Limits...
A Simulation Study of Categorizing Continuous Exposure Variables Measured with Error in Autism Research: Small Changes with Large Effects.

Heavner K, Burstyn I - Int J Environ Res Public Health (2015)

Bottom Line: Variation in the odds ratio (OR) resulting from selection of cutoffs for categorizing continuous variables is rarely discussed.Cutoffs chosen for categorizing continuous variables can have profound effects on study results.When measurement error is not too great, the shape of the OR curve may provide insight into the true shape of the exposure-disease relationship.

View Article: PubMed Central - PubMed

Affiliation: Department of Environmental and Occupational Health, School of Public Health, Drexel University, Philadelphia, PA 19104, USA. karynkh@aol.com.

ABSTRACT
Variation in the odds ratio (OR) resulting from selection of cutoffs for categorizing continuous variables is rarely discussed. We present results for the effect of varying cutoffs used to categorize a mismeasured exposure in a simulated population in the context of autism spectrum disorders research. Simulated cohorts were created with three distinct exposure-outcome curves and three measurement error variances for the exposure. ORs were calculated using logistic regression for 61 cutoffs (mean ± 3 standard deviations) used to dichotomize the observed exposure. ORs were calculated for five categories with a wide range for the cutoffs. For each scenario and cutoff, the OR, sensitivity, and specificity were calculated. The three exposure-outcome relationships had distinctly shaped OR (versus cutoff) curves, but increasing measurement error obscured the shape. At extreme cutoffs, there was non-monotonic oscillation in the ORs that cannot be attributed to "small numbers." Exposure misclassification following categorization of the mismeasured exposure was differential, as predicted by theory. Sensitivity was higher among cases and specificity among controls. Cutoffs chosen for categorizing continuous variables can have profound effects on study results. When measurement error is not too great, the shape of the OR curve may provide insight into the true shape of the exposure-disease relationship.

No MeSH data available.


Related in: MedlinePlus

ROC curves for selected cutoffs for W1.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4555337&req=5

ijerph-12-10198-f009: ROC curves for selected cutoffs for W1.

Mentions: The sensitivity and specificity for each cutoff used for dichotomization is illustrated in Figure 4 and Figure A4. There was a clear pattern of trade-off between sensitivity and specificity, with cutoffs at higher apparent exposure values producing classifiers that favored specificity over sensitivity and vice versa. Cutoffs in the center of the apparent exposure distribution produced values of sensitivity and specificity that were similar to each other. The curves for sensitivity and specificity cross below the mean for controls and above the mean for cases. The specificity is unstable for smaller cutoff values and the sensitivity is unstable for larger cutoff values. Sensitivity tended to be higher among cases and specificity higher among controls. The deviation from non-differential misclassification was most pronounced when dichotomization was done at the extremes of the exposure distribution. Furthermore, the degree of divergence from non-differential misclassification increased with the strength of the true exposure-outcome association as well as the magnitude of the measurement error variance. Receiver operator characteristic (ROC) curves for selected cutoffs (−3, −2, −1, 0, 1, 2, 3) are shown in Figure A5. In general, the ROC curves varied little by the cutoff used to dichotomize W1, particularly when the true association was weak.


A Simulation Study of Categorizing Continuous Exposure Variables Measured with Error in Autism Research: Small Changes with Large Effects.

Heavner K, Burstyn I - Int J Environ Res Public Health (2015)

ROC curves for selected cutoffs for W1.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4555337&req=5

ijerph-12-10198-f009: ROC curves for selected cutoffs for W1.
Mentions: The sensitivity and specificity for each cutoff used for dichotomization is illustrated in Figure 4 and Figure A4. There was a clear pattern of trade-off between sensitivity and specificity, with cutoffs at higher apparent exposure values producing classifiers that favored specificity over sensitivity and vice versa. Cutoffs in the center of the apparent exposure distribution produced values of sensitivity and specificity that were similar to each other. The curves for sensitivity and specificity cross below the mean for controls and above the mean for cases. The specificity is unstable for smaller cutoff values and the sensitivity is unstable for larger cutoff values. Sensitivity tended to be higher among cases and specificity higher among controls. The deviation from non-differential misclassification was most pronounced when dichotomization was done at the extremes of the exposure distribution. Furthermore, the degree of divergence from non-differential misclassification increased with the strength of the true exposure-outcome association as well as the magnitude of the measurement error variance. Receiver operator characteristic (ROC) curves for selected cutoffs (−3, −2, −1, 0, 1, 2, 3) are shown in Figure A5. In general, the ROC curves varied little by the cutoff used to dichotomize W1, particularly when the true association was weak.

Bottom Line: Variation in the odds ratio (OR) resulting from selection of cutoffs for categorizing continuous variables is rarely discussed.Cutoffs chosen for categorizing continuous variables can have profound effects on study results.When measurement error is not too great, the shape of the OR curve may provide insight into the true shape of the exposure-disease relationship.

View Article: PubMed Central - PubMed

Affiliation: Department of Environmental and Occupational Health, School of Public Health, Drexel University, Philadelphia, PA 19104, USA. karynkh@aol.com.

ABSTRACT
Variation in the odds ratio (OR) resulting from selection of cutoffs for categorizing continuous variables is rarely discussed. We present results for the effect of varying cutoffs used to categorize a mismeasured exposure in a simulated population in the context of autism spectrum disorders research. Simulated cohorts were created with three distinct exposure-outcome curves and three measurement error variances for the exposure. ORs were calculated using logistic regression for 61 cutoffs (mean ± 3 standard deviations) used to dichotomize the observed exposure. ORs were calculated for five categories with a wide range for the cutoffs. For each scenario and cutoff, the OR, sensitivity, and specificity were calculated. The three exposure-outcome relationships had distinctly shaped OR (versus cutoff) curves, but increasing measurement error obscured the shape. At extreme cutoffs, there was non-monotonic oscillation in the ORs that cannot be attributed to "small numbers." Exposure misclassification following categorization of the mismeasured exposure was differential, as predicted by theory. Sensitivity was higher among cases and specificity among controls. Cutoffs chosen for categorizing continuous variables can have profound effects on study results. When measurement error is not too great, the shape of the OR curve may provide insight into the true shape of the exposure-disease relationship.

No MeSH data available.


Related in: MedlinePlus