IMPORTANCE Cognitivebehavioraltherapyforinsomnia(CBT-I)isthemostprominent nonpharmacologictreatmentforinsomniadisorders.Althoughmeta-analyseshaveexamined primaryinsomnia,lessisknownaboutthecomparativeefficacyofCBT-Ioncomorbidinsomnia. OBJECTIVE ToexaminetheefficacyofCBT-Iforinsomniacomorbidwithpsychiatricand/or medicalconditionsfor(1)remissionfrominsomnia;(2)self-reportedsleepefficiency,sleep onsetlatency,wakeaftersleeponset,totalsleeptime,andsubjectivesleepquality;and (3)comorbidsymptoms. DATA SOURCES AsystematicsearchwasconductedonJune2,2014,throughPubMed, PsycINFO,theCochraneLibrary,andmanualsearches.Searchtermsincluded(1)CBT-IorCBT orcognitivebehavioral[anditsvariations]orbehavioraltherapy[anditsvariations]or behavioralsleepmedicineorstimuluscontrolorsleeprestrictionorrelaxationtherapyor relaxationtrainingorprogressivemusclerelaxationorparadoxicalintention;and(2)insomnia orsleepdisturbance. STUDY SELECTION Studieswereincludediftheywererandomizedclinicaltrialswithatleast oneCBT-Iarmandhadanadultpopulationmeetingdiagnosticcriteriaforinsomniaaswellas aconcomitantcondition.Inclusioninfinalanalyses(37studies)wasbasedonconsensus between3authors’independentscreenings. DATA EXTRACTION AND SYNTHESIS Datawereindependentlyextractedby2authorsand pooledusingarandom-effectsmodel.Studyqualitywasindependentlyevaluatedby 2authorsusingtheCochraneriskofbiasassessmenttool. MAIN OUTCOMES AND MEASURES Apriorimainoutcomes(ie,clinicalsleepandcomorbid outcomes)werederivedfromsleepdiaryandotherself-reportmeasures. RESULTS Atposttreatmentevaluation,36.0%ofpatientswhoreceivedCBT-Iwereinremission frominsomniacomparedwith16.9%ofthoseincontrolorcomparisonconditions(pooledodds ratio,3.28;95%CI,2.30-4.68;P < .001).Pretreatmentandposttreatmentcontrolledeffect sizesweremediumtolargeformostsleepparameters(sleepefficiency:Hedgesg = 0.91[95% CI,0.74to1.08];sleeponsetlatency:Hedgesg = 0.80[95%CI,0.60to1.00];wakeaftersleep onset:Hedgesg = 0.68;sleepquality:Hedgesg = 0.84;allP < .001),excepttotalsleeptime. Comorbidoutcomesyieldedasmalleffectsize(Hedgesg = 0.39[95%CI,0.60-0.98]; P < .001);improvementsweregreaterinpsychiatricthaninmedicalpopulations(Hedges g = 0.20[95%CI,0.09-0.30];χ 2 testforinteraction = 12.30;P < .001). CONCLUSIONSANDRELEVANCE Cognitivebehavioraltherapyforinsomniaisefficaciousfor improvinginsomniasymptomsandsleepparametersforpatientswithcomorbidinsomnia.A smalltomediumpositiveeffectwasfoundacrosscomorbidoutcomes,withlargereffectson psychiatricconditionscomparedwithmedicalconditions.Large-scalestudieswithmore rigorousdesignstoreducedetectionandperformancebiasareneededtoimprovethequality oftheevidence.