Is the apparent cardioprotective effect of recent alcohol consumption due to confounding by prodromal symptoms?

Previous research has reported a protective association between alcohol drinking and acute coronary heart disease in the 24-hour period after drinking. This study investigated whether this apparent protective association resulted from confounding due to the effect of prodromal symptoms on drinking behavior. In 1992, the authors conducted a case-control study that measured recent alcohol consumption and reasons for recent abstention from alcohol among patients with acute coronary heart disease identified from a community-based disease register and a representative control sample from the same community (Auckland, New Zealand). Cases were significantly more likely than controls to report recent abstention from drinking because they felt unwell. In unadjusted analyses, a protective association was observed between recent alcohol consumption and acute coronary heart disease; however, this association was weakened considerably after adjustment for the effect of prodromal symptoms on drinking behavior. The previously reported protective association between recent alcohol consumption and acute coronary heart disease appears to be largely due to the confounding effect of prodromal symptoms on drinking.

Hospitalization for pneumonia in children in Auckland, New Zealand.

OBJECTIVE: To describe the epidemiology of hospitalizations for pneumonia in children in Auckland, New Zealand. METHODS: A consecutive sample of children hospitalised with pneumonia at the Starship Childrens Hospital from 1 July 1993 to 30 June 1996. Subjects were Pacific Island, Maori, and European/other children aged 0-14 years resident in north, west and central Auckland who were hospitalized with pneumonia. Comparisons were made of the number of hospitalisations by year, ethnicity, age and season; and of the hospitalisation rates by year, ethnicity and age. RESULTS: There were 681 children who were hospitalized with pneumonia during 1993-94, 731 during 1994-95 and 630 during 1995-96. The average annual hospitalization rate was 5.0 per 1000 children aged 0-14 years (95% CI 4.8-5.2). The average annual hospitalisation rate for Pacific Island children was 14.0 per 1000 (95% CI 13.0-14.9), for Maori children 6.7 per 1000 (95% CI 6.0-7.4) and for European/other children was 2.7 per 1000 (95% CI 2.6-2.9). Fifty-three per cent of the hospitalised children were less than 2 years of age. A larger percentage of Pacific Island (61%) and Maori (60%) children were aged less than 2 years compared to European/other (42%) children (P < 0.001). There was marked seasonal variability in the number of hospitalizations, with peaks in hospitalizations corresponding to peaks in positive respiratory viral isolates. CONCLUSIONS: Pneumonia was a consistent cause of hospitalisation for a large number of Auckland children during this 3-year period. Hospitalisation rates and age distribution varied with ethnicity. Hospitalization rates were highest for Pacific Island. intermediate for Maori and lowest for European/other children. Based on these hospitalisation data, pneumonia is a significant cause of morbidity for children in Auckland, New Zealand.

Trends in the use of blood pressure lowering medications in Auckland, and associated costs, 1982-94.

AIM: To describe the trends to the use of blood pressure lowering medication and associated costs in Auckland, New Zealand between 1982 and 1994. METHODS: Three cross sectional surveys of cardiovascular risk factors in people aged 35-64 years have been conducted in the Auckland region in 1982, 1986-8 and 1993-4, with random selection of 3804 European men and women from Auckland electoral rolls. RESULTS: Mean systolic and diastolic blood pressure fell significantly in both sexes between 1982 and 1993-4. There was a possible trend towards a decrease in the proportion of the survey population taking blood pressure lowering drugs with 9.3% on medication in 1982 and 8.0% in 1993-4, while the number of drugs prescribed per person for blood pressure control declined from 1.41 to 1.15. As a percentage of the total antihypertensive drug use in the population, diuretic use dropped from 40.3% to 11.7%, and beta blockers decreased from 36% to 27%. Angiotensin converting enzyme (ACE) inhibitors that were unavailable in 1982, were the most commonly prescribed antihypertensive at 35.8% in 1993-4 and calcium antagonists increased from 2% in 1982 to 22.1% in 1993-4. In 1995 dollars the average daily cost of blood pressure lowering medication per person has increased from 35 cents in 1982 to 76 cents in 1994. CONCLUSION: Over the 12 year survey period ACE inhibitors, beta blockers and calcium antagonists have replaced diuretics as the major antihypertensive drugs used in the Auckland population. This has resulted in an increase in the average daily cost of antihypertensive drug therapy per person of approximately 100% in the period 1982-94.

