Medical history statistics were barely of interest to the local public in past decades. Since the mid1980s, however, the seemingly unstoppable rise of Acquired Immune Deficiency Disease (AIDS) has again shown how devastating infectious diseases can be in the absence of an effective antidote. And even diseases which have long been considered curable still claim numerous victims. A 1993 polio epidemic in Holland, for example, vividly demonstrated that only permanent, active resistance guards against resurrection of horrors long considered overcome. A look past the borders of the European continent makes the enormous challenges faced by modern medicine even more obvious. Strictly speaking, only smallpox, which due to the population’s lack of confidence in immunization still ravaged this country with epidemics in the 1920s, can be considered overcome on a worldwide basis. Though other infectious diseases, such as cholera, pneumonia, typhoid and puerperal fever, but also children’s diseases such as measles, scarlet fever and rubella, also the twins diphtheria and whooping cough, “angels of death” in common parlance, and finally tuberculosis, one of the banes of the 19th and early 20th century, have largely lost their earlier prominent places in the cause of death statistics, one could easily argue that they have merely been replaced by modern “civilization diseases” such as cancer and circulation problems. It must be taken into account, however, that the average life expectancy especially among younger people has risen dramatically in the period of 1890–1990 (see table D.1.). For this reason, caution is in order when, for example, quantitatively interpreting the long term development of a disease as imprecisely defined as “old age”. The same caution should be applied to long term statistical analysis of deaths caused by circulation problems and cancer. In general, this raises the question to what extent the – compared to the present – much lower life expectancy in the 19th century is responsible for the lower instead of higher incidence of certain diseases.
This chapter does not present a comprehensive historical-statistical picture of public health in Switzerland; rather, it deals exclusively with the Swiss epidemiological conditions in the 19th and 20th century. For instance, we refrained from collecting data on the development of public health costs, patient levels of individual hospitals, or medical insurance membership levels. The section concentrates on the three areas health care personnel levels, officially registered incidents of disease, and causes of death, with particular emphasis on the last mentioned due to the availability of pertaining data. The following elaborations reflect these three asymmetrically divided themes.
Health care personnel
The number and variety of health care professions continuingly decreased during the 19th century, with barbers, barber-surgeons, wound doctors, and lower level surgeons increasingly being replaced by professionally trained physicians who had to prove their medical knowledge to canton committees. It is virtually impossible to compare official health care statistics between cantons in the time between 1810 and 1860 since the description of individual health care professions varied from canton to canton. After that, inter-canton consensus began to form on the unacceptability of this situation. After a few cantons reached agreement on mutual acceptance of their respective medical diplomas, a treaty on mutual acceptance of health care personnel was entered into by Zurich, Berne, Schwyz, Glarus, Solothurn, Schaffhausen, Appenzell Ausserrhoden, St. Gall, and Thurgovia in 1867. The treaty was later joined by Basle-Country, Lucerne, Uri, Zoug, Appenzell Innerrhoden, Argovia, Neuchâtel, and Grisons. Other cantons, though, retained their earlier laws and did not join the treaty. This led to several “hideouts” where health care personnel banned from the treaty cantons were able to continue to practice their craft. At the federal level, the freedom of movement of health care personnel has been regulated since 1877 by a law which makes freedom of establishment contingent upon the passing of a federally accepted professional exam. In 1886, this exam, which up to that point had only applied to physicians, pharmacists, and veterinarians, was extended to include dentists as well. Individuals in the possession of a diploma of the treaty of 1867, or a canton equivalent, were also given the freedom of establishment by the Federal law of 1877. Subsequent treaties between Switzerland and the German Empire (1884), the Principality of Liechtenstein (1885), Austria-Hungary (1885), Italy (1888), and France (1888) also guaranteed foreign physicians freedom of establishment.
The quality of the statistical surveys of health care personnel apparently benefited from these developments a great deal. Even before the turn of the century, the number of practicing physicians was tallied by canton in a set of very credible statistics. Changes in the number of practicing pharmacists and dentists, too, can easily be documented in long term statistical series since approximately 1890 while a proper tally of veterinarians apparently continued to elude governments. Midwives had commonly been counted as health care personnel together with physicians, pharmacists, dentists, and veterinarians in the sanitary medical reports of the 19th century. However, the Swiss Statistical Year Book of Switzerland only reports on them by canton until 1928. For the following decades, we had to rely on the population counts of the years 1930, 1941, 1950, 1960, 1970, and 1980 for a tally of employed and self-employed midwives.
