This is a longitudinal intervention study based on a quasi-experimental design [16
]. Due to some constraints, a randomized controlled trial is impossible to realize. The constraints are mostly practical and ethical since all frail older people who wish to benefit from an innovative project cannot be denied access to it. In addition, researchers cannot have total control on the intervention since projects are delivering the services in the ‘real world’. Projects may organize their intervention according to specific types of services. Non randomization and variety of projects impose a thorough description and follow up of beneficiary population and intervention content. The Researchers use three comparison strategies to evaluate intervention – comparison of the arm benefiting from the intervention with older people benefitting from a different type of project, older people who are not benefiting from a project and are still living at home and older people who are already institutionalized.
Typology of projects
Projects are clustered according to the type of intervention they perform (see Table ). The typology definition is based on a previous study on the expressed intention and the perspective of actors involved in the project implementation [12
Typology of the projects according to main intervention
Sample size considerations
Beneficiaries will be grouped by type of projects as shown in Table in order to get a reasonable sample size for each side of the comparison. Using an effect size of 0.50, a two-sided significance level of 0.05 and a power of 85%, 75 participants are needed in each group. We expect each project to include at least 45 participants per year because this is the minimum size of case load projects reported at the beginning of their activities. The clusters of projects will have then a larger sample size than indicated by the sample size calculation.
The study will include older people receiving home care services who are at least 65
years old. Projects are asked to include only clients who are frail and at risk of institutionalization. Due to practical reasons, frailty is measured in the study with two different scales and caregivers are allowed to choose between them. One of these scales is the Edmonton Frail Scale [17
] with a cutoff point of 6. Another scale is an adapted version of the Katz scale for Belgian home care and residential care [19
]. Older people with a Katz score equal to A, B or C will be included [20
]. Clients who have been diagnosed with dementia by a geriatrician, neurologist or psychiatrist may also be included in the study.
Each project has to collect data on at least 75% of their total client population in order to ensure representativeness. This means that caregivers have to complete questionnaires for at least 75% of clients benefiting from their projects.
Ethical approval and informed consent
This observational study was approved by the Belgian Privacy Commission and by the Ethics committee of the Belgian Universities - Université Catholique de Louvain and Katholieke Universiteit Leuven with dossier number B40320108337. A formal procedure was implemented so that caregivers could fill out the questionnaires on a secured website.
For this study, older persons are asked to sign an informed consent agreement. In case they are not capable of signing this document, a family member or another legal representative will sign it on their behalf, as stipulated by Belgian law [21
]. Clients do not undergo the intervention because of the research. The research however evaluates the effects of the intervention. Clients have the right not to participate in the research and they may withdraw their consent at any time. In that case, all data for this person is removed. Their consent or refusal to consent does not affect their participation in the intervention.
All data are anonymized and analyzed according to the rules of the Belgian Privacy Commission [22
The aim of the study is to answer the following research questions:
1. Do characteristics of populations differ between projects?
2. Which types of projects are associated with less decrease of the functional and cognitive performance of frail older persons?
3. Which types of projects contribute to a better quality of life of frail older persons?
4. Which types of projects are associated with a decreased risk of institutionalization of frail older persons?
5. Is the interRAI Maple scale a good predictor of institutionalization in our study?
Selection of Variables
A previous systematic review identified the variables most frequently used in the evaluation of interventions aiming at maintaining frail older persons at home [14
]. In our study, variables from the interRAI Home Care instrument which correspond to these identified variables will be used in the evaluation of interventions [23
For data without any correspondence with the InterRAI instruments, other tools have been selected by a consensus amongst experts and according to their international validity. These complementary tools are the WHOQoL-8 and Zarit Burden scale (12-item) [24
interRAI HC instrument
The interRAI HC instrument is an internationally validated instrument consisting of several domains such as cognitive functioning, ADL, social and psychological wellbeing, health status, environmental characteristics, etc. The use of a comprehensive geriatric assessment such as the interRAI HC is essential for this study in order to have an in-depth view of the client’s health and functional status [27
]. The follow up of these characteristics will give insight in the evolution of the client’s situation and will enable researchers to show the effect of innovation projects. In the study we use the validated Belgian version of this instrument named BelRAI HC.
The WHO defines Quality of Life as “individuals’ perception of their position in life in the context of the culture and value systems in which they live and in relation to their goals, expectations, standards and concerns” (24). The WHO-QOL-8 scale measures the perceived overall quality of life of the clients. It is derived from the WHO-QOL-Bref and contains each of the domains: physical and psychological health, social relations and environment [30
]. It is currently used as an instrument in the WHO’s SAGE Survey [32
]. This scale was designed for use where researchers need a short and concise instrument to measure quality of life. It is a self report instrument, so it has to be filled out by the clients themselves.
