Alan P. Bender

Data on prior pesticide use collected from self- and proxy respondents

Proxy respondents have often been used in case-control studies of cancer and pesticides. To evaluate the effect of exposure misclassification, we compared data collected during 1981–1983 from participants interviewed for a case- control study of leukemia and non-Hodgkins lymphoma with data collected during 1990–1991 from proxy respondents for participants who died or became incompetent since the initial interview (328 self-proxy pairs). As questions increased in detail, agreement percentages decreased. Agreement percentages were highest for demographic and general farming information (averages = 88–90%) and lowest for specific pesticide use (averages = 68–74%). Generally, odds ratios calculated from proxy respondent data were less than those from self-respondent data; however, several exceptions occurred. The findings indicate that pesticide data provided by proxy respondents will not necessarily result in the same estimate of risk and/or lead to the same conclusions as data provided by self-respondents.

Minnesota population cancer risk

Background. The Minnesota Cancer Surveillance System (MCSS) provides information on the occurrence of newly diagnosed cancers among Minnesota residents. Cancer is a major cause of death and morbidity in older persons. Population cancer risk (PCR) was assessed as a measure of the number of cancers that will occur in the lifetime of 1000 persons.

A standard person-years approach to estimating lifetime cancer risk

Several methods have been used to estimate the lifetime probability of cancer, such as simple cumulative incidence or competing risk models. These methods are characterized by either simplistic assumptions or detailed computations. The standard person-years method, that parallels cohort analyses, offers a simpler and more accurate method of approximating the lifetime risks of cancer. Since lifetime cancer risk refers to the aggregate risk to a cohort rather than the risk to an individual, it is suggested that a new term, population cancer risk, be used in describing these risks. A reasonable definition for the population cancer risk is the expected number of cancers in the lifetimes of 1,000 people. Estimates of lifetime risks of cancer are interpreted best as a composite measure of the joint forces of cancer morbidity and all-cause mortality at a point in time. The overall population cancer risk (lifetime risk) for residents of Minnesota (United States) is calculated to be 465 cancers/1,000 people; national data provide similar results.

Feasibility study of a statewide pathology-based Cancer surveillance system in Minnesota: 1. information characteristics

The Minnesota Department of Health has completed a 2-year feasibility study comparing the completeness and accuracy of information from pathology-based cancer ascertainment with that of the traditional surveillance method based on hospital discharge records. Overall, for incident cancers, the primary site designation of the pathology-based system was correct for 94.5% of the cancers, and the histologic designation was correct for 97.0% of the cancers. For prevalent cancers the accuracy of both site and histology designation was inadequate at 81.0 and 76.8% respectively. Pathology-based ascertainment was more complete than discharge-based surveillance (98.4% vs. 96.6%), which reflected the growing number of cancers diagnosed in hospital outpatient departments and medical clinics. The major limitation of the pathology-based system was the inability to determine from written pathology reports whether the cancer was newly diagnosed. However, when asked, pathologists correctly determined the incidence status for approximately 75% of the cancers. In light of the results of the feasibility study, Minnesota is implementing a pathology-based system as a cost-effective, scientifically valid method to meet the states current and future needs for cancer surveillance.

C-peptide response to meal challenge in nondiabetic and diabetic adults living in Wadena, Minnesota

OBJECTIVE The goal of the study was to provide cross-sectional descriptive data on the response of C-peptide to a vigorous meal stimulus in a population-based sample of nondiabetic adults compared with a population-based sample of adults with NIDDM. Available information is scanty, especially in subjects > 50 yr old. RESEARCH DESIGN AND METHODS The group under study included 377 adults without previously known diabetes randomly chosen from the population of the city of Wadena, Minnesota, almost all of northern European background, and 88 adults with known diabetes. PCP was measured 90 min after ingestion of 480 ml liquid meal Ensure-Plus, which includes 95 g dextrose, 26 g protein, and 25 g fat. C-peptide also was measured in a 260-min urine collection after the meal challenge. Novo antibody M1221 was used for C-peptide assay throughout the study. Participants whose medical record indicated insulin-dependent diabetes with a history of acetone production were excluded from analyses. RESULTS The distribution of UCP and PCP in this group of subjects appears very broad. Both the highest and lowest values for C-peptide were observed in individuals with diabetic glucose tolerance. The mean and median values in the nondiabetic group are higher than in previously published reports. After statistical adjustment for age, sex, BMI, and concomitant plasma glucose, participants with IGT produced significantly more C-peptide than the group with NGT (3.48 vs. 2.96 nM PCP, P < 0.05). Participants with diabetic glucose tolerance and who were not taking insulin produced as much or more C-peptide than either the NGT or IGT groups, depending on the statistical model used for adjusting for plasma glucose. Diabetic participants who were taking insulin produced significantly lower amounts of C-peptide than any of the non-insulin-taking groups (∼ 30% of the C-peptide produced by the non-insulin-taking diabetic participants). A decline in PCP production with increasing years since diagnosis (5.7%/yr) was observed exclusively in the insulin-taking NIDDM participants. Effect modification by glucose tolerance classification was observed on the relationship between plasma glucose and PCP: PCP increased with increasing plasma glucose in NGT and IGT groups, but a nonsignificant negative relationship was exhibited in diabetic participants. CONCLUSIONS The data suggest that two forms of NIDDM may exist, crudely distinguished by the clinical decision to use insulin to control blood glucose levels. The insulin-taking diabetic individuals may experience a greater likelihood of pancreatic failure, whereas non-insulin-taking diabetic individuals probably experience stable pancreatic function over the course of their disease. Longitudinal observation of the Wadena cohort will provide more insight into this possibility.

A nonparametric approach for determining significance of county cancer rates compared to the overall state rate: illustrated with Minnesota data

BackgroundThe study of the geographical distribution of disease has expanded greatly with GIS technology and its application to increasingly available public health data. The emergence of this technology has increased the challenges for public health practitioners to provide meaningful interpretations for county-based state cancer maps.MethodsOne of these challenges—spurious inferences about the significance of differences between county and overall state cancer rates—can be addressed through a nonparametric statistical method. The Wilcoxon’s signed rank test (WSRT) has a practical application for determining the significance of county cancer rates compared to the statewide rate. This extension of the WSRT, developed by John Tukey, forms the basis for constructing a single confidence interval for all differences in county and state directly age-adjusted cancer rates. Empirical evaluation of this WSRT application was conducted using Minnesota cancer incidence data.ResultsThe WSRT procedure reduced the impact of statistical artifacts that are frequently encountered with standard normal significance testing of the difference between directly age-adjusted county and the overall state cancer rates.ConclusionAlthough further assessment of its performance is required, the WSRT procedure appears to be a useful complement for mapping directly age-adjusted state cancer rates by county.