A randomised trial of population screening for melanoma.

OBJECTIVES: Melanoma is a significant cause of morbidity and mortality worldwide and incidence is increasing. Survival after treatment is inversely related to the thickness of the tumour at diagnosis. Population screening has the potential to reduce mortality but there is no conclusive evidence of benefit. Such evidence can come best from a randomised trial. Here we describe the design of a community based randomised trial of a population screening programme for melanoma in Queensland, Australia and early results of the first phase of the trial. METHODS: A total of 44 communities (aggregate population 560 000 adults aged 30 years or more) will be randomised to receive either a community based screening programme for 3 years or normal practice. The screening programme promotes thorough skin self examination and whole body skin examination by a doctor and provides open access skin cancer screening clinics. In its first phase, the trial is underway in nine intervention and nine control communities. The primary outcome measure is mortality from melanoma during 15 years of follow up. RESULTS: The first phase of the trial has shown the feasibility of implementing a population skin screening programme including regular skin cancer screening clinics, and has shown the strong support of communities and doctors for the programme. There has been a significant 2.5-fold increase in participation in screening in the intervention communities in this first phase after the first 12 months of the trial and no significant increase in participation in screening in control communities during this period. CONCLUSIONS: The design of a community based randomised trial of screening for melanoma has been successfully peer reviewed and the intervention has been shown to be feasible in practice. This randomised trial may be one of the last opportunities to develop the evidence required for public health recommendations for population screening for melanoma.

Factors associated with Pap smear taking in general practice: focusing public health initiatives.

This study ascertained the relationships of patient, practitioner and consultation factors with: 1. taking a Pap smear or referral to a specialist in the consultation; and 2. an unsuccessful offer to take a Pap smear or making an appointment for a smear. In a cross-sectional study of 3478 women presenting to 230 randomly selected general practitioners in Brisbane and Toowoomba, information about most recent Pap smear, screening in the consultation, and independent variables were collected from patients and doctors. Relationships between three levels of outcome variable (no action, Pap smear taken or referral, appointment or refusal), and independent variables (practitioner variables, consultation variables, patient variables) were modelled using polytomous logistic regression. Presenting for a routine checkup and breast cancer screening were associated with all types of action. Younger age, longer consultations and consultations with a female practitioner were associated with Pap smear taking and referrals, and not appointments or refusals. Being due for a Pap smear, having blood pressure measured, consulting a younger general practitioner and one who ascribed to current guidelines on screening were associated with an appointment or a refusal, when compared with no action. Results identify different profiles of those who get a Pap smear and those who do not, the former indicating a more proactive patient group, while the latter suggest more active general practitioners who attempted opportunistic screening of passive patients, or women who do not specifically seek Pap smears. We have identified factors that have significance for developing public health programs focused on consumers and providers.

Telephone administration of the SF-36 health survey: validation studies and population norms for adults in Queensland.

The Rand Corporation medical outcomes short-form 36 health survey (SF-36) is a multidimensional measure of self-perceived general health status, which has been validated in adult populations in the United States and Great Britain, and, more recently, in an Australian population. The SF-36 is increasingly being used in health outcomes research internationally, mainly as a self-administered tool, and clearly has potential for use in Australia. This study aimed to assess the acceptability, reliability and validity of telephone administration of the instrument in the Queensland adult population, and to provide reliable population norms. We report the results of a telephone survey in which we interviewed 12,793 adults. It was the first large-scale, statewide application of the SF-36 in Australia. A response rate of 82 per cent was achieved, and the SF-36 satisfied psychometric criteria for reliability and construct validity. Population norms broken down by age and sex are provided. They will be important for the interpretation of future studies using the SF-36 in particular population or patient groups.

Measurement of human exposure to ultraviolet-B solar radiation using a CR-39 dosimeter.

CR-39 (allyl diglycol carbonate) dosimeters were tested for measurements of ultraviolet-B (erythemal dose) solar radiation with a rotating model head and a human survey during solar exposure. The ratio of summer dose to winter dose for the vertex is 3.3, and for the eye and lower face is greater than 1. A survey suggests the relation of the ultraviolet-B dose to outdoor activities.

Prevention and management of chronic disease in Aboriginal and Islander Community Controlled Health Services in Queensland: a quality improvement study assessing change in selected clinical performance indicators over time in a cohort of services.

