Population-based linkage of health records in Western Australia: development of a health services research linked database.

OBJECTIVES: To introduce the Western Australian Health Services Research Linked Database as infrastructure to support aetiologic, utilisation and outcomes research. To compare the study population, data resources, technical systems and organisational supports with international best practice in record linkage and health research. METHOD AND RESULTS: The WA Linked Database systematically links the available administrative health data within an Australian State of 1.7 million people. It brings together, initially, six core data elements (birth records, midwives' notifications, cancer registrations, in-patient hospital morbidity, in-patient and public out-patient mental health services data and death records). It will be updated regularly and is designed, in future extensions, to include data on primary, residential and domiciliary care and health surveys. Linkage uses probabilistic matching of patient names and other identifiers. Geocodes for spatial analysis are assigned using address linkage and mapping software. By June 1997, the project had taken 2 1/2 years to develop the system and link seven million core data records from 1980 to 1995. CONCLUSIONS: The system is consistent with international benchmarks, from four centres of excellence, for the study population, core datasets, matching and geocoding, and collaborative networks. There are prospects to redress deficiencies in primary medical contact and other data resources, validation studies, tracing systems and a more supportive legal framework. IMPLICATIONS: The WA Linked Database will be used in combination with medical record audits to provide a comprehensive evaluation of health system performance.

Trends in repeat prostatectomy after surgery for benign prostate disease: application of record linkage to healthcare outcomes.

OBJECTIVE: To compare the risk of repeat prostatectomy for benign prostatic hyperplasia (BPH) in a population-based cohort of 19 598 men in Western Australia treated by transurethral resection of the prostate (TURP) or open prostatectomy over a 16-year period. PATIENTS AND METHODS: The Western Australian Health Services Research Linked Database was used to extract all hospital morbidity data, death records and prostate cancer registrations for men who had prostate surgery for BPH in 1980-95. The cumulative incidence of first repeat prostatectomy calculated using the actuarial life-table and incidence-rate ratios of the first repeat prostatectomy, comparing TURP and open prostatectomy, were obtained using Cox regression. RESULTS: The cases comprised 18 464 TURPs and 1134 open prostatectomies, from which there were 1095 subsequent repeat prostatectomies. After adjustment for calendar time, age and admission type, the incidence rate of the first repeat prostatectomy was up to 2.30 times higher (95% confidence interval, 1.62-3.27) after initial TURP than for initial open prostatectomy. The absolute risks at 8 years for TURP was 6.6%, and was 3.3% for open prostatectomy. CONCLUSION: The absolute risk of a repeat prostatectomy for TURP and open prostatectomy were consistent with the best reported international experience. There was evidence that the risk in 1990-95 had declined compared with earlier periods, despite a shift towards more closed procedures. The differential risks of repeat prostatectomy should be explained to patients and considered in the development of clinical guidelines, notwithstanding the advantages of TURP over open prostatectomy in terms of surgical morbidity and cost.

Mortality and prostate cancer risk in 19,598 men after surgery for benign prostatic hyperplasia.

OBJECTIVES: To examine postoperative mortality and prostate cancer risk after the first prostatectomy for benign prostatic hypertrophy over a 17-year period in a population-based cohort of men in Western Australia, using improved methods to adjust for comorbidity. PATIENTS AND METHODS: The relative survival from death and prostate cancer incidence was calculated against the background population rates. The outcomes of transurethral resection of the prostate (TURP) and open prostatectomy (OP) were compared adjusting for calendar year, age, admission type and comorbidity using Cox regression. Fractional polynomials were used to take account of nonlinearity in confounder effects. RESULTS: At 10 years, the relative survival was 116.5% in TURP patients and 123.5% after OP. Adjusting only for confounding by age, calendar year and admission type, TURP had a higher mortality rate than OP (rate ratio, RR, 1. 20; 95% confidence interval 1.08-1.34). The RR fell to 1.10 (0.99-1. 23) after adjustment for comorbidity and to 1.07 (0.95-1.19) when accounting for nonlinearity. The relative survival from the incidence of prostate cancer at 10 years was 103.7% after TURP and 104.5% after OP. The RR adjusted for age and calendar year was 1.44 (0.94-2.21) for incidence and 1.37 (0.81-2.29) for prostate cancer mortality. CONCLUSION: There is at most a small and clinically unimportant excess mortality risk from TURP; any difference could be due to a protective effect of OP on the long-term risk of prostate cancer and a lower rate of repeat prostatectomy.

