Persons in the homes for the elderly suffer greatly from (multiple) chronic diseases and associated disablement 3 Over the last decades, Dutch residents of homes for the elderly have become older and more disabled and show more and more resemblance with nursing home patients 45. Primary care physicians (PCP) are responsible for the medical treatment of persons living in homes for the elderly. However, primary care physicians are often unable to handle the complex medical problems 67. Many health problems go unnoticed by the primary care physicians 8.

The Disease Management Model is based on 3 elements: coordination of care, guiding of the care process and empowerment of the patient 10. This model is strongly recommended to improve the health and quality of life of the chronically ill 111213. Beneficial effects of disease management were reported among stroke patients and among diabetes mellitus type 2 patients 141516. However no studies were performed yet to evaluate the effects on functional health, quality of care and the cost-effectiveness of disease management for disabled persons in homes for the elderly. We use the concept of disease management but not focused on the diseases level but on the disabilities and handicaps they cause.

Already in 1995 the National health Council of the Netherlands stressed the importance of improving the quality of care for chronic patients by a shared disease management of the health professionals involved, with clearly defined medical responsibilities, and the development of shared management protocols 17.

In residential homes, implementation of the three elements of the Disease Management Model demands specific adaptation and agreement across the responsible players. For example, who is best suited to do the guidance and coordination? Primary care physicians are responsible but do not regard themselves suited for systematic management and long-term monitoring for chronic diseases and disabilities associated with frail health 9. In residential homes, nurse helpers have daily contact with the frail elderly and are well positioned to coordinate the care. Appropriate care coordination demands up to date and state of the art medical and social input. Guidance and medical input is needed by regular contact with the PCP. However, PCP often skip team meetings as they can be at impractical times, or have unclear agenda's which greatly limits its potential value. To increase the quality of the team meetings, PCP should be present, and nurse helpers could be trained care to bring forward relevant medical observations. Also, a consultant such as a geriatrician or other old age physician may be invited to provide state of the art advice. This approach demands empowerment of residential personnel by systematic observation and effective communication with medical professionals.

In addition, the issue of patient empowerment can be problematic. For example, about half of the residents suffer from dementia which greatly limits the potential for patient empowerment.

In our aging population the number of persons with a chronic disease is expected to increase from 1994–2015 with 25–60% 18. In the Netherlands there are about 110.000 residents in homes for the elderly 4. Professional care is needed for 71% of the residents such as assistance with activities of daily living or mobility, nursing care (medication, wound care etc) and domestic help. Twelve percent of persons of 75 years and older live in a home for the elderly and 4% in a nursing home 19

The quality of care in these homes is frequently discussed in national and international newspapers. The care organizations responsible for the quality of care given in homes for the elderly often do not have the tools to measure and improve quality of care. Scientific studies of quality of care for the elderly are rare.

Aging is costly for health care systems. About one third of the health care expenditures in industrialized countries involves persons of 70 years and older. Elderly are massive consumers of medication and occupy most of hospital beds (3). Studies of comparable interventions and associated costs in residential homes are absent. Nevertheless, we reanalyzed two meta-analyses of Stuck 2002 and Elkan 2000 on preventive effects of home visits to community dwelling elderly and selected only studies that focused on frail elderly 12 of 27 trials 2021. Six of these studies that reported on costs, found that preventive visits or outreaching geriatric management reduced care costs 22.

Objectives for this article were to describe the design of an evaluation study on the clinical and economical effects of a new disease management model for residents in homes for the elderly.

A cluster randomized controlled clinical trial is carried out among five intervention homes and five usual care homes for the elderly in the north-west of the Netherlands that comprise over five hundred residents. There is a follow up of six months (Figure 1). The ethical committee of the VU medical center approved the study.

The randomization is carried out on home matched by percentage of psycho geriatric (i.e. cognitive impaired) residents. The care services and type of disability in homes with a high percentage of psycho geriatric patients are likely to differ depending on how many residents need psycho geriatric care. So, the homes were first ranked on percentage of psycho geriatric patients. The two homes with the highest percentage of psycho geriatric patients were than matched, and so on.

Next, we checked the risk of imbalance in numbers following Pocock's recommendations 23. If the difference in number of intervention and control residents would be >15% (75 or more) the randomization should be repeated until the imbalance was 15% or less.

Homes were all ordered on the percentage psycho-geriatric patients and numbered from rank one to rank 10. In this way matched homes are ranked after each other, one having an even and the other an uneven number. Randomization was carried out using Pocock's first column in his random numbers table 23. If the table's first number is even, the even number of first matched home is assigned the intervention. If the next table number is uneven, the uneven number of the next matched couple is assigned the intervention. And so on until all matched couples are assigned.

All residents were eligible except the terminally ill. Terminally ill is defined as death expected within six months. A family proxy of cognitive impaired persons was approached to provide outcome information.

All residents from the usual care homes as well as from the intervention homes receive an invitation letter and an informed consent form two weeks before the start of the study. If the resident is not able to understand the information and/or to sign the informed consent papers a close family member will be invited to participate and provide proxy information on the outcomes.

All eligible persons who sign an informed consent are going to be visited by an interviewer of the VU medical centre for an interview on their health and resource use. Table 1 provides an over-view of the measurements.

Table 1 shows an overview of all outcome measures and instruments.

Sample size calculations are based on the expected effects of the intervention on the main outcome measures concerning quality of care and functional health. In the following sample size calculations we used an alpha of 0.05, power of 80% and inflation of 10% because of anticipated intra-cluster correlation in the homes for the elderly. Regarding health related quality of life, Cohen's D effect size ranged from 0.5 to 3.8 in our meta-analysis 22. To detect a fair benefit, i.e. effect size = 0.5, a minimum of 64 persons is needed in each group 35. For functional health and disability we anticipate on a comparable effect-size and consequently identical sample size. Furthermore if we assume a dropout rate of 15% during the 6 months follow-up we need to include at least 100/85 × 64 × 110% = 82 persons in each group.

Effect analyses will be performed both on 'intention to treat' and per protocol principles. Differences between intervention and usual care patients at 6 months on the outcome measures (risk adjusted interRAI LTCF based Quality indicators, EuroQoL, functional health and disability) will be compared between the intervention and control group by both univariate and multivariate techniques. We will use the multivariate technique to adjust for possible differences in baseline scores and background variables between the intervention and control groups. Dropout and loss to follow up will be described. Potential effect-modification will be explored.

Especially, possible differential effects of disease management will be explored across residents with complex and simple health problems.

The process evaluation involves assessing the extent to which the intervention program is performed according to protocols, the nature of the recommendations made to the participants of the MM, compliance with these recommendations, physicians and therapists judgments about the intervention program and recommendations. Data on these topics are collected using structured registration forms during the intervention. Finally, semi-structured interviews will be held with the participating nurses, primary care physicians, and nursing home physicians at the end of the intervention period in order to record their experiences and opinions on the new disease management model.

Cost data are collected by resident interview at base line, and at 6 months from a societal perspective and supplemented with resource use as registered within the home for the elderly. In case residents are cognitively impaired or not able anymore to be interviewed, proxies will be sought, preferably close family members. Only direct healthcare costs will be considered, because all patients have retired. Included cost categories are costs of consultations with the general practitioner, the nursing home physician and medical specialists, hospitalizations and admissions to the medical department of the nursing home and use of medication and medical aid. Medication data are retrieved from the centralized pharmacy files in the research region. Care consumption will be valued according to guidelines for economic evaluation in health care in the Netherlands 3637.