The study was conducted in 10 nursing homes in
southern Taiwan. Inclusion criteria were: (1) older adults
aged 65 years and over, (2) using wheelchairs for mobility,
(3) living in the facility for at least three months, (4)
cognitively intact (a score of eight or higher on the Short
Portable Mental Status Questionnaire) (Pfeiffer, 1975), and
(5) heavily or moderate dependency in their activities of
daily living (a score of 21–90 on the Barthel index). Exclusion
criteria were: (1) having severe or acute cardiovascular,
musculoskeletal, or pulmonary illnesses, or (2) suffering
from a spinal cord injury with no rehabilitation potential.
Based on the statistical software Sample Power 2.0, the
required sample size was 56 participants for each group
(power = 0.8; alpha = 0.05; R-square of covariate in medium
level 0.13; effect size in medium level 0.25).
Following approval by the Institutional Review Board of
the university hospital and the agency administrators, a
convenience sample of 127 participants was recruited and
randomly assigned to two groups based on the nursing
homes where they lived: five nursing homes in the
experimental group (n = 64) and five nursing homes in
the control group (n = 63). A total of 115 participants
completed the study (retention rate: 90.55%): nine parti-
cipants withdrew by the end of the three-month interval
(experimental group n = 3; control group n = 6) and three
more participants withdrew by the end of the six-month
interval (experimental group n = 1; control group n = 2). The
reasons for withdrawal included: deceased (experimental
group n = 3; control group n = 3) and discharged from
nursing homes (experimental group n = 1; control group
n = 5). To ensure the consistency of the intervention received
by the participants, one experimental group participant
with an attendance rate of less than 50% was excluded from
the data analysis (N = 114; experimental group n = 59;
control group n = 55) (Fig. 1). The characteristics of the
participants who withdrew from the study were similar to
those participants who remained in the study.