Time-space trends in cancer incidence in the Netherlands in 1989-2003.

Incidence of cancer may vary within a country and over time because of previous differences in exposure to risk factors or interventions for early detection (screening). This study describes time-space trends of incidence of common cancer sites across the Netherlands during the period 1989-2003 and speculates on the reasons for the observations. From the Netherlands Cancer Registry, World standardized incidence rates per municipality were smoothed calculating weighted averages for each 2 km by 2 km grid of the population mid-points of neighbouring municipalities and presented as map animations. Spatial relative changes in incidence were estimated by comparing the periods 1989-1994 and 1998-2003. Complete time-space trends can be found as map animations on http://maps.ikcnet.nl. The incidence of cervical and stomach cancer (for both sexes) decreased, being higher in the cities than in the rural areas during all periods and contrasting the trends in colorectal and breast cancer. The relative increase in incidence of lung cancer among females was highest in the rural north, but the incidence remained higher in the cities of the mid-west Netherlands. For males, there was a marked decrease in lung cancer incidence across the country since 1991. Incidence of melanoma increased, rates being twice as high in the coastal area than in the cities. Prostate cancer maps largely replicated the known history of PSA-testing in the Netherlands. Time-space cancer incidence patterns gave insight into effects of changes in exposure to risk determinants and early detection. The maps illustrate marked potential for cancer prevention at the national and regional level.

Misclassification of exposure is high when interview data on drug use are used as a proxy measure of chronic drug use during follow-up.

BACKGROUND AND OBJECTIVE: In many observational studies, the association between drugs and disease is analyzed with information from a baseline interview. We investigated the magnitude and direction of exposure misclassification by comparing interview data at baseline with prospectively gathered pharmacy data. METHODS: The study population for this study consisted of a cohort of 2,487 participants aged 71 years or older from the Rotterdam Study. Data on drug use were gathered at the baseline interview and through pharmacies during the follow-up period between January 1, 1991, and January 1, 1999. We assessed the sensitivity, specificity, and positive and negative predictive value of interview data as proxy measures of chronic use of calcium channel blockers (CCB) in comparison with longitudinal medication records from the pharmacy. RESULTS: Only 3 of the 206 subjects (1.5%) who reported use at baseline did not use CCBs during follow-up. Of the 2,281 persons who reported no use of CCBs at baseline, however, 354 actually used CCBs during follow-up (15.5%). The difference between interview data and pharmacy records corresponded to a misclassification bias of 0.73 (95%CI: 0.52-1.02). CONCLUSION: Misclassification of exposure was high when interview data were used as a proxy measure of chronic use during follow-up.

Prescription medications associated with a decreased risk of non-Hodgkin's lymphoma.

Earlier epidemiologic studies have suggested an inverse association between non-Hodgkin's lymphoma and exposure to histamine(2) (H(2)) blockers, nonsteroidal anti-inflammatory drugs, cholesterol-lowering drugs, and antibiotics. Data from the PHARMO database were used to conduct a nested, population-based case-control study that included 1985-1998 drug-dispensing records for 300,000 residents of six Dutch cities. Included were those subjects without a previous history of cancer who were aged >/=20 years and were registered with an incident primary discharge diagnosis of non-Hodgkin's lymphoma between 1991 and 1998. This paper includes data on 211 cases and 800 controls individually matched on sex, age, community pharmacy, calendar time, and duration of follow-up. Conditional logistic regression analysis was used to evaluate the association between non-Hodgkin's lymphoma and categories of cumulative drug use in days. In multivariate analyses, nonsignificant risk reductions were found for all drugs tested, and the negative association tended to increase with increasing duration of use. For women, the odds ratio for H(2) blockers was 0.29 (95% confidence interval: 0.12, 0.69) and for analgesics was 0.40 (95% confidence interval: 0.22, 0.71). Results support an inverse association between occurrence of non-Hodgkin's lymphoma and use of H(2) blockers and analgesics among women, and they warrant confirmation in larger studies.