Biomedical subjects

M W Knuiman

Publications and source records attributed to M W Knuiman.

At least 37 records · Page 2Linked to original sources

PubMed1998

Adjustment for regression dilution in epidemiological regression analyses.

PURPOSE: The term "regression dilution" describes the dilution/attenuation in a regression coefficient that occurs when a single measured value of a covariate is used instead of the usual or average value over a period of time. This paper reviews the current knowledge concerning a simple method of adjusting for regression dilution in single and multiple covariate situations and illustrates the adjustment procedure. METHODS: Formulation of the regression dilution problem as a measurement error problem allows existing measurement error theory to be applied to developing methods of adjustment for regression dilution. This theory leads to a precise method of adjustment for linear regression and approximate methods for logistic and Cox proportional hazards regression. The method involves obtaining the naive estimates of coefficients by assuming that covariates are not measured with error, and then adjusting these coefficients using reliability estimates for the covariates. Methods for estimating the reliability of covariates from the reliability and main study data and a method for the calculation of standard errors and confidence intervals for adjusted coefficients are described. RESULTS: An illustration involving logistic regression analysis of risk factors for death from cardiovascular disease based on cohort and reliability data from the Busselton Health Study shows that the different methods for estimating the adjustment factors give very similar adjusted estimates of coefficients, that univariate adjustment procedures may lead to inappropriate adjustments in multiple covariate situations, whether or not other covariates have intra-individual variation, and when the reliability study is moderate to large, the precision of the estimates of reliability coefficients has little impact on the standard errors of adjusted regression coefficients. CONCLUSIONS: The simple method of adjusting regression coefficients for "regression dilution" that arises out of measurement error theory is applicable to many epidemiological settings and is easily implemented. The choice of method to estimate the reliability coefficient has little impact on the results. The practice of applying univariate adjustments in multiple covariate situations is not recommended.

Analysis of Variance↗

PubMed1998

Multivariate risk estimation for coronary heart disease: the Busselton Health Study.

Coronary heart disease (CHD) is a multifactorial disease and CHD risk should be estimated by assessing all cardiovascular risk factors simultaneously. Simply adding up the number of factors with 'at risk' values fails to identify high-risk subjects with multiple risk factors at moderately elevated values. A more efficient approach is to use a quantitative multivariate risk score. A number of overseas studies have produced CHD risk scoring systems for men. There are few risk scores developed for women and no CHD risk scores have been developed from Australian data. This study used data on CHD risk factors and morbidity/mortality follow-up for the 1978 Busselton Health Survey participants to provide age-specific estimates of absolute risk of CHD hospitalisation or death, and to develop multivariate CHD risk scoring systems for men and women. The scores are based on age, blood pressure, anti-hypertensive medication, total and HDL cholesterol, smoking, diabetes, left ventricular hypertrophy and previous history of CHD. The generalisability and applicability of these risk estimation systems to Australian populations in the late 1990s is discussed.

Adult↗

PubMed1998

Longitudinal analysis of the effect of smoking cessation on cardiovascular risk factors in a community sample: the Busselton Study.

BACKGROUND: Cessation of smoking is known to have health benefits, but individuals who give up smoking can also suffer adverse effects. This study investigated the effects of smoking cessation on cardiovascular risk factors in a community-based sample. METHODS: Longitudinal risk factor data were analysed in 1372 women and 888 men who took part in population-based mass health screenings in the rural town of Busselton, Western Australia. Age-adjusted changes in body weight, forced expiratory volume in 1 s (FEV1), forced vital capacity, blood pressure and total cholesterol over 3-year and 6-year follow-up periods were compared for 235 ex-smokers ('quitters'), 1499 never-smokers and 526 continuing smokers. RESULTS: In women, weight gain was larger in 'quitters' than in continuing smokers after 3 years (1.74 versus 0.32 kg, P = 0.015) and after 6 years (2.39 versus 1.24 kg, P = 0.085). Male 'quitters' had gained significantly more weight after both 3 years (2.84 versus 0.93 kg, P < 0.001) and 6 years (4.46 versus 2.40 kg, P < 0.001). Although FEV1 declined less in 'quitters' than in continuing smokers, this was statistically significant only for men and women aged less than 45 years. Smoking cessation was associated with smaller increases in cholesterol level in women, but not in men. Smoking cessation had no significant effect on changes in forced vital capacity or blood pressure. CONCLUSIONS: Smoking cessation has different consequences for cardiovascular risk factors in men and women. Although giving up smoking is associated with weight gain in both men and women, the magnitude of the gain is larger in men. Smoking cessation has beneficial effects on lung function, especially in younger people, and on lipid levels in women, but this requires further study. Smoking cessation does not appear to influence blood pressure.

