drug treatment of older people with affective disorders in the community: lessons from an attempted...

DRUG TREATMENT OF OLDER PEOPLE WITHAFFECTIVE DISORDERS IN THE COMMUNITY:

LESSONS FROM AN ATTEMPTEDCLINICAL TRIAL

TIM STEVENS1*, CORNELIUS KATONA2, MONICA MANELA3, VIVIENNE WATKIN4 AND GILL LIVINGSTON5

1Lecturer in Psychiatry of Old Age, UCLMS Dept of Psychiatry and Behavioural Sciences, London, UK2Professor of Psychiatry of the Elderly, UCLMS Dept of Psychiatry and Behavioural Sciences, London, UK

3Research Doctor, UCLMS Dept of Psychiatry and Behavioural Sciences, London, UK4Specialist Registrar in Psychiatry, UCLMS Dept of Psychiatry and Behavioural Sciences, London, UK

5Senior Lecturer in Psychiatry, UCLMS Dept of Psychiatry and Behavioural Sciences, London, UK

SUMMARY

Background. Depression and phobic anxiety disorders are the most common psychiatric disorders in people aged 65and over. SSRI antidepressants are e�ective in treating both conditions in younger people, and in treating depressionin hospital samples of older subjects. No studies have investigated the e�cacy of SSRIs in older people with theseconditions living in the community.

Objectives. To evaluate the e�cacy and feasibility of treating older people su�ering from depression and/or phobicanxiety in the community with ¯uoxetine alone.

Design. Subjects identi®ed as depressed and/or anxious at screening were o�ered open-label ¯uoxetine and werereassessed for a�ective illness at 3 and 6 months.

Measures. Outcome was assessed using the depression subscale of the Short Comprehensive Assessment andReferral Evaluation (Short-CARE) Scale and the Anxiety Disorder Scale.

Results. Of 67 subjects with depression and/or phobic anxiety, 55 (81%) were eligible to take ¯uoxetine. Fifty-four(98%) of these agreed to follow-up but only six (11%) agreed to take medication. No subject was still takingmedication by the end of the study. Among those subjects on whom follow-up data were available, 70% of subjectsdepressed at screening and 97% of those with phobic anxiety retained their diagnoses at 3 months; at 6 months, the®gures were 65% and 92% respectively.

Conclusions. Drug treatment alone is not acceptable to older patients in the community with depression and phobicanxiety disorders. Discussion of symptoms with an appropriate professional is insu�cient therapy on its own. Furtherwork is needed to evaluate the e�ectiveness of a key worker such as a mental health nurse in coordinating treatment ofpatients with these disorders. Copyright # 1999 John Wiley & Sons, Ltd.

KEY WORDSÐanxiety; depression; older people; antidepressant; ¯uoxetine; acceptability; feasibility; key worker

Depressive illness is common in older people livingin the community, with reported prevalence ratesof 10±15% (eg Manela et al., 1996; Ben-Arie et al.,1990). Anxiety disorders in this population are alsocommon, with recent community studies reportingrates of 10% and 12% for phobic, and 3.7% and4.7% for generalized anxiety disorder respectively

(Lindesay et al., 1989; Manela et al., 1996). Thelatter study reported an overall prevalence foranxiety disorder of 15%, equivalent to that fordepression. A review of the literature on anxietydisorders in the elderly by Flint (1994) concludedthat there was a high rate of comorbidity of thesedisorders with depression. Together, they con-stitute a major source of morbidity among theolder people in the community. Depression is alsoassociated with excess mortality, eg 34% at 4 years(Murphy et al., 1988) and 45% at 5 years (Kivela,1995).

CCC 0885±6230/99/060467±06$17.50 Received 28 May 1998Copyright # 1999 John Wiley & Sons, Ltd. Accepted 11 November 1998

INTERNATIONAL JOURNAL OF GERIATRIC PSYCHIATRY

Int. J. Geriat. Psychiatry 14, 467±472 (1999)

*Correspondence to: Dr T. Stevens, Department of Psychiatryand Behavioural Sciences, University College London MedicalSchool, Wolfson Building, Riding House Street, LondonW1N 8AA, UK. Tel: 0171 530 2337. Fax: 0171 323 1459.

