In common with mental healthcare in general the provision of services for elderly people is shifting from in-patient treatment to the community (Reference Gatz and SmyerGatz & Smyer, 1992; Cambpell, 1994). There is evidence that functionally mentally ill, former long-stay elderly patients who are resettled into community provision do well (Reference Holloway, Ruther Ford and CarsonHolloway et al, 1994; Reference Anderson and TriemanAnderson & Treiman, 1995). However, less is known about the fate of those becoming ‘older people with enduring mental illness’ in contemporary community-oriented services (Abdul-Hamid et al, 1998). ‘Graduate’ is a term that describes people with severe psychiatric disorders who have graduated into old age (Reference Arie, Jolley, Levy and PostArie & Jolly, 1982).

The pragmatic approach to needs for care assessment has informed the development of a series of needs assessment tools (Wykes et al, Reference Wykes, Creer, Sturt and Wing1982; Reference Wykes, Sturt and Creer1985, Brewin et al, Reference Brewin, Wing and Mangen1987; Reference Brewin, Wing and Mangen1988; Reference Brugha, Wing and BrewinBrugha et al, 1988; Reference Abdul-HamidAbdul-Hamid, 1991; Reference Marshall, Hogg and LockwoodMarshall et al, 1995). The Camberwell Assessment of Need (CAN) was developed by Phelan et al (Reference Phelan, Slade and Thornicroft1995). It provides comprehensive needs assessment for adults and has been shown to be easy to use, valid and reliable. McWalter et al (Reference McWalter, Toner and McWalter1998) described the development and reliability and validity of the Care Needs Assessment Pack for Dementia (CARENAPD). There is, however, a lack of simple and effective tools to assess the needs of elderly people with a functional mental illness (older people with enduring mental illness). In the current study we report the development and application of such an assessment tool to a population of older people with enduring mental illness.

Method

The aims of this study were to: determine the reliability and validity of a novel needs assessment method, the Elderly Psychiatric Needs Schedule (EPNS); to undertake a point prevalence survey of older people with enduring mental illness in contact with specialist mental health services (both generic and elderly); and to undertake a needs assessment of this population.

The study was carried out in Norwood, which had a total population of 41740 and is an ethnically diverse and socially deprived part of inner London, of whom 5472 were over 65 years (1991 census). The study population included all people at or over the age of 60 with a functional mental illness who were in contact with the catchment area psychiatric services (both old age and general adult services) and who had an onset of illness before the age of 60.

Along with people who were over 65 years of age, a pre-elderly group of people who were 60-65 years of age were included since these ‘pre-elderly group’ people are likely to become older people with enduring mental illness within 5 years. Hospital medical notes of more than 300 Norwood patients who were at or over the age of 60 were reviewed and 77 older people with enduring mental illness were identified. The definition of older people with enduring mental illness used was that of Arie & Jolley (Reference Arie, Jolley, Levy and Post1982): those who have functional psychiatric disorder before the age of 60 and continue to be in contact with psychiatric services. To make the survey more representative, this included both those under the community team and those seen at the out-patient department only. A recent study showed that these two groups of patients could need different levels of intervention provided between ‘complex’ and ‘routine’ patients (Reference Mehta, Salvaji and AbedMehta et al, 2007). Of the 77 people identified, 75 were interviewed (2 were untraceable) and their psychiatric keyworker was also interviewed separately.

Results

Table 1 presents basic demographic features of the sample. The ethnic origin reflected the demographic structure of the elderly population of Norwood. In contrast with samples of younger people with severe mental illness, only a minority (34%) were single and 35% were married or cohabiting. The majority of participants had a case-note diagnosis of a psychotic illness (schizophrenia 41%, schizoaffective psychosis 5% and bipolar affective disorder 7%). Forty-one per cent of the sample had a diagnosis of depression without psychotic features.

Table 1.

