304 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 1166 1.6 (1.1 2.4)** 1.1 (0.8 1.5) 1.2 (0.7 2.0) 0.9 (0.6 1.3) Know about mosquito breeding place No 685 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 785 1.7 (1.3 2.3)*** 1.0 (0.8 1.3) 0.7 (0.5 1.0) 0.8 (0.6 1.1) Know about malaria diagnostic No 402 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 1068 1.2 (0.9 1.6) 1.4 (1.0 1.8)* 1.0 (0.7 1.5) 1.6 (1.2 2.2)** Know about place of malaria diagnosis No 756 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 714 1.9 (1.4 2.5)*** 1.6 (1.3 2.0)*** 1.9 (1.3 2.7)*** 2.0 (1.5 2.7)*** Took treatment within 24 h No 755 1 (ref) 1 (ref) 1 (ref) 1 (ref) Yes 715 1.9 (1.4 2.5)*** 1.6 (1.3 2.0)*** 1.0 (0.7 1.5) 0.9 (0.7 1.3) Distance from health facility Less than 5 km 181 1 (ref) 1 (ref) 1 (ref) 1 (ref) 5 10 km 290 1.1 (0.9 1.7) 0.8 (0.3 1.1) 1.2 (0.7 1.3) 1.2 (0.8 1.5) 11 15 km 448 1.6 (1.1 2.3)* 0.6 (0.4 0.9)*** 1.5 (0.9 2.4) 0.9 (0.6 1.3) Above 15 km 551 4.0 (2.6 5.8)*** 0.6 (0.2 0.8)*** 2.1 (1.7 4.3)*** 0.8 (0.6 1.3) OR, odds ratio; HH, household; ref, reference category. *P < 0.05; **P < 0.01; ***P < 0.001. 1382 2017 John Wiley & Sons Ltd

diagnostic tools inclined towards preference of government providers. Long distances to health facilities increased the preference for private providers who are easily accessible. Discussion The study reveals the interesting fact that contrary to traditional understanding, households in highly endemic areas prefer to visit either private or government health posts. This may be an effect of extensive programmatic interventions. Tribal groups are less likely to use formal health providers than other social groups. A qualitative investigation by Sabin et al. 2010 [6] of pregnant women in two tribal districts of Jharkhand in India also revealed that traditional remedies were very common among the community for treatment of malaria. Further, the higher the education the better is the knowledge on diagnosis and preventive measures against malaria, and the higher the preference for private and government health posts. Delayed treatment seeking, self-medication, preference given to traditional healers are common among uneducated and poor people. This is not only critical to the patient s health but also contributing to the disease burden in the community [14]. Family income also influences health care-seeking behaviours [7]. Distance to health facilities emerged as detrimental to utilisation of government health posts as the private health posts are easily available at accessible reach, though expensive. Further, a study by Das and Ravindran (2010) [15] in Odisha, India revealed that visiting a health provider more than 5 km away was more likely to result in delayed or inappropriate treatment. Dev et al. 2006 [16], in their study reported that in rural and remote areas, no or lesser availability of public transport coupled with economic poverty is the major constraints for delayed treatment seeking. Besides convenience and accessibility to health post, the faith of the community also plays an important role in selection of the health providers. Government health workers often have insufficient or no diagnostic kits and antimalarial drugs to cope up with the crisis and thus promote loss in faith on the health workers [11, 17 20]. Thus in all probability, informal sources of treatment are not preferred for malaria. Thus, diagnosis and treatment for malaria should be restricted to existing government healthcare providers such as village-level community health workers like ASHAs, malaria-specific peripheral workers, ANMs by providing further training, motivation and timely incentives. It is also essential to procure and distribute antimalarials and diagnostic test kits at grass root level in a timely fashion. There is also a need for mass sensitisation for social mobilisation [21]. Our study was conducted during the peak malaria transmission season so that the extent of various sources of health utilisation was captured. It is not based on reported febrile illness, but on microscopically confirmed malaria cases, so their behaviour related to health utilisation was covered and is expected to strengthen the programme s prompt diagnosis and treatment arm. However, seasonal variation of malaria and health service utilisation was not captured, nor was the views of health providers taken into account. Designing an appropriate strategy for local behaviour change will not only generate awareness on malaria, but also produce demand for health services. The programmatic approach of indoor residual spraying, mosquito breeding source reduction and judicious use of integrated community case management of malaria must be strengthened. Similarly, the National Vector Borne Disease Control Programme (NVBDCP) of India focuses on prompt diagnosis by use of bivalent RDTs and immediate