Table 4.
Secondary analysis results
KFQ-c | SCAS-c | SCAS-p | CAIS-c | CAIS-p | ADIS-c/p CSR | ||
---|---|---|---|---|---|---|---|
Treatment | |||||||
GPD-CBT vs SFBT (ref) | 0·817 (−2·785 to 4·418); p=0·65 | 4·944 (−0·932 to 10·820); p=0·098 | –2·072 (−5·733 to 1·590); p=0·27 | 1·129 (−2·668 to 4·927); p=0·56 | –0·832 (−4·008 to 2·344); p=0·60 | –0·177 (−0·783 to 0·429); p=0·56 | |
Timepoint | |||||||
After treatment (ref) vs 6 month follow–up | 2·737 0·847 to 4·627); p=0·0049 | 5·267 (2·491 to 8·042); p=0·0003 | 2·939 (1·078 to 4·798); p=0·0022 | 0·617 (−1·106 to 2·340); p=0·48 | 1·931 (0·411 to 3·451); p=0·013 | 0·765 (0·432 to 1·097); p<0·0001 | |
Treatment by timepoint interaction | –1·917 (−5·613 to 1·778); p=0·31 | 3·72 (−1·596 to 9·053); p=0·17 | 1·99 (−1·617 to 5·602); p=0·28 | –1·665 (−4·963 to 1·632); p=0·32 | 1·839 (−1·111 to 4·789); p=0·22 | –0·210 (−0·860 to 0·440); p=0·52 | |
GPD-CBT vs SFBT (ref) after treatment | –0·142 (−3·99 to 3·703); p=0·94 | 6·808 (0·312 to 13·305); p=0·040 | –1·705 (−4·885 to 2·734); p=0·57 | 0·296 (−3·414 to 4·008); p=0·87 | 0·087 (−3·521 to 3·696); p=0·96 | –0·282 (−0·958 to 0·394); p=0·41 | |
GPD-CBT vs SFBT (ref) at 6 month follow-up | 1·775 (−2·46 to 6·015); p=0·41 | 3·08 (−3·33 to 9·487); p=0·34 | –3·067 (−7·406 to 1·271); p=0·16 | 1·962 (−2·568 to 6·493); p=0·39 | –1·751 (−5·143 to 1·640); p=0·30 | –0·072 (−0·770 to 0·627); p=0·84 | |
Child's baseline endpoint score | 0·764 (0·597 to 0·931); p<0·0001 | 0·637 (0·487 to 0·787); p<0·0001 | 0·602 (0·478 to 0·726); p<0·0001 | 0·456 (0·305 to 0·608); p=<0·0001 | 0·724 (0·570 to 0·878); p<0·0001 | 0·515 (0·221 to 0·809); p=0·0007 | |
Sex | |||||||
Male (ref) vs female | –1·440 (−5·152 to 2·272); p=0·44 | –4·340 (−10·372 to 1·693); p=0·16 | 2·022 (−1·707 to 5·750); p=0·29 | –2·587 (−6·257 to 1·083); p=0·17 | –0·296 (−3·527 to 2·934); p=0·86 | 0·118 (−0·499 to 0·734); p=0·71 | |
Age | 0·018 (−1·017 to 1·054); p=0·97 | 0·364 (−1·628 to 2·355); p=0·72 | –0·040 (−1·025 to 0·944); p=0·94 | –0·132 (−1·304 to 1·039); p=0·82 | 0·632 (−0·271 to 1·534); p=0·17 | –0·141 (−0·309 to 0·028); p=0·1005 | |
Parental level of anxiety | p=0·68* | p=0·53* | p=0·29* | p=0·80* | p=0·12* | p=0·99* | |
Borderline (ref) vs normal | –0·957 (−5·733 to 3·819); p=0·69 | –2·340 (−10·146 to 5·467); p=0·55 | –3·523 (−8·334 to 1·288); p=0·15 | –0·115 (−5·046 to 4·818); p=0·96 | –4·135 (−8·432 to 0·163); p=0·059 | 0·000† (−0·820 to 0·820) | |
Borderline (ref) vs high | –3·600 (−11·715 to 4·516); p=0·38 | –7·075 (−19·447 to 5·297); p=0·26 | –4·980 (−13·208 to 3·248); p=0·23 | –2·374 (−10·162 to 5·414); p=0·55 | –0·075 (−8·383 to 8·233); p=0·99 | 0·078 (−1·319 to 1·475); p=0·91 | |
Child's primary anxiety disorder | p=0·54* | p=0·92* | p=0·023* | p=0·55* | p=0·077* | p=0·063* | |
GAD (ref) vs SAD | –2·844 (−7·893 to 2·205); p=0·27 | –2·701 (−11·151 to 5·749); p=0·53 | 2·549 (−2·585 to 7·682); p=0·33 | 2·678 (−2·538 to 7·894); p=0·31 | 3·559 (−1·109 to 8·226); p=0·13 | 0·488 (−0·366 to 1·342); p=0·26 | |
GAD (ref) vs social phobia | 2·384 (−4·498 to 9·267); p=0·49 | 0·207 (−13·541 to 13·955); p=0·98 | –5·200 (−12·182 to 1·781); p=0·14 | –3·086 (−10·375 to 4·203); p=0·40 | –4·188 (−10·208 to 1·831); p=0·17 | –0·794 (−1·889 to 0·302); p=0·15 | |
GAD (ref) vs other | 0·698 (−3·782 to 5·178); p=0·76 | 0·410 (−6·652 to 7·472); p=0·91 | −5·770 (−10·422 to −1·118); p=0·016 | 1·055 (−3·382 to 5·492); 0·64 | –2·299 (−6·311 to 1·712); p=0·26 | –0·676 (−1·454 to 0·102); p=0·088 |
Data are estimate (95% CI); p value. KFQ-c=Koala Fear Questionnaire-child report. SCAS-c=Spence Children's Anxiety Scale-child report. SCAS-p=Spence Child Anxiety Scale–parent report. CAIS-c=Child Anxiety Impact Scale–child report. CAIS-p=Child Anxiety Impact Scale-parent report. ADIS-c/p=Anxiety Disorders Interview Schedule for DSM-IV child and parent version. CSR=Clinical Severity Rating. GPD-CBT=guided parent-delivered cognitive behavioural therapy. SFBT=solution-focused brief therapy. GAD=generalised anxiety disorder. SAD=social anxiety disorder.
p values assessing the overall effect of categorical factors with three or more levels do not have a corresponding single effect measure so estimates and 95% CIs are not presented.
Final figure as provided by the statistical programme.