Demographic and clinical characteristics of Norwood's older people with enduring mental illness

Characteristic	Participants (n=77)
Age group: years, n (%)
60–65	20 (26)
65+	57 (74)
Gender, n (%)
Male	30 (39)
Female	47 (61)
Marital status, n (%)
Single	26 (34)
Married	23 (30)
Divorced/separated	15 (19)
Widowed	9 (12)
Cohabitating	4 (5)
Ethnic origin, n (%)
White	69 (89)
Caribbean	4 (5)
African	1 (1)
Asian	3 (4)
Diagnosis, n (%)
Schizophrenia	32 (41)
Bipolar disorder	5 (7)
Schizoaffective disorder	4 (5)
Depression	32 (41)
Psychopathology scores, mean (s.d.)
Brief Psychiatric Rating Scale score	35.5 (10.7)
Clifton Assessment Procedures for the Elderly score	6.5 (6.5)
Mini-Mental State Examination score	26.5 (4.4)
Geriatric Depression Scale scorea	5.1 (4.0)

The EPNS was used to assess the needs of Norwood's older people with enduring mental illness. This schedule covers 20 items that constitute the different needs of elderly people with psychiatric illness. The global scoring of each item identifies the need as either non-existent (no need), unmet because of lack of provision of service, unmet because of a patient's refusal to utilise services, partially met need when the service used was not effective in meeting the need, or need met when an effective intervention was used to alleviate the need (Table 2).

Table 2.

The psychiatric needs of Norwood's older people with enduring mental illness

Item	No need, n (%)	Unmet need, n (%)	Patient refused, n (%)	Partially met need, n (%)	Met need, n (%)
1. Psychiatric assessment and treatment needs	0	0	7 (10.6)	16 (24.2)	43 (65.2)
2. Dementia assessment and management needs	60 (90.9)	2 (3)	1 (1.5)	2 (3)	1 (1.5)
3. Management of medication	42 (63.6)	0	6 (9.1)	2 (3)	16 (24.2)
4. Substance misuse services needs	60 (90.9)	1 (1.5)	4 (6.1)	0	1 (1.5)
5. Physical health needs	36 (54.5)	1 (1.5)	1 (1.5)	13 (19.7)	15 (22.7)
6. Communication needs	63 (95.5)	1 (1.5)	0	2 (3)	0
7. Mobility needs	42 (63.6)	3 (4.5)	2 (3)	11 (16.7)	8 (12.1)
8. Incontinence service needs	55 (83.3)	3 (4.5)	2 (3)	0	6 (9.1)
9. Self-care needs	40 (60.6)	1 (1.5)	2 (3)	8 (12.1)	15 (22.7)
10. Domestic management needs	37 (56.1)	3 (4.5)	2 (3)	5 (7.6)	19 (28.8)
11. Food and eating needs	39 (59.1)	0	1 (1.5)	3 (4.5)	23 (34.8)
12. Safety and protection needs	50 (75.8)	3 (4.5)	1 (1.5)	5 (7.6)	7 (10.6)
13. Social activity needs	33 (50)	5 (7.6)	18 (27.3)	4 (6.1)	6 (9.1)
14. Day-care needs	32 (48.5)	5 (7.6)	18 (27.3)	5 (7.6)	9 (13.6)
15. Money and benefits needs	47 (71.2)	0	1 (1.5)	2 (3)	16 (24.2)
16. Transport needs	44 (66.7)	6 (9.1)	1 (1.5)	3 (4.5)	12 (18.2)
17. Housing needs	48 (72.7)	3 (4.5)	4 (6.1)	3 (4.5)	8 (12.1)
18. Ethnic and cultural needs	62 (93.9)	1 (1.5)	0	0	3 (4.5)
19. Management of patient's affairs	54 (81.8)	0	1 (1.5)	0	11 (16.7)
20. Other needs	61 (90.9)	1 (1.5)	2 (3)	1 (1.5)	1 (1.5)

The reliability the EPNS

Two interviewers assessed 20 of the study participants independently using the EPNS. The percentage agreement and Kappa (κ) scores were calculated for each item (Table 3). The percentage agreement varied between 73% in the food and eating needs to 100% on ten of the items. The mean percentage agreement was 96%. Kappa varied from κ=0.41 in food and eating needs to κ=1 on ten of the items; mean κ=0.90. The internal consistency of the items of the scale using Cronbach's alpha reliability test has revealed a value of 0.84, which indicates that the scale has a good internal consistency and justifies the summing across items.

Table 3.

