|Down Load PDF|
APPENDIX 6 - PROTOCOL
A NATIONAL SAMPLE SURVEY TO ESTIMATE
ANNUAL RISK OF TUBERCULOUS INFECTION IN INDIA
Tuberculosis (TB) continues to be a major public health problem in India. Of the various epidemiological parameters for assessment of TB situation in any community, annual risk of tuberculous infection (ARTI) is currently preferred since disease surveys are too expensive and health information system suffers from inadequacies. Most of the tuberculin surveys carried out so far in the country for estimation of ARI have been localized and thus do not represent TB situation either in the country as a whole or in the respective regions. Therefore, it is proposed to conduct a nation-wide tuberculin survey, in order to assess the current epidemiological situation of TB in different zones of the country, by computing ARTI.
The analysis for estimating prevalence of infection and ARTI was performed among 17811 satisfactorily test read children without BCG scar.
The survey shall be conducted among children 1-9 years of age, residing in rural as well as urban areas of the districts selected from different parts of the country by appropriate statistical methods.
The survey shall be conducted in 24 districts (about 5% sample of the total number of districts in the country - 466 according to 1991 census). For purpose of the survey, the country has been divided into four zones with equitable population as per the 1991 census, viz. north, east, south and west. The states included in each of these zones are given in Annexure I.
All the districts in a particular zone are listed in the ascending order of population size and six districts are selected by systematic sampling. The required number of rural and urban clusters is based on the ratio of rural and urban population in the entire zone. The number of clusters (rural, urban) to be covered in individual districts are determined on pro-rata basis of the selected districts.
Selection of clusters: For purpose of the survey, a village is considered as a rural cluster and a census enumeration block in the urban areas is considered as an urban cluster. The appropriate number of clusters shall be selected using PPS (population proportion to size) method. In each cluster, 85 eligible children shall be registered following EPI survey technique.
In each cluster, 85 children shall be registered for tuberculin testing. It is expected that at least 25 (30%) of them shall be without BCG scar. Considering a drop out of 20% between registration and reading of the tuberculin reaction sizes, there shall be 20 testread children without BCG scar, from each of the clusters. Therefore, in each of the four zones, about 51,000 children shall be registered. This will include a minimum of 12,000 test-read children without BCG scar. This sample size has been calculated considering an expected prevalence of infection of about 8% in the study population and is expected to give an estimate of prevalence of infection for each zone within 10% of the true value at 5% level of significance. (Please see section-2 of this report for computation of the sample size). Thus in the entire country, about 2,04,000 children shall be registered into the study which shall include a minimum of 48,000 test-read children without BCG scar.
The Team Leader in conjunction with Supervisor from NTI/other Collobarating Agency shall draw the survey plan for each district. The children residing in a particular cluster for more than six months shall be registered on house to house visits by trained field workers, as per the Work Instructions. Each registered child will be injected intradermally on the volar aspect of the left forearm with 1TU of PPD RT23 with Tween-80, using a suitable disposable tuberculin syringe and needle. The BCG scar status of each tested child shall be ascertained by examination of both the shoulders. The tuberculin vials shall be obtained from BCG Lab, Guindy and transported to field under proper storage conditions. Trained readers will read the test results three days after the tuberculin test and record the maximum transverse diameter of induration. However, in case of exigencies, the reading may be performed after 2 or 4 days of the test. The data shall be recorded on pre-structured proformae.
The fieldwork including planning, registration, tuberculin testing & reading and recording of the data shall be performed by field workers trained in such work, as per the Work Instructions prepared by NTI, Bangalore.
A pilot study to pre-test the procedures and data entry formats shall be conducted among 7 rural and 3 urban clusters.
The quality control of tuberculin testing shall be performed by direct observation of the tests given by field workers. The Supervisors from NTI and other Collaborating Institutions shall do this observation. The proportion of unsatisfactory tests shall not be allowed to exceed 2-3%. For quality control of reading of the reactions, 5% of the reaction sizes shall be read independently by these Supervisors. Corrective actions shall be undertaken as and when required. The quality control of data entry shall be undertaken by scrutiny of the day's work including individual entries in child cards at the end of each day's work. The data shall again be rechecked at NTI and in case of gross discrepancies from any clusters, appropriate number of new clusters shall be selected from that district to complete the required number of clusters from the district.
Staff and Training
For each zone, 18 field workers shall be recruited on contractual basis for the period of the survey. These workers shall be divided into two teams each consisting of 8 field workers and 1 team leader. Each of the teams shall be accompanied during fieldwork by a supervisor from NTI or from other collaborating institutions.
In all, 64 field workers, 8 team leaders and 8 health supervisors shall undergo training under the project in field planning, registration, tuberculin testing and reading and recording of the data. These workers shall be trained as teams consisting of 8 field workers, one team leader and one health supervisor for a minimum of 40 working days either at Tuberculosis Research Centre, Chennai or at National Tuberculosis Institute, Bangalore.
Collaboration with other Institutions
Institutes with previous experience in conducting epidemiological surveys shall be identified and assigned the responsibility of day to day implementation, supervision and control of the field work in a proportion of the districts. The cooperation from the State TB Officer and District TB Officers shall be sought in terms of local support such as deputing health supervisors for the period of the survey who shall be responsible for eliciting community participation and coordinating the activities within their districts.
A period of approximately 18 months is required for each zone including training and fieldwork, in case of two teams being deployed for each zone. The study period shall increase in case only one team is available for a particular zone.
All the reagents and procedures to be employed in these studies have been widely used and constitute no hazard to the subjects except transient itching and swelling among a small proportion of the children.
The purpose of the survey shall be explained to community leaders and parents/guardians of the children, from whom a written consent shall be obtained, on the form designed by the Ethical Committee. No individual shall be forced to participate. The children with suspicion of tuberculosis shall be referred to the nearest Government health center for further investigations. They shall be appropriately treated through the district health services, if found to be suffering from tuberculosis.
The data collected will be analyzed centrally at NTI. The data will be double entered into the computer to minimize keypunch errors.
The data analysis shall be performed primarily for the children without BCG scar. The tuberculin reaction sizes obtained in a particular zone shall be arranged in the form of frequency distribution tables. Using the frequency distribution curves, a demarcation level for identification of the children with tuberculous infection and estimating prevalance of infection shall be decided. Prevalance of infection will also be estimated by mirror image technique by identifying the mode of reactors attributable to tuberculous infection. The annual risk of tuberculous infection from each zone shall be estimated using the following formula:
ARTI = 1 - (1-P) 1/A
where P is the estimated prevalence of infection for that zone and A is the mean age of the children test-read.
An attempt shall also be made to estimate ARTI among children with BCG scar, if found feasible.
The tuberculin reaction sizes obtained among smear positive cases of pulmonary tuberculosis shall be analyzed in order to find out the sensitivity of the test at different levels of demarcation.
The total budget requirement for implementation of the project is 1.5 crore Indian rupees.
Zones and List of States
* 1991 Census
Total Population = 846303