HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 63:275-282 (2008)
© 2008 The Gerontological Society of America

SPECIAL SECTION

Multidimensional Geriatric Assessment: Back to the Future Cost-Effectiveness of a Multidisciplinary Intervention Model for Community-Dwelling Frail Older People

René J. F. Melis, Eddy Adang, Steven Teerenstra, Monique I. J. van Eijken, Anders Wimo, Theo van Achterberg, Eloy H. van de Lisdonk and Marcel G. M. Olde Rikkert

Departments of 1 Geriatric Medicine, 2 Epidemiology, Biostatistics, and Health Technology Assessment, 3 Centre for Quality of Care Research, and5 Department of General Practice, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands.
4 Neurotec, Division of Geriatric Epidemiology, Karolinska Institutet, Stockholm, Sweden.

Address correspondence to René J. F. Melis, MD, Radboud University Nijmegen Medical Centre, Department of Geriatric Medicine 925, PO Box 9101, NL, 6500 HB, Nijmegen, The Netherlands. E-mail: r.melis{at}ger.umcn.nl


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Background. There is growing interest in geriatric care for community-dwelling older people. There are, however, relatively few reports on the economics of this type of care. This article reports about the cost-effectiveness of the Dutch Geriatric Intervention Program (DGIP) compared to usual care in frail older people at 6-month follow-up from a health care system's point of view.

Methods. We conducted this economic evaluation in an observer-blind randomized controlled trial (Dutch EASYcare Study: ClinicalTrials.gov Identifier NCT00105378). Difference in treatment effect was calculated as the difference in proportions of successfully treated patients (prevented functional decline accompanied by improved well-being). Incremental treatment costs were calculated as the difference in mean total care costs. The incremental cost-effectiveness ratio (ICER) was expressed as total cost per successful treatment. Bootstrap methods were used to determine confidence intervals (CI) for these measures.

Results. The average cost of the intervention under study (DGIP) was 998 euros (95% CI, 888–1108). The increment in total cost resulting from DGIP was a little over 761 euros (–3336 to 4687). Hospitalization and institutionalization costs were less; home care, adult day care, and meals-on-wheels costs were higher. There was a significant difference in proportions of successful treatments of 22.3% (4.3–41.4). The number needed to treat was approximately 4.7 (2.3–18.0). The ICER is 3418 euros per successful treatment (–21,458 to 45,362). The new treatment is cost-effective at a willingness-to-pay of 34,000 euros.

Conclusion. The results of this economic evaluation suggest that DGIP is an effective addition to primary care for frail older people at a reasonable cost.

There is growing interest in geriatric care for community-dwelling older people (1,2). Both growing awareness of patient autonomy and population aging drive this development (3,4). Community-based geriatric intervention models aim to improve functioning, well-being, coping styles, and mortality among community-dwelling older persons. The idea is that increasing older people's autonomy will also increase their quality of life. Despite some promising effects on disabilities, mood, and mortality, there is still much controversy about important determinants of success, such as which patients to include, the setting, and the intensity of intervention (1,5,6). However, for geriatric primary care to be truly valuable to society it needs to provide effective interventions that give ‘value for money.’ There are relatively few reports on the economics of outpatient comprehensive geriatric assessment or preventive home visits (7,8). A review of comprehensive geriatric assessments concluded that evidence suggests no increase of total costs of care (7). Another review of preventive home visits stated that there is a potential for producing net cost savings (8).

From April 2003 until July 2005 we carried out a randomized comparison of a multidisciplinary home-based intervention program for frail older people with usual care (Dutch EASYcare Study) (9). The comparison showed that patients' well-being and functional abilities improved compared to control conditions (10). This article reports about health care utilization and the cost-effectiveness of the Dutch Geriatric Intervention Program (DGIP) compared to usual care in frail older people from a health care system's point of view.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Study Design and Setting
We conducted an economic evaluation in an observer-blind randomized controlled trial (ClinicalTrials.gov Identifier NCT00105378) of which the design was previously published (9). We used a health care system's perspective to identify all relevant costs. The local ethical committee approved this study.

Patients
Primary care physicians referred frail older patients to our intervention model when there was a problem in cognition, nutrition, behavior, mood, or mobility. The problem had to call for nursing assessment, coordination of care, or therapeutic monitoring and case management. Patients were living in their own home or in a home for the aged and were 70 years old or older. We restricted patient inclusion to those with limitations in cognition, (instrumental) activities of daily living, or mental well-being.

Randomization
Patients were randomized over two treatment arms: DGIP and usual care. We used a two-step pseudo cluster randomization procedure, because we expected methodological problems when using either individual randomization or cluster randomization (9,11,12).

