Search

The association between the use of inhaled beta-agonists and the risk of death and near-death from asthma has previously been reported. It was based on a nested case-control study of 129 cases and 655 control subjects selected from a cohort of 12,301 users of asthma drugs followed during the period 1980 through 1987. In this paper we examine the question of asthma and non-asthma mortality using data from the entire cohort of 12,301 asthmatics. There were 46 asthma and 134 non-asthma deaths in this cohort, for which there were 47,842 person-years of follow-up. The overall rate of asthma death was 9.6 per 10,000 asthmatics per year. This rate varied significantly according to the use of fenoterol, albuterol, or oral corticosteroids in the prior 12 months and the number of asthma hospitalizations in the prior 2 years. The rate decreased significantly, by 0.6 asthma deaths per 10,000 asthmatics per year over the study period, after controlling for the effect of the four other risk factors. It also increased significantly with the use of all beta-agonists, and more so for fenoterol than for albuterol, although this difference was partly explained by the dose inequivalence of the two drugs. Change-point dose-response curves showed that the risk of asthma death began to escalate drastically at about 1.4 canisters (of 20,000 micrograms each) per month of inhaled beta-agonist, the recommended limit. For non-asthma death, the overall rate of 28 deaths per 10,000 asthmatics per year was not related to the use of inhaled beta-agonists.(ABSTRACT TRUNCATED AT 250 WORDS)

We have previously reported an increasing dose-response relationship between the regular use of beta-agonist inhalers and the risk of asthma death and near death among a cohort of 12,301 subjects who had been dispensed 10 or more prescriptions of asthma drugs from January 1980 to April 1987. That analysis was based solely on information obtained from linkable computerized data bases. Such an association might be explained in part by the tendency of patients with more severe asthma, that is, those at greatest risk for an adverse outcome, to use more beta-agonist medication. To further examine this potential confounding by severity, we gathered clinical information independently from the field on the 129 case patients and their 655 control patients from the matched case-control analysis of 12,301 subjects. In 68% of the control patients with a life-threatening episode and 75% of the matched control subjects, we obtained a valid questionnaire from at least one physician who had seen the patient during the previous 2 yr. Acceptable information on hospitalizations because of asthma was obtained in 87% of those hospitalized. Clinical features associated with an increased risk of fatal and near-fatal asthma were: a history of loss of consciousness or seizures during a previous asthma attack (odds ratio, 10.2; 95% CI, 3.9 to 26.7), a history of attacks of asthma precipitated by eating certain foods (odds ratio, 5.1; 95% CI, 2.4 to 11.1), a clinical score designed to reflect the severity of prior attacks of asthma leading to hospitalization, and prior respiratory acidosis among those in whom a blood gas determination was recorded.(ABSTRACT TRUNCATED AT 250 WORDS)

To determine the frequency of and risk factors for falls and injurious falls in the noninstitutionalized elderly, the authors conducted a follow-up study of 409 community-dwelling persons aged 65 years or more in west-central Montreal, Quebec, Canada, from May 1987 to October 1988. Following an initial at-home interview, each subject was telephoned every 4 weeks for 48 weeks for collection of data on falls experienced since the last contact. Each of the 12 follow-up interviews was completed by at least 90% of the subjects eligible for interview. Data were also collected in the follow-up interviews on time-varying exposures. Twenty-nine percent of the subjects fell during follow-up; 17.6% fell once, and 11.5% fell two or more times. The incidence rate for falls was 41.4 falls per 1,000 person-months. The majority of falls resulted in no injury or in minor injury only. Potential risk factors investigated included sociodemographic variables, physical activity, alcohol consumption, acute and chronic health problems, dizziness, mobility, and medications. Multivariate analyses showed that the following factors were statistically significantly associated with an increased rate of falls: dizziness (incidence rate ratio (IRR) = 2.0), frequent physical activity (IRR = 2.0), having days on which activities were limited because of a health problem (IRR = 1.8), having trouble walking 400 m (IRR = 1.6), and having trouble bending down (IRR = 1.4). Factors which were protective included diversity of physical activities (IRR = 0.6), daily alcohol consumption (IRR = 0.5), having days spent in bed because of a health problem (IRR = 0.5), and taking heart medication (IRR = 0.6). Risk factors for injurious falls were similar.

Objective: To examine the relationship between patterns of use of inhaled beclomethasone dipropionate and the risk of fatal and near-fatal asthma., Design: Nested case-control analysis of a historical cohort; a further analysis., Setting: The 12,301 residents of Saskatchewan aged 5 to 54 years who were dispensed 10 or more asthma drugs from 1978 to 1987., Patients: The 129 persons who experienced asthma death (n = 44) and near-death (n = 85) and their 655 controls matched as to age and date of entry into the cohort, with the additional matching criteria of at least one hospitalization for asthma in the prior 2 years, region of residence, and having received social assistance., Main Outcome: Life-threatening attacks of asthma defined as death due to asthma or the occurrence of hypercarbia, intubation, and mechanical ventilation during an acute attack of asthma., Results: After accounting for the risk associated with use of other medications and adjustment for markers of risk of adverse events related to asthma, subjects who had been dispensed, on average, one or more metered-dose inhalers of beclomethasone per month over a 1-year period had a significantly lower risk of fatal and near-fatal asthma (odds ratio, 0.1; 95% confidence interval, 0.02 to 0.6)., Conclusion: These data support recent guidelines from several countries that recommend the use of inhaled corticosteroids in moderate and severe asthma.

