Cross-sectional design assessing baseline data in a sample drawn from the REDIC-NH project.
The study population was a sub-sample of participants in the Resource Use and Disease Course in dementia – Nursing Home (REDIC-NH) project. The REDIC-NH study is a longitudinal observational study that includes newly admitted patients from 47 small and large nursing homes in four Norwegian counties and follows them from admission to the NH over a course of 5 years or death . Patients older than 65 years, or younger than 65 years but with established dementia, were included. In addition, the expected stay in the NH had to be more than four weeks. Patients with a life expectancy shorter than six weeks were not eligible. The study included a convenience sample of 695 persons, and recruitment took place between January 2012 and August 2014.
To increase homogeneity and describe the resource use in a dementia population exclusively, patients without dementia, or not permanently admitted to NH where excluded from the present study, as were participants without complete Resource Utilization in Dementia (RUD) questionnaire.
To be admitted to a long-term NH stay in Norway, the person must apply to the municipality. The application is evaluated based on a need assessment, and available places are allocated based on urgency. If there are no available places, the applicant usually enters a waiting list, with a waiting period from a few days up to several weeks.
Data were collected by healthcare workers at the nursing home, mainly registered nurses (74%), under the supervision of 10 research nurses after admission to NH. The research nurses completed five days of training prior to the start of the study, and the data collectors completed two days of training. Data were collected through cognitive and physical tests and structured interviews with the patients, their family caregivers, and the health workers.
Demographic data included the patients’ age and gender and were collected by reviewing the patients’ journals kept at the NHs. A diagnosis of dementia according to the ICD-criteria  was independently established by two of the authors (SB and GS) based on all available information about the patients. Both SB and GS are specialists in psychiatry and experienced in elderly psychiatry and research. If no consensus was reached, a third psychiatrist was consulted.
The clinical measures dementia severity and severity of physical health were obtained using the following instruments: The General Medical Health Rating (GMHR) , a four-category, reliable, and valid global bedside assessment tool for rating the severity of physical health. The score is based on an overall assessment by health care workers. The Clinical Dementia Rating Scale (CDR) assesses the severity of dementia as no dementia, possible dementia, and mild, moderate, or severe dementia. CDR comprises six items (memory, orientation, judgement and problem solving, community affairs, home and hobbies, and personal care). A score is calculated according to an algorithm where the memory item is given more weight. For statistical purposes, we calculated the CDR sum of boxes (CDR-SB), which offers an extended range of values and is calculated by adding the item scores (range 0-18). Higher scores indicate more severe dementia .
The extent of formal care and the extent and providers of informal care were recorded by the RUD questionnaire and included the following information about the family caregiver : Age, gender, relation to the patient, co-residential status, work status, hours worked last month, and lost work hours due to care tasks in the last month. Information about the extended social network included: Relation to patient and hours of provided informal care last month. The extent of informal care provided by the family caregiver last month was recorded in regard to the following three aspects: 1) the time used to help the patients with personal activities of daily living (PADL), 2) the time used on instrumental ADL (IADL), 3) and the time used on supervision, like helping the patient with orientation or preventing behavior that is distressing to the patient. We calculated the total informal care time by summarizing the amounts of time for these three types of care. If this sum exceeded 24 hours per day, a total informal care time of 24 hours per day was set. The time provided by professional home care service was used to calculate formal care due to insufficient or lacking data on services like home help, meal delivery, day care centres, or respite care.
Data are described by the means and standard deviations (SD) or frequencies and percentages. Differences between groups were assessed by Independent Samples t-test for continuous variables and the χ2-test for categorical variables. Associations between predefined covariates (gender, age, caregiver gender, caregiver relation to patient, caregiver in work, co-residency with family caregiver, GMHR, and CDR) and the three outcome variables were assessed by estimating bivariate and multiple linear mixed models. Random effects for nursing homes were included in the models to adjust the estimates for possible within-nursing-home correlations. Stratification by living with or without a caregiver was performed by including interactions between the dichotomous variable (co-residency with or without caregiver) and all covariates. Interactions with p<0.1 were kept in the model. Missing values in variables co-resident, carer relation, and carer in work were imputed by logical rules whenever possible. The analyses were performed using IBM SPSS Statistics for Windows version 25.0 (Armonk, NY: IBM Corp.) and SAS v 9.4. Results with p-values less than 0.05 were considered statistically significant.
The patients’ capacity to consent was assessed by the nursing home staff, including a physician. Written informed consent was obtained from patients with the capacity to consent or from the family caregivers on behalf of the patients in cases of reduced capacity to consent. The study was approved by the Regional Committee for Medical and Health Research Ethics (2011/1738).