Aspirin but not ibuprofen use is associated with reduced risk of prostate cancer: a PLCO study.

BACKGROUND: Although most epidemiological studies suggest that non-steroidal anti-inflammatory drug use is inversely associated with prostate cancer risk, the magnitude and specificity of this association remain unclear. METHODS: We examined self-reported aspirin and ibuprofen use in relation to prostate cancer risk among 29 450 men ages 55-74 who were initially screened for prostate cancer from 1993 to 2001 in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Men were followed from their first screening exam until 31 December 2009, during which 3575 cases of prostate cancer were identified. RESULTS: After adjusting for potential confounders, the hazard ratios (HRs) of prostate cancer associated with <1 and ≥ 1 pill of aspirin daily were 0.98 (95% confidence interval (CI), 0.90-1.07) and 0.92 (95% CI: 0.85-0.99), respectively, compared with never use (P for trend 0.04). The effect of taking at least one aspirin daily was more pronounced when restricting the analyses to men older than age 65 or men who had a history of cardiovascular-related diseases or arthritis (HR (95% CI); 0.87 (0.78-0.97), 0.89 (0.80-0.99), and 0.88 (0.78-1.00), respectively). The data did not support an association between ibuprofen use and prostate cancer risk. CONCLUSION: Daily aspirin use, but not ibuprofen use, was associated with lower risk of prostate cancer risk.

Relation between policies and work related assault: Minnesota Nurses' Study.

AIMS: To assess the relation between violence prevention policies and work related assault. METHODS: From Phase 1 of the Minnesota Nurses' Study, a population based survey of 6300 Minnesota nurses (response 79%), 13.2% reported experiencing work related physical assault in the past year. In Phase 2, a case-control study, 1900 nurses (response 75%) were questioned about exposures relevant to violence, including eight work related violence prevention policy items. A comprehensive causal model served as a basis for survey design, analyses, and interpretation. Sensitivity analyses were conducted for potential exposure misclassification and the presence of an unmeasured confounder. RESULTS: Results of multiple regression analyses, controlling for appropriate factors, indicated that the odds of physical assault decreased for having a zero tolerance policy (OR = 0.5, 95% CI 0.4 to 0.8) and having policies regarding types of prohibited violent behaviours (OR = 0.5, 95% CI 0.3 to 0.9). Analyses adjusted for non-response and non-selection resulted in wider confidence intervals, but no substantial change in effect estimates. CONCLUSIONS: It appears that some work related violence policies may be protective for the population of Minnesota nurses.

An epidemiological study of the magnitude and consequences of work related violence: the Minnesota Nurses' Study.

AIMS: To identify the magnitude of and potential risk factors for violence within a major occupational population. METHODS: Comprehensive surveys were sent to 6300 Minnesota licensed registered (RNs) and practical (LPNs) nurses to collect data on physical and non-physical violence for the prior 12 months. Re-weighting enabled adjustment for potential biases associated with non-response, accounting for unknown eligibility. RESULTS: From the 78% responding, combined with non-response rate information, respective adjusted rates per 100 persons per year (95% CI) for physical and non-physical violence were 13.2 (12.2 to 14.3) and 38.8 (37.4 to 40.4); assault rates were increased, respectively, for LPNs versus RNs (16.4 and 12.0) and males versus females (19.4 and 12.9). Perpetrators of physical and non-physical events were patients/clients (97% and 67%, respectively). Consequences appeared greater for non-physical than physical violence. Multivariate modelling identified increased rates for both physical and non-physical violence for working: in a nursing home/long term care facility; in intensive care, psychiatric/behavioural or emergency departments; and with geriatric patients. CONCLUSIONS: Results show that non-fatal physical assault and non-physical forms of violence, and relevant consequences, are frequent among both RNs and LPNs; such violence is mostly perpetrated by patients or clients; and certain environmental factors appear to affect the risk of violence. This serves as the basis for further analytical studies that can enable the development of appropriate prevention and control efforts.

Metabolites of a tobacco-specific lung carcinogen in the urine of elementary school-aged children.