Interventions
A geriatric specialist nurse visited the patients randomized to the intervention arm. During a maximum of 3 months, up to six visits for additional geriatric evaluation and management were performed. The nurse, geriatrician, and primary care physician had frequent consultations on individual cases. No restrictions were imposed on the care patients received in the usual care group. However, no model of care comparable to DGIP was regularly available in usual care.

Data Collection
Effect measures.-- The patients were interviewed to obtain written informed consent and to collect baseline demographic characteristics, data on general health conditions, and data on use of home care. These measurements were repeated 6 months after inclusion. The outcome measures used for this economic evaluation were functional performance in (instrumental) activities of daily living measured using the Groningen Activity Restriction Scale (GARS-3) and mental well-being using the mental health subscale of the Medical Outcomes Study Short Form (MOS-20MH) (13,14).

Cost analysis.-- Costs were measured by registration of volumes of care consumed and multiplied by appropriate prices per unit of care. To be able to calculate the costs of DGIP, DGIP nurses registered the time spent on the intervention using a Microsoft Outlook agenda. They registered the visits per patient and the time spent on consultation, phone calls, traveling, and administration. Data on the workload of the primary care physician and the geriatrician resulting from DGIP were derived from the workload of the nurses.

We also established the amount of care needed during the follow-up period of 6 months. To calculate incremental costs, data collection of cost measures focused on cost drivers, that is, factors that were likely to cause the most important cost differences between strategies. These were: utilization of family physician care (number of practice visits, house calls, and telephone calls), the number of referrals to other health care providers (e.g., outpatient specialist care, paramedical disciplines such as physical therapist, occupational therapist), and the number of days in hospital. All the above information was taken from the primary care physician's information system. The number of days hospitalized was cross-checked against the information provided by the patient using questionnaires at 3 and 6 months of follow-up. The patients provided information on the amount of home care, adult day care, and meals-on-wheels they used during follow-up. They also provided information on number of days institutionalized in a nursing home or home for the aged; this information was cross-checked against the information available from the primary care physician's information system.

Cost prices per unit of care (euros) were taken from Dutch guidelines for economic health care evaluations (15). In accordance with these guidelines, 45% overhead costs were added to the total direct costs to calculate a cost price per DGIP visit. Cost prices of referrals to outpatient specialist care and paramedical disciplines were estimated because this information was not readily available. All prices were indexed to the price level of 2005, using the Dutch consumer price index figures for health care costs (16).

Statistical Analysis
To obtain an interpretable effect measure, we considered the treatment a success if a patient's MOS-20MH score increased by more than 10 points and the GARS-3 score declined no more than 4.5 points. We used this definition because the course of GARS-3 scores showed a decline in our control group, whereas fairly steady MOS-20MH scores were observed over a 6-month period. Relative to the score range, both cutoffs represent approximately a 10% change in score.

Difference in treatment effect was calculated as the difference in proportions of successfully treated patients. Incremental treatment costs were calculated as the difference in mean total care costs. The incremental cost-effectiveness ratio (ICER) was expressed as total costs per successful treatment. Bootstrap methods were used to determine confidence intervals for these treatment arms' differences that take into account the clustering of patients at the level of the primary care physician. From the bootstrap sample, a cost-effectiveness acceptability curve was drawn. This curve gives an estimate of the proportion of the bootstrap distribution favoring one strategy over the other given a willingness-to-pay (WTP) for a gained unit of effect (17). We performed a sensitivity analysis for missing data; missing values were substituted with the mean of the other group (18).


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
In this study, 151 patients were included. Our study population consisted of a majority of widowed women born in The Netherlands. A majority (85%) lived in their own home; the rest lived in homes for the aged. The patients had a mean age of 82.2 years, a Cumulative Illness Rating Scale-Geriatrics of approximately 10 points, considerable cognitive deterioration, and low mental well-being. Approximately half the study group had home care available at baseline.

In the group that received DGIP, 85 patients were included, and 66 were included in the usual care group. Baseline characteristics and baseline measures of primary outcomes showed no significant differences between study groups, except for the score on loneliness (Table 1). Figure 1 shows the participant flow through the study. A nurse visited DGIP patients almost four times (Table 2).


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

  		Table 1. Demographic Characteristics and Outcome Measures of the Study Population at Enrollment.

 

Figure 01
View larger version (22K):
[in this window]
[in a new window]

  		Figure 1. Study flow chart. *A number of participants withdrew their consent for the visit for data acquisition by interview, but agreed to fill in the questionnaires. {dagger}Differences between numbers assessed at follow-up and numbers included in the analysis result from the fact that, although assessment was executed, participants did not always completely fill in the instruments. Therefore, responder status could not be established for everyone assessed

 

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

  		Table 2. Number of Health Care Units Used Per Patient During 6-Month Follow-Up.