Eighty patients undergoing elective thoracotomy were studied to assess the possibility of predicting arterial oxygenation (PaO2) during one-lung anaesthesia (OLA). The first 50 patients were studied retrospectively. The method of multiple linear regression was used to construct a predictive equation for PaO2 during OLA. Potential predictors of PaO2 during OLA which were considered were: age, side of operation, preoperative pulmonary flow rates, preoperative and intraoperative PaO2 during two-lung ventilation. The three most significant predictors for PaO2 during OLA were: side right of operation (P < 0.05), preoperative FEV1% (P < 0.01) and intraoperative PaO2 during two-lung ventilation (P = 0.0001). The predictive equation for PaO2 after ten minutes of OLA was: PaO2 = 100 - 72 (side) - 1.86 (FEV1%) + 0.75 (two-lung) PaO2; (for side insert 0 for left-sided thoracotomy and 1 for right-sided thoracotomy). The remaining 30 patients were studied prospectively and the predicted PaO2 correlated with the observed PaO2 after ten minutes of OLA (r = 0.73, P < 0.01). Four of 30 patients had a predicted PaO2 at ten minutes of OLA < 150 mmHg. Of these, 2/4 subsequently required abandonment of OLA for pulse oximetric saturation < 85%. We conclude that although it is not possible to predict an individual patient's PaO2 during OLA with a high degree of accuracy, it is possible, before the initiation of OLA, to identify those patients whose arterial oxygenation is likely to decrease to low levels during OLA.

We conducted a study of general practitioners in Torino, Italy, to determine their patterns of preventive practice. We examined a set of primary and secondary preventive interventions and their determinants. We explored the determinants of primary preventive interventions (antismoking and alcohol abuse counseling, counseling for prevention of accidents at home among the elderly, flu vaccination for the elderly, and counseling on contraceptive use) and cancer early detection techniques (chest x rays and sputum cytology for lung cancer; mammography, physical examination, teaching of breast self-examination, and Papanicolaou [Pap] smear for gynecological cancers). Grouping determinants in cognitive, sociodemographic, and organization factors, we found different patterns for each maneuver. Cognitive factors played an important role, but their importance varied for each intervention. We found that smoking behavior of physicians predicted antismoking counseling. Further, the availability of other primary care services is an important factor in the early detection of gynecological cancers.

This study was undertaken to follow up an incidence rate study of spinal injuries carried out from a cross-sectional random sample consisting of 2,342 workers who were compensated at least once in the year 1981 by the Quebec Worker's Compensation Board (QWCB) for an absence from work. These workers were followed for 3 years using the QWCB information system to record any recurrence of compensated absence from work. A total of 850 (36.3%) had at least one recurrence, and had longer episodes of absence than those without a recurrence (P less than 0.0001). Of these 850 workers, 824 (96.9%) had less than five recurrences and showed a systematic trend of gradual increasing duration of absence on each subsequent recurrence (P less than 0.05 in a repeated measures analysis of variance). A positive relationship was found between the duration of the initial episode of absence from work and the subsequent history of absence from work, both in terms of risk of recurrence (P less than 0.001) and of cumulated absence from work (P less than 0.0001), after controlling for age, sex, and site of symptoms. The computed instantaneous risk of entering a recurrence in a 31 year-old man experiencing lumbar symptoms, was 19.9% (95% confidence interval = 19.8-19.9) in the year following an initial episode of one day duration, and 26.7% (95% confidence interval = 24.3-29.3) after an initial episode of 6 months.

It has been previously demonstrated that serum lactate dehydrogenase is elevated among HIV patients with pneumocystis carinii pneumonia (PCP). To evaluate the clinical utility of this test we analyzed the admission LDH levels of patients hospitalized for the first time due to the secondary complications of AIDS. Among 76 patients without a prior history of PCP, 41 (54%) had PCP diagnosed during their hospitalization while 35 (46%) did not have PCP. Serum LDH was significantly higher among PCP patients than in patients without PCP (mean = 423 IU/L vs 234 IU/L). Receiver operating characteristic curve analysis demonstrated that at an optimal cutoff point of LDH greater than or equal to 240 IU/L, the test sensitivity and specificity were 0.78 and 0.74 respectively among all hospitalized patients. However, when only patients with dyspnea were considered, the optimal test sensitivity and specificity improved to 0.94 and 0.78 at a cutoff point of LDH greater than or equal to 220 IU/L. Comparing the areas under fitted ROC curves, serum LDH was a significantly better discriminator among patients with dyspnea than among those who were not short of breath. We conclude that while serum LDH is strongly associated with the presence of PCP among AIDS patients, it is a poor screening test for PCP when applied to all hospitalized AIDS patients with and without respiratory complaints. Serum LDH is no substitute for appropriate microbiological studies. However, with further evaluation, it may prove to be a useful test in guiding the clinical management of dyspneic patients in whom sputum or bronchial examinations are negative or not immediately available.

Objective: The purpose of this study was to determine whether the severity o clinical presentation, sex, age, HLA type, and the presence of IAs and ICAs could predict the variation of residual insulin secretion as measured by the serum C-peptide response to a Sustacal meal., Research Design and Methods: A cohort of 151 newly diagnosed IDDM children (mean age 10.2 +/- 4.6 yr) was followed prospectively for 3 yr. Thirty-five patients (12 males, 23 females) were still secreting C-peptide after 36 mo., Results: We found that age (P = 0.0001), sex (P = 0.003), presence of ICA (P = 0.006), severity of clinical presentation (P = 0.001), and symptom duration (P = 0.002) significantly predicted the rate of loss of C-peptide secretion. The risks of accelerated C-peptide disappearance decreased with increasing age, the risk ratios being 0.25 for the older group (greater than 12 yr) compared with the younger group (less than 6 yr) and 0.50 for the intermediate group (6-12 yr) compared with the younger group. The risk for the presence of ICA was 1.7, and the risk for males was 1.7 also. There was a significant negative correlation between ICA titers and C-peptide at 18 and 24 mo after diagnosis (P = 0.04). There were no significant differences in HbA1 values between patients who secreted C-peptide and those who did not., Conclusions: We conclude that younger age of onset, male sex, high titers of ICA, severe clinical presentation, and shorter symptom duration significantly predict accelerated rates of loss of C-peptide secretion.