Trends and determinants of blood pressure in Auckland, New Zealand 1982-94.

AIM: To describe blood pressure trends in Auckland, New Zealand from 1982 to 1994 and assess possible explanations for the trends. METHODS: Three cross sectional surveys of cardiovascular risk factors were undertaken in 1982, 1986-8 and 1993-4, with a total of 3806 European men and women aged 35-64 years randomly selected from Auckland electoral rolls. RESULTS: Mean systolic blood pressure fell in males from 132.2 mmHg in 1982 to 126.3 mmHg in 1993-4, and in females from 125.9 mmHg in 1982 to 121.7 mmHg in 1993-4. Both male and female diastolic mean blood pressure decreased more than 6 mmHg over the 12 years. The prevalence of antihypertensive drug use fell over the 12 year period. Regression analysis revealed a positive association between blood pressure and blood lipids. Body mass index (BMI) was also positively related to blood pressure while cigarette smoking was inversely related. However, concurrent trends in blood lipids, BMI and cigarette smoking could account for less than 6% of the average decline in systolic blood pressure over the 12 year period. CONCLUSION: There has been a substantial fall in mean blood pressure levels in Auckland adults aged 35-64 years which appears to be due to a shift in the general population blood pressure. The reduction in blood pressure is not explained by changes in pharmaceutical interventions and only a small part of the decline can be explained by concurrent trends in cardiovascular risk factors.

The increasing prevalence of obesity in New Zealand: is it related to recent trends in smoking and physical activity?

AIMS: To describe recent trends in body mass in Auckland, and to determine whether associated changes in cigarette smoking and physical activity could explain these trends. METHODS: Risk factors for coronary heart disease were measured in three age-stratified random samples of Europeans aged 35-64 years using a standard protocol in 1982, 1986-8 and 1993-4. Body mass index (BMI) was calculated as weight in kg/(height in m2), with overweight or obese being defined as BMI>25. Results. The age-standardised mean BMI increased from 26.6 (25.4-25.8) to 26.4 (26.2-26.7) in men and from 24.6 (24.2-24.9) to 25.1 (24.8-25.5) in women between 1982 and 1993-4. The prevalence of overweight and obese people increased from 52.8% (48.1-57.7) to 64.2% (58.1-70.3) in men and from 36.5% (31.5-41.5) to 44.9% (39.7-50.1) in women. Self reported cigarette smoking declined between 1982 and 1993-94 with an associated increase in the prevalence of ex-smokers. The change in smoking status accounted for only 7% and 10% of the observed increase in BMI in men and women respectively. Most of the observed change in BMI was due to a general increase in BMI in all smoking categories in both sexes. The prevalence o leisure time physical activity increased in both sexes between 1986-88 and 1993-94 while work physical activity decreased. The change in physical activity should have decreased mean BMI by approximately 4% in men and 14% in women. As with smoking there was a trend towards increasing mean BMI irrespective of the activity undertaken. CONCLUSION: Recent trends in smoking cessation explain only a small percentage of the increase in body mass while the trends in physical activity would have predicted a small decrease in the prevalence of obesity, contrary to the observed trends. By exclusion, an increase in total energy intake, is the most likely explanation for the observed trends.

Trends in coronary heart disease risk factors in Auckland 1982-94.

