Methods
Patients admitted to hospitals in North Nottinghamshire with acute
stroke (first or recurrent) were eligible for inclusion. They were
identified from medical admission wards, a specialist stroke ward,
and rehabilitation wards. The medical notes of patients were checked
to confirm the presence of stroke. Other patient details pertinent to
the study were also obtained from the notes. Patients were excluded
if they were unconscious on admission to hospital, were admitted
from a nursing home, lived outside a 30-mile radius of the research
center, had a main medical problem not related to stroke, or had a
prestroke Barthel Index15 score of 10. Those who had a planned
discharge to a nursing home were also excluded.
No limit was set on elapsed time between stroke and recruitment,
although the aim was to reach patients within 4 weeks of the stroke.
Patients were included if they had no identifiable caregiver. The
closest caregiver was defined as “the person, other than a healthcare
professional, perceived by the patient or family as normally being
most responsible for day-to-day decision making and care.”13
Patients were approached by an assistant psychologist, usually in
hospital, who explained the nature and purpose of the study.
Consenting patients and caregivers were randomized by the FSO by
a telephone call to an independent center, which held a computergenerated
random sequence. Patients and their respective caregivers
were allocated to receive the FSO service (intervention group) or to
a standard care control group.
Patients and caregivers allocated to the intervention group were
contacted by the FSO within 2 weeks of recruitment. The initial
contact by the FSO was undertaken to make introductions; to make
patients and caregivers aware of the support being offered; to provide
an information pack produced by the Stroke Association, UK; and to
identify unmet needs in terms of information required about stroke
and other concerns about adapting to disability such as benefit
entitlement and emotional support. The FSO made an initial contact
with the patient in hospital, attended case conferences, and acted as
a liaison with the rehabilitation team regarding discharge from
hospital. The FSO also made contact with caregivers to assist them
with preparations for patients’ discharge from hospital. After discharge,
the FSO visited patients and caregivers at home to discuss
problems, to offer information and emotional support, and to direct
them to appropriate services. The content and frequency of the visits
were left to the discretion of the FSO and the specific needs and
requests of patients and caregivers, but the service provided was
based on that of other FSO services.16
The intervention was provided for up to 9 months after recruitment.
Patients and caregivers were asked to complete and return
postal questionnaires at 4 and 9 months after recruitment. If they had
difficulty completing questionnaires, an independent assessor who
was blinded to the group allocation visited them at their place of
residence to assist with completion.
The outcome questionnaire included the General Health Questionnaire
1217 (GHQ-12) to assess patients’ mood. Personal self-care was
assessed by the Barthel Index,15 and instrumental activities of daily
living were assessed by the Nottingham Extended Activities of Daily
Living scale (EADL).18 A questionnaire was designed specifically to
assess patients’ perceived knowledge about whom to contact if they
required any assistance in obtaining information on stroke, reducing
the risk of a future stroke, benefits, community services, practical
help, and emotional support. A Likert scale was used that consisted
of responses that ranged from 0 (“I have no idea whom to contact”)
to 3 (“I know whom to contact and have received such information”).
A questionnaire was also designed specifically to assess patients’ and
caregivers’ satisfaction with information on stroke, recovery, benefits,
reducing the risk of a future stroke, community services,
practical help, emotional support, and overall experience with
hospital and community services. The Likert scale ranged from 0 (“I
am very dissatisfied”) to 4 (“I am very satisfied”). Test-retest
analyses over a month showed these questionnaires to have satisfactory
reliability (rs0.84). Caregivers received a questionnaire comprising
the GHQ-12, the same measures of knowledge and satisfaction,
and the Caregiver Strain Index.19 This is a scale designed to
assess the psychological impact of providing care.
These measures were repeated at 9 months after recruitment.
Caregivers’ own independence was also assessed with the EADL at
9 months. Except for the GHQ-12 and Caregiver Strain Index, higher
scores indicate better outcome. When patients were unable to answer
questions, the EADL and Barthel Index scores were obtained from
discussions with their informal caregiver or, when appropriate, with
allocated nurses in hospital or residential care homes.