Articol Special 2:Articol Special 2.qxd.qxd

78

ARTICOLE DE SINTEZÃ

This article describes the first-ever clinic to usefour computer-aided cognitive behavior therapy(CCBT ) systems to guide interactive self-help forsufferers from anxiety and depressive disorders. Theclinic ran a free service in London for 15 months, in thelast few months of which clients were to use all four,not just two, of the systems at home supplemented bybrief live support on a phone helpline as needed. Thisarticle gives three detailed case illustrations to givepractitioners a taste of CCBT. It also summarizes theoutcome of a cohort reported in detail elsewhere(Marks, Mataix-Cols, & Gega, 2003) and draws lessonsfor the future.

Why Have Computer-Aided Self-Help? Thereare at least three reasons:

1. The demand for cognitive behavior therapy(CBT ) for anxiety and depressive disorders exceedsthe supply of suitably trained therapists, so waitinglists are often long.

2. Many sufferers prefer to avoid the stigma com-monly incurred by seeing a therapist.

3. Many people prefer to confide sensitive infor-mation to a computer rather than to a human.

COMPUTER-AIDED CBT SELF-HELP FOR ANXIETYAND DEPRESSIVE DISORDERS: EXPERIENCE OF ALONDON CLINIC AND FUTURE DIRECTIONS

Lina Gega, Isaac Marks, David Mataix-Cols*

* London UniversityCorrespondence concerning this article should be addressed to: Isaac Marks, 43 Dulwich Common, LondonSE217EU, United Kingdom; e-mail: [email protected].

Rezumat:

Acest articol descrie o clinicã unde un programde auto-ajutor, cu spectru larg realizat prin intermedi-ul calculatorului a reuºit sã creascã numãrul pa cien -þilor anxioºi/depresivi vãzuþi de fiecare medic, odatãcu scãderea timpului alocat de fiecare clinician pentrufiecare pacient, fãrã afectarea eficacitãþii. Starea mul-tor pacienþi s-a ameliorat dupã folosirea unuia dintrecele patru sisteme computerizate de auto-ajutor ter-apie cognitive comportamentalã (TCC) de auto-ajutorpentru fobie/panicã, depresie, tulburare obsesiv-com -pul sivã ºi anxistate generalizatã. Sistemele sunt accesi-bile la domiciliu, douã prin telefon, douã prin internet.O evaluare iniþialã minimã poate fi fãcutã prin telefonde cãtre un clinician, iar dacã pacientul se blocheazã înparcurgerea programului, atunci poate obþine un scurtsfat direct de la un terapeut, prin telefon. Astfel de ser-vicii de extindere a activitãþii terapeutice oferã speranþeîn creºterea comfortului în utilizare ºi a confidenþiali tã -þii intervenþiilor de auto-ajutor ghidat, ceea c ear ducela scãderea costului per pacient al aplicãrii TCC ºi di -mi nuând stigma. Cazurile exemplificate ilustreazã pro-cesul clinic ºi rezultatele unui astfel de tratament asistatde calculator.

Cuvinte cheie: tratament asistat de calculator;auto-ajutor; terapie cognitiv comportamentalã; fobie;panicã; tulburare obsesiv-compulsivã; depresie; anxi-etate generalizatã; terapie prin internet; rãspuns vocalinteractiv

Abstract:

This article describes a broad-spectrum, com-puter-aided self-help clinic that raised the throughputof anxious/depressed patients per clinician and low-ered per-patient time with a clinician without impair-ing effectiveness. Many sufferers improved by usingone of four computer-aided systems of cognitivebehavior therapy (CBT) self-help for phobia /panic,depression, obsessive-compulsive disorder, and gener-al anxiety. The systems are accessible at home, twoby phone and two by the Web. Initial brief screening bya clinician can be done by phone, and if patients getstuck they can obtain brief live advice from a ther-apist on a phone helpline. Such clinician- extendersystems offer hope for enhancing the convenience andconfi- dentiality of guided self-help, reducing the per-patient cost of CBT, and lessening stigma. The caseexamples illustrate the clinical process and out-comes of the computer-aided system. © 2003 WileyPeriodicals, Inc. J Clin Psychol/In Session 60:147–157, 2004.