AIMS: This paper describes trends in major coronary heart disease risk factors over the period 1982 to 1994 in the city of Auckland, New Zealand. METHODS: Coronary heart disease risk factor levels were measured in three cross-sectional surveys in Auckland in 1982, 1986-8 and 1993-4, following a standardised protocol. Random samples of nonMaori, nonPacific Island adults aged 35-64 years were selected from the Auckland general electoral rolls using similar methods in all three surveys. Participants attended a study centre for interview and risk factor measurement. RESULTS: The analyses presented are based on samples of 1029 men and 569 women in 1982, 541 men and 365 women in 1986-8, and 712 men and 685 women in 1993-4. The data are directly age-standardised to the 1986 New Zealand population. Over the 12 year period the prevalence of self-reported cigarette consumption declined significantly from 28.6% to 16.9% in men and from 24.5% to 14.8% in women. Mean serum total cholesterol showed little change between 1982 and 1986-8 but declined significantly between 1986-8 and 1993-4 by approximately 6% from 6.12 mmol/L to 5.73 mmol/L in men and by 9% from 6.17 mmol/L to 5.60 mmol/L in women. Mean serum high density lipoprotein (HDL) cholesterol showed a modest increase between 1982 and 1986-8 but declined significantly between 1986-8 and 1993-4 by 12% from 1.25 mmol/L to 1.10 mmol/L in men and by 9% from 1.55 mmol/L in women to 1.40 mmol/L in women. The total cholesterol to HDL cholesterol ratio increased significantly between 1986-8 and 1993-4 by 6% in men but showed little change in women. Mean blood pressure levels fell by 4-6 mmHg systolic and 6-7 mmHg diastolic over the 12 year period in men and women. Mean body mass index increased significantly from 25.6 to 26.4 in men and from 24.5 to 25.1 in women during this period. CONCLUSIONS: Over the 12 year period, 1982-94 there have been substantial reductions in the prevalence of self reported cigarette smoking, mean serum total cholesterol levels and mean blood pressure levels in middle aged Aucklanders. Of concern, the prevalence of obesity has increased and mean serum HDL cholesterol levels have fallen over the period. Coronary heart disease prevention and control programmes appear to have been successful in reducing the prevalence of most major coronary heart disease risk factors however some reorientation will be required to redress the adverse trends in HDL cholesterol levels and obesity.

Do cardiovascular disease risk factors predict all-cause mortality?

BACKGROUND: The purpose of this study is to describe associations between a number of standard cardiovascular risk factors and all-cause mortality. METHOD: Mortality data were collected for a randomly selected cohort of 1029 New Zealand men aged 35-64 years, followed up over a 9-year period. A proportional hazards regression model was used to estimate the relative risks (RR) for all-cause mortality associated with a number of cardiovascular risk factors. RESULTS: In all, 96 deaths occurred over the 9-year period, of which 50% were due to cardiovascular causes. All-cause mortality was positively associated with cigarette smoking (age-adjusted RR = 2.01, 95% CI:1.15-3.53, current versus never), systolic blood pressure (age-adjusted RR = 2.18, 95% CI:1.23-4.44, upper versus lower tertile), and body mass index (age-adjusted RR = 1.59, 95% CI:0.94-2.66, upper versus lower tertile) and inversely associated with high density lipoprotein (HDL)-cholesterol (age-adjusted RR = 0.45, 95% CI:0.25-0.80, upper versus lower tertile). All-cause mortality was only weakly associated with serum total cholesterol (age-adjusted RR = 1.19, 95% CI:0.70-1.99, upper versus lower tertile), and there was no evidence of a U-shaped relationship for this risk factor. There was an inverse association between all-cause mortality and socioeconomic status (age-adjusted RR = 1.70, 95% CI:1.03-2.80, lower versus upper). Light alcohol consumption was associated with reduced all-cause mortality (age-adjusted RR = 0.63, 95% CI:0.37-1.05, light versus teetotal), but this benefit did not persist for alcohol consumption above about three standard drinks per day. CONCLUSIONS: The findings of this study indicate that the standard cardiovascular risk factors are likely to have a beneficial impact on all-cause mortality as well as cardiovascular disease in middle-aged and older men.

Heterosexual behaviour and HIV risk in New Zealand: data from a national survey.

The pattern of any future major heterosexual epidemic of acquired immunodeficiency syndrome (AIDS) will depend partly on sexual behaviour and condom use among heterosexuals. This survey was designed to provide information on patterns of sexual behaviour in New Zealand. A national sample aged 18 to 54 was selected using a random method and telephone interviews were administered to 2361 people, using a questionnaire based on the protocol developed by the Global Program on AIDS of the World Health Organization. The reported mean lifetime number of partners increased with age up to 25 to 29 years for women and 30 to 34 years for men, and declined at older ages. Fifteen or more lifetime partners were reported by 17 per cent of men and 4 per cent of women. Multiple partnerships in the previous 12 months were commonest in those aged 20 to 24. In this age group, 32 per cent of men and 20 per cent of women reported two or more partners. Recent condom use for contraception was reported by 23 per cent of men and 19 per cent of women. Use was highest amongst those aged 18 to 24, and decreased sharply with age. The true proportion of the population with many sexual partners may be higher than reported. These data will be useful in modelling approaches to estimate the likelihood of future heterosexual spread of AIDS. The data on lifetime numbers of partners suggest that sexual decisions depend not just on age and sex but also on the era, and thus on changing social values about sexual behaviour.