The ability of four experienced clinicians to predict short-term outcome (serum creatinine level at 1 year) and long-term outcome (renal insufficiency) was evaluated in 87 patients with lupus nephritis. The correlational agreement and the accuracy of their predictions were contrasted with the actual outcomes observed and with statistically generated prognostic regression models. In contrast to previously published data, all four clinicians predicted both short-term outcomes (P < 0.001) and long-term outcomes (P < 0.02) well. The clinicians' predictions approximated that of a statistically generated computer model for both agreement and accuracy for renal function at 1 year. The four clinicians identified nearly identical clinical variables as important in determining prognosis. Provision of biopsy data to the clinicians improved short-term and long-term prediction slightly. The value of the statistical models was 'validated' by demonstrating that three of the four clinical variables identified by the models, but not by the clinicians, could enhance clinical prediction (P < 0.05). In addition, the extent of tubulo-interstitial involvement on biopsy, a predictor that has recently received increased attention, could improve the long-term clinical predictions of all four clinicians (P < 0.05).

Between October 1981 and December 1989, 44 patients with cutaneous T-cell lymphoma (CTCL) were treated with a single field rotational total skin electron irradiation (RTSEI) technique developed in the McGill University, Department of Radiation Oncology. Only 11 (25%) of the 44 patients had received no prior treatment. Three-quarters (33/44) had advanced (T3 or T4) disease. Complete responses were seen in 32/44 (73%) of patients (91% T2, 71% T3 and 58% T4), but only 3/11 (27%) of patients with T2 disease and 3/21 (14%) of patients with T3 disease remain in continuous complete remission in the skin, after median intervals of 58 and 35 months, respectively. Median cause-specific survival for the whole group is 43 months and survival at 5 years is 38%. Survival was significantly better for patients with T2 disease than for patients with T3 disease (relative risk 4.3; 95% CI 1.4-13.2) and patients with T4 disease (relative risk 3.1; 95% CI 0.8-12.1). The RTSEI technique used at McGill has depth-dose characteristics and photon contamination similar to other commonly used TSEI techniques. It is relatively simple and provides a homogenous dose distribution over the entire skin surface in a short treatment time. Results of treatment are similar to those obtained with other techniques. For T2 disease, TSEI is an effective treatment modality with a possibility of long-term tumor control. For more advanced disease, more aggressive treatment, which may include TSEI, is necessary.

Background: Morbidity and mortality from asthma appear to be increasing, and it has been suggested that medications used to treat asthma are contributing to this trend. We investigated a possible association between death or near death from asthma and the regular use of beta 2-agonist bronchodilators., Methods: Using linked health insurance data bases from Saskatchewan, Canada, we conducted a matched case-control study of subjects drawn from a cohort of 12,301 patients for whom asthma medications had been prescribed between 1978 and 1987. We matched 129 case patients who had fatal or near-fatal asthma with 655 controls (who had received medications for asthma but had not had fatal or near-fatal events) with respect to region of residence, age, receipt of social assistance, and previous hospitalization for asthma., Results: The use of beta-agonists administered by a metered-dose inhaler was associated with an increased risk of death from asthma (odds ratio, 2.6 per canister per month; 95 percent confidence interval, 1.7 to 3.9) and of death or near death from asthma, considered together (odds ratio, 1.9; 95 percent confidence interval, 1.6 to 2.4). For death from asthma, use of the beta-agonist fenoterol was associated with an odds ratio of 5.4 per canister, as compared with 2.4 for the beta-agonist albuterol. On a microgram-equivalent basis, the odds ratio for this outcome with fenoterol was 2.3, as compared with 2.4 with albuterol., Conclusions: An increased risk of death or near death from asthma was associated with the regular use of inhaled beta 2-agonist bronchodilators, especially fenoterol. Regardless of whether beta-agonists are directly responsible for these adverse effects or are simply a marker for more severe asthma, heavy use of these agents should alert clinicians that it is necessary to reevaluate the patient's condition.

Objective: To evaluate the lifetime benefits of reducing total serum cholesterol levels to prevent coronary heart disease (CHD)., Design: We developed a CHD primary prevention computer model to estimate the benefits associated with lifelong risk factor modification. We validated the model by comparing the computer estimates with the observed results of three primary CHD prevention trials., Patients: Men and women age 35 to 65 years who are free of CHD, with total serum cholesterol levels ranging from 5.2 to 7.8 mmol/L (200 to 300 mg/dL), with or without additional CHD risk factors., Interventions: Serum cholesterol reduction through dietary modification or diet and medications., Main Outcome Measures: Changes in life expectancy and the delay of symptomatic CHD., Results: The computer forecasts for CHD end points closely matched the observed results of the Lipid Research Clinics Trial, the Helsinki Heart Study, and MRFIT. We then applied the computer model to low-risk and high-risk men and women with total serum cholesterol levels between 5.2 and 7.8 mmol/L (200 and 300 mg/dL) and estimated that, after reducing serum cholesterol levels 5% to 33%, the average life expectancy would increase by 0.03 to 3.16 years. We also forecast that the average onset of symptomatic CHD would be delayed among these patient groups by 0.06 to 4.98 years., Conclusion: We conclude that this computer model accurately estimates the results of clinical trials and can be used to forecast the changes in life expectancy and morbidity (the development of CHD) associated with specific CHD risk reduction interventions. The wide variation surrounding these estimates underscores the need to better define which groups of individuals will gain the most from cholesterol reduction.

Two recent epidemiologic case-control studies suggested that fenoterol, a selective beta-adrenergic agonist, was associated with an increase in the risk of asthma death. The results of these studies were criticized because of methodologic problems in the choice and selection of control subjects; the different methods used to gather exposure data in cases and control subjects; and because of inadequate classification and adjustment for asthma severity. In response to this controversy, a new study is underway, the Saskatchewan Asthma Epidemiology Project. The SAEP includes two complementary studies, an historic cohort and a case-control analysis, that employ the computerized databases of the Saskatchewan Health Department. A unique aspect of the SAEP is the attempt to incorporate knowledge of asthma physiology and management into the design of the studies. Specifically, the study design recognized the role of antiinflammatory drugs in asthma treatment; the distinction between asthma death and near-fatal asthma; the severity of asthma; patterns of drug use; and the distinction between inadequate clinical care and disease severity. The strategies we employed in the SAEP may prove helpful to investigators whenever clinical and biologic processes create sources of potential bias requiring special procedures for the design and analysis of epidemiologic studies.