Reliability of the Elderly Psychiatric Needs Schedule

Item	Agreement, %	κ
1. Psychiatric assessment and treatment needs	99	0.95
2. Dementia assessment and management needs	100	1.00
3. Management of medication	100	1.00
4. Substance misuse services needs	99	0.78
5. Physical health needs	100	1.00
6. Communication needs	100	1.00
7. Mobility needs	96	0.89
8. Incontinence service needs	100	1.00
9. Self-care needs	98	0.94
10. Domestic management needs.	97	0.94
11. Food and eating needs.	73	0.41
12. Safety and protection needs	100	1.00
13. Social activity needs	100	1.00
14. Day-care needs	99	0.97
15. Money and benefits needs	85	0.65
16. Transport needs	100	1.00
17. Housing needs	98	0.95
18. Ethnic and cultural needs	100	1.00
19. Management of patient's affairs	88	0.56
20. Other needs	100	1.00

Discussion

The literature on older people with enduring mental illness raises more questions than it answers (Reference Abdul-Hamid and SilvermanAbdul-Hamid & Silverman, 1998). In this era of community care there is still very little information on the numbers of these people in the community, their problems, needs or the service provision already in place to satisfy their needs.

It is known that a significant number of older people with enduring mental illness who were placed in the community are living in nursing homes. There are conflicting reports about the suitability of these homes as places to meet the needs of these people (Reference Dencker and GottfriesDencker & Gottfries, 1991; Reference Holloway, Ruther Ford and CarsonHolloway et al, 1994). We do not know how much psychiatric input there is into these homes and how far the services are involved in the assessment, monitoring, rehabilitation, treatment and follow-up of these individuals.

There is clearly a need for studies to locate and quantify the elderly graduate population. This study has aimed to assess the psychiatric and medical needs of older people with enduring mental illness using a needs schedule derived from previous needs tools but specifically developed for this purpose. The first aim of this study was to assess the reliability and validity of this scale. The interrater reliability was generally very good and on 10 out of the 20 items the agreement was complete. The two raters (W.A-H. and K.L-C.) undertook training on the assessment and a manual was developed (the EPNS and manual are available from the first author). Construct validity of the EPNS was assessed by comparing it with a variety of well-known assessment tools used in the elderly population. The total needs score of the EPNS correlated highly with the Clifton Assessment Procedures for the Elderly (CAPE; Reference Pattie and GilleardPattie & Gilleard, 1979) total score, which is a measure of overall disability. It correlated less well with the Brief Psychiatric Rating Scale (BPRS; Reference Ventura, Green and ShanterVentura et al, 1993), which is a measure of psychiatric symptoms, and the Mini-Mental State Examination (MMSE; Reference Folstein, Folstein and McHughFolstein et al, 1975), which is a measure of cognitive state. The total score of unmet needs showed a similar pattern. The unmet needs scores showed the highest correlation with the BPRS, which suggests that people who are not receiving the full range of services are exhibiting more symptoms.

It is important to note that one of the study limitations is the lack of control for different diagnoses. As recent work has shown that the level of intervention provided could differ between ‘complex’ and ‘routine’ patients (Reference Mehta, Salvaji and AbedMehta et al, 2007), a larger study might be needed to control for diagnosis. The findings on the prevalence of need identifies the range of problems that older people with enduring mental illness present to community mental health services, which differs significantly from their younger counterparts. Unsurprisingly, all participants had needs for psychiatric treatment and care. Although the mean MMSE score for the sample was 26.5, clinically significant dementia was rated as being uncommon in the sample that had an EPNS rating (9%). However, overall, more than 40% of the sample had needs in the domains of day care (54%), social activity (50%), physical health (46%), domestic management (44%) and food (41%). A small but significant proportion had a need for incontinence services (17%).

Encouragingly, none of the psychiatric and treatment needs of the older people with enduring mental illness were unmet due to lack of service provision. One may say that this is expected in a sample that is chosen from people in contact with the service. However, 11% had unmet needs due to refusal and 24% had only partly met psychiatric and assessment needs. The service should develop ways of reaching and meeting the needs of these individuals.

The main unmet needs due to lack of service provision were in the domains of transport, social activities and day care, mobility needs, incontinence service needs, self-care needs, housing needs and dementia assessment needs. That the unmet needs for dementia assessment were higher than the met needs (though small in numbers) highlights the problem of underestimation of dementia in the elderly graduate population (Reference Harvey, Davidson and PowchikHarvey et al, 1992; Reference Anderson and TriemanAnderson & Trieman, 1995).

Refusal was, in general, more common than unmet need because of a lack of service or failure to carry out an assessment. The main unmet needs identified in the study that resulted from a patient's refusal were social activity needs, day-care needs, management of medication, substance misuse service needs and housing needs. Clearly these are important aspects of the care of older people with enduring mental illness and the services should address the reasons for refusal and seek ways of making services more acceptable.

Declaration of interest

None. Funding detailed in Acknowledgements.