Adult↗

PubMed1997

Prediction of coronary heart disease mortality in Busselton, Western Australia: an evaluation of the Framingham, national health epidemiologic follow up study, and WHO ERICA risk scores.

STUDY OBJECTIVES: To evaluate the performance of the Framingham, national health epidemiologic follow up study, and the WHO ERICA risk scores in predicting death from coronary heart disease (CHD) in an Australian population. DESIGN: Cohort follow up study. SETTING AND PARTICIPANTS: The cohort consisted of 1923 men and 1968 women who participated in health surveys in the town of Busselton in Western Australia over the period 1966-81. Baseline assessment included cardiovascular risk factor measurement. Mortality follow up to 31 December 1994 was used. MAIN RESULTS: Risk scores for death from CHD within 10 years based on age, systolic blood pressure, cholesterol, smoking, and BMI were derived from the Busselton study data using logistic regression analysis. Similar risk scores developed from the Framingham, the national health epidemiologic follow up study, and the WHO ERICA cohorts were found to perform just as well in Busselton as the Busselton-derived scores, both before and after controlling the effect of age. There was considerable overlap across the different risk scores in the identification of individuals in the highest quintile of risk. Those in the top 20% of scores included about 41% of deaths from CHD among men and about 63% of deaths from CHD among women. CONCLUSION: Although there is variation in risk score coefficients across the studies, the relative risk predictive performance of the scores is similar. The use of Framingham and other similar risk scores will not be misleading in white Australian populations.

Adult↗

PubMed1997

Influence of lifestyle, coping, and job stress on blood pressure in men and women.

We designed this study to clarify the role of work stress on long-term blood pressure control and in particular to investigate whether perceived work stress directly affected resting blood pressure levels or whether there were indirect effects mediated by coping mechanisms and lifestyle. Men (n = 337) and women (n = 317) working in a government tax office completed questionnaires for assessment of work-related stress, coping strategies, and lifestyle. Seven resting blood pressure measurements were recorded serially on each of two occasions a week apart. Men had higher blood pressures (119.6/68.6 versus 110.9/65.6 mm Hg) than women; they used more "maladaptive" coping strategies, drank more alcohol, and ate less healthily but exercised more than women. There were no direct associations between measures of work stress and blood pressure. In univariate and regression analyses, both body mass index and lifestyle factors in the form of alcohol consumption, exercise, and diet were related to blood pressure in men and women. Various "adaptive" or "maladaptive" coping mechanisms were identified and independently related to both job stress and blood pressure levels. Women were more likely to use "healthier" or adaptive coping mechanisms than men. Thus, work stress per se had no direct effect on blood pressure, but the ways that individuals reported coping with stress were significantly related to blood pressure, with blood pressure elevation effects appearing to be mediated largely by dietary and drinking habits and physical inactivity. The results point to the need to target individual coping strategies and lifestyle as much as the working environment in workplace cardiovascular health promotion programs.

Adaptation, Psychological↗

PubMed1997

An empirical comparison of multivariable methods for estimating risk of death from coronary heart disease.

