On the Analysis of Crossover Designs Dallas E. Johnson Professor Emeritus Kansas State University dejohnsn@ksu.edu 785-532-0510 (Office) 785-539-0137 (Home) Dallas E. Johnson 1812 Denholm Dr. Manhattan, KS 66503-2210 CROSSOVER DESIGNS Two Period - Two Treatment Crossover Design Period 1 SEQ 1 A SEQ 2 B Period 2 B A We have n1 subjects randomly assigned to sequence 1, and n2 subjects randomly assigned to sequence 2. Suppose nj subjects are assigned to the jth sequence of treatments. Let yijk denote the observed response from the kth subject in the jth sequence during the ith period; i=1,2; j=1,2; k = 1,2,...,nj. Means Model: yijk = ij + jk + ijk for all i,j,k Ideal Conditions: jk ~ iid N(0,2), ijk ~iid N(0,2), and all jk’s and ijk’s are independent. Main Advantage: Each subject serves as his/her own control. Most appropriate for conditions that reoccur. Disadvantages: The designs cannot be used for treatment comparisons when the condition being treated is cured during the first period. The two period/two treatment crossover design should not be used when carryover effects exist unless one can include a “washout period” prior to administering a treatment in the second period. IDEAL TREATMENT STRUCTURE MODEL Model Parameters Period 1 Period 2 SEQ 1 SEQ 2 CARRYOVER MODEL: Model Parameters Period 1 Period 2 SEQ 1 SEQ 2 Model Parameters Period 1 Period 2 NOTE: SEQ 1 SEQ 2 Therefore , , . If carry-over exists, the difference in treatment effects can only be obtained from Period 1 data. Then . Usual Analysis SV Seq df 1 MS F S12 S12/S22 Sub(Seq) n1+n2-2 S22 Trt 1 S32 S32/S52 Per 1 S42 S42/S52 Error n1+n2-2 S52 Hyp Tstd Question: What if the treatments have different variances? Let yjk be the vector of responses for the kth subject in the jth sequence of treatments, and suppose we assume that yjk ~ N( j, ) where 1 j j 2j *11 *12 2 *11 2 *12 11 12 1 1 * * 2 * 2 * 1 1 21 22 21 22 21 22 2 ij i S j * and where and are determined by i and j. Two period/two treatment crossover designs where the treatments have unequal variances. Here we assume that A2 A B A B B2 for the AB sequence and that B2 A B A B A2 for the BA sequence. Note that y11k y21k ~ N ( A B 1 2 , A2 2 A B B2 ), k 1, 2,..., n1 and y22 k y12 k ~ N ( A B 2 1 , A2 2 A B B2 ), k 1, 2,..., n2 Shanga (2003) showed that in the two period/two treatment crossover design without carryover effects, the estimated standard errors for testing both between subject and within subject contrasts of the ij’s when the treatments have unequal variances are exactly the same as they are when the treatments have equal variances. Thus if one is only interested in the difference between the two treatments, then there is no need to worry about unequal variances in this design. Other questions: Are there any consequences to ignoring carryover and/or unequal variances in the two period/two treatment crossover design? In particular - does unequal variances show up as carryover or does carryover show up as unequal variances? To answer these kinds of questions, Shanga simulated two period/two treatment crossover experiments satisfying four different conditions: (1) no carryover and equal variances (C0V0), (2) no carryover and unequal variances(C0V1), (3) carryover and equal variances (C1V0), and (4) carryover and unequal variances (C1V1). Each of 1000 sets of data under each of these conditions was analyzed four different ways assuming: (1) (2) (3) (4) no carryover and equal variances (C0V0), no carryover and unequal variances(C0V1), carryover and equal variances (C1V0), and carryover and unequal variances (C1V1). PROC MIXED; TITLE2 'EQUAL VARIANCES'; CLASSES SEQ PERIOD TRT PERSON; MODEL PEF=SEQ TRT PERIOD/DDFM=SATTERTH; REPEATED TRT/SUBJECT=PERSON(SEQ) TYPE=CS; LSMEANS TRT /PDIFF; RUN; PROC MIXED; TITLE2 'UNEQUAL VARIANCES'; CLASSES SEQ PERIOD TRT PERSON; MODEL PEF=SEQ TRT PERIOD/DDFM=SATTERTH; REPEATED TRT/SUBJECT=PERSON(SEQ) TYPE=CSH; LSMEANS TRT /PDIFF; RUN; Tests for equal treatment effects. Analysis Assumptions N= 6, =.5, B=2 Simulation C0V0 C0V1 C1V0 C1V1 