1 / 35

Exploring the Shape of the Dose-Response Function

Exploring the Shape of the Dose-Response Function. Outline. Traditional approach to dose-response analysis The “step function” Alternative: “Flexible” regression line Spline regression Examples: logistic/linear/Cox. Example: Sleep-Disordered Breathing and Stroke.

Download Presentation

Exploring the Shape of the Dose-Response Function

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Exploring the Shape of the Dose-Response Function

  2. Outline • Traditional approach to dose-response analysis • The “step function” • Alternative: “Flexible” regression line • Spline regression • Examples: logistic/linear/Cox

  3. Example: Sleep-Disordered Breathing and Stroke • Study: the Sleep Heart Health Study • Data set: cross-sectional • Exposure variable: apnea-hypopnea index (AHI) • Dependent variable: self-reported stroke • Potential confounders: known stroke risk factors

  4. Data set • Observations: N=5,192 • Self-reported stroke: N=204 Apnea- Hypopnea Index (AHI) Mean Percentile Distribution 5th 25th 50th 75th 95th 8.9 0.2 1.4 4.5 11.3 34.1

  5. Traditional Approach: Categorical Analysis • Categorization dummy coding AHI Q2 Q3 Q4 0 - 1.4 0 0 0 1.5 - 4.5 1 0 0 4.6 - 11.3 0 1 0 >11.3 0 0 1

  6. Traditional Approach: Step Function Model: Log odds (stroke) = 1 + 2Q2 + 3Q3 + 4Q4 + Z Maximum Likelihood Estimates: Log odds (stroke) = (-9.924) + (0.301)Q2 + (0.344)Q3 + (0.454)Q4 + Z

  7. Adjusted Odds Ratios of PrevalentSTROKEby Quartile of the Apnea-Hypopnea Index AHI Quartile I II III IV 1.0 (ref.) 1.35 (0.84 - 2.18) 1.41 (0.88 - 2.26) 1.57 (0.98 - 2.53)

  8. Traditional Approach: Step Function

  9. Traditional Approach: “Step Function” Log odds (stroke) = 1 + 2Q2 + 3Q3 + 4Q4 + Z AHI Fitted Model 0 - 1.4 Log (odds of stroke) = 1 + Z 1.5 - 4.5 Log (odds of stroke) = 1 + 2 + Z 4.6 - 11.3 Log (odds of stroke) = 1 + 3 + Z > 11.3 Log (odds of stroke) = 1 + 4 + Z

  10. Traditional Approach: “Step Function” Log odds (stroke) = 1 + 2Q2 + 3Q3 + 4Q4 + Z AHI Fitted Model 0 - 1.4 Log (odds of stroke) = -9.924 + Z 1.5 - 4.5 Log (odds of stroke) = -9.623 + Z 4.6 - 11.3 Log (odds of stroke) = -9.580 + Z > 11.3 Log (odds of stroke) = -9.470+ Z

  11. Traditional Approach: Step Function Log odds (stroke) -9.470 + Z -9.580 + Z -9.623 + Z -9.924 + Z 0 1.4 4.5 11.3 AHI

  12. Step Function: Problems • Unrealistic assumptions • A “step function” • We actually don’t believe it; our mind tries to draw an imaginary smooth line through the step • Choice of categories could influence the shape • Test for trend • Not a test for monotonic dose-response • Statistical hypothesis testing

  13. Alternative: “Flexible” Regression Line • Spline Regression • Categorize (specify cutoff points) (as in categorical analysis) • Fit the regression line in segments (as in categorical analysis) • Enforce continuity at the junctions (knots) (new)

  14. EXAMPLE: Linear Spline Regression Log odds (stroke) 0 1.4 4.5 11.3 AHI

  15. Linear Spline Regression Log odds (stroke) 0 1.4 4.5 11.3

  16. Linear Spline Regression • Fit two straight regression lines • Ensure continuity at the knot (AHI=1.4) Method: • Define a new variable, S S=0, if AHI<1.4 S=AHI-1.4, if AHI>1.4