treatment particularly in remote rural areas. However, frontline health workers (ASHAs) are rarely equipped in terms of knowledge, RDTs and antimalarial drugs, and thus, the commitment of the programme to controlling malaria is defeated. A concerted effort on the part of service providers in the village and people s timely acceptance of the services and compliance will improve the malaria condition in the study area. Acknowledgements The authors thank Dr. Mantosh S. Malhotra, Consultant, ICMR-National Institute of Malaria Research (Indian Council of Medical Research), New Delhi, for critical reviewing the interview schedule. The manuscript was approved by the ICMR-National Institute of Malaria Research Publication Screening Committee (01/07/2017). The study was supported by intramural funds of ICMR- National Institute of Malaria Research (Indian Council of Medical Research) New Delhi. References 1. World Health Organization. World Malaria Report 2015. World Health Organization. Geneva. Switzerland. 2. NVBDCP. Malaria situation in India. Available from: http:// nvbdcp.gov.in/doc/malaria-situation-june17.pdf. [15 July 2017]. 3. Singh N, Chand SK, Bharti PK et al. Dynamics of forest malaria transmission in Balaghat district, Madhya Pradesh, India. PLoS ONE 2013: 8: e73730. 2017 John Wiley & Sons Ltd 1383

4. Sharma RK, Thakor HG, Saha KB et al. Malaria situation in India with special reference to tribal areas. Indian J Med Res 2015: 141: 537 545. 5. Saha KB, Saha UC, Sharma RK et al. Indigenous and tribal peoples health. Lancet 2016: 388: 2867. 6. Sabin LL, Rizal A, Brooks MI et al. Attitudes, knowledge, and practices regarding malaria prevention and treatment among pregnant women in Eastern India. Am J Trop Med Hyg 2010: 82: 1010 1016. 7. Yadav SP. A study of treatment seeking behaviour for malaria and its management in febrile children in rural part of desert, Rajasthan, India. J Vector Borne Dis 2010: 47: 235 242. 8. Singh N, Singh MP, Saxena A, Sharma VP, Kalra NL. Knowledge, attitude, beliefs and practices (KABP) study related to malaria and intervention strategies in ethnic tribal s of Mandla (Madhya Pradesh). Curr Sci 1998: 75: 1386 1390. 9. Kumar S. Much health care in rural India comes from unqualified practitioners. BMJ 2004: 328: 975. 10. Andersen R, Newman JF. Societal and individual determinants of medical care utilization in the United States. Milbank Mem Fund Q Health Soc 1973: 51: 95 124. 11. Das A, Das Gupta RK, Friedman J, Pradhan MM, Mohapatra CC, Sandhibigraha D. Community perceptions on malaria and care-seeking practices in endemic Indian settings: policy implications for the malaria control programme. Malar J 2013: 12: 39. 12. Hussain A, Ali KSM, Kvale G. Determinants of mortality among children in the urban slums of Dhaka city, Bangladesh. Trop Med Int Health 1999: 4: 758 764. 13. Census. HH-01 Normal Households by Household Size. Office of the Registrar General, Census Commissioner, India. Ministry of Home Affairs, Government of India. Available from: http://www.censusindia.gov.in/2011census/ hh-series/hh01.html [6 Feb 2017]. 14. Chaturvedi HK, Mahanta J, Pandey A. Treatment-seeking for febrile illness in north-east India: an epidemiological study in the malaria endemic zone. Malar J 2009: 8: 301. 15. Das A, Ravindran TS. Factors affecting treatment-seeking for febrile illness in a malaria endemic block in Boudh district, Orissa, India: policy implications for malaria control. Malar J 2010: 9: 377. 16. Dev V, Phookan S, Sharma VP, Dash AP, Anand SP. Malaria parasite burden and treatment seeking behavior in ethnic communities of Assam, Northeastern India. J Infect 2006: 52: 131 139. 17. Ndyomugyenyi R, Magnussen P, Clarke S. Malaria treatment seeking behaviour and drug prescription practices in an area of low transmission in Uganda: implication for prevention and control. Trans R Soc Trop Med Hyg 2007: 101: 209 215. 18. Uzochukwu BSC, Onwujekwe OE. Socio-economic differences and health seeking behaviour for the diagnosis and treatment of malaria: a case study of four local government areas operating the Bamako initiative programme in southeast Nigeria. Int J Equity Health 2004: 3: 6. 19. McCombie S. Treatment seeking for malaria: a review of recent research. Soc Sci Med 1996: 43: 933 945. 20. Ahorlu CK, Koram KA, Ahorlu C et al. Sociocultural determinants of treatment delay for childhood malaria in southern Ghana. Trop Med Int Health 2006: 11: 1022 1031. 21. Saha KB, Sharma RK, Mishra R et al. Establishing communication mechanism for malaria prevention in Baiga tribal villages in Baiga Chak area of Dindori district, Madhya Pradesh. Indian J Med Res 2015: 141: 576 583. Corresponding Author Kalyan B. Saha, ICMR-National Institute for Research in Tribal Health, Nagpur Road, Post Garha, Jabalpur 482003, Madhya Pradesh, India. E-mail: kalyansaha_icmr@yahoo.com 1384 2017 John Wiley & Sons Ltd