Search

This paper deals with statistical situations which are typical of many planned experiments in medical statistics where the effect of certain treatments are measured by mortality, recurrence of attacks, recovery etc. The classical methods in mortality statistics developed by actuaries and demographers are predominantly concerned with the problem of obtaining unbiased estimates in face of varying duration of exposure to risk and several causes of departure. In recent litterature this situation with departure for other reasons than the one with which the investigation is concerned, is sometimes referred to as “missing observations”. Now, it is obvious that if the rates of death, rates of recovery etc. (with corrected denominators to achieve unbiasedness) are used and if the corresponding probabilities are high, there will be a loss of information due to the fact that the time which elapse to deaths and recoveries are wholly or partially disregarded. To overcome the difficulties of several causes of ...

DURING recent years, there have been substantial increases in the uses of vital records for research at the local, state, and national levels. Vital records are often used in retrospective or followback studies where the record serves as a focal point in sample surveys for collecting supplementary vital statistics. In these studies, a sample of records is selected from a frame of registered events, and the information available from the records is supplemented by information collected by surveys. The supplementary data are usually collected from one or more sources of information identified on the vital record itself. In the course of conducting several follow-back studies to collect supplementary mortality statistics during the past two or three years, the National Office of Vital Statistics (U. S. Department of Health. Education, and Welfare, PHS) has experimented with survey procedures for conducting these studies.' Some results pertaining to survey procedures and response rates are summarized in this report. The form and content of the official death records in use in the continental United States are determined by the 53 registration areas and therefore vary in certain details. However, these records conform in most essential respects to the standard certificate recommended by the U. S. Public Health Service. The standard death certificate identifies selected persons and institutions which can be utilized as sources of information in follow-back studies. These sources are (1) the funeral director, (2) the physician who certifies the death, (3) the informant who supplies personal particulars about the deceased person for the death record, and (4) the institution in which death occurred. The results presented in this report are based primarily on two followback studies anchored to death records. Response rates and related information pertaining to the physician, informant, and funeral director as sources of information are derived from the Pennsylvania Mortality Study. Information pertaining to the institution of death is based on the Illness Study of Deceased Persons. The Pennsylvania Mortality Study was undertaken collaboratively by the National Office of Vital Statistics and

TN considering statistics of disease we have become accustomed to include mortality statistics with morbidity statistics. Morbidity statistics are obtained by the compilation of data relating to cases of disease, and, as disease is present wherever there are cases, morbidity statistics relate to the prevalence and geographic distribution of disease and the conditions under which it occurs. Mortality statistics are the statistics of deaths. They are obtained by the collection and numerical classification of data relating to the frequency of deaths, their causes and the conditions under which they occur. Most deaths are caused by disease; and, in so far as they are thus caused, mortality statistics relate to those cases of disease which terminate life. To illustrate, in a city in which there had been 500 cases of typhoid fever, of which 45 had terminated fatally, morbidity statistics would treat of the facts relating to the 500 cases, while mortality statistics would include data relating to the 45 deaths.

The causes of death in England and Wales have been published in tabular form in the annual reports of the Registrar General since 1837 when Dr William Farr first applied statistical methods to the analysis of the English death roll. The classification of causes of death which Farr developed for this purpose (General Register Office, 1845) included the "shaking palsy", and deaths ascribed to this entity tabulated according to age and sex may be found in the annual reports for the years 1855 to 1880 inclu sive. Farr's successor, William Ogle, introduced in 1881 an improved classification based on the new nomenclature of disease of the Royal College of Physicians of London (Idem, 1881). In the new nosology specific medical terms largely replaced the common or popular names previously employed; "paralysis agitans" thus superseded the "shaking palsy". The Registrar General continued to publish the number of deaths assigned each year to this cause up to the year 1900. One notes with interest that Charcot (1880) and Sanders (1879) made reference to these reports in their clinical writings. As an unfortunate by-product of changes in the recording of mortality statistics coincident with the introduction of the first international statistical classification in 1901, the mortality ascribed to paralysis agitans was no longer tabulated separately. Publication of data for this disease was resumed in 1921 when the third revision of the International List was adopted. There is thus available, except for the 20-year gap 1901-20, a continuous record of mortality due to paralysis agitans in England and Wales from 1855 to the present representing the most extensive available mortality data on this disease.