  17. Linear Spline Regression Log odds (stroke) = 0 + 1(AHI)+ 2(S)+ Z To the left of the knot: S=0 Log odds (stroke) = 0 + 1(AHI) + Z To the right of the knot: S=AHI-1.4 Log odds (stroke) = 0 + 1(AHI) + 2(AHI-1.4) + Z = 0 -1.4 2 + (1+ 2)AHI + Z • Different slopes • Identical predicted value at the knot (AHI=1.4)

  18. More Flexible Spline Regression • Quadratic spline AHI + AHI2 • Cubic spline AHI + AHI2 + AHI3

  19. Basic quadratic spline: Step #1 • Determine cutpoints (C1, C2, C3) on the exposure scale (4 categories) • These are either percentiles or some other values. That is, decide on the values of C1, C2, C3 of your choice C1=?; C2=?; C3=?;

  20. Step #2 S1 = EXP2; S2 = 0; S3 = 0; S4 = 0; IF EXP > C1 THEN S2 = (EXP-C1)2; IF EXP > C2 then S3 = (EXP-C2)2; IF EXP > C3 then S4 = (EXP-C3)2;

  21. Step #3 Regress the dependent variable on EXP S1 S2 S3 S4 covariates And find the four regression equations: one per exposure category (together they form a continuous dose-response function) Step #4 Compute and display the dose-response function

  22. C1=14; C2=29; Example: pack-years of smoking and CHD C3=43; EXP = pack-years S1 = EXP**2; S2=0; S3=0; S4=0; IF EXP > C1 THEN S2 = (EXP-C1)**2; IF EXP > C2 then S3 = (EXP-C2)**2; IF EXP > C3 then S4 = (EXP-C3)**2;

  23. PROCLOGISTIC; MODEL DIS = EXP S1 S2 S3 S4;

  24. Maximum Likelihood Estimates Parameter DF Estimate Intercept 1 -1.7022 (α) EXP 1 -0.0203 (β0) S1 1 0.00252 (β1) S2 1 -0.00265 (β2) S3 1 -0.00047 (β3) S4 1 0.000305 (β4)

  25. Log odds (CHD) =α + 0(EXP)+ 1(S1) + 2(S2) + 3(S3) + 4(S4) EXP Four regression equations < 14 Log odds (CHD) = S1=EXP2, S2=0, S3=0, S4=0 15-29 Log odds (CHD) = S1=EXP2, S2=(EXP-14)2, S3=0, S4=0 30-43 Log odds (CHD) = S1=EXP2, S2=(EXP-14)2, S3=(EXP-29)2, S4=0 >43 Log odds (CHD) = S1=EXP2, S2=(EXP-14)2, S3=(EXP-29)2, S4=(EXP-43)2

  26. Cubic Spline RegressionLog odds (stroke) vs. AHI3 Knots: 0.2, 4.5, 34.1

  27. Cubic Spline RegressionLog odds (stroke) vs. AHI4 knots: 0.2, 1.4, 11.3, 34.1

  28. Spline Regression: Applications Regression Dependent SAS Procedure Model Variable Logistic log odds (Y=1) PROC LOGISTIC Linear mean Y PROC REG Cox log (hazard) PROC PHREG All models are linear functions of the predictors

  29. Spline Regression (within PROC REG) Systolic BP vs. AHI3 knots: 0.1, 3.6, 29.1

  30. Spline Regression (within PROC REG) Systolic BP vs. AHI4 knots: 0.1, 1.1, 9.5, 29.1

  31. Spline Regression (within PROC REG) Systolic BP vs. AHI5 knots: 0.1, 1.1, 3.6, 9.5, 29.1

  32. Spline Regression Key Advantages • Less restrictive assumptions • More regional flexibility • Does not rely on statistical hypothesis testing • Not as sensitive to the choice of cutoff points • Visual inspection of the dose-response pattern • Might be used to guide the choice of categories for traditional categorical analysis

  33. Spline Regression Key Issues • Moderately sensitive to the number of knots (especially if only 3 are specified) • What do the “bumps and valleys” really mean? • Visual (subjective) interpretation • Consider the scale of the Y-axis • Consider the amount of data at the tail(s) • Straight line at the outermost segments

More Related