BACKGROUND: Logistic regression and, more recently, Cox regression have been the predominant methods for identifying risk factors and developing risk estimation equations for coronary heart disease (CHD). Software for the regression tree method is now available for binary and survival outcomes and thus offers an alternative methodology. This paper compares these four methods for identifying significant risk factors from among a set of candidate factors and for estimating the risk of death from CHD using baseline and mortality follow-up data on 1,701 men participating in the Busselton Health Study. The candidate risk factors were age, body mass index, systolic and diastolic blood pressure, treatment for hypertension, cholesterol and smoking. METHODS: Logistic regression, Cox proportional hazards regression, binary regression tree, and survival regression tree analyses have been applied to data obtained from the same cohort of men for CHD death risk estimation and prediction. The four methods are compared in terms of the variables selected, goodness-of-fit of models, similarity of cross-validated estimated risks for individuals, and ability to discriminate between those who died from CHD and those who did not die from CHD during the follow-up period, including the comparison of Receiver Operating Characteristic (ROC) curves. RESULTS: Although age and a blood pressure variable were selected by all four methods, body mass index was also selected by the regression tree methods and smoking was also selected by Cox regression. There was good, but not excellent, agreement between methods in estimates of risk for individuals, the areas under the ROC curves were 0.66 for the binary tree, 0.72 for logistic regression, 0.71 for the survival tree method and 0.78 for Cox regression. The average differences in estimated risk between those who died from CHD and those who did not die from CHD during the follow-up period were 0.051 for logistic regression, 0.070 for the binary tree method, 0.073 for the survival tree method and 0.088 for Cox regression. CONCLUSION: For a moderately sized cohort typical of many applications of these methods in the literature, the two methods which used the survival outcome performed better than the methods using a binary outcome. Despite selecting some different variables and showing moderate differences in risk estimates for individuals, the two binary approaches were similar in performance. Cox regression appeared to be superior to the survival tree method, but further larger studies of completely separate samples for model development and evaluation of prediction performance are required to confirm this finding.

Age Factors↗

PubMed1996

Spouse correlations in cardiovascular risk factors and the effect of marriage duration.

Spouse correlations in cardiovascular risk factors were investigated using data on 2,836 spouse pairs collected in the Busselton Population Health Surveys over the period 1966-1981. The risk factors considered were systolic blood pressure, diastolic blood pressure, body mass index, triceps fatfold, cholesterol, and forced expiratory volume (1 second). Statistically significant positive correlations (p < 0.01) were found for all (age-adjusted) variables. There was a statistically significant decreasing trend in the correlations for systolic blood pressure with marriage duration (trend p < 0.01). Although no other variables showed statistically significant trends, the correlations for diastolic blood pressure (p = 0.29), body mass index (p = 0.14), and forced expiratory volume (p = 0.16) also decreased with marriage duration, and correlations for cholesterol (p = 0.61) and triceps fatfold (p = 0.99) increased with marriage duration. These results suggest that there is spousal concordance in cardiovascular risk factors. The lack of consistent increasing trends in the correlations with marriage duration suggests that assortative mating may be a more likely explanation than the sharing of a common environment.

Adult↗

PubMed1996

An evaluation of road crash injury severity measures.

Reliable and consistent measures of injury severity are necessary for the study of environmental, crash and personal factors involved in road traffic crashes. This study was designed to evaluate measures of injury severity derived from computerized hospital discharge records, using 3609 road crash casualties admitted to hospital in Western Australia in 1988. External cause of injury codes were used to identify injuries from road traffic crashes. The ICDMAP software was used to convert the diagnosis codes into Abbreviated Injury Scale (AIS) scores for each body region. The maximum AIS values were derived using (1) all 12 diagnosis codes; (2) the first six diagnosis codes; and (3) the principal diagnosis code alone. Other measures of injury severity evaluated were the number of body regions with at least one injury; the number of regions with Abbreviated Injury Severity score of three or more; and total number of days spent in hospital. Discriminant analysis suggested that the AIS could be separated into minor and major injuries at a score of three and the Injury Severity Score at a score of nine. The measures derived from the AIS were all strongly correlated with each other and with the length of hospital stay and the dichotomized values gave similar results to the other scores when used in regression analyses of the injury experience of different types of road users. It was concluded that measures incorporating elements of both severity and number of injuries were preferable but length of hospital stay would be a suitable proxy if no other injury information was available.

