Multicomponent Geriatric Intervention for Elderly Inpatients With Delirium: Effects on Costs and Health-Related Quality of Life

Journal of Gerontology: MEDICAL SCIENCES 2008, Vol. 63A, No. 1, 56 61 Copyright 2008 by The Gerontological Society of America Multicomponent Geriatric Intervention for Elderly Inpatients With Delirium: Effects on Costs and Health-Related Quality of Life Kaisu H. Pitkala, 1,2 Jouko V. Laurila, 3 Timo E. Strandberg, 4 Hannu Kautiainen, 5 Harri Sintonen, 6,7 and Reijo S. Tilvis 3 1 Unit of General Practice and 3 Clinics of Internal Medicine and Geriatrics, Helsinki University Hospital, Finland. 2 Department of General Practice and Primary Health Care, and 6 Department of Public Health, University of Helsinki, Finland. 4 Department of Public Health Science and General Practice, University of Oulu, and Oulu University Hospital, Unit of General Practice, Finland. 5 Rheumatism Foundation Hospital, Heinola, Finland. 7 Finnish Office for Health Technology Assessment (FinOHTA/Stakes), Helsinki. Background. The detrimental effects of delirium on functioning and mortality are well known, but health-related quality of life (HRQoL) and costs of care have rarely been investigated among patients with delirium. We studied the effects of multicomponent geriatric treatment on costs of care and HRQoL in delirious inpatients. Methods. A randomized, controlled trial of 174 inpatients with delirium was performed in an acute geriatric hospital. The intervention was individually tailored geriatric treatment. The HRQoL was measured by the 15D instrument and subjective health by a four-level ordinal scale. Health care costs including intervention costs were calculated for 1 year after the delirium episode. Results. Mean age of the patients was 83 years; 31% had prior dementia. After the index hospitalization for delirium, a greater proportion in the intervention group than in the control group stated that they felt healthy (71% vs 49%, p ¼.050). HRQoL deteriorated in both groups as a consequence of delirium. Deterioration was, however, slower in the intervention group ( 0.026, 95% confidence interval [CI], 0.051 to 0.001) than in the control group ( 0.065, 95% CI, 0.09 to 0.040; p ¼.034). Counting all costs of hospital care, long-term care, skilled home nursing visits, and costs related to intervention, the intervention group used, on average, 19,737 e during the follow-up year, whereas the respective figure for the control group was 19,557 e. The difference between the groups was nonsignificant (180 e [95% CI, 5,006 to 5,064 e]). Conclusions. Comprehensive geriatric intervention improved HRQoL without increasing overall costs of care. Key Words: Delirium Multicomponent geriatric treatment Quality of life Cost of care. DURING their acute care, about 25% of older medical inpatients suffer from delirium (1). Delirium increases mortality (2,3) and the need for permanent institutional care (3 6). It also leads to poor functional outcome (6,7) and dementia (2,8). However, to our knowledge, only few studies have investigated quality of life (QoL) among patients with delirium. To date, only one uncontrolled study has reported a deteriorating health-related QoL (HRQoL) among delirious hip-fracture patients (9). The increased need for hospital and institutional days related to delirium produces substantial costs for health and social care (10). It has been estimated in the United States that the annual inpatient costs alone due to delirium and its complications are $4 billion (11). One study showed that the short-term costs for delirious patients in intensive care units are much higher than the costs for most other patient groups (12). In particular, patients with delirium superimposed on dementia have the highest costs for care (13). One study demonstrated that long-term nursing-home costs for delirium are substantial and that they represent a potential primary source of savings in delirium prevention (14). However, there are few studies that follow-up delirious patients long-term and investigate rigorously their subsequent health care costs. Moreover, only a few randomized studies have investigated the effects of comprehensive geriatric care on delirium (15 17). We have previously studied, in a randomized controlled trial, the effects of comprehensive geriatric assessment and treatment of delirium according to current guidelines (18). This intervention led to faster alleviation of delirium and slower deterioration in cognition as compared to the control group. However, no impact on mortality or on admission to long-term institutional care was observed (17). In the present report we assessed the effects of this same intervention on HRQoL and costs for health care during a 1-year follow-up. METHODS The details of participant recruitment and delirium screening have been described in detail (17). Potential participants were consecutive patients (aged. 69 years) admitted to the general medicine service at one Helsinki city hospital. 56

