James A. Hanley

Streptozocin Alone Compared with Streptozocin plus Fluorouracil in the Treatment of Advanced Islet-Cell Carcinoma

To evaluate the treatment of advanced islet-cell carcinoma, we randomly assigned 84 patients to streptozocin alone or streptozocin plus fluorouracil. Each regimen was given in five-day courses. The most frequent toxic effects were nausea and vomiting, mild and reversible renal toxicity, and bone-marrow depression with the combination regimen. The combination had advantages over streptozocin alone in overall rate of response (63 vs. 36 per cent) and in rates of complete response (33 vs. 12 per cent). There was no evidence of a preferential response among types of functional tumors. Objective responses were generally of long duration (median, 17 months) and of substantive clinical benefit. Treatment with the combination also yielded a survival advantage over treatment with streptozocin alone (medians, 26 and 16 1/2 months), but this difference is not statistically significant. In spite of gastrointestinal side effects, streptozocin combined with fluorouracil appears to be a valuable treatment for advanced islet-cell carcinoma.

Statistical Approaches to the Analysis of Receiver Operating Characteristic (ROC) Curves

In this article we review published and some unpublished work in statistical analyses of ROC curves. We describe both single and joint indices and indicate the approaches that have been taken to consider between-reader variations and correlations, within-reader variations, and variations and correlations between cases. We then discuss in detail a single index, the TP ratio at a fixed FP ratio (designated TPFP), or the FP ratio at a fixed TP ratio (designated FPTP). We show how to calculate confidence limits around any point on the curve; we further show, using the conventional Dorfman and Alf program and the jackknifing technique, how to calculate these confidence limits for multiple curves derived from the same sample of patients.

Problem of immortal time bias in cohort studies: example using statins for preventing progression of diabetes

Immortal time in observational studies can bias the results in favour of the treatment group, but it is not difficult to identify and avoid

The psychological and physical health of family members caring for an elderly person with dementia

Most elderly persons with dementia are cared for at home, usually by the spouse or an adult child. The objective of the present study was to determine whether there is an excess of psychological and physical health problems among family caregivers (CGs) of elderly persons with dementia. Data were obtained by interview from close family members of dementia patients (CGs), and from a comparison group made up of close family members of patients undergoing cataract surgery (non-caregivers, NCGs). CGs had significantly higher levels of depression and physical symptoms than NCGs. The association between caregiving and the health variables was stronger among subjects who were the patients spouse than among those who were the patients child. Furthermore, greater behavioral disturbance in the demented patient was associated with higher levels of morbidity in the CG. The results suggest that CGs might benefit from careful monitoring of their health status, and from greater access to specialized support services.

Random measurement error and regression dilution bias

Random measurement error is a pervasive problem in medical research, which can introduce bias to an estimate of the association between a risk factor and a disease or make a true association statistically non-significant. Hutcheon and colleagues explain when, why, and how random measurement error introduces bias and provides strategies for researchers to minimise the problem

The effect of measuring error on the results of therapeutic trials in advanced cancer

In this study, 16 experienced oncologists each measured 12 simulated tumor masses employing their usual clinical methods. Unknown to the oncologists, two pairs of these tumors were identical in size. This permitted a total of 64 measurement comparisons of the same investigator measuring the same size mass and 1920 comparisons of different investigators measuring the same size mass. If a 50% reduction in the product of perpendicular diameters is accepted as a criterion, the objective response rate due to measuring error alone was 7.8% by the same investigator and 6.8% by different investigators. If a 25% reduction criterion is used, the respective “placebo” response rates were 19% and 25%. In the clinical setting it is recommended that the 50% reduction criterion be employed and that the investigator should anticipate an objective response rate of 5 to 10% due to human error in tumor measurement.

Nicotine-dependence symptoms are associated with smoking frequency in adolescents