Accidents, Traffic↗

PubMed1996

Familial correlations, cohabitation effects, and heritability for cardiovascular risk factors.

Familial correlations in cardiovascular risk factors were investigated with use of data from a community-based sample of 1319 nuclear families involving 4178 adult persons collected in the Busselton Population Health Surveys over the period 1966 to 1981. The risk factors considered were systolic blood pressure, diastolic blood pressure, body mass index, triceps fatfold, and cholesterol. All risk factors showed positive familial correlations, with correlations generally being lower for spouses than for parent-offspring pairs or for siblings. Spouse correlations showed little variation with age, suggesting that observed correlations are primarily due to assortative mating and not to cohabitation. The parent-offspring correlations tended to decline with age of (adult) offspring; this observation suggests that the effect of a shared household environment during childhood and adolescence diminishes over time when living apart during adulthood. The sibling correlations decreased with age for blood pressure and serum cholesterol and increased with age for body mass index and triceps fatfold. The estimated heritabilities were 27% for systolic and diastolic blood pressure, 37% for serum cholesterol, 52% for body mass index, and 23% for triceps fatfold. These results confirm that substantial familial aggregation of cardiovascular risk factors occurs and that much of this aggregation has a genetic basis, although assortative mating (in spouses) and environmental influences (in offspring and siblings) are also present. The nuclear family should be considered as a point of intervention in cardiovascular disease prevention programs.

Adult↗

PubMed1996

Microalbuminuria in type 2 diabetes: an independent predictor of cardiovascular mortality.

BACKGROUND: Microalbuminuria has been shown to be associated with cardiovascular mortality in type 2 diabetic subjects. It is unclear to what extent this is due to the increased prevalence of other cardiac risk factors. AIMS: To examine the relationship of urine albumin excretion to cardiovascular mortality and to determine its status as an independent risk factor. METHODS: In a prospective longitudinal study from 1986-1993 we followed 666 type 2 diabetic subjects from a diabetes outpatient service. Cardiovascular risk factors including urine albumin concentration were measured at study entry. Cox proportional hazards regression was used to determine risk factors for mortality. The hazard ratios of microalbuminuria and macroalbuminuria for all cause, cardiovascular and coronary heart disease mortality were determined after accounting for other cardiac risk factors including blood pressure, glycated haemoglobin, total cholesterol, HDL cholesterol, triglycerides, urea, smoking, body mass index, patient age and disease duration. RESULTS: The prevalence of urine albumin of 30-300 mg/L at study entry was 31.7%. A total of 167 deaths occurred (80 from cardiovascular disease). Mortality hazard ratios in subjects with urine albumin of 30-300 mg/L as compared to < 30 mg/L, adjusted for age, sex and other cardiovascular risk factors were 1.77 (95% CI 1.22-2.57, p = 0.002) for all causes, 2.34 (95% CI 1.38-3.99, p = 0.002) for cardiovascular and 1.78 (95% CI 0.97-3.26, p = 0.061) for coronary heart disease (CHD) mortality. Other factors significantly associated with cardiovascular mortality included diastolic blood pressure, HDL cholesterol and glycated haemoglobin. Total cholesterol and log triglyceride were significantly associated with CHD mortality. Disease duration, age at diagnosis, smoking and body mass index were not related to cardiovascular or CHD mortality. CONCLUSIONS: We confirm microalbuminuria as an independent predictor of mortality in type 2 diabetes despite its association with a number of conventional cardiovascular risk factors.

Albuminuria↗

PubMed1996

Self-reported health and use of health services: a comparison of diabetic and nondiabetic persons from a national sample.