INTERVENTION ON DELIRIUM: EFFECTS ON COSTS AND HRQoL 57 Exclusion criteria included refusal, life expectancy of, 6 months (e.g., metastatic cancer, severe stroke), inability to obtain informed consent within 2 working days, or admission from permanent institutional care to the hospital. Thus, all patients were home-dwelling at baseline. Because all participants were delirious, informed consent was obtained from each patient s closest proxy (100%). The study was approved by Helsinki University Hospital and the Helsinki City ethics committees. Patients were screened within 2 working days of their admission by administration of the Confusion Assessment Method (CAM) test (19), Mini-Mental State Examination (MMSE) (20), and Digit span (21), and by a proxy interview and a review of medical records. Among participants with a positive CAM test, delirium was further confirmed by the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (22). Patients were randomly allocated by means of computer-generated random numbers to undergo the intervention (n ¼ 87) or to receive the usual care in the same hospital (n ¼ 87). The study nurse assessed the HRQoL of all participants at baseline and at hospital discharge by using the 15D questionnaire (23). The 15D is a generic 15-dimensional measure developed according to the feedback of experts and patients (24) and has been validated systematically internationally since the 1970s in various population and patient samples. 15D correlates well with other HRQoL measures such as SF-36 (RAND-36), EuroQol 5, and the Nottingham Health Profile (25,26). The 15D instrument may be used not only as a profile measure but also as a single index, which is a specific strength of the 15D for this type of study. The index varies between 0 (poorest HRQoL) and 1 (excellent HRQoL). Dimensions of the 15D are mobility, vision, hearing, breathing, sleeping, eating, speech, elimination, usual activities, mental function, discomfort and symptoms, depression, distress, vitality, and sexual activity. Usually the 15D questionnaire is filled in by the individual whom it concerns, but it may also be administered in an interview with the participant or his/her proxy. The 15D shows very good discriminant validity among various aged populations (27). It has sensitivity to change after a health care intervention (28), and also shows prognostic validity (27). Answers to the 15D questionnaire came from the patient or a proxy interview. When using the 15D, we excluded the question concerning sexual activity, because a large proportion of delirious patients ignored that question. Subjective health has been considered an important dimension of HRQoL (29 31). In our study, subjective health was inquired about at discharge with a four-level ordinal scale and dichotomized as good (feeling healthy or quite healthy) or poor (feeling quite unhealthy or unhealthy). Because of their delirious state, we were unable to acquire this data from all patients at baseline. The patients medical records were carefully reviewed to obtain data on medical comorbidities. Comorbidity was assessed by the Charlson comorbidity index, a weighted index taking into account the number and severity of comorbid conditions (32). Each patient s premorbid dementia status was based on information from interviews of proxies, the Clinical Dementia Rating (CDR) scale (33), DSM-IV criteria of dementia (22), and reviews of medical records confirming whether these patients had undergone full assessment for diagnosis of dementia. The Barthel index was used as described (34). Intervention Our previous article (17) describes all details of the intervention, which was based on good detection of delirium and included a comprehensive geriatric assessment at baseline as well as careful diagnosis of the underlying etiological conditions behind the delirium. Atypical antipsychotics were used if necessary, and effective general treatments were implemented for all patients. After the acute phase of delirium, all patients not recovering from impaired cognition (MMSE, 24) in the intervention group underwent detailed diagnostics of dementia and thereafter received acetylcholinesterase inhibitors (AChEI). Outcome Measures The use and costs of health services were measured in detail for 1 year after the index hospitalization for delirium. All days spent in institutions (various hospitals, nursing homes, or long-term care hospitals) and visits to specialists during the 1-year follow-up were retrieved from patients medical records in all their area hospitals and from social care registers. The frequency of skilled home nursing visits was determined from participants, their caregivers, and home nurses at 3 months after discharge from the hospital, and their total number was extrapolated for the whole time period spent at home. These resources were valued at their average unit costs in Finland in 2001 (35). We also included the use of atypical antipsychotics, AChEIs, vitamin D calcium supplements, nutritional supplements, and hip protectors as parts of the intervention and valued them at their retail prices in pharmacies, or at wholesale prices for hospitals. Costs were expressed in the European monetary unit euro (e, corresponding to USD 1.33 in March 2007). Mortality data were determined from medical records and central registers. Statistical Methods The results were expressed as means with standard deviations (SD) and 95% confidence intervals (CI). As the data for costs were highly skewed, bias-corrected and accelerated bootstrap estimation was used to derive 95% CIs (5000 replications). The statistical significance between groups was evaluated by Mann Whitney test, chi-square, or Fischer s exact test. Analysis of covariance (ANCOVA) was also used to compare the changes in HRQoL between the groups as a consequence of the intervention. In these analyses, baseline values were used as covariates. The normality of variables was evaluated by the Shapiro Wilk W test. Median regression analysis was used to model the relationship between the health care costs during the subsequent year and of predictor variables. Analysis was performed according to the intention-to-treat principle with the last observation carried forward (LOCF).

