at SciVerse ScienceDirect

Sleep Medicine Reviews 17 (2013) 453e464

Contents lists available

Sleep Medicine Reviews

journal homepage: www.elsevier .com/locate /smrv

CLINICAL REVIEW

Adherence to cognitive behavioral therapy for insomnia: A systematicreview

Ellyn E. Matthews a,*, J. Todd Arnedt b, Michaela S. McCarthy a, Leisha J. Cuddihy c,Mark S. Aloia d

aUniversity of Colorado Denver, College of Nursing, Aurora, CO 80045, USAbUniversity of Michigan, Departments of Psychiatry and Neurology, Ann Arbor, USAcDepartment of Behavioral Medicine, Pine Rest Christian Mental Health Services, Grand Rapids, USAdNational Jewish Health, Department of Medicine, Denver, CO, USA

a r t i c l e i n f o

Article history:Received 1 August 2012Received in revised form2 January 2013Accepted 2 January 2013Available online 16 April 2013

Keywords:AdherenceCognitive behavioral therapyInsomniaBehavioral intervention

* Corresponding author. Tel.: þ1 303 724 8552; faxE-mail address: Ellyn.matthews@ucdenver.edu (E.

1087-0792/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.smrv.2013.01.001

s u m m a r y

Chronic insomnia is a significant public health problem worldwide, and insomnia has considerablepersonal and social costs associated with serious health conditions, greater healthcare utilization, workabsenteeism, and motor-vehicle accidents. Cognitive behavioral therapy for insomnia (CBTI) is an effi-cacious treatment, yet attrition and suboptimal adherence may diminish its impact. Despite theincreasing use of CBTI, surprisingly little attention has been devoted to understanding the role ofadherence. This review describes a comprehensive literature search of adherence to CBTI. The searchrevealed 15 studies that evaluated adherence to CBTI in adults using valid and reliable measures of sleep,and measure of adherence other than study withdrawals. The primary purposes of this review were to 1)synthesize current study characteristics, methodology, adherence rates, contributing factors, and impacton outcomes, 2) discuss measurement issues, and 3) identify future practice and research directions thatmay lead to improved outcomes. Strong patterns and inconsistencies were identified among the studies,which complicate an evaluation of the role of adherence as a factor and outcome of CBTI success. Theimportance of standardized adherence and outcome measures is discussed. In light of the importance ofadherence to behavior change, this systematic review may better inform future intervention efforts.

� 2013 Elsevier Ltd. All rights reserved.

Introduction

Chronic insomnia is a public health crisis affecting approximately10e15% of adults worldwide.1e3 In addition to the social and finan-cial toll,4 insomnia is associated with serious health conditions,5

greater healthcare utilization,6 work absenteeism,7 and motor-vehicle accidents.8 With accumulating evidence of the effective-ness of cognitive behavioral therapy for insomnia (CBTI),3,9e11

adherence to this treatment is a fundamental public health, clini-cal, and scientific concern.

Little is known about adherence rates and factors related to pa-tients’ adherence to CBTI. The lack of attention to adherence to CBTIstands in stark contrast to adherence to treatments for other sleepdisorders12 and other medical conditions, which have been the focusof theoretical and empirical research for decades.13 The term com-pliance, used extensively in the past, has drawn criticism for itsemphasis on medical authority and implication that patients are

: þ1 303 724 8560.E. Matthews).

All rights reserved.

passive recipients of care. In response, the term “adherence” wasintroduced to call attention to the importance of patient agreementwithmedical recommendations.14 Adherence is generally referred toas the extent to which a person’s behavior coincides with medical orhealth advice.15,16 For the purpose of this paper, adherence is definedas persistence in the practice and maintenance of desired healthbehaviors, and is the result of patient’s active participation in andagreement with treatment recommendations.17e19

Insomnia, characterized by difficulty initiating, maintaining, orobtaining good quality sleep, often occurs in the presence ofa medical or psychiatric condition (comorbid insomnia) or maystand alone as a single condition (primary insomnia).1 Growingevidence supports the effectiveness of CBTI for the management ofboth primary and comorbid insomnia without the associatedhabituation, and cognitive and psychomotor impairments ofpharmacotherapy.11,20 CBTI, a multi-component intervention, mostoften features sleep restriction (SR), stimulus control (SC), sleephygiene education (SHE), cognitive therapy (CT), and can includerelaxation techniques. The specifics of these interventions are dis-cussed in detail elsewhere.9,21,22 Based on substantial evidenceprimarily from randomized controlled trials (RCT), the American

AbbreviationsBT behavioral therapyCBT cognitive behavioral therapyCBTI cognitive behavioral therapy for insomniaCT cognitive therapyDSM-IV Diagnostic and statistical manual of mental disorders,

4th editionBT behavioral therapyISI insomnia severity indexISPP individualized sleep promotion planPAP positive airway pressurePSG polysomnographyPSQI Pittsburgh sleep quality indexRCT randomized control trialRM repeated measuresRT relaxation therapySC stimulus controlSHE sleep hygiene education

SIGN Scottish Intercollegiate Guidelines NetworkSR sleep restrictionSRAQ spousal-rated adherence questionnaireTPB theory of planned behaviorTRAQ therapist-rated adherence questionnaireTST total sleep timeTTM transtheoretical model of behavior changeVAMC veteran’s administration medical centerWASO wake after sleep onset

Glossary of termsActigraphy: a technique using a small device (i.e., actigraph)

worn on the wrist or ankle to measure bodymovement and identify activity and rest patternsthat is useful in assessing sleepewake cycles acrossmany consecutive days and nights

Self-efficacy: one’s own judgment of competence to completetasks and reach goals

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464454

Academy of Sleep Medicine has established that combined SR, SC,relaxation training, and CT are efficacious therapies for chronicinsomnia.10 Furthermore, CBTI, with and without relaxation therapy,is superior to pharmacological therapy in maintaining treatmentgains beyond the completion of treatment.20,23,24 RCTs support theefficacy of CBTI under ideal controlled conditions, however, theeffectiveness of CBTI in real world situations is less certain.

Adherence to medical treatment is a challenging problem formany clinicians, including behavioral therapists.25 Overall, adher-ence among patients with chronic conditions is disappointinglylow compared to patients with acute medical problems, partic-ularly after several months of therapy. It is not surprising thatadherence is suboptimal to treatment of chronic sleep disorders. Ina meta-analysis of 569 studies examining adherence to commonmedical treatments, the average adherence rate was 75% and themean adherence to sleep-related treatments (65.5%) was the low-est of the disorders studied.25 Most studies reflect poor adherenceto positive airway pressure (PAP) treatment for sleep apnea, butalso include studies of adherence to behavioral treatment forinsomnia. Poor adherence to sleep-related treatments may be dueto barriers specific to PAP or to a low relative importance assignedto managing sleep problems in general.

In contrast to the PAP literature, CBTI studies rarely includeadherence information. If included, it is often limited to averagesessions attended or overall study attrition. Available estimates ofwithdrawal during CBTI suggest that 14e40% of study participantsdrop out of individual or group treatment before mid-treatment,26

thus potentially diminishing the opportunity for improvement ininsomnia. Dropout rates for internet-based treatment fall withinthe same range as group or individual CBTI. For example, 17% ofadults with chronic insomnia (n ¼ 94) dropped out of weeklyinternet modules.27 In another study, 33% withdrew before the endof five weekly CBTI online modules.28 Research conducted in nat-ural clinical settings may have even higher dropout rates29,30

compared with RCTs, which typically have more rigorous mon-itoring and homogenous participant samples.31,32

Examining the proportion of individuals who discontinue CBTIallows researchers to understand the characteristics of this groupcompared to treatment completers. Much can also be gained,however, by considering adherence beyond attrition rates, both interms of factors that shape adherent behavior, and in refining CBTIto reflect the most effective treatment in a range of vulnerablesubgroups. To date, previous reviews of CBTI have inadequately

addressed the issue of adherence. The goal of this systematic re-view is to analyze the best available scientific evidence related toCBTI adherence and to identify gaps in the literature. Specifically,we will 1) describe current study characteristics, methodology,adherence rates, contributing factors, and impact on outcomes, 2)discuss measurement issues, and 3) identify future practice andresearch directions that may lead to improved outcomes.