Population-based epidemiological and health service utilisation information on diabetes and other noncommunicable diseases is still scarce in Australia. Such information is needed by health economists, policy makers and service providers. Data from the 1989-90 National Health Survey conducted by the Australian Bureau of Statistics have been used to obtain estimates of the prevalence of cardiovascular morbidity, life-style factors, use of hospital and medical services, and self-assessed health and happiness for Australian persons with diabetes. Prevalences are compared with those for persons without diabetes. Those with diabetes had two to three times the prevalence of most cardiovascular conditions, similar levels of exercise (except for diabetic women over 40 years of age who exercised less than their nondiabetic counterparts), lower levels of alcohol consumption (except for younger men, who had a similar frequency of heavy drinking as their nondiabetic peers), similar levels of smoking, a higher prevalence of overweight, and significantly greater frequency of hospital admissions, use of outpatient services and general practitioner consultations. About half of the people with diabetes assessed their health as good or excellent but 90 per cent stated that they were happy or very happy. No differences between diabetic people living in capital cities and other areas were found. These results have implications for education and life-style behaviour modification programs for people with diabetes. Research into the prevention and more effective management of diabetes and its complications is required in order to contain the escalating health care burden associated with diabetes in Australia.

Adolescent↗

PubMed1996

Risk factors for stroke mortality in men and women: The Busselton Study.

BACKGROUND: Although there have been a number of prospective studies of risk factors for stroke, the Busselton study remains the only Australian study to report such data. The Framingham stroke risk appraisal function has been evaluated in very few populations. This paper presents results of stroke mortality risk factor analyses from the Busselton study and compares the Busselton risk factor coefficients with those based on Framingham data. METHODS: Mortality follow-up to 31 December 1993 on 4872 participants in the Busselton Health Survey 1966-1981 was used for Cox proportional hazards regression analysis of risk factors for death from stroke. Risk factors investigated were systolic and diastolic blood pressures, antihypertensive therapy, diabetes, smoking, body mass index, cholesterol, alcohol, coronary heart disease, atrial fibrillation and left ventricular hypertrophy. RESULTS: The average duration of follow-up was 20 years and a total of 213 individuals died from stroke during the follow-up period. Age, systolic blood pressure, antihypertensive therapy, smoking, cholesterol, alcohol (negatively), coronary heart disease and atrial fibrillation were significantly and independently associated with stroke modality rate. The Cox regression coefficients were generally similar to those from the Framingham study, but there were important differences. CONCLUSIONS: This report has confirmed age, blood pressure, antihypertensive therapy, smoking, and pre-existing cardiovascular disease and diabetes as important risk factors for stroke. The suggestion of a protective effect from moderate alcohol consumption requires further study. The Framingham stroke risk appraisal function should be validated before being recommended for individual stroke risk assessment in other populations.

Adult↗

PubMed1995

1989-90 National Health Survey: prevalence of self-reported diabetes in Australia.

OBJECTIVE: To estimate the prevalence of self-reported diabetes in a representative population sample in Australia; to identify major differences in prevalence by State and Territory and by geographic origin; and to compare trends in prevalence over time. DESIGN: and setting: Analysis of data collected by the Australian Bureau of Statistics in the 1989-90 National Health Survey. The sampling frame (22,202 households and 54,241 people) ensured adequate representation from each State and Territory in Australia. RESULTS: Prevalences of self-reported diabetes/high blood sugar levels were 1.9% for males and 2.0% for females over all ages; 3.1% for men and 2.9% for women aged over 25 years; and 8.1% for men and 6.9% for women aged over 65 years. Interstate differences of borderline significance were found and will require validation in larger samples. Age-standardised prevalences were higher in subjects born in Southern Europe and lower in those born in Western Europe compared with Australian-born subjects. Prevalences were high in households where the language spoken was other than English. The frequency of self-reported diabetes appears to be rising. CONCLUSIONS: Australia has comparatively low total prevalences of self-reported diabetes/high blood sugar levels. High prevalences subsections exist, including the elderly and certain migrant groups

Adolescent↗

PubMed1995

Age and secular trends in risk factors for cardiovascular disease in Busselton.

