Several methods have been proposed for standardizing risks, risk ratios, and risk differences based on the results of logistic regression. These methods provide an alternative to direct standardization, a particularly useful approach when there are many covariates. In this paper, methods for calculating approximate confidence limits for these standardized measures are presented. A simple example, in which published data are used, illustrates the techniques and allows comparison with confidence limits calculated from the directly standardized risk ratio.
Clinical trials are commonly performed in surgery to assess the efficacy of one or more treatments. Many therapies result in only partial or temporary improvement, rather than cure. Others sharply affect the quality of patients' lives or of their deaths. For most interventions, it is important to document effects on quality of life as well as morbidity and mortality rates. Yet, a review of the literature reveals that very few surgical trials consider quality of life variables as outcome measures. Surgical investigators in areas like cancer, inflammatory bowel disease, end stage renal disease, and cardiac disease have examined quality of life issues extensively using a variety of scales and indices. However, most studies on quality of life are hampered by poor design and inadequate methods of assessment. Failure to evaluate quality of life variables prevents the recognition and full use of potentially beneficial therapies and the rejection of potentially harmful ones.
Observations of cancer risk in irradiated human populations over time after exposure suggest that there are at least two, and perhaps more, very different patterns of temporal distribution of risk for radiation-induced cancer. The first, exemplified by bone sarcoma following therapeutic injection of 224Ra and chronic granulocytic leukemia in Japanese A-bomb survivors, is an early, wave-like pulse consisting of an increase in risk followed by a gradual decline back to baseline levels. The second, exemplified by breast cancer following a brief exposure to external gamma ray or X ray, and by lung cancer and stomach cancer in A-bomb survivors, is an increase in relative risk over about 10 years to a value which appears to remain constant over time thereafter. The first pattern suggests that tumor growth kinetics may play a central role in the temporal distribution of risk following exposure, while the second seems more consistent with multi-event models for carcinogenesis, in which radiation or some other cause of early events must be followed by one or more later events whose frequencies depend mainly on attained age. There are, however, other data that appear to conform to neither of the two models just mentioned. Influences of other cancer causes, like tobacco smoking, are potentially serious confounding factors in studies of induction period.
The accuracy of routinely collected mortality data for ischemic heart disease (IHD) as indicators of death from acute myocardial infarction (AMI) was assessed in ages 25-64 years, according to the WHO criteria defined in 1983. Cases were identified from computer records (linked for individuals) of all death certificates and hospital discharges in Western Australia between 1971 and 1982. Where the official cause was IHD about 90% of deaths fulfilled the WHO criteria for definite or possible AMI. Up to 10% of fatal cases of definite or possible AMI were coded to other causes in the official death statistics, however it appeared that variations in this figure with changes in coding practices could cause appreciable bias in the estimation of secular trends in IHD mortality. This problem could largely be overcome by reviewing fatal events where the death certificate was coded to one of a limited number of other ICD rubrics.
This paper discusses the advantages and disadvantages of using administrative data for longitudinal research, focusing on loss to follow-up. Comparisons between research relying on primary data collection and that using data bases are made. After development of a suitable framework, follow-up in several well-known projects based on primary data collection (the Seven Countries project on coronary heart disease, the Massachusetts research on long-term care and the Pittsburgh clinical trial of tonsillectomy) is compared with follow-up using the Health Services Commission data base in Manitoba, Canada. Overall follow-up in the Manitoba research compares favorably with participation and follow-up rates in other studies based on primary data collection. Initial nonresponse and nonlocation are major problems with studies using primary data; failure to locate earlier respondents in subsequent waves results in a wide range of overall response rates. Data bases do not require researchers to contact individuals and hence follow-up is simplified. Eight year follow-up rates in the Manitoba data base are almost always over 80% and often over 90%. Because records can be flexibly summarized for each individual over time, data bases facilitate certain types of longitudinal studies which would be difficult, if not impossible, to perform using other methodologies. If the desired data are available and recorded with acceptable accuracy, administrative data banks hold considerable promise for the health care researcher.