Methods

Search methods and study selection

A systematic review of empirical literature having to do withadherence to CBTI was conducted with assistance from a trainedhealth librarian. Databases included PubMed, PsycInfo and MED-LINE. Search terms included “sleep disorders”, “insomnia”, “cogni-tive behavioral therapy”, “sleep restriction”, “stimulus control”,“sleep hygiene”, “sleep education”, “relaxation”, “cognitive ther-apy”, “adherence”, and “compliance” in all applicable combina-tions. The initial search was inclusive of all published articleswritten, regardless of date, to ensure a more comprehensive result.All CBTI studies that evaluated adherence as a primary or secondaryoutcome qualified for inclusion. Publications that met the followingcriteria were included in the final review: a) study sample of adultswith insomnia, b) interventions comprised of any cognitivebehavioral therapy components aimed at treating insomnia, c) validand reliable measures of sleep, and measure of adherence otherthan study withdrawals, d) written in English, e) published in peerreviewed journals. Case studies, opinions and editorials were notconsidered. Studies were evaluated for methodological andreporting quality using the Scottish Intercollegiate GuidelinesNetwork (SIGN) checklists for controlled trials and cohort studies.33

The checklist items were selected because they can be used indifferent study types to assess the studymethodology and potentialbias with respect to study question, participant selection, com-parability of groups, clarity of outcomes, and statistical analysis,among other criteria.33

The titles and abstracts of retrieved studies were independentlyreviewed by two authors (EEM andMSM) against the inclusion andexclusion criteria. The full text article was consulted if there wereany aspects of the abstracts that were unclear and for the finalselection of articles. A third author (MSA) arbitrated if there weredisagreements between the initial reviewers; all authors reviewed

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464 455

the selected studies. Studies that did not explicitly report adher-ence data were excluded.

Evaluation of study characteristics, CBTI, and measures

Studies that met criteria were evaluated for sample character-istics, CBTI features, design, and measurement. CBTI was evaluatedbased on several components of the intervention, includingwhether CBTI was standardized or tailored, the mode of delivery(e.g., individual, group, internet modules), as well as the type ofprovider (e.g., nurse, psychologist). The dose (i.e., number of hoursexposed to the intervention), duration (i.e., length of CBTI deliveryin weeks) and presence of homework were also assessed. Studydesign, type of control group, theoretical underpinnings, and po-tential adherence factors were evaluated. We assessed the type andcomprehensiveness of sleep and adherence measures (i.e., subjec-tive/objective, daily log recording). The impact of adherence onsleep improvement was noted in available studies, however, nometa-analysis of adherence on sleep outcomes was conducted dueto the paucity of intervention studies focused on adherence to CBTIin improving sleep quality.

Results and synthesis

A total of 821 publications from 1981 to 2012 were identifiedfrom the database search. Of these, 227 articles were furtherreviewed; 212 articles were excluded based on the criteria listedabove and duplicate results were noted. The most common reasonfor article exclusion was absence of adherence data. Table 1 sum-marizes the remaining 15 studies published between 1992 and 2012that measured adherence to sleep restriction (SR), stimulus control(SC), and/or sleep hygiene education (SHE) using methods otherthan studywithdrawals. As shown in Fig.1, the number of publishedstudies increased in the last decade. Initial studies focused onadherence in primary insomnia. More recently, an increasing num-ber of studies have included patients with insomnia in associationwith medical and psychiatric disorders. Fourteen out of 15 studieswerepublishedafter2001,whichclearlydemonstrates that researchaddressing adherence to CBTI is still in its infancy.

Sample characteristics

Sample size for the reviewed studies varied widely, with a rangeof 21e301 participants. The mean age ranged from 39.934 to54.2 y35. Only one study focused exclusively on older adults(M ¼ 67.96, SD ¼ 7.07, range 60e81) 36. Twelve studies includedboth genders, composed of 50% women 28 to 73% women.36,37

Studies of insomnia comorbid with breast cancer (n ¼ 4) con-tained female participants only. Overall, study samples representedthe higher reported incidence of insomnia in women.38 The oneexception was a sample recruited from a veteran’s administrationmedical center (VAMC) in which only 14% of the participants werefemale.35 This is likely due to the largelymale population of VAMCs.Interestingly, it has been suggested that psychoeducational inter-vention studies including a high proportion of womenmay producelarger effects39 and preventive care utilization is higher in womenthan in men.40 Thus, gender disparity needs to be explored ingreater detail to understand the role of gender in relation toadherence to CBTI.

Participants were described as having primary insomnia in sixstudies,34,36,37,41e43 insomnia comorbid with breast cancer in fourstudies,44e47 and mixed samples of primary and insomnia comor-bid with psychiatric or medical disorders in five studies27,28,35,48,49

(Fig. 1). Insomnia inclusion criteria for the majority of studies wasbased on either the research diagnostic criteria for an insomnia

disorder27,28,49,50 the International classification of sleep disorderscriteria37,47,51 or criteria for primary insomnia based on the Diag-nostic and statistical manual of mental disorders, 4th edition (DSM-IV).43,52 Participants were self-referred or physician-referred andeligibility related to insomnia was typically determined througha clinical interview and/or the use of established measures such asthe insomnia severity index (ISI)41,45,47e49 or the Pittsburgh sleepquality index (PSQI).34,36,42,45,53 The ISI yields a total score from 0 to28 and scores are interpreted as follows: absence of insomnia (0e7); sub-threshold insomnia (8e14); moderate insomnia (15e21);and severe insomnia (22e28). Only four studies of the reviewedstudies41,45,48,49 reported average baseline ISI assessments. Inwomen with insomnia and breast cancer, scores ranged from 17.5(4.21)45 to 17.9 (4.3).41 Slightly higher scores of 18.08 (0.59)49 and18.78 (5.0)48 were reported in patients with primary insomnia orinsomnia comorbid with psychiatric and medical disorders. Thesebaseline scores suggest many study participants experience mod-erate insomnia prior to therapy. It is difficult, however, to comparethe severity of the insomnia symptoms among the reviewedstudies due to the wide variety of screening instruments used todetermine eligibility.

Evaluation of CBTI characteristics

The majority of studies combined at least three standard CBTIstrategies; one study delivered SHE and SR only.36 CBTI sessionsfollowed a standard protocol with the exception of one study, inwhich the therapy plan was individualized to meet the needs ofwomen during chemotherapy treatment.44 The mode of CBTI de-livery varied. Seven studies administered CBTI individually34e36,42,44e46; and five studies utilized a small group for-mat.28,37,43,47,48 Two studies used internet-based modules.27,49 Dueto small sample size in one study,41 telephone, individual and groupCBTI participants were combined.

CBTI was delivered by nurses in four studies37,44e46; and psy-chologists,28,34,35,43 PhD level clinicians,42 or psychology graduatestudents36 in six studies. Five studies did not include a descriptionof the therapy providers.27,41,47e49

The total maximum CBTI dose and duration of the reviewedstudies were assessed. Maximum doses were highly variableranging from 3 h34 to 13.5 h.47 The median dose was 6.5 h. MostCBTI interventions were delivered in 6e8 sessions, however, someparticipants received two or fewer sessions.34,35 Internet-basedstudies specifically required homework assignments in additionto the use of a daily sleep diary.27,49 Homework assignments mayenhance adherence, especially for longer interventions by con-solidating acquired knowledge, improving skill acquisition, andbridging the gap between learning and real life.39

Study design

Of the studies reviewed, over half were RCTs (8 out of 15;53%).27,35,36,42,44e47 Only one of these RCTs had three arms.42 Six outof the 15 studies (40%) used a one-group preepost or repeatedmeasures design.28,34,41,43,48,49 One study reported cross sectionalanalysis of adherence to CBTI at 12 mo post-treatment.37 Severalstudies were identified as pilots with 21e34 participants.36,42,44,45

Adherence to CBTI was frequently a secondary aim of a largerstudy,27,36,37,41,44e47,49 suggesting that adherence is rarely the pri-mary focus of CBTI studies.

The control groups across the studies varied widely with respectto the type of intervention, including various active controls or waitlist control.27,47 Active controls included a healthy eating educa-tion,44,46 sleep hygiene education alone,35 desensitization,45 and anunspecified behavioral treatment.36

Table 1Summary of studies reviewed.