BACKGROUND Although many sociodemographic and psychosocial factors have been identified as related to adolescent smoking, few studies have examined the role of nicotine-dependence (ND) symptoms. The objective was to study the association between ND symptoms and smoking status among adolescents in the early stages of the smoking onset process. METHODS The McGill University Study on the Natural History of Nicotine Dependence is an ongoing 6-year prospective investigation of the natural history of ND among 1267 grade 7 students in ten Montreal high schools. The baseline response was 55.4%. Subjects for this cross-sectional analysis of baseline data, collected in 1999, included 241 past 3-month smokers (mean age [SD]=13.0+/-0.7 years at baseline). ND symptoms were measured in five indicators, including a measure based on the criteria for tobacco dependence in the International Classification of Diseases-10th Revision (ICD-10), the Hooked on Nicotine Checklist, and three symptom clusters (withdrawal, self-medication, and ND/cravings symptoms). The association between ND symptom indicators and each of sporadic, monthly, weekly, and daily smoking relative to less frequent smoking was investigated in multiple logistic regression analysis. RESULTS Despite low cigarette exposure, 16.6% (95% confidence interval [CI], 11.9%-21.3%) of past 3-month smokers were tobacco dependent. The proportion increased from 0%, 3.1% (95% CI, 0.0%-9.2%), and 4.6% (95% CI, 0.2%-9.0%) among triers, sporadic smokers, and monthly smokers, respectively, to 19.4% (95% CI, 5.5%-33.3%) and 65.9% (95% CI, 51.9%-79.9%) among weekly and daily smokers, respectively. ND/cravings consistently distinguished each smoking category from less frequent smokers; the odds ratios (95% CI) for ND/cravings symptoms were 1.16 (0.99-1.35) in sporadic smokers; 1.17 (1.06-1.29) in monthly smokers; 1.34 (1.19-1.50) in weekly smokers; and 1.39 (1.22-1.59) in daily smokers. CONCLUSIONS These data challenge current smoking onset models, which suggest that ND develops only after several years of heavy or daily smoking. ND symptoms are associated, at least cross-sectionally, with increased smoking in adolescents. To increase the likelihood of being effective, tobacco-control programs for children and adolescents will need to take early ND symptoms into account.

Blood Pressure and Adiposity in Children and Adolescents

Background—Although obesity is associated with important hemodynamic disturbances, there are few data on population-wide blood pressure (BP) distribution in children and adolescents in this era of endemic pediatric obesity. Methods and Results—We conducted a school-based survey of a representative sample of youth aged 9, 13, and 16 years in Quebec, Canada. Resting BP was measured with an oscillometric device in 3589 subjects (80% response). Additional measures included height, weight, and subscapular and triceps skinfold thickness, an age-appropriate questionnaire, and a fasting blood draw. Mean (SD) systolic/diastolic BP (SBP/DBP) levels in 9-, 13-, and 16-year-olds were 103 (9)/57 (6), 113 (12)/58 (7), and 124 (14)/61 (7) mm Hg in males and 103 (10)/57 (6), 111 (11)/60 (7), and 114 (11)/62 (7) mm Hg in females. The prevalence of high-normal or elevated SBP was 12%, 22%, and 30% among 9-, 13-, and 16-year-old males, respectively, and 14%, 19%, and 17% among same-aged females. The prevalence of high-normal or elevated DBP was <1%. In multiple linear regression analysis, body mass index was consistently associated with SBP and DBP in all age-gender groups. Conclusions—Mean SBP and the prevalence of high-normal and elevated SBP are elevated in children and adolescents. Public policy, public health programs, and clinical preventive measures are urgently needed to address the obesity epidemic and its hemodynamic consequences.

A heuristic approach to the formulas for population attributable fraction

BACKGROUND As the definitional formula for population attributable fraction is not usually directly usable in applications, separate estimation formulas are required. However, most epidemiology textbooks limit their coverage to Levins formula, based on the (dichotomous) distribution of the exposure of interest in the population. Few present or explain Miettinens formula, based on the distribution of the exposure in the cases; and even fewer present the corresponding formulas for situations with more than two levels of exposure. Thus, many health researchers and public health practitioners are unaware of, or are not confident in their use of, these formulas, particularly when they involve several exposure levels, or confounding factors. METHODS/RESULTS A heuristic approach, coupled with pictorial representations, is offered to help understand and interconnect the structures behind the Levin and Miettinen formulas. The pictorial representation shows how to deal correctly with several exposure levels, and why a commonly used approach is incorrect. Correct and incorrect approaches are also presented for situations where estimates must be aggregated over strata of a confounding factor.

Nested case-control study of the effects of non-steroidal anti-inflammatory drugs on breast cancer risk and stage.

We carried out a nested case–control study to measure the rate ratio (RR) for invasive female breast cancer in relation to non-steroidal anti-inflammatory drug (NSAID) use. The source population consisted of the female beneficiaries of the Saskatchewan Prescription Drug Plan from 1981 to 1995 with no history of cancer since 1970. Four controls/case, matched on age and sampling time, were randomly selected. Dispensing rates during successive time periods characterized NSAID exposure. RRs associated with exposure during each period were adjusted for exposure during the others. Confounding by other determinants was studied in analyses adjusted with data obtained by interviewing samples of subjects accrued from mid-1991 to mid-1995. We accrued 5882 cases and 23 517 controls. Increasing NSAID exposure 2–5 years preceding diagnosis was associated with a trend towards a decreasing RR (P -trend = 0.003); for the highest exposure level RR = 0.76, 95% confidence interval 0.63–0.92. This protective effect could not be attributed to confounding by other determinants. In analyses involving only the cases, NSAID exposure 2–5 and 6–10 years preceding diagnosis was associated with significantly reduced risks of presenting with a large tumour (> 5 cm diameter) or distant metastasis, but not regional lymph node metastasis. The use of NSAIDs may retard the growth of breast cancers and prevent distant metastasis.