Pulmonary heart failure, or cor pulmonale, is common in some parts of Britain, and Flint (1954) found that it accounted for a quarter of the cases of heart failure admitted to one Sheffield hospital (the City General Hospital). Yet, as was pointed out by Wood (1956), there are few comparative figures, and in the mortality statistics quoted by Southwood (1959) it receives scant mention. During 1954 to 1956 a study was made in hospitals of the Sheffield Hospital Region in order to compare the relative frequency of the various causes of heart failure in Sheffield and in other areas. A secondary purpose of the study was to compare the symptoms and signs, personal habits, occupation, and residence in the patients suffering from various forms of heart failure.

SynopsisThe paper, which was written at the invitation of the Council of the Faculty for submission on the occasion of the Centenary celebrations, deals with the broad trend of mortality rates in the hundred years 1850-1950.The paper is in five main parts :—1. The political and social background of the period, in so far as it might affect mortality rates, is briefly discussed.2. The trend of mortality rates during the hundred years is discussed principally with reference to the following six countries which, besides being relatively stable politically during the period, have recorded mortality rates throughout— England & WalesDenmarkFranceNetherlandsNorwaySweden.The experience in more recent years of a number of other countries is also discussed.3. The relationship between the mortality rates of males and females is examined at various ages for the six countries above mentioned.4. The more recent mortality statistics are analysed according to cause of death in broad groups.5. The paper concludes with a statement on the points which seem to be of special interest, setting out in the form of questions some of the thoughts which spring from them.In the pages which follow (Appendices 1-18) the statistics referred to in Parts 2 and 3 are illustrated by a series of graphs.

Summary 1 The history of appendicitis mortality prior to 1933 is briefly reviewed. 2 Mortality statistics for appendicitis in Australia from 1933 to 1957 are presented, both for the population as a whole, and for the various age groups. A remarkable reduction in mortality rates is in evidence. 3 Possible reasons for the improving prognosis are advanced. These possibilities ar.e examined in the light of two series of appendicitis cases from the Royal Adelaide Hospital (Series A from 1936-8, Series B from 1956-7). Antibiotics, intestinal decompression and intravenous infusions appear to have been the main factors contributing to the improved prognosis. This improvement has occurred principally in the group of perforative-appendicitis. 4 The graver prognosis of appendicitis in the aged is shown; and factors responsible for this are summarized. This older group remains a challenge for further improvement. 5 Post-operative complications in the Royal Adelaide Hospital Series are reviewed.

1. 1. The literature on injuries to the spine has been reviewed with especial reference to morbidity and mortality statistics both in military and civil practice. 2. 2. The pathologic physiology occurring in the spinal cord and in the urinary tract is described briefly. 3. 3. Infection of the urinary tract is found to be the most important cause of death in these injuries. Catheterization of the bladder is accepted as the cause for infection in the opinion of almost all commentators. 4. 4. The management of the urologic complications has been outlined, stress being laid upon the absolute necessity for the avoidance of catheterization. We are aware that some cases are not infected by catheterization, but we feel that this is a matter of luck rather than sound judgment. We feel that twenty years of clinical experience justifies our views on this subject. 5. 5. An analysis is made of 54 cases treated in Harlem Hospital without a single infection in the urinary tract.