In summary, from the epidemiologic perspective, the conference made it clear that: In modern occurrence research in medicine, quality of life is an important aspect of health outcome, along with duration of life; and it is of interest as a determinant of outcome as well. The concept of quality of life remains ill-defined but is, in the medical sense, some aggregate representation of disability discomfort, and distress resulting from illness and/or medical action. In medicine, quality of life is, in concept, specific to various types of clinical situation; and empirical scales need to be similarly situation-specific. Whereas clinical research on quality of life is already being conducted successfully, further development of scales specific to particular types of clinical situation remains a challenge to theorists of quality of life, as well as to experts on clinimetrics who collaborate with clinical specialists.
The Primary Preventive Trial in Göteborg, Sweden, a study of a random population sample of middle-aged men, made it possible to analyse the risk factor pattern cross-sectionally in 166 men with uncomplicated angina pectoris (AP) and compare with 5735 men without angina pectoris or myocardial infarction (MI). A prospective analysis was also performed concerning the risk factor pattern in 128 cases with uncomplicated AP and 34 cases with complicated AP (following an MI) respectively, appearing during a follow-up time of 4 years. At cross-sectional analysis, uncomplicated AP was related to elevated serum cholesterol, elevated systolic and diastolic blood pressure, increased relative body weight, smoking, diabetes mellitus, low physical activity during leisure time, dyspnea and mental stress. However at multivariate, prospective analysis only dyspnea, stress, diabetes mellitus and increased relative body weight were predictors for uncomplicated AP. In contrast, elevated serum cholesterol, high blood pressure, smoking, and high physical activity at work were predictors for complicated AP. Possible reasons for the apparent risk factor differences and different mechanisms in AP and MI are discussed.
Adherence to standards for judging the content validity of health measures and for labeling them is needed for the field of health assessment to proceed in an orderly fashion. This paper discusses the dimensionality of health and the range of health states that can be measured within each dimension. These two attributes of published definitions of health are used to derive minimum standards for judging the validity of health measures in terms of their content. Five generic health concepts are defined: physical health, mental health, social functioning, role functioning, and general health perceptions. Items from widely used health measures are presented to clarify distinctions among these concepts and the different health states they encompass. It is recommended that labels be assigned to health measures in a manner consistent with their content and other evidence of validity.
Type A behavior was assessed in Rotterdam in 3171 males, aged 45-59 years, by the Jenkins Activity Survey (JAS) as part of the Kaunas-Rotterdam Intervention Study (KRIS). During a follow-up period of 9 1/2 years, 112 fatal and 157 non-fatal cases of myocardial infarction occurred. The JAS did not predict future cases of fatal or non-fatal myocardial infarction. Persons scoring highest on the hard-driving scale or the Dutch adaptation of the JAS tended to have higher incidences of angina pectoris. However, overall the validity of this test as a predictor of CHD was not substantiated in this population.
This paper analyzes the mortality of Mexican Americans and Puerto Ricans residing in the Chicago metropolitan area. In a comparison of these two groups with nonHispanic whites in the same area, it was found that Hispanic mortality is unusually low for the two leading causes of death in the mainstream U.S. population. Explanations discussed include underdiagnosis, underreporting of illness, the "healthy migrant" thesis and possible return migration of elderly ill.
A twenty year debate about the appropriate surgery for breast cancer has resulted in two clinical trials comparing radical vs more conservative operations. Despite the favorable results of these trials, the majority of breast cancer patients in North America still undergo mastectomy. We investigated the psychological and social adjustment following total and partial mastectomy in a group of patients randomly assigned to one or the other operation (National Surgical Adjuvant Breast Protocol—B-06). Total mastectomy patients showed higher levels of depression and less satisfaction with body image. Partial mastectomy patients did not display any measurable increase in fear of recurrence. Patients undergoing radiation therapy showed surprising increase in depressive symptoms. Radiation therapy could well be more frightening to breast surgery patients than had been anticipated. These patients may benefit from some anticipatory counselling.