Differences in intake of specific food plants by Polynesians may explain their lower incidence of colorectal cancer compared with Europeans in New Zealand.

Epidemiological studies have implicated obesity; high intakes of alcohol, fat, and energy; and low intakes of food plants as risk factors for colorectal cancer. In New Zealand, Polynesians (including Maoris and people from several Pacific Islands) are more likely to be overweight and have higher intakes of fat and energy than Europeans, and they are likely to have similar total intakes of food plants. Yet, in New Zealand, Polynesians have a significantly lower incidence of colorectal cancer than the Europeans. It is possible that the difference in incidence of colorectal cancer is due to differences in consumption of specific food plants by Polynesians and Europeans in New Zealand. Here we have compared the consumption of specific food plants by 429 Maoris, 643 Pacific Islanders, and 4,451 Europeans in paid employment in New Zealand. Of the 51 food plants eaten by New Zealanders, 6 were eaten significantly more frequently and 17 significantly less frequently by the two Polynesian groups than by Europeans. The quantity of any protective chemical components (or other as yet unknown protective factors) in food plants is likely to be related to their botanical classification. Differences in the intake of specific food plants may at least partly explain differences in the incidence of colorectal cancer between Polynesians and Europeans.

Accounting for medical variation: the case of prescribing activity in a New Zealand general practice sample.

Medical practice variation is extensive and well documented, particularly for surgical interventions, and raises important questions for health policy. To date, however, little work has been carried out on interpractitioner variation in prescribing activity in the primary care setting. An analytical model of medical variation is derived from the literature and relevant indicators are identified from a study of New Zealand general practice. The data are based on nearly 9,500 completed patient encounter records drawn from over a hundred practitioners in the Waikato region of the North Island, New Zealand. The data set represents a 1% sample of all weekday general practice office encounters in the Hamilton Health District recorded over a 12-month period. Overall levels of prescribing, and the distribution of drug mentions across diagnostic groupings, are broadly comparable to results drawn from international benchmark data. A multivariate analysis is carried out on seven measures of activity in the areas of prescribing volume, script detail, and therapeutic choice. The analysis indicates that patient, practitioner and practice attributes exert little systematic influence on the prescribing task. The principal influences are diagnosis, followed by practitioner identity. The pattern of findings suggests also that the prescribing task cannot be viewed as an undifferentiated activity. It is more usefully considered as a process of decision-making in which 'core' judgements--such as the decision to prescribe and the choice of drug--are highly predictable and strongly influenced by diagnosis, while 'peripheral' features of the task--such as choosing a combination drug or prescribing generically--are less determinate and more subject to the exercise of clinical discretion.(ABSTRACT TRUNCATED AT 250 WORDS)

The New Zealand Partner Relations Survey: methodological results of a national telephone survey.

OBJECTIVE: To implement and evaluated a national survey of sexual behaviour using computer-assisted telephone interviewing (CATI). DESIGN: A two-stage stratified national sample survey in which households were selected by random digit-dialing (RDD), with a single eligible interviewee per selected household, followed by subsample surveys of non-contacts and refusals to determine eligibility. METHODS: A 15-minute questionnaire based on the Global Programme on AIDS (GPA)/World Health Organization (WHO) protocol was administered by telephone to a nationally representative sample of 2361 respondents in the 18-54-year age group. RESULTS: The overall response rate was 63%, but lower in the cities, in the 18-24 age group, and among men. Three-quarters of surveyed non-contacts, and a quarter of re-surveyed refusals, did not meet the eligibility criteria for the study. Less than 20% of refusals cited the subject matter of the survey as the reason for refusal. Item non-response (< 1%) increased with question sensitivity, and varied by respondent age, ethnicity and partnership status. Men reported twice as many adult lifetime partners as women. CONCLUSIONS: The GPA/WHO protocol can be successfully adapted to administration by telephone, with adequate response rates and exceptionally low levels of item non-response. CATI is a cost-effective method for collecting national information on sexual behaviour in countries where there is a high level of telephone ownership. Used in conjunction with RDD, it can overcome problems of sample design in settings where there is no comprehensive population-sampling frame. Checks on item sensitivity and partner estimates suggest that acceptable levels of reliability can also be achieved.

Decreased blood selenium and risk of myocardial infarction.