ABSCESS OF THEBRAIN.DR.FRANCISC.GRANT. The extent of infection depends entirely on the ability of the surrounding tissue to wall it off and the result of the protective reaction is the formation of a capsule about the abscess. The two most important factors for a successful result are: the presence of adhesions obliterating the subarachnoid space and the degree of encapsulation about the infectious process. Dr. Grant considers the mortality statistics in his series discouraging, but he obtains satisfaction from the fact that a diagnosis of abscess of the brain was made in twenty-six of the thirty cases reported. The abscess was found and drained in the designated area in each of the twenty-six cases. The relatively high percentage of correct localization of the lesion of the brain was due mainly to careful neurologic investigation and exacting history taking. In twenty-seven of the thirty cases the point

Social concern for the children of mothers who work outside of their homes has had a long history. It began with the industrial revolution in England and Europe when the invention of machines outmoded home industries and forced not only men but their wives and children into factories and mills for their pittance living. The problems of the latter part of the eighteenth century and of the nineteenth century were the problems of the "poor working class." Mortality statistics indicated that the babies of factory women died in alarming numbers. There was evidence too of increasing juvenile crime developing in the overcrowded new industrial centers (15, p. 4'). Mixed motives instigated the zeal of philanthropists to do something about the deplorable state of infants and children, left without care while the mother worked long hours away from home. There were those who wished to keep the working class alive because industry needed them and

One of the questions water works men are frequently called upon to answer is "How does the hardness of a drinking water affect health?" The answer to this question may be sought by means of physiological and medical research or by a study of mortality statistics. The following paper deals with the latter line of inquiry and consists of a comparison of statistical data relative to mortality from certain selected diseases amongst populations supplied with "soft" water with similar data for populations supplied with "hard" water.

Two methods of classification of age-adjusted mortality statistics had indicated that in a band of counties on the high plains in eastern Colorado, the 1959–1961 death rate had persisted at the 1949–1951 level in contrast to the steep decline manifest in all other rural and urban regions of the state. For confirmation, prevalence of hypertension was studied among all 33,484 Colorado Selective Service System registrants born in 1939–1941, of whom 11,003 had been examined during 1957–1964. Hypertension was identified in 275 individuals, 8.2 per 1,000, and their geographic distribution agreed with the geographic mortality pattern. Thus, relative to the rest of the state, a high-risk hypertension region may have emerged in Colorado for reasons that are not yet clear. Since there is a geographic pattern, there is reason to suspect a changing environment.

1. The results obtained in a series of 21 cases of bleeding peptic ulcer treated with the Meulengracht regime are presented. These are compared with a control group of 72 cases treated with the older “star-vation-Sippy” method. 2. Mortality statistics of all cases treated with the Meulengracht regime, thus far reported in the literature, are presented. These are compared with similar statistics for the older methods of treatment. 3. The mortality rate in the Meulengracht series was 4.76 per cent. This is a decided improvement over the 11.1 per cent mortality in the control group. 4. The patients receiving early liberal feedings manifested a well-being not noted in those under the older method of treatment. 5. The average period of hospitalization was not decreased under the Meulengracht regime. 6. Two of our 21 cases (9.5 per cent) were further complicated by the occurrence of perforation, whereas this occurred in only one patient in our control series (1.4 per cent). The possible role played by the increased feedings is considered.

INTRODUCTION MORTALITY statistics of a flock of Single Comb White Leghorn chickens in which there was no culling or discarding prior to 300 days of age indicate that the female is more susceptible than the male to lymphomatosis (Burmester 1945). Marine and Rosen (1940) obtained a higher incidence of lymphomatosis in complete castrates than in incomplete castrates (slips) and suggested that the development of lymphomatosis was closely associated with the deficiency of the male hormone. Results of two experiments to be reported here confirm the findings that males are more resistant to lymphomatosis than capons and further show that the administration of the male sex hormone to males and capons increases their resistance to this disease. Experiment I tested the effect of early gonadectomy in the male and female among birds not inoculated and others inoculated with lymphomatotic material. In experiment II the effect of castration of males was again . . .

