Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37
http://www.biomedcentral.com/1471-2261/10/37
Page 7 of 8
undertaken, the model will be populated with costs of
the therapies prescribed in each strategy and acute and
long term costs of further cardiovascular events.
In order to explore uncertainties in the analyses,
deterministic sensitivity analysis is proposed to test the
robustness of the model when varying key model para-
meters and structural assumptions. Probabilistic sensitiv-
ity analysis will be undertaken to incorporate the
uncertainty around parameter values and quantify the
overall decision uncertainty, and inform whether further
research is required.
Qualitative Study
Sampling
A purposively selected sample of 30 patients (10 each
from intervention and control and 10 patients who
declined the invitation to participate), and 20 healthcare
professionals (Health Care Assistants (HCAs), nurses
and GPs) will be selected for interview.
Patient Sampling Strategy Sampling will be carried out
on the basis of study arm (intervention or control), with
a further group of people who did not consent to parti-
cipate in the trial also being invited to attend for an
interview. Within each group, participants will be
selected on the basis of: age (tertiles); socio-economic
status (using IMD scores); number of different classes of
medications; and whether they have had a stroke or a
TIA. A researcher will randomly select patients from
these categories, ensuring that similar numbers of
patients in all categories are included.
Health care professionals sampling strategy Practices
participating in the study were selected to ensure a
range of practice characteristics are represented, includ-
ing practice size and socio-economic status. A
researcher will randomly select 20 practitioners from
these practices and send an invitation letter and infor-
mation sheet, inviting them to take part in an interview.
Patients and practitioners who fail to respond to invi-
tation or who do not wish to participate will be replaced
by another patient/practitioner with similar characteris-
tics. This process will continue until theoretical satura-
tion is achieved and interviews cease.
Interviews
Semi-structured, face-to-face, in-depth interviews will be
carried out in patients’ own homes or in other suitable
locations, or with healthcare professionals in the sur-
gery, and will be conducted by a researcher trained in
qualitative interviewing techniques. Fully informed con-
sent will be obtained from interviewees at the start of
the interview, and a consent form signed. An interview
topic guide will be used (see additional file 1) which will
then be modified and refined during the first interviews.
Each interview is expected to last between 60 and 90
minutes, and will be audio taped and transcribed
verbatim.
Data Analysis
Data collection and analysis will be iterative, occurring
as data collection in the interviews proceeds. Data will
be analysed using a thematic approach, based on the
principles of ‘Framework’ analysis [25] and using Frame-
work software. The research team will actively contri-
bute to the development of the analysis and conceptual
framework and their different disciplinary and profes-
sional backgrounds will maximise theoretical sensitivity
[16].
Time plan
Patient recruitment began in July 2009 and is planned to
continue until February 2011. By October 2009, 23
patients (4% of target) have been recruited into the trial.
Interviews will commence in January 2010 and are
expected to be completed by February 2011.
Discussion
The results of this trial and the health economic analysis
will provide insight into the role of intensive blood pres-
sure targets for people who have had a stroke or TIA. If
the trial is negative and a significant difference in systo-
lic blood pressure is not observed between the two
study arms, then the embedded qualitative work will be
of importance to determine why low blood pressure tar-
gets did not lead to lower blood pressure. If the trial is
positive, then the critical question remains as to whether
striving for lower blood pressure targets is appropriate.
If we observe no difference in adverse event rates or
quality of life between the two arms of the trial, then it
is likely that aiming for lower blood pressure targets will
be worthwhile, given the benefits of reduced stroke risk
that were observed in the PROGRESS trial [5]. This will
be tested by our economic analysis. If, on the other
hand, the lower blood pressures are at the cost of higher
adverse event rates, then it may be that a further trial
powered to detect differences in clinical end-points will
be required to guide clinical practice.
Additional material
Additional file 1: Interview guides. This file contains a copy of the
interview guides for Patients and Health Care Professionals.
Acknowledgements
We would also like to acknowledge the help and support provided by the
Stroke Research Network, the Primary Care Research Network and the West
Midlands Research Consortium (MidReC).
This programme receives financial support from the National Institute for
Health Research (NIHR) Programme Grants for Applied Research funding
scheme. The views and opinions expressed in this editorial are those of the