Assessment of frailty in old people with multimorbidity

Abstract: Background: As the population grows older, understanding and management of multimorbidity and frailty will become increasingly important. There is no consensus on how to define or measure frailty and awareness of frailty is limited in many healthcare settings. This impedes implementation of frailty assessment and management in routine care. The overall aim of this thesis was to contribute to the understanding of assessment and management of frailty in old people with multimorbidity. Study I: Study I aimed to evaluate the effect of outpatient Comprehensive Geriatric Assessment (CGA) on frailty in community dwelling older people with multimorbidity and high health care utilization. Methods: The Ambulatory Geriatric Assessment—Frailty Intervention Trial (AGe-FIT) was a randomized controlled trial (intervention group n=208, control group n=174) with a follow-up period of 24 months. Frailty was a secondary outcome. Inclusion criteria: age ≥75 years, ≥3 current diagnoses in ICD-10, and ≥3 hospitalisations in the last 12 months. The intervention group received CGA-based care and tailored interventions by a multidisciplinary team in addition to usual care. The control group received usual care. Frailty was measured with the frailty phenotype (FP). At 24 months, frail and deceased participants were combined in the analysis. Results: Ninety percent of the population were frail or pre-frail at baseline. After 24 months, there was a significant smaller proportion of frail and deceased (p=0,002) and a significantly higher proportion of prefrail patients in the intervention group (p=0,004). Mortality was high, 18% in the intervention group and 26% in the control group. Conclusion: Outpatient CGA may delay the progression of frailty and may contribute to the improvement of frail patients in older persons with multimorbidity. Study II: Study II aimed to evaluated differences in health-related factors between older persons with multimorbidity who were discordantly classified by five frailty instruments, with focus on the Clinical Frailty Scale (CFS) and the Frailty Phenotype (FP). Methods: A cross-sectional study in a community-dwelling setting. Inclusion criteria were: ≥75 years old, ≥3 visits to the emergency department in the past 18 months, and ≥3 diagnoses in 3 different ICD-10 chapters. Frailty was assessed by CFS, FP, Short Physical Performance Battery (SPPB), grip strength and walking speed. Results: 385 participants had data on all frailty instruments. Prevalence of frailty ranged from 34% (CFS) to 75% (SPPB). Nine percent of participants were non-frail by all instruments, 20% were frail by all instruments and 71% had discordant frailty classifications. Those who were frail according to CFS but not by the other instruments had lower cognition and functional status. Those who were frail according to FP but not CFS were, to a larger extent, women, lived alone, had higher cognitive ability and functional status. Conclusion: The CFS might not identify physically frail women in older community-dwelling people with multimorbidity. They could be at risk of not receiving the attention motivated by their frail condition. Study III Aim: The aims of study III was to translate TFI into Swedish and study its psychometric properties in community-dwelling older people with multimorbidity. The Tilburg Frailty Indicator (TFI) is a questionnaire with 25 questions (10 on background factors, 15 on frailty) suitable for frailty-screening in Primary Care. TFI measures physical, psychological, and social frailty. Method: A cross-sectional study with participants with the same inclusion criteria as study II. International guidelines for back-translation were followed. Reliability (inter-item correlations, internal consistency, test–retest) and validity (concurrent, construct, structural) was evaluated. Results: 315 participants (58% women) were included with a mean age of 83,3 years. The reliability coefficient KR-20 was 0,69 for the total sum. 39 individuals were re-tested, and the weighted kappa was 0,7. TFI correlated moderately with other frailty measures. The individual items correlated with alternative measures mostly as expected. In the confirmatory factor analysis (CFA), a three-factor model fitted the data better than a one-factor model. Conclusion: We found evidence for adequate reliability and validity of the Swedish TFI and potential for improvements. Study IV The aim of study IV was to explore Primary Health Care Professionals’ (PHCP) experiences of frailty assessment with the TFI with focus on feasibility factors and their experiences of assessment and management of frailty. Methods: A qualitative semi-structured interview study. Participants were PHCPs (physicians, nurses, and physiotherapists). Interviews were thematically analysed using qualitative content analysis according to Graneheim and Lundman. Results: Nine interviews were performed. The PHCPs experiences were expressed in two themes: “Assessment of frailty with TFI is feasible if healthcare adjusts to older people” and “When the concept of frailty is unclear, interventions are uncertain.” TFI was described as easy to use and holistic. It was suggested that the TFI could be used as a self-administered questionnaire for the patient or by the PHCPs. The TFI provided useful information for care planning, although it took a long time. Some questions raised concerns. Conclusion: There was an overall positive perception of TFI’s clinical usefulness. The result highlights frailty as an elusive but important concept in care management of old people, which indicates a need of educational efforts aiming to increase knowledge about frailty in primary care.