OUTCOME OF AFFECTIVE DISORDERS INTHE COMMUNITY

Several studies have investigated the outcome ofdepression and/or anxiety in the community andfound that many subjects have not recovered onfollow-up. This may be related to undertreatment,as the two studies which reported the number ofdepressed people taking antidepressants bothfound very low rates (2/123, Green et al., 1994;and 11%, Livingston et al., 1997). Many olderpeople living at home with anxiety and depressiondo not receive treatment for these disorders and donot improve with time alone.

PHARMACOLOGICAL TREATMENT OFAFFECTIVE DISORDERS

There has been considerable research into thee�ectiveness and tolerability of pharmacologicaltherapy in depression in older people, mainly inpatients referred to secondary services, though lessinto medication for anxiety disorders. Severalstudies have challenged the view that the outcomeof depression in older people is worse than inyounger patients (eg Baldwin and Simpson, 1997;Alexopoulos et al., 1996; Brodaty et al., 1993). Arecent review article by Butler et al. (1997) reportedthat most patients over the age of 65 with depres-sion respond well to antidepressant treatment, andthat a combination of medication with psycho-therapy appears more e�ective than either alone.

The choice of antidepressant also appearsimportant in in¯uencing outcome, partly becauseof the di�erent side-e�ect pro®les of the di�erentclasses, which in turn a�ect compliance. Peabodyet al. (1986) identi®ed cardiac e�ects such asarrhythmias and antiadrenergic e�ects includingpostural hypotension as common and dangerous ina review of trials of tricyclic antidepressants(TCAs) and monoamine oxidase inhibitors(MAOIs) in the elderly. The newer antidepressants,such as the selective serotonin reuptake inhibitors(SSRIs) and selective noradrenaline reuptakeinhibitors (SNRIs), as well as atypical antidepres-sants such as trazodone and mianserin, appear tobe tolerated better by older subjects (Montgomery,1994; Hotopf et al., 1996). This is illustrated by onestudy, of 62 older acute medical inpatients withdepression, which found that not only was¯uoxetine more e�ective than placebo, but alsopatients with serious physical disease responded

signi®cantly better to drug treatment (Evans et al.,1997). In contrast, a trial by Koenig et al. (1989)attempted to examine the e�cacy and safety ofnortriptyline in the treatment of major depressionin older medical inpatients, but was halted atmidpoint as more than 80% of eligible subjectswere unable to participate because of medicalillness. In general, however, their advantages of asimple regimen, a clinically e�ective starting dose,fewer contraindications and a good safety pro®lemay be more likely to be observed in clinicalpractice than in controlled trials (Katona, 1995).

While the role of pharmacotherapy in theanxiety disorders is less researched, SSRIs havebeen shown to be useful in panic disorder, generaland phobic anxiety in some patients (den Boeret al., 1995).

No studies have yet been completed to examinethe acceptability, feasibility and e�cacy of o�eringdrug treatment to older people living at home andsu�ering from anxiety and/or depression. The aimof the present study was to establish the feasibilityand acceptability of pharmacological treatment,namely the SSRI ¯uoxetine, for older communitysubjects with persistent depression or phobicanxiety.

METHODS

Sample selection

The study sample was selected from a popula-tion identi®ed as su�ering from depression, phobicanxiety or both in a community epidemiologicalstudy carried out by Manela et al. (1996). Thisstudy had been carried out in Islington, an area ininner London shown by the Jarman underprivi-leged area score of 49 as the sixth most sociallydeprived region in England and Wales (Jarman,1983). Of 774 people over 65 who were approached,700 were interviewed using the shortened version ofthe Comprehensive Assessment and ReferralEvaluation (Short-CARE) (Gurland et al., 1984),a valid and reliable semi-structured questionnairewhich has screening scales to measure sleepdisorder, somatic symptoms and subjective mem-ory. There are also two diagnostic scales fordepression (validated against psychiatrists' judge-ment of depression of severity requiring psychiatrictreatment) and for dementia and a scale for activitylimitation (designed to identify those who needhelp with day-to-day living). The Anxiety DisorderScale, a semi-structured anxiety disorder scale