Limited data are available in the literature on carcinogen uptake by children exposed to environmental tobacco smoke (ETS). In this study, we quantified metabolites of the tobacco-specific lung carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) in the urine of elementary school-aged children participating in the School Health Initiative: Environment, Learning, Disease study, a school-based investigation of the environmental health of children. The metabolites of NNK are 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and its glucuronide (NNAL-Gluc). We also measured cotinine and its glucuronide (total cotinine). Urine samples were collected from 204 children. Seventy (34.3%) of these had total cotinine > or =5 ng/ml. NNAL or NNAL-Gluc was detected in 52 of 54 samples with total cotinine > or =5 ng/ml and in 10 of 20 samples with total cotinine < 5 ng/ml. Levels of NNAL plus NNAL-Gluc and total cotinine were significantly higher when exposure to ETS was reported than when no exposure was reported. However, even when no exposure to ETS was reported, levels of NNAL, NNAL-Gluc, and NNAL plus NNAL-Gluc were higher than in children with documented low exposure to ETS, as determined by cotinine levels < 5 ng/ml. Levels of NNAL, NNAL-Gluc, and cotinine were not significantly different in samples collected twice from the same children at 3-month intervals. Levels of NNAL plus NNAL-Gluc in this study were comparable with those observed in our previous field studies of adults exposed to ETS. There was a 93-fold range of NNAL plus NNAL-Gluc values in the exposed children. The results of this study demonstrate widespread and considerable uptake of the tobacco-specific lung carcinogen NNK in this group of elementary school-aged children, raising important questions about potential health risks. Our data indicate that objective biomarkers of carcinogen uptake are important in studies of childhood exposure to ETS and cancer later in life.

The effect of fecal occult-blood screening on the incidence of colorectal cancer.

BACKGROUND: Both annual testing for fecal occult blood and biennial testing significantly reduce mortality from colorectal cancer. However, the effect of screening on the incidence of colorectal cancer remains uncertain, despite the diagnosis and removal of precancerous lesions in many persons who undergo screening. METHODS: We followed the participants in the Minnesota Colon Cancer Control Study for 18 years. A total of 46,551 people, most of whom were 50 to 80 years old, were enrolled between 1975 and 1978 and randomly assigned to annual screening, biennial screening, or usual care (the control group). Those assigned to the screening groups were asked to prepare and submit two samples from each of three consecutive stools for guaiac-based testing. Those with at least one positive slide in the set of six were offered a diagnostic examination that included colonoscopy. Screening was conducted between 1976 and 1982 and again between 1986 and 1992. Study participants have been followed with respect to newly diagnosed cases of colorectal cancer and deaths. Follow-up has been more than 90 percent complete. RESULTS: During the 18-year follow-up period, we identified 1359 new cases of colorectal cancer: 417 in the annual-screening group, 435 in the biennial-screening group, and 507 in the control group. The cumulative incidence ratios for colorectal cancer in the screening groups as compared with the control group were 0.80 (95 percent confidence interval, 0.70 to 0.90) and 0.83 (95 percent confidence interval, 0.73 to 0.94) for the annual-screening and biennial-screening groups, respectively. For both screening groups, the number of positive slides was associated with the positive predictive value both for colorectal cancer and for adenomatous polyps at least 1 cm in diameter. CONCLUSIONS: The use of either annual or biennial fecal occult-blood testing significantly reduces the incidence of colorectal cancer.

A school-based strategy to assess children's environmental exposures and related health effects in economically disadvantaged urban neighborhoods.