The short-term prognosis of lupus nephritis was evaluated by assessing serum creatinine 12 months after renal biopsy in 87 patients with lupus nephritis. On univariate analysis, significant clinical and laboratory predictors of this outcome included clinical signs of renal injury (serum creatinine, 24-hour urinary protein, prolonged renal disease, nephrotic syndrome, serum albumin), as well as thrombocytopenia, older age, and coexisting illness or hypertension at the time of biopsy. On renal biopsy, diffuse proliferative nephritis, higher activity, chronicity, or tubulointerstitial scores, or subendothelial or subepithelial electron dense deposits predicted a higher serum creatinine 12 months after biopsy. A clinical predictive model was developed which included as independent predictors serum creatinine, age, platelet count and 24-hour urinary protein. Any one of three biopsy variables added information to the clinical prediction model: a marked quantity of subendothelial deposits (p = 0.02), a higher activity index score (p = 0.02), or the presence of diffuse proliferative lupus nephritis (p = 0.05). However, the relative predictive accuracy of the clinical model did not improve with the addition of any of the biopsy variables. The value of renal biopsy in lupus nephritis is discussed based on the ability of biopsy information to confirm the prognosis, to add new predictive information for a group of subjects, and to improve predictive accuracy for individual patients.

To successfully appraise the significance of epidemiological data on drug risk and safety requires a good understanding of the errors involved in the design and analysis of pharmacoepidemiological studies. A proper comprehension of the repercussions of these errors and of the strengths and limitations of the tools used to measure their magnitude are essential to sound decision making by the regulatory, industry or clinical consumers of these data. In this paper, we examine the role of statistics in managing the quantifiable errors present in pharmacoepidemiological data analysis and interpretation. Some epidemiological principles on the measurement of risk are first introduced. The influences of controllable systematic error and random error on our assessment of epidemiological data are then presented, along with the prevailing statistical principles and measures necessary to control these errors. To illustrate the various issues addressed, published data on the risks of NSAIDs, focusing particularly on upper gastrointestinal bleeding (UGIB), the risks of replacement estrogens for endometrial cancer and the safety of allopurinol for cataracts are used as examples throughout.

An exact unconditional method for the design and analysis of the 2 x 2 matched-pairs trial is presented. Unlike the exact conditional method, which is based on only the number of discordant pairs, the exact unconditional method uses the total number N of sampled pairs. This unconditional test, based on a simple Z statistic, yields sample sizes that are generally smaller than those produced by the exact conditional test for the cases tabulated herein, namely one-sided alpha = .01, .025, and .05 along with 80% and 90% power. Moreover, it is found to be uniformly more powerful than the latter for all the combinations of parameters considered in this paper, namely alpha = .01, .025, and .05, and N = 10(1)200. The method is illustrated by assessing various design options of an in vitro study of the effectiveness of antimicrobial agents. Some numerical examples of matching efficiency are also given.

Objective: To test three methods of introducing health charts into the medical records of six family medicine units., Design: Quasi-experiment., Participants: The staff physicians and family medicine residents in all six units and the nurses in two units., Interventions: Group 1 (minimal intervention): health charts, a user's guide and one training session. Group 2 (intermediate intervention): same intervention as for group 1 plus two feedback sessions at 3 and 6 months. Group 3 (maximum intervention): same intervention as for group 2 plus promotion of the team concept (nurses were included). The intervention phase lasted from September 1987 to August 1988., Outcome Measures: The frequency with which the health charts were used, the item scores of each preventive care activity and the overall unit scores. Data were gathered through chart audits at baseline and at the end of the intervention phase., Results: The frequency with which the health charts were used varied from 3.9% to 26.9%. The greatest increases in item scores were observed in the use of mammography (20.0%), counselling on lifestyle (19.4%) and breast examination (17.2%). Although the overall improvement in the unit scores was statistically significant (p less than 0.05) the hypothesis of an increasing gradient of effect across the three intervention groups could not be tested because of the variation in scores across the units., Conclusion: Health charts and other similar tools are useful; however, they are not sufficient to change practice behaviours. The support of a "champion" on the health care team might well be a determining factor of success for the delivery of preventive services in primary care practice.

Objective: To determine the cost of screening all Canadians aged 30 years or more without coronary heart disease (CHD) for hypercholesterolemia., Data Sources: The expected results of initial screening of the serum cholesterol level were estimated on the basis of 1986 Canadian census data and the 1978 Canada Health Survey. The results of repeat testing were estimated on the basis of data from the Lipid Research Clinics Prevalence Study. Lipid profile results were extrapolated from tests at the Montreal General Hospital's clinical chemistry laboratory. Laboratory costs and primary care practitioner costs were provided by the Canadian Society of Clinical Chemists and provincial fee schedules respectively., Main Results: Among 12,479,356 Canadians free of CHD 48.7% would be identified as being at high risk, 4.8% would be identified as being at moderate risk, and 46.6% would be reassured that their lipid risk for CHD was low. The total cost of implementing the program in the first year would be $432 million to $561 million ($325 million for laboratory tests and $107 million to $236 million for visits to primary care practitioners)., Conclusion: The substantial cost of implementing a nationwide screening program must be weighed against the expected benefits to ensure that the final result is both practical and economically feasible.