Study Sample Study design CBTI components Total dose, description Control group Measures (bold ¼ measuresusedfor adherence)

Adherence outcomes

Berger et al.,200344

21 women withinsomnia comorbidwith breast cancer(stage IeII) (U.S.)Age range 43e66 y100% female

RCT pilot;secondary analysisof1-group;RM

SR (modified)SCRTSHE

5e6 ½ h over 4 CTX cycles;30, 60, 90 d follow-up;tailored sleep promotion plan(ISPP) in home

Healthy eatingeducation

Sleep diary, 12-itemquestionnaire to record dailyadherence to ISPPPSQI, wrist actigraphy, Piperfatigue scale (PFS),

[ adherence for most components ofISPP; patterns of adherence stable overtime (SC most variable); [ adherence atthe end of chemotherapy treatments;average nightly adherence for ISPPcomponents at 30, 60, 90 d postintervention ¼ SR (83e88%); SC (36e56%); RT (83e88%); SHE (77e88%)

Bouchardet al.,200341

39 primaryinsomnia patients(Canada)Mean age ¼ 41.4 y59% female

Correlation,descriptive;secondary analysis

SRSCCTSHE

4e12 h over 8 wk; variousmodalities: telephone (4 h);individual (7 h); group (12 h)

None Sleep diaryISI, Beck anxiety inventory, BDI,self-efficacy scale

[ self-efficacy associated with [

adherence behaviors; self-efficacy andadherence [ from the 2nd to 8th and Y

1-wk post CBTI

Chamber andAlexander,199234

103 primaryinsomnia patients;self or physicianreferred (U.S.)Mean age ¼ 39.9 y67% female

1-group;preepost

SRSCCTSHE

2e3 h individual initial sessionby a clinical psychologist; somepatients returned for 1e3monitoring

None SQAW and items related torecall of compliance with CBTISleep history; at 6-mofollowup: current medications,other treatment strategies,change in sleep.

Mean satisfaction with CBTI was 3.58/5and [ satisfaction associated with [

adherence; self-reported adherence didnot predict treatment outcomes

Edinger et al.,200935

41 primaryinsomnia patientsor insomniaassociated withpsychiatricdisorders recruitedfrom VAMC (U.S.)Mean age ¼ 54.2 y14% female

RCT, parallel group SRSCSHE

2e4 h over 4 biweekly sessions;individual CBTI or SHE by clinicalpsychologists

SHE alone Electronic sleep diary, post-CBTI questionnaire assessinghow many days/week 6 coreelements of CBTI wereenacted, and the usefulness ofeach strategyPSG (screening), actigraphy,ISQ, PSQI, DBAS, therapyevaluation questionnaire (TEQ),

CBTI group adherence to 6 coreelements averaged 6.23 days/week,compared with the 5.80 days/week inthe control group (SHE alone); CBTIgroup reported [ average usefulnessscores, compared with controls; post-treatment sleep diary data showed theCBTI group had significantly Y

variability in bedtime and rise time thancontrols

Harvey et al.,200237

90 primaryinsomnia patientsrecruited froma sleep clinic at 1 ypost-CBTI(Scotland)Mean age ¼ 53.9 y73% female

Secondary,longitudinal dataanalysis from largeclinicaleffectiveness trial

SRSCCTSHE

5 h over 6 wk; group sessionsconducted by a primary carenurse

Questionnaire related to theuse of 10 CBTI components at1 y follow-up post-CBTI (yes/no responses)Sleep diary during CBTI

Reported use of SC and SR was the bestpredictor of clinical improvement in SLand WASO. Use of CT contributed toa significant Y in WASO; relaxation wasthe most highly endorsed component(74%) but its use did not predictimprovement in SL or WASO; imagery(19% used) was not associated with a Y

in SL or WASO. Average % of selfreported use of CBTI components:49% altering lifestyle (e.g., caffeine, diet)32% altering bedroom (e.g.,temperature)41% stimulus control/sleep restriction59% not taking naps52% pre-bed routine74% relaxation (abbreviated method)21% cognitive control19% imagery41% blocking thoughts41% cognitive restructuring

E.E.Matthew

set

al./Sleep

Medicine

Reviews17

(2013)453

e464

456

Herbert et al.,201027

94 primaryinsomnia patients(Canada); media orphysician referred;55.3% had a co-morbid medical orpsychiatricconditionage data unavailable61.7% female

RCT; secondaryanalysis of 1-group;RM

SRSCCTSHERTmedicationtapering

Total dose unknown;5 weekly Internet moduleswith homework assignments

Wait control Sleep diary, weekly electronichomework checks (adherencedefined as homeworkpractice > 4 nights per week)Medical history, mini-international neuropsychiatricinterview (screening);computer attitudes scale,multidimensional scale ofperceived social support,attitude toward completion ofthe internet programquestionnaire (adapted), PSOCQ(adapted to sleep)

17% dropout rate; [ attrition predictedby [symptom severity and psychiatriccomorbidity; [ perceived behavioralcontrol, social support, and intention tocomplete the program weresignificantly associated with [

adherence to sleep hygiene homework.Participant adherence tocomponents ¼ avoidance of clock-watching (70%),caffeine taper (60.9%),alcohol taper (95.6%),avoiding heavy meals at bedtime (87%),sleeping in separate bedroom fromnoisy bed partner (87%),exercising (39%),taper liquids before bed (60.9%),temperature control in bedroom (87%),avoidance of napping (82.6%),regular sleep schedule (82.6%),avoiding reading in bed (82.6%),abdominal breathing (60%),progressive muscle relaxation (44.4%),imagery induced relaxation (38.8%),hypnosis (41.2%),sleep restriction (44.4%).

Manber et al.,201148

301 patientsreferred forinsomnia, withcomorbidpsychiatric, sleep,and medicaldisorders (U.S.)age data unavailable58% female

Pre-to posttreatment casereplication

SRSCSHECTRTScheduled worrytimeRelapse prevention

10 ½ h over 7 sessions (1e5 weekly,6e7 biweekly); group sessions

e Sleep diary, treatmentcomponents adherence scale(TCAS)ISI, BDI, TSS

Two behavioral element of CBTI(keeping a fixed rise time andrestricting time in bed) and onecognitive element (changingexpectations about sleep) were Y

among patients with greater depressionseverity.

Matthewset al.,201245

34 women withstage IeIII breastcancer recruited atcancer centers(U.S.)Mean age ¼ 53.56 y100% female

RCT; secondaryanalysis of 1-group;RM

SRSCCTSHE

3e4 h over 6 wk; individualsessions conducted by a CBTItrained advanced practice nurse

Behavioral placebo(desensitization)

Sleep diary (sleep restrictioncoding grid calculation),motivation to change sleepbehaviors (single item atbaseline screening)Demographic/medicalcharacteristics; ISI, HADS, PFS

Overall, adherence to prescribedbedtime remained constant butprescribed rise time and total time inbed Y in weeks 5e6 of CBTI. Adherenceto rise time was poor, but [ forprescribed bedtime, which wasassociated with a Y number ofnocturnal awakenings. [ adherencewas associated with [self-reportedmotivation to change sleep behaviors.And Y fatigue was associated with [

adherence. Past chemotherapy wasrelated to [ adherence to prescribedbedtime, and prescribed total time inbed

McChargueet al.,201046

113 women, stage IeIIIA breast cancerrecruited at cancercenters (U.S.)Mean age ¼ 51.6 y100% female

RCT; secondarydata analysis

SR (modified)SCRTSHE

5e6 ½ h over 4 CTX cycles and 30,60, 90 d follow-up; ISPP inparticipant’s home

HEC 12-item adherencequestionnaire measured dailyadherence to ISPPDemographic/medicalcharacteristics, hospital anxietyand depression scale (HADS),symptom experience scale(SES)

Average self-reported adherence ratesto the ISPP were 51e52% across 4 CTXcycles but adherence varied bycomponent with Y adherence to SR and[ adherence to RT; as sleep improvedand depression [ across chemotherapytreatments, participants progressivelyadhered Y to the ISPP

(continued on next page)

E.E.Matthew

set

al./Sleep

Medicine

Reviews17

(2013)453

e464

457

Table 1 (continued )

Study Sample Study design CBTI components Total dose, description Control group Measures (bold ¼ measuresusedfor adherence)

Adherence outcomes

Perlis et al.,200442

27 primaryinsomnia patients(U.S.); physician orself-referred fromadsMean age ¼ 41.3 y70% female

RCT; 3-arm SRSCCTSHErelapse prevention

4e12 h over 8 wk; individualsessions varied in lengthbetween 30 and 90 min; bya single PhD-level clinician

placebo þ CBTImodafinil þ CBTImodafinil þ control

Sleep diaryPSG, medical history inventory,and interview (screening), PSQI,Beck anxiety inventory; BDI;schedule for affective disordersand schizophrenia-lifetimeversion, computer-basedstructure clinical interview forDSM-IV-TR, symptomschecklist

Modafinil did not positively alter sleepoutcomes, however, themodafinil þ CBTI group tended to [

adherence (80% vs 51%, P ¼ 0.059) forthe first 4-wk period of activetreatment; consistently [ adherence tothe prescribed delay in bedtime.Adherence was most robust in week 1e3 of CBTI.The placebo þ CBTI group showeda linear trend toward [ adherence overtime.

