Transdiagnostic group cognitive behavioural therapy for emotional disorders in primary care: The results of the PsicAP randomized controlled trial

Psychological Medicine camb idge.o g/psm O iginal A icle Ci e his a icle: Cano-Vindel A, Muñoz- Na a o R, Mo iana JA, Ruiz-Rod íguez P, Med ano LA, González-Blanch C (2021). T ansdiagnos ic g oup cogni i e beha iou al he apy o emo ional diso de s in p ima y ca e: he esul s o he PsicAP andomized con olled ial. Psychological Medicine 1–13. h ps://doi.o g/10.1017/S0033291720005498 Recei ed: 31 July 2020 Re ised: 28 No embe 2020 Accep ed: 30 Decembe 2020 Key wo ds: Anxie y; dep ession; emo ional diso de s; g oup psycho he apy; andomized clinical ial; soma iza ion; ansdiagnos ic Au ho o co espondence: Roge Muñoz Na a o, E-mail: [email p o ec ed] © The Au ho (s), 2021. Published by Camb idge Uni e si y P ess. This is an Open Access a icle, dis ibu ed unde he e ms o he C ea i e Commons A ibu ion- NonComme cial-NoDe i a i es licence (h p:// c ea i ecommons.o g/licenses/by-nc-nd/4.0), which pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any medium, p o ided ha no al e a ions a e made and he o iginal a icle is p ope ly ci ed. The w i en pe mission o Camb idge Uni e si y P ess mus be ob ained p io o any comme cial use and/o adap a ion o he a icle. T ansdiagnos ic g oup cogni i e beha iou al he apy o emo ional diso de s in p ima y ca e: he esul s o he PsicAP andomized con olled ial An onio Cano-Vindel1, Roge Muñoz-Na a o2, Juan A. Mo iana3, Paloma Ruiz-Rod íguez4, Leona do Ad ián Med ano5and Césa González-Blanch6 1 Facul y o Psychology, Complu ense Uni e si y o Mad id, Campus de Somosaguas, s/n, 28223 Mad id, Spain; 2 Depa men o Psychology and Sociology, Facul y o Social and Human Sciences, Uni e si y o Za agoza, C/Cdad. Escola , S/N, 44003, Te uel, Spain; 3 Depa men o Psychology, Uni e si y o Có doba/ Maimónides Ins i u e o Resea ch in Biomedicine o Co doba-IMIBIC/Reina So ía Uni e si y Hospi al, A . Menéndez Pidal, s/n, 14004 Có doba, Spain; 4 Cas illa La Nue a P ima y Ca e Cen e, Heal h Se ice o Mad id, Calle Te uel, 4, 28941 Fuenlab ada, Mad id, Spain; 5 Facul y o Psychology, Uni e si y Siglo 21, De los La inos 8555, 5008 Có doba, A gen ina and 6 Men al Heal h Cen e, Uni e si y Hospi al “Ma qués de Valdecilla”- IDIVAL, A . Valdecilla, 25, 39008 San ande , Can ab ia, Spain Abs ac Backg ound. Emo ional diso de s a e highly p e alen in p ima y ca e. We aimed o de e - mine whe he a ansdiagnos ic psychological he apy plus ea men -as-usual (TAU) is mo e e icacious han TAU alone in p ima y ca e adul pa ien s. Me hods. A andomized, wo-a m, single-blind clinical ial was conduc ed in 22 p ima y ca e cen es in Spain. A o al o 1061 adul pa ien s wi h emo ional diso de s we e en olled. The ansdiagnos ic p o ocol (n= 527) consis ed o se en 90-min sessions (8–10 pa ien s) deli e ed o e a 12–14-week pe iod. TAU (n= 534) consis ed o egula consul a ions wi h a gene al p ac i ione . P ima y ou come measu es we e sel - epo ed symp oms o anxie y, dep ession, and soma iza ions. Seconda y ou come measu es we e unc ioning and quali y o li e. Pa ien s we e assessed a baseline, pos - ea men , and a 3, 6, and 12 mon hs. In en ion- o- ea and pe -p o ocol analyses we e pe o med. Resul s. Pos - ea men p ima y ou comes we e signi ican ly be e in he ansdiagnos ic g oup compa ed o TAU (anxie y: p< 0.001; Mo is’sd=−0.65; dep ession: p< 0.001; d= −0.58, and soma ic symp oms: p< 0.001; d=−0.40). These e ec s we e sus ained a he 12- mon h ollow-up (anxie y: p< 0.001; d=−0.44; dep ession: p< 0.001; d=−0.36 and soma ic symp oms: p< 0.001; d=−0.32). The ansdiagnos ic g oup also had signi ican ly be e ou - comes on unc ioning (d= 0.16–0.33) and quali y o li e domains (d= 0.24–0.42), wi h sus- ained imp o emen a he 12-mon h ollow-up in unc ioning (d= 0.25–0.39) and quali y o li e (d= 0.58–0.72). Reliable eco e y a es showed la ge be ween-g oup e ec sizes (d> 0.80) in a ou o he ansdiagnos ic g oup a e ea men and a he 12-mon h ollow-up. Conclusions. Adding a b ie ansdiagnos ic psychological in e en ion o TAU may signi i- can ly imp o e ou comes in emo ional diso de s ea ed in p ima y ca e. T ial Regis a ion. is c n.o g iden i ie : ISRCTN58437086 In oduc ion Emo ional diso de s –dep ession, anxie y, and soma o o m diso de s –a e all highly p e a- len in he communi y, imposing an eno mous bu den on socie y (Whi e o d, Fe a i, Degenha d , Feigin, & Vos, 2015). In pa icula , dep ession and anxie y ep esen a global bu - den ha is e en g ea e han he impac o ch onic physical condi ions (Vigo, Tho nic o , & A un, 2016). Gi en he la ge nega i e consequences o hese diso de s, i is clea ha men al heal h should