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

Odds ratios (OR) for different cutoffs used to create 5 categories for the causal yet mismeasuered exposure (W1) (all scenarios for Figure 3) (δ is identified as “W1 delta” in the figure). The 4 W1 cutoffs were centered at 0 with δ from 0.5 to 1.5 standard deviations, in increments of 0.1. Reference category for W1 is mean +/− (0.5 × δ × standard deviation).
© Copyright Policy
Related In: Results  -  Collection

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

ijerph-12-10198-f006: Odds ratios (OR) for different cutoffs used to create 5 categories for the causal yet mismeasuered exposure (W1) (all scenarios for Figure 3) (δ is identified as “W1 delta” in the figure). The 4 W1 cutoffs were centered at 0 with δ from 0.5 to 1.5 standard deviations, in increments of 0.1. Reference category for W1 is mean +/− (0.5 × δ × standard deviation).

Mentions: The effect of changing the cutoffs used to create five categories of the causal exposure is illustrated in Figure 3 and Figure A2. The graphs were distinct for each of the three exposure-response models. For the linear models, as the width of the reference group (δ) increased, the ORs were farther from the for all scenarios. In addition, as the measurement error variance increased, the ORs for all four categories moved closer to the . Again, we note that whereas under low measurement error the graphs can be used to distinguish between associations of different shapes, this identification of shapes becomes problematic as measurement error increases, even with the increase in number of categories from two to five. In practice, this may lead to both under- and over-interpretation of information contained in the data about the shape of the exposure-response relationship (e.g., dismissing non-linearity when it is present, or positing a curvilinear shape when the exposure-response is linear, respectively), even though it would leave little doubt that there is an association between the exposure and outcome.


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)

Odds ratios (OR) for different cutoffs used to create 5 categories for the causal yet mismeasuered exposure (W1) (all scenarios for Figure 3) (δ is identified as “W1 delta” in the figure). The 4 W1 cutoffs were centered at 0 with δ from 0.5 to 1.5 standard deviations, in increments of 0.1. Reference category for W1 is mean +/− (0.5 × δ × standard deviation).
© Copyright Policy
Related In: Results  -  Collection

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

ijerph-12-10198-f006: Odds ratios (OR) for different cutoffs used to create 5 categories for the causal yet mismeasuered exposure (W1) (all scenarios for Figure 3) (δ is identified as “W1 delta” in the figure). The 4 W1 cutoffs were centered at 0 with δ from 0.5 to 1.5 standard deviations, in increments of 0.1. Reference category for W1 is mean +/− (0.5 × δ × standard deviation).
Mentions: The effect of changing the cutoffs used to create five categories of the causal exposure is illustrated in Figure 3 and Figure A2. The graphs were distinct for each of the three exposure-response models. For the linear models, as the width of the reference group (δ) increased, the ORs were farther from the for all scenarios. In addition, as the measurement error variance increased, the ORs for all four categories moved closer to the . Again, we note that whereas under low measurement error the graphs can be used to distinguish between associations of different shapes, this identification of shapes becomes problematic as measurement error increases, even with the increase in number of categories from two to five. In practice, this may lead to both under- and over-interpretation of information contained in the data about the shape of the exposure-response relationship (e.g., dismissing non-linearity when it is present, or positing a curvilinear shape when the exposure-response is linear, respectively), even though it would leave little doubt that there is an association between the exposure and outcome.

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