Riedel andLichstein,200136

22 adults >60 y oldwith primaryinsomnia recruitedvia media, seniorgatherings (U.S.)Mean age ¼ 68.0 y73% female

RCT; secondaryanalysis of 1-group;RM

SRSHE

Total dose unknown;6 individualweekly sessions conducted bypsychology graduate students

Unspecifiedbehavioraltreatment

Sleep diarySTAI-T, GDS, mini-mental stateexam (screening), Epworthsleepiness scale

Time spent in bed per night at posttreatment exceeded therapistrecommendations by an average of27.89 min (SD ¼ 31.72); 14 participants(64%) were within 30 min and 8participants (36%) were within 15 minof prescribed time in bed.Mean adherence to SR was 68.99%Self-reported time in bed post-CBTI wassignificantly Y than time in bed atbaseline p < 0.01.Night to night consistency of time spentin bed and time arising variance weresignificantly [ at post treatment;[ consistency of time spent in bed pernight and [ consistent arising timepredicted sleep improvements

Tremblayet al.,200947

57 women withbreast cancer post-treatment (Canada)Mean age ¼ 54.0 y100% female

RCT; secondarydata analysis

SRSCCTSHEfatigue and stressmanagement

13.5 h over 8 wk þ boostersession; groupsessions of 4e6 patients3 nights of PSG precededCBTI.

Wait list control Sleep diary (adherence tobehavioral strategies codingcalculations)PSG, ISI, DBAS, HADS, TEPCQ,TAPQ

Post-CBTI, subjectivesleep improvementswere best predicted by [ initial levels oftreatment expectancies andY dysfunctional beliefs about sleep.At 6-mo follow-up, subjectively assessedsleep improvements were best predictedby adherence to behavioral strategies butnot with PSG-assessed sleepimprovements.

Vincent andHameed,200343

50 primaryinsomnia patients(Canada)Mean age ¼ 51.4 y66% female

1-group; RM SRSCCTSHEmedication taperrelaxation trainingstress managementand problemsolving

10 ½ h over 7 wk;groups sessions of 4e6 patients,by PhD level psychologist withsleep training

e Sleep diary; TRAQ, SRAQ,attendance recordInsomnia interview schedule(IIS) (screening), PSQI, BELIEF,DBAS, BDI, structured interviewfor DSM disorders (SCID)

Therapists’ rated nearly half of theparticipants as "very much" to"extremely" adherent, which was themost consistent relationship to sleepoutcomes (i.e., Ypost-CBTI maladaptivebeliefs, Y sleep-related impairment, and[ sleep quality).No adherence variables predicted post-CBTI SL, TST or SE.Comorbid dysthymia was associatedwith Y adherence and Y improvementin SL and SE. There was a trend towardY adherence associated with anxietydisorders.

E.E.Matthew

set

al./Sleep

Medicine

Reviews17

(2013)453

e464

458

Vincent andLewycky,200949

118 primaryinsomnia patients(28%) or insomniaco-morbid withmedical/psychiatricdisorders (66.9%)(Canada)age data unavailable67.8% female

1-group; RMsecondaryadherence data

SRSCCTSHEmedicationtapering

Total dose unknown;5 weekly internetmodules withhomework assignments

e Weekly online homeworkchecks of adherence totreatment componentsSleep diary, ISI, MFI and GF,DBAS, PSAS, clinical globalimprovement scale (CGI),

Reported adherence to treatmentcomponents (% of participants) was asfollows: avoidance of clock-watching(73.9%), sleep hygiene (76.8%), stimuluscontrol (64.2%), relaxation training(67.6%), sleep restriction (51.6%),hypnotic tapering (22.6%).There was a 33% attrition rate by 5 wkand additional 8.5% dropped out at 1-mo follow-up. Attrition was related toreferral status (i.e., dropouts were morelikely to have been referred fortreatment rather than recruited fromthe community).TST did not predict attrition from onlinetreatment

Vincent et al.,200828

40 adults withprimary orcomorbid insomnia(Canada)Mean age ¼ 46.9 y50% female

Correlational-descriptivedesign

SRSCCTSHE

6 weekly small group(6e8 participants) sessionsby psychologists

Sleep diary (adherence towake-up time consistency),modified medical outcomesstudy general adherence scale(MOS-A)GTQ (assessed likes/dislikes andusefulness of SC/SR),motivation survey (assessedbeliefs about SC/SR, perceivedbarriers, and perceivedbehavioral control_)

Y age was associated with [ wake-uptime variance, which was associatedwith [sleepiness; participants with Y

perceived barriers were more likely toreport [ adherence and discomfort,annoyance and boredom were robustlyassociated with Y adherence. Genderand educational level were notassociated with adherence; a [

education was correlated with [

perceived control over SC/SRTYperceived barriers to SC/SRT wasassociated with [ liking of SC/SRT andperceived usefulness of SCParticipants with [ overall adherencehad improved TST, Y nocturnalawakenings, and Ydaytime impairment.Participants attended an average of 5.5out of 6 sessions (91.7% attendance);compared to non-users, sleep aid usersattended fewer sessions

Note: BDI ¼ Beck depression inventory; BELIEF ¼ beliefs and attitudes about sleep scale; BT ¼ behavioral therapy; CBTI ¼ cognitive behavioral therapy for insomnia; CGI ¼ clinical global improvement scale; CT ¼ cognitivetherapy; CTX ¼ chemotherapy; DBAS ¼ dysfunctional attitudes about sleep scale; GDS ¼ geriatric depression scale; GF ¼ general fatigue; GTQ ¼ group therapy questionnaire; HADS ¼ hospital anxiety and depression scale;HEC ¼ healthy eating education; IIS ¼ insomnia interview schedule; ISI ¼ insomnia severity index; ISPP ¼ individualized sleep promotion plan; ISQ ¼ insomnia symptom questionnaire; MFI ¼ multi-dimensional fatigueinventory; MOS-A ¼ modified medical outcomes study general adherence scale; PSAS ¼ pre-sleep arousal scale; PFS ¼ Piper fatigue scale; PSG ¼ polysomnography; PSOCQ ¼ pain stages of change questionnaire;PSQI ¼ Pittsburgh sleep quality index; RCT ¼ randomized control trial, RM ¼ repeated measures; RT ¼ relaxation therapy; SC ¼ stimulus control; SCID ¼ structured interview for DSM disorders; SE ¼ sleep efficiency;SES ¼ symptom experience scale; SHE ¼ sleep hygiene education; SL ¼ sleep latency; SRAQ ¼ spousal-rated adherence questionnaire; SQAW ¼ sleep questionnaire and assessment of wakefulness; SR ¼ sleep restriction; STAI-T¼ state-trait anxiety inventory; TAPQ¼ therapeutic alliance perception questionnaire; TCAS¼ treatment components adherence scale; TEPCQ¼ treatment expectancies and perceived credibility questionnaire; TEQ¼ therapyevaluation questionnaire; TRAQ ¼ therapist-rated adherence questionnaire; TSS ¼ treatment satisfaction scale; TST ¼ total sleep time; [ higher; Y lower.

E.E.Matthew

set

al./Sleep

Medicine

Reviews17

(2013)453

e464

459

Fig. 1. Studies by year and insomnia type.