While studying the trend of Negro mortality from various causes, I have endeavored for the purpose of comparison to secure a large amount of statistical material from the same geographical area. This is especially important in studying the statistics of cancer. For some reason cancer rates vary under different climatic and other environmental influences, and the comparison of data over a series of years from a region of varying extent, such as the expanding U. S. Registration Area for Deaths, involves several sources of error. In order to avoid these and still have a large body of statistical material, I have compiled from the volumes on mortality statistics the data on cancer mortality in the Registration States of 1920, from 1920 to 1930. The mere compilation of the number of deaths in the white and colored population of this region is of course a very simple matter. The calculation of death rates, however, presents certain complications, and for the sake of the critical student of vital statistics an account of the procedure followed is given.

A review of mortality statistics has been made in cases of acute perforations of the stomach and duodenum, covering large groups in both this country and abroad. The question of extensive surgery, such as gastroenterostomy and pyloroplasty performed at the time of closing the perforation, was discussed, Report was made covering 36 cases of acute perforation with 5 deaths in the service of Dr. Stanton and myself.

The certifiers of 3193 U.S. deaths, aged 35–74, were queried by mail on the presence and severity of several chronic respiratory diseases (excluding tuberculosis, lung cancer and occupational pneumoconioses), in order to evaluate published vital statistics which indicate a marked increase since 1950 in the U.S. death rate from these diseases. The prevalence at death of the chronic respiratory diseases was found to be more than nine tunes greater than its underlying cause death rate. It was also found that U.S. death certifiers in 1963 listed on the death certificates only half of the severe chronic respiratory disease present at death, suggesting that vital statistics based on death certificates may seriously underestimate the contribution of chronic respiratory diseases to U.S. deaths. Such epidemiologic observations on the chronic respiratory diseases as their rapidly increasing U.S. mortality rate, their relatively small contribution to mortality in the U.S. when compared to the U.K., and their relatively high mortality rates in several states could be statistical artifacts arising from variation in habits of death certification.

In these days of continuous change in mortality, with the necessity of forecasting in many of our operations, a paper concerned with the graduation of mortality statistics may appear to be academic, the more so if attention is restricted to the fitting of mathematical curves, with or without any attempt to advance that elusive and, as some think, delusive conception, the law of mortality. Nevertheless, it is felt that if progress is to be made in any branch of science, theoretical research and experiment are essential, as also is the publication of the results, unless the work is completely abortive. It has become the fashion to speak of graduation as though its function were restricted to the provision of a practical instrument for monetary calculations; as Todhunter is reputed to have said, the tabulated values of lx are merely successive terms in a commutation column. Notwithstanding this, most of us retain, consciously or unconsciously, a feeling that, underlying all the roughnesses in our data referable to errors of observation and an ever-changing environment, there may be an inherent mathematical system of law and order, which if it could but be discovered would give such insight into the meaning of the unadjusted figures that a considerable advance would be made in the practical application of our science.

One of the most interesting questions in mortality statistics is that of hereditary influences, but very little light has hitherto been thrown on this subject. What effect will the good or bad health of the present generation have on that of the future one? In which way will nature eliminate the fatal germs of a family disposition? Will they disappear through a gradual dying out of the persons with such a disposition, or will they in the course of time grow less dangerous till at last they have lost entirely their original character? All these questions have hitherto been left almost unanswered in spite of the great scientific and economic interest attached to them.

An important part of the work of an actuary consists in making and testing graduations of mortality statistics. Much space has been devoted in the Journal to methods of graduation, but for many years, apart from the recent paper by H. L. Seal (J.I.A. Vol. lxxi, p. 5), little attention seems to have been paid to the testing of graduations. Although statistical methods are usually employed in making the tests, consideration has not often been given to ascertaining whether mortality statistics fulfil the conditions necessary for the tests to be strictly applicable. It is the purpose of this paper to set out some investigations into this question. Such matters may be thought to be of only theoretical interest, since, in practice, the tests work satisfactorily as a general rule; but actuaries should not employ their tools in a mechanical fashion without realizing fully their limitations and implications, and the question therefore takes on a practical as well as a theoretical aspect. It is hoped that this paper will show how far the tests of a graduation are satisfactory in theory and practice and indicate where modifications are required.

