Crisis resolution and home treatment teams for older people with mental illness

AIMS AND METHOD

We examined the impact of a crisis resolution and home treatmentteams (CRHTT) on hospital admission rates, bed days and treatmentsatisfaction among older people with mental illness and theircarers. We compared these factors in the 6 months before theservice started and 6 months after its introduction.

RESULTS

The CRHTT significantly reduced admissions (P<0.001), butthere was no significant difference in the length of hospitalstay as compared before and after the introduction of thisservice. There was a trend towards carers, but not patients,being more satisfied with treatment after the introductionof the CRHTT.

CLINICAL IMPLICATIONS

The CRHTT reduced hospital admissions for older people by 31%and carers preferred the service. Further research on crisisteams in older people with mental illness is needed using randomisedcontrolled methodology.

Introduction

TOP

Introduction

The psychiatric services of West Suffolk County cover a populationof approximately 280 000, of which 46 384 is over 65 yearsof age. There are five adult and five older age community mentalhealth teams. Current government policy has made compulsorythe introduction of crisis resolution and home treatment teams(CRHTT) for working-age adults (Department of Health 2000, 2001). The CRHTT in West Suffolk was launched in June 2005to cover working-age adults, namely those aged 17-65 yearsold. This is a practitioner-led service which provides short-termassessment and management at the time of a crisis as an alternative to hospital admission and/or facilitates early discharge fromhospital through intensive home treatment post-discharge. Todate the teams have reduced adult admissions to mental healthwards on average by 18%.

In the UK there are very few crisis teams covering older people’s mental health services (Cooper et al, 2007). An electronicdatabase search revealed only one study that has consideredthe use of an outreach team for older people on a waiting listfor hospital admission (Richman et al, 2003). The existingCRHTT in Suffolk was extended to cover older people from March 2006, in line with the West Suffolk Hospital National HealthService (NHS) Trust policy to provide equitable services forall age groups. This gave us a unique opportunity to examinethe impact of the crisis team on hospital admissions, lengthof stay and satisfaction with the service, in the period beforeand after the introduction of the CRHTT.

When the CRHTT was extended to cover service users over theage of 65, there were a number of other service changes, includingthe closure of a dementia care ward and two day hospitals,and introduction of an old age intermediate care team.

Method

TOP

Method

The study was approved by the Suffolk local research ethicscommittee. We compared the factors outlined above 6 monthsprior to the start of the crisis team (September 2005-March2006) and 6 months after its introduction (March 2006-September2006). All service users aged 65 years old or over who presentedin a crisis during this period were included in the study unless they were out-of-area. A ‘crisis’ was defined asan event where admission was being considered. If the individualpresented within 2 months of the original crisis presentationthen this was considered one and the same crisis, but if theypresented again after 2 months then this was counted as a newcrisis event. Prior to the introduction of the CRHTT every individual presenting in a crisis was admitted to hospital; the CRHTTonly provided a service to those who would otherwise have beenadmitted to hospital.

For every individual the following were ascertained: age, gender,marital status, place of residence, whether they lived alone,diagnosis, past psychiatric history and current Mini-MentalState Examination score (Folstein et al, 1975). We lookedat total number of admissions, number of compulsory admissions,average length of hospital stay and number of deaths over thestudy period.

After the CRHTT was introduced, we also collected the followingdata: total number of days of home treatment and whether anindividual needed to be admitted within 2 months of a crisisintervention.

Service user satisfaction was assessed by the Client Satisfaction Questionnaire (Larson et al, 1979), a validated 8-item self-reportquestionnaire using a 4-point scale (1=very dissatisfied, 4=verysatisfied). It has been applied in previous studies on crisisteams (Taachi et al, 2003; Johnson et al, 2005a, 2005b) andwe also adapted it for use in carers (available from authors).

The treating consultants were asked whether the service userhad capacity to complete the satisfaction questionnaire andwhether they had a suitable carer. Users were excluded if theywere detained under the Mental Health Act, if they lacked capacityor were admitted out-of-area; carers were excluded if theylacked capacity or if the person they cared for had died. Writtenconsent was obtained from the participants and the questionnairewas sent by post.

We analysed data using SPSS version 13 for Windows. Continuousdata were analysed using two-tailed t-tests and categoricaldata were compared using the chi-squared test.

Results

TOP

Results

Demographics
The baseline characteristics of the service user groups beforeand after the introduction of the crisis team are presentedin Table 1. There were no significant differences. In total,21 service users died (14%); 13 in the pre-CRHTT (22.6%) and8 in the post-CRHTT group (8.6%); all the deaths were attributedto natural causes.

