Multi-criteria decision analysis of a proactive person-centred integrated primary care program care for frail elderly in the Netherlands: U-PROFIT

Background: The aim of the current study is to perform a multi-criteria decision analysis (MCDA) of the proactive, person-centred integrated primary care programme U-PROFIT for frail elderly in the Netherlands. Although, the cost-effectiveness of U-PROFIT has been assessed in a large trial, this MCDA includes a more comprehensive set of outcomes that are summarized in a single value to inform decision making from multiple perspectives. A unique feature of U-PROFIT is the two-steps frailty screening combining information from the Electronic Medical Record and self-reported data. The frail elderly receive a holistic assessment from a registered nurse that is translated into an individualised care plan. The nurse provides integrated care in collaboration with the GP and other relevant disciplines. Methods: The population consisted of frail elderly aged ≥ 69 years from three general practices in the Utrecht region. This study was a 12-month prospective cohort study applying a regression-discontinuity design. The cut-off score to define an intervention and control group was an age of 75; frail patients aged ≥ 75 years from the three GP practices received the U-PROFIT intervention, and frail patients aged 69-74 years were assigned to the control group based on the two-steps frailty screening. In both age groups elderly had to meet the same frailty criteria. Data on the performance of U-PROFIT was collected by questionnaires and is ongoing until September 2018. Outcomes included physical functioning, psychological well-being, social participation and relationships, enjoyment of life, person-centeredness and continuity of care. Inverse probability weighting was applied and linear regression analyses were performed including the treatment variable, age, baseline outcome and the interaction between treatment and age. The weights for the MCDA were obtained by a discrete choice experiment among patients, partners (informal caregivers), professional care providers, payers and policy makers (n~150 in each stakeholder group). The performance was multiplied by the weights to obtain the overall value score for U-PROFIT and usual care. Results: In the MCDA-analysis, the overall value for integrated care was higher than for usual care due to higher partial value scores of psychological well-being, enjoyment of life, person-centeredness and continuity of care. Results were consistent across stakeholder groups and also deterministic sensitivity analysis with Swing Weights show