58 PITKALA ET AL. Table 1. Sample Characteristics Variable Intervention (N ¼ 87) Control (N ¼ 87) p Value At baseline Mean age, (SD) 84 (5.6) 83 (6.2).53* Females, (%) 76 71.49 y Prior dementia, (%) 26 35.25 y Charlson comorbidity index, mean (SD) 2.6 (2.0) 2.3 (1.9).27* Physical functioning: Barthel index, (mean (SD) 79.0 (20.9) 79.1 (18.4).63* At 1 y Admitted to permanent institutional care, (%) 43 52.22 y Deceased, (%) 35 30.52 y Notes: *Difference between the tested by t test. y Differences between the groups test with chi-square test. SD ¼ standard deviation. RESULTS Patients were old and frail with a mean age of 83 years, and 31% had prior dementia (Table 1). During the follow-up year, 42.5% of the intervention patients and 51.7% of control patients were admitted to permanent institutional care (p ¼.22). Respective mortality rates were 35% and 30% (p ¼.52). At discharge from the index hospitalization for delirium, a greater proportion in the intervention group than in the control group felt healthy or quite healthy (71% vs 49%, p ¼.050). At the same time, HRQoL was significantly higher in the intervention than in the control group (0.68 [SD 0.12] vs 0.62 [SD 0.15], p ¼.020). Although the HRQoL deteriorated in both groups as a consequence of delirium, deterioration was slower in the intervention ( 0.026, 95% CI, 0.051 to 0.001) than in the control group ( 0.065, 95% CI, 0.09 to 0.040). Using the baseline measurements as covariates, the difference in changes between the groups was significant (ANCOVA, Figure 1. Changes with 95% confidence intervals in various dimensions of health-related quality of life (HRQoL) in the intervention and control groups from baseline to discharge. p ¼.034). Among those patients having both HRQoL assessments available, the median time between the baseline assessment and the discharge assessment was 18 days in the intervention group (N ¼ 79) and 16 days in the control group (N ¼ 77) (p ¼.58). The significant differences between the groups in HRQoL appeared in several dimensions: mental function corresponding to cognition and alertness, usual activities corresponding to functioning in activities of daily living, vitality, depression, and speech (Figure 1). Table 2 and Figure 2 indicate how the intervention and the control groups used various health services during the 1- year follow-up. Counting all costs of care, the intervention group used a mean 19,737 e per patient during the follow-up year, whereas the respective figure for the control group was 19,557 e. The difference between the groups was not significant (180 e, 95% CI, 5006 to 5064 e). The total cost for the whole intervention group was 1,722,763 e, and the total cost for the control group was 1,702,258 e. The extra costs of intervention including atypical antipsychotics, AChEIs, vitamin D calcium supplements, hip protectors, Service Table 2. Use of Health Services During the Follow-Up Year Price per Unit (Euros)* Use in Units Intervention Group (N ¼ 87) Control Group (N ¼ 87) Days in primary hospitals 176.90 Initial admission days during delirium 2550 1951 Readmission days during the follow-up year 1837 1487 Days at specialist hospitals 285.00 176 240 Specialist consultations, units 147.30 43 25 Days spent in psychiatric hospitals 247.70 78 37 Days spent in nursing homes 115.00 3864 5127 Days spent in long-term care hospitals 113.50 2507 3363 Skilled home nursing, visits 46.00 2259 831 Atypical antipsychotics, daily doses 1.0 1898 794 Use of AChEIs, daily doses 3.5 7605 462 Vitamin D calcium supplements, daily doses 0.15 2158 275 Hip protectors 40 30 1 Nutritional supplements, daily doses 1.0 2444 0 Notes: *Hujanen (35). AchEIs ¼ acetylcholinesterase inhibitors.