Keywords: computer-aided treatment; self-help;cognitive behavior therapy; phobia; panic; obses-sive-compulsive disorder; depression; generalizedanxiety; Internet-accessed therapy; interactive voiceresponse.

Revista Românã de Psihiatrie, seria a III-a, vol. IX, nr. 2-3, 2007

79

Why Have Computer-Aided Self-Help at Home?Again, there are at least three reasons:

1. It is more convenient to do CCBT at homerather than to travel somewhere to have it. This diffi-culty is compounded in sufferers who become house-bound due to agoraphobia, social anxiety, obsessions,or compulsions.

2. Doing CCBT at home guided by a system on adistant computer which is accessed by Internet or byphone eases the incorporation of new advances ontothat system so that sufferers can benefit from thembefore they are widely known. Even in the mentalhealth field, a surprising number of professionals do notknow that anxiety and depressive disorders have longbeen treated successfully by appropriate CBT, and thatwith only brief input from a professional, self-help hasbeen effective.

3. Doing CCBT at home guided by a system on adistant central computer reached by Internet or phoneeases audit by that system of the outcome of mass-es of patients doing CBT.

None of these reasons would suffice if CCBTyielded inferior results to traditional CBT either face-to-face or by phone. However, evidence is accu-mulating that CCBT works well ( National Institute ofClinical Excellence of England, 2002).

A BROAD-SPECTRUM CCBT CLINIC INLONDON

Definition and Choice of CCBT

The clinic defined CCBT systems as those whichhelp the patient rather than the therapist make most ofthe decisions about how to devise, execute, and com-plete CBT, including appropriate homework andrelapse prevention. The clinic chose to use FearFighterfor phobia /panic (Kenwright, Liness, & Marks, 2001;Marks, Kenwright, McDonough, Whit- taker, &Mataix-Cols, in press), Cope for nonsuicidal depression(Osgood-Hynes et al., 1998), and BTSteps for obses-sive-compulsive disorder (Greist et al., 2002). TheseCCBT systems all (a) allowed a therapist to delegate tothem most of the tasks required to guide a patientthrough CBT self-help for anxiety or depressive disor-ders with a savings of at least two thirds of the thera-pist’s time per patient, (b) could give patients access athome either immediately (by phoning Cope andBTSteps ) or eventually (via FearFighter on the Web),and (c) were already of proven value in past researchtrials.

The clinic added a fourth CCBT system calledBalance (Yates, 1996) for generalized anxiety/milddepression. This is a shorter, less interactive system andtakes over fewer CBT tasks than the other three sys-tems. In CD-ROM form, patients also could use Bal-

ance at home; a modified Internet version is now avail-able as well.

Though CCBT is sometimes wrongly described asInternet therapy and FearFighter can be accessed viathe Internet, none of the clinic’s systems constitutedInternet therapy. “Internet therapy” best denotes treat-ment where the patient and therapist communicate byInternet (thus easing patient–therapist communicationwhen it is convenient for one party to send a messageand the other to answer it after a delay). In this regard,it resembles a phone voice-mail system. Internet thera-py has its therapeutic decisions made not by a comput-er but by a live therapist in real time just as inface-to-face or phone therapy. Internet therapy thusdoes not save much of the therapist’s time.

In contrast to Internet therapy, the FearFighter,Cope, and BTSteps computer systems help the patientmake most CBT self-help decisions and thus save agreat deal of therapists’ time. The therapist’s role in theclinic was restricted to briefly screening the patient andoffering live advice (latterly solely by phone) if thepatient got stuck during CCBT.

The clinic’s patients accessed the Cope andBTSteps CCBT systems by phoning a computer on aninteractive voice response (IVR) system. This was notphone therapy in the usual sense of that term. Patientsphoning the Cope and BTSteps IVR systems read amanual before phoning the computer to do CCBT, andthe computer helped them make most therapy decisionsduring the calls, thus saving therapist time.

Design and Operation of the CCBT Clinic

The clinic’s broad-spectrum design was intended togive anxiety and depression sufferers access to one offour CCBT self-help systems. It publicized its servi ce inlocal general practitioner offices, community men talhealth centers, psychiatric outpatient clinics, lo cal news-papers, Yellow Pages, patient organizations, and else-where. The clinic accepted self-referrals who com pleteda screening questionnaire they had obtained from one ofthe aforementioned facilities or by phoning the clinic.

