PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 12369083

Interval estimation for Poisson capture-recapture models in epidemiology.

C P Farrington
View original source ↗Export metadata ↓

Abstract

Capture-recapture studies in epidemiology are frequently undertaken to adjust underlying rates for underreporting. This contrasts with applications in ecology, in which the aim is to estimate population size. In consequence, the Poisson model provides a more appropriate framework than the multinomial in some epidemiological applications. A method for constructing profile Poisson likelihood interval estimates using log-linear modelling, avoiding the need for constrained optimization, is described for several relevant parameters, including the unobserved disease rate, the total disease rate, and the case ascertainment probability. Asymptotic properties of these estimates are studied and contrasted with those derived from the multinomial model. The method is illustrated with a published data set on spina bifida.

Explore related subjects

Epidemiologic MethodsHumansModels, StatisticalNew YorkPoisson DistributionPrevalenceSpinal DysraphismJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

C P Farrington. 2002-10-30. Interval estimation for Poisson capture-recapture models in epidemiology.. https://doi.org/10.1002/sim.1223

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Cancer mortality in a synthetic spinning plant in Besançon, France.

OBJECTIVES: To assess the mortality of a cohort of workers in a synthetic textile spinning plant and to evaluate the relationship between mortality from lung, liver and bladder cancer and the processes or the products used. METHODS: The study population consisted of male workers present for at least 6 months in the plant from 1968 to 1984. The cohort was followed until 1999. Vital status and the causes of death were determined by consulting national registries. The population of the Franche-Comté region was used for comparison. In total, 17 groups of exposure were assessed by the industrial hygienist, based on the consensus of an expert group that determined the exposure levels of each job to selected occupational hazards. Each worker was assigned to one or several groups, according to his occupational history. Confounding factors could not be assessed. Standardised mortality ratios (SMR) and 95% bilateral confidence intervals were calculated based on an assumed Poisson distribution of the number of cases to compare the plant mortality and the population mortality. Internal analyses were performed with Cox models in order to assess the risks of death related to the various exposures. RESULTS: In the whole cohort, mortality from all malignant neoplasms was lower than expected, but this was not significant. All the estimated SMRs were lower than or close to 1. The "hot -line fitters" (RR = 2.13; n = 9; 1.06 to 4.29) and the "fibre-drawing workers" (RR = 1.83; n = 20;1.09 to 3.07) experienced a statistically significant excess in mortality from lung cancer. A slightly elevated but not significant risk of death related to lung cancer (RR = 1.5; n = 41; 0.8 to 2.7) was observed in the groups with the highest exposure to mineral fibres. A statistically significant increase in cancer deaths was observed for workers with high exposure to dust (higher intensity: RR = 1.42; n = 79; 1.06 to 1.89). CONCLUSION: Some findings, mainly of lung cancer, justify further exploration in other plants in this industry.

Epidemiologic Methods↗
PubMed2007

Mortality in Iraq.

Explore the source record for details and available documents.

Epidemiologic Methods↗