Medical cause of death certification provides vital information regarding public health indicators for monitoring of health policies. They are an important source of information for assessing the patterns of diseases in a given population. Lack of reliable cause of death data will impede monitoring and evaluation of health related activities, research and thereby can potentially lead to misleading information regarding health care.
MCCD are an important source of information for population and hospital based cancer registries. In India, active cancer registration is done, where trained registry staff visits different places (diagnostic laboratories, hospitals and vital statistics departments) for collection of cancer related information on a standard format. The death certificate notifications are followed-up and those cases where only death certificate is available and no other medical records are found, are assigned as death certificate only (DCO). It is often observed during data collection in cancer registries of lower income countries that the death certificates can be the only source of cancer statistics for the country. Varying patterns of DCO % and Mortality: Incidence ratio were observed among different population based cancer registries which is dependent on the quality of death certification. In some registries very few DCO is obtained due to non-availability or poor quality of certification of cause of death. Hence, poorly prepared death certificates will lead to inadequate data collection for the registries, which in turn will affect the cancer related statistics and thereby influence monitoring and evaluation of cancer prevention and control activities.5
We observed that in almost half (49.3%) of the MCCDs there were errors related to demographic information of the deceased. The MCCDs either had no/ incomplete/incorrect entry regarding age, gender and residence of the deceased. Inaccurate age will affect the age related epidemiological parameters such as age related incidence and mortality rates. Incomplete/ incorrect residential address will affect the geographical distribution of cancer and tracing of the families of the deceased for data collection. Similarly, incorrect/lack of gender information can affect gender wise patterns of cancer burden and mortality. Inaccurate demographic information and wrong legal nominee may lead to difficulty in medical reimbursements from the insurance organizations and future medico-legal claims. One possible explanation for demographic errors may also be due to several patients being brought dead or the accompanying attendant not closely related to the deceased.
We observed that nearly one-fifth (19.4%) of the MCCDs had technical errors such as incorrect/absence of name of the physician, absence of signature, seal of the hospital, signing date, time of death, date of death which makes the MCCDs invalid for medico-legal purposes. For registering a death, identity of the deceased, date and time of death and cause of death are mandatory for the death registering authorities. Absence of any of these details will not allow registration of the death.
In some of the death certificates, illegible writing and use of abbreviations/short forms were seen in 11% and 12% of the certificates, respectively. Abbreviations and short forms of disease should not be used as they are likely to lead to confusion in statistical office. Illegible writing and abbreviation of medical terms makes the understanding of MCCDs difficult for ICD coding.11
Only 30 (4 %) death certificates had no errors. We observed Grade IV errors in 75.9% and Grade V errors in 75.1% certificates. Grade IV and V errors are the major errors in death certification. Wrong diagnosis/incorrect filing up of the certificate will result in gross errors in mortality statistics thus directly affecting formulation of national health policy. In addition, the certified cause of death is subjected to legal scrutiny in medico-legal deaths. The importance of recording the sequence correctly lies in the fact that appropriate strategies can be adopted to cut the chain at its most vulnerable point and thus prevent death.
Terminal events leading to death like circulatory failure, respiratory failure etc. as modes of dying should be avoided as they are no more than signs of death and provide no useful information for the underlying disease process.8 In our study, 231(29.7%) death certificates have labelled modes of death as immediate cause of death. Most clinicians confuse the cause of death with the modes of death.12 This finding is observed frequently in lower-middle income countries.1, 13 In the Indian context, many a times these deaths are unattended, hence, exact time, date and terminal event symptomatology is not available so the cause of death is also not known. However, the same can be established by the clinical post-mortem examination after the consent of the relatives.8
Globally, similar outcomes have been reported in the past. Adjacent countries like Pakistan, Nepal and Bangladesh have also reported a high percentage of errors in the cause of death certificates.1,14,15 These errors are not just limited to the developing countries. More than 50% of general practitioners in the United Kingdom and in the US reported being insufficiently instructed about the process of death certification.1 It is difficult to compare our findings with previous studies done in India due to differences in the definitions and interpretations of errors between studies. However, there is consistent finding among most of the previous studies, including ours, that the majority of the MCCDs have wrong cause of death certification, which qualify as major errors.16–23 The current study included death certification for only cancer related deaths. Hence, our findings may not be generalizable to other establishments.
It was observed that majority of the death certificates are prepared by the MBBS doctors and the proportion certification errors committed by the MBBS doctors was significantly higher as compared to the post-graduate residents and doctors who have higher clinical experience. Another possible reason of higher proportion of errors by junior doctors is that the treating physician, who is generally the senior doctor and has all the details of the deceased medical conditions, may not be available at the time of death certification.
Lastly, a total of sixteen COVID-19 related death took place in the study period and all of them had major certification errors. Inaccurate COVID-19 related death certification will contribute to misleading data for COVID-related statistics.