The School Health Initiative: Environment, Learning, Disease (SHIELD) study is a novel school-based investigation of children's environmental health in economically disadvantaged urban neighborhoods of Minneapolis. This article describes the study design and summarizes lessons learned about recruiting and monitoring this historically understudied population. The SHIELD study focused on measuring children's exposures to multiple environmental stressors [volatile organic chemicals (VOCs), environmental tobacco smoke, allergens, bioaerosols, metals, pesticides, polychlorinated biphenyls (PCB), phthalates] and exploring related effects on respiratory health (e.g., lung function) and learning outcomes (e.g., standardized test scores, academic achievement). It involved intensive exposure monitoring, including environmental measurements inside and outside the children's schools and inside their homes, personal measurements with passive dosimeters worn by the children, and biological marker measurements in blood and urine. The SHIELD participants comprised a stratified random sample of 153 "index" children and 51 of their siblings enrolled in grades 2-5 at two adjacent elementary schools. The Minneapolis Public Schools (MPS) assisted with identifying, contacting, recruiting, and monitoring this population, which traditionally is difficult to study because families/children are highly mobile, speak a diversity of languages, frequently do not have a telephone, endure economic hardships, often do not trust researchers, and have a spectrum of unconventional lifestyles and living arrangements. Using a school-based approach, the overall SHIELD enrollment (response) rate was 56.7%, with a wide disparity between English-speaking (41.7%) and non-English-speaking (71.0%) families/children. Most children remained involved in the study through both monitoring sessions and exhibited an acceptable degree of compliance with study protocols, including providing blood and urine samples. Results indicate that it is both practical and affordable to conduct probability-based exposure studies in this population, but that it is also important to improve our understanding of factors (e.g., cultural, economic, psychological, social) affecting the willingness of families/children to participate in such studies, with special emphasis on developing cost-effective recruitment methods.

A novel form of ascertainment bias in case-control studies of cancer screening.

In case-control studies of cancer screening, some have generally admonished investigators against case definitions based on diagnosis dates because of lead-time bias. However, perhaps partly due to vagueness, the admonitions have been frequently ignored. A recurrence-time model simulates case ascertainment when diagnosis must occur within a specific calendar period. The model depends on screening test sensitivity and rate, age-specific preclinical incidence rates, and preclinical duration time and survival time distributions. For one study of sigmoidoscopic screening for colorectal cancer, when the true odds ratio is 1, its estimate is 0.50 to 0.75 under plausible assumptions. This bias can affect any observational study wherein case definition depends on diagnosis times (e.g., health-plan enrollment data). To avoid bias in observational investigations of cancer screening wherein the case definition depends on the diagnosis date, one must ensure that both screening and preclinical incidence do not occur before the case definition period.

Colorectal cancer deaths as determined by expert committee and from death certificate: a comparison. The Minnesota Study.

For deaths during the first 13 years of follow-up of the Minnesota Colon Cancer Control Study, an expert committee using numerous medical documents, and a nosologist using only the death certificate independently determined whether colorectal cancer caused the death and, if not, whether the disease was present at death. Deaths due to colorectal cancer numbered 318 according to the nosologist and 323 according to the committee, a discrepancy of 1.5%, which is similar in magnitude to that in three previous studies. The nosologist and committee agreed that colorectal cancer caused the death in each of 290 individual cases; they disagreed widely on the number of deaths from other causes but with colorectal cancer. If it is important to know only the gross number of deaths from colorectal cancer, then the death certificate alone appears to be sufficiently accurate; if it is important to know the cause of death of individual subjects or the number dying from other causes but with colorectal cancer, then the expert committee method provides more accurate information.

Colorectal cancer mortality: effectiveness of biennial screening for fecal occult blood.

BACKGROUND: In 1993, a randomized controlled trial in Minnesota showed, after 13 years of follow-up, that annual fecal occult blood testing was effective in reducing colorectal cancer mortality by at least 33%. Biennial screening (i.e., every 2 years) resulted in only a 6% mortality reduction. Two European trials (in England and in Denmark) subsequently showed statistically significant 15% and 18% mortality reductions with biennial screening. Herein, we provide updated results-through 18 years of follow-up--from the Minnesota trial that address the apparent inconsistent findings among the trials regarding biennial screening. METHODS: From 1976 through 1977, a total of 46551 study subjects, aged 50-80 years, were recruited and randomly assigned to an annual screen, a biennial screen, or a control group. A screen consisted of six guaiac-impregnated fecal occult blood tests (Hemoccult) prepared in pairs from each of three consecutive fecal samples. Participants with at least one of the six tests that were positive were invited for a diagnostic examination that included colonoscopy. All participants were followed annually to ascertain incident colorectal cancers and deaths. RESULTS: The numbers of deaths from all causes were similar among the three study groups. Cumulative 18-year colorectal cancer mortality was 33% lower in the annual group than in the control group (rate ratio, 0.67; 95% confidence interval [CI] = 0.51-0.83). The biennial group had a 21% lower colorectal cancer mortality rate than the control group (rate ratio, 0.79; 95% CI = 0.62-0.97). A marked reduction was also noted in the incidence of Dukes' stage D cancers in both screened groups in comparison with the control group. CONCLUSION: The results from this study, together with the other two published randomized trials of fecal occult blood screening, are consistent in demonstrating a substantial, statistically significant reduction in colorectal cancer mortality from biennial screening.