Zarit Burden Interview 12 (ZBI-12)
In the international literature caregiver burden is perceived as a multidimensional response to stressors associated with care giving. These stressors may be physical, psychological, emotional, social and financial [33
]. In this research we use a shorter version of the original 22-item scale Zarit Burden Interview. This scale contains 12 items with a score from 0 to 4. The total score can vary from 0 to 48 and the higher the total score, the higher the caregiver burden. The short form was chosen, in order not to put a heavier weight on the burden of the caregiver and to increase the response rate.
Professional caregivers are asked to fill out the interRAI Home Care instrument. Clients fill out the WHO-QOL-8 and the main informal caregivers complete the ZBI-12. These questionnaires are filled out at inclusion of the frail older person in the project (baseline), at exit from the project and 6
months after baseline. Additionally, if clients stay longer than 6
months in the project, caregivers have to do a follow-up every 6
months until the moment clients leave the project. Follow-up of clients happens from March 2011 until July 2013. A client follow-up period will vary between 6 to 36
months. Criteria to exit the cohort will be institutionalization longer than 3
months and death. Follow up will also end for clients who leave the projects.
Professional caregivers are instructed on how to fill out the questionnaires during a 2-day training and a follow up training of 1
day. They fill out the interRAI HC based on observation and on interview of the older person and of the main informal caregiver.
The interRAI-data will be linked with data from the national database IMA (Inter-mutuality Agency). This database contains data on national health service consumption reimbursed by NIHDI such as medication use, hospital days, medical, nursing and physical therapy use, etc. This data will be added to the analysis for the clients in the projects.
The “permanent sample” of people over 65 extracted from IMA database will also serve as a control group for our study population. By using this database we will be able to compare older persons who are benefiting from the innovative projects with older persons who are not benefiting from these projects (control group). In this control group we will have two major groups – clients who are still living at home and clients who are institutionalized. All clients who are in the projects and have not given their informed consent will be excluded from this database.
– Research question 1 - Do characteristics of populations differ across projects?
In order to answer this first research question comparisons will be made between the populations in the different types of projects to explore whether they present particular characteristics. In these comparisons we will use the variables shown in Table Frequencies will be calculated for categorical variables and means and medians with confidence intervals for continuous variables. Other possible comparison will be done by analysis of variance. All the statistical analysis in the study will be performed with the software Stata 11.1.
Variables to characterize the population of the study
– Research question 2 - Which types of projects are associated with less decline of the functional and cognitive performance of frail older persons?
– Research question 3 - Which types of projects contribute to a better quality of life of frail older persons?
Longitudinal analyses will be performed to answer the second and third research questions. Outcome variables will be the interRAI ADL hierarchy scale, the interRAI CPS2 scale and the WHOQOL scale. Mean will be compared between arms of the study and covariate analysis (ANCOVA) will be done to control results for potentially confounding variables.
The ADL hierarchy scale is a measure of a person’s functional performance. It is based on the concept of early, middle and late loss of ADLs and consists of 4 items from the interRAI instruments: personal hygiene, toilet use, locomotion and eating. The scores vary from 0 to 6 and the higher the score, the greater the dependency [36
The cognitive performance scale (CPS2) is an internationally validated scale that describes a person’s cognitive status. The scores vary from 0 to 6 and the scale consists of 5 items from the interRAI HC instrument: skills for daily decision-making, making oneself understood, short-term memory recall, procedural memory and eating impairment. Higher scores indicate a greater degree of cognitive impairment. Several studies have shown a significant correlation between the CPS scale and the Mini-Mental State Examination (MMSE) [38
– Research question 4 - Which types of projects are associated with a decreased risk of institutionalization of frail older persons?
In order to answer the fourth research question, the definitive institutionalization (more than 90 consecutive days in residential care) will be used as event. The analysis consists of the calculation of incidence rates, cumulative incidence and hazard rates of definitive institutionalization through survival analyses for each type of project. Survival analysis measures the time to a certain event or change of state, in this case, institutionalization. The multivariate Cox model which will be calculated from this analysis will allow us to isolate the effects of the intervention from the effects of confounding variables. These variables are extraneous variables which correlate (positively or negatively) with both the dependent variable and the independent variable. In this analysis we will use the variables of Table as confounding variables. The use of a Cox model may improve the estimate of the intervention effect by narrowing the confidence interval. Another important advantage of this analysis is that it takes into account the variability of covariates across time which makes it very appropriate for longitudinal studies [42
– Research question 5 - Is the interRAI Maple scale a good predictor of institutionalization in our study?
The fifth research question will be approached by using a multiple logistic regression model with the Methods for Assigning Priority Levels scale (Maple) as one of the independent variables. This interRAI scale is known to be a predictor for institutionalization and can be used as an indicator for allocation of home care resources and prioritization of clients needing community or residential care. MAPLe is an internationally validated predictor of nursing home placements, caregiver distress and ratings that the client would be better off elsewhere [45