Prostate cancer in Western Australia: trends in incidence and mortality from 1985 to 1996.

OBJECTIVE: To measure trends in recorded incidence and mortality rates of prostate cancer in Western Australia from 1985 to 1996 and to relate these to prostate-specific antigen (PSA) testing for prostate cancer. DESIGN: Descriptive study based on data from the Western Australian Cancer Registry, the Australian Bureau of Statistics and the Health Insurance Commission. DATA: All newly diagnosed cases of prostate cancer and all deaths from prostate cancer in Western Australia from 1985 to 1996. MAIN OUTCOME MEASURES: Recorded incidences and mortality rates for prostate cancer. RESULTS: After increasing steadily from 42 per 100,000 person-years in 1985 to 61 in 1992, the recorded incidence more than doubled to 134 per 100,000 person-years in 1994, then fell sharply to 87 in 1996. Among men aged 50 years or more, those aged 50-54 years had the largest annual increases: 14% (95% confidence interval [CI], 10%-19%) from 1985 to 1992 and 108% (95% CI, 84%-134%) from 1992 to 1994. They also had the smallest annual decline between 1994 and 1996 (8%; 95% CI, 1% increase to 16% decrease). The mortality rate showed no sudden increases or decreases. In men aged 60 years or older, the mortality rate increased annually by 2.9% (95% CI, 2%-4%) from 1985 to 1996. The number of Medicare reimbursements for PSA tests increased until May 1995, then fell. There was a significant correlation between the monthly number of PSA tests and new cases of prostate cancer (P < 0.01). CONCLUSIONS: Following a period of steady increase, the recorded incidence of prostate cancer increased dramatically in 1992 because of screening by PSA testing. From 1994, these incidence figures declined almost as sharply, partly because of reductions in testing. The mortality rate has not shown any systematic deviation from its long-term trend.

The Quality of Surgical Care Project: a model to evaluate surgical outcomes in Western Australia using population-based record linkage.

BACKGROUND: The aim of this study is to establish a model to evaluate surgical outcomes and, where indicated, recommend changes to improve the quality of surgical care in Western Australia (WA). Open resection for aneurysm of the abdominal aorta was the first procedure evaluated and the results are reported in an accompanying paper. METHODS: The Quality of Surgical Care Project (QSCP) is conducted under the aegis of the Royal Australasian College of Surgeons (RACS) in WA, and brings together a multidisciplinary team of surgeons, public health researchers and health service administrators. The Western Australia Health Services Research Linked Database (the WA Linked Database) is used to provide linked chains of patients records residing in the state health department from the following sources: hospital morbidity data system, birth and death records, mental health services data, cancer registrations and midwives' notifications. This links 16 years of population-based patient records from 1980, including all public and private hospital admissions and re-admissions. The Quality of Surgical Care Project was established to use and to correlate the data from the WA Linked Database. RESULTS: The result is a powerful database for a contained population that is available for scientific analysis by a multidisciplinary team of clinical epidemiologists, surgeons and health service managers. Users will have the ability to establish benchmark standards for the outcomes of surgical procedures in WA for use in quality improvement programmes run by the College and will facilitate self-directed performance auditing activities as a commitment to greater community accountability. CONCLUSIONS: The Quality of Surgical Care Project provides a potential model of benefits to be realized by both the medical profession and the community through multidisciplinary collaboration supported by adequate information. Although migration from WA is relatively low, future linkage to the state electoral roll will allow correction for any population change.