be a majo p io i y o all heal h sys ems (Pa el e al., 2018). One app oach o imp o ing access o e idence-based ea men s would be o in eg a e men al heal h ca e in o he p ima y ca e se ing (Wo ld Heal h O ganiza ion, 2018), whe e mos pa ien s wi h mild o mode a e emo ional diso de s a e ea ed (Ko ess-Mas e y e al., 2007). Many s udies ha e shown ha psychological he apy, mainly cogni i e beha iou al he apy (CBT), is an e ec i e ea men o emo ional diso de s in p ima y ca e (Cuijpe s e al., 2019b; Seekles e al., 2013). In ecen yea s, se e al ambi ious heal h ca e ini ia i es, such as he Imp o ing Access o Psychological The apies (IAPT) p ojec in he UK, ha e been ca ied ou o expand access o e idence-based psychological ea men s o common men al diso de s h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a o a wide popula ion. The esul s o he IAPT and simila p o- jec s show ha hese ini ia i es a e cos -e ec i e (Laya d & Cla k, 2015) and highly bene icial o socie y (Cla k, 2018; Wake ield e al., 2020). The pionee ing IAPT p ojec p o ided a model o simila in e na ional p ojec s, such as hose ca ied ou in Aus alia (C oma y, D ummond, F ancis, Wa son, & Ba e sby, 2016), No way (Knaps ad e al., 2018,2020), and Canada (Naeem, Pika d, Rao, Ayub, & Munshi, 2017). None heless, se - e al ba ie s o dissemina ion o psychological ea men s ha e been iden i ied (Ha ey & Gumpo , 2015), including he g owing numbe o diso de -speci ic ea men guidelines (Newby, McKinnon, Kuyken, Gilbody, & Dalgleish, 2015), which may be pa icula ly ele an in p ima y ca e whe e indi iduals wi h emo- ional diso de s equen ly p esen mixed symp oms a ibu able o di e en men al heal h diso de s, and p ima y ca e p o ide s o en a e no able o make a p ecise di e en ial diagnosis due o ime and o he cons ain s (Tylee & Wal e s, 2007). In he pas wo decades, he e has been a g owing e o o de elop psychological ea men s based on a ansdiagnos ic app oach, suppo ed by e idence showing ha many men al diso de s sha e he same psychological p ocesses implica ed in he onse and main enance o psychopa hology (Aldao, Nolen-Hoeksema, & Schweize , 2010; No on & Hope, 2005; Pa el e al., 2018). The a ailable e idence sugges s ha ansdiag- nos ic psychological he apy can be a highly e ec i e ea men o emo ional diso de s (Newby e al., 2015). This he apeu ic app oach ocuses on ea ing he common ac o s in ol ed in many emo ional diso de s such as cogni i e biases (Beck, 2019; Eysenck & De akshan, 1997) and dys unc ional emo ion egula ion s a egies (Saki is & Be le, 2019; Sloan e al., 2017), which a e bo h cogni i e and beha iou al (Na agon-Gainey, McMahon, & Chacko, 2017). Mos ecen esea ch has ocused on indi idual ea men s –o , in some cases, i ual he apy (i.e. in e ne -based) –bu ew s udies ha e es ed he e icacyo ansdiagnos ic g oup cog- ni i e beha iou al he apy (TD-GCBT) (Chambe lain & No on, 2013; No on & Ba e a, 2012). The a ailable e idence, al hough limi ed, sugges s ha he ansdiagnos ic g oup app oach migh be a pa icula ly use ul, cos -e ec i e ea men gi en he high como bidi y among men al diso de s, especially in p ima y ca e. As in many o he Eu opean coun ies, he Na ional Heal h Sys em in Spain (Spanish ac onym: SNS) is based on he p inci- ples o uni e sal co e age, ee access, and ai ness. The SNS is p ima ily unded by axes. The sys em i sel is cen ed a ound p i- ma y ca e cen es, wi h he popula ion assigned o a gi en cen e based on geog aphic p oximi y. Consequen ly, in his model, he gene al p ac i ione s (GP) and nu sing s a ac as he ga ekeepe s o he SNS, excep o eme gency ca e. P ima y ca e cen es a e s a ed by mul idisciplina y eams comp ised by GPs, paedia i- cians, nu ses, and adminis a i e s a ; some cen es also ha e social wo ke s, midwi es, and/o physio he apis s. The co e package o p ima y heal h ca e bene i s comp omises all heal h ca e p e en ion, diagnosis, ea men and ehabili a ion se ices, including men al heal h. Un o una ely, in Spain, p ac ically none o he egions has men al heal h p o essionals on s a a he p ima y ca e le el. Ra he , clinical psychologis s wo king in specialized ca e can pe iodically be sen o p ima y ca e cen es o men al heal h ca e issues, bu his p ac ice is e y uncommon. The e o e, gi en he minimal p esence o clinical psychologis s in p ima y ca e, he p o ision o psychological he apy mainly elies on e e al o specialized ca e, o which wai ing imes a e ypic- ally long. Thus, mos pa ien s wi h emo ional diso de s a e ea ed di ec ly by hei GPs, and his ea men ends o be medica ion- cen ic, despi e he ecommenda ions p o ided by mos clinical guidelines. The use o psycho opic d ugs (mainly an idep essan s, anxioly ics and hypno ics) has