Fig. 2. Adherence measures by study population.

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464460

A limited number of studies used explicit theoretical models toexamine adherence to CBTI. Hebert and colleagues27 utilized thetheory of planned behavior (TPB)54 and the transtheoretical modelof behavior change (TTM)55 as a framework to maximize adher-ence; however, their main findings suggest that neither TPB norTTM did a good job of predicting participant attrition. Bouchard andcolleagues41 proposed self-efficacy, a concept derived from socialcognitive theory56,57 is useful in predicting adherence to CBTI,while Vincent et al. indicate the health beliefs model58 may explainnon-adherence to CBTI. Measures of adherence in three otherstudies43,45,47 were based on the cognitive behavioral model ofinsomnia.59

Potential predictors of CBTI adherence

Demographic, medical and psychological characteristics werenot consistently associated with adherence in the studies in thisreview. The majority of studies failed to show any relationshipbetween adherence and demographic factors (e.g., age, gender,education). In one study, gender and education had no significantimpact on adherence, but younger participants were less adherenttowake time recommendations than older participants.28 Similarly,medical factors were not related to adherence in most studies. Ina small pilot study, the use of stimulant medication (e.g., modafinil)in conjunction with CBTI enhanced adherence to prescribed bed-time compared to placebo plus CBTI.42 In women with breastcancer, better adherence was reported toward the end of chemo-therapy treatments (when treatment adverse effects are less se-vere)44 and in women with lower levels of subjective fatigue.45

Other factors such as depressive symptoms, may moderate theeffect of sleep improvement on adherence and study attrition. Afew studies identified potential psychological factors that decreaseadherence including higher anxiety43 and greater depressivesymptoms.46,48 Hebert and colleagues noted the presence of psy-chiatric disorders (e.g., depression, generalized anxiety disorder)predicted attrition from internet-based CBTI.

Attitudes toward treatment and past experiences appear toimpact willingness to follow CBTI recommendations. For example,greater pre-CBTI motivation to change sleep behaviors,45 highertreatment expectancies,47 higher levels of self-efficacy,41 andgreater satisfaction with CBTI34 were found to be related to betteradherence. In one study, those with fewer perceived barriers (dis-comfort, annoyance, boredom) and less pretreatment sleepinesswere more adherent, as measured by global self-report scale andwake time consistency from sleep diary.28 In a study of adults withprimary insomnia, higher perceived behavioral control, socialsupport, and intention were associated with better adherence as

measured by online weekly reports.27 Overall, few studies haveadequately addressed the impact of patient expectations andtreatment barriers on adherence; these relationships merit furtherclinical consideration and investigation.

Proposed barriers to following CBTI recommendations mayinclude ease of assimilation in daily life, adaptability in the homesetting, perceived relevance, and perceived effectiveness.37 Adher-ence to sleep restriction recommendations and faithfulness to pre-scribed bedtime and wake-up times is difficult because it oftenrequires a considerable alteration in lifestyle.36 It has been suggestedthat poor adherence to SR and treatment drop-out may be due to 1)the fundamentally counterintuitive recommendation to limit sleep,2) objections to increased daytime sleepiness early in treatment, 3)complaints of boredom and lack of activities because of delayedbedtime, and4) resistance toprescribed timeoutof bed, especially onthe weekends.36 Resistance to prescribed sleep times may also rep-resent patient ambivalence about changing behavior. Few studieshave investigated the role of these barriers or suitable interventions.

It is plausible that individuals become less adherent and stopattending appointments because they feel better or the symptomshave resolved. However, Ong et al.26 reported that the best pre-dictors of early attrition from group CBTI were short sleep durationcoupled with depressive symptoms. Other investigations havefound mixed results regarding the impact of sleep improvement onadherence. Matthews et al.45 reported that subjective overall ratingof sleep was inversely related to adherence to prescribed bedtimeand total time in bed at the beginning of CBTI, however, significantcorrelations did not continue to the end of treatment. Conversely,McChargue and colleagues46 reported decreasing sleep distur-bances contributed to lower total adherence rates during the courseof chemotherapy. The level of sleep disturbance before treatmentmay be as important an indicator of adherence as how quickly sleepimproves during treatment.37,43,46 Additional study is needed tounderstand how the severity of disease and symptoms affectsadherence.

Measurement issues in adherence to CBTI

Measures of adherence to CBTI included multiple and singleitem patient-reported,27,28,34,35,37,44,46,48 therapist/spouse-repor-ted43 adherence questionnaires, and adherence to prescribedsleep schedules as derived from sleep diaries.27,28,35e37,41e45,47,48

Internet-based CBTI incorporated weekly homework checks ofadherence.27,49,60 Combined self-reported adherence and sleepdiary measures were used in four studies,28,35,43,44 thus providinga more complete assessment of adherence. Although several

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464 461

studies employed actigraphy or polysomnography,35,44e47 analysisof adherence using objective measures was rare. Fig. 2 illustratesthe type and frequency of use of adherence measures by studypopulations.

Participants in CBTI trials are frequently categorized as eitheradherent or non-adherent to a given behavior based on subjectiveresponses.When adherence is measured as a dichotomous variable,the complexity inherent in adherence to a behavioral interventionmay not be captured adequately. Behaviors related to stimuluscontrol, sleep hygiene, and cognitive interventions are difficult tomeasure objectively. Some studies have included self-report itemsabout the behaviors or derived information directly from sleep di-aries using an operationally defined criterion for adherence. Whileinclusion of self-report adherence information appears obvious, itshould be recognized that bias may result if self-report is the onlysource of adherence information. This bias underscores the need tocompare adherence data from multiple sources. Information frombed partners and therapists may provide additional validation.

Actigraphy may be useful as an objective verification of dailyreports of bedtime, rise time and may even improve self-reportaccuracy or adherence. Light monitors, available on some acti-watches, may provide insight into adherence to recommendedbehaviors. Even with actigraphy it is difficult to verify that patientsare, in fact, getting out of bed during prolonged nocturnal awak-enings. Future studies would benefit from comparing the degree towhich various sources of information correspond.

Overall, measurement and reporting of adherence measures inthe literature is inconsistent and insufficient, leading to challengesin interpreting treatment and process outcomes. Studies thatattempt to predict or modify adherence often face the problem thatadherence as a dependent variable is complex and non-normallydistributed.61 It is apparent that both standardized definitionsandmeasures of adherence are needed. To that end, the “consensussleep diary” was developed as the result of collaborations withinsomnia experts and potential users. The adoption of this stan-dardized patient-informed sleep diary will facilitate comparisonsacross studies and advance the field.62 Similarly, understanding theminimal level of adherence needed for positive outcomes and thetime point in treatment (if any) at which adherence is most criticaland discerning adherence to individual components will advancethe field. For example, is it essential to have strict adherence duringthe initial phase of treatment when patients are learning thestrategies and rationale, or toward the closing sessions to sustaintreatment gains? One approach to measurement is to combinevarying rates of adherence to the different components of treat-ment into a composite score. Morgenthaler et al.10 suggests SC isthe most effective stand-alone behavioral intervention for insom-nia, therefore, may need to be weighted heavily in a compositemeasure. As an alternative, more attention might be given toattendance at CBTI sessions. In the absence of other data, the per-centage of treatment sessions attended could be regarded asa limited proxy for adherence. Confounding factors such as trans-portation would have to be considered.

Relationship between adherence and treatment outcome

Adherence to nonpharmacological regimens has been found tobe strongly related to treatment outcome in other psychologicalsettings, particularly when measures of adherence are continuousand when the disease is chronic.25 Evidence of the associationbetween adherence to CBTI and sleep outcomes, however, is diffi-cult to interpret due to the diverse measurement of both adherenceand sleep improvement. Higher levels of adherence were asso-ciated with various sleep improvements in some studies36,37,43 butnot in all.34

One study of adherence assessed by an author-constructed self-report scale showed greater adherence to CBTI components wereassociated with lower post treatment ISI scores in 301 adults withinsomnia.48 In contrast, Chambers et al.34 found that adherenceassessed through recall of adherence to CBTI recommendations didnot predict improvements in daytime sleepiness, sleep latency,total sleep time (TST), awakenings or wake after sleep onset(WASO) at six months after CBTI in 103 adults with chronicinsomnia. This may be due in part to low variance in adherence, aspatients reported high adherence to recommendations. Sleep im-provements and adherence were subjectively assessed by patients,thus subject to bias and recall inaccuracy.

