Models for temporal variation in cancer rates. II: Age-period-cohort models

Trends in lung cancer, chronic obstructive lung disease, and emphysema death rates for England and Wales 1941-85 and their relation to trends in cigarette smoking

Trends in smoking associated respiratory diseases in England and Wales during 1941-85 have been studied, with careful attention to problems caused by changes in classification of cause of death. Three diseases were selected for analysis: lung cancer, emphysema, and chronic obstructive lung disease. During 1971-85 deaths that would previously have been certified under chronic bronchitis have increasingly tended to be classified under chronic airways obstruction. The definition of chronic obstructive lung disease that was used includes both terms to avoid the artificial decline caused by consideration of chronic bronchitis in isolation. Age specific rates for all three diseases show a pronounced cohort (period of birth) pattern, rates for men rising up to the rates for those born shortly after the turn of the century and then declining, and rates for women peaking in the cohort born 20-25 years later. For chronic obstructive lung disease, but not for lung cancer and emphysema, the cohort peak is superimposed on a sharply declining downward trend. In both sexes cohort patterns of cumulative cigarette consumption peak at a time broadly similar to those seen for the three diseases. Trends in cigarette consumption, however, cannot explain the underlying steeply declining rate of chronic obstructive lung disease. Nor can they fully explain the declining trends in lung cancer and emphysema rates in younger men and women.

Analysis of cohort mortality from prostatic cancer in Spain, 1951-1983

Prostatic cancer mortality rates in Spain were analysed for the period 1951-1983 by age groups. Five year interval age cohort trends were also studied. A rising trend was seen over this period of time, although there has been some stabilisation during the recent past. Age cohort analysis shows a generation effect in those cohorts born before 1896, consistent with an increased exposure to environmental and/or occupational factors.

Suicide in England and Wales 1946-1985: an age-period-cohort analysis

Four decades of suicide mortality data for England and Wales were analysed within an age-period-cohort framework. Marked differences in results were found by sex. The analysis of the male data showed that the risk of suicide increased with age until men were in their mid-fifties (after allowing for period and cohort effects), and that suicide risk declined for males born from 1876 to 1915, but had increased over 4-fold for certain later-born cohorts. Female suicide risk was found to decline with later nativity (after adjusting for age and period effects). The identification of period effects in terms of known changes in the carbon monoxide (CO) content of domestic gas successfully accounted for temporal variation in male suicide risk, but fitted less well to the data for females.

Asthma deaths in England and Wales 1931-85: evidence for a true increase in asthma mortality

A recent increase in asthma mortality has been reported in several countries. Such increases can be brought about by changes in factors acting close to the time of death, but they may also be caused by risk factors determined by the generation into which a person is born, as indicated by the year of birth. Asthma mortality rates since 1931 are independently associated with birthdate as well as date of death. In particular there has been an increase in asthma mortality in birth cohorts born since the 1940s. Such changes are unlikely to be due to a change in reporting of asthma deaths, and other evidence including the reported increase in the prevalence of eczema in succeeding National Birth cohorts supports the view that these changes may be due to an increased prevalence of atopy. In the absence of any further improvements in the management of asthma, such an increased prevalence of atopy implies that the mortality rate is unlikely to decline substantially for some years to come.

Trends in cutaneous malignant melanoma in Denmark 1943-1982 by anatomic site

Time trends in the incidence of cutaneous malignant melanoma were examined by analysing data from the Danish Cancer Registry, 1943-1982, by sex and subsite for 3509 male and 5305 female cases. The age-standardized incidence rate for cutaneous malignant melanoma for men and women in Denmark increased by five- to six-fold between 1943 and 1982. The increase varied by site; a particularly pronounced elevation was seen for tumours of the trunk in both men and women and of the leg in women. A statistical analysis of the effects of age, time and cohort showed steep increases in risk for progressively younger birth cohorts. This association was particular pronounced for sites other than the face, scalp and neck; for the latter sites, there was a much smaller increase for younger cohorts.

Trends in breast cancer incidence and mortality in Denmark, 1943-1982

The trend in breast cancer incidence and mortality was examined using data from the Danish Cancer Registry and the national mortality statistics respectively. The study population comprised 65,870 incident cases and 33,817 deaths from breast cancer in Denmark between 1943 and 1982. The incidence rate remained almost constant up to around 1960, whereafter it rose steadily. Comparatively little change was observed in mortality. Possible explanations for the differing trends in incidence and mortality include under-reporting of breast cancer from death certificates and improvement in survival. A multivariate statistical analysis showed that the increase in incidence was due mostly to a cohort effect, though calendar time may have had a slight effect as well.

Age, period and cohort analysis of trends in mortality from major diseases in Japan, 1955 to 1979: peculiarity of the cohort born in the early Showa Era

Japanese researchers have reported that recent mortality rates from diabetes mellitus, ischemic heart disease, peptic ulcer, cirrhosis of the liver and suicide for middle-aged Japanese men have increased by comparison with those for other age groups. There has been some controversy over the etiology of this unusual trend, and in particular whether it is due primarily to recent undesirable socio-economic factors (period effects) or to factors specific to these cohorts born in the early Showa Era, around 1925 to 1940 (cohort effects). A possible source of this controversy lies in the methods which have been used to describe the trends; these are mostly descriptive and graphical. To elucidate which factors are responsible for these trends, we analysed the mortality data quantitatively applying an age-period-cohort model modified so that period effects remain constant within certain age groups but may vary from one age group to the next. Although the identifiability problem still occurs in the modified model, estimable curvature components of time effects may be used to examine these unusual trends. In fact, the peculiarity of the cohort born in the early Showa Era was clearly detected by the curvature components of cohort effects for these major diseases. These findings are consistent with the 'cohort hypothesis' for the recent peculiar trend in Japanese male mortality.

Models for temporal variation in cancer rates. I: Age-period and age-cohort models

A main concern of descriptive epidemiologists is the presentation and interpretation of temporal variations in cancer rates. In its simplest form, this problem is that of the analysis of a set of rates arranged in a two-way table by age group and calendar period. We review the modern approach to the analysis of such data which justifies traditional methods of age standardization in terms of the multiplicative risk model. We discuss the use of this model when the temporal variations are due to purely secular (period) influences and when they are attributable to generational (cohort) influences. Finally we demonstrate the serious difficulties which attend the interpretation of regular trends. The methods described are illustrated by examples for incidence rates of bladder cancer in Birmingham, U.K., mortality from bladder cancer in Italy, and mortality from lung cancer in Belgium.