Mortality rates from heart disease and stroke in Australia have been falling for more than 20 years. No completely satisfactory explanations for this trend exist. However, it is believed to be due, at least in part, to changes in the incidence of cardiovascular disease arising from changes in the prevalence and severity of risk factors for cardiovascular disease. The adult community of Busselton in Western Australia participated in cross-sectional health surveys every three years from 1966 to 1981. This paper describes secular trends from 1966 to 1981 and age trends from 25 to 80 years for cardiovascular risk factors in Busselton men and women. Downwards secular trends were observed for mean blood pressure and smoking for men and women, upwards trends were observed for body mass index in men, and mean cholesterol was approximately constant over this period. The age and secular trends were consistent with other Australian studies conducted in the 1980s and with overseas studies. An estimated 67 per cent of the decline in cardiovascular mortality rates among Busselton men and 22 per cent of the decline among Busselton women may be attributed to changes in the prevalence of risk factors for cardiovascular disease.

Adult↗

PubMed1994

A comparison of hospital and police road injury data.

In order to gather as much information as possible on road crashes and outcomes, routinely collected police reports of traffic accidents and hospital discharge files were individually matched or "linked" using a computerised iterative procedure on name-identified data from both sources. The two groups of linked and unlinked hospital records were compared. Within the linked dataset, a comparison of like variables was made and showed good agreement between the two sources on accident type and road user type. However, police-reported levels of injury severity were shown to be less reliable. In addition, the proportion of hospital inpatient records that linked to a police record was found to be influenced by several factors. The overall linkage rate from hospital to police was 64% but varied from 29% for motorcyclists in single-vehicle accidents to 79% for motor vehicle drivers. The linkage rate increased with increasing levels of injury severity and was substantially lower for casualties of certain ethnic groups. It was deduced that for most instances where a hospital record did not link to a police record, the crash had not been reported. These findings confirm that there was considerable underreporting of hospitalised road casualties to the police and that the extent of underreporting was greater for those less severely injured.

Accidents, Traffic↗

PubMed1994

Serum insulin is a risk marker for coronary heart disease mortality in men but not in women.

Serum insulin 1 h post-glucose load is examined in this prospective study of 2971 Caucasoid subjects aged > 20 years in 1966 and followed to 1989. The serum insulin levels as a continuous variable show no significant linear association with coronary heart disease (CHD) deaths in either sex after accounting for age by Cox proportional hazards analysis. In males the quintile classes of serum insulin show a striking U-shaped pattern with both the highest and lowest quintiles having significant associations with CHD deaths. In females the insulin quintiles show no direct association. Analysis for interactions of risk variables indicate that in females the relative protection of low cholesterol levels is abolished by hyperinsulinaemia after 12 years. Thus, serum insulin is not a direct aetiological risk factor for CHD. The findings suggest that the associations are likely to be due to confounding effects of unmeasured variables including lipid subfractions.

Adult↗

PubMed1994

Mortality and coronary heart disease in diabetes mellitus.

A clinic-based study of 1,063 patients with Type 2 diabetes recruited from 1973 to 1982 identified 533 deaths (attributed to coronary heart disease in 268 cases) by 31 December 1989. When compared to the general population of Australia the overall standardised mortality ratio was 1.42 (95 per cent confidence interval (CI) 1.26 to 1.58) for females and 1.19 (CI 1.03 to 1.35) for males. Cox regression analysis showed that having coronary heart disease or absence of foot pulses at the time of entrance to the study were the major independent risk factors for overall mortality after adjustment for initial age. Elevated cholesterol and blood pressure were found to be major independent risk factors for death from coronary heart disease.

Adult↗

PubMed1994

Mortality trends, 1965 to 1989, in Busselton, the site of repeated health surveys and interventions.

The community of Busselton in Western Australia has participated in repeated cross-sectional health surveys and other health interventions since 1966. Health surveys of adults were conducted every three years from 1966 to 1981. Health interventions, including programs for high blood pressure, smoking, high blood cholesterol levels and obesity, were implemented in the community. Mortality rates for the Busselton area are compared with rates for the remainder of the southwest region of Western Australia in an attempt to determine if the surveys and associated interventions have had any impact on survival. Statistical comparisons via Poisson regression analysis showed that mortality rates for males over the period 1965 to 1989 in Busselton declined at a similar rate to the southwest. However, for females, especially those aged 45 to 74 years, mortality rates declined significantly faster in Busselton than in the southwest, suggesting a beneficial impact on survival of the surveys and associated interventions.

Adult↗