Because blacks make up a small proportion of the hip fracture population, little is known about how blacks' experience following hip fracture compares to that of whites. This study, a retrospective review of the medical records of 119 community residing subjects, 60 years of age and older, 37% of whom were black, admitted with a diagnosis of hip fracture to a large urban teaching hospital, investigated differences between black and white patients in factors associated with outcome following hip fracture and outcomes at time of hospital discharge. Blacks were significantly more likely than whites to exhibit a high total number of diagnoses, urinary incontinence following surgery, low admission hemoglobins and mental impairment. Blacks experienced significantly longer hospitalizations than whites, were significantly more likely to be nonambulatory at discharge and showed different patterns of discharge destination. Multiple regression and logistic regression indicated that the greater amount of illness of blacks is a consistent significant predictor of these differences in outcomes, but that race, delays to surgery and non-surgical treatment also make independent significant contributions. These findings indicate the importance of planning for the in-hospital care of black hip fracture patients, and the examination of the financial consequences of DRG's for hospitals serving black or poor populations.
A validation study was carried out on self-reported smoking for 1177 people in Sydney and Melbourne in 1983. Because of its long half life and the fact that smoking is its only source in body fluids, saliva cotinine was chosen as the validation measure. Cotinine levels above 250 nmol/1 were used to classify people as smokers. The sensitivity of self-reported smoking was 92.6% and the specificity was 93.4%. There was some evidence that people in the process of changing their smoking status might be slow in updating their self-classification. The smoking prevalence estimate based on cotinine levels was found to be 1.7% lower than that for self-reported smoking status. The small proportion of false negatives and false positives suggests that commercially collected data banks can be valid sources of prevalence data. Correlation between cotinine level and reported cigarette consumption was not affected by sample volume, and was similar to that achieved for carbon monoxide and thiocyanate at a low 0.34. Regression analysis using self-reported cigarette consumption filter/non-filter cigarettes, and time since last cigarette as predictors, explained 13.6% of the variance in cotinine level.
Clinical features of different types of stroke were investigated in a sample of 409 patients representative of all cases admitted for acute stroke, except subarachnoidal hemorrhages, within a well defined population. A specific cerebrovascular diagnosis was obtained by detailed clinical investigation, including CT scan. In people > 50 years old, men/women risk for stroke was estimated to be 1.40:1. The risk was higher in men up to the age of 80; above this age similar risk for the two genders was observed. Eleven per cent had intracerebral hemorrhage, 13% TIA, 51% non-embolic and 25% embolic brain infarction. In all diagnostic categories there were similar proportions of patients who had a history of hypertension and previous stroke, neither did hemoglobin and hematocrit levels differ between the different stroke disorders. TIA preceded intracerebral hemorrhage in 11% and brain infarction in 15–20%. As opposed to patients with ischemic lesions, subjects with intracerebral hemorrhage had higher systolic blood pressure levels and more severe symptoms on admission to hospital. Ischemic stroke was associated with male predominance, different ischemic manifestations of heart diseases and diabetes.
The most central epidemiological feature of cancers of the breast, ovary and endometrium is the enormous protective effect of menopause. Simple mathematical “models” of the age-incidence of these tumours suggests that their incidence by age 70 would be between 4 and 8-fold increased if women continued to menstruate into old age. Using the notion of “effective cell-cycle time” to explain this phenomenon, different “models” are constructed for each of these three tumour sites and shown to provide an excellent description of most of the known epidemiology of each of the cancers, and to provide a basis for predicting the long-term effects of various factors on the risk of these diseases. In particular, the model suggests that 5 years of combination-type oral contraceptive use will approximately halve the lifelong risk of both ovarian cancer and endometrial cancer.