IN spite of the fact that mortality statistics rank malignancy as one of the leading causes of death, the incidence of multiple malignant tumors in the same individual is comparatively rare. Cases in which three morphologically dissimilar tumors occur are regarded as curiosities. The case herewith presented illustrates the occurrence of three such malignant tumors of the uterus. Billroth (1) is credited with the first report on multiple primary malignant growths in the same individual. He emphasized the importance of certain basic criteria before the diagnosis of such a condition could be established. His postulates are as follows: (1) each tumor must have an independent histologic appearance; (2) the tumors must arise in different situations, and (3) each tumor must produce its own metastasis. Mercanton added a fourth prerequisite—that there be no recurrence of the tumors following their removal. The demand for the fulfillment of the third postulate of Billroth, as well as that of Mercanton, is quite as ...

Ages at death of Common Egrets were computed from band recovery records in the files of the U. S. Fish and Wildlife Service, Laurel, Maryland. Most of the records represent egrets banded by Ben B. Coffey, Jr., near Glen Allan, Mississippi, between 1935 and 1949. All recoveries recorded up to December, 1960, were analyzed, but only birds banded before 1950 were used in computing mortality rates. Birds reported shot were excluded from the sample. The method used in the construction of life tables follows that pioneered by Lack (1943) which has been reviewed and expanded by Deevey (1947) and Hickey (1952). First-year mortality in birds should be carefully evaluated, for it is generally higher than the average adult mortality and can strongly influence population dynamics. There has been some difference of opinion as to the most suitable starting date for calculating first-year mortality. Lack (1949) used June 1 and Hickey (1952) used August 1 as the starting date in compiling mortality statistics of first-year herons. I have followed the pattern of Verheyen and LeGrelle (1952), Olsson (1958), and Owen (1959) in calculating first-year mortality after July 1. Several sources of potential bias in band recovery data have been discussed by Hickey (1952). Recognition of these makes the figures and interpretations presented here tentative.

n/a

Summary 1. Mortality statistics in thirty-three cases of acute staphylococcalempyema in infants and children are presented. 2. The disease is most common in the first year of life. 3. Age is the single most important factor influencing mortality.From birth to 4 months of age, the mortality was 66.7 per cent. From 4 months to 12 years of age, the mortality was 4.8 per cent. 4. Differences in the pathologic reactions and immune responseshave been demonstrated between the very young infant and the older age groups. 5. In the fatal cases in early infancy, progression of the underlyingstaphylococcal pneumonic process has appeared to be the cause of death. It is suggested that improvement in mortality rates in this age group probably must derive from a more effective therapy of the staphylococcal pneumonia.

The paper deals chiefly with those mortality statistics, compiled from the experience of insured persons, that have been used as a guide in life insurance underwriting. Properly interpreted, these statistics shed considerable light on the factors influencing human longevity. The use made of other mortality statistics in the underwriting of life insurance risks is also indicated. * A paper presented at the 107th Annual Meeting of the American Statistical Association, New York City, December 30, 1947.

An attempt has been made to demonstrate, using the admission and mortality statistics over three quinquennia from the one hospital, that the routine use of prompt massive intravenous thiamine is lifesaving in a significant number of patients suffering from alcoholic psychoses and that the former common cause of death was in fact an acute thiamine deficiency cardiomyopathy.

This study of rather a small group of cases of diabetics shows clearly enough, I think, that diabetics do not die early in life as a rule, but largely late in life when life is already on the descent; that they do not die because they have diabetes, but largely because of some other condition which can befall any non-diabetic; that the older they are at the time that diabetes develops, the better chance they have for a longer span of life. I wish to emphasize the need for improved mortality statistics.