been inc easing in Spain and hei use is cu en ly highe han in many o he Eu opean coun ies (OECD, 2015). In his con ex , we conduc ed a andomized con olled ial (RCT) o compa e TD-GCBT (7 sessions) plus ea men -as-usual (TAU) o TAU alone in adul pa ien s wi h emo ional diso de s in he p ima y ca e se ing. We hypo hesized ha TD-GCBT + TAU would be mo e e ec i e han TAU alone in educing anxie y, dep ession, and soma ic symp oms, and ha hese bene i s would be sus ained 3, 6, and 12 mon hs a e ea men inaliza- ion. We u he hypo hesized ha TD-GCBT + TAU would, compa ed o TAU alone, educe disabili y, imp o e quali y o li e, and inc ease ea men sa is ac ion and ha hese bene i s would also be sus ained o e ime. Subjec s and me hods S udy design The PsicAP ial was a mul icen e, wo-a m, single-blind, RCT. Pa ien s we e ec ui ed in he p ima y ca e se ing wi hin he SNS wi h symp oms o an emo ional diso de (dep ession, anx- ie y diso de , o soma iza ion) and andomized o ecei e ei he TAU alone (con ol g oup) o combined ea men in ol ing TD-GCBT + TAU. Pa icipan s Pa ien s we e ec ui ed om 22 p ima y ca e cen es in eigh di - e en egions in Spain (Andalusia, Basque egion, Can ab ia, Cas illa la Mancha, Galicia, Mad id, Na a a, and Valencia) (Cano-Vindel e al., 2016). All pa ien s who isi ed hei GPs wi h signs o symp oms o nega i e o unpleasan emo ional p o- blems, mode a e dep ession, anxie y, o soma ic symp oms wi h- ou any clea biological basis we e conside ed candida es o s udy inclusion. Pa ien s ecei ing ea men wi h an idep essan s, anxioly ics, and/o hypno ics we e also eligible and in i ed o pa - icipa e by hei ea ing GP. Rec ui men The GPs explained he s udy o po en ial pa icipan s du ing he cou se o a ou ine clinical isi . Pa ien s who ag eed o pa icipa e we e p o ided wi h pa ien in o ma ion shee wi h w i en de ails abou he s udy and hen asked o sign an in o med consen o m, a e which an ini ial session wi h a psychologis was scheduled. A his isi , he s udy pa icipan s comple ed a ba e y o elec- onic ques ionnai es. The main s udy inclusion c i e ia we e (1) age be ween 18 and 65 yea s and (2) he p esence o symp- oms sugges i e o an emo ional diso de , whose p esence was ini- ially assessed wi h he Pa ien Heal h Ques ionnai e (PHQ; Spi ze e al., 1999). Exclusion c i e ia included any o he ollowing se e e men al diso de s: ea ing diso de s; alcohol o subs ance abuse; bipola diso de ; se e e majo dep essi e diso de ; ecen suicide a emp , o o he se e e men al diso de s diagnosed by he GP. The spe- ci ic PHQ modules ha assess o he p esence o ea ing diso de s and alcohol abuse we e used o de ec hese condi ions and o exclude pa ien s wi h hese diso de s. Pa icipan s ecei ing psy- chological ea men o any men al diso de we e also excluded. 2 An onio Cano‐Vindel e al. h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a Randomiza ion and masking A compu e -gene a ed alloca ion sequence was used o andomly assign pa ien s (1:1) o ecei e ei he TD-GCBT + TAU o TAU alone. S udy pa icipan s we e con ac ed by email o phone o in o m hem o hei ea men alloca ion. A clinical psychologis was assigned o lead a speci ic TD-GCBT in e en ion. Pa ien s alloca ed o TAU we e ins uc ed o e u n o hei GPs o ea - men . None o he pa icipan s o clinicians was blinded o he ea men alloca ion. Howe e , in acco dance wi h he single- blind s udy design, he assesso s in ol ed in he p e- and pos - ea men assessmen phases we e blinded o he alloca ion and did no pa icipa e in he in e en ions (TD-GCBT o TAU). The GPs did no ecei e any in o ma ion om he esea che s ega ding baseline assessmen s, andomiza ion, p og ess in he in e en ion, o o he ou comes. Medical eco ds and no es we e no sha ed be ween he in e en ion eam and p ima y ca e p o ide s. P ocedu es In e en ions The TD-GCBT p o ocol (Cano-Vindel e al., 2016; González- Blanch e al., 2018b) was a planned p og am consis ing o se en 90-min he apy sessions held o e a 12–14-week pe iod in small g oups (8–10 pa ien s) in he p ima y ca e cen e. These sessions we e led by ained clinical psychologis s, which we e no pa o he p ima y ca e s a . In o de o ensu e he ideli y and consis ence o he TD-GCBT ea men a all pa icipa ing si es, all he apis s we e equi ed o unde go an 8-h aining p og am in he ea men p o ocol. This aining session was led by a senio clinical psychologis . All he apis s ecei ed a de ailed, session-by-session ou line o he ea men . The he apeu ic app oach was based on he ansdiagnos ic app oach o emo ional diso de s, which assumes ha mos emo ional diso de s sha e se e al common ac o s (Aldao e al., 2010; Ho mann & Ba low, 2014), and ha he onse and main enance o emo ional diso de s a e due o