Sample size evaluation for a multiply matched case–control study using the score test from a conditional logistic (discrete Cox PH) regression model

Lachin, John M.

doi:10.1002/sim.3057

Cited by 45 publications

(31 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…35 Statistical power calculations indicated that the sample size is sufficient for detecting a linear dose-response with an OR of 1.05 or greater per 1 mGy increase in cumulative RBM dose with a statistical power of 80% using asymptotic conditional logistic regression. 36 …”

Section: Methodsmentioning

confidence: 99%

Radiation exposure from computerized tomography and risk of childhood leukemia: Finnish register-based case-control study of childhood leukemia (FRECCLE)

Nikkilä

Raitanen²,

Lohi

et al. 2018

Haematologica

View full text Add to dashboard Cite

The only well-established risk factors for childhood leukemia are high-dose ionizing radiation and Down syndrome. Computerized tomography is a common source of low-dose radiation. In this study, we examined the magnitude of the risk of childhood leukemia after pediatric computed tomography examinations. We evaluated the association of computed tomography scans with risk of childhood leukemia in a nationwide register-based case-control study. Cases (n=1,093) were identified from the population-based Finnish Cancer Registry and three controls, matched by gender and age, were randomly selected for each case from the Population Registry. Information was also obtained on birth weight, maternal smoking, parental socioeconomic status and background gamma radiation. Data on computed tomography scans were collected from the ten largest hospitals in Finland, covering approximately 87% of all pediatric computed tomography scans. Red bone marrow doses were estimated with NCICT dose calculation software. The data were analyzed using exact conditional logistic regression analysis. A total of 15 cases (1.4%) and ten controls (0.3%) had undergone one or more computed tomography scans, excluding a 2-year latency period. For one or more computed tomography scans, we observed an odds ratio of 2.82 (95% confidence interval: 1.05 – 7.56). Cumulative red bone marrow dose from computed tomography scans showed an excess odds ratio of 0.13 (95% confidence interval: 0.02 – 0.26) per mGy. Our results are consistent with the notion that even low doses of ionizing radiation observably increase the risk of childhood leukemia. However, the observed risk estimates are somewhat higher than those in earlier studies, probably due to random error, although unknown predisposing factors cannot be ruled out.

show abstract

Section: Methodsmentioning

confidence: 99%

Radiation exposure from computerized tomography and risk of childhood leukemia: Finnish register-based case-control study of childhood leukemia (FRECCLE)

Nikkilä

Raitanen²,

Lohi

et al. 2018

Haematologica

View full text Add to dashboard Cite

show abstract

“…Effect modification i.e ., difference between subgroups was assessed based on the significance of an interaction term added to a model with the main effects. Statistical power calculations indicated that the material was sufficient for detecting a linear dose‐response with OR of 1.06 or greater per 10 nSv h −1 increase in dose rate with statistical power of 80% …”

Section: Methodsmentioning

confidence: 99%

Background radiation and childhood leukemia: A nationwide register‐based case‐control study

Nikkilä

Erme

Arvela

et al. 2016

Intl Journal of Cancer

View full text Add to dashboard Cite

High doses of ionizing radiation are an established cause of childhood leukemia. However, substantial uncertainty remains about the effect of low doses of radiation, including background radiation and potential differences between genetic subgroups of leukemia have rarely been explored. We investigated the effect of the background gamma radiation on childhood leukemia using a nationwide register-based case-control study. For each of the 1,093 cases, three age- and gender matched controls were selected (N = 3,279). Conditional logistic regression analyses were adjusted for confounding by Down syndrome, birth weight (large for gestational age), and maternal smoking. Complete residential histories and previously collected survey data of the background gamma radiation in Finland were used to assess the exposure of the study subjects to indoor and outdoor gamma radiation. Overall, background gamma radiation showed a non-significant association with the OR of childhood leukemia (OR 1.01, 95% CI 0.97, 1.05 for 10 nSv/h increase in average equivalent dose rate to red bone marrow). In subgroup analyses, age group 2-<7 years displayed a larger effect (OR 1.27, 95% CI 1.01, 1.60 for 1 mSv increase in equivalent cumulative dose to red bone marrow). Suggestive difference in OR by genetic subtype was found. Our results provide further support to the notion that low doses of ionizing radiation increase the risk for childhood leukemia, particularly at age 2-<7 years. Our findings suggest a larger effect of radiation on leukemia with high hyperpdiploidy than other subgroups, but this result requires further confirmation.

show abstract

“…This “efficiency‐rule” reappears in a sample size formula Lachin () derived for multiply matched nested case‐control data. In the situation of “one case per risk set” Lachin () just adds factor

(m + 1) / m

to the right hand side of the famous formula by Schoenfeld (). However, as shown by Ohneberg et al.…”

Section: Discussionmentioning

confidence: 94%

A simulation approach for power calculation in large cohort studies based on multistate models

Jenny

Beyersmann²,

Schumacher

2018

Biometrical J

View full text Add to dashboard Cite

Realistic power calculations for large cohort studies and nested case control studies are essential for successfully answering important and complex research questions in epidemiology and clinical medicine. For this, we provide a methodical framework for general realistic power calculations via simulations that we put into practice by means of an R-based template. We consider staggered recruitment and individual hazard rates, competing risks, interaction effects, and the misclassification of covariates. The study cohort is assembled with respect to given age-, gender-, and community distributions. Nested case-control analyses with a varying number of controls enable comparisons of power with a full cohort analysis. Time-to-event generation under competing risks, including delayed study-entry times, is realized on the basis of a six-state Markov model. Incidence rates, prevalence of risk factors and prefixed hazard ratios allow for the assignment of age-dependent transition rates given in the form of Cox models. These provide the basis for a central simulation-algorithm, which is used for the generation of sample paths of the underlying time-inhomogeneous Markov processes. With the inclusion of frailty terms into the Cox models the Markov property is specifically biased. An "individual Markov process given frailty" creates some unobserved heterogeneity between individuals. Different left-truncation- and right-censoring patterns call for the use of Cox models for data analysis. p-values are recorded over repeated simulation runs to allow for the desired power calculations. For illustration, we consider scenarios with a "testing" character as well as realistic scenarios. This enables the validation of a correct implementation of theoretical concepts and concrete sample size recommendations against an actual epidemiological background, here given with possible substudy designs within the German National Cohort.

show abstract

Sample size evaluation for a multiply matched case–control study using the score test from a conditional logistic (discrete Cox PH) regression model

Cited by 45 publications

References 26 publications

Radiation exposure from computerized tomography and risk of childhood leukemia: Finnish register-based case-control study of childhood leukemia (FRECCLE)

Radiation exposure from computerized tomography and risk of childhood leukemia: Finnish register-based case-control study of childhood leukemia (FRECCLE)

Background radiation and childhood leukemia: A nationwide register‐based case‐control study

A simulation approach for power calculation in large cohort studies based on multistate models

Contact Info

Product

Resources

About