C0V0 =.040 (1) =.87 =.040 (1) =.87 = .050 (1) =.38 =.050 (1)=.38 C0V1 =.045 (1) =.43 =.045 (1) =.43 = .050 (1) =.18 =.046 (1)=.17 C1V0 =.124 (1) =.66 =.124 (1) =.66 = .050 (1) =.38 =.050 (1)=.38 C1V1 =.066 (1) =.26 =.066 (1) =.26 = .050 (1) =.18 =.046 (1)=.17 C0V0 =.048 (1) =1.0 =.048 (1) =1.0 = .055 (1) =.68 =.055 (1)=.66 C0V1 =.055 (1) =.79 =.055 (1) =.80 = .055 (1) =.32 =.054 (1)=.31 C1V0 =.214 (1) =.95 =.214 (1) =.95 = .055 (1) =.67 =.055 (1)=.66 C1V1 =.102 (1) =.54 =.102 (1) =.54 = .055 (1) =.32 =.054 (1)=.31 N= 12, =.5, B=2 Tests for equal treatment effects. Analysis Assumptions N= 18, =.5, B=2 Simulation C0V0 C0V1 C1V0 C1V1 C0V0 =.046 (1) =1.0 =.046 (1) =1.0 = .045 (1) =.83 =.045 (1)=.83 C0V1 =.040 (1) =.92 =.040 (1) =.92 = .034 (1) =.47 =.034 (1)=.46 C1V0 =.297 (1) =.99 =.297 (1) =.99 = .045 (1) =.83 =.045 (1)=.83 C1V1 =.117 (1) =.69 =.117 (1) =.69 = .034 (1) =.47 =.034 (1)=.46 C0V0 =.051 (1) =1.0 =.051 (1) =1.0 = .061 (1) =.96 =.061 (1)=.96 C0V1 =.055 (1) =.99 =.055 (1) =.99 = .057 (1) =.67 =.055 (1)=.67 C1V0 =.507 (1) =1.0 =.508 (1) =1.0 = .061 (1) =.96 =.061 (1)=.96 C1V1 =.230 (1) =.92 =.230 (1) =.92 = .054 (1) =.67 =.054 (1)=.67 N= 30, =.5, B=2 NOTE: Failing to assume carryover when carryover exists invalidates the tests for equal treatment effects and the invalidation generally gets worse as the THREE TREATMENT - THREE PERIOD CROSSOVER DESIGN Period 1 1 2 A A SEQUENCE 3 4 5 B B C 6 C Period 2 B C A C A B Period 3 C B C A B A Carryover Case: Table 1. Model parameters for the 3 period/3 treatment crossover design with carryover. Sequence Per 1 2 3 + 1+ A + 1+ B 4 5 +1+B +1+ C +2+A+C 1 + 1+A 2 +2+ B+A + 2+C+A + 2+A+B + 2+C+B 3 + 3+C+B + 3+B+C + 3+C+A + 3+A+C + 3+ B+A 6 + 1+C + 2+B+C + 3+A+B Let represent the expected response in Period i of Sequence j. that Note Suppose nj subjects are assigned to the jth sequence of treatments. Let yijk denote the observed response from the kth subject in the jth sequence during the ith period; i=1,2,3; j=1,2,...,6; k=1,2,...,nj. Means Model: yijk = ij + jk + ijk for all i,j,k where represents the expected response in Period i of Sequence j. Ideal Conditions: jk ~ iid N(0,2), ijk ~iid N(0,2), and all jk’s and ijk’s are independent. Remarks: A crossover experiment is really a special type of a repeated measures experiment where the treatment is changing over time. We know that traditional ANOVA analyses of repeated measures experiments are only valid when the repeated measures satisfy compound symmetry and tests for differences across time points are valid if and only if the repeated measures satisfy the Hyuhn-Feldt Conditions. Question: What do the above remarks have to do with the validity of our analyses of crossover experiments described previously? The analyses that we have performed up to this point in time using our ideal conditions are valid if the vector of measurements on a subject within each sequence satisfies compound symmetry. That is, the variance of the measured response is the same for each treatment*period combination and the correlation between measurements in different periods is the same for all pairs of periods. Again, let yjk be the vector of responses for the kth subject in the jth sequence of treatments, and suppose we assume that yjk ~ N( j, ) where 1 j 2j j pj , 11 12 22 21 p1 p 2 1p 2p , and pp ij i S j * and where and are determined by i and j. Remark: The covariance matrix is said to possess compound symmetry if = 2[(1- )I + J]for some 2 and . Remark: The covariance matrix satisfy the H-F conditions if is said = I + j’ + j ’ for some and . to Goad and Johnson (2000) showed: (1) If satisfies the H-F conditions, then the traditional tests for treatment and period effects are valid for all crossover experiments both with and without carryover. (2) There are cases where the ANOVA tests are valid even when does not satisfy the H-F conditions. (a) In the no carryover case, tests for equal treatment effects are valid for the six sequence three period/three treatment crossover design when there are an equal number of