 
Of all the patients, 34 were successfully treated: They experienced an improvement in well-being without a decline in functional performance (Table 3). Of these patients, 7 were in the control group, and 27 were in the intervention group. The significant difference in proportions of successfully treated patients was 22.3% (95% confidence interval [95% CI], 4.3-41.4) in favor of DGIP, and the number needed to treat was 4.7 (95% CI, 2.3-18.0). The DGIP intervention cost 998 euros per patient (95% CI, 888–1108). The incremental cost of the nursing program was 761 euros (95% CI, –3336 to 4687). Hospitalization (–675 euros) and institutionalization in homes for the aged and nursing homes (–841 euros) cost less in the DGIP group, whereas home care (+952 euros), day care (+241 euros), and meals-on-wheels (+91 euros) were more expensive (Table 3). The ICER is 3418 euros per successful treatment (95% CI, –21,458 to 45,362). The probability of DGIP being the "dominant" intervention was estimated to be 34.6% (Figure 2, proportion of ICERs falling in the Southeast quadrant). The vertical axis of the cost-effectiveness acceptability curve represents the probability that the ICER of DGIP compared with usual care is acceptable for a range of values of the WTP per successful treatment (Figure 3). If society is willing to pay 34,000 euros or more for a successful treatment, then there is a 95% probability that DGIP is efficient.


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

  		Table 3. Treatment Effect and Health Care Usage and Costs Per Patient at 6-Month Follow-Up.

 

Figure 02
View larger version (19K):
[in this window]
[in a new window]

  		Figure 2. Cost-effectiveness plane for the incremental costs and effects of Dutch Geriatric Intervention Program (DGIP) compared to usual care

 

Figure 03
View larger version (6K):
[in this window]
[in a new window]

  		Figure 3. Cost-effectiveness acceptability curve of Dutch Geriatric Intervention Program (DGIP) compared to usual care

 
During follow-up, 25 patients had missing values for treatment success; 10 patients were in the intervention group, and 15 were in the control group (Figure 1). However, the results from the sensitivity analysis for missing data did not essentially differ from the results presented (data not shown).


   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
In this group of frail patients, the mean total care costs in the control group were approximately 9000 euros over a 6-month period. The yearly health care costs in the age group 75–84 years is 8408 euros in The Netherlands (19), but a vulnerable population—in which higher health care costs can be expected—was included in this study. The average cost of the intervention under study (DGIP) was almost 1000 euros. The increment in total costs resulting from DGIP was a little over 750 euros. Hospitalization and institutionalization costs were lower; home care, adult day care, and meals-on-wheels costs were higher. There was a difference in proportions of successful treatments (prevented functional decline accompanied by improved well-being) of about 20%. The number needed to treat was approximately 5. The ICER is roughly 3500 euros per successful treatment. The new treatment is cost-effective at a WTP of 34,000 euros.

Comparison of these results with the literature is difficult, because there are not many cost-effectiveness studies in this field. Most articles only provide data on incremental costs without calculating ICERs (7). The available reviews of literature hint at the possibilities of cost savings, but request more research into this matter at the same time (7;8). One article on preventive home visits concluded that the visits produced net cost savings in the third year (20). However, no information on the uncertainty of this estimation was given, which limits interpretation. A sound cost-effectiveness evaluation of outpatient geriatric evaluation and management estimated an ICER of 10,600 dollars per quality-adjusted life year (QALY), and 26,500 per QALY if the time horizon of 5 years was limited to the follow-up period of 64 weeks (21). Another randomized study of outpatient geriatric evaluation and management concluded that there were no increases in costs (22). Our study did show a small increment in costs, although the cost-effectiveness seems reasonable. The uncertainty in the estimation is large, because the differences in total care costs are large. Total care costs were < 50 euros for some patients, but for others the amount was > 40,000 euros. This enormous range in individual health care costs reflects the heterogeneity in the care needed.

An explanation for the positive results in some of the earlier studies might be that those studies focused on certain aspects of health care, such as hospital costs (7). Our study took a wide health care system's perspective as a starting point. Our study showed lower hospital and institutionalization costs, but home care and adult day care costs were higher. This finding is probably a direct intervention effect, because often nurses organized more home and adult day care. As independent living is considered an important aim (3), these increases seem reasonable and justifiable.

