Scientific Papers

Lifestyle counselling as secondary prevention in patients with minor stroke or transient ischemic attack: a randomized controlled pilot study | Pilot and Feasibility Studies

Description of Image

We conducted a parallel group randomized controlled pilot and feasibility trial to test the feasibility of implementing an individual face-to-face health behavioural counselling with post-discharge follow-up sessions against usual care in a fast-track stroke care setting. A protocol article outlining the study design and procedures has previously been published [15].

Setting and participants

The target population was patients hospitalized with recent minor stroke or TIA who were discharged home. We included patients admitted to the Department of Neurology at Nordsjællands Hospital, Denmark, from October 2018 to January 2020, with minor stroke (ICD-10 I61, I63, I64; Scandinavian Stroke Scale 45–58) or TIA (ICD-10 G45.9) within the previous 7 days. In Denmark, all patients suspected of acute stroke or TIA undergo admission through a standardized fast-track patient pathway. This pathway includes rapid diagnostic procedures, treatment, and evaluation of risk factors for recurrent strokes. Typically, patients are hospitalized for 3 days. Patients were eligible if they were ≥ 18 years old, could consent to participation, and were discharged home. Exclusion criteria were severe communication barriers, inability to use a telephone, severe disability prior to the stroke (WHO performance status > 2; incapable of self-care and mobilized less than 50% of the day), active abuse of alcohol or narcotics, severe psychiatric illness (affective disease, dementia, schizophrenia), or inability to participate due to severe medical frailty. All new patients with suspected stroke or TIA were screened for eligibility by the primary researcher (J. L.). Potential participants were invited through verbal and written information before discharge. All participants gave written informed consent before participation.

A sample size of 40 randomized participants was chosen weighing the precision of the feasibility measures against the available resources. With this sample size, a theoretical attrition rate of 10% could be estimated with a precision of 3–24% (95% confidence) and a theoretical adherence rate of 5% with a precision of 1–17% (95% confidence) using the Exact method [16].


Baseline assessment

Demographic and health behavioural information was collected using standardized questions from the Danish National Health Survey [17], in addition with assessment of stroke severity [18, 19], prior health problems [20], vital signs [21], spirometry (FEV1/FVC) [22], and biochemistry (HbA1c and blood lipids) [4].

Randomization and group allocation

Participants were allocated to either intervention or usual care after baseline testing using a simple non-stratified 1:1 randomization. A computer-generated (using the rbinom-function in R [23]) randomization sequence was implemented into the Research Electronic Data Capture (RedCap) software [24], which secures concealment of future allocations and prohibits post hoc changes to the allocation. The participants remained in the allocation group for the entire study period.


In brief, participants randomized to the intervention arm received nurse-led targeted health behaviour counselling focusing on smoking cessation, physical activity, and adherence to preventive medication, in addition to usual care. When possible, we tested the participants aerobic capacity prior to the initial counselling using the Astrand-Rhyming cycle test [25]. The initial counselling session was provided face to face by the primary investigator before discharge from the hospital. The counselling employed a patient-centred approach [12] and was structured around the 5A’s model [10, 11] (Fig. 1) with the intention of engaging the participant in healthy behaviour and adhering to preventive medication to reduce their risk of recurrent cerebrovascular events. Participants in the intervention arm were issued a VivoFit activity tracker to count daily steps and aerobic walking time.

Fig. 1
figure 1

The 5A’s approach modified from Vallis et al. (2013) [10] and Sherson et al. (2014) [11], as proposed in Liljehult et al. (2020) [15]

Post-discharge counselling was provided by J. L. via telephone 4 and 8 weeks after discharge and aimed at maintaining motivation and adjusting goals and strategies if necessary. A standardized question guide was used to ensure the consistency of the telephone counselling with questions on overall well-being, persistence of symptoms or functional deficits, adherence to medication, side effects, physical activity, smoking, and use of health care services. The intervention is described in details in the published protocol [15].