OBJECTIVE: To evaluate clinical healthcare performance in Aboriginal Medical Services in Queensland and to consider future directions in supporting improvement through measurement, target setting and standards development. DESIGN: Longitudinal study assessing baseline performance and improvements in service delivery, clinical care and selected outcomes against key performance indicators 2009-2010. SETTING: 27 Aboriginal and Islander Community Controlled Health Services (AICCHSs) in Queensland, who are members of the Queensland Aboriginal and Islander Health Council (QAIHC). PARTICIPANTS: 22 AICCHS with medical clinics. INTERVENTION: Implementation and use of an electronic clinical information system that integrates with electronic health records supported by the QAIHC quality improvement programme-the Close the Gap Collaborative. MAIN OUTCOME MEASURES: Proportion of patients with current recording of key healthcare activities and the prevalence of risk factors and chronic disease. RESULTS: Aggregated performance was high on a number of key risk factors and healthcare activities including assessment of tobacco use and management of hypertension but low for others. Performance between services showed greatest variation for care planning and health check activity. CONCLUSIONS: Data collected by the QAIHC health information system highlight the risk factor workload facing the AICCHS in Queensland, demonstrating the need for ongoing support and workforce planning. Development of targets and weighting models is necessary to enable robust between-service comparisons of performance, which has implications for health reform initiatives in Australia. The limited information available suggests that although performance on key activities in the AICCHS sector has potential for improvement in some areas, it is nonetheless at a higher level than for mainstream providers. IMPLICATIONS: The work demonstrates the role that the Community Controlled sector can play in closing the gap in Aboriginal and Torres Strait Islander health outcomes by leading the use of clinical data to record and assess the quality of services and health outcome.

Reducing indigenous mortality in Australia: lessons from other countries.

Mortality rates from all causes in Maoris in New Zealand and Native Americans have fallen substantially since the early 1970s. Comparable mortality rates for Australian Aboriginals and Torres Strait Islanders in 1990-1994 were at or above the rates observed 20 years ago in Maoris and Native Americans, being 1.9 times the rate in Maoris, 2.4 times the rate in Native Americans, and 3.2 times the rate for all Australians. Circulatory diseases, respiratory diseases, injuries and endocrine diseases (mostly diabetes) are responsible for almost 70% of these excess deaths. Mortality rate trends in indigenous populations in other countries suggest the feasibility of substantial and rapid reductions in mortality rates of Australia's indigenous people.

Childhood cancer incidence in Australia, 1982-1991.

The data of the Australian Paediatric Cancer Registry on childhood cancer incidence in Australia for the 10-year period 1982-1991 are presented. The crude average annual incidence of cancer in children under the age of 15 years was 13.8 per 100,000. The incidence of childhood cancer in Australia is rising. Significant increases were seen in acute non-lymphoblastic leukaemia, astrocytoma and melanoma. The age-standardised incidence of 14.4 per 100,000 is about 34% higher than in the UK. Most types of cancer had a higher incidence in Australia than in the UK, and the difference was significant for acute lymphoblastic leukaemia, astrocytoma and melanoma. Of particular interest is malignant melanoma, whose incidence in Australia is more than 5 times that in the UK, as a result of excessive UV exposure. Australia has a higher incidence of Ewing's tumour than osteosarcoma, nearly twice that of the UK. International comparative studies may help to elucidate the aetiology of these tumours.

Risk prevalence and screening for cancer by general practitioners.

BACKGROUND: This study presents cancer risk prevalence and screening rates of patients attending general practitioners. Conditions addressed include smoking, skin cancer, Pap smears, clinical breast examinations, and mammography. METHODS: A cross-sectional study of 7,160 patients 18-75 years presenting to 230 general practitioners in a metropolitan and country region in Queensland, Australia, was used. Risk prevalence was assessed from patients' self-report of risk behaviors and screening prior to the consultation. Details of preventive care provided in the consultation were based on doctors' self-report at the conclusion of the consultation. RESULTS: Twenty-seven percent of men and 23% of women smoked, rates being highest in the younger age groups. Using national guidelines, prior screening had not occurred as recommended for men and women as follows: skin cancer screening--66% of men, 70% of women; cervical cancer screening--27%; clinical breast examination--45%; mammography--75%. These patients were thus categorized as "at risk," and screening rates for these patients in the consultation were as follows: smokers--34%; skin cancer--5%; cervical cancer--16%; mammography--4%; clinical breast examination--8%. CONCLUSIONS: Significant numbers of patients presenting to general practitioners were at risk of developing cancer, according to national guidelines. Of these, only small proportions of all groups had a preventive intervention by their GP at the consultation identified in this study. While GPs generally see their role in prevention as important, these results raise questions for future cancer prevention policies and training of GPs.

Screening for cardiovascular disease and risk reduction counselling behaviors of general practitioners.

