Proper interpretation of non-differential misclassification effects: expectations vs observations

Jurek, Anne M.; Greenland, Sander; Maldonado, George; Church, Timothy R.

doi:10.1093/ije/dyi060

Cited by 315 publications

(222 citation statements)

References 29 publications

Supporting

Mentioning

213

Contrasting

Unclassified

Order By: Relevance

“…Additionally, a more precise risk estimate resulted (i.e., tighter confidence interval) using the C-CASA. These findings support the notion that misclassification may lead to overestimation of true risk (37). Such a change in risk estimation has clinical implications and likely affects risk-benefit analyses.…”

Section: Discussionsupporting

confidence: 78%

Columbia Classification Algorithm of Suicide Assessment (C-CASA): Classification of Suicidal Events in the FDA’s Pediatric Suicidal Risk Analysis of Antidepressants

Posner

Oquendo²,

Gould³

et al. 2007

Am J Psychiatry

485

243

View full text Add to dashboard Cite

Objective-To evaluate the link between antidepressants and suicidal behavior and ideation (suicidality) in youth, adverse events from pediatric clinical trials were classified in order to identify suicidal events. The authors describe the Columbia Classification Algorithm for Suicide Assessment (C-CASA), a standardized suicidal rating system that provided data for the pediatric suicidal risk analysis of antide-pressants conducted by the Food and Drug Administration (FDA).Method-Adverse events (N=427) from 25 pediatric antidepressant clinical trials were systematically identified by pharmaceutical companies. Randomly assigned adverse events were evaluated by three of nine independent expert suicidologists using the Columbia classification algorithm. Reliability of the C-CASA ratings and agreement with pharmaceutical company classification were estimated.Results-Twenty-six new, possibly suicidal events (behavior and ideation) that were not originally identified by pharmaceutical companies were identified in the C-CASA, and 12 events originally labeled as suicidal by pharmaceutical companies were eliminated, which resulted in a total of 38 discrepant ratings. For the specific label of "suicide attempt," a relatively low level of agreement was observed between the C-CASA and pharmaceutical company ratings, with the C-CASA reporting a 50% reduction in ratings. Thus, although the C-CASA resulted in the identification of more suicidal events overall, fewer events were classified as suicide attempts. Additionally, the C-CASA ratings were highly reliable (intraclass correlation coefficient [ICC]=0.89).Address correspondence and reprint requests to Dr. Posner, Division of Child Psychiatry, Columbia University/New York State Psychiatric Institute, 1051 Riverside Dr., Unit 74, New York, NY 10032; posnerk@childpsych.columbia.edu. CME Disclosure Dr. Posner has received funding from the FDA to develop the suicidality classification system used in their antidepressant safety analysis. Subsequently, as part of an effort to help execute the FDA adult suicidality classification mandates that followed this FDAsponsored pediatric study, Dr. Posner has had research support from GlaxoSmith-Kline, Forest Laboratories, Eisai, Inc., AstraZeneca, Johnson and Johnson, Abbott Laboratories, Wyeth, Organon USA, Bristol-Myers Squibb, Sanofi-Aventis, Cephalon, Novartis, Shire Pharmaceuticals, and UCB Pharma. Dr. Oquendo has received funding from NIMH, National Institute on Alcohol Abuse and Alcoholism, Moody's Foundation, American Foundation for Suicide Prevention, and Eli Lilly (unrestricted educational grant); she has also served as a consultant to Pfizer. Dr. Stanley has received fluoxetine/placebo for a randomized controlled trial from Eli Lilly. Mr. Davies is a shareholder with Merck, Pfizer, Wyeth, GlaxoSmithKline, Johnson and Johnson, Amgen, and Bard. Dr. Gould reports no competing interests. APA policy requires disclosure by CME authors of unapproved or investigational use of products discussed in CME programs. Offlabel use of m...

show abstract

Section: Discussionsupporting

confidence: 78%

Columbia Classification Algorithm of Suicide Assessment (C-CASA): Classification of Suicidal Events in the FDA’s Pediatric Suicidal Risk Analysis of Antidepressants