Usual care

Participants in the control arm received standard care, which included a review of prescribed medication, and verbal and written encouragement to cease use of tobacco, diminish the intake of alcohol, be physically active to the extent possible, and eat a healthy diet (reduce the consumption of red meat and salt and increase the consumption of fish, fruit, and vegetables) [26].

Outcome measures

The primary outcomes were measures of feasibility: (1) eligibility including the eligibility rate (proportion of eligible patients compared to the total number of stroke patients), (2) acceptance including the study participation rate (proportion of patients accepting participation in the study), (3) demand and practicality (proportion of study elements that could be delivered according to the protocol), (4) adherence including the degree of adherence to the programme (proportion of attendance in follow-up sessions), (5) attrition (drop-out and drawn-out), and (6) implementation and integration evaluation the ability of delivering the intervention per protocol.

Secondly, we tested the feasibility of potential clinical outcome measures for use in future studies including the following: arterial blood pressure measured in accordance with US guidelines [27]; self-reported smoking, physical activity level (International Physical Activity Questionnaire-Short Form [28, 29]); adherence to medication (number of missed/consumed doses in the past 7 days); anthropometrics (body weight, waist/hip ratio); fatigue (Fatigue Assessment Scale [30]); long-term readmission with stroke, TIA, or ischemic heart disease; and fatality within 1 year.


Participants from both allocation groups were reassessed by the primary investigator (J. L.) in a hospital-based outpatient clinic 12 weeks after discharge. To evaluate the feasibility of potential outcome measures, we measured the following: arterial blood pressure, body weight, and waist/hip ratio, fatigue, smoking, physical activity, and adherence to prescribed medication at 12 weeks. Information on 1-year mortality and stroke recurrence was obtained from electronic hospital records.


We used the model for designing feasibility studies in preventive medicine proposed by Bowen et al. [31] to evaluate the feasibility. The eligibility rate was calculated as the proportion of all screened patients who met the eligibility criteria and were available for invitation. The acceptance rate was calculated as the proportion of invited patients who accepted participation, in contrast to patients who declined participation or cases where the invitation to participate was retracted before inclusion. Practicality was calculated as follows: (1) the proportion of included participants who completed the baseline measurements (baseline interview, SSS, spirometry) or (2) the proportion of participants allocated to the intervention arm who completed the initial elements of the intervention (VivoFit, face-to-face counselling, Astrand-Rhyming cycling test). The adherence and attrition rates were calculated as follows: (1) the proportion of participants in the intervention arm who participated in the follow-up telephone counselling (4 and 8 weeks after discharge) and used the VivoFit, in contrast to participants who were either drawn out or unavailable/non-adherent, and (2) the proportion of all participants who completed the 12-week follow-up and were included in the 1-year follow-up, in contrast to participants who were either drawn out or lost to follow-up. The implementation rate was calculated as the proportion of participants in the intervention arm who completed all elements of the original protocol or a modified version (omitting the Astrand-Rhyming cycling test) in contrast to participants who were non-adherent with one or more elements and the proportion of the participants in the control arm who completed the control protocol in contrast to participants who did not complete the 12 week follow-up. Specific progression criteria were not specified a priori — but the progression of the study was monitored through regular meetings with the programme steering committee at which data on recruitment, protocol adherence, and follow-up were reported and discussed in plenum. Confidence intervals of all proportions were calculated using the Exact method (Clopper-Pearson).

Data was collected in real time using RedCap electronic case report forms. Statistical analyses were carried out in R 3.3.1/R Studio 0.99.

Ethical considerations

The study was conducted in accordance with the Helsinki Declaration [32] including respect for the participants’ autonomy and right to informed consent. The participants were informed that participation was voluntary, and that further participation could be declined at any time without explanation.

The study protocol was approved by the Scientific Committee of the Capital Region (H-17040484) and the Danish Data Protection Agency ( VD-2018–306, I-6552). The study protocol was registered at (NCT03648957).

Description of Image

Source link