Often a "disease" or "state of disease" is defined by a subdomain of a continuous outcome variable. For example, the subdomain of diastolic blood pressure greater than 90 mmHg has been used to define hypertension. The classical method of estimating the risk (or prevalence) of such defined disease states is to dichotomize the outcome variable according to the cutoff value. The standard statistical analysis of such risk of disease then exploits methods developed specifically for binary data, usually based on the binomial distribution. We present a method, based on the assumption of a Gaussian (normal) distribution for the continuous outcome, which does not resort to dichotomization. Specifically, the estimation of risk and its variance is presented for the one- and two-sample situations, with the latter focusing on risk differences and ratios, and odds ratios. The binomial approach applied to the dichotomized data is found to be less efficient than the proposed method by 67% or less. The latter is found to be very accurate, even for small sample sizes, although rather sensitive to substitutions of the underlying distribution by thicker tailed distributions. Canadian total cholesterol data are used to illustrate the problem. For the one-sample case, the approach is illustrated using data from a study of the arterial oxygenation of 20 patients during one-lung anesthesia for thoracic surgery. For the two-sample case, data from a prognostic study of the renal function of 87 lupus nephritic patients are used.

Forty patients undergoing elective thoracotomy were studied to assess the possibility of predicting PaO2 during one-lung ventilation (OLV) when continuous positive airway pressure (CPAP) was applied to the nondependent lung. The first 20 patients were studied retrospectively and the three most significant independent variables that correlated with PaO2 during OLV with CPAP were: side of operation (P = 0.04), FEV1/FVC ratio (P = 0.01), and the intraoperative PaO2 during two-lung ventilation (P = 0.0002). By the method of multiple linear regression, these three variables were used to construct a predictive equation for PaO2 during OLV with CPAP. The second 20 patients were studied prospectively and the predicted PaO2 correlated significantly with the observed PaO2 during OLV with CPAP (r = 0.86, P less than 0.001). Therefore, it is concluded that the PaO2 during OLV with CPAP can be predicted using routinely available data.

Concerns about excesses in a wide array of adverse health outcomes have been expressed for over 25 years by a rural population in southwestern Alberta, living downwind from natural gas refineries. Among these has been the perception that deaths have occurred more frequently than ought to have been expected. As part of a large field epidemiologic study undertaken during the summer of 1985 to investigate possible health effects in this area, a residential cohort study was carried out to study mortality. The cohort was defined as all those individuals who resided in the area in 1970. A total of 30,175 person-years of risk within Alberta were experienced by this cohort during 1970-84. The deaths during this period were enumerated by resident reports and by manual record linkage with the death records of the Alberta Bureau of Vital Statistics. Age- and sex-standardized mortality ratios, based on expected rates from 2 pre-specified demographically similar, non-metropolitan Southern Alberta populations, were 0.88 and 0.84 respectively, neither of which was significantly different from unity. These data cannot address the question of etiology but they can do much to allay the anxieties of a community convinced it had experienced an epidemic of death.

Six human haematopoietic cell lines, five of leukaemic origin, including cells with myeloid, lymphoid and undifferentiated phenotype have been studied with respect to radiation response. The intrinsic radiosensitivity of the cells varied widely, the D0s ranging from 0.53 to 1.39 Gy. Five of the cell lines showed some capacity to accumulate sublethal damage; in three of these, enhanced survival was demonstrated in split-dose experiments. One cell line (HL-60) was anomalous in that although little accumulation of sublethal damage was demonstrable, survival was enhanced by fractionation of the dose. Five of the six cell lines studied were of leukaemic origin. The results support the belief that, in contrast to the almost constant radiosensitivity of normal haematopoietic cell progenitors, leukaemic cell progenitors may show a wide range of radiosensitivites.

A random sample including 2342 cases representative of all occupational back injuries in Quebec (1981) was followed up prospectively over three years to assess the recurrence rate of back problems (lumbar, thoracic, and cervical). Each medical and accident report was reviewed to obtain the site of symptoms and occupation. Age, sex, industrial sector, and number of episodes of absence from work were abstracted from the computerised Quebec Compensation Board files. The recurrence rate was 20.0% at one year follow up and 36.3% at three years. A multivariate analysis using a Poisson regression, was performed to model the risk of recurrence over time. Men had a higher chance of recurrence (risk ratio = 1.85, 95% CI = 1.50-2.27) but among recurrent cases, the average total number of episodes was comparable between men and women. Age showed a protective effect on the probability of recurrence (10 years: RR = 0.93, 95% CI = 0.88-0.98) due to the lower recurrence rate in the 45-64 year old group (31.8%). Cervical and lumbar symptoms had identical recurrence profiles whereas thoracic symptoms had a significantly lower recurrence rate. Drivers had the highest recurrence rate (42.1%) and nurses had the highest average number of recurrences (2.03) among recurrent cases. Both occupations had statistically significant excesses after controlling for the other variables.

Using a simulation game designed to teach children to obey certain traffic safety rules, an experimental study was conducted with 136 five-year-old children in four Quebec schools. Within each classroom, subjects were randomly divided into four groups: three intervention groups and one control group. Each of the experimental groups was subjected to a different intervention with outcome measured using three instruments related to attitudes, behavior, and transfer of learning of pedestrian traffic safety. Results suggest that simulation games including role-playing/group dynamics and modeling/training can change attitudes and modify behavior in the area of pedestrian traffic safety in children of this age.

The incidence of severe hypoglycemia was determined in a 1-yr prospective study of 350 insulin-dependent diabetic (IDDM) children. There were no significant differences in mean glycosylated hemoglobin, age, and duration of disease between the patients who had severe hypoglycemia and those who did not. There were 25 episodes in 24 patients (6.8%). Their insulin doses at the time of the episode (U.kg-1.day-1) were significantly higher than those of the nonhypoglycemic group (mean +/- SD 1.01 +/- 0.30 vs. 0.89 +/- 0.29; P = .04). The hypoglycemic group had a significantly higher mean number of previous episodes of severe hypoglycemia than the nonhypoglycemic group (0.92 +/- 1.18 vs. 0.25 +/- 0.68; P = .01). In only 64% of the episodes, an unusual circumstance such as strenuous physical activity or missed or delayed meals preceded the event. Multivariate analysis of the data by logistic regression showed risks of developing hypoglycemia of 2.5 per 0.5 U/day insulin and of 2.0 per previous episode of severe hypoglycemia. We conclude that severe hypoglycemia may be a recurrent problem in some diabetic children, but it does not appear to be related to age or blood glucose control. The presence of previous episodes may be a guide to identify patients at greater risk of developing severe hypoglycemia. Adherence to regular testing, strict spacing and consistency of meals, and extra food for extra activity may reduce this serious complication.