Do vouchers improve breast cancer screening rates? Results from a randomized trial.

OBJECTIVE: To determine the effect of a voucher for free mammography on compliance with recommended mammography screening guidelines. STUDY DESIGN: Vouchers for free mammography distributed to a random sample of women over the age of 50 in two rural southern Minnesota counties. The vouchers were good for one year. Baseline and follow-up data were collected and rates of compliance with current mammography guidelines were observed for the voucher group and a control group of women living in the same counties. METHODS: Logistic regression models were used to estimate the effect of the voucher on compliance with mammography guidelines and the impact of factors potentially influencing the effectiveness of the voucher. PRINCIPAL FINDINGS: The voucher improved mammography rates primarily through increasing screening among women who were out of compliance at baseline. CONCLUSIONS: Vouchers, even when distributed randomly within a population of rural Midwestern women, can significantly improve compliance rates. Vouchers are no less effective a means of increasing screening among vulnerable women than among other women.

Fecal occult blood screening in the Minnesota study: role of chance detection of lesions.

BACKGROUND: In the Minnesota Colon Cancer Control Study, annual fecal occult blood testing reduced mortality from colorectal cancer by at least 33.4%. Some attribute a large part of this reduction to chance detection of cancers by colonoscopies; rehydration of guaiac test slides greatly increased positivity and consequently the number of colonoscopies performed. This study was conducted to determine how much of the reduction resulted from chance detection. METHODS: We used a mathematical model developed by Lang and Ransohoff to estimate the proportion of the 33.4% mortality attainable by chance alone. Applying the model requires the specification of five parameters: duration of follow-up, rate of compliance with fecal occult blood testing, rate of compliance with colonoscopy, positivity rate, and efficacy of colonoscopy in reducing colorectal cancer mortality. We took values for four of the five parameters directly from the Minnesota study. For the fifth parameter, efficacy of colonoscopy, we selected a value of 60%, based on the conclusions of another study. Whereas the Lang-Ransohoff model selects persons for colonoscopy by chance alone, those with bleeding cancers would also be selected by sensitive fecal occult blood testing. We therefore adjusted the result of the Lang-Ransohoff model for this dual detectability. RESULTS: We found that 16%-25% of the reduction in colorectal cancer deaths effected by fecal occult blood testing in the Minnesota study was due to chance detection; the remainder was due to sensitive detection. CONCLUSION: Chance played a minor role in the detection of colorectal cancers by fecal occult blood testing in the Minnesota study.

Fecal occult blood screening in the Minnesota study: sensitivity of the screening test.

BACKGROUND: In the Minnesota Colon Cancer Control Study, which used guaiac slides to annually screen stool samples for blood, mortality from colorectal cancer was reduced by 33.4%. The reported sensitivity of this test for colorectal cancer was about 90%. However, results from another study estimated the sensitivity to be 25%-33%; other investigators have reported intermediate values. Given these contradictions, we examined screening sensitivity for colorectal cancer in the Minnesota study by several direct and indirect methods. METHODS: In this reanalysis of data from the Minnesota study, we distinguished between sensitivity for colorectal cancer of the screening test (composed of six slides) and of the screening program (a series of such tests). We estimated screen sensitivity by adjusting the crude estimate from the final tests in each screening phase for colorectal cancer incidence in 5 years of follow-up, by modeling guaiac slide results at each screen as a function of the presence of occult blood, and by incorporating sensitive detection into a modification of a mathematical model developed by Lang and Ransohoff. Program sensitivity was estimated from the fraction of screen-detected cancers among all cancers diagnosed in screened individuals. RESULTS: The crude estimate of program sensitivity was 89.4%, whereas the modified Lang-Ransohoff model estimates screen sensitivities at 94.1%-96.2%, consistent with the estimates from the other methods. Indirect measures, such as the association between the number of positive slides among the six slides in each set and the positive predictivity for colorectal cancer, are consistent with these estimates. CONCLUSIONS: The Minnesota study reduced mortality from colorectal cancer through use of a screening test with average screen and program sensitivities of about 90%.