Copyright # 1999 John Wiley & Sons, Ltd. Int. J. Geriat. Psychiatry 14, 467±472 (1999)

468 T. STEVENS, C. KATONA, M. MANELA ET AL.

originally developed for the Guy's/Age Concernsurvey (Lindesay et al., 1989) which generatesdiagnoses of generalized anxiety and phobicanxiety and has been validated against clinicaldiagnosis, was also administered. Further detailsare given in an earlier paper (Manela et al., 1996).

A total of 694 subjects had completed the fullinterview; 84 subjects (12%) had been identi®ed ashaving phobic anxiety disorder and depression hadbeen present in 104 (15%). Twenty-three subjects(27%) with phobic disorder were also depressed.

In the present study, those subjects ful®lling`caseness' criteria for depression or phobic anxietywere approached after an interval of 2±3 years(mean 2.6) and invited to undergo a screeninginterview after providing verbal consent.

Study design

At baseline, all subjects providing verbal consentwere interviewed using the Short-CARE and theAnxiety Disorder Scale. Those subjects ful®llingany of the exclusion criteria (Table 1) were con-sidered ineligible to receive ¯uoxetine.

Subjects who ful®lled `caseness' criteria fordepression (score 56 on the depression subscaleof the Short-CARE) or phobic anxiety (AnxietyDisorder Scale) were then asked to sign informedconsent. They were seen by a psychiatrist (VW),who explained the study and the purpose of takingmedication. A complete medical and psychiatrichistory was obtained and a physical examinationperformed. Subjects willing to accept medicationwere then commenced on ¯uoxetine 20 mg daily.All subjects, whether taking tablets or not, were

asked to consent to follow-up and monitoring overup to 6 months.

All consenting subjects were re-interviewed at3 months and 6 months, whether on medication ornot, using the instruments employed in the ®rstperiod. Those receiving ¯uoxetine were againphysically examined.

Analysis

The data were recorded on SPSS-PC� (Norusis,1991).

RESULTS

Demographics

Of the 165 subjects eligible for interview, 25 haddied and 23 had moved out of the area. There were117 subjects still living in the area. Generalpractitioners were concerned about the health of®ve others and asked us not to contact them. Afurther 23 refused. A total of 86 (73.5%) of thoseavailable) subjects from the original study wereinterviewed, 65 of whom (76%) were female;11 (13%) were single, 26 (30%) married, 43 (50%)widowed and the remainder (7%) divorced orseparated. The median age of the sample was75 years with a range from 67 to 91 years.

Morbidity and follow-up data

At follow-up, 67 subjects were still classi®ed asdepressed and/or having a phobic anxiety disorderand one as having generalized anxiety only.Eighteen had neither depression nor an anxietydisorder. Fifty-one subjects had phobic anxietyand 34 were depressed (18 had both). Of these, 32su�ered from agoraphobia, 27 from speci®c phobiaand one from social phobia (some subjects hadmore than one phobic disorder).

Exclusion criteria

Thirteen of the 67 subjects (19%) were noto�ered pharmacotherapy because they ful®lled oneor more of the exclusion criteria. Nine were alreadyon antidepressants (two paroxetine, two ¯uoxetine,two amitryptyline, one lofepramine, one clomipra-mine, one dothiepin). Two were taking carbama-zepine, one of whom also had a history of recentextensive myocardial infarction and the other wasalso taking lofepramine as noted. One subject was

Table 1. Exclusion criteria for receiving ¯uoxetine

1. Participation in any clinical pharmacological study within

1 month preceding start of active treatment

2. Cardiovascular, hepatic or renal disease su�ciently severe

to preclude the use of ¯uoxetine

3. Depression of su�cient severity to require ECT

4. Allergy to ¯uoxetine

5. Signi®cant organic brain disease or a history of seizures

6. Schizophrenia or other psychotic disorders

7. Concurrent treatment with lithium, carbamazepine or

tryptophan

8. Use of MAOIs within 4 weeks prior to starting treatment

9. History within the previous 18 months of drug (including

alcohol) abuse

10. Presence of a terminal illness

11. Current use of antidepressant medication


DRUG TREATMENT OF OLDER PEOPLE IN THE COMMUNITY 469

abusing alcohol and one had advanced respiratorydisease.

