The present study was a community based cross-sectional study conducted in area under Mohanpur Community Health Centre (CHC), the rural field practice area of Department of Community Medicine, Agartala Government Medical College. All 22 sub-centres were selected for the purpose of the study and the study population comprised of children between 12 to 23 months, residing in area under Mohanpur Community Health Centre (CHC). The study was conducted for a period of one year (November 2013 to October 2014).

Sample size calculation: Sample size was calculated by following methods-

A decision value of 2 was calculated from pre-designed standard statistical table from WHO guideline [3].

Informed consent: A duly explained, written consent, translated by trained personnel into local language (Bengali) was taken from all the parents before including them in the study.

Ethical consideration was taken from the Institutional Ethics committee, Agartala Government Medical College before commencement of the study.

Sampling technique: Out of all the eight districts in Tripura, West Tripura district was selected for the purpose of the present study. Out of total three sub-divisions under West Tripura, Mohanpur sub-division was selected using lottery method. Again by lottery method 15 children of 12 to 23 months age from each sub-centre were selected from the immunization register available at each sub-centre. And information regarding their immunization status was verified by visiting their homes.

Full immunization: Defined as immunization of a child with one dose of Bacille Calmette Guerin (BCG), 3 doses of Diphtheria Pertussis and Tetanus (DPT), Oral Polio Vaccine (OPV), Hepatitis B Vaccine and one dose of Measles vaccine within the age of one year.

Lot: Each Sub-centre within Mohanpur CHC area was considered as one Lot in the present study.

Data was collected by using a pre-designed and pre-tested, semi-structured questionnaire and immunization cards have been verified physically to confirm appropriate date of vaccination. In those patients where immunization card was not available verification was done based on examining BCG scar and interviewing the respondent during the home visit for every child.

Analysis was done using software SPSS version 21.0. Chi square test, Fisher’s-Exact test have been used to see the significant association between different variables as and when required.

The present study showed that, out of the total 330 children between 12-23 months age group, 197 (59.7%) were male and 133 (40.3%) were female. Mother was the respondent in 319 (96.7%) cases (father was interviewed where mother was not available at the time of home visit). Among them, 230 (69.7%) were Hindu by religion and 188 (57%) belonged to Scheduled Tribe (ST) caste. Majority of the fathers (163, 49.40%) had primary education and mothers, secondary education (156, 47.30%). Father’s occupation for majority of them (133, 40.30%) was farmer/fisherman/agriculture, followed by unskilled labourer (85, 25.80%) whereas, 258 (78.20%) of their mothers were housewives. According to modified B G Prasad’s scale, maximum (189, 57.30%) children belonged to Socio-Economic Status (SES) class IV. The birth order of majority (190, 57.60%) of the children were first and 251 (76.1%) of them were delivered at institution.

[Table/Fig-1] shows that out of the total 22 lots under Mohanpur CHC, eight (36.36%) were rejected. It also reveals that the total estimated coverage of full immunization for children under Mohanpur CHC was, 0.9167 x 100% = 91.67%.

Main reasons for failure of full immunization as shown in [Table/Fig-2] were, unawareness of need to return for second and third dose of vaccines (26.7%), illness of the child- not brought for immunization (26.7%), followed by fear of side reactions (20%) etc. The variables, found to have significant association with full immunization status of children in univariate analysis were religion of the child, social caste, Father’s literacy status, Father’s occupation and place of delivery of child [Table/Fig-3]. Multivariate logistic regression analysis [Table/Fig-4] with the variables found significant in univariate analysis showed that, children who belonged to General caste, had poor educational status of father (primary education) and were delivered at home have significantly lesser odds of being fully immunized then those belonging to Scheduled Caste (SC) and Other Backward Caste (OBC) caste, whose father’s education were secondary level and above and those who are institutionally delivered respectively.

This present study revealed that out of the total 330 study children, majority (59.7%) were males unlike a previous similar study conducted in Iran [9]. Mother was the respondent in 96.7% of cases in the present study, similar to the study by Karinagannanavar A et al., in Bellary district [10]. In their study, majority of the children were hindu (85.30%), similar to the present study, but 14.3% were muslims which is none in case of the present study, probably because of very less muslim population in the study area and difference in ethnicity of the study population. Educational status of the respondents in the present study was found to be better than similar other studies.

In this study, most of the children belonged to SES class IV (57.30%) as per modified BG Prasad’s scale, still institutional delivery was found to be much higher (76.1%) than that in the study by Karinagannanavar A et al., in Bellary District [10].

The present study showed that under Mohanpur CHC, 36.36% lots were under-performing even though full immunization coverage was 91.67% whereas, similar other studies [11–15] revealed, full immunization coverage to be 62.7%, 63.3%, 44.1%, 75% and 84.21% respectively which clearly shows that the full immunization coverage of area under Mohanpur CHC assessed by using LQAS technique is much higher than the overall state, national figures and similar various other studies done throughout the country. To assess the application of LQAS technique to measure immunization coverage identify the areas of high and low coverage, the study conducted by Pradeep BS et al., in Bangalore concluded that LQAS can be used as an effective tool to monitor routine immunization activity and it would decrease the time taken for evaluation of immunization coverage [15]. Punith K et al., in their study at Mathikere Urban Health Centre of Bangalore, have showed that, considering the time and resources required, LQAS was better in evaluating primary immunization coverage than cluster sampling [16].

The principal reasons for failure of full immunization found in the present study were, unawareness of need to return for second and third dose of vaccines (26.7%) and illness of the child (26.7%), similar to what Vohra et al., and Bholanath et al., revealed in their studies [11,13]. Again similar results were cited by Prabhakaran Nair et al., Punith et al., and Saxena et al., in their respective studies [14,16,17]. General caste, poor educational status of father and home delivery are the risk factors found significantly associated with failure of full immunization of the child in the present study. This shows even though study settings are different, the primary reasons for failure of full immunization status remains similar all over the country.

The target population was not adequate as per the precision set for this study using LQAS technique as the revised sampling fraction was more than 10% (usually in LQAS technique <10% sampling fraction is taken). To overcome this, a larger target population has to be considered in future for similar studies. Another limitation was the sampling frame of different lots were designed based on the immunization register available at each sub-centre and not by home visit, which might have led to bias if children without full immunization were not registered properly and could have overestimated the coverage of full immunization. Recall bias from the respondents who could not produce immunization card and no BCG scars found was another limitation that could be identified.

This present study shows that although the overall full immunization coverage among the children of Mohanpur CHC area is high, there are many (36.36%) pockets of under immunization, the primary reasons being unawareness of need to return for second and third dose of vaccines and illness of the child. This reveals that LQAS method is an effective one to identify specific areas of under-performance even if the overall performance is satisfactory to evaluate health programmes at community level.