Posner

Oquendo²,

Gould³

et al. 2007

Am J Psychiatry

485

243

View full text Add to dashboard Cite

show abstract

“…However, differential misclassification with respect to case/control status is unlikely because ETS exposure has never been known to be a risk factor for bladder cancer. Nondifferential misclassification of the exposure status tends to bias the results toward the null assuming independence of the errors and absence of other biases (38). Thus, underestimation of a true effect rather than creation of a spurious association is more likely to result from misclassification of ETS exposure status in our study subjects.…”

Section: Discussionmentioning

confidence: 88%

Environmental Tobacco Smoke and Bladder Cancer Risk in Never Smokers of Los Angeles County

et al. 2007

View full text Add to dashboard Cite

Cigarette smoking is a major risk factor for bladder cancer and a prominent point source of 4-aminobiphenyl (4-ABP), a recognized human bladder carcinogen. 4-ABP-hemoglobin (Hb) adducts are established biomarkers of 4-ABP exposure in humans. The role of environmental tobacco smoke (ETS) in the etiology of bladder cancer is largely unknown. As part of a large population-based bladder cancer study in Los Angeles County, California, lifetime exposure to ETS was ascertained for 148 cases and 292 control subjects who had never used any tobacco products over their lifetime. 4-ABP-Hb adducts were quantitatively measured on 230 control subjects. Female lifelong nonsmokers living with two or more smokers during childhood were significantly related to risk of bladder cancer [odds ratio (OR), 3.08; 95% confidence interval (95% CI), 1.16-8.22]. During adulthood, f2-fold risks were seen among women living with a spouse/domestic partner who smoked for z10 years or having a coworker who smoked in an indoor environment for z10 years. When all sources of ETS exposure were combined, a statistically significant, dose-dependent association (P for trend = 0.03) was noted in women, with the OR for the highest category of ETS exposure being 5.48 (95% CI, 1.06-28.36). Levels of 4-ABP-Hb adducts varied by ETS exposure status among female control subjects. Mean level was lowest in women never exposed to ETS (16.4 pg/g Hb) and highest in those with current ETS exposure (23.6 pg/g Hb). ETS exposure was associated with neither bladder cancer risk nor 4-ABP-Hb adduct levels in male lifelong nonsmokers. In conclusion, ETS is a risk factor for bladder cancer in women who were lifelong nonusers of any tobacco products. [Cancer Res 2007;67(15):7540-5]

show abstract

“…We recognize that some bias occurred because we used claims to determine early stage of disease at the time a woman was actually enrolled. However, the measures of effect were more likely to be biased toward the null if the non-differential misclassification happened only on identifying early versus non-early stage of cancer (Jurek et al 2005;Jurek et al 2008) and if there was no difference in this misclassification preversus post-BCCPTA.…”

Section: Limitationsmentioning

confidence: 99%

Medicaid Enrollment at Early Stage of Disease: The Breast and Cervical Cancer Prevention and Treatment Act in Georgia

Chien

Adams

Zhou

2011

INQUIRY

View full text Add to dashboard Cite

This study linked data from the Georgia Comprehensive Cancer Registry to Medicaid enrollment and claims to test whether the Breast and Cervical Cancer Prevention and Treatment Act (BCCPTA), which provided a new and quicker pathway to Medicaid eligibility for low-income breast cancer patients, led to more patients enrolling at an early stage of disease. Results based on difference-in-differences analysis indicated that Georgia's BCCPTA increased by 11 percentage points the probability of breast cancer patients enrolling in Medicaid at an early stage (p = .024). This effect could mean more treatment options and higher survival rates for these patients.

show abstract

Proper interpretation of non-differential misclassification effects: expectations vs observations

Abstract: Non-differentiality of exposure misclassification does not justify claims that the observed estimate must be an underestimate; further conditions must hold to get bias towards the null, and even when they do hold the observed estimate may by chance be an overestimate.

Cited by 315 publications

References 29 publications

Columbia Classification Algorithm of Suicide Assessment (C-CASA): Classification of Suicidal Events in the FDA’s Pediatric Suicidal Risk Analysis of Antidepressants

Columbia Classification Algorithm of Suicide Assessment (C-CASA): Classification of Suicidal Events in the FDA’s Pediatric Suicidal Risk Analysis of Antidepressants

Environmental Tobacco Smoke and Bladder Cancer Risk in Never Smokers of Los Angeles County

Medicaid Enrollment at Early Stage of Disease: The Breast and Cervical Cancer Prevention and Treatment Act in Georgia

Contact Info

Product

Resources

About