Diseases registered with the police
In 1886, the Swiss population voted for the introduction of an obligatory registration of “epidemics dangerous to the public”. Only four diseases were initially covered under this term, those being smallpox, cholera, typhus, and the plague. One year later, there was agreement to extend obligatory registration to other infectious diseases, and four years after that, the Sanitary Demographic Weekly Bulletin (“Sanitarisch-demographisches Wochenbülletin” / “Bulletin hebdomadaire démographique et sanitaire suisse”), a precursor to today’s bulletin by the Federal Department of Health began publishing disease reports from the cantons. Nonetheless, even after the caesurae of 1886, 1887, and 1891, the official figures on the incidence of diseases only managed to convey a vague guess at the actual epidemiological picture. Responsible for the comparatively low reliability of Swiss morbidity statistics of the late 19th century were most likely the low physician density, negligently performed surveys, and insufficiently educated diagnostics. On the other hand, in subsequent decades, the disease registration error quota clearly decreased. It follows that, at least for certain diseases, the secular morbidity reduction progressed on a much steeper curve than suggested by statistical registrations. This does not entirely eliminate the interpretation value of the data, however; given proper caution, it can well be used to arrive at an approximate impression of the morbidity progression of certain diseases during the past century.
Unfortunately, the same does not hold true for the Swiss hospital statistics, and even less for recordings by physicians in private practice. Even today, the statistics of the Union of Swiss Hospitals (VESKA/ASEH) are based on the voluntary participation of a few hundred hospitals which cannot be said to have representative character. As for physicians in private practice, there simply were not enough to attempt to use their data for a national extrapolation or even one at the canton level.
Causes of death
When drafting the first national statistics on causes of death, the Swiss Physicians Commission looked towards England which had introduced such a compilation already in 1836. In 1875, the Physicians Commission succeeded in getting the Federal Assembly to mandate the Federal Statistical Office with reporting on causes of death as certified by physicians in addition to issuing annual statistics on births, deaths, and marriages, a task the Bureau had done since 1867. However, the issuance of a certificate of death including a precise description of the primary cause of death did not become obligatory on a national level, rather it was vaguely described as a duty which physicians were to perform “whenever possible”. Before 1875, official certification of the cause of death was mandatory in only a few cantons, those being Zurich, Basle-City, Neuchâtel, and Geneva. Since there was no unified nomenclature for causes of death before 1875, for most diseases it makes little sense to compare causes of death among cantons for the 1850s, 1860s, and early 1870s. Statistical interpretation of mortality curves can only be performed, at best, for victims of particular diseases in certain cantons. We arrived at such series – of uneven length and quality – for Zurich, Schwyz, Glarus, Solothurn, Basle-City, Schaffhausen, Appenzell Innerrhoden, St. Gall, Thurgovia, and Neuchâtel by either using the annual audit reports of the government or the sanitary surgeon general, or by using other sources that were considered reasonably reliable.
It seems advisable to refrain from blindly connecting those series originating from the early and advancing 19th century with data published since 1876 in the “Statistical Deliveries” (“Statistische Lieferungen” / “Livraisons statistiques”). Not only did fundamental changes in the terminology of diseases occur during that period of time, it’s even more important to realize that traditional terms such as “gall fever”, “mucous fever”, or “bloated throat” trace back to a concept of medicine which had nothing in common with later systems reflecting the more advanced knowledge of modern medicine. The following two examples dramatically illustrate what this situation implies: “Typhus”, today a synonym for stomach typhus, was used to describe almost any condition that caused confusion or stomach-related nausea. And what the sanitary physician generals of Zurich and Thurgovia listed as “dropsy” (“Wassersucht”) in the 40’s, 50’s, and 60’s of the last century does not describe some sort of uncommon addiction in the sense of “Sucht” which had been combated so successfully that there wasn’t even a need to include it in the revised cause-of-death statistics of 1876, but merely described certain heart and kidney problems, and chronical malnourishment symptoms. The comparison of the Federal Statistical Office’s cause-of-death statistics with earlier canton figures regarding birth and puerperal mortality, tuberculosis, and certain other acute epidemic diseases seems less problematical. Clearly comparable, on the other hand, are statistics reporting on death by external force (suicide, accidents, murder and manslaughter, respectively).
