[Death due to tuberculosis requiring treatment or an accompanying disease? A contribution to the lethality and mortality of tuberculosis in Germany].

BACKGROUND

Tuberculosis (TB) is a curable disease. Nevertheless, patients in Germany also die of TB. Although mortality is decreasing, there are indications for an increase in lethality. This observation provided the impetus for a detailed analysis that sought to investigate the validity of the statistics on deaths caused by TB.

METHOD

The study population consists of the 926 fatal cases that were classified either as "death from TB" or as "death due to other causes" out of the 6044 TB patients in the DZK study. For the analysis, health authorities were asked to provide additional information and such documents as the death certificate, the autopsy protocol and the final medical report. In 778 cases, there was at least one additional piece of information available. Three teams of two experts each conducted independent evaluations of the documents.

RESULTS

Based on the findings of the experts, every second death caused by TB in 1997 and 1998 was not recorded correctly during the post-mortem examination. Every third TB death was not diagnosed during the patient's lifetime. Patients who died due to TB were, on average, older and more likely to be born in Germany. This indicates that age-related comorbidity among the native German population plays a relevant role. Yet, the unicausal death registration did not acknowledge comorbidity as a contributing factor to the fatal outcome. Pulmonary TB with positive microscopy and culture, miliary TB and meningeal TB were more common among the deaths due to TB than among the general study population, and led more often to a fatal outcome than other organ manifestations. However, the two groups did not differ with regard to multi-drug resistant TB. Alcohol abuse was a leading risk factor for death caused by TB in patients under 65 years. Patient's delay ranged from six to 34 days, and doctor's delay from eight to 46 days. For example, alcohol abusers, on average, visited a physician much later, but were diagnosed more rapidly after the first visit than patients who were not alcohol-dependent. A period of 32 - 200 days elapsed between diagnosis and death caused by TB.

CONCLUSIONS

The post-mortem examination often missed TB as the cause of death. Many native German TB patients showed age-related comorbidity. Pulmonary TB with positive microscopy, miliary TB and meningeal TB led more often to a fatal outcome than other organ manifestations. Alcohol abuse was a leading risk factor for TB deaths in patients younger the 65 years. The average period between the onset of symptoms and the diagnosis was significantly longer than the one month generally considered acceptable. The experts could not confirm an increase in lethality for the period under investigation.