Eligible subjects

Fifty-®ve subjects (82%) were eligible to receive¯uoxetine. Of these, 54 (98%) agreed to follow-up.One patient's psychiatrist asked that she not beincluded. Four (7.4%) subjects refused ¯uoxetinebecause they disliked taking medication; two(3.7%) stated that their symptoms were natural inthe circumstances and that medication was there-fore inappropriate; and one (1.9%) declined on thegrounds that her symptoms were not su�cientlydistressing to warrant drug therapy. An additional41 (76%) refused medication but gave no reason.Only six (11%) accepted medication.

Subjects on medication

Of the six subjects who were eligible and con-sented to take medication, two withdrew consentprior to taking any. Of the four remaining subjects,two had side-e�ects: one developed an urticarialrash and one developed pruritus. One subject waswithdrawn as her tablets went missing withoutexplanation and it was thought that she was eitherthrowing them away or taking too many. The ®nalsubject stopped taking tablets after 6 months butbefore the ®nal evaluation as he felt that `it doesn'thelp answering questions'. He had improved butwas still depressed. His suicidal thoughts haddisappeared.

Three-month outcome

Of the depressed subjects, 14 subjects remaineddepressed at 3 months and six were no longerdepressed. There were no data on 14. Of thosewith phobic anxiety, 30 remained cases andone recovered. There are no data on the other20 subjects.

Six-month outcome

Of the depressed subjects, 11 subjects remaineddepressed at 6 months and six were no longerdepressed. There were no data on 17. Of thosewith phobic anxiety, 24 remained cases andtwo recovered. There are no data on the other25 subjects.

DISCUSSION

In this study, designed to examine the acceptabilityand feasibility of treating a�ective disorders with awell-tolerated and e�ective antidepressant, ¯uox-etine, in 67 older people living in the community,we found that no subject completed the drug study.Some subjects were excluded mainly because theywere already on antidepressants (although those onTCAs were not on therapeutic British NationalFormulary (BNF) doses; Livingston et al., 1997).Very few subjects were excluded because the drugwas contraindicated and this suggests that it isfeasible to use ¯uoxetine in this setting. However,the vast majority of subjects excluded themselves.Drug treatment was therefore not acceptable tothem. This may have been because the subjectswere reluctant to be interviewed as they had beenapproached earlier, but this seems unlikely as alarge number of subjects agreed to further follow-up. The impression of the interviewers was thatalthough most subjects wanted to be visited, theyhad not presented to psychiatrists and some did notrecognize themselves as depressed. Those who didwere nonetheless reluctant to take medication. Thismay have been because those people in thecommunity who ful®l the diagnostic criteria fordepression are still unlike those who are usuallyseen in hospital and are regarded as clinical cases.It also accords with our clinical experience thatolder people, in contrast to younger people, arereluctant to take medication, a ®nding not con®nedto antidepressant drugs; for example, patientsrefusing antipsychotic medication in acuteinpatient units are typically older than those whoaccept treatment (Hoge et al., 1990).

The results at 3- and 6-month follow-up indic-ated that discussion of symptoms with an experi-enced professional was not a su�cient therapyalone, at least in subjects with already persistentdepression or anxiety, as of those patients on whomfollow-up data were available, 70% of thoseidenti®ed as depressed and 97% of those identi®edas having phobic anxiety at screening retained theirdiagnoses at 3-month follow-up. The ®gures at6-month follow-up were 65% and 92% respect-ively. These follow-up results are hampered by thehigh attrition rate, but it is unlikely that refusershad a better outcome.