We focused on all cost drivers relevant to the viewpoint taken. Medication, laboratory, and diagnostics costs were not taken into account, because the influence of DGIP on these cost parameters was judged to be negligible. For the same reason, no societal perspective was used. The effect of the intervention on patient productivity is very limited, given their age. The effect on caregiver productivity is probably fairly small as well; caregiver burden and time spent on care was hardly affected by the intervention.

No QALYs were used in our study, which limits comparability with other interventions. In contrast, the use of QALYs in gerontological research is delicate and its interpretation difficult. Generic instruments like EuroQoL, from which QALYs can be calculated, probably show floor effects in frail patients as in our study. Also, health states of frail older people are complex and often not sufficiently graded in terms of QALYs.

Of course, every operationalization of "successful treatment" is arbitrary to a certain degree. However, the difference in successfully treated patients, which was 22.3% in favor of DGIP with the currently used dichotomization, is rather insensitive to change of cutoffs. Varying the cutoffs on the GARS-3 and MOS-20MH in different combinations resulted in differences in treatment success rates, roughly ranging from 15% to 25% all in favor of DGIP (data not shown).

This economic evaluation was conducted alongside a randomized trial, which is an important safeguard against threats to internal validity. Study arm sizes were somewhat different, but patients were comparable at baseline. The number of participants with missing values (20%) was expected given their frailty. Participants who were lost to follow-up were older and had worse GARS-3, MOS-20MH, and Mini-Mental Status Examination (MMSE) scores at baseline. Moreover, when we used a conservative imputation strategy—assigning "mean of the other group" to missing patients—a sensitivity analysis showed about the same results (18).

External validity benefited from the wide health care system perspective that this evaluation used. The study conditions were similar to current practice; as in usual care, many physicians were involved in referring patients. Also, no strict treatment protocols were used; the control group received usual medical care without any restrictions, and guidelines led the DGIP intervention without restraining free selection from the available treatment options.

The results of this economic evaluation suggest that DGIP is an effective addition to primary care for frail older people at a reasonable cost. However, there is a need for larger and more rigorously designed economic evaluations of this type of health care model, because uncertainty in the cost-effectiveness of these models still exists.


   Acknowledgments
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Note: The Dutch EASYcare Study is labeled as Clinicaltrials.gov Identifier NCT00105378.

This work was supported by ZonMw (The Netherlands Organization for Health Research and Development) and by the Radboud University Nijmegen Medical Centre.

We thank Marleen Lenkens and Hanny Hordijk, geriatric specialist nurses, who executed the home visiting program; Sascha van de Poll, Jurgen Claassen, and Marieke Hartgerink, geriatricians, who supervised the home visits; Mebeline Boon, who executed a large part of data entry and data editing; and Hans Bor for his statistical support.


   Footnotes
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Decision Editor: Darryl Wieland, PhD, MPH

