The population worldwide is aging. The demographic change is challenging to health care organizations and highlights the need for effective preventive and proactive care models in primary care, especially for older people. This study, "Secure and focused primary care for older people" (SAFE), investigates the effectiveness of a new proactive care model based on comprehensive geriatric assessment (CGA) in primary care in a population with high risk of hospitalisation.

Background and Aim

The number of old people is rising and in Sweden, approximately 20% of the population is older than 65 years. In 2050, the number of people 85 years and above, is expected to have doubled [1]. The demographic change is challenging to future healthcare systems [2]. The prevailing strategy for reducing the healthcare impact of an ageing population has been 'compression of morbidity', to extend the healthy period of life, and delay disability until a brief period at the end of life [3]. Thus, preventative and proactive primary care is central in meeting future challenges such as an aging population and highlights the need for effective preventive and proactive care models in primary care, especially for older people. This study, "Secure and focused primary care for older people" (SAFE), investigates the effectiveness of a new proactive care model in primary care in a population with high risk of hospitalisation.

Comprehensive geriatric assessment (CGA) is considered gold standard in evaluation and caring for old in-hospital patients [4]. CGA is described as a multidimensional, multidisciplinary and holistic evaluation of the health status of an older person, together with the formation of a care plan based on individual needs and preferences [4]. Data from outpatients in geriatric care has shown that CGA may delay the progression of frailty, but the study population was quite small [5].There is evidence that CGA can decrease the need of inpatient care and nursing home admissions. Some studies suggest that comprehensive care programs can be cost-effective. They also seem to be widely accepted and increase patient satisfaction [6,7]. However, results are conflicting and meta-analyses suffer from the lack of a universal definition of frailty and the great variation of interventions, outcome measures and scales to measure frailty. Still, both NICE guidelines and the ICOPE recommendations of the WHO include CGA for older people with frailty and/or multimorbidity [8].

In a previous study, "Proactive healthcare for frail elderly persons", a predictive statistical model that identified individuals, 75 years and above, with high risk for hospitalisation during the coming 12 months, was validated [9]. The effectiveness of CGA adapted to primary care using the new CGA tool: the Primary care Assessment Tool for Elderly (PASTEL), delivered to older adults with identified high risk for hospitalisation, was evaluated [10]. This pragmatic multicenter trial comprised nine intervention practices and ten matched control practices in the county council of Östergötland, Sweden, in 2017-2019. No specific intervention measures beyond the CGA assessment with PASTEL were described or analyzed in the study. Follow-up was part of ordinary clinical routine. Which specific interventions in primary care that really make a difference to reduce risk for hospitalisation and improve quality of life for old outpatients remains to be explored.

The main aim of the present study is to examine whether a proactive care model with comprehensive geriatric assessment (CGA) in primary care with additional long-term care coordination and increased patient participation, contributes to reduced inpatient care and/or increased quality of life among community-dwelling older people. The CGA used in this study will be based on the instrument "the Primary Care Assessment Tool for Elderly" (PASTEL).

Design and Method

This is a prospective, multi-center trial that will be carried out in two regions in Sweden (Östergötland and Jönköping). The study will include 26 intervention primary care practices and 25 matched control primary care practices. Data will be collected at baseline, at one year follow-up (12 months) and at two year follow-up (24 months).

Participants will be identified through a statistical prediction model based on age, health care use and diagnostics data covering the previous year. A risk score for future morbidity and health care need will be calculated and the participants with the highest risk scores (top 15%) will be invited to participate in the study.

Inclusion Criteria:

• 75 years or older
• community dwelling (living in own home)
• Top 15% of risk score calculation

Exclusion Criteria:

- Persons living in nursing homes

Data may be made available upon request when the study is completed following the laws for sharing patient data.