Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 15550597

An incidence density sampling program for nested case-control analyses.

D B Richardson
View original source ↗Export metadata ↓

Abstract

BACKGROUND: The nested case-control design can be a very efficient approach to an epidemiological investigation. In order to obtain unbiased estimates of relative risk, controls should be selected by incidence density sampling, which involves matching each case to a sample of those who are at risk at the time of case occurrence. METHODS: This paper presents a simple computer program for incidence density sampling. This program was evaluated using data derived from a cohort study of mortality among workers employed in the nuclear weapons industry. Controls were selected for cases via incidence density sampling; an estimate of the exposure-mortality association was obtained via conditional logistic regression. After 100 iterations of this procedure, the average effect estimate was compared to the risk estimate obtained via proportional hazards regression. The same methods were used to evaluate a program for incidence density sampling that was proposed previously by Pearce in 1989. RESULTS: Relative risk estimates obtained from nested case-control analyses conducted using the incidence density sampling program reported in this paper are unbiased. In contrast, the program for incidence density sampling proposed by Pearce tended to produce biased relative risk estimates; the magnitude of bias increased with increasing numbers of controls selected per case. CONCLUSIONS: The computer program described in this paper offers a simple approach to incidence density sampling for nested case-control analyses with exact matching on attained age and appropriate enumeration of the pool of eligible controls for each case. This method overcomes problems of bias inherent in a previously proposed program for incidence density sampling.

Explore related subjects

BiasCase-Control StudiesHumansIncidenceOccupational ExposureRisk AssessmentSampling StudiesSoftwareJournal ArticleResearch Support, U.S. Gov't, P.H.S.

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

D B Richardson. 2004. An incidence density sampling program for nested case-control analyses.. https://doi.org/10.1136/oem.2004.014472

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

KEEP EXPLORING

Related citations

PubMed2026

Assessment of blinding in pharmacotherapy and noninvasive neuromodulation randomized controlled trials for neuropathic pain in adults.

In randomized controlled trials (RCTs), study participants and research personnel are often blinded to minimize biases related to knowing treatment allocation. To determine if blinding was effective, participants may be asked which treatment they believe they received ("treatment guess"). This descriptive review characterized blinding assessment (BA) reporting in pharmacotherapy and neuromodulation neuropathic pain RCTs. Of 288 papers, 36 (12.5%) reported a BA. One paper reported the results of 2 studies, so in total 37 studies with a BA were assessed. Of these, 19 were crossover, 17 parallel, and 1 partial crossover in design. All 37 studies assessed participant blinding, and 10 also assessed investigator blinding. Approximately 27% included an "unsure" answer option for treatment guess, and 38% asked the reason for the guess. There were no clear patterns in BA reporting across time nor based on treatment type. Seventeen trials provided sufficient data to calculate Bang Blinding Index (BI) to determine blinding success. Participants remained blinded (BI = 0 &#xb1; 0.2) in 10/17 placebo and 10/17 treatment arms, 6 placebo and 5 treatment arms had a BI > 0.2 suggesting possible unblinding, whereas 1 placebo and 2 treatment arms had a BI < -0.2 suggesting misinformed guessing. Overall, we found that BAs are done in a minority of published neuropathic pain trials and with variable methodology. Given the importance of minimizing risk of bias because of treatment unblinding, future studies should consider including BAs, and further consensus building is necessary to determine if and how BAs should be conducted and interpreted in analgesic clinical trials.

Bias↗
PubMed2006

Multiple imputation for correcting verification bias.

In the case in which all subjects are screened using a common test and only a subset of these subjects are tested using a golden standard test, it is well documented that there is a risk for bias, called verification bias. When the test has only two levels (e.g. positive and negative) and we are trying to estimate the sensitivity and specificity of the test, we are actually constructing a confidence interval for a binomial proportion. Since it is well documented that this estimation is not trivial even with complete data, we adopt multiple imputation framework for verification bias problem. We propose several imputation procedures for this problem and compare different methods of estimation. We show that our imputation methods are better than the existing methods with regard to nominal coverage and confidence interval length.

Bias↗
PubMed2006

Rater bias in a blinded randomized placebo-controlled psychiatry trial.

Rater bias occurs when rater knowledge of treatment assignment modifies the outcome assessment. Raters may be unconsciously influenced by inclinations for or against a particular treatment and consequently may give a more or less generous assessment depending upon these biases. Blinding of raters by keeping raters unaware of treatment assignment is one way to limit bias influencing assessment due to knowledge of treatment assignment. Unblinding may be particularly problematic in efficacy studies comparing placebo to drugs and/or non-drug psychotherapy treatments where subjects may reveal drug side-effects or mention their therapist by name, thus unblinding their treatment assignment. We present a new instrumental variable statistical approach for assessing the association between success in blinding and its impact on efficacy estimates of active drug and/or cognitive behavioural psychotherapy versus placebo in the multicentre comparative treatment study of panic disorder. Despite the uncertainty involved in assessing bias that may be unobserved and unconscious, we will show how to derive a bound for the impact of rater bias.

Bias↗