dys egula ed cogni i e-beha iou al emo ion egula ion s a egies (Aldao e al., 2010). Any pa icipan who missed a ain- ing session was con ac ed by elephone by an assis an esea che and o e ed o a end he nex session. A he s a o each aining session, he con en o he p e ious session was b ie ly e iewed. The TAU in e en ion consis ed o egula consul a ions wi h he ea ing GP, who assessed he pa ien ’s physical and/o psy- chosocial complain s. GPs we e ins uc ed o ea pa ien s in bo h a ms in acco dance wi h hei bes clinical judgmen . In gene al, hese ea men s in ol ed he p esc ip ion o anxioly ics, an idep essan s, o hypno ics, and/o in o mal counselling/ suppo . P ima y ou comes The p ima y ou come measu e was he se e i y o symp oms o emo ional diso de s (anxie y, dep ession, and soma ic symp- oms), which we e assessed by he ele an PHQ modules (Spi ze e al., 1999). The PHQ has been alida ed as a sensi i e and speci ic es o de e mine he p esence o hese diso de s based on he DSM-IV c i e ia. In a p e ious s udy ha included a subse (15%, n= 178) o he pa ien s in he cu en ial, we compa ed wo PHQ modules o assess symp oms o anxie y (Muñoz-Na a o e al., 2017a) and dep ession (Muñoz-Na a o e al., 2017a) wi h clinical in e iews ( he gold s anda d) o alid- a e hese ins umen s in he p esen ial. Symp oms o Anxie y (GAD-7). The p esence o anxie y was de e mined acco ding o he Gene alized Anxie y Diso de -7 scale (Spi ze e al., 2006). To al GAD-7 sco es ange om 0 o 21. Fo his ial, he cu -o o clinical signi icance on he GAD-7 was se a ⩾10. In e nal consis ency o he scale was good (α= 0.87). A his cu -o poin , bo h sensi i i y (0.87) and speci ici y (0.78) we e accep able (Muñoz-Na a o e al., 2017a). Mean sco es we e 12.3 (4.6) and he in e nal consis ency o he scale was good (α= 0.87). Symp oms o Dep ession (PHQ-9). This PHQ module (K oenke e al., 2001) was used o de ec dep ession wi h a cu -o poin o ⩾10. To al sco es anged om 10 o 23. Based on a p e ious s udy o his cu -o sco e (Muñoz-Na a o e al., 2017a), pa ici- pan s who sco ed om 20 o 23 we e assigned o unde go a second-o de assessmen (in e iew wi h a clinical psychologis ) o con i m he p esence o mode a e o se e e dep ession dis- o de . Based on his clinical assessmen , he pa ien s we e included o no in he ial (Cano-Vindel e al., 2016). The mean sco e on he PHQ-9 was 13.6 (5.4) and he scale showed a good in e nal consis ency (α= 0.86). The mos widely used cu -o sco e in he li e a u e (⩾10) was used as he h eshold o caseness, as his cu poin p esen s a eliable balance be ween sensi i i y and speci ici y (K oenke e al., 2001). Symp oms o soma iza ions (PHQ-15). On he Spanish e sion o he PHQ (Spi ze e al., 1999), pa ien s a e 13 soma ic symp oms on a scale om 0 o 2, as ollows: 0 (no bo he ed), 1 (bo he ed a li le), o 2 (bo he ed a lo ). Two i ems om he dep ession mod- ule (sleep and i edness) a e added and sco ed as ollows: 0 (no a all), 1 (se e al days), o 2 (mo e han hal he days o nea ly e e y day). The eliabili y o his scale was accep able (α= 0.80), wi h a mean sco e o 14.0 (4.8). Soma ic symp oms we e assessed acco ding o he PHQ-15 sum sco e, wi h a maximum sco e o 30. Pa ien inclusion was based on an algo i hm wi h he 13 som- a ic symp oms (pa ien s wi h ⩾3 symp oms we e classi ied as ‘bo he ed a lo ’). A p e ious s udy used his algo i hm o de e - mine he sensi i i y (78%) and speci ici y (71%) o he PHQ-15 (K oenke e al., 2010). Seconda y ou comes Func ional S a us. Seconda y ou come measu es included he le el o disabili y on daily li e domains (wo k, social, and amily li e) measu ed wi h he Sheehan Disabili y Scale (SDS; Luciano e al., 2010). The in e nal consis ency o his scale was good (α= 0.80). Quali y o Li e. Quali y o li e domains (physical, psychological, social, and en i onmen al) was e alua ed wi h he Wo ld Heal h O ganiza ion Quali y o li e Ins umen -Abb e ia ed e - sion (WhoQoL-B e ; Lucas-Ca asco, 2012). In e nal consis ency o all ac o s was accep able (α> 0.70): (physical: α= 0.77; psy- chological: α= 0.79; social: α= 0.70; en i onmen al: α= 0.79). T ea men Sa is ac ion. T ea men sa is ac ion o he g oups was assessed h ough a single ques ion (‘Ra e you sa is ac ion wi h he ea men ecei ed’) wi h esponses gi en on a 10-poin scale an- ging om 0 ( o ally unsa is ied) and 10 ( o ally sa is ied). S a is ical me hods Powe and Sample size. To de ec a minimally impo an e ec size o ⩾0.2 wi h an alpha e o p obabili y o 0.05 and a powe Psychological Medicine 3 h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a (1–β) o 0.80 using G*Powe 3.1 o SPSS (Faul, E d elde , Lang, & Buchne , 2007), a sample size o 394 pa icipan s pe g oup would be equi ed. An icipa ing