BACKGROUND: This study presents prevalence of risk factors for cardiovascular disease in general practice patients and screening rates for risk factors. Conditions addressed include smoking, weight, alcohol intake, blood pressure, and cholesterol. METHODS: Adult patients (7,160) attending 230 general practitioners in a metropolitan (Brisbane) and rural (Toowoomba) region in Queensland were recruited to a cross-sectional study. A self-administered patient questionnaire provided self-reported information on lifestyle risk factors and rates of previous screening. A doctor's questionnaire completed at the conclusion of the consultation provided information about physician knowledge of patient risk factors and details of preventive care provided in the consultation. RESULTS: Twenty-five percent of patients reported that they smoked, 2% drank beyond defined safe limits, 40% had body mass index > 24.9 (kg/m2). Doctors identified 66% of self-reported smokers, 40% of heavy drinkers, and 59% of overweight patients. Over 90% of patients reported prior blood pressure measurements in agreement with national recommendations; cholesterol screening within the past 5 years was reported by 51% of patients. Screening and/or counselling of patients in the consultation was highest for blood pressure (47%) and smoking (34%) and considerably lower for overweight (22%), alcohol (19%), and cholesterol (6%). CONCLUSIONS: Although preventive activities are being undertaken in general practice, performance of these activities is less than ideal. The barriers to undertaking these activities need to be addressed for change to occur.

Public hospital activity, expenditure and staffing levels for indigenous and nonindigenous settlements in remote Queensland.

Aborigines and Torres Strait Islanders who live in discrete communities have poor health. Their life expectancy is relatively low and the underlying rates of morbidity, mortality and hospitalisation are much higher than the Australian average. For Aborigines and Torres Strait Islanders in the remote communities, the mortality rate is some three times higher than that of the total population. By comparing remote Aboriginal and Torres Strait Islander communities with similarly remote nonindigenous settlements, the paper shows that there was greater hospital activity per person in remote Aboriginal and Torres Strait Islander communities. Paradoxically, whereas there was greater hospital activity in discrete indigenous communities, there was lower expenditure per person on hospital services. The paper finds also that there were fewer medical and nursing staff per person in remote Aboriginal and Torres Strait communities.

Information-seeking behaviour and sources of health information: associations with risk factor status in an analysis of three Queensland electorates.

Making a special effort to gain health information has been associated with healthy behaviour. This paper identifies the information-seeking dimension and the sources of information people use and examines whether there is an association between these and individual health status. In November 1989 a randomly selected sample of 2134 adults in the electorates of Maryborough, Toowoomba and Port Curtis (Gladstone) completed a self-administered questionnaire which aimed to measure health-related knowledge, attitudes and behaviour. Factor analysis identified clusters of items representing different dimensions of risky health behaviour as well as different perceived barriers to seeking preventive health care. Multiple regression analysis was then used to determine sociodemographic, attitudinal and behavioural variables related to a tendency not to seek health information and to be at risk for preventable diseases. The findings have implications for the development of future health promotion programs in provincial Australian cities. Both age and sex were significant in predicting where people obtain information on health. Women were more likely than men to seek health information and their sources of information reflected this proactive attitude to health. While there was no significant difference between the perceived health status of males and females, reported behavioural risk factors were much more prevalent in males. Most respondents had consulted a general practitioner in the previous twelve months and stated that they would change their behaviour on the advice of a medical practitioner. However, few nominated medical practitioners as their main source of health information.

Reducing systematic bias in studies of general practitioners: the use of a medical peer in the recruitment of general practitioners in research.

Reducing systematic bias in any group of study participants should be a priority of any researcher. This can be achieved by ensuring the sampling framework is adequate and by increasing response rates. Response rates in studies of general practitioners have to date tended to be low. Generalization of results to the wider population of GPs is therefore reduced. This paper systematically examines those factors which can reduce bias, recognising accurate identification of the target population, gaining good access to respondents, and maximising response rates as crucial factors. The importance of a medical peer in recruitment is examined. Applying these factors to a study situation, three different recruitment strategies were tested. As the strategy improved, there was an incremental improvement in the response rate (44%, 67%, 78%). These results indicate that by specifically addressing strategies which facilitate access to the target population, and increase the legitimacy and credibility of the study, significant improvements in response rates can be achieved.

Mortality from myocardial infarction in different types of hospitals.

Hospitals ranging from large urban teaching hospitals to small country hospitals were stratified into four levels of care and examined for their effectiveness of coronary care in relation to these levels. The crude hospital mortality among 2265 patients admitted for definite or possible acute myocardial infarction was 21% at level 1 (the most elaborate level), 22% at level 2, 21% at level 3, and 19% at level 4 (the least elaborate). Adjustment for age or other prognostic factors produced no significant differences across levels either for coronary care unit care or for combined coronary unit and ward care. Success in resuscitation was also similar across levels. These findings suggest that increased resources for coronary care units--whether for new services or for upgrading existing ones--may not be required.