Received July 30, 2006

Accepted December 4, 2006


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Stuck AE, Egger M, Hammer A, Minder CE, Beck JC. Home visits to prevent nursing home admission and functional decline in elderly people: systematic review and meta-regression analysis. JAMA. 2002;287:1022-1028.[Abstract/Free Full Text]
  2. Melis RJ, Olde Rikkert MG, Parker SG, van Eijken MI. What is intermediate care? BMJ. 2004;329:360-361.[Free Full Text]World Health Organization (WHO). Active ageing: a policy framework. Geneva, Switzerland: WHO; 2002.
  3. Salvage AV, Jones DA, Vetter NJ. Options of people aged over 75 years on private and local authority residential care. Age Ageing. 1989;18:380-386.[Abstract/Free Full Text]
  4. Stuck AE, Siu AL, Wieland GD, Adams J, Rubenstein LZ. Comprehensive geriatric assessment: a meta-analysis of controlled trials. Lancet. 1993;342:1032-1036.[Medline]
  5. Fletcher AE, Price GM, Ng ES, et al. Population-based multidimensional assessment of older people in UK general practice: a cluster-randomised factorial trial. Lancet. 2004;364:1667-1677.[Medline]
  6. Wieland D. The effectiveness and costs of comprehensive geriatric evaluation and management. Crit Rev Oncol Hematol. 2003;48:227-237.[Medline]
  7. Elkan R, Kendrick D, Hewitt M, et al. The effectiveness of domiciliary health visiting: a systematic review of international studies and a selective review of the British literature. Health Technol Assess. 2000;4:(13): i-v 1–339.[Medline]
  8. Melis RJ, van Eijken MI, Borm GF, et al. The design of the Dutch EASYcare study: a randomised controlled trial on the effectiveness of a problem-based community intervention model for frail elderly people [NCT00105378]. BMC Health Serv Res. 2005;5:65.[Medline]
  9. Melis RJ, van Eijken MI, Teerenstra S, et al. Multidimensional Geriatric Assessment. A randomised study of a multidisciplinary program to intervene on geriatric syndromes in frail older people who live at home (Dutch EASYcare Study) [Clinicaltrials.gov Identifier NCT00105378]. J Gerontol Biol Sci Med Sci. 2008;63A:283–290.
  10. Teerenstra S, Melis RJ, Peer PG, Borm GF. Pseudo cluster randomization dealt with selection bias and contamination in clinical trials. J Clin Epidemiol. 2006;59:381-386.[Medline]
  11. Borm GF, Melis RJ, Teerenstra S, Peer PG. Pseudo cluster randomization: a treatment allocation method to minimize contamination and selection bias. Stat Med. 2005;24:3535-3547.[Medline]
  12. Kempen GI, Miedema I, Ormel J, Molenaar W. The assessment of disability with the Groningen Activity Restriction Scale. Conceptual framework and psychometric properties. Soc Sci Med. 1996;43:1601-1610.[Medline]
  13. Kempen GI. Het meten van de gezondheidstoestand van ouderen. Een toepassing van een Nederlandse versie van de MOS-schaal. [Assessment of health status of the elderly. Application of a Dutch version of the MOS scale]. Tijdschr Gerontol Geriatr. 1992;23:132-140.[Medline]
  14. Oostenbrink JB, Bouwmans CA, Koopmanschap MA, Rutten FF. Handleiding voor Kostenonderzoek, Methoden en Standaard Kostprijzen voor Economische Evaluaties in de Gezondheidszorg (Manual for Research on Costs, Methods and Standardized Cost Prices for Economic Evaluation in Health Care). Diemen, The Netherlands: Health Care Insurance Board (College voor Zorgverzekeringen; CVZ); 2004.
  15. StatLine (Electronic databank of Statistics Netherlands). Available at: http://statline.cbs.nl/StatWeb/Start.asp?lp=Search/Search&LA=EN&DM=SLEN.
  16. Lothgren M, Zethraeus N. Definition, interpretation and calculation of cost-effectiveness acceptability curves. Health Econ. 2000;9:623-630.[Medline]
  17. Unnebrink K, Windeler J. Intention-to-treat: methods for dealing with missing values in clinical trials of progressively deteriorating diseases. Stat Med. 2001;20:3931-3946.[Medline]
  18. Polder JJ, Takken J, Meerding WJ, Kommer GJ, Stokx LJ. Costs of illness in the Netherlands. The "care euro" unraveled (Kosten van ziekten in Nederland. De zorgeuro ontrafeld.) 2002. Bilthoven, National Institute voor Public Health and the Environment (Rijksinstituut voor Volksgezondheid en Milieu; RIVM).
  19. Stuck AE, Minder CE, Peter-Wuest I, et al. A randomized trial of in-home visits for disability prevention in community-dwelling older people at low and high risk for nursing home admission. Arch Intern Med. 2000;160:977-986.[Abstract/Free Full Text]
  20. Keeler EB, Robalino DA, Frank JC, Hirsch SH, Maly RC, Reuben DB. Cost-effectiveness of outpatient geriatric assessment with an intervention to increase adherence. Med Care. 1999;37:1199-1206.[Medline]
  21. Cohen HJ, Feussner JR, Weinberger M, et al. A controlled trial of inpatient and outpatient geriatric evaluation and management. N Engl J Med. 2002;346:905-912.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Journals of Gerontology Series A: Biological Sciences and Medical SciencesHome page
R. J. F. Melis, M. I. J. van Eijken, S. Teerenstra, T. van Achterberg, S. G. Parker, G. F. Borm, E. H. van de Lisdonk, M. Wensing, and M. G. M. O. Rikkert
Multidimensional Geriatric Assessment: Back to the Future A Randomized Study of a Multidisciplinary Program to Intervene on Geriatric Syndromes in Vulnerable Older People Who Live at Home (Dutch EASYcare Study)
J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2008; 63(3): 283 - 290.
[Abstract] [Full Text] [PDF]

Home page
 Journals of Gerontology Series A: Biological Sciences and Medical SciencesHome page
A. Huss, A. E. Stuck, L. Z. Rubenstein, M. Egger, and K. M. Clough-Gorr
Multidimensional Geriatric Assessment: Back to the Future Multidimensional Preventive Home Visit Programs for Community-Dwelling Older Adults: A Systematic Review and Meta-Analysis of Randomized Controlled Trials
J. Gerontol. A Biol. Sci. Med. Sci., March 1, 2008; 63(3): 298 - 307.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS