Table 2 First Pilot: Selecting frail patients

Background:

Measuring frailty is subject to debate and various operational definitions were proposed [15]. For our purpose we sought a valid easy administrable self-report instrument.

Objective:

To determine the feasibility of multidimensional frailty screening instruments that could be sent by mail.

Methods:

After a literature search three multidimensional screening instruments were selected and tested in one general practice among all 75+ patients: 1) VES-13, 2) Strawbridge's frailty screen, and 3) COOP-WONCA charts. Feasibility was expressed in percentage complete item response [20–22]. Our goal was to identify the worst quarter. This point of departure was based on studies by Fried and Rockwood who reported between 20–30% of 75+ people to be frail according to their measures [14,17].

Results:

Of 116 patients 85 (81%) agreed to participate and 69 actually returned the questionnaire. The complete item response on the COOP-WONCA, Strawbridge screen, and VES-13 were 87%, 60% and 56% respectively. In order to identify a quarter of persons with the worst health on the COOP-WONCA, all persons were selected who scored in the worst quartile of at least two of the six charts (overall health ≥4; physical fitness ≥5; changes in health ≥4; daily activities ≥4; Feelings ≥3; social activities ≥3). This resulted in 23 persons who were further assessed at home by the RAI-HC. 90% had at least one chronic disease, two thirds had at least one ADL limitation, 60% had depressive symptoms (CESD>16) and 30% had cognitive impairment (MMSE<24) [37].

Conclusion:

The COOP-WONCA was the most feasible screener. Our selection rule identified a frail group. The geriatric assessment identified new potentially treatable problems.