Differences in mortality from acute myocardial infarction between coronary care unit and medical ward: treatment or bias?

To analyse the effectiveness of coronary care units in reducing mortality from myocardial infarction 18 hospitals ranging from large urban teaching hospitals to small country hospitals were stratified into four levels of care. Previous analysis had failed to show significant differences in the overall mortality in hospital among levels. There were significant differences in mortality, however, between those patients allocated to be cared for in the coronary care unit and those in the medical wards in the more advanced hospitals. The differences were largest in the hospitals with the most elaborate facilities (level 1) and non-existent in those with the least (level 4). Several analytical approaches to these observed differences indicated that they were: (a) reduced by adjustment for age and severity of infarction; (b) paralleled by differences in coexisting disease recorded on death certificates; (c) no longer significant at level 1 after allowing for differences in coexisting disease; and (d) not significant at any level after exclusion of patients first diagnosed at necropsy. These findings suggest that the observed differences in mortality between coronary care units and medical wards are largely due to bias in selection and diagnosis.

Conducting regional health surveys using a computer-assisted telephone interviewing method.

After the creation of thirteen health regions within Queensland in 1991, the need arose for an information base at the regional level to assist regions with their role in planning, monitoring and evaluating health services. A series of regional health surveys was conducted in 1993 to provide this information, using a computer-assisted telephone interviewing method. Over 10 400 interviews were conducted throughout the state. This is the first time a computer-assisted method has been used on a large scale to collect health-related information in Australia. Interviews used list-directed or random-digit dialing, depending on the rate of unlisted numbers in a region. Response rates were not significantly different for the two methods, although the number of contactable numbers attempted and the noncontact rates were significantly higher for random-digit regions. The last-birthday method was used to select the adult for interview in each household. The method resulted in a bias toward female respondents.

Use of Queensland Hospital services by interstate and overseas visitors.

In response to concerns about the number of interstate and overseas visitors using Queensland hospital services, the present study examined a sample of 1,295 hospital records to determine the proportion of patients who were incorrectly identified as Queensland residents. Across six hospitals the overall detection rate was 4.6%. Rates varied between hospitals, with the highest detection recorded for Goondiwindi near the Queensland/New South Wales border; and the lowest for Prince Charles in Brisbane. There were also important variations across hospitals based on specific holiday periods. In particular, Goondiwindi and the Gold Coast had substantially higher detection rates for the Christmas holiday period (December-January) than for the mid-year period (June-August). These findings are discussed in terms of their implications for hospital services, especially lost revenue and increased patient load. Health information managers are identified as a key group for addressing some of the current problems in this area.

Correlates of physician counseling associated with obesity and smoking.

BACKGROUND: Patient, physician, and consultation variables associated with overweight and smoking counseling in general practice consultations were examined. METHODS: A random sample of full-time general practitioners was used. The sample consisted of 7,160 patients from 230 GPs who attended for consultations on consecutive days, and self-reported information from the doctor and the patient was collected via questionnaire. The aim of this paper is to identify variables associated with the doctor's identification of overweight and smoking status and with the occurrence of counseling for these two behavioral risk factors. RESULTS: Forty percent of patients were overweight (BMI > 24) and 25% were self-reported smokers. Doctors identified 59% of overweight patients and 66% of smokers. Doctors only counseled patients identified as having the risk factor, counseling 36% of identified overweight patients and 49% of identified smokers. Identification of overweight was associated with being female, being heavier, having been previously counseled, being less well educated, presenting with an associated condition, and visiting a doctor who is younger and knows the patient's medical history well. Counseling for overweight was associated with being younger, being previously counseled, presenting with an associated condition, presenting for a routine checkup, visiting a GP who generally has longer consultations, having BP measured in the consultation, visiting an older doctor and visiting a doctor who considers identification of risk behaviors important. Identification of smokers was associated with being a heavier smoker, with those who had been previously counseled, with marital status other than single or married, with a BP measurement being taken in the consultation, and with a doctor who believed it possible to influence lifestyle change. Counseling for smoking was associated with younger patients, longer consultations, previous counseling, BP measurement, presenting with an associated condition, and not presenting frequently. CONCLUSIONS: We have identified factors associated with counseling about behavioral risk factors which provide a framework for planning education programs to increase the level of primary preventive activities within general practice.

Bacterial invasion of oral tissues of immunosuppressed rats.

The periodontia of immunosuppressed rats were examined by light and electron microscopy. Bacterial invasion was noted in the periodontal tissues of septicemic ligature-treated animals. Invading bacteria consisted of gram-negative rods which were morphologically similar to the Pseudomonas aeruginosa strain isolated from the mouth of one of the septicemic animals.