Previous studies have shown that a communitypsychiatric nurse acting as case manager to work inconjunction with general practitioners and com-munity agencies can signi®cantly improve the



mental state of depressed elderly people, evenwhere the recommended introduction of anti-depressant medication is not possible (Mannet al., 1993; Waterreus et al., 1994; Blanchardet al., 1995). One study of intervention in depressedpeople receiving social services home care foundthat an individualized package of care designed bythe old age psychiatry team and implemented by aresearcher was e�ective in treating depression ascompared to usual GP care and that antidepressantprescription alone did not have a signi®cant e�ect(Banerjee et al., 1996). Similarly, a controlledintervention study in which key workers wereallocated to carers of elderly spouses with psychia-tric illness revealed signi®cant improvement in themental health of those carers who were mentally illat outset. This improvement followed an indi-vidually tailored intervention programme, but nosingle intervention such as antidepressant medica-tion or carer education was signi®cantly associatedwith improvement on its own (Hinchli�e andLivingston, 1995; Murray et al., 1997).

We conclude that the use of a speci®c serotoninreuptake inhibitor by itself, in this population, inwhich depression is associated with individualsu�ering, increased risk of suicide and higheconomic cost, is feasible but was not acceptablein the context of this study. The passage of timeand opportunity to talk about symptoms alone donot, in the main, lead to improvement. The use of acase management system with key worker imple-mentation, rather than physical treatment alone, ismore appropriate in this population.

ACKNOWLEDGEMENTS

We would like to thank Lilly Industries Ltd forpaying the salary of MM during this study, and thesubjects and their carers who took part.

REFERENCES

Alexopoulos, G. S., Meyers, B. S., Young, R. C.,Kakuna, T., Feder, M., Einhorn, N. and Rosendahl,E. (1996) Recovery in geriatric depression. Arch. Gen.Psychiat. 53(4), 305±312.

American Psychiatric Association (1994) Diagnosticand Statistical Manual of Mental Disorders, edn. 4American Psychiatric Association, Washington, DC.

Baldwin, R. C. and Simpson, S. (1997) Prognosis andoutcome studies in latelife depression. Clin. Neurosci.4, 16±22.

Banerjee, S., Shamash, K., Macdonald, A. J. and Mann,A. (1996) Randomised controlled trial of the e�ect ofintervention by psychogeriatric team on depression infrail elderly people at home. Brit. Med. J. 126(313),1058±1061.

Ben-Arie, O., Welman, M. and Teggin, A. F. (1990) Thedepressed elderly living in the community: A follow-upstudy. Brit. J. Psychiat. 157, 425±427.

Blanchard, M. R., Waterreus, A. and Mann, A. H.(1995) The e�ect of primary care nurse interventionupon older people screened as depressed. Int. J.Geriatr. Psychiat. 10, 289±298.

Brodaty, H., Harris, L., Peters, K., Wilhelm, K., Hickie,I., Boyce, P., Mitchell, P., Parker, G. and Eyers, K.(1993) Prognosis of depression in the elderly. Acomparison with younger patients. Brit. J. Psychiat.163, 589±596.

Butler, R. N., Cohen, G., Lewis, M. I., Simmons-Clemmons, W. and Sunderland, T. (1997) Latelifedepression: Treatment strategies for primary carepractice. Geriatrics 52(4), 51±52, 58±60, 63±64.

Copeland, J. R. M., Kelleher, M. J., Kellett, J. M. andGourlay, A. J. (1976) A semi-structured clinicalinterview for the assessment of diagnosis and mentalstate in the elderly: The Geriatric Mental StateSchedule. Psychol. Med. 6, 439±449.

den Boer, J. A., Westenberg, H. G., De Leeuw, A. S. andvan Vliet, I. M. (1995) Biological dissection of anxietydisorders: The clinical role of selective serotoninreuptake inhibitors with particular reference to ¯uvox-amine. Int. Clin. Psychopharmacol. 9(Suppl.4), 47±52.

Evans, M., Hammond, M., Wilson, K., Lye, M. andCopeland, J. (1997) Placebo-controlled treatment trialof depression in elderly physically ill patients. Int. J.Geriatr. Psychiat. 12, 827±824.

Flint, A. J. (1994) Epidemiology and comorbidity ofanxiety disorders in the elderly. Am. J. Psychiat 151(5),640±648.

