Tuberculosis and War. Группа авторов
deaths (all forms) per 100,000 population, Norway, 1924–1950. Data to reproduce figures were abstracted from graph in [71], reproduced with the permission of the European Respiratory Society.
Norway, as mentioned earlier, is one of the few countries that has collected comprehensive TB morbidity data since the beginning of the 20th century. Figure 9 shows national notification rates of infectious respiratory TB and TB mortality rates (all forms) from 1924 through 1950 [71]. Tuberculosis cases declined to a nadir in 1941, then increased slowly to a peak in 1945, subsequently to resume the pre-war decline. The increase in cases during war years was a relatively modest 17.5%. A similar increase is not discernible for mortality. Calculating the case-to-death ratio from these data (Fig. 10) [71] shows a fairly flat course hovering around an average of about 1.6 from 1920 through 1943. Subsequently, the ratio gradually increases to 3.5 in 1950 in the wake of the progressively successive introduction of chemotherapy that reduced case fatality. The data from Norway do not confirm Rist’s hypothesis of accelerated progression from disease to death, which might be attributable to the relatively small effect the war seems to have had on TB in Norway.
Like Norway, Denmark introduced systematic notification of TB cases early on [72]. Figure 11 shows the comparison of notified pulmonary TB cases and TB deaths from 1921 through 1957 [72]. The first nadir of cases was reached in 1940, then increased to a peak in 1946, resulting in an increase of 53.4%; in other words, a much larger increase than reported from Norway during the war years. Over the same period, the number of deaths remained essentially unchanged. Calculating the case-to-death ratio, the ratio hovers around an average of slightly over 1.5 during the 11 years from 1921 through 1931. Subsequently, there is a regular, virtually exponential increase in the ratio by an average 5.2% annual increase until the end of the observation period (Fig. 12). The ratio had substantially increased to 3.1 in 1946 when the peak morbidity was reported. By the end of the observation period, the ratio had risen to over 5.5, testifying the power of chemotherapy in saving lives. In the period from 1940 to 1946, when cases increased, the ratio also increased. This is quite the opposite of what would be expected if Rist’s hypothesis for France holds good. In other words, there is no evidence in Denmark that cases of TB succumbed at an accelerated pace during the war years; in fact, it was just the reverse.
Fig. 10. Case-to-death ratio of notified infectious respiratory tuberculosis case rates to tuberculosis mortality (all forms) rates, Norway, 1924–1950. Underlying data to calculate ratios were abstracted from the graph in [71].
Fig. 11. Reported notifications of pulmonary tuberculosis cases and tuberculosis deaths (all forms), Denmark, 1921–1957. Data to draw graph were abstracted from [72], reproduced with the permission of the World Health Organization.
Fig. 12. Case-to-death ratio of notified pulmonary tuberculosis cases to tuberculosis deaths (all forms), Denmark, 1921–1957. Data to calculate ratios were abstracted from graph in [72], reproduced with the permission of the World Health Organization. The dashed line to the left of the 1931 vertical line is the average ratio during that period, the line to the right the regression from that year onwards through the end of the observation period.
Fig. 13. Reported cases and deaths from pulmonary tuberculosis, per 100,000 population, London and Paris, 1934–1944. Data to reproduce graph were abstracted from [73], reproduced with permission from Elsevier. For better proportional comparison, data are shown semi-logarithmically.
Finally, we turn to Paris and London for which comparative rates for pulmonary cases and deaths are available for the period 1934 through 1944 (Fig. 13) [73]. The reported burden is larger in Paris than in London, and the course is parallel for the observation period, except that in the final year of observation TB cases in Paris declined while in London they continued to increase. The case-to-death ratio (Fig. 14) is generally higher in London than in Paris, but there are major fluctuations in the ratio over the observation period. However, there seems to be no indication in either city that, while there are fluctuations, the ratio is getting systematically smaller during the war years compared to the pre-war years. On the contrary, should one like to construe an interpretation, if anything, the ratio may tend to get larger with time. Again this does not suggest accelerated progression to death during WWII.
Fig. 14. Case-to-death ratio of notified pulmonary tuberculosis cases and deaths, per 100,000 population, London and Paris, 1934–1944. Underlying data to calculate ratios were abstracted from graph in [73], reproduced with permission from Elsevier. For better proportional comparison, data are shown semi-logarithmically.
In summary, the examination of comparative morbidity and mortality data in various settings does not give any indication that TB tends to progress faster to death at times when TB morbidity increases during wartime. The data examined here were almost exclusively from Western European countries of the inter-war years and WWII, because morbidity data from WWI seem to be rarely available. Tuberculosis epidemiology during and in the aftermath of WWII is, however, not at all comparable with that during and after WWI. The reported excess peak in mortality in Germany reported in WWII was not anywhere nearly as large as that reported for WWI. Second, the period after WWII brought chemotherapy, which had a substantial impact on the duration of infectiousness of cases and case fatality. The 2 post-war situations cannot thus possibly be compared. The phenomenon reported by Redeker [62] of a seeming resumption of the pre-war (WWI) secular trend without a ratchet effect after the hyperinflation period (Fig. 4 chapter 1) stands out as a unique puzzle. A similar phenomenon has been reported for some countries, but few had actually a sufficiently long observation period before WWI and/or did collect data during the war years [74, 75]. Among those countries with longer series