For 25 years residents of a rural area in southwestern Alberta have complained of health problems attributed to sour gas emissions from nearby natural gas refineries. We undertook a large epidemiologic study of the current health status and the selected morbidity rates among 2152 people in the exposed area. We established two comparison groups: one was a demographically similar unexposed population and the other a demographically different group also exposed to sour gas emissions in another region. The methods included a cross-sectional survey of current residents and separate historical cohort studies involving registry linkage to investigate cancer incidence and all-cause mortality. The cross-sectional survey involved a comprehensive health questionnaire, standardized clinical examinations by physicians blinded to the subjects' symptoms and concerns, and several laboratory tests. We were able to contact just under 60% of the people who we knew had moved from each area since 1958 and found no evidence of selective migration for health reasons. Although the residents of the exposed area reported an excess number of symptoms and health problems there were no significant differences in the mortality rate, incidence of cancer, reproductive problems, major ailments, hair levels of arsenic and certain metals or respiratory function between the groups.

The percentage change over time in a variable such as serum creatinine or blood pressure is a potentially appealing outcome measure for longitudinal studies of prognosis or therapy. We demonstrate that when the baseline value of the variable used in the calculation of the percentage change outcome is included in a linear regression model as an independent predictor, a quadratic relationship between the baseline level and the predicted outcome level of this same variable is inevitably masked by the linear regression model. Misinterpretation of the percentage change model has resulted in conflicting results regarding the value of serum creatinine as a prognostic marker in lupus nephritis. We illustrate the problem and resolve the conflict using data on 87 patients with lupus nephritis. A straightforward alternative approach to the use of percentage change as an outcome is proposed: it is not subject to the problems of interpretation noted in the percentage change model.

Airways responsiveness to a bronchodilator is frequently measured to assist in determining the cause of respiratory symptoms. Clinically, a greater than 15% improvement in the FEV1 is often used to define the "increased" response indicative of asthma. However, unlike other tests of lung function, reference standards derived from "healthy" members of a general population sample have never been reported. As part of a health survey carried out in Alberta, Canada, 2,609 subjects completed a standardized respiratory symptom questionnaire and had FEV1 measured before and 20 min after inhaling terbutaline sulfate via a 750-ml spacer device. Among asymptomatic never-smoking subjects with a FEV1 greater than 80% of predicted, the upper 95th percentile of bronchodilator response (BDR), when expressed as 100 x (FEV1 postBDR - FEV1 preBDR)/predicted baseline FEV1 averaged 9%. This value remained remarkably stable across gender, age (7 to 75 yr), and height groups, and deviated to 6% only when baseline FEV1 was greater than 120% of predicted. Consistent with other respiratory function variables, in which the upper limit of normal is often defined as the upper 95th percentile, our population-derived reference values provide a conceptual definition of BDR that can easily be applied to define "increased" response in the clinical setting.

A historical cohort of 68 female breast cancer patients from one institution who were enrolled in a multicenter randomized controlled trial between 1971 and 1973 were followed up to the beginning of 1986. Weight and height at the time of mastectomy were transformed into two indices of body size, namely the Quetelet Index and a weight to "ideal weight" ratio. These two indices were analyzed for their relation with overall and disease-free survival, while controlling for the effect of several potential confounding variables. While neither index was linearly related to the hazard of death or recurrence, a significant quadratic (curvilinear) relation was found for both indices and both hazards. In all cases the hazard function was concave up, indicating that not only overweight but also underweight status is predictive of an unfavorable prognosis of breast cancer. This finding offers a possible explanation for the discrepancies among previous studies on this topic.

It has been proposed that the number and extent of tumors formed after chronic exposure to dimethylhydrazine (DMH) can be predicted by the indigenous number and distribution of DNA-synthesizing cells in the murine colonic mucosa, and that this sensitivity to DMH is genetically determined. In order to test this hypothesis we studied two genetically distinct inbred strains of mice; the DMH-sensitive A/J (A) mouse, and the relatively DMH-resistant C57BL/6J (B) mouse before and after a single exposure to DMH. The untreated A strain had the longer crypt column [33.2 +/- 0.8 (SD) cells versus 28.8 +/- 0.9 cells], a higher absolute number of labeled cells per crypt column (4.4 +/- 0.6 versus 2.6 +/- 0.9), a greater labeling index (13.4 +/- 1.6% versus 9.1 +/- 2.9%), a wider proliferative compartment, and a greater number and percentage of labeled cells in the middle and upper thirds of the crypt than the untreated B strain. After acute exposure to DMH the A strain lost 14 +/- 3% of their total body weight, while the B strain lost 0.5 +/- 2% total body weight 48 h post-DMH. There was an initial loss of cryptal cells, a drop in the labeling index, and a subsequent increase and overshoot in the number of labeled cells and the labeling index. This pattern of cell loss and recovery over time was parallel in both strains, and thus cannot explain the differences in ultimate tumor formation after chronic exposure to the carcinogen. The data are consistent with the theory that the susceptibility to DMH carcinogenesis can be predicted by the indigenous proliferative characteristics of the murine colonic mucosa. The acute proliferative response to DMH in these strains is similar and parallel; thus ultimate tumor load may depend on long term effects such as the establishment of stable transmissible mutations.

Assertiveness training and its relationship to smoking behavior and how young adolescent boys and girls differ with respect to assertion was investigated. A total of 161 seventh-grade students from six health education classes participated. Two classes received an innovative smoking education program, two classes received assertiveness training plus an innovative smoking education program, and two classes received only the smoking education program usually offered by the school. Locus of control and levels of assertion and smoking behavior were assessed at pre- and posttest. The results indicate that boys are significantly more assertive than girls at age 12 (p = .0018), and this assertiveness increases equally over the next four years (p less than .0001). There were no significant differences in smoking behavior among the three groups; however, trends in smoking behavior in the desired direction were observed. There is no significant difference in smoking behavior between boys and girls. In this respect the sexes have reached equality, that is, girls are initiating cigarette smoking as frequently as are boys. Finally, there were no significant changes in assertion among the groups. It is concluded that the utility of assertiveness training for young adolescents is questionable.