The relationship between whole blood selenium levels and risk of acute myocardial infarction was investigated in a community-based control study in Auckland, New Zealand. A pilot study in 14 patients admitted to hospital within 4 hours of onset of symptoms demonstrated that selenium levels were stable in the first 16 hours after admission for an acute myocardial infarction. Some 252 cases (199 men, 53 women) presenting to hospital within 20 hours of onset of acute myocardial infarction were compared with 838 controls (500 men, 338 women), group-matched for age and sex. Myocardial infarction patients had significantly lower mean selenium levels: 82.8 and 87.9 micrograms/l in male cases and controls (p = 0.003) and 82.1 and 88.5 micrograms/l in female cases and controls (p = 0.02) respectively. The relative risks of myocardial infarction in participants with selenium levels below the median level (85 micrograms/l) in comparison with participants above the median were 1.6 (95% CL 1.1-2.2) and 1.7 (95% CL 0.9-3.5) in men and women respectively. The effects of a low selenium level on risk of myocardial infarction were confined to cigarette smokers. These results suggest the hypothesis that a decreased blood selenium in the presence of cigarette smoking is a risk factor for coronary heart disease.

Trends in cardiovascular risk factors in Auckland, 1982 to 1987.

This paper examines trends in Auckland over a five year period (1982-87) in the major cardiovascular risk factors: serum total cholesterol, blood pressure and cigarette smoking; trends in body mass index are also presented. The data came from two independent random samples of European people aged 40-64 years chosen from the central Auckland electoral rolls in 1982 (915 men and 476 women) and 1986-88 (503 men and 359 women); response rates were over 80% in both surveys. In the five year period self reported smoking declined by 22% in men and 10% in women; the decline was particularly marked in upper socioeconomic men and women aged 55-64 years. There were no consistent changes in either blood pressure or body mass index levels. Serum cholesterol levels, after adjustment for the change in laboratory methods, declined by approximately 1% in both men and women. Cardiovascular risk factor levels remain high in Auckland; comprehensive population based prevention programmes, such as Heartbeat (New Zealand), are urgently required in New Zealand.

Patterns of care and professional decision making in a New Zealand general practice sample.

It is hypothesised that professional and organisational factors will significantly influence the pattern of care in general practice. This hypothesis is assessed for four key activities: laboratory test orders, prescriptions, referrals, and request for follow up. The data are drawn from a 1% sample of all weekday general practice office encounters in the Hamilton health district recorded over a twelve-month period from February 1979. Overall, 16% of patients receive a laboratory test, under two-thirds a prescription, 8% are referred, and under half receive a specific request for follow up. Interpractitioner variation in activity is substantial for each service: orders for laboratory tests vary between 1% and nearly half of all patients seen, the upper limit of activity for both prescribing and follow up is four patients in five, while rates of referral vary between 0 and 20%. Professional and organisational attributes exert a limited influence on this variation in service activity. Nevertheless, older practitioners, those trained outside New Zealand or without postgraduate qualifications, solo practitioners and practices with high volumes, with low fees and located outside Hamilton, have slightly lower rates of laboratory test orders and slightly higher rates of prescribing. These effects persist after controlling for diagnosis. In a multivariate analysis, diagnostic criteria are seen to be of overwhelming importance in the determination of activity levels. The identity of the practitioner is also a significant influence. Overall, the model accounts for only a fifth of the variation in activity level.

The influence of practitioner and practice characteristics on service activity levels in a New Zealand general practice sample.

The impact of practitioner and practice-related characteristics on service activity levels in general practice is assessed. It is hypothesized that while initial medical contact is largely patient-initiated and hence influenced by broader social factors, the subsequent provision of services within the medical system is more likely to be determined by organisational and professional factors. The data are drawn from a one per cent sample of all weekday general practice office encounters in the Hamilton Health District recorded over a twelve-month period from February 1979. The average encounter involved three services: typically a history or examination, a prescription and a request for follow-up. Activity levels were measured as the number of services received from the general practitioner by patients in the course of the medical encounter. The broad social characteristics of patients had little impact on activity level. Descriptors of patient health condition, and related attributes such as age and gender, were only moderately predictive of the number of services received. Practitioner and practice-related characteristics seemed to exert little influence, even after controlling for case mix. The strongest predictor of the number of services provided was the identity of the patient's doctor. Of the 30 per cent of variation in total service use explained by the model, half was accounted for by practitioner identity.