Inclusion criteria were presence of an anxiety ordepressive disorder, motivation to do self-help, and nosubstance abuse, psychosis, or active suicidal plans.From the screening questionnaire, the staff judgedreferrals’ likely suitability for CCBT and offered thema30-min screening interview by phone, or face-to-face atthe clinic in the case of earlier referrals. Broad diag-noses were made using a checklist summarizing rel-evant ICD-10 diagnostic criteria. Diagnoses were: 71depression, 60 phobia /panic disorder, 35 generalizedanxiety disorder, 35 obsessive-compulsive disorder(OCD), 26 stress/adjustment disorder, 7 mixed anxi-ety/depression, and 6 somatoform disorder. The clinic’sstaff were mainly two nurse practitioners totaling onlyone full-time-equivalent clinician; in addi- tion, aresearch psychologist took on a mainly research ratherthan a clinical role.

Lina Gega, Isaac Marks, David Mataix-Cols: Computer-Aided CBT

80

How Clinic Patients Did CCBT

Staff gave patients who proved suitable at thescreening interview an identification number allowingaccess to the clinic’s four CCBT self-help systems thatmost suited them: FearFighter for phobia /panic, Copefor nonsuicidal depression, BT-Steps for (OCD), orBalance for general anxiety/mild depression. Patientsknew that information given to the CCBT system wasconfidential to staff and could not be accessed withoutknowing the patient’s identification number and pass-word ( Many said they told the computer sensitivethings they would not confide to a human.) None of thesystems stored personal names or addresses.

Patients were told they could use their system asmuch as they wished. They were advised to useFearFighter, Cope, or BT-Steps at least six times over12 weeks. During office hours, they also had six briefscheduled therapist contacts by phone or face-to-facefor advice. Users of Balance (which is more basic thanthe other three systems) were asked to use it at leastthree times over four weeks and to have three brieftherapist contacts by phone or face-to-face over the fourweeks.

Where Patients Accessed CCBT

Patients accessed FearFighter on a PC mostly atthe clinic, and occasionally at a free Internet café ormedical center. When FearFighter became available onthe Web, some could access it around the clock on acomputer at home or elsewhere linked to the Internet.

The clinic gave Cope and BT-Steps users self-helpbooklets to guide their free phone calls made mostlyfrom home to either Cope’s or BT-Steps’s IVR systemin a computer in Madison, WI, U.S.A. Users couldphone the computer from home at any time for as longand as often as they desired, and drove their interviewsby key presses on their telephone keypad. The comput-er faxed to the clinic weekly reports of patients’ phonecalls, their duration, and the modules accessed, and forCope patients, suicide risk. Had risk become high,which never happened, this would have been immedi-ately faxed or phoned to the clinic.

Balance users accessed the system by a PC with aCD-ROM drive at the clinic, their home, a free Internetcafé, or a physician’s office.

CASE ILLUSTRATION 1: COPE FORDEPRESSION

Presenting Problem /Client Description

Jo was a 40-year-old woman, unemployed,divorced, and living on her own. Prompted by a postershe saw at her general practitioner’s office, she con-tacted the self-help clinic. She completed and sent in ascreening questionnaire and was offered a screeninginterview the same week. She chose to have it face-to-face rather than by phone as she preferred to disclosepersonal information to a professional whom she couldsee and judge as trust- worthy. At her 30-min screening

interview, she described racing thoughts, sleeplessness,agitation and inability to relax, fear of death, tearful-ness, suicidal ideas, tiredness, and heartburn. She drankover 50 units of alcohol a week, smoked 35 cigarettes aday, and used cocaine about once a month. She alsotook prescribed and black-market sedatives.

Jo had an unsupportive partner and financial prob-lems.

