We have examined the cause of death reporting system in India, using the nine aforementioned criteria to assess the usability of the cause of death statistics generated thereof . In Table -3.12 we summarize the findings and give our own rating of the contemporary Indian cause of death reporting system. We have followed a three category rating consisting of satisfactory, tolerable and poor. The prime factors affecting usability of the cause of death statistics in India are (a) poor coverage, (b) tendency to assign deaths to residual and "unclassifiable" categories, (c) long delay and irregular publication of statistics, and (d) lack of systematic screening. We present herewith our subjective assessment of factors contributing to various aspects of poor performance, and then discuss possible measures that we think will improve the usability of cause of death statistics in India.

           Poor coverage has two aspects, namely (a) a total lack of reporting from certain areas, and (b) under- reporting from other areas. These areas are sample villages in case of SCD-Rural and non-reporting municipalities in case of the MCCD. Total non-compliance is a result of lack of awareness. A visit to a few SCD-Rural training programmes in Andhra Pradesh during the year 1998 revealed that some PHC medical officers were totally unaware of the cause of death reporting system that was supposed to operate in their area. Total apathy reigns supreme at all levels, regardless of whether the areas concerned are urban or rural. Most health care institutions, clinical teams, medical attendants and municipal offices are manifestly guilty of either ignorance or complacency, thus contributing to very poor coverage by cause of death reporting systems. Much of this apathy and managerial inattention could be attributed perhaps to the fact that the data is being analysed at the national level. There is neither any mechanism or any effort to analyse cause of death data at the state level and use the results for state level health policy analysis. Consequently, field agents and medical practitioners neither have any means of direct feedback about the nature of utilisation of data collected by them nor do they perceive any stake in the latter. This contributes to a gradual deterioration in the accuracy of cause of death statistics.

           A review of the state of cause of death reporting systems and revamping of the is needed. It is imperative for each state to build an infrastructure for local analysis of causes of death. Sponsored research to analyse cause of death statistics and the implications for health policy, will, it is hoped, generate the requisite enthusiasm for usable statistics. In addition, state departments of health and municipal administration need to pay some managerial attention and periodically review the performance of cause of death reporting systems. We feel that if the initiative is taken up by the health and municipal administration departments, consecutively for a period of, say, five years, the coverage of the MCCD scheme would improve substantially. It is our conjecture that once coverage is increased substantially to about 80% of estimated deaths, it is likely to sustain itself without the need for much managerial and supervisory resources. A drive is in order, so that there is widespread awareness of the links between filing of individual cause of death reports and the utility of cause of death statistics in planning and policy analysis. The RBD Act provides for a fine of upto Rs. 50 for the absence of, or incorrect of filling of cause of death reports. Our experience in Andhra Pradesh proves that this provision has not been used at all. The fine prescribed by the RBD Act is more of a token amount than a real financial burden on health care providers. We feel that this fact can be suitably exploited to increase awareness among health care providers about filing cause of death report so that the fine is an effective reminder to defaulting health care providers, enabling them to comply with the legal requirement of filing a cause of death report.

           The high incidence of unclassifiable deaths can be attributed to report-writing skills. Chiefs of clinical units do not appropriately emphasize the importance of writing up the cause of death report. Short term training programmes that help build cause of death report writing skills are necessary to remedy this lacuna. Non maintenance or poor maintenance of medical records also contribute to inaccurate assignment of cause of death. Faced with a situation of inadequate information from medical records, the physician writing the cause of death report tends to the assign the death to unclassifiable category or to some miscellaneous codes. Hence the RBD Act needs further amendments requiring health care providers to maintain appropriate medical records in order to facilitate accurate classification of cause of death.

           Delay in compilation and publication of cause of death statistics can be further reduced by computerising the operations. At present a lot of the tabulation work is done manually. Some amount of computerisation has already been implemented in the office of the RGI, which in turn has also been subcontracting data entry to private computer service providers. Computerisation needs to be enforced at the state level, so that state level statistics can be published locally and utilised to inform state health policies. The tabulation and publication of state level statistics should be decentralised to State Vital Statistics offices. If the operations are computerised at the state and municipal level, this decentralisation can be achieved without any significant addition to current staff. Suffice to say that computerisation of cause of death report filling and collation is essential.