The majority of studies assessing diary-based sleep improve-ments relative to adherence, suggest that adherence to key therapyrecommendations improves sleep outcomes in a variety of ways,however, divergent measures of adherence prevent definitive con-clusions. For example, Harvey and colleagues37 used a self-reportmeasure of adherence to ten different CBTI components, adminis-teredoneyearafter completionof onlineCBTI toadultswithprimaryinsomnia. They found that continuedadherence toSCandSRwas thebest predictor of clinical improvement in sleep latency and night-time wakefulness, and ongoing use of cognitive restructuring con-tributed significantly to reduction in nocturnal wakefulness.37

Similarly, in 40 outpatients with primary or comorbid insomniaparticipating in online CBTI, high levels of adherence to SC and SRusing a global self-report scale and diary-based wake time consis-tencywasassociatedwith improvements in total sleep time,numberof awakenings, ISI and less daytime impairment.28

In a study of group-based CBTI in 50 adults with chronicinsomnia, adherence was measured by an investigator developedtherapist-rated adherence questionnaire (TRAQ), a spousal-ratedadherence questionnaire (SRAQ), and attendance record.43 Partici-pants who were judged as more adherent by therapists reportedsubjective improvement in sleep-related impairment, sleep quality,and maladaptive beliefs/attitudes about sleep. Therapist-ratedadherence, but not consistency of bedtime/wake-up time, attend-ance, or spousal-rated adherence, was positively associated withdiary-based sleep outcomes.

In older adults with good overall adherence to 6-wk CBTI,greater consistency with the recommended sleep schedule pre-dicted less nocturnal wake time and higher sleep efficiency. Aconsistent rise time predicted better subjective sleep quality andfewer awakenings, however, three other sleep diary-based adher-ence calculations did not predict improvements in sleep diarymeasures of sleep latency, WASO, or number of awakenings.36

Using several methods to measure adherence to SR, these in-vestigators concluded that the sleep schedule consistency thatoccurs by way of SR may underlie improved outcomes. Methods ofmeasuring adherence consisted of 1) subtracting the amount oftime in bed prescribed at the final treatment from mean reportedtime in bed at post-treatment, 2) calculating adherence percentageby dividing reported time in bed reduction since baseline by pre-scribed time in bed prediction� 100, 3) change inmean time in bedfrom baseline to post-treatment, 4) time in bed variance calculatedby computing the variance of nightly time in bed (in minutes) foreach participant at baseline and post-treatment, and 5) computingthe variance of time arising (leaving bed) in themorning at baselineand post-treatment (arising variance).36

In breast cancer survivors (n ¼ 58), five dichotomous indices ofadherence to SR were calculated from sleep diary data.47 Indicesincluded 1) adherence to prescribed bedtime hour (i.e., not morethan 15 min before the set time), 2) adherence to prescribed arisingtime (i.e., not more than 15 min after set time), 3) arising from bedwithin 30 min following a nocturnal awakening, 4) avoidance ofdaytimenaps, and 5) adherence to total time spent in bedprescribed

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464462

(i.e., not more than 30 min greater than the set amount of time. Aproportion (%) of the number of days where the participant com-plied with CBTI recommendations was determined for each index.Higher levels of adherence to arising during night awakenings wereassociated with a reduction in clinician-rated ISI scores. Higheradherence to prescribed rise times was correlated with reducedWASO and total sleep time. In addition, actigraphy-based improve-ments in WASO, total sleep time, and sleep efficiency were sig-nificantly predicted by a greater adherence to the recommendationtoavoiddaytimenapping.47Using a similar sleepdiary-based codinggrid in breast cancer survivors,Matthews and colleagues45 reportedthat higher adherence to prescribed bedtime in week 5e6 of CBTIwas associated with fewer nighttime awakenings. Adherence toprescribed bedtimeandprescribed total time in bedwere associatedwith increased total sleep time throughout CBTI.

Conclusions

Insomnia is a serious health concern. CBTI is an efficacious treat-ment, yet attrition and suboptimal adherence potentially diminishesits impact. AlthoughCBTI has grown inutility since the 1960s, there isa dearthof researchdevoted tomeasuringadherence toCBTI.Generallimitations of reviewed studies include small samples, reliance onself-reported data, and a restricted range of adherence scores.Inconsistency in adherence findings among the studies may beexplained in part by diverse sample characteristics (e.g., education,socioeconomic status), varied CBTI components, and lack of a stan-dardized definition and measurement of adherence.

The type of provider, mode of delivery, dose and duration oftreatment are also inconsistent across studies, limiting the ability tocompare findings. Although there have been several studies ofinsomnia inwomenwith breast cancer, very little research has beendone on other medical and psychiatric comorbidities. Expandingstudies of adherence to CBTI to other chronically ill populations iswarranted to fill this knowledge gap.

Despite the utility of behavioral change models, CBTI adherencestudies are largely atheoretical. In fact, only three of the reviewedstudies used an explicit conceptual model or theory. Some mayargue that theory does not necessarily lead to better interventionsand, indeed, this has not yet been the case with PAP adherencestudies. Theories do, however, provide insights into the behavioralconstructs that predict adherence. Advantages of theory guidedresearch include the ability to make systematic predictions basedon the theory, utilize measures derived from the theory, andinterpret findings in support or against the theory. To that end,theory-based studies specifically designed with adherence asa primary outcome are needed. With this information, treatmentapproaches could be designed to target factors and processes thatare most likely to impact treatment adherence.

Clinical and research implications

There is a paucity of evidence identifying the impact of adher-ence on sleep parameters and subjective sleep quality in both pri-mary and comorbid insomnia. Few studies have evaluated theimpact of adherence on sleep outcomes, yet some findings suggestthat there may be an association between adherence to SR, SHE, SCand sleep improvements. Patients and providers invest substantialtime and effort in CBTI, despite being in an environment of scarcehealthcare resources. It is plausible that even modest improve-ments in adherence could result in better patient retention, greaterpatient and provider satisfaction, and most importantly, improvedoutcomes. It is possible that minimal investment in adherence mayrealize significant outcomes.

Initial steps to advance practice and research, therefore, include1) establish standard measures and definitions of adherence; 2)develop standard measures and definitions of treatment response;3) evaluate adherence and its relationship to treatment outcomeacross a wide-range of insomnia populations in different settingswith different CBTI delivery modalities; and 4) consider the role oftheory/models of behavior change in explaining relationship be-tween adherence and outcomes.

Uniform measures of adherence and treatment outcomes areneeded to enhance the comparability of studies. Standardized in-struments are needed for screening and outcome assessment. Forexample, Morin and colleagues63 examined psychometric indicesof the ISI to detect cases of insomnia in a population-based sampleand to evaluate treatment response. A cutoff score of 10 was opti-mal for detecting insomnia cases in the community sample. Ina clinical sample, a change score of�8.4 points was associated withmoderate improvement.63

Investigations are warranted that evaluate adherence to CBTIprotocols that contain additional treatment components, or aredelivered via other modalities. While the studies reviewed here areinformative, it is difficult to arrive at a definitive conclusion aboutthe state of the science, in part due to the wide variation in studymeasures. To address measurement issues, adherence must bemeasured in multiple ways, not simply inferred,64 and futurestudies should include some type of adherence evaluation for allCBTI components. Studies that regularly measure perceivedimprovement, may discern whether participants’ decliningattendance and adherence are due to mitigation of the problem orsymptom.

Adherence metrics are largely based on clinical judgment of theallowable time before or after a prescribed bedtime or rise time thatcould be considered “adherent” to the therapist’s recommendation.The validity of different time intervals as the cut off for adherence toprescribed sleep times should be analyzed in large samples of thosewith comorbid as well as primary insomnia. Further analysesshould determine whether the factors that predict sleep restrictionadherence also predict adherence to stimulus control, sleep hy-giene, and cognitive therapy. Larger adherence studies are neededto validate sleep diary data against objective data.

Large-scale RCTs focused on adherence are needed to fill theknowledge and clinical practice gaps identified in this review. Italso would be valuable to target a variety of clinically heteroge-neous groups of patients who experience insomnia. Studies guidedby theories or models of behavior change to explain the relation-ship between adherence and outcomes, utilize measures derivedfrom the theory, and interpret findings are needed to advance thefield.