She had been physically and sexually abusedrepeatedly during childhood and mar- riage, and hadattempted suicide many years before. She also had beendepressed since age 16 years. Her physicians pre-scribed diazepam until she was 25, and she becameaddicted to sedatives. Jo felt dismissed by and a nui-sance to her physicians, saying that medication wastheir easy option and they were unaware of psycholog-ical treatments for depression. She was angry that shecould not afford private treatment and was on a longwaiting list. She worried that she could not choosewhether her therapist was a man or a woman, a quali-fied therapist or a trainee, and the type of treatment shewould receive.

Case Formulation

Jo was diagnosed with severe generalized anxietydisorder and recurrent moderate depression suitablefor CBT with Cope.

Course of Treatment

The clinic offered Jo phone access to Cope’sphone-IVR self-help system for depression and a set ofCope explanatory booklets with sections to read beforemaking each Cope call. She completed Cope over 12weeks, during which time she worked on all of itsmain self-help modules: constructive thinking, behav-ioral activation, and assertiveness training. She spent atotal of 2.5 hr making 16 Cope phone calls. She alsoreceived two hours (seven phone contacts) of live thera-pist support, divided equally between progress reviews(asking her to complete pen-and-paper ratings and mon-itoring her state), general support (including discussingrelationship difficulties and referring her to a rela- tion-ship counselor), and treatment advice (weekly monitor-ing of her alcohol and sedative consumption, and listingpros and cons of using these as a way of coping).

Outcome and Prognosis

By a three-month follow-up, Jo had improved con-siderably. Her pre, post, and follow-up ratings were60% improvement in depression [Beck DepressionInventory; Beck, Ward, Mendelson, Mock, & Erbaugh,1961 (30, 15, and 10, respectively)]; 37% reduction inanxiety [Beck Anxiety Inventory; Beck, Epstein,Brown, & Steer, 1988 (46, 29, and 27, respectively)];and 30% improvement in work and social adjust-ment [Mundt, Marks, Greist, & Shear, 2002 (22, 12,and 15, respectively; score range 0 – 40)]. She feltmuch better, less anxious and depressed, and no longerdesperate. She drank 30% less alcohol and smoked


50% less, but relationship and financial problems stilldistressed her. She did not take antidepressant med-ications due to side effects, and during her three-month follow-up tried two outside counseling sessions,which she stopped as she found them unhelpful. Jo saidthat the clinic had helped her when she was desperate,and that Cope gave her focus and a goal to worktowards. She also valued her brief regular phonesupport with a clinician. She thought many peoplecould benefit from the Cope system but probably didnot know about it. She therefore became involved inpatient advocacy and local campaigns to raise aware-ness of treatment for depression, including Cope. Sheappeared empowered to deal better with the futurerelapses that seemed likely.

CASE ILLUSTRATION 2: FEARFIGHTER FORAGORAPHOBIA / PANIC DISORDER


A physician advised Dee, a single woman of 28years, to contact the clinic. Dee had not worked forthree years due to panic attacks when out alone or antic-ipating doing so. The attacks had begun five yearsearlier, seven months after nearly drowning onvacation abroad. Since then, she had not traveledabroad. She reported severe general anxiety everymorning and agitated depression. An anxiety manage-ment course and fluoxetine had not improved her con-dition for long. She also had begun dothiepin (75 mg/day) two weeks before screening. Dee completed andreturned a screening questionnaire. During her 30-minscreening interview face to face, she said she felt tenseand agitated in the morning and tired and sluggish inthe evening. She had been tearful for most days duringthe past few weeks and worried whether she would beable to get a job or enjoy social outings and vacationsin the future. Dee avoided using public transportand going to shops, pubs, and restaurants unaccom-panied as well as going to unfamiliar places farfrom home. She feared that if unaccompanied shewould panic with nobody available to help her, and iffar from home, she would be unable to get back to asafe place quickly.

Case Formulation

Dee was diagnosed with agoraphobia with panicdisorder suitable for self-exposure guided byFearFighter.