a d opou a e o 25%, he neces- sa y sample size would be 525 pa icipan s pe g oup. Fo all esul s, we applied 95% con idence in e als (CI). Main Analyses. Fo p ima y ou comes, we conside ed g oup di - e ences in anxie y (GAD-7), dep ession (PHQ-9), and soma ic symp oms (PHQ-15) compa ing baseline and pos - ea men sco es using a mixed-e ec model. A mixed-e ec s model was compu ed, including ime and ea men g oup as ixed e ec s and single pa icipan s as a andom e ec ; g oup di e ences we e analysed a e con olling o baseline cha ac e is ics: gende , age, and ea men cen e. Fo seconda y ou comes, we simila ly checked be ween-g oup di e ences in he le el o disabili y on daily li e domains (wo k, social, and amily li e) and quali y o li e domains (physical, psychological, social, and en i onmen al). We pe o med an in en ion- o- ea (ITT) analysis ha included all andomized pa ien s using he chained equa ions mul iple impu a ion p ocedu e in he SPSS s a is ical so wa e p og am, wi h i e impu a ions. The e ec sizes o he ea men on p ima y and seconda y ou comes (mean sco es) we e calcula ed by apply- ing Mo is’sds a is ic. Mo is (2008) desc ibed an e ec size o he p e-pos change (PPC) design, whe e he s anda dized e ec o he ea men is de ined as he di e ence be ween g oups in mean PPC alues, di ided by he common s anda d de ia ion. The o mula o he Mo is’e ec size is as ollow: [δ PPC = (μ T2 –μ T1 )−(μ C2 –μ C1 )/σ], whe e μ g is he mean o g oup g a ime , and σis he s anda d de ia ion o he un ea ed popu- la ion. The main ad an age o his o mula is ha i akes in o accoun he mean and s anda d de ia ion o he sample bo h a he inal assessmen and a baseline, leading o a mo e ep esen- a i e e ec size when he alues di e ed om baseline. Addi ional analyses: Reco e y, eliable eco e y, and de e io - a ion a es we e calcula ed. The eco e y index was de ined as p e- ea men sco es abo e he h eshold on any o he h ee scales and below he h eshold on all scales a ei he he pos - ea men o 12-mon h ollow-up assessmen . The eliable eco e y a e was calcula ed using a change sco e based on he s anda d de ia ion (S.D.) and C onbach’s alpha o each measu e (as desc ibed in he IAPT p ojec ) o accoun o scale measu emen e o s (Cla k e al., 2009). Thus, we used a change sco e o ⩾5 o he GAD-7 and ⩾6 o he PHQ-9 and he PHQ-15. Among he indi iduals who me he eco e y c i e ia, indi iduals who sco ed below he caseness h eshold on all h ee measu es a e ea men and showed eliable imp o emen on ⩾one o he h ee measu es we e conside ed o ha e achie ed a eliable eco e y. By con as , de e io a ion was de ined as an inc ease in he sco e on any o he h ee scales based on he c i e ia o he scale in ques ion. All s a is ical analyses we e wo- ailed; gi en he mul iplici y o compa isons o p ima y and seconda y ou comes, he inal alpha le el was se a 0.01. E ec sizes o hese analyses we e calcula ed using Cohen’sd.Inaddi ion,a pe -p o ocol (PP) analysis o p ima y and seconda y ou comes was pe o med only in he pa ien s who comple ed all ollow-up measu emen s. Resul s A o al o 1691 pa ien s assessed be ween 14 Janua y 2014 and 30 July 2018, we e conside ed po en ial candida es o pa icipa ion in his ial and we e en olled. O hose, 630 (37.3%) ailed o mee all inclusion c i e ia, lea ing a o al o 1061 pa ien s (62.7%) who me he ial inclusion c i e ia. These pa ien s we e hen andomized o he ea men (n= 527) o con ol (n= 534) g oups, and his sample (n= 1061) was used o he ITT analyses. O hese 1061 pa ien s, 316 in he TAU g oup and 314 in he TD-GCBT + TAU g oup comple ed all pos - ea men assess- men s (d op-ou a e: 40.8 and 40.4%, espec i ely). The inal 1-yea ollow-up assessmen s we e comple ed by 30 July 2019. The s udy low diag am is shown in Fig. 1. Pa icipan s in he expe imen al g oup a ended a mean o 4.5 (S.D. = 2.6) o he se en sessions; 67.7% o he pa icipan s a ended 4 o mo e sessions and we e hus conside ed complian . The numbe o TD-GCBT sessions a ended was signi ican ly co - ela ed wi h pos - ea men ou come measu es: PHQ-9 ( =−0.14; p= 0.022), GAD-7 ( =−0.16; p= 0.008), and PHQ-15 ( =−0.16; p= 0.007), indica ing ha ea men exposu e was associa ed wi h be e ou comes. The sociodemog aphic cha ac e is ics o he sample a e shown in Table 1. The mos common pa ien p o ile was a ma ied woman in he ea ly 40s (mean age, 43.6; S.D., 12.3), employed pa - ime, ea ning less han €24 000 annually and p esen ing symp oms o a leas h ee common diso de s (dep ession, anx- ie y, and soma ic). The e we e no signi ican be ween-g oup di - e ences in TAU in e ms o medica ions o dosage a he 12-mon h ollow-up and isi s o he GPs (da a no shown, a ail- able upon eques ). P ima y ou comes Anxie y symp oms (GAD-7) On he ITT analyses, he e we e signi ican be ween-g oup di e - ences in anxie y symp oms a all pos - ea men ime poin s ( p< 0.001), wi h be e esul s in he TD-GCBT + TAU g oup, wi h small o medium e ec sizes (Mo is’sd=−0.38 o −0.65). On he PP analysis, signi ican di e ences we e also obse ed on all measu es, bu wi h medium o la ge e ec sizes (Mo is’sd= −0.62 o −1.01). Table 2 and Fig. 2 p o ide de ailed esul s on hese measu es. Dep ession symp oms (PHQ-9) The ITT analyses e ealed signi ican be ween-g oup di e ences (p< 0.001) a e ea men inaliza ion and a all ollow-up assess- men s, wi h be e ou comes in he TD-GCBT + TAU g oup, wi h small o medium e ec sizes (Mo is’sd=−0.36 o −0.58). The PP analyses also showed signi ican di e ences be ween he g oups on all measu es, wi h medium o la ge e ec sizes (Mo is’sd=−0.60 o −0.92). See Table 2 and Fig. 2 o mo e de ails. Soma iza ion symp oms (PHQ-15) Signi ican be ween-g oup di e ences we e de ec ed a all pos - ea men ime poin s, indica ing a g ea e educ ion in soma ic symp oms in he TD-GCBT + TAU g oup ( p< 0.001). E ec sizes anged om small o medium (Mo is’sd=−0.31 o −0.40). The PP analyses showed signi ican di e ences be ween he g oups a all ime poin s, wi h medium e ec sizes (Mo is’s d=−0.49 o −0.65). Table 2 and Fig. 2 p o ide mo e de ails. Reco e y, eliable eco e y, and de e io a ion a es In he ITT analyses (da a shown as alues and 95% CI), eco e y a es o he TAU g oup a he pos - ea men and 12-mon h 4 An onio Cano‐Vindel e al. h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a assessmen s we e 18% (14–22%) and 29% (22–36%), espec - i ely. Fo he TD-GCBT + TAU g oup, he eco e y a es a hose same ime poin s we e 51.7% (46–57%) and 52% (45– 60%), yielding a be ween-g oup e ec size o 0.76 (0.60–0.92) and 0.51 (0.36–0.67), espec i ely. The p opo ion o indi i- duals conside ed o ha e achie ed a eliable eco e y a he pos - ea men and 12-mon h assessmen s in he TAU g oup was 13.3% (9–17%) and 11% (5–16%), espec i ely; in he TD-GCBT + TAU g oup, he eliable eco e y a es we e 49.5% (44–55%) and 45% (37–48%), yielding a be ween-g oup e ec size o 0.84 (0.68–1.05) and 0.83 (0.67–0.99), espec - i ely. De e io a ion a es a he pos - ea men and 12-mon h assessmen s we e 14% (10–17%) and 12% (7−14%) o he TAU g oup, and 3% (1–5%) and 3% (1–5%) o he TD-GCBT + TAU, yielding a be ween-g oup e ec size o 0.41 (0.26–0.57) and 0.35 (0.19–0.50), espec i ely. Simila esul s we e ob ained on he PP analysis (da a a ailable on eques ). Fig. 1. Flow o pa icipan s h ough he ial. Psychological Medicine 5 h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a Seconda y ou comes We obse ed (ITT analysis) a signi ican ly g ea e dec ease in he ea men g oup . con ols on all he h ee disabili y dimensions (wo k, social, and amily li e), wi h small e ec sizes (Mo is’d= −0.16 o −0.39). These di e ences we e no signi ican a he 3-mon h ollow-up bu we e signi ican a subsequen assessmen s (mon hs 6 and 12). A simila esul was ound on he PP analysis, bu wi h g ea e e ec sizes, anging om small o medium (Mo is’d=−0.26 o −0.51) (Table 3). Small o medium e ec sizes (Mo is’d= 0.17–0.42) we e ound on he ou quali y o li e dimensions assessed, wi h some a ia ions a mon hs 3 and Table 1. Demog aphics cha ac e is ics o sample Cha ac e is ics To al (n= 1061) In en ion o ea sample Pe p o ocol sample TAU (n= 534) TD-GCBT (n= 527) TAU (n= 316) TD-GCBT (n= 315) Gende Female 861 (81.1) 437 (81.8) 424 (80.5) 261 (82.6) 251 (79.7) Male 200 (18.9) 97 (18.2) 103 (19.5) 55 (17.4) 64 (20.3) Age g oup, yea s ⩽19 16 (1.5) 6 (1.1) 10 (1.9) 4 (1.3) 2 (0.6) 20–39 386 (36.4) 197 (36.9) 189 (35.9) 92 (29.1) 102 (32.4) 40–59 581 (54.8) 286 (53.6) 295 (56.0) 186 (58.9) 186 (59.0) ⩾60 78 (7.4) 45 (8.4) 33 (6.3) 34 (10.8) 25 (7.9) Ma i al s a us Ma ied 513 (48.4) 248 (46.4) 265 (50.3) 159 (50.3) 177 (56.2) Di o ced 87 (8.2) 34 (6.5) 53 (10.1) 21 (6.6) 28 (8.9) Widowed 29 (2.7) 14 (2.6) 15 (2.8) 10 (3.2) 6 (1.9) Sepa a ed 58 (5.5) 37 (6.9) 21 (4.0) 18 (5.7) 10 (3.2) Ne e ma ied 212 (20.0) 102 (19.1) 110 (20.0) 56 (17.7) 51 (16.2) Unma ied 162 (15.3) 99 (18.5) 63 (12.0) 52 (16.5) 43 (13.7) Le el o educa ion No schooling 11 (1.0) 7 (1.3) 4 (0.8) 3 (0.9) 3 (1.0) Basic educa ion 267 (25.2) 140 (26.2) 127 (24.1) 87 (27.5) 65 (20.6) Seconda y educa ion 233 (22.0) 122 (22.8) 111 (21.1) 68 (21.5) 59 (18.7) High School 262 (24.7) 123 (23.0) 139 (26.4) 79 (25.0) 94 (29.8) Bachelo 242 (22.8) 119 (22.3) 123 (23.3) 70 (22.2) 84 (26.7) Mas e /doc o a e 46 (4.3) 23 (4.3) 23 (4.4) 9 (2.8) 10 (3.2) Employmen si ua ion Employed ull- ime 248 (14.7) 87 (16.3) 93 (17.6) 44 (13.9) 48 (15.2) Employed pa - ime 633 (37.4) 209 (39.1) 183 (34.7) 122 (38.6) 