Respiratory disease in non-smoking Western Australian goldminers.

Respiratory symptoms, spirometry, and transfer factor were measured in 208 non-smoking Western Australian underground goldminers (mean age 32) to identify the presence of respiratory abnormalities resulting from underground work. These subjects were part of a larger group of 771 subjects attending for statutory periodic chest x ray examinations in the industry. They had worked underground for a median of three years. The prevalence odds ratios of bronchitis, dyspnoea, wheeze, and asthma all tended to be related to duration of underground employment, even after adjusting for age, those for wheeze and asthma reaching statistical significance. After adjusting for age and height the duration of employment also had a significant effect on TL/VA but not on FEV1, FVC, or TL. These changes are consistent with the presence of airway narrowing and non-specific lung fibrosis or emphysema in non-smoking underground goldminers.

Comparison of diets supplemented with fish oil or olive oil on plasma lipoproteins in insulin-dependent diabetics.

This study was designed to compare changes in high-density lipoprotein (HDL)- and low-density lipoprotein (LDL)-cholesterol in normolipidaemic male insulin-dependent diabetics (IDD) following dietary supplementation with either the fish oil concentrate Max EPA or olive oil. The contribution of the small quantity of cholesterol in Max EPA to these changes was also examined. Twenty-seven subjects were matched in groups of three and randomly allocated to one of three treatment groups of nine subjects each. Subjects were given 15 1-g capsules of oil daily for 3 weeks, consisting of either Max EPA, olive oil, or olive oil to which was added the same amount of cholesterol as contained in Max EPA, respectively. There was a significant increase in eicosapentaenoic acid, and a decrease in arachidonic acid, in the platelet membrane phospholipids of subjects taking Max EPA. In this group, there was an approximately 30% increase in serum HDL2-cholesterol (0.59 +/- 0.07 to 0.77 +/- 0.11 mmol/L, mean +/- SEM; P less than .01) and a corresponding decrease in HDL3-cholesterol (0.79 +/- 0.03 to 0.71 +/- 0.03 mmol/L; P less than .05). Although total and LDL-cholesterol concentrations were also higher after Max EPA, the changes were not significant. Triglycerides were significantly decreased by Max EPA. There were no significant changes in lipids in the groups given olive oil. These results show that compared with olive oil, dietary supplementation with Max EPA substantially increases HDL2-cholesterol in insulin-dependent diabetics. This is most likely due to a selective effect of omega 3 fatty acids.(ABSTRACT TRUNCATED AT 250 WORDS)

Cervical cytology in Western Australia. Frequency, geographical and socioeconomic distributions and providers of the service.

Basic data were obtained from the records of 16,069 women who had smears taken for cervical cytological examination in Western Australia during an eight-week period in 1983. The peak smear rate was 340.7 per 1000 at 25-29 years of age and fell thereafter with age. The estimated peak frequency of smears that were designated as "screening" smears was 178.3 per 1000 at 30-34 years of age. Screening smears comprised 39% to 66% of the total number of smears, depending on age. After correction for the estimated prevalence of past hysterectomy, only in the age range 20-34 years did the rate of all smears approach the rate of screening smears that would be obtained under a recommended frequency of once every three years. The frequency of screening smears was 20% less in rural areas of Western Australia than in the capital city, Perth. In Perth it fell with decreasing socioeconomic status. General practitioners took 62.4% of all smears and 70.3% of screening smears. On average, female general practitioners took twice as many smears than did male general practitioners.

Evidence for a direct effect of alcohol consumption on blood pressure in normotensive men. A randomized controlled trial.