Course of Treatment

The clinic gave Dee immediate access to theFearFighter self-help system for phobia / panic. Shespent five hours at the clinic on the system over fivesessions, plus 55 min of live therapist support. Shecompleted FearFighter’s nine steps, including educa-

tion about the nature of fear and the principles of expo-sure, advice on how to get a friend or relative to be acotherapist, guidance on how to set specific and meas-urable goals and then carry out effective exposure,suggestions on anxiety management and how totroubleshoot common difficulties, and reward for ongo-ing exposure and monitoring anxiety. Therapist supportwas divided between progress review (homeworkachieved, monitoring mood and anxiety) and treatmentadvice (fine tuning exposure tasks to maximize gains).The therapist taught Dee diaphragmatic breathing tocontrol her anxiety symptoms enough to make her ini-tial exposure tasks tolerable. As she moved up heranxiety hierarchy, the therapist encouraged her to dofocused exposure without such anxiety control by revis-iting the habituation rationale.


By a one month follow-up, Dee felt muchimproved, used public transport, had a full- time job,and had been to a crowded concert and abroad whereher problem had begun. She no longer had general anx-iety and was not depressed. Her pre, post, and follow-up ratings were Fear Questionnaire ( FQ ) agora-phobia: 75% improvement (33, 8, and 9, respec-tively); FQ blood-injury: 83% improvement (12, 2, and3, respectively); FQ social phobia: 67% improvement(18, 6, and 8, respectively); FQ global phobia: 50%improvement (2, 1, and 1, respectively); FQ anxi-ety/depression: 80% improvement (35, 7, and 8, respec-tively); and work & social adjustment: 88% improve-ment (25, 3, and 4, respectively).

Before trying FearFighter, Dee had said thatshe would prefer a therapist over computer-guidedtreatment, but at posttreatment evaluation she said shewould choose 75% computer-guided and 25% therapist-guided treatment. She rated the clinic as 0 (“very good”on a scale in which 8 was “very poor”) and thought thata similar approach could help people overcome prob-lems similar to hers. Prognosis appeared good.

CASE ILLUSTRATION 3: BTSTEPS FOR OCD


Jim, a 51-year-old married man, contacted theclinic after seeing a poster in a physician’s office. Hehad been severely disabled by OCD and social phobiafor over 15 years, and had lived on disability benefitssince then. An intense fear of contamination led Jim toavoid touching most objects at home unless he com-pleted a self-cleaning routine of hand washing. Hespent more than three hours a day on these rituals. Mostof all, he dreaded and avoided touching the telephone;he feared contamination would spread to anything hetouched afterwards and to other parts of his body. Hestrongly avoided public toilets; if he ever had to useone, he felt compelled to change and wash his clothes

81


82

upon arriving home. He always carried a bottle ofcleaning fluid on him to wash his hands. The OCDimpaired his daily life and family relationships marked-ly and depressed him. Jim felt a “prisoner” to his OCDand had seriously considered suicide in the past. Bothan elder brother (reported to have had schizophrenia)and a younger brother (who abused drugs) had commit-ted suicide, which had greatly upset him.

At the time of his screening interview. Jim was onclomipramine (40 mg /day) and diazepam (15 mg /day).Past unsuccessful treatments had included “talkingtherapy” with psychologists, psychiatrists, and coun-selors and a one-week inpatient admission. He hadnever had a trial of exposure therapy.

Case Formulation

He was diagnosed as having OCD suitable forself-exposure, with self-imposed ritual prevention to beguided by BTSteps. Despite his avoidance of touchingthe telephone, he nevertheless agreed to use it to accessBTSteps.

Course of Treatment

At the end of the 65-min screening interview,the clinician explained the rationale of exposure andritual prevention. Jim decided to confront his fear ofusing the phone by using BTStep’s phone self-help sys-tem. The clinician gave him a BTSteps manual withseveral sections to read before making correspondingcalls to the BTSteps phone-IVR self-help system. Jimcould access BTSteps’s computer-guidance systemfrom the com- fort of his home any time day or night.By pressing keys on his telephone keypad, Jim decidedwhich of 800 different voice files of individually tai-lored advice the computer would play for him.

Jim phoned the BTSteps computer-guidance sys-tem 62 times, for a total of 513 min over 10 weeks. Healso had weekly phone support contacts with theclinician, which totaled 153 min. Of the 10 phone sup-port calls, eight were from the clinician to Jim and twofrom Jim to the clinician. Of the 153 min of cliniciantime, 60 were spent reviewing progress, 53 on generalsupport (e.g., How are you today? How’s your fami-ly?), and 40 on treatment advice (e.g., how to preventrituals more effectively).

