Longitudinal cohort study

All Three Rounds Findings


Acute Diarrheal Diseases (ADD) is one class of diseases for which morbidity data is collected and reported by the Integrated Disease Surveillance Project (IDSP). Diarrhoea being the second most killer of childhood deaths worldwide, it is essential to study the incidence of diarrhoea in the intervention to understand the severity of the problem.

The study has been designed as a longitudinal cohort study in 6 states being followed over three cycles in a timespan of three months to explore different social, cultural, behavioural and systemic issues linked and contributing to high incidence of diarrhoea in the proposed intervention states/ locations. A cohort of 600 children (0-59months) affected with diarrhoea, were identified and surveyed.


To estimate the incidence of the diarrhoea cases in children under the age of 5 years, a one-sample formula was used.

The two-stage sampling methodology adopted was:

  1. Selection of Primary Sampling units
  2. Selection of households


Diarrhoea Prevalence
  • Diarrhoea prevalence in urban areas is much more than rural areas. Bihar and Jharkhand have same figures while Delhi and UP have figures that deviate the most from the secondary findings.
  • By the Chi-Square test, it was found that there is a significant relationship between residence type (rural/ urban) and diarrhoea prevalence at 90% level of confidence.
  • The association between diarrhoea prevalence and type of ration cards ownership was significant.


Social Demographic and Economic status of cohort children
  • The kaccha HHs are 1.1 times more likely to have children suffering from Diarrhoea as compared to pucca HHS.
  • About 62% families reported owing a ration card out of which more than 50% owned BPL cards.
  • By the Standard of Living Index (SLI) calculation, it was found that there is a strong correlation between type of house structure and SLI category of HHs i.e. HH that fall in a higher SLI category had more probability of residing in a pucca house structure.


Diarrhoea Incidence
  • In both the 2nd and 3rd round data collection, state-wise distribution of diarrhoea proportion indicates that this figure is maximum in the state of Bihar.
  • Considering both 2nd and 3rd rounds, it was seen that 30.9% of all cohort children had suffered from either during 2nd or 3rd round & Bihar again exhibited the maximum proportion.
  • As per SLI data collection, the repeat occurrence of diarrhoea has maximum proportion in the low SLI category.
  • It was found that the higher proportion of boys suffered from repeat occurrence of diarrhoea than girls.
  • It was also found that in the state of Bihar average episode of diarrhoea per child is much higher as compared to other states.


Symptoms, Treatment and Feeding practices
  • In 1st, 2nd and 3rd round, loose water stool, vomiting and stomach cramps were majorly quoted symptoms by more than half of the respondents.
  • On an average in all the three rounds, majority of the mothers of children (both below and above the age of 6 months) suffering from diarrhoea had either reduced the quantity of milk and other food items, or maintained the same quantity during the time their child was suffering from diarrhoea.
  • In all the 3 rounds, it was quoted that maximum no. of respondents relied on the private facilities for treatment.
  • Medicines in consultation with doctor were stated to be used by 55% of all HHs for the treatment of diarrhoea in all the 3 rounds. Commercially available ORS was used by only 14% of all HHs.


  • Diarrhoea is a seasonal disease, so factor of seasonality should also be taken into account.
  • Concerted efforts should be made to deal with the severe situation prevalent in Bihar.


  • The prevalence rate in West Bengal was found to be quite low because of the unavailability of the respondents during survey.
  • Since the follow up period was quite short, diarrhoea incidence amongst the selected cohort was found to be low in 2nd and 3rd round.
  • The study only stretched from Early-September to Mid-December, yearly estimates of incidence rates cannot be given, as seasonality effects cannot be taken into consideration.