A direct pressor effect of alcohol is proposed as the basis for the association between regular alcohol consumption and an increase in blood pressure found in population studies. To examine this further, a randomized controlled crossover trial of the effects of varying alcohol intake on blood pressure in 46 healthy male drinkers was conducted. From an average of 336 ml of ethanol per week, alcohol consumption was reduced by 80% for 6 weeks by drinking a low alcohol content beer alone. This reduction was associated with a significant reduction in systolic and diastolic blood pressure (p less than 0.001 and p less than 0.05 respectively). The mean difference in supine systolic blood pressure during the last 2 weeks of normal or low alcohol intake was 3.8 mm Hg, which correlated significantly with change in alcohol consumption (r = 0.53, p less than 0.001). Reduction of alcohol intake also caused a significant decrease in weight (p less than 0.001). After adjustment for weight change, an independent effect of alcohol on systolic but not diastolic blood pressure was still evident, with a 3.1 mm Hg fall predicted for a decrease in consumption from 350 ml of ethanol equivalent per week to 70 ml per week (p less than 0.01). Systolic blood pressure rose again when normal drinking habits were resumed. These results provide clear evidence for a direct and reversible pressor effect of regular moderate alcohol consumption in normotensive men and suggest that alcohol may play a major role in the genesis of early stages of blood pressure elevation.

Plasma noradrenaline and its relationship to plasma oestradiol in normal women during the menstrual cycle.

In order to investigate a possible relationship between sympatho-adrenal neuronal activity and the endocrine changes during the menstrual cycle, free and sulphate-conjugated plasma catecholamines and oestradiol were measured under carefully controlled conditions in 26 normal menstruating women. Plasma oestradiol levels were generally higher during the luteal compared with the follicular phase which corresponded to the self-reported day of the cycle. Free plasma noradrenaline concentration was higher during the luteal phase (P = 0.02) and was positively correlated with plasma oestradiol concentration (r = 0.40, P = 0.023). These relationships were not present for plasma adrenaline. It is conceivable that the higher luteal phase noradrenaline is causally related to the higher oestradiol levels, leading to incomplete inactivation by reducing tissue uptake or competitive inhibition of catechol-O-methyl transferase. As sulphated noradrenaline was not significantly different between the follicular and luteal phases, competitive inhibition of phenolsulphotransferase by oestradiol was considered unlikely.

Cutaneous malignant melanoma and fluorescent lighting.

A population-based case-control study of cutaneous malignant melanoma occurring during 1980-81 was conducted in Perth, Western Australia. Three hundred and thirty-seven cases and 349 matched controls were reinterviewed in 1983 with regard to their fluorescent light exposure. The incidence rate of all melanomas was not associated with rate of exposure or cumulative exposure to all fluorescent lights or just those without diffusers. Separate analyses by histogenetic type and, where possible, body site of melanoma showed, in most instances, no consistent association between incidence rate of melanoma and exposure to fluorescent lights without diffusers. Incidence of melanomas of unclassifiable histogenetic type, however, increased with increasing duration of exposure (P-value for trend .02). This association was weaker and the P-value higher (.11) when exposure was considered only in residential rooms and offices, where light fittings are closest to the subject. Adjustment for the effects of total and intermittent sun exposure on melanoma rates did not alter the above trends appreciably.

Vegetarian diet, blood pressure and cardiovascular risk.

This paper reviews the association between a vegetarian diet and a number of risk factors for cardiovascular disease investigated in a series of epidemiological and experimental studies. Ninety-eight Seventh-day Adventist "vegetarians" were similar to 113 Mormon omnivores for strength of religious affiliation, consumption of alcohol, tea and coffee and use of tobacco, but were significantly less obese and had significantly lower blood pressures adjusted for age, height and weight. A random sample of forty-seven Adventist vegetarians had significantly lower home blood pressures, serum cholesterol levels and blood pressure responses to a cold-pressor test than Mormon omnivores carefully matched for age, sex and Quetelet's index. In a controlled dietary intervention study mean systolic and diastolic blood pressures and serum cholesterol fell significantly during feeding with a vegetarian diet--an effect unrelated to changes in other lifestyle factors. Dietary analysis indicated that a vegetarian diet provided more polyunsaturated fat, fibre, vitamin C, vitamin E, magnesium, calcium and potassium and significantly less total fat, saturated fat and cholesterol than an omnivore diet. There was no evidence for a difference between vegetarians and omnivores in levels of catecholamines, plasma renin activity, angiotensin II, cortisol or serum and urinary prostanoids.

