Indmedica Home | About Indmedica | Medical Jobs | Advertise On Indmedica
Search Indmedica Web
Indmedica - India's premier medical portal

Indian Journal of Community Medicine

Risk Factors for Sick Children in a Fisherman Community in Pondicherry

Author(s): M. B. Sudharsanam

Vol. 31, No. 4 (2006-10 - 2006-12)

M. B. Sudharsanam


India Under-five children constitute about 13% of theTotal population of India. WHO report of 2002 estimates the probability of dying for under-five children to be around 91 per 1000. In India, the mortality of under- five children is mainly due to acute respiratory infections (ARI) (23%) and diarrheal diseases (18%) as per WHO report 2002. National Family Health Survey (NFHS) of 1998 used acute respiratory infections, diarrhea and fever as essential parameters for the morbidity profile of children as these constituted a major part of sickness in this population. This study is an attempt to find out the prevalence sickness in under-five children and to identify the risk factors for sickness in these children. The knowledge about these risk factors will help the administrators to decide about interventions.

Material and Methods

It was a case control study undertaken in a fishermen village, Veerampattinarry in Pondicherry Union Territory. The study was carried from March 2004 to June 2004 in the under five population of the village. Enumeration list done for Pulse Polio immunization during January 2004 was used for enlisting the children. All children between 2 and 59 months of age were enrolled and the entire population was screened for sickness without any sampling. A Sick child was defined as any child between 2 and 59 months, with fever (lasting at least 2 days) or acute respiratory infection or earache or diarrhea (2 episodes of watery loose stools for at least 1 day). A period of 2 weeks was chosen as the criterion to ensure better recall on the part of the mothers about the illness and also because these illnesses are supposed to recur frequently among children. A pre-tested interview schedule was used to obtain the requred information after getting informed consent from the mother. Details were collected from the father or other members of the family only if the mother was not available at the time of the interview. Out of 441 children, 406 were contacted for sickness screening by enquiring the mother. In those houses, which were locked at the time of the first visit, attempt was made to contact the children by making at least two more visits to the house in the next one week. In spite of these only 406 children could be met and 35 children (8%) were excluded from study. None of the parents denied participating in the study. We selected an age, matched child (+/- 2months) control without sickness in the past 2 weeks for each case. Details of information collected from both cases and controls included age, sex, birth order; mother’s education and occupation; father’s occupation; number of person and, children in the family; monthly per capita income; persons with a similar illness in the family; smoking among family members; coughing practices of family members; person who usually takes care of the child; housing type, ventilation, fuel for cooking; smoke vent, sullage disposal, toilet availability; breast feeding duration, age at weaning; previous episodes of diarrhea, fever, ARI; immunization status for age; hygienic practices like washing hands before eating and after defecation and anthropometrical measurements like height, weight and mid-arm circumference. This study was done after getting clearance from JIPMER research council and ethical committee.


Sickness was found more in 3-12 and 13-24 months (both contributing to 17.9% of cases) and least in 37-48 months (8% of cases). The prevalence of sickness in the study population was 14%. Prevalence of ARI was the highest (5.6%), followed by that of diarrhea (4.7%) and then fever (3.7%). Table I shows the association of various risk factors by univariate analysis. Children of fathers who worked outside the village (gas company worker, daily wages construction worker, small companies worker) were at 2.3 times higher risk than of fisherman children. Children in whom any other member in the same family had similar illness within the previous 2 weeks were at a 31 fold higher risk of being sick than who did not have. Children with worm infestation were at 4 times higher risk of being sick than children without worm infestation. Children of mothers who did not practice any method of contraception were at three times higher risk than children of mothers who practiced any method of contraception. Other factors considered were not significant.

Table I. Risk Factors for Sick Children by Univariate Analysis

Risk Factors Cases Control Odds Ratio
(95% CI)
No. % No. %
Other 24 42 13 22.4 2.35 (1.04-5.31)
Fishermen 33 58 42 77.6 1.00 (Indicator)
Total 57 100 55 100  
with illness
Present 21 36.8 1 1.9 31.50(4.05-244.67)
Absent 36 63.2 54 98.1 1.00 (Indicator)
Total 57 100 55 100  
Present 13 22.8 5 9.1 4.25(1.44-12.52)
Absent 44 77.2 50 90.9 1.00 (Indicator)
Total 57 100 55 100  
Practiced 32 56.1 17 30.9 2.861(1.32-6.21)
25 43.9 38 69.1 1.00 (Indicator)
  Total 57 100 55 100  

Table II shows the results of logistic regression for four significant risk factors identified in the univariate analysis Children with other family member with similar illness in the past 15 days had 30 times more risk than those who did not have the disease and children with worm infestation had 7 times more risk than children without worm infestation.