Finally, qualitative research with a focus on the individualmeaning of adherence can provide important insight to the un-derlying adherence puzzle. Qualitative methods may uncover theanswers to questions such as whether individuals do not adherebecause they find CBTI is too difficult? or they do not believe it canhelp? or they give up after several weeks without improvement?Perhaps, they find one specific component of CBTI so effective thatthe other components are unnecessary. Understanding the nuancesof behaviors and adherence to insomnia therapy and other sleepdisorders is warranted.

Adherence factors and guidelines from other sleep disordersand populations utilizing cognitive behavioral therapy (CBT) maybe helpful to frame the findings in this review. For example, a re-view of adherence to positive airway pressure (PAP) treatment forobstructive sleep apnea indicates that PAP is effective, but adher-ence is restricted by several disparate factors.12 Among these fac-tors are those related to the efficacy of the treatment and itscomfort as well as psychological and social factors that likely reflect

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464 463

a person’s approach to health management and behavior change.Interventions to improve adherence to PAP provide some insight tothe limited effects of comfort management on adherence. Thepsychosocial aspects that contribute to behavior change providepromise for improving adherence, however, studies of social sup-port, relationship-based approaches, and other theory-driven in-terventions, are warranted for PAP adherence as well as CBTI.12

Guidelines for adherence problems in those with serious andpersistent mental illness65 suggest that psychological/program-matic interventions should include ongoing symptom/side effectmonitoring and management; psychoeducation (e.g., counselingand written/audiovisual materials); and environmental supports(e.g., more frequent and/or longer visits with providers/therapiststo improve therapeutic alliance). For example, adherence to inter-net CBT for anxiety and depression, as measured by “homework”completion, was improved with email/phone reminders and morefrequent clinician contact in a study of over 2000 patients referredfor generalized anxiety, depression, or social phobia.66 In-vestigators concluded that improved adherence to internet CBT isan important determinant of its effectiveness. Adherence to treat-ment for other sleep disorders, as well as CBT for mental illness andinsomnia appear to involve many different factors, thus, a variety ofstrategies may be needed.

Steiner64 recently suggested that the design of adherence in-terventions has been guided by the mistaken assumption thatadherence is a simple behavior that can be predicted from patientcharacteristics. Instead, adherence should be conceptualized asa set of interacting behaviors influenced by individual, social, andenvironmental forces. Thus, counseling for behavioral change iscomplex and needs to be complemented by identification andremoval of multifaceted barriers to treatment.

Practice points

1) Insomnia is a significant problem in adults worldwide

2) Evidence supports the efficacy of CBTI, however, evi-

dence of the impact of adherence is inconclusive

3) CBTI interventions are highly variable in terms of

components, mode of delivery, sleep measures used,

and type of therapist

4) Methods of measuring adherence are also inconsistent

across studies, contributing to difficulty in guiding

practice

5) Evidence suggest that greater adherence to CBTI rec-

ommendations may lead to greater improvements in

sleep

Research agenda

Future CBTI studies should aim to:

1) Establish standard measures and definitions of

adherence

2) Develop accepted measures and guidelines for treat-

ment response

3) Target adherence and its relationship to treatment

outcome across a wide-range of insomnia populations

in different settings with different CBTI delivery

modalities

4) Evaluate the role of theory/models of behavior change

in explaining relationship between adherence and

outcomes

Acknowledgments

This study was funded by the National Institute of Health andNational Institute of Nursing Research (1K23NR010587) and theAmerican Nurses Foundation (#2010-049).

References

1. Kraus SS, Rabin LA. Sleep America: managing the crisis of adult chronicinsomnia and associated conditions. J Affect Disord 2011;136:192e212.

2. Morin CM, LeBlanc M, Belanger L, Ivers H, Merette C, Savard J. Prevalence ofinsomnia and its treatment in Canada. Can J Psychiatry 2011;56:540e8.

3. National Institutes of Health State of the Science Conference Statement.Manifestations and management of chronic insomnia in adults. Sleep 2005;28:1049e57.

4. Leger D, Bayon V. Societal costs of insomnia. Sleep Med Rev 2010;14:379e89.5. Taylor DJ, Mallory LJ, Lichstein KL, Durrence HH, Riedel BW, Bush AJ. Comor-

bidity of chronic insomnia with medical problems. Sleep 2007;30:213e8.6. Kapur VK, Redline S, Nieto FJ, Young TB, Newman AB, Henderson JA. The

relationship between chronically disrupted sleep and healthcare use. Sleep2002;25:289e96.

7. Kleinman NL, Brook RA, Doan JF, Melkonian AK, Baran RW. Health benefitcosts and absenteeism due to insomnia from the employer’s perspective:a retrospective, case-control, database study. J Clin Psychiatry 2009;70:1098e104.

8. Daley M, Morin CM, LeBlanc M, Gregoire JP, Savard J, Baillargeon L. Insomniaand its relationship to health-care utilization, work absenteeism, productivityand accidents. Sleep Med 2009;10:427e38.

9. Ebben MR, Spielman AJ. Non-pharmacological treatments for insomnia.J Behav Med 2009;32:244e54.

10. Morgenthaler T, Kramer M, Alessi C, Friedman L, Behlecke B, Brown T, et al.Practice parameters for the psychological and behavioral treatment ofinsomnia: an update. An American Academy of Sleep Medicine report. Sleep2006;29:1415e9.

*11. Morin CM, Bootzin RR, Buysse DJ, Edinger JD, Espie CA, Lichstein KL. Psycho-logical and behavioral treatment of insomnia: update of the recent evidence(1998e2004). Sleep 2006;29:1398e414.

12. Matthews EE, Aloia MS. Continuous positive airway pressure treatment andadherence in obstructive sleep apnea. Sleep Med Clin 2009;4(4):473e85.

13. Dimatteo MR. Variations in patients’ adherence to medical recommendations:a quantitative review of 50 years of research. Med Care 2004;42:200e9.

14. World Health Organization. Adherence to long-term therapies: evidence foraction. Geneva: World Health Organization; 2003.

15. Haynes EB. Introduction. In: Haynes EB, Sackett DL, Taylor DW, editors. Com-pliance in health care. Baltimore, MD: The Johns Hopkins Press; 1978. p. 1e18.

16. Sackett DL. Introduction and magnitude of compliance and noncompliance.In: Sackett DL, Haynes EB, editors. Compliance with therapeutic regimens.Baltimore, MD: The Johns Hopkins University Press; 1976. p. 1e25.

17. Benner JS, Glynn RJ, Mogun H, Neumann PJ, Weinstein MC, Avorn J. Long-termpersistence in use of statin therapy in elderly patients. J Am Med Assoc2002;288:455e61.

18. Carpenter R. Perceived threat in compliance and adherence research. Nurs Inq2005;12:192e9.

19. Chatterjee JS. From compliance to concordance in diabetes. J Med Ethics2006;32:507e10.

20. Morin CM, Vallieres A, Guay B, Ivers H, Savard J, Merette C, et al. Cognitivebehavioral therapy, singly and combined with medication, for persistentinsomnia: a randomized controlled trial. J Am Med Assoc 2009;301:2005e15.

21. Siebern AT, Manber R. Insomnia and its effective non-pharmacologic treat-ment. Med Clin North Am 2010;94:581e91.

22. Yang CM, Spielman AJ, Glovinsky P. Nonpharmacologic strategies in themanagement of insomnia. Psychiatr Clin North Am 2006;29:895e919.

23. Jacobs GD, Pace-Schott EF, Stickgold R, Otto MW. Cognitive behavior therapyand pharmacotherapy for insomnia: a randomized controlled trial and directcomparison. Arch Intern Med 2004;164:1888e96.

24. Sivertsen B, Omvik S, Pallesen S, Bjorvatn B, Havik O, Kvale G, et al. Cognitivebehavioral therapy vs zopiclone for treatment of chronic primary insomnia inolder adults: a randomized controlled trial. J Am Med Assoc 2006;295:2851e8.

25. Dimatteo MR, Giordani PJ, Lepper HS, Croghan TW. Patient adherence andmedical treatment outcomes: a meta-analysis. Med Care 2002;40:794e811.