Risk assessment and risk stratification in sudden cardiac death: a biostatistician's view.

Determining individual probabilities of developing lethal arrhythmia over time (risk assessment) and grouping individuals by that probability (risk stratification) are similar to, yet differ in purpose from, screening, diagnosis, risk factor identification, and prognostic staging. Methods of handling bias, use of multiple predictors, and evaluation of results provide challenges. A key purpose of risk assessment and stratification is examined. The role of operational definitions of predictors and events and of methods that account for multiple predictors and known confounding factors is analyzed. Constructed examples illustrate potential pitfalls in assessment and how multivariate techniques can deal with multiple predictors. A trial design to evaluate risk stratification for the identified purpose is elaborated and potential results are interpreted. Bias from predictors regressing to the mean can be minimized either by averaging a number of measurements or by equalizing the bias in comparison groups. An analysis of two predictors and two risk strata illustrates how the discrimination of combined predictors may be greater than the sum of the individual variables' discrimination. Risk stratification can be evaluated in trials that randomize competing interventions within different risk strata. Results of such trials indicate whether the risk strata adequately distinguish individuals by their responsiveness to particular intervention. Potential pitfalls, not easily recognized in risk stratification, can be avoided in the methods and in studies for evaluating those methods. Multivariate techniques maximize the discrimination of multiple predictors, but may increase complexity. Randomized trials of treatment provide evidence for utility of risk stratification.

Derivation of a defibrillator implant criterion based on probability of successful defibrillation.

Common criteria for implant of a cardioverter defibrillator include verification of a 2:1 energy safety margin or a fixed safety margin of 10 joules. These criteria have been established empirically. We present a statistically model based on defibrillation efficacy curves which may be used to establish a criterion which would meet a predetermined target. As an example, an implant criterion is derived based on a goal of 1-year sudden cardiac death survival of at least 99% by selecting an expected first-shock efficacy to meet that target. Logistic regression was performed on data from over 1,500 defibrillator implants including successful epicardial and transvenous electrode system implants as well as data from unsuccessful implants. A random sample from these curves was used to generate a representative sample of 1,000 potential implant candidates. By assuming successful defibrillation using a series of shocks at specified energies, i.e., choosing an implant criterion, the probability of successful defibrillation of the patient by a single shock at a predetermined maximum output can be established. Independent data are used to validate the model's accuracy in predicting defibrillation efficacy within the derived example.

Computation modes of multivariate positive predictive characteristics.

Multivariate receiver operator characteristics (ROCs) and positive predictive characteristics (PPCs), based on a combination of two or more clinical variables, are usually computed by varying dichotomy limits for each variable independently. This approach has a similar number of degrees of freedom (i.e., uses the same number of programmable parameters) as the approach which defines test positive cases based on a linear combination of all the clinical variables involved. Either approach can be implemented without any assumption about the underlying probability distributions by using an exhaustive computer search. Both approaches were compared in a demonstration study of predicting 2-year all cause mortality after acute myocardial infarction, based on applying various time- and spectral-domain indices of signal-averaged ECGs from a research survey. The results showed that the optimum mode for the computation of ROCs and PPCs depends on the character of the particular data used. Therefore, in order to increase the precision of the retrospective multifactoral studies, both approaches to ROC and PPC computation should be used and compared in each individual investigation.

Mammogram utilization among farm women.