subjects assigned to each sequence. (b) In the no carryover case, tests for equal period effects are valid only when the H-F conditions be satisfied (b) The traditional tests for equal treatment effects and equal period effects are valid for a crossover design generated by t-1 mutually orthogonal tt Latin squares when there are equal numbers of subjects assigned to each sequence. (c) The traditional tests for equal treatment effects, equal period effects, and equal carryover effects are likely to be invalid in the four period/four treatment design regardless of whether carryover exists or not. Cases where the validity of ANOVA tests are still in doubt. (4) When carryover exists, the tests for equal carryover effects are not valid unless satisfies the H-F conditions. (5) When there are unequal numbers of subjects assigned to each sequence, the ANOVA tests are unlikely to be valid unless satisfies the H-F conditions. Goad and Johnson (2000) provide some alternative analyses for crossover experiments. Consider again, the three period/three treatment crossover design in six sequences. Question: Suppose the variance of a response depends on the treatment, but that the correlation is the same between all pairs of sequence cells. That is, for Sequence 1, the covariance matrix is: A2 Σ1 B A C A A B B2 C B A C B C C2 Shanga simulated three period/three treatment crossover experiments satisfying four different conditions: (1) (2) (3) (4) no carryover and equal variances (C0V0), no carryover and unequal variances(C0V1), carryover and equal variances (C1V0), and carryover and unequal variances (C1V1). Each of 1000 sets of data under each of these conditions was analyzed four different ways assuming: (1) (2) (3) (4) no carryover and equal variances (C0V0), no carryover and unequal variances(C0V1), carryover and equal variances (C1V0), and carryover and unequal variances (C1V1). TITLE1 'CRSOVR EXAMPLE - A THREE PERIOD/THREE TRT DESIGN'; TITLE2 'ASSUMES CARRYOVER AND UNEQUAL VARIANCES'; PROC MIXED; CLASSES SEQ PER TRT PRIORTRT SUBJ; MODEL Y = SEQ TRT PER PRIORTRT/DDFM=KR; LSMEANS TRT PER PRIORTRT/PDIFF; REPEATED TRT/SUBJECT=SUBJ TYPE=CSH; RUN; Tests for equal treatment effects. N= 6 =.5 B=2 C=4 Analysis Assumptions Simulation C0V0 C0V1 C1V0 C1V1 C0V0 =.053 (1) =1.0 =.057 (1) =1.0 = .057 (1) =1.0 =.051 (1)=1.0 C0V1 =.066 (1) =.50 =.057 (1) =.88 = .049 (1) =.42 =.049 (1)=.67 C1V0 =.138 (1) =1.0 =.149 (1) =1.0 = .057 (1) =1.0 =.051 (1)=1.0 C1V1 =.070 (1) =.32 =.069 (1) =.73 = .049 (1) =.42 =.049 (1)=.67 Tests for equal treatment effects. N= 12 =.5 B=2 C=4 Analysis Assumptions Simulation C0V0 C0V1 C1V0 C1V1 C0V0 =.049 (1) =1.0 =.052 (1) =1.0 = .054 (1) =1.0 =.055 (1)=1.0 C0V1 =.070 (1) =.89 =.053 (1) =.99 = .055 (1) =.77 =.046 (1)=.94 C1V0 =.227 (1) =1.0 =.232 (1) =1.0 = .054 (1) =1.0 =.055 (1)=1.0 C1V1 =.081 (1) =.70 =.100 (1) =.97 = .055 (1) =.77 =.046 (1)=.94 Tests for equal treatment effects. N= 18 =.5 B=2 C=4 Analysis Assumptions Simulation C0V0 C0V1 C1V0 C1V1 C0V0 =.054 (1) =1.0 =.056 (1) =1.0 = .048 (1) =1.0 =.053 (1)=1.0 C0V1 =.071 (1) =.99 =.051 (1) =1.0 = .054 (1) =.91 =.051 (1)=.99 C1V0 =.370 (1) =1.0 =.378 (1) =1.0 = .048 (1) =1.0 =.053 (1)=1.0 C1V1 =.094 (1) =.90 =.125 (1) =1.0 = .054 (1) =.91 =.051 (1)=.99 Tests for Carryover Simulation C1V0 C1V1 N= 6 A=1 B=1 C=1 = .043 (.5) =.52 (1) =.99 =.040 (.5)=.53 (1) =.99 N= 6 A=1 B=.5 C=.25 = .054 (.5) =.86 (1) =1.0 =.044 (.5)=.99 (1) =1.0 N= 6 A=1 B=2 C=4 = .048 (.5) =.10 (1) =.25 =.044 (.5)=.13 (1) =.35 Tests for Carryover Simulation C1V0 C1V1 N= 12 A=1 B=1 C=1 = .040 (.5) =.85 (1) =1.0 =.042 (.5)=.85 (1) =1.0 N= 12 A=1 B=.5 C=.25 = .047 (.5) =1.0 (1) =1.0 =.046 (.5)=1.0 (1) =1.0 N= 12 A=1 B=2 C=4 = .064 (.5) =.15 (1) =.45 =.056 (.5)=.20 (1) =.62 In the three treatment/three period/six sequence crossover design, Shanga also considered testing H 0 : A2 B2 C2 Shanga claimed that his tests were LRTs, but Jung (2008) has shown that they are not LRTs. Nevertheless, Shanga's tests had good power for detecting unequal variances. That’s All For Now!