Motivational Interviewing Early After Acute Stroke
A Randomized, Controlled Trial
Background and Purpose— The purpose of this study was to determine whether motivational interviewing, a patient-centered counseling technique, can benefit patients’ mood 3 months after stroke.
Methods— A single-center, open, randomized, controlled trial was conducted at a single hospital with a stroke unit. Subjects consisted of 411 consecutive patients on the stroke register who were over 18 years of age and who did not have severe cognitive and communication problems that would prevent them from taking part in an interview; were not known to be moving out of the area after discharge; and were not already receiving psychiatric or clinical psychology intervention. All patients received usual stroke care. Patients in the intervention group received 4 individual, weekly sessions of motivational interviewing with a trained therapist in addition to usual stroke care. The primary outcome was the proportion of patients with normal mood at 3 months poststroke measured by the 28-item General Health Questionnaire (normal, <5; low ≥5) using a mailed questionnaire.
Results— Eighty-one of 207 (39.1%) patients in the control group and 100 of 204 (49.0%) patients in the intervention group had normal mood at follow up. A significant benefit of motivational interviewing over usual stroke care (OR: 1.60, 95% CI: 1.04 to 2.46, P=0.03) was found.
Conclusion— Our results suggest motivational interviewing leads to an improvement in patients’ mood 3 months after stroke.
There is a strong relationship between early psychologic problems and the rate and extent of recovery after stroke. Depressed patients with stroke lack the motivation to participate in rehabilitation, making less progress,1 staying in the hospital longer,2 failing to engage in leisure pursuits and social activities,3 and surviving for less time.4 These negative outcomes could be related to a failure to adjust or adapt to the effects of the stroke.
Several studies have attempted to address psychologic problems directly using either pharmacological5 or conventional cognitive-behavioral therapies.6–9 However, the results thus far have failed to give a clear message10 as have studies aiming to reduce psychologic problems indirectly, for example, by improving social support.11 Alternative approaches to addressing psychologic issues after stroke need to be explored.
Motivational interviewing is a specific talk-based therapy originally developed to help people with addictions.12 More recently, it has been used successfully with a wide range of health problems characterized by poor motivation and the necessity to make some form of health behavior change.13
Our aim is to intervene at an early stage after stroke using motivational interviewing to support and build patients’ motivation to adjust and adapt to having had a stroke. Through the use of motivational interviewing techniques, patients will be helped to recognize the importance of making psychologic adjustments and practical adaptations. Subsequently, they will be able to develop confidence in their ability to adjust and adapt and to identify realistic personal goals for their recovery. This will address low expectations14 and provide the psychologic impetus to engage in rehabilitation and improve recovery.15
This was a single-center, open, randomized, controlled trial. Ethical approval was obtained from the local research ethics committee. This study is registered as an International Standard Randomized Controlled Trial, no. ISRCTN54465472.
This study was conducted at a hospital serving an urban population of approximately 250 000 people.
All patients with suspected acute stroke admitted to the hospital are identified on a stroke register. Patients admitted to the hospital between July 2002 and January 2005 and who met the inclusion criteria were invited to participate in the study. Inclusion criterion was patients over 18 years of age. Exclusion criteria were severe cognitive and communication problems preventing them from being interviewed; known to be moving out of the area after discharge; and already receiving psychiatric or clinical psychology intervention. Patients who agreed to participate and provided written informed consent were randomized between days 5 and 28 poststroke.
A research nurse randomized the patients (one-to-one ratio) to either usual stroke care (control) or motivational interviewing (intervention) using a minimization computer program on a personal computer. Minimization was based on the following variables: age (<65 years; ≥65 years); sex (male; female); baseline Barthel as a marker of dependence (severe, ≤10; moderate, 11 to 17; mild/no, 18 to 20); and location (stay; no stay on acute stroke unit). To randomize a patient, data from each variable are entered in the program, which then allocates group membership. The same nurse then assigned patients in the intervention group to one of 4 therapists using an opaque sealed envelope so that no therapist’s caseload exceeded 6 patients per week. The therapist was informed of the patient’s name and location, arranged the first appointment, and subsequently, 3 further weekly appointments. If the patient was in the hospital, they were interviewed in a private room; if discharged, they were interviewed as an outpatient in a private room in the hospital or in their own home. One of the therapists left the project in April 2003 and was not replaced. The therapists were not involved in the initial assessment of the patients, the randomization procedure, or the assignment of patients to therapists.
Between days 5 and 7 poststroke, routine clinical data were recorded on age; sex; history of stroke or transient ischemic attack; type (ischemic stroke or intracerebral hemorrhage) and class16 of stroke; and function (Barthel17). Patients who consented to participate completed a baseline assessment with a research nurse between days 5 and 28 at the time of randomization. The assessment included mood (28-item General Health Questionnaire19 [GHQ-28]); beliefs and expectations of recovery (Stroke Expectations Questionnaire14 [SEQ]); and function (Barthel).
Usual Stroke Care
Patients in the control group received usual medical, nursing, and therapy input, including inpatient care and discharge planning, through regular multidisciplinary team meetings and a stroke review clinic appointment at 3 months poststroke. There is no clinical psychology service for patients with stroke; if a patient has psychologic problems, it is reported to the clinician in charge of their care who reassesses the patient and refers them to a psychiatrist if appropriate.
Patients in the intervention group received up to 4 individual sessions of motivational interviewing, one per week, with the same therapist; sessions took place in a private area and lasted between 30 and 60 minutes. In the initial session, the therapist set the agenda so that the patient talked about adjustment to having had a stroke and their current concerns, for