View this table:
[in this window]
[in a new window]

Table 1. Baseline characteristics

Admissions and bed days
There was a significant reduction in admission after introductionof the crisis team (P<0.001). In the pre-CRHTT period therewere 65 crisis events which resulted in 65 admissions. Afterthe introduction of the CRHTT there were 102 crisis eventsof which only 70 required admissions. Of these, 66 crisis eventsled to direct hospital admission while four required admissionafter a brief period of home treatment. The crisis resolvedwith home treatment alone in 32 instances. Overall, the CRHTTreduced admissions by 31%. Nine individuals were detained underthe Mental Health Act in both the pre- and post-CRHTT periods.

There was no reduction in length of hospital stay or in bedusage according to functional and organic diagnostic groupsduring the post-CRHTT period (Table 2).

View this table:
[in this window]
[in a new window]

Table 2. Bed use

Of the 70 individuals admitted, 17 also had contact with theCRHTT either before admission or after discharge. The crisisteam treated people on average for 6.25 days pre-admissionand 19.71 days post-discharge. In the group (n=32) who receivedonly home treatment, 17 had depression, 8 dementia, 3 schizophreniaand 2 other psychotic illness. Twenty-three were referred inhours and 9 were referred out of hours. The mean number of days that the crisis team were involved for was 10.52 (s.d.=14.91).

Patient and carer satisfaction
Of the 143 service users only 59 had capacity to give consentto participate in the study. Of these, 28 returned their postalquestionnaires (response rate 47.45%). In the carer satisfactionsurvey, 39 out of 143 service users did not have any suitablecarer: 21 users died and so carers were not contacted, 15 usersdid not have a next of kin, 2 carers lacked capacity and for1 contact address could not be found. This left 104 potentiallyeligible carers of which 56 replied to our questionnaire (response rate 53.85%). There was no statistical difference in serviceuser or carer satisfaction between the pre- and post-CRHTTgroups, but there was a trend for greater satisfaction in carersin the post-CRHTT group (Client/Carer Satisfaction Questionnaire,maximum score 32 indicating most satisfaction, mean 25.38 v.25.51 for service users and carers respectively; Table 3).

View this table:
[in this window]
[in a new window]

Table 3. Satisfaction of service users and carers¹

Discussion

TOP

Discussion

Since the introduction of the crisis team in older people’smental health service there has been a 31% reduction in admissions,and service users and carers were in general satisfied withthe service. However, there was no difference in involuntaryadmissions. These findings are in keeping with studies conductedon working-age adult crisis teams (Johnson et al, 2005a, 2005b;Joy et al, 2006).

It may be argued that individuals who received home treatmentonly were below the ‘admission threshold’ and thereferrals to the CRHTT had been generated by the availabilityof this new service. Most of these individuals had depression,were referred in-hours and their episode settled quickly. Still,we could argue the CRHTT played an important role in preventingpossible future admissions of this subgroup of service usersby treating them early.

Although the average length of contact with the CRHTT was greater post-discharge (19.71 days) than pre-admission (6.25 days),the total length of hospital stay showed no reduction. Thismay suggest that the CRHTT was not fulfilling its other roleof enabling early discharge and may be working with users longerthan is necessary for an acute service. However, length of stay may be largely affected by factors out of the crisis team’scontrol. Discharge of in-patients on elderly wards is frequentlydelayed because of difficulties that social services have findingsuitable placements and arranging care packages. In addition,due to a reduction in the total number of in-patient beds,only severely unwell individuals who need to stay in hospitalfor longer are admitted.

Limitations
During the study period the local services were undergoing changeswhich may have had an impact on admission rates. However, theintermediate care team were not dealing with ‘requestsfor admissions’ per se. Another limitation would be seasonalvariation in the number and type of referrals, as the studyperiod covered different seasons of the year. Yet another limitationwas lack of randomisation. Nevertheless, there were no differencesbetween the study groups at baseline. The satisfaction surveydid not include referrers, which could be looked at in futurestudies. We also did not do a formal evaluation of the cost-effectivenessof the CRHTT. All these limitations need to be addressed infuture studies.

Acknowledgments

We thank the MHS Research and Training Fund charity for fundingthe administrative cost of the study, Manuel Regueira (CRHTTmanager), Jon Cooper (CRHTT community psychiatric nurse), DrDarley, Dr Chipperfield, Dr Riddell, Dr Ferrera (old age consultants),and in particular the service users and their carers who gavetheir time to help with this study.

References

TOP