The four hours the clinician spent with Jim onface-to-face screening and phone support was at least50% less than is usual with severe and chronic OCD,although it is four times more than the total of an hourof screening and support that is usual for users ofBTSteps (Greist et al., 2002).


After 10 weeks of treatment, Jim was usingthe phone regularly without washing his hands, nolonger divided his house into clean and unclean objects,and used public toilets anywhere he went. His scores onthe Yale–Brown Obsessive-Compulsive Scale (Good-man et al., 1989) fell markedly to almost normal levels

( Total: 20 to 4; Compulsions: 14 to 3; Obsessions: 6to 1). His mood also improved, with BeckDepression Inventory scores falling from 27 to 17.

At the end of self-help for his OCD, Jim’s socialphobia remained unchanged, but he said he felt confi-dent he could use the principles of self-exposure thera-py learned during his OCD treatment to tackle hissocial anxiety. At a two-month follow-up, his OCDremained much improved, and he had begun doing vol-untary work as a deliberate method to expose himself tosocial situations. Prognosis was promising.

RESULTS WITH CCBT

Over 12 months of intake, the self-help clinicreceived 355 screening questionnaires. Of these 355referrals, 8% were unsuitable on the questionnaire. Theremainder (327 referrals) were offered a screeninginterview with a clinician, of whom 266 attended; ofthese, 210 (79%) were suitable for and offered CCBT.Of the 210 suitable clients, 42 (20%) refused CCBTand 60 (29%) dropped out early or gave no posttreat-ment data. Unsuita- bles, refusals, dropouts, and com-pleters did not differ on initial severity, demographicvariables, or computer literacy. Of all referrals, slightlyover half were women, and a third were unemployed orstudents. Where information was available, over halfhad a current partner, and half had a postschool educa-tional qualification.

The sample was chronic (mean problem dura-tion eight years) with moderately severe problems.Where information was available, 39% had given upwork or were on long-term sick leave due to theirproblem, almost half were having current treatmentfrom their physician or a mental health professional,and about half were on psychotropic medication. Thevast majority had had past treatment for their problem,although only 20% had had CBT; 35% used computersmost days at work.

By posttreatment, improvement on work /socialadjustment was significant for FearFighter, Cope,and Balance users. Completers of each self-help systemalso improved significantly from pre- to posttreatmenton measures specific to their problem. The clin- icallymeaningful effect size of 0.8 or more was exceeded byFearFighter users on the FQ’s global phobia and anxi-ety/depression scores, by Cope users on depression andon work /social adjustment, and by BTSteps users onthe obsessive-compulsive Total and Obsessions andCompulsions subscores. Balance users did not attainthis clinically mean- ingful effect size on any meas-ure. Completers improved comparably to completersin other studies that used the same CCBT systems andmeasures.

Patients were fairly satisfied with their CCBT sys-tem, and even more satisfied with their live support andthe self-help clinic as a whole. They rated a marginal


83

preference for therapist over computer guidance.Satisfaction and preference (therapist vs. computer) rat-ings were similar among users of the four different sys-tems.

A mean of 58 days elapsed from patients’ startingto ending CCBT. Over that period, they had a mean of64 min of support from a clinician. About half thepatients had live support by phone and half face-to-faceat the clinic. The clinic’s patients who accessed thecomputer by phone spent very similar total times call-ing the computer as in previous studies—two hours onCope calls and four hours on BTSteps calls.

CLINICAL IMPLICATIONS

CCBT plus brief access to live advice enabledtherapists to treat many more patients per hour than ispossible without CCBT. Used in this way, CCBT is aclinician extender, nota clinician replacer. Apart from30 min of screening, staff gave a per-patient overallmean of about an hour of support distributed over threemonths. This support seems vital for most sufferers ifthey are to complete self-help successfully. The reduc-tion of per-patient time with a clinician is achieved bydelegating to a computer self-help system most of theroutine tasks involved in therapy, reserving for the cli-nician only those tasks which are not manageable by acomputer at present.