Green, B. H., Copeland, J. R. M., Dewey, M. E.,Sharma, V. and Davidson, I. (1994) Factors associatedwith recovery and recurrence of depression in olderpeople: A prospective study. Int. J. Geriatr. Psychiat.9, 789±795.

Gurland, B., Golden, R., Teresi, J. A. and Challop, J.(1984) The Short-CARE. An e�ective instrument forthe assessment of depression and dementia. J.Gerontol. 39(2), 166±169.

Hinchli�e, A. C. and Livingston, G. (1995) The mentalhealth of carersÐcan it be improved? Prim. CarePsychiat. 1, 249±253.

Hoge, S. K., Appelbaum, P. S., Lawlor, T., Beck, J. C.,Litman, R., Greer, A., Gutheil, T. G. and Kaplan, E.(1990) A prospective, multicentre study of patients'refusal of antipsychotic medication. Arch. Gen. Psy-chiat. 47(10), 949±956.


DRUG TREATMENT OF OLDER PEOPLE IN THE COMMUNITY 471

Hotopf, M., Lewis, G. and Normand, C. (1996) AreSSRIs a cost-e�ective alternative to tricyclics? Brit. J.Psychiat. 168(4), 404±409.

Jarman, B. (1983) Identi®cation of underprivilegedareas. Brit. Med. J. 286, 1705±1709.

Katona, C. (1995) Rationalising antidepressants forelderly people. Int. Clin. Psychopharmacol.10(Suppl. 1), 37±40.

Kivela, S.-L. (1995) Long-term prognosis of majordepression in old age: A comparison with prognosisin dysthymic disorder. Int. Psychogeriat. 7(Suppl.),69±82.

Koenig, H. G., Veeraindar, G., Shelp, F., Kudler, H. S.,Cohen, H. J., Meador, K. G. and Blazer, D. G. (1989)Antidepressant use in elderly medical inpatients:Lessons from an attempted clinical trial. J. Gen. Int.Med. 4, 498±505.

Lindesay, J., Briggs, K. and Murphy, E. (1989) TheGuy's/Age Concern Survey prevalence rates of cogni-tive impairment, depression and anxiety in an urbanelderly community. Brit. J. Psychiat. 155, 317±329.

Livingston, G., Hawkins, A., Graham, N., Blizard andMann, A. H. (1990) The Gospel Oak study: Prevalencerates of dementia, depression and activity limitationamong elderly residents in inner London. Psychol.Med. 20, 137±146.

Livingston, G., Watkin, V., Milne, B., Manela, M. andKatona, C. (1997) The natural history of depressionand the anxiety disorders in older people. J. A�ect.Dis. 46, 255±262.

Manela, M., Katona, C. and Livingston, G. (1996) Howcommon are the anxiety disorders in old age? Int. J.Geriatr. Psychiat. 11, 65±70.

Mann, A., Blanchard, M. and Waterreus, A. (1993)Depression in older people. Some criteria for e�ectivetreatment. Encephale 19(Spec. No. 3), 445±450.

Montgomery, S. A. (1994) Long-term treatment ofdepression. Brit. J. Psychiat. 26(Suppl.), 31±36.

Murphy, E., Smith, R., Lindesay, J. and Slattery, J.(1988) Increased mortality rates in late-life depression.Brit. J. Psychiat. 152, 347±353.

Murphy, J., Manela, M., Shuttleworth, A. and Living-ston, G. (1997) An intervention study with husbandand wife carers of older people with a psychiatricillness. J. A�ect. Dis. 46, 279±284.

Norusis, M. J. (1991) SPSS/PC 4.SPSS Inc.Peabody, C. A., Whiteford, A. A. and Hollister, L.(1986) Antidepressants and the elderly. J. Am. Geriatr.Soc. 34(12), 869±874.

Waterreus, A., Blanchard, M. and Mann, A. (1994)Community psychiatric nurses for the elderly: Well-tolerated, few side-e�ects and e�ective in the treatmentof depression. J. Clin. Nurs. 3(5), 299±306.



drug treatment of older people with affective disorders in the community: lessons from an attempted...

Documents