Table II. Risk Factors for Sick Children by Logistic Regression.

Risk Factors B S.E Odds Ratio with
(95% CI)
No contraceptive
0.579 0.475 1.79(0.70-4.53)
Presence of other
member with
similar illness
3.396 1.070 29.86(3.667-243.10)
Father’s occupation
other than fishing
0.600 0.500 0.55(0.21-1.46)
Presence of worm
1.909 0.592 6.75(2.12-21.51)


The consideration of “sick child” has brought to focus some of the common morbidities affecting the children underfi ve years of age. Some geographical differences can be determined by carefully conducted epidemiological studies. They have the advantage of well-defined denominators.

Our study showed a prevalence of 14% of sickness, which included respiratory infections (5.6%), acute diarrhea (4.7%), and fever (3.7%). These figures are very low when compared to all the other studies both in India and other developing countries. The National Family Health Survey 1998 showed that in India, among children upto three years, at least 19% of the children suffered from diarrhea, 19% from ARI and 30% from fever, two weeks prior to to the survey. In Tamilnadu, the nearby state, the survey revealed that the prevalence of fever was 23%, diarrhea 14% and ARI 10%.

The higher prevalence of sickness was observed among children under one year of age in our study. This finding is consistent with the NFHS 1998, and Fuchs et al have revealed no significant difference in morbidities among male and female children and our study also showed the same1. Both these studies revealed that children of higher birth order suffered more from diarrhea but our study showed no significant association between birth order and sickness1. Sickness was more in children of mothers who were illiterate or with just primary education as revealed by NFHS1998, but our study showed no significant association of maternal literacy with sickness. De Francisco in Gambia found no associations between mortality from ARI and maternal education socioeconomic status or with the age of the mother2. Our study also shows no significant association of age of the mother, literacy status, and maternal occupation with sickness. Hirata et al in his case control study on diarrhea has reported that children of farmer or gum planter [Odds ratio (OR) 6.6] had more risk than others and our study revealed that children of fathers who worked outside the village (Delivery boys, company worker, daily wages construction workers, Government class 3 and 4 workers) had more risk of being sick when compared with the children of fishermen but failed to be an independent risk factor after logistic regression analysis3. Sur et al and Simanjuntak et al revealed that children had 3.8 times and 11.1 times more risk respectively of having diarrhea if another member in the family had diarrhea and our study showed similar finding with risk of sickness being 30 times more if a household person had sickness in past 2 weeks4,5.


  1. Fuchs SC, Victora CG Risk and prognostic factors for diarrheal disease in Brazilian infants: a special case-control design application Cad Saude Publica. 2002; 18: 773-82.
  2. De Francisco A, Morris J, Hall AJ, Armstrong Schellenberg JR, Greenwood BM. Risk factors for mortality from acute lower respiratory tract infections in young Gambian children. Int J Epidemiol. 1996; 22:1174-82.
  3. Hirata M, Kuropakornpong V, Arun S, Sapchatura M, Kumnurak Sukpipatpanont B, Chongsuvivatwong V, Funahara Y, Sato S. A case control study of acute diarrheal disease among school age children in southern Thailand. Southeast Asian J Trop Med Public Health. 1997; 28 : Suppl 3:18-22.
  4. Sur D, Manna B, Deb AK, Deen JL, Danovaro-Holliday MC, von Seidlein L. Factors associated with reported diarrhoea episodes and treatment seeking in an urban slum of Kolkata, India: J Health Popul Nutr. 2004; 22:119-29.
  5. Simanjuntak CH, Punjabi NH, Wangsasaputra F, Nurdin D, Pulungsif SP, Rofiq A Diarrhoea episodes and treatmentseeking behaviour in a slum area of North Jakarta, Indonesia: J Health Popul Nutr. 2004; 22:119-29.

Department of Commurnity Medicine, Jawaharlal Institute of
Postgraduate Medical Sciences and Research (JIPMER), Pondicherry
E-mail: [email protected]
Received: 16.6.05

Access free medical resources from Wiley-Blackwell now!

About Indmedica - Conditions of Usage - Advertise On Indmedica - Contact Us

Copyright © 2005 Indmedica