The mean of about one hour’s live support from aclinician is well below the mean of at least eight hoursper clinician usually needed by chronicanxious/depressed patients, although total treatmenttime per client differs from one CBT therapist to anoth-er. During the clinic’s year of intake, the full-time-equivalent of one clinician dealt with 355 referrals anddelegated most therapy tasks to CCBT. Throughput perclinician at the clinic thus far exceeded the 50 refer-rals a year that CBT therapists on average screenand treat (Marks, 1985), although therapists vary great-ly in this regard.

The greater throughput of patients per therapistwith the help of CCBT did not appear to sacrifice effec-tiveness. Anxiety and depression sufferers at the clinicimproved significantly and clinically meaningfully,and were fairly satisfied with CCBT despite a pref-erence for face-to-face care. When nearby physi-cians and a secondary CBT service recommendedthe clinic to many patients, this markedly reducedpatients’ consultations with the physicians and thephysicians’ referrals to secondary mental health servic-es. It also slightly shortened the waiting list for face-to-face CBT in secondary care.

During most of the clinic’s period of opera-tion, clients accessed two of the four systems (Copeand BTSteps ) by phone at home, but the rest attendedthe clinic in person to use a stand-alone PC for the othertwo systems. Eventually, FearFighter and a modified

form of Balance also became accessible at home on theWeb. It then became possible to offer most patientsCCBT self-help entirely at home without having toattend the clinic in person. In this final phase, the clin-ic became a virtual center, with clients obtaining CBTself-help advice at home via one of the four CCBT sys-tems that were available any time of day or night. Whenusers got stuck, they sought brief advice by phone fromstaff during office hours. Patients only obtained accessto CCBT after they had been deemed suitable in a 30-min screening interview with CBT staff by phone.Thus, staff were able to treat more patients than hadbeen possible before they used CCBT.

A rough cost comparison of CCBT with purelyface-to-face CBT was calculated. It assumed the samethroughput of patients managed per therapist usingCCBT as in the clinic, a U.S. $97-per-hour cost of aCBT therapist ( Netten & Curtis, 2000) and licensecosts of CCBT as noted by a U.K. regulatory body.Assuming administrative costs like those of the clinicand 15% overheads, the estimated per-patient costadvantage of CCBT over face-to-face CBT would risefrom about 15% per patient for 350 patients a year to41% per patient for 1,350 patients per year. This advan-tage rises with volume savings as the number ofpatients rises, and discounts any value from CCBT athome giving clients immediate rather than delayedaccess to CBT, unrestricted access, easier disclosure ofsensitive information, and removal of the need to trav-el to a therapist. This rough esti- mate of cost effective-ness needs to be validated.

In contrast to its lower per-patient cost, the totalcost of CCBT nationally might rise if so many userswho were previously untreated sought CCBT to off-set savings from lower per-patient costs. Widespreaddissemination of CCBT might eventually reducedemands on primary and secondary services and lessenmedication use and chronicity. Despite its apparent costeffectiveness, the self-help clinic eventually had toclose due to lack of funding—a problem common withnew healthcare technologies. It may take years forhealthcare funders and clinicians to widely agree tofund CCBT to reap its benefits.

Although a pragmatic evaluation such as our studymay reveal more about imple- mentation issues than arandomized controlled trial, it cannot tell us how muchthe patients may have improved due to the passage oftime, contact with a service, CBT, CCBT, the clini-cian’s brief help, or the psychotropic drugs which someclients took, nor is it known if similar gains might haveaccrued from offering an appropriate CBT self-helpbook plus access to a helpline. The amount of improve-ment should be regarded with caution because almosthalf of the clients were noncompleters (refusers plusdropouts), even though completers and noncompleterswere indistinguishable at the start.


84

CONCLUSION

CCBT is developing rapidly. Patients can now bescreened and, if suitable, help them- selves entirely athome by accessing two of the four CCBT systems usedby the clinic by phone and two on the Internet. Thosewho get stuck during self-help can receive support froma clinician on a live helpline. As referrals can nowbe screened for CCBT and supported by cliniciansentirely by phone while doing CCBT at home, self-helpclinics can act as call centers for wide areas.

The model suggested is stepped care, with CCBTself-help as a potential first port of call for most anxi-ety/depression sufferers. Those who fail to improvesufficiently with CCBT could go on to have live clini-cian-guided help.