26. Ong JC, Kuo TF, Manber R. Who is at risk for dropout from group cognitive-behavior therapy for insomnia? J Psychosom Res 2008;64:419e25.

*27. Hebert EA, Vincent N, Lewycky S, Walsh K. Attrition and adherence in theonline treatment of chronic insomnia. Behav Sleep Med 2010;8:141e50.

*28. Vincent N, Lewycky S, Finnegan H. Barriers to engagement in sleep re-striction and stimulus control in chronic insomnia. J Consult Clin Psychol2008;76:820e8.

* The most important references are denoted by an asterisk.

E.E. Matthews et al. / Sleep Medicine Reviews 17 (2013) 453e464464

*29. Espie CA, Inglis SJ, Tessier S, Harvey L. The clinical effectiveness of cognitivebehaviour therapy for chronic insomnia: implementation and evaluation ofa sleep clinic in general medical practice. Behav Res Ther 2001;39:45e60.

30. Morgan K, Thompson J, Dixon S, Tomeny M, Mathers N. Predicting longer-term outcomes following psychological treatment for hypnotic-dependentchronic insomnia. J Psychosom Res 2003;54:21e9.

*31. Edinger JD, Wohlgemuth WK, Radtke RA, Marsh GR, Quillian RE. Cognitivebehavioral therapy for treatment of chronic primary insomnia: a randomizedcontrolled trial. J Am Med Assoc 2001;285:1856e64.

32. Morin CM, Colecchi C, Stone J, Sood R, Brink D. Behavioral and pharmaco-logical therapies for late-life insomnia: a randomized controlled trial. J AmMed Assoc 1999;281:991e9.

33. Harbour R, Miller J. A new system for grading recommendations in evidencebased guidelines. BMJ 2001;323:334e6.

*34. Chambers MJ, Alexander SD. Assessment and prediction of outcome fora brief behavioral insomnia treatment program. J Behav Ther Exp Psychiatry1992;23:289e97.

35. Edinger JD, Olsen MK, Stechuchak KM, Means MK, Lineberger MD, Kirby A,et al. Cognitive behavioral therapy for patients with primary insomnia orinsomnia associated predominantly with mixed psychiatric disorders: a ran-domized clinical trial. Sleep 2009;32:499e510.

*36. Riedel BW, Lichstein KL. Strategies for evaluating adherence to sleep restric-tion treatment for insomnia. Behav Res Ther 2001;39:201e12.

*37. Harvey L, Inglis SJ, Espie CA. Insomniacs’ reported use of CBT components andrelationship to long-term clinical outcome. Behav Res Ther 2002;40:75e83.

38. Phillips BA, Collop NA, Drake C, Consens F, Vgontzas AN, Weaver TE. Sleepdisorders and medical conditions in women. Proceedings of the women &sleep workshop, National Sleep Foundation, Washington, DC, March 5e6,2007. J Womens Health (Larchmt) 2008;17:1191e9.

39. Van Daele T, Hermans D, Van Audenhove C, Van den Bergh O. Stress reductionthrough psychoeducation: a meta-analytic review. Health Educ Behav 2012;39:474e85.

40. Vaidya V, Partha G, Karmakar M. Gender disparities in utilization of pre-ventive care service in United States. J Womens Health (Larchmt) 2012;21:140e5.

*41. Bouchard S, Bastien C, Morin CM. Self-efficacy and adherence to cognitive-behavioral treatment of insomnia. Behav Sleep Med 2003;1:187e99.

42. Perlis ML, Smith MT, Orff H, Enright T, Nowakowski S, Junquist C, et al. Theeffects of modafinil and cognitive behavior therapy on sleep continuity inpatients with primary insomnia. Sleep 2004;27:715e25.

43. Vincent NK, Hameed H. Relation between adherence and outcome in thegroup treatment of insomnia. Behav Sleep Med 2003;1:125e39.

*44. Berger AM, VonEssen S, Kuhn BR, Piper BF, Agrawal S, Lynch JC, et al.Adherence, sleep, and fatigue outcomes after adjuvant breast cancer chemo-therapy: results of a feasibility intervention study. Oncol Nurs Forum 2003;30:513e22.

*45. Matthews EE, Schmiege SJ, Cook PF, Berger AM, Aloia MS. Adherence tocognitive behavioral therapy for insomnia (CBTI) among women followingprimary breast cancer treatment: a pilot study. Behav Sleep Med 2012;10:217e29.

46. McChargue DE, Sankaranarayanan J, Visovsky CG, Matthews EE, Highland KB,Berger AM. Predictors of adherence to a behavioral therapy sleep interventionduring breast cancer chemotherapy. Support Care Cancer 2010;20:245e52.

*47. Tremblay V, Savard J, Ivers H. Predictors of the effect of cognitive behavioraltherapy for chronic insomnia comorbid with breast cancer. J Consult ClinPsychol 2009;77:742e50.

48. Manber R, Bernert RA, Suh S, Nowakowski S, Siebern AT, Ong JC. CBT forinsomnia in patients with high and low depressive symptom severity:adherence and clinical outcomes. J Clin Sleep Med 2011;7:645e52.

49. Vincent N, Lewycky S. Logging on for better sleep: RCT of the effectiveness ofonline treatment for insomnia. Sleep 2009;32:807e15.

50. Edinger JD, Bonnet MH, Bootzin RR, Doghramji K, Dorsey CM, Espiee CA, et al.Derivation of research diagnostic criteria for insomnia: report of an AmericanAcademy of Sleep Medicine Work Group. Sleep 2004;27:1567e96.

51. American Academy of Sleep Medicine. The international classification of sleepdisorders: diagnositic and coding manual. 2nd ed. Westchester, IL: AmericanAcademy of Sleep Medicine; 2005.

52. American Psychiatric Association. Diagnostic and statistical manual of mentaldisorders. 4th ed. Washington, DC: American Psychiatric Association; 2000.

53. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh sleepquality index: a new instrument for psychiatric practice and research. Psy-chiatry Res 1989;28:193e213.

54. Ajzen I. The theory of planned behaviour. Organizational Behav Hum DecisProcess 1991;50:170e211.

55. Prochaska JO, Diclemente CC. Stages and processes of self-change of smoking:toward an integrative model of change. J Consult Clin Psychol 1983;51:390e5.

56. Bandura A. Self-efficacy: toward a unifying theory of behavioral change.Psychol Rev 1977;84:191e215.

57. Bandura A. Social foundations of thought and action: a social cognitive theory.Englewood Cliffs, N.J: Prentice Hall; 1986.

58. Rosenstock IM. Historical origins of the health belief model. Health EducMonogr 1974:1e8.

59. Morin CM. Insomnia: psychological assessment and management. New York:Guilford Press; 1993.

60. Watzke B, Rueddel H, Koch U, Rudolph M, Schulz H. Comparison of thera-peutic action, style and content in cognitive-behavioural and psychodynamicgroup therapy under clinically representative conditions. Clin Psychol Psy-chother 2008;15:404e17.

61. Aloia MS, Goodwin MS, Velicer WF, Arnedt TJ, Zimmerman M, Skrekas J, et al.Time series analysis of treatment adherence patterns in individuals withobstructive sleep apnea. Ann Behav Med 2008;36:44e53.

62. Carney CE, Buysse DJ, Ancoli-Israel S, Edinger JD, Krystal AD, Lichstein KL, et al.The consensus sleep diary: standardizing prospective sleep self-monitoring.Sleep 2012;35:287e302.

63. Morin CM, Belleville G, Belanger L, Ivers H. The insomnia severity index:psychometric indicators to detect insomnia cases and evaluate treatmentresponse. Sleep 2011;34:601e8.

64. Steiner JF. Rethinking adherence. Ann Intern Med 2012;157:580e5.65. Velligan DI, Weiden PJ, Sajatovic M, Scott J, Carpenter D, Ross R, et al. Strat-

egies for addressing adherence problems in patients with serious and per-sistent mental illness: recommendations from the expert consensusguidelines. J Psychiatr Pract 2010;16:306e24.

66. Hilvert-Bruce Z, Rossouw PJ, Wong N, Sunderland M, Andrews G.Adherence as a determinant of effectiveness of internet cognitivebehavioural therapy for anxiety and depressive disorders. Behav Res Ther2012;50:463e8.