INTERVENTION ON DELIRIUM: EFFECTS ON COSTS AND HRQoL 59 Table 3. Median Regression Model for the Total Health Care Costs During the Subsequent 1-Year Follow-Up Variable Coefficient (95% CI) p Value Age 302 ( 416 to 1021).41 Sex (male) 2300 ( 11,490 to 6889).62 Change in HRQoL* 37,410 ( 70,941 to 3880).029 Control group 3654 ( 11,620 to 4312).37 Constant 3248 Notes: *During hospitalization of delirium. CI ¼ confidence interval; HRQoL ¼ health-related quality of life. Figure 2. Costs (with 95% confidence intervals) of various health services used by the intervention and control groups during the 12-month follow-up. Other: costs of atypical antipsychotics, acetylcholinesterase inhibitors, vitamin D calcium supplements, nutritional supplements, and hip protectors. and nutritional supplements accounted for only 2.3% (38,789 e) of the total cost in the intervention group. There was substantial variation between individuals both in costs of care (from 373 to 54,400 e) and in HRQoL changes ( 0.3905 to 0.213), which occurred during the hospitalization for delirium. In-hospital changes in the HRQoL score independently predicted total health care costs during the subsequent 1-year follow-up (Table 3). DISCUSSION Comprehensive, multicomponent geriatric treatment led to better subjective health as well as better HRQoL for survivors at discharge from the hospital. We reported earlier that the delirium also resolved faster, and cognition was improved by the intervention. The mean health care costs of one patient with delirium are high, nearly 20,000 e (corresponding to USD 26,600) annually. However, the favorable outcomes reported here were achieved without increasing the costs of care. To our knowledge, this is one of the first studies investigating HRQoL with a validated measure among patients with delirium. Our results show that delirium seriously affects HRQoL, and following a delirium episode the trend is toward decreasing HRQoL among survivors. A difference of 0.02 0.03 in the 15D score between patient groups, or as an effect of health care intervention, has been considered clinically significant (23). Both groups showed a decreasing trend in the HRQoL score, but the intervention group showed a slower decrease ( 0.026 in the invention group vs 0.065 in the control group). We have previously measured the 15D scores in various samples of elderly patients. The mean score for delirious patients (0.694) falls between that of the population-based home-dwelling sample (age 80 100 years) (mean 15D index 0.796) and that of long-term care patients (mean 15D score 0.580) (24). However, it must be emphasized that our participants score is not directly comparable to these previous scores, because we omitted the question of sexual activity, which was left unanswered by the delirious patients. Previous studies have pointed out that it is difficult to measure HRQoL among very old and frail individuals. For example, the well-validated measure SF-36 may be unsuitable for frail elderly persons because a large proportion of them were unable to complete the questionnaire (36). The construct validity of 15D has not been especially tested among patients with delirium or dementia. However, 15D shows good discriminative and prognostic validity among patients with delirium or dementia (27). The favorable outcome of the intervention in terms of HRQoL is supported by the fact that subjective health also showed favorable differences between these groups. Subjective health is considered an important dimension of HRQoL (29 31,37), which is not included in the 15D. These findings are also in line with our earlier finding that the intervention had positive effects in shortening the delirium and in improving cognition (17). The dimensions of HRQoL showing differences favoring the intervention group were mental function, usual activities, vitality, depression, and speech. These findings suggest that the improvements in well-being in the intervention group were not solely due to improvements in cognitive functioning. This multidimensionality of response supports the notion that our intervention and control groups had true differences in HRQoL. There was no difference between the groups in the costs of care, which suggests that improved cognition, faster alleviation of delirium, and better HRQoL may be achieved without increasing total costs of care. Actually, the extra costs produced by our intervention were very low, only 2.3% of the total costs of care. Among frail elderly patients, such as individuals suffering from delirium, HRQoL and cognition may be far more significant aims for care than merely postponement of death. It seems that our intervention succeeded in these aims. In intervention studies of elderly patients, counting total costs of care is still quite rare even among patients with dementia (38). The high cost of delirious patients is due to the fact that almost half (47%) of these patients needs permanent institutional care. Even if they stay at home, they need many communal services due to their decreased physical functioning and cognition. The range in costs of care, as well as the changes in HRQoL achieved, among patients with delirium is very wide. Annual costs of care varied between 373 and 54,400 e (USD 496 72,352), whereas the changes in the HRQoL scores among patients discharged from the hospital ranged from 0.3905 to 0.213. When adjusted for age, sex, and treatment group, the HRQoL only weakly predicted follow-