Utilization of preventive health care services is lower in rural populations than in urban populations, possibly as a result of barriers to preventive health care that are characteristic of rural settings. This study was conducted to identify factors associated with mammogram utilization among farm women. Mammogram utilization among farm women from six southern Minnesota counties was examined as part of a larger community-based cancer intervention study. Farm women aged 40 and older were randomly selected from a list of farm households and interviewed by telephone to determine mammogram utilization and factors related to utilization. Of the 606 respondents, 78 percent reported ever having a mammogram and 49 percent reported a mammogram within the past year. Physician recommendation for a screening mammogram and family history of breast cancer were found to be associated with ever having a mammogram. Correct knowledge of mammogram screening guidelines was associated with a mammogram within the past 12 months. Overall, physician recommendation was the most influential determinant of utilization. As more emphasis is placed on prevention, patient education by physicians can have the greatest impact on mammogram utilization.

Longitudinal compliance with annual screening for fecal occult blood. Minnesota Colon Cancer Control Study.

In a randomized, controlled trial of fecal occult blood screening for colorectal cancer, the Minnesota Colon Cancer Control Study carried out 11 annual mail screens in two phases between 1976 and 1992. This long-term compliance record of 15,476 individuals is summarized and related to demographic characteristics as well as to the screening experience of the participants. There was a strong and consistent effect of age, with peak compliance among participants about 70 years old, and lower compliance among the youngest (< or = 55 years) and oldest (> or = 80 years) participants. There was a significantly higher rate of screen compliance among participants who lived with other participants, compared with households where only one individual participated in the study. Finally, participants who underwent a diagnostic colorectal examination with negative results had significantly lower odds of compliance.

Reducing mortality from colorectal cancer by screening for fecal occult blood. Minnesota Colon Cancer Control Study.

BACKGROUND: Although tests for occult blood in the feces are widely used to screen for colorectal cancers, there is no conclusive evidence that they reduce mortality from this cause. We evaluated a fecal occult-blood test in a randomized trial and documented its effectiveness. METHODS: We randomly assigned 46,551 participants 50 to 80 years of age to screening for colorectal cancer once a year, to screening every two years, or to a control group. Participants who were screened submitted six guaiac-impregnated paper slides with two smears from each of three consecutive stools. About 83 percent of the slides were rehydrated. Participants who tested positive underwent a diagnostic evaluation that included colonoscopy. Vital status was ascertained for all study participants during 13 years of follow-up. A committee determined causes of death. A single pathologist determined the stage of each tissue specimen. Differences in mortality from colorectal cancer, the primary study end point, were monitored with the sequential log-rank statistic. RESULTS: The 13-year cumulative mortality per 1000 from colorectal cancer was 5.88 in the annually screened group (95 percent confidence interval, 4.61 to 7.15), 8.33 in the biennially screened group (95 percent confidence interval, 6.82 to 9.84), and 8.83 in the control group (95 percent confidence interval, 7.26 to 10.40). The rate in the annually screened group, but not in the biennially screened group, was significantly lower than that in the control group. Reduced mortality in the annually screened group was accompanied by improved survival in those with colorectal cancer and a shift to detection at an earlier stage of cancer. CONCLUSIONS: Annual fecal occult-blood testing with rehydration of the samples decreased the 13-year cumulative mortality from colorectal cancer by 33 percent.

Sample sizes for prevention trials have been too small.

Planners of several large prevention trials have overestimated the expected incidence of events in the control group, largely because they failed either to recognize or to adequately correct for various effects of population selection. Consequently, the studies have been too small in size or too short in duration to achieve their stated objectives. The selection effects include those engendered by the choice of the target population, the self-selection of volunteers, and protocol exclusions. This paper presents a taxonomy of these effects and the likely direction of their influence on the incidence of events and on mortality rates from other causes. Little information is available to help sample-size planners in adjusting for these effects. A few studies have provided information on the extent to which control group incidence rates have fallen short of expectations. In particular, researchers from the University of Minnesota's Colon Cancer Control Study have provided a detailed comparison of event incidence and all-cause mortality rates with general population rates. (AM J Epidemiol 1993;137:797-810). Other studies should publish similarly detailed information to assist sample-size planners of prevention trials. Until more information is published, this paper provides preliminary guidelines for prevention trial sample-size planning.