In a trial to determine the effect of zeranol implants on weight gains from birth to 180 days of age, 269 male Holstein calves were assigned randomly to one of three groups: a) 56 intact calves, b) 106 castrated calves implanted at birth and again at 90 days with 36 mg zeranol, and c) 107 castrated controls. All calves were weighed at 0, 28, 56, 90, and 180 days of age. Zeranol-implanted steers averaged 9.2% greater average daily gain at 90 days and 9.5% greater at the end of the trial than control steers. Intact male calves' weight gains were intermediate between the two steer groups. There was no correlation between implant status, total protein of blood serum, mortality, birth weight, or parity of the dam. During the first 180 days of age, implanting the first 180 days of age, implanting Holstein calves with zeranol at birth and reimplanting at 90 days can return upwards of $14.89 per dollar invested.

We conducted a prospective study to describe the course of the pancreatic beta-cell function from the time of clinical diagnosis of insulin-dependent (type I) diabetes to determine whether DR type, presence of islet cell antibodies (ICA), presence of insulin antibodies (IA), age at onset, and sex could help in the prediction of residual endogenous insulin secretion. A cohort of 68 children was followed for 18 mo after diagnosis of type I diabetes. The outcome variables selected for analysis were 1) serum C-peptide peak concentration after a Sustacal meal, 2) time of disappearance of the serum C-peptide response, and 3) time after diagnosis at which the maximal serum C-peptide response was observed. After institution of insulin therapy, serum C-peptide peak concentrations rose temporarily for 1-6 mo and declined thereafter. Multivariate analysis of the data showed that DR type (P = .2488) and presence of IA (P = .1604) had no effect on serum C-peptide over time, but sex (P = .0146), age at onset (P = .0002), and presence of ICA (P = .0147) significantly contributed to the variation of serum C-peptide over time. Furthermore, age at onset, presence of ICA, and sex were also the only significant predictors of the time of disappearance of the beta-cell function. The relative risks of beta-cell-function disappearance were 0.87 (P = .0015), 9.43 (P = .0181), and 2.25 (P = .0468), respectively. In conclusion, there are distinct variations in the natural course of the beta-cell function in type I diabetes. beta-Cell-function survival is significantly shortened the younger the subject is at disease onset, if ICA are present at diagnosis, and if the subject is male.

The objectives of the study were to estimate the rate of 1-year incidence of occupational back pain, describe the duration of absence from work, and assess medical and economic aspects of this problem. A total of 2,523 files of occupational back pain, representative of the Province of Quebec for the year 1981, were examined by a multidisciplinary team. Census data from 1981 were used for the denominators. The overall 1-year incidence in 1981 was 1.37%. The age-specific incidence rates show an increase-decrease pattern that could be interpreted as a "healthy worker effect". The male to female relative risks by age range from 5.6 (20 to 24 years old) to 1.8 (55 to 64 years old). In the 1-year follow-up, 74% of the workers were absent for less than 1 month. The 7.4% of cases who were absent from work for more than 6 months accounted for nearly 75% of lost days, medical costs, and indemnity payments. We pinpoint some methodologic issues of studies on occupational back pain.

The incidence of low nocturnal blood glucose values (i.e., less than 65 mg/dl) was assessed in 20 insulin-dependent diabetic patients treated with continuous subcutaneous insulin infusion supported by capillary blood glucose monitoring before each meal and the evening snack. Patients were randomly assigned to a control or experimental group. Both groups followed an identical protocol for the first part of the study (baseline). Patients were instructed to determine capillary blood glucose measurements five times during the night for three consecutive nights. The same procedure was repeated one week later, but this time the subjects in the experimental group were instructed to have an extra snack if capillary blood glucose levels at 10:30 P.M. were 120 mg/dl or less. The control group continued with the usual routine of one evening snack at 9 P.M. At baseline, the incidence of capillary blood glucose values of less than 65 mg/dl was 13 percent. The ingestion of an extra snack at bedtime resulted in the absence of capillary blood glucose values of less than 65 mg/dl in the experimental group, whereas the incidence of capillary blood glucose values of less than 65 mg/dl in the control group remained 13 percent (p = 0.038). The capillary blood glucose concentration at 10:30 P.M. was highly predictive of the risk of nocturnal blood glucose values below 65 mg/dl (p = 0.015) and fasting capillary blood glucose values above 140 mg/dl (p = 0.0001). These data show that nocturnal hypoglycemia may be a considerable problem during continuous subcutaneous insulin infusion therapy even if the basal infusion rate is adjusted in the hospital on the basis of nocturnal blood glucose concentrations. The ingestion of an extra snack at bedtime for capillary blood glucose values below 120 mg/dl has the potential to minimize this risk. The capillary blood glucose concentration at 10:30 P.M. is a significant predictor of nocturnal hypoglycemia.

