Prevalence and risk factors of frailty among home care clients

This study is part of the larger Nutrition, Oral Health and Medication (NutOrMed) study. The participants were 75-year-old or older home care clients living in three cities in Eastern and Central Finland Home care clients [15]. The study population in two communities was a randomised sample of home care clients and in one community, a total sample of home care clients. In all, 440 participants were selected and home care nurses asked these clients if they were willing to participate. If the client was cognitively impaired, his or her proxy made the decision on participation. A total of 300 home care clients or their proxies gave written consent for the study. After that, 25 refused to participate, 4 died and 3 moved to another residence, while 11 had missing data. Complete baseline data on aCGA frailty was gathered from 257 participants. A more detailed description of the NutOrMed data is given in a previous article [15].

Nurses, nutritionists and pharmacists conducted interviews in the subjects’ homes. The nurses knew their clients and the nutritionists and pharmacists had previous experience in interviewing and assessing older persons. The interviews concerning sociodemographic factors, activities of daily living and instrumental activities of daily living, cognitive functioning, depressive symptoms and health status were carried out by a home care nurse. If a participant was not able to reply to the interview questions, a caregiver or nurse was interviewed.

Functional ability was assessed with Barthel’s activities of daily living (ADL) index (scale 0–100) [16] and Lawton’s instrumental activities of daily living (IADL) (scale 0–8) [17] and cognitive status was measured with the Mini-Mental State Examination (MMSE) on a scale of 0–30 [18], with higher scores indicating better functioning. The 15-item Geriatric Depression Scale (GDS-15) [19] was used to assess depressive symptoms. Self-rated health was determined using a 5-step scale (very good, good, moderate, poor, very poor), with the two last steps resenting poor health, and self-rated ability to walk 400 m using a 4-step scale (able to walk without difficulties, able to walk with help, unable to walk independently, unable to walk), with the two first steps representing independently [20].

A validated and standardised Mini Nutritional Assessment (MNA) test was used to measure nutritional status, with scores of 24.0–30.0 indicating normal nutritional status and scores <23.5, malnutrition or its risk [21]. The MNA test contains anthropometric measurements, questions on dietary habits, a global assessment and a self-assessment. A pharmacist collected information on drug use; this information was complemented with medication lists, packages and prescriptions. We used a cut-off of 10 drugs/day (excessive polypharmacy) according Jyrkkä et al. [22] because of the high number of drugs used among this population (mean 8.6), thereby depicting drug use better.

Comorbidity was determined using a modified version of the Functional Comorbidity Index (FCI) [23]. The geriatricians identified diagnoses (rheumatoid arthritis and other inflammatory connective tissue diseases, osteoporosis, diabetes, chronic asthma or chronic obstructive pulmonary disease (COPD), coronary artery disease, heart failure, myocardial infarction, stroke, depressive disorder, visual impairment, hearing impairment, Parkinson’s disease or multiple sclerosis and obesity) based on primary care medical records and the investigator determined FCI index, where a higher FCI sum score indicates greater comorbidity.

Frailty status was assessed using the aCGA [13], which is based on the full CGA and developed to detect frailty in vulnerable population. The aCGA can detect problems that may otherwise go undetected, such as difficulty in bathing, dressing and shopping, reduced cognition or depression symptoms [13]. These problems are common among home care clients, although poorly detected. The validated aCGA consists of 15 questions including 3 questions on ADL; bathing, toilet use and transport, 4 questions on IADL; shopping, food preparation, housekeeping and laundry, 4 questions on the MMSE; calculation, reading, writing and image copying and 4 questions on the GDS-15; “Do you feel that your life is empty?”, “Do you feel happy most of the time?”, “Do you often feel helpless?” and “Do you feel pretty worthless the way you are now?”) (Table 1) [13].

The prevalence of frailty was assessed in all three aCGA domains: functional status, cognitive status and depression. The cut-off value for functional status was ≥1, for cognitive status, ≤ 6 and for depression, ≥ 2. All three domains were unified into one frailty variable by which the participants were categorised into two groups. The frailty variable indicates frailty with a positive score in the ≥1 domain. The frailty variable was also calculated with a positive score in the ≥2 domain [14]. Statistical comparisons between the groups were done using a chi-square or t-test and a Mann-Whitney U-test for non-parametric variables, with a p-value of 0.05 considered significant. The results were expressed as frequencies with percentages or means with standard deviations (SD). Univariate and multivariate regression analyses were performed to identify demographical, clinical and functional factors associated with frailty. The analyses were done using SPSS version 21.0 (SPSS, Inc., Chicago, IL).