60 PITKALA ET AL. up costs. Those patients achieving a higher HRQoL used somewhat less health care during their follow-up. A limitation of this study is that we valued the resource use measured in physical units at average unit costs. Therefore, the costs reflect the differences between the groups in the use of physical units alone, not the possible differences between the groups in the intensity of resource use within a physical unit, say within a hospital day. However, during the index hospital stay for delirium, all extra costs related to intervention as well as specialist consultations, which are usually included in the average unit costs, are counted here as extra costs of the intervention group. Another limitation is that the costs of social services could not be calculated because the use of these services could not be reliably retrieved from official records. Neither could we reliably trace all the primary care doctor office visits, because the patients could have potentially used so many health centers and private doctors. However, these costs of physician visits probably account for a minority of the total costs: An elderly patient in Finland visits a doctor on average 2 3 times a year, and, for example, patients with dementia 5 times a year. There were on average five more patients in the intervention group at home any time during the follow-up year than in the control group. This equates to a total of 25 extra doctor visits in the intervention group. The costs for these visits would be, 30 e/patient/year accounting for, 0.2% of the total health care costs of these patients. The primary hypothesis of this study was to postpone institutional care; therefore, the sample size was originally calculated to measure the difference between the groups in the proportions of patients deceased or institutionalized (17). The study was not powered to detect significant differences in costs of care. Besides, what constitutes a clinically important difference in costs is hard to define. Conclusion Our multicomponent geriatric intervention showed favorable outcomes in terms of HRQoL as compared to control treatment. In addition, the intervention improved cognition, and delirium symptoms resolved more rapidly. All this was achieved without increasing health care costs. ACKNOWLEDGMENT This trial was supported by the Lions Organization (Punainen Sulka Red Feather), Helsinki University Central Hospital, Helsinki City, and the Academy of Finland (Grant 48613). CORRESPONDENCE Address correspondence to Kaisu H. Pitkälä, MD, PhD, University of Helsinki, Department of General Practice and Primary Health Care, PO Box 41, 00014 University of Helsinki, Finland. E-mail: kaisu. pitkala@helsinki.fi REFERENCES 1. Bucht G, Gustafson Y, Sandberg O. Epidemiology of delirium. Dement Geriatr Cogn Disord. 1999;10:315 318. 2. McCusker J, Cole M, Abrahamowicz M, Primeau F, Belzile E. Delirium predicts 12-month mortality. Arch Intern Med. 2002;162:457 463. 3. Pitkala KH, Laurila JV, Strandberg TE, Tilvis RS. Prognostic significance of delirium in frail older people. Dement Geriatr Cogn Disord. 2005;19:158 163. 4. Francis J, Martin D, Kapoor WN. A prospective study of delirium in hospitalized elderly. JAMA. 1990;263:1097 1101. 5. Levkoff SE, Evans DA, Liptzin B, et al. Delirium. The occurrence and persistence of symptoms among elderly hospitalized patients. Arch Intern Med. 1992;152:334 340. 6. O Keeffe S, Lavan J. The prognostic significance of delirium in older hospital patients. J Am Geriatr Soc. 1997;45:174 178. 7. McCusker J, Cole M, Dendukuri N, Belzile E, Primeau F. Delirium in older medical inpatients and subsequent cognitive and functional status: a prospective study. CMAJ. 2001;165:575 583. 8. Rahkonen T, Luukkainen-Markkula R, Paanila S, Sivenius J, Sulkava R. Delirium episode as a sign of undetected dementia among community dwelling elderly subjects: a 2 year follow up study. J Neurol Neurosurg Psychiatry. 2000;69:519 521. 9. Duppils GS, Wikblad K. Cognitive function and health-related quality of life after delirium in connection with hip surgery. A six-month follow-up. Orthop Nurs. 2004;23:195 203. 10. Inouye SK, Bogardus ST Jr, Charpentier PA, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med. 1999;340:669 676. 