IN the literature, the occurrence of cancer of the A thyroid gland has been studied at numerous teaching hospitals and large clinics by comparison, over a period, with the type of goiter seen clinically, with the number of thyroid operations, with the number of deaths or with the number of autopsies. Sokal,1 after estimating prevalence and incidence by combining clinical data, mortality statistics and autopsy statistics, presents his conclusions in a more tangible and vivid manner. He states that in a hypothetical city of 1,000,000 there would be 25 patients with thyroid cancer at any given time, that on the . . .

completeness of registration of 90 per cent or over were admitted to this area. Transcripts of the certificates were purchased by the United States Government from these states and cities. Yearly mortality statistics have been published from 1900 to date. A corresponding area for birth registration was started in 1915, admission to which was subject to the same provisions. Both registration areas were completed for the continental United States in

We have presented a series of cases, numbering 425, termed loosely and possibly incorrectly “toxemia of pregnancy.” No attempt is made to discuss the etiology. The classification used is standard. Our methods of management, past and present, are outlined, admittedly conservative, from the earliest prenatal period to the final stage of delivery and puerperium. Complications occurring in the mother and infant are presented and discussed; mortality statistics for both are given. From our follow-up, we have felt that recurrence is more frequent than formerly thought. It is to be stressed that damage to the cardiovascular renal system is often of far-reaching consequence, even in supposedly mild cases of toxemia of pregnancy.

SINCE 1901, the International List of Causes of Death has been used in the United States for classifying causes on certificates of death. Because many certificates include two or more causes of death, it became necessary to formulate rules to decide the single cause to which the death should be ascribed. Such rules were drawn up under the supervision of the Bureau of the Census, of which the National Office of Vital Statistics was then a part. Since 1913, these rules, published as the Joint Cause Manual, have been in general use throughout the country. Both the International List and the Joint Cause Manual have been revised every ten years and every revision has resulted in breaks in the comparability of mortality statistics. The sixth revision of the International List (1948) resulted in a new list known as the International Statistical Classification of Diseases, Injuries and Causes of Death, which differs radically in arrangement, and to some extent in inclusions, from the old classifications. At the same time, the rules of the Manual of Joint Causes were discarded and rules new to this country were adopted. Previous revisions of the International

Suspensions of cells in the bone marrow and in the spleen of normal mice were prepared. Two groups of lethally irradiated mice were given cells from bone marrow and two groups of lethally irradiated mice were given cells from the spleen of the same donor mouse. One of the groups injected with bone marrow and one injected with spleen cells were killed at 9 days for determination of CFU. The remaining animals were followed for 100 days to determine the mortality statistics. The mortality among mice not given cells was 100 per cent; among mice injected with cells, the mortality was reduced. The data permit a comparison of the reduction in mortality (protection) following injection of bone marrow CFU with protection following injection of spleen CFU. Protection with CFU of bone marrow was different from protection with CFU of spleen. It seems possible that both hemopoietic pluripotent "stem cells" and "early differentiated progenitors" can form spleen colonies and that the ratio of these CFU may be different for bone marrow and spleen.

CHICK mortality similar to that described by Temperton and Bythell (1944) and termed “six day chick disease” has been prevalent for several years among the chickens of certain inbred lines maintained at the Regional Poultry Research Laboratory. Temperton and Bythell state that “commencing on the sixth day, heavy losses occurred and continued until the group was 12 to 14 days old with a peak in mortality on the eighth, ninth and tenth days. The manner of death was characteristic and not symptomatic of any known pathological condition.” This statement describes satisfactorily the situation observed among the chickens in the present report. Analysis of mortality statistics from hatching to 21 days of age for a four-year period (1945–1948) indicates that excessive mortality occurred in the chickens of certain lines, while the losses in other lines were low. Late in the fall of 1948, chick mortality in some lines became alarmingly . . .