115 (36.5) Unemployed, in sea ch o wo k 366 (21.6) 123 (23.0) 107 (20.3) 64 (20.3) 59 (18.7) Unemployed, no looking o wo k 202 (11.9) 60 (11.2) 77 (14.6) 40 (12.7) 49 (15.6) Tempo a y incapaci y o wo k 129 (7.6) 32 (6.0) 41 (7.8) 25 (7.9) 26 (8.3) Pe manen incapaci y o wo k 37 (2.2) 10 (1.9) 13 (2.5) 9 (2.8) 9 (2.9) Re i ed 76 (4.5) 13 (2.4) 13 (2.5) 12 (3.8) 9 (2.9) Le el o income (pe yea ) Less han €12 000 eu os 670 (39.6) 214 (40.0) 195 (37.0) 118 (37.3) 98 (30.5) Be ween €120 000 and €24 000 690 (40.8) 215 (40.2) 218 (41.4) 129 (40.8) 139 (44.6) Be ween €240 000 and €36 000 218 (12.9) 74 (13.9) 73 (13.9) 48 (15.2) 48 (15.7) Mo e han €36 000 113 (6.7) 31 (5.8) 41 (7.8) 21 (6.6) 30 (9.2) TAU, ea men -as-usual; TD-GCBT, ansdiagnos ic g oup cogni i e-beha iou al he apy. Resul s a e p esen ed as numbe and pe cen ages. 6 An onio Cano‐Vindel e al. h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a Table 2. Summa y o be ween-g oup di e ences in p ima y ial ou come Ou come measu e In en ion o ea sample Pe p o ocol sample TAU TD-GCBT Di e ence TAU TD-GCBT Di e ence No. M (S.D.) No. M (S.D.) Mo is’dp No. M (S.D.) No. M (S.D.) Mo is’dp GAD-7 Baseline 534 –527 –– – – 534 12.1 (4.7) 527 12.5 (4.6) –0.264 Pos - ea men 534 9.5 (5.4) 527 6.8 (4.7) −0.65 <0.001 316 10.2 (5.5) 315 6.0 (4.3) −1.01 <0.001 3 mon hs 534 8.7 (5.3) 527 7.3 (5.0) −0.38 <0.001 238 8.9 (5.4) 273 6.7 (4.9) −0.62 <0.001 6 mon hs 534 8.6 (5.4) 527 6.9 (5.1) −0.45 <0.001 204 8.8 (5.7) 229 6.2 (4.9) −0.78 <0.001 12 mon hs 534 8.3 (5.7) 527 6.6 (5.4) −0.44 <0.001 180 8.7 (5.8) 208 5.8 (5.3) −0.91 <0.001 PHQ-9 Baseline 534 –– 527 –– – – 534 13.5 (5.4) 527 13.7 (5.3) –0.443 Pos - ea men 534 10.8 (6.4) 527 8.0 (5.7) −0.58 <0.001 316 11.5 (6.6) 315 7.0 (5.2) −0.92 <0.001 3 mon hs 534 10.2 (6.4) 527 8.4 (6.0) −0.39 <0.001 238 10.3 (6.5) 273 7.8 (6.0) −0.60 <0.001 6 mon hs 534 9.8 (6.4) 527 7.9 (6.1) −0.40 <0.001 205 10.0 (6.6) 228 7.3 (6.1) −0.75 <0.001 12 mon hs 534 9.4 (6.3) 527 7.8 (5.9) −0.36 <0.001 180 9.7 (6.5) 208 7.1 (6.2) −0.61 <0.001 PHQ-15 Baseline 534 –– 527 –– – – 534 14.0 (4.8) 527 14.3 (4.9) –0.388 Pos - ea men 534 11.7 (5.2) 527 9.9 (5.4) −0.40 <0.001 316 12.1 (5.2) 315 9.1 (5.3) −0.65 <0.001 3 mon hs 534 11.4 (5.1) 527 10.1 (5.3) −0.32 <0.001 238 11.7 (5.0) 273 9.5 (5.4) −0.49 <0.001 6 mon hs 534 11.1 (5.3) 527 9.8 (5.6) −0.31 <0.001 205 11.5 (5.3) 228 9.2 (5.7) −0.59 <0.001 12 mon hs 534 10.7 (5.6) 527 9.4 (5.6) −0.32 <0.001 180 11.7 (5.6) 208 8.8 (5.7) −0.57 <0.001 GAD-7, gene alized anxie y diso de -7; M, mean; PHQ-9, Pa ien Heal h Ques ionnai e-9; PHQ-15, Pa ien Heal h Ques ionnai e-15, TAU, ea men -as-usual; TD-GCBT, ansdiagnos ic g oup cogni i e-beha iou al he apy; S.D., s anda d de ia ion. Psychological Medicine 7 h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a 6, and wi h medium e ec sizes (Mo is’d=0.55–0.72) a he inal assessmen (Table 3). The PP esul s we e simila , bu wi h la ge e ec sizes on almos all measu es (Table 3). O e all, pa ien s in he ea men a m we e mo e sa is ied han hose in he TAU a m [9.75 (1.49) . 7.72 (2.68)], which ep esen s a la ge e ec size (Mo is’d> 0.90); howe e , his e ec size dec eased a subsequen ollow-ups om a la ge o a medium-sized e ec (Mo is’d= 0.65 o 0.51). Discussion The main inding o his RCT is ha adding TD-GCBT o TAU esul ed in a g ea e educ ion in he symp oms o emo ional diso de s a all pos - ea men ime poin s (immedia ely a e ea men inaliza ion, and a mon hs 3, 6, and 12). The e ec sizes o he p ima y ou comes we e la ge o anxie y symp oms and medium o dep essi e and soma ic symp oms. Fu he mo e, hese he apeu ic e ec s we e sus ained h oughou he 12-mon h ollow-up pe iod, wi h small o la ge e ec sizes. Impo an ly, he eliable eco e y a es showed la ge e ec sizes in a ou o he expe imen al g oup a he immedia e pos - ea men assessmen and a he 12-mon h- ollow-up. Pa ien s in he ea men g oup expe ienced a g ea e dec ease in measu es o disabili y signi ican ly, a g ea e imp o emen in he quali y o li e and –as a consequence –highe ea men sa is ac ion imme- dia ely (sco es >9 on a 10-poin scale) a e ea men comple ion and a he 12-mon h ollow-up, hus suppo ing ou seconda y hypo hesis ha TD-GCBT + TAU would yield supe io esul s o TAU alone on hese measu es. O he pa ien s in he TD-GCBT a m, app oxima ely 70% a ended ou o mo e o he se en ses- sions, a inding ha indica es he accep abili y o he expe imen al in e en ion. These esul s a e consis en wi h indings om p e ious s udies, con i ming he g ea e e icacy and e ec i eness o adding CBT o TAU in he ea men o emo ional diso de s (Ca pen e e al., 2018; Cuijpe s e al., 2019a). Some s udies ha e ound ha TD-CBT is highly e ec i e in educing he symp oms