Some might benefit from posted self-help instruc-tions (Burgess, Gill, & Marks, 1998) or self-helpbooks, perhaps with access to a live helpline.Books may cost less than CCBT, but are less inter-active and harder to modify on a large scale. It also ishard to track patients’ progress with books whereasCCBT on the Internet or a central IVR computer easesthe assessment of outcome on a mass scale.

Major hurdles at present include the reluctanceof healthcare funders to pay for CCBT and the lackof personnel trained to support it. It can take manyyears for new technology to become routine in thehealth services.

Select References/ Recommended Readings

Beck, A.T., Epstein, N., Brown, G., & Steer, R.A. (1988).An inventory for measuring clinical anxiety: Psychometric proper-ties. Journal of Consulting and Clinical Psychology, 56, 893–897.

Beck, A.T., Ward, C.H., Mendelson, M., Mock, J., & Erbaugh,J. (1961). An inventory for measuring depression. Archives ofGeneral Psychiatry, 4, 561–571.

Burgess, M., Gill, M., & Marks, I.M. (1998). Postal self-expo-

sure treatment of recurrent night- mares: Randomised controlled trial.British Journal of Psychiatry, 172, 257–262.

Goodman, W.K., Price, L.H., Rasmussen, S.A., Mazure, C.,Fleischman, R.L., Hill, C.L., Heninger, G.R., & Charney, D.S.(1989). The Yale–Brown Obsessive Compulsive Scale: I.Develop- ment, use and reliability. Archives of General Psychiatry,46, 1006–1011.

Greist, J.H., Marks, I.M., Baer, L., Kobak, K.A., Wenzel,K.W., Hirsch, M.J., Mantle, J.M., & Clary, C.M. (2002). Behaviourtherapy for obsessive compulsive disorder guided by a computer or

by a clinician compared with relaxation as a control. Journal ofClinical Psychiatry, 63, 138–145. Kenwright, M., Liness, S., &Marks, I.M. (2001). Reducing demands on clinicians by offer-ing computer-aided self-help for phobia /panic. Feasibility study.British Journal of Psychiatry, 179, 456– 459.

Marks, I.M., Kenwright, M., McDonough, M., Whittaker, M.,& Mataix-Cols, D. (in press). Saving clinicians’ time by delegatingroutine aspects of therapy to a computer: A randomised con- trolledtrial in phobia /panic disorder. Psychological Medicine.

Marks, I.M., Mataix-Cols, D., Kenwright, K., Cameron, R.,Hirsch, S., & Gega, L. (2003). Prag- matic evaluation of computer-aided self help for anxiety and depression. British Journal ofPsychiatry, 183, 57– 65.

Marks, I.M., & Mathews, A.M. (1979). Brief standard self-rat-ing for phobic patients. Behaviour Research and Therapy, 17,263–267.

Mundt, J.C., Marks, I.M., Greist, J.H., & Shear, K. (2002).Work and Social Adjustment Scale: A simple accurate measure ofimpairment in functioning. British Journal of Psychiatry, 180,461– 464.

Netten, A., & Curtis, L. (2000). Unit cost of health and socialcare 2000. Personal Social Services Research Unit, University ofKent, UK. Retrieved September 2002, from http://www.ukc.ac.uk / PSSRU/

NICE ( National Institute of Clinical Excellence of England).(2002). Appraisal consultation doc- ument & final appraisal determi-nation on computer-aided CBT. Retrieved September 2002, fromhttp://www.nice.org.uk

Osgood-Hynes, D.J., Greist, J.H., Marks, I.M., Baer, L.,Heneman, S.W., Wenzel, K.W., Manzo, P.A., Parkin, J.R., Spierings,C.J., Dottl, S.L., & Vitse, H.M. (1998). Self-administered psy-chotherapy for depression using a telephone-accessed computersystem plus booklets: An open U.S.–U.K. study. Journal of ClinicalPsychiatry, 58, 358–365.

Yates, F. (1996). Evaluation of the Balance computer interven-tion. Unpublished report to the Men- tal Health Foundation, London.

***

Documents

Articol Special 2:Articol Special 2.qxd.qxd