Alcohol, personality and predisposition to essential hypertension.

The nature of the relationship between alcohol, personality and blood pressure levels was examined in 491 working men who completed detailed questionnaires which included Eysenck's personality inventory. Alcohol had an effect on systolic blood pressure levels independent of all other factors studied. However, in 152 non-smoking moderate to heavy drinkers (greater than 18 g ethanol per day) the extroversion/introversion trait was the most significant predictor of systolic blood pressure levels, and in introverted drinkers the prevalence of hypertension (greater than or equal to 140 mmHg systolic or greater than or equal to 90 mmHg diastolic) was three times that of extroverted drinkers and nine times that of teetotallers. This association between introversion and 'hypertension' was not seen in drinkers who also smoked cigarettes. The interactions between environmental stimuli (alcohol, smoking) and presumably genetically determined personality characteristics may have an important bearing on concepts of essential hypertension and point to new approaches for investigation.

Relationships between age, body weight, physical fitness and sex-hormone-binding globulin capacity.

The associations between sex-hormone-binding globulin capacity (SHBG), age, body mass index (BMI), and physical fitness have been studied in 34 men and 36 women. Multivariate analysis was used to look for independent associations with SHBG. The data indicate that when controlled for a number of other factors SHBG levels are related, in men but not in women, to age (positively, p less than 0.001) and BMI (negatively, p less than 0.001).

The relationship of blood pressure to diet and lifestyle in two religious populations.

The association between blood pressure and a vegetarian diet was studied in relation to obesity, sex, age and lifestyle in 98 Seventh-day Adventist (SDA) lacto-ovo vegetarians, 82 SDA omnivores and 113 Mormon omnivores aged 25 to 44 years. Mean blood pressures adjusted for age, height and weight were significantly lower in SDA vegetarians than in Mormon omnivores (115.6/68.7 and 121.2/72.2, respectively, in males and 109.1/66.7 and 114.9/72.6, respectively, in females) and were not related to past or current use of alcohol, tobacco, tea and coffee, physical activity, personality or religious observance. Mean blood pressures in SDA omnivore males (121.7/71.7) were similar to those in Mormon males, while those in SDA omnivore females (109.9/67.4) were similar to SDA vegetarian females. Quetelet's Index in these subgroups demonstrated the same pattern as blood pressure and may reflect, in part, the high level of physical activity in female SDA omnivores. The prevalence of mild hypertension (greater than or equal to 140 mmHg systolic or greater than or equal to 90 mmHg diastolic) was 10 and 8.5% in Mormon and SDA omnivores, respectively, compared with 1 to 2% in SDA vegetarians. Analysis of diet records showed that vegetarians ate significantly more dietary fibre, polyunsaturated fat, magnesium and potassium and significantly less total fat, saturated fat and cholesterol than did Mormon omnivores. SDA omnivores had a dietary pattern which was less homogeneous, and which lay between those of the other groups. Which, if any, of these dietary differences were responsible for the blood pressure differences could not be determined in this study.

Vegetarian diet, lifestyle and blood pressure in two religious populations.

1. The association between vegetarianism and blood pressure was studied in relation to obesity, sex and aspects of lifestyle in 180 Seventh-day Adventists and 113 Mormons aged 25-44 y. 2. Volunteers completed a questionnaire, a 1-day diet record and submitted to standardized measurements of blood pressure, heart rate and body size. 3. Ninety-eight Adventist "vegetarians' were comparable to the 113 Mormon omnivores for strength of religious affiliation, consumption of alcohol, tea and coffee and use of tobacco, but were significantly less obese. 4. Obesity correlated positively with blood pressures in males and females of both diet classes. Age showed a positive correlation with blood pressure in females only. 5. Adjustment of blood pressures for age and Quetelet Index indicated that there is an additional blood pressure reducing effect associated with a vegetarian diet.