In certain cantons, the recording of causes of death remained lacking even after 1875, as evidenced by the statistics on non-certified and inaccurately diagnosed deaths for each canton. One of the large surveys dedicated to causes of death in its “Marriage, Birth, and Death” series of special publications, the Federal Statistical Office reveals the frustration officials experienced over certificates of death submitted by arrogant physicians and processed by ignorant civil servants. The major problem was the proverbial illegibility of physicians’ handwriting; a second the fact that certain physicians insisted on reporting the identified cause of death in Latin. Thirdly, physicians in certain parts of the country, especially in Wallis, were not particularly cooperative, as evidenced by the disappointingly high percentage of non-certified or inadequately described causes of death in those cantons. Exceptionally high even by international standards, this rate was surely affected also by the fact that, unlike other countries, Switzerland mandated the issuance of a physician’s official certificate of death by law.
It was the task of the civil servants to transcribe the physicians’ reports into final form and to pass them on to the Federal Statistical Office. There, it quickly became obvious that certain civil servants were not up to the task of successfully handling the physicians’ Latin terminology as certain lists reported diseases which were not part of any official nomenclature. Through the performance of double checks, the Office hoped to alleviate the damage done to the overall statistics through this kind of errors. This plan might have been successful had it not been for a fourth source of errors, one completely beyond any sort of examination – the accuracy of the physicians’ diagnoses. Considering the result of recent scientific studies which indicate that a contradiction between pathological analysis and the physician’s determination of the primary cause of death is not all that rare, one can imagine the number of misdiagnosed causes of death a hundred years ago.
Even when the cause of death was properly diagnosed, the question comes up whether or not a second disease, one also fatal in the long term, had been overlooked. How many accident victims, for example, also suffered from tuberculosis? How many who were suffering from cancer died “prematurely” of typhus? And was the wish for quick relief from an incurable disease an irrelevant or statistically significant suicide motive?
And there is yet another area of concerns with regard to excessively liberal use of cause-of-death statistics: analogous to the recording of births, marriages, and divorces (see chapter C.), the number of victims of a certain disease relates to the residential population since 1890, whereas from 1876 to 1890 the number of reference had been the population present at the time of the respective count. In 1889 and 1890, the Federal Statistical Bureau published two tables for the large and medium sized cities, one of which records actual location at the time of death and the other the deceased’s official place of residence. A comparison of the tables reveals substantial deviation between the two series for certain diseases.
It is quite obvious that the tables in this chapter do not speak for themselves. Nonetheless, it would be inadvisable to simply dismiss Switzerland’s statistical cause of death data of the late 19th and early 20th century. As much as the application of major qualifications and reservations is in order, it would be inappropriate to summarily deny the validity and usefulness of this source of information. The need for such radical judgment is alleviated even more by the Federal Statistical Office’s commendable efforts to continuingly improve the quality of its surveys since the publication of the very first volume of the “Statistical Deliveries”. One result of these efforts was the complete change in the causes of death nomenclature in the year 1901. At the same time, the Bureau’s introduction of anonymous certificates of death broke the argument that a precise description of the cause of death represented a violation of medical confidentiality.
In the first half of this century, each census had a parallel revision of the cause of death nomenclature. There were no further changes in 1960 as there were already efforts underway to adopt the “International Code of Diseases” (ICD). In 1970, Switzerland adopted the code; since then, Swiss mortality statistics enjoy the advantage of international comparability. The disadvantage of a four digit international code substantially inferior in accuracy to the Swiss classification must be seen as irrelevant as at the same time, a five digit code was introduced which maintains the finer differentiations of the Swiss nomenclature below the ICD’s four levels. Since this represented the eighth Swiss revision of the cause of death statistics, the classification, which is still in use today, was named ICD-8.