o anxie y and dep ession (Newby e al., 2015), as e idenced by he s udies ca ied ou by No on and Ba e a (No on & Ba e a, 2012) and Chambe lain and No on (Chambe lain & No on, 2013), bo h o which epo ed good esul s o TD-GCBT (mainly o anxie y diso de s). In he p ima y ca e se ing, howe e , ela i ely ew s udies ha e been conduc ed o e alua e indi idual o g oup TD-CBT, wi h he no able excep ion o a ecen p e-pos obse a- ional s udy (K is jánsdó i e al., 2018) ha assessed a 6-week TD-GCBT in e en ion o adul pa ien s wi h dep ession and/ o anxie y diso de . Reliable eco e y a es in ou ial we e simi- la o hose epo ed in o he simila p ojec s, such as he IAPT p ojec in he UK, which achie ed eliable eco e y a es close o 50% on measu es o anxie y and dep ession (Cla k, 2018; Wake ield e al., 2020) and sligh ly lowe han hose epo ed in he No wegian e sion o he IAPT (58.5%) (Knaps ad e al., 2020). Howe e , i is wo h no ing ha we applied s ic e , mo e conse a i e c i e ia, as we included h ee main ou comes (anxie y, dep ession, and soma iza ions) . only wo (anxie y and dep ession) in he IAPT p og ams. In ac , he be ween-g oup e ec size in ou s udy (>0.80) was la ge han ha obse ed in he No wegian s udy (0.61). I is also impo an o emphasize ha eliable de e io a ion a es in he expe imen al a m in ou ial we e no ably lowe han in he con ol g oup (3% . 12–14%) as well as lowe han he 5–10% de e io a ion a es com- monly ound in adul pa ien s pa icipa ing in clinical ials o psycho he apeu ic he apies (Lambe & Ogles, 2004). In sho , he indings o his la ge RCT suppo he e icacy o adding TD-GCBT o TAU in he p ima y ca e se ing o ea pa ien s wi h di e en emo ional diso de s. Ou esul s show ha his combined app oach imp o es symp oma ology and Fig. 2. P ima y Ou comes a Baseline and Follow-up. TAU, ea men -as-usual; TD-GCBT, ansdiagnos ic g oup cogni i e-beha iou al he apy. E o ba s ep esen s anda d e o s. 8 An onio Cano‐Vindel e al. h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a Table 3. Summa y o be ween-g oup di e ences o seconda y ial ou come Ou come measu e In en ion o ea sample Pe p o ocol sample TAU TD-GCBT Di e ence TAU TD-GCBT Di e ence No. M (S.D.) No. M (S.D.) Mo is’dp No. M (S.D.) No. M (S.D.) Mo is’dp Wo king li e a Baseline 534 3.5 (3.1) 527 3.6 (3.2) –0.600 534 3.5 (3.1) 527 3.6 (3.2) –0.604 Pos - ea men 534 3.0 (3.1) 527 2.6 (3.0) −0.16 0.002 316 3.1 (3.1) 315 2.4 (2.9) −0.26 0.002 3 mon hs 534 2.7 (3.0) 527 2.4 (3.0) –0.753 238 2.6 (3.0) 273 2.5 (2.9) –0.814 6 mon hs 534 2.7 (83.0) 527 2.1 (2.9) −0.22 0.001 204 2.8 (3.0) 229 1.9 (2.7) −0.32 0.001 12 mon hs 534 3.1 (3.3) 527 2.4 (3.2) −0.25 <0.001 180 3.3 (3.3) 208 2.0 (2.7) −0.44 <0.001 Social li e a Baseline 534 4.6 (3.0) 527 4.7 (3.0) –0.965 534 4.6 (3.0) 527 4.7 (3.0) –0.946 Pos - ea men 534 4.1 (3.1) 527 3.2 (3.0) −0.33 <0.001 316 4.1 (3.1) 315 2.9 (2.8) −0.41 <0.001 3 mon hs 534 3.5 (3.1) 527 3.2 (2.9) –0.281 238 3.4 (3.2) 273 3.1 (2.9) –0.291 6 mon hs 534 3.4 (3.2) 527 2.7 (3.1) −0.27 <0.001 205 3.6 (3.2) 228 2.6 (2.8) −0.36 <0.001 12 mon hs 534 3.8 (3.4) 527 2.9 (3.4) −0.33 <0.001 180 4.0 (3.3) 208 2.6 (3.1) −0.48 <0.001 Family li e a Baseline 534 4.6 (3.1) 527 4.8 (3.0) –0.437 534 4.6 (3.1) 527 4.8 (3.0) –0.430 Pos - ea men 534 3.9 (3.1) 527 3.1 (2.9) −0.33 <0.001 316 4.0 (3.1) 315 2.8 (3.1) −0.43 <0.001 3 mon hs 534 3.5 (3.1) 527 3.1 (3.1) –0.061 238 3.5 (3.1) 273 3.0 (3.0) –0.067 6 mon hs 534 3.6 (3.2) 527 2.7 (3.1) −0.36 <0.001 205 3.6 (3.1) 228 2.6 (2.7) −0.41 <0.001 12 mon hs 534 3.8 (3.3) 527 2.8 (3.2) −0.39 <0.001 180 3.9 (3.3) 208 2.5 (2.8) −0.51 <0.001 Physical b Baseline 534 22.4 (4.3) 527 22.1 (4.3) –0.327 534 22.4 (4.3) 527 22.1 (4.3) –0.327 Pos - ea men 534 23.2 (4.5) 527 24.7 (4.6) 0.42 <0.001 316 22.7 (4.6) 315 25.1 (4.7) 0.61 <0.001 3 mon hs 534 23.5 (4.6) 527 24.2 (4.8) 0.23 0.017 238 23.2 (4.8) 273 24.4 (4.9) 0.34 0.004 6 mon hs 534 23.6 (4.5) 527 24.3 (4.4) 0.23 0.008 204 23.1 (4.8) 229 24.7 (4.9) 0.44 0.001 12 mon hs 534 24.2 (5.1) 527 26.4 (5.3) 0.58 <0.001 180 22.7 (5.1) 208 25.6 (5.3) 0.73 <0.001 Psychological b Baseline 534 16.9 (3.8) 527 16.9 (3.8) –0.578 534 16.9 (3.8) 527 16.9 (3.8) –0.578 Pos - ea men 534 17.7 (3.9) 527 19.2 (4.0) 0.39 <0.001 316 17.4 (4.2) 315 19.6 (4.0) 0.61 <0.001 3 mon hs 534 18.1 (3.9) 527 18.9 (4.2) 0.21 0.001 238 18.0 (4.0) 273 19.3 (4.2) 0.38 <0.001 6 mon hs 534 18.5 (3.8) 527 19.1 (3.9) 0.16 0.010 205 18.3 (4.2) 228 19.3 (4.2) 0.31 0.008 (Con inued) Psychological Medicine 9 h ps://www.camb idge.o g/co e/ e ms. h ps://doi.o g/10.1017/S0033291720005498 Downloaded om h ps://www.camb idge.o g/co e. Uni e sidad de Za agoza, on 04 Ma 2021 a 11:02:34, subjec o he Camb idge Co e e ms o use, a ailable a