In heart failure secondary to chronic mechanical overload, cardiac sympathetic neurons demonstrate depressed catecholamine synthetic and transport function. To assess the potential of sympathetic neuronal imaging for detection of depressed transport function, serial scintigrams were acquired after the intravenous administration of metaiodobenzylguanidine [131I] to 13 normal dogs, 3 autotransplanted (denervated) dogs, 5 dogs with left ventricular failure, and 5 dogs with compensated left ventricular hypertrophy due to a surgical arteriovenous shunt. Nine dogs were killed at 14 hours postinjection for determination of metaiodobenzylguanidine [131I] and endogenous norepinephrine content in left atrium, left ventricle, liver, and spleen. By 4 hours postinjection, autotransplanted dogs had a 39% reduction in mean left ventricular tracer accumulation, reflecting an absent intraneuronal tracer pool. Failure dogs demonstrated an accelerated early mean left ventricular tracer efflux rate (26.0%/hour versus 13.7%/hour in normals), reflecting a disproportionately increased extraneuronal tracer pool. They also showed reduced late left ventricular and left atrial concentrations of tracer, consistent with a reduced intraneuronal tracer pool. By contrast, compensated hypertrophy dogs demonstrated a normal early mean left ventricular tracer efflux rate (16.4%/hour) and essentially normal late left ventricular and left atrial concentrations of tracer. Metaiodobenzylguanidine [131I] scintigraphic findings reflect the integrity of the cardiac sympathetic neuronal transport system in canine mechanical-overload heart failure. Metaiodobenzylguanidine [123I] scintigraphy should be explored as a means of early detection of mechanical-overload heart failure in patients.

This paper presents an exact significance test procedure for unidirectional multiple comparisons of Poisson variables. It is sufficiently general for application to situations where the 'denominator', which might be the amount of follow-up time in cohort studies or the expected number of events in studies that use standardized rates, differs from one group to the other, and where a group-specific alternative hypothesis is of interest. We use unidirectional Z statistics, of the form proposed by Dunnett, along with the conditional distribution needed in computing the exact attained significance levels, to address three distinct unidirectional hypotheses. In one case, we compare one exposed with several reference groups, while in the other two cases, we compare several exposed groups with one reference group. We also provide a multiple stage testing approach. Data from studies in cancer and trauma epidemiology illustrate the procedures.

Patients with concomitant ovarian cancer and bowel obstruction were studied in an effort to find objective prognostic factors predictive of patient outcome. A total of 62 patients were followed from 31 to 354 weeks, and a total of 20 variables were considered in the analyses. At the end of the study 49 patients were dead of their disease, and 13 were alive (six disease free and seven with persistent disease). Survival probabilities of the sample were 79% at 6 weeks, 48% at 20 weeks, and 24% at 104 weeks. Univariate analyses revealed no significant difference in the survival times of medically versus surgically treated patients; age greater than 60 years at diagnosis of cancer, presence of ascites, low serum albumin levels, elevated blood urea nitrogen levels, elevated alkaline phosphatase levels, lack of previous radiotherapy (p less than 0.002 for all), advanced tumor stage, normal/ileus x-ray results, and a short diagnosis to obstruction interval (p less than 0.04 for all) resulted in lower survival probabilities.

Three hundred twelve diabetic children and adolescents were seen in our diabetic clinic and instructed to test their capillary blood glucose (CBG) twice daily and to use an algorithm to adjust their short-acting insulin. Of this group, 219 youngsters had a full 3-yr period of observation. At each clinic visit, blood was obtained for fasting blood glucose and HbA1 and, once a year, cholesterol and triglycerides were also measured. Patient and parent accuracy in measuring CBG was found to be adequate. The changes over time in HbA1 were nondifferential across age and sex, and there was no difference in the level of HbA1 between age and sex groups, the number of tests reported to have been done by the patients, the number of injections of insulin per day, or the serum cholesterol. There was a significant relationship between the HbA1 and the fasting blood glucose (P less than .001) measured by the laboratory as well as with the serum triglyceride (P less than .01). The failure to improve diabetic control, despite measures that would have been expected to do so, was believed to relate more to a lack of compliance than to a flaw in the therapeutic approach. It was interesting to note that the adolescent patients in the study were in no worse control than the younger children in the group. Although better technical skills are available today to manage diabetes, the psychosocial approach to patient motivation requires improvement.

Since 1958 there has been a perception of excess illness in a rural Canadian population living downwind from two natural gas refineries, the emissions of which contain mostly sulfur dioxide but also hydrogen sulfide. To determine if there was an excess of adverse health outcomes in the population exposed (defined by place of residence), a health survey was undertaken in 1985 in this area and in one unexposed to emissions but demographically similar. Participation was 92% from both the exposed population (n = 2,157) and a representative sample (n = 839) of the main reference population. More respiratory symptoms were reported in the exposed group than in the non-exposed group among those 5 to 13 yrs of age (28% versus 18%) and among never-smokers greater than or equal to 14 yrs of age (35% versus 24%). FEV1, FVC, and FEV1/FVC were similar in both areas. Dichotomizing the level of exposure (high, low) within the exposed area revealed a trend in the high exposure area toward increased respiratory symptoms in the younger age group (39% versus 24%), but decreased symptoms in the older age group (33% versus 36% among never-smokers). FEV1 was similar between the two areas. The excess of respiratory symptoms in the exposed area unassociated with impaired spirometric values would be compatible with increased awareness of health or a small biologic environmental effect.

The reliability of the 30-item Modified Rathus Assertiveness Schedule (MRAS) was examined using the test-retest method over a three-week period. The MRAS, which was administered to 103 white middle-class seventh graders, yielded correlations of .74 using the Pearson product and Spearman Brown correlation coefficient. The correlations for males yielded .77 and .72 using the Pearson Product and Spearman Brown techniques, respectively. For females the correlations for both tests were .72. Paired t tests by sex indicated no significant change in MRAS scores between test administrations. These findings add to the existing information on the reliability of the MRAS.

Behavior of a number of Quality of Life measures gathered from two samples of terminal cancer patients over the last weeks of their lives are reported. Samples represent patients in the 26 hospices participating in a nationwide U.S. demonstration project and patients in the palliative care units of two Montreal hospitals. The U.S. data reported are quality of life measures made by a lay principal care person (PCP) or trained interviewer; the Montreal measures were made by both an attending doctor and an attending nurse. The general finding, as expected, is one of increasing deterioration in quality of life, with accelerated deterioration between 3 and 1 week of death. Pain follows a somewhat different pattern than other measures. More patients are in either of the extreme categories at an earlier point in time than found for other measures, and there are fewer changes as death is approached. Finally, about 20% of the patients do not fall into extremely low quality of life categories, even in the week prior to death.