11. Inouye SK. The dilemma of delirium: clinical and research controversies regarding diagnosis and evaluation of delirium in hospitalized elderly medical patients. Am J Med. 1994;97:278 288. 12. Milbrandt EB, Deppen S, Harrison PL, et al. Costs associated with delirium in mechanically ventilated patients. Crit Care Med. 2004;32: 955 962. 13. Fick DM, Kolanowski AM, Waller JL, Inouye SK. Delirium superimposed on dementia in a community-dwelling managed care population: a 3-year retrospective study of occurrence, costs, and utilization. J Gerontol A Biol Sci Med Sci. 2005;60A:748 753. 14. Leslie DL, Zhang Y, Bogardus ST, Holford TR, Leo-Summers LS, Inouye SK. Consequences of preventing delirium in hospitalized older adults on nursing home costs. J Am Geriatr Soc. 2005;53: 405 409. 15. Cole MG, Primeau FJ, Bailey RF, et al. Systematic intervention for elderly inpatients with delirium: a randomized trial. CMAJ. 1994;151: 965 970. 16. Cole MG, McCusker J, Bellavance F, et al. Systematic detection and multidisciplinary care of delirium in older medical inpatients: a randomized trial. CMAJ. 2002;167:753 759. 17. Pitkala KH, Laurila JV, Strandberg TE, Tilvis RS. Multicomponent geriatric intervention for elderly inpatients with delirium: a randomized, controlled trial. J Gerontol A Biol Sci Med Sci. 2006;61A:176 181. 18. Young LJ, George J. Do guidelines improve the process and outcomes of care in delirium? Age Ageing. 2003;32:525 528. 19. Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: the confusion assessment method. A new method for detection of delirium. Ann Intern Med. 1990;113:941 948. 20. Folstein MF, Folstein SE, McHugh PR. Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189 198. 21. Wechsler D. The Bellevue intelligence tests. In: Wechsler D. The Measurement of Adult Intelligence. Baltimore: Williams & Wilkins; 1944. 22. American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders. 4th Ed. (DSM-IV). Washington, DC: APA; 1994. 23. Sintonen H. The 15D instrument of health-related quality of life: properties and applications. Ann Med. 2001;33:328 336. 24. Sintonen H. The 15-D Measure of Health Related Quality of Life: Reliability, Validity and Sensitivity of Its Health State Descriptive System. Working Paper 41. National Centre for Health Program Evaluation. Melbourne, 1994. Available at: http://www.buseco.monash. edu.au/centres/che/pubs/wp41.pdf 25. Hawthorne G, Richardson J, Day NA. A comparison of the Assessment of Quality of Life (AQoL) with four other generic utility instruments. Ann Med. 2001;33:358 370. 26. Stavem K, Froland SS, Hellum KB. Comparison of preference-based utilities of the 15D, EQ-5D and SF-6D in patients with HIV/AIDS. Qual Life Res. 2005;14:971 980.

INTERVENTION ON DELIRIUM: EFFECTS ON COSTS AND HRQoL 61 27. Strandberg T, Pitkälä K, Sintonen H, Huusko T, Kautiainen H, Tilvis R. Usability, discriminant and prognostic validity of 15D instrument for health related quality of life in older population samples. In: Huusko T, Strandberg T, Pitkala K, eds. Can Older People s Quality of Life Be Measured? (in Finnish). Helsinki: The Central Union for the Welfare of the Aged; 2006:42 61. 28. Kattainen E, Sintonen H, Kettunen R, Merilainen P. Health-related quality of life of coronary artery bypass grafting and percutaneous transluminal coronary artery angioplasty patients: 1-year follow-up. Int J Technol Assess Health Care. 2005;21:172 179. 29. Frytak JR. Assessment of quality of life in older adults. In: Kane RL, Kane RA. Assessing Older Persons. Measures, Meaning and Practical Applications. New York: Oxford University Press; 2000:200 236. 30. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30:473 483. 31. WHO. WHOQOL-BREF. Introduction. Administration, scoring and generic version of the assessment. Field Trial Version December 1996. In: Rapley M. Quality of Life Research. A Critical Introduction. London: Sage Publications; 2003. 32. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373 383. 33. Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A new clinical scale for the staging of dementia. Br J Psychiatry. 1982;140: 566 572. 34. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J. 1965;14:61 65. 35. Hujanen T. The Health Care Unit Costs in Finland at 2001. Aiheita 1/2003. (In Finnish). Helsinki: STAKES National Research and Development Centre for Welfare and Health; 2003. 36. Hayes V, Morris J, Wolfe C, Morgan M. The SF-36 health survey questionnaire: is it suitable for use with older adults? Age Ageing. 1995;24:120 125. 37. Cummins RA. Assessing quality of life. In: Brown RI, ed. Quality of Life for People With Disabilities. Models, Research and Practice. Padstow, Australia: Stanley Thornes Ltd.; 1997:116 150. 38. Wimo A, Winblad B. Economic aspects on drug therapy of dementia. Curr Pharm Des. 2004;10:295 301. Received January 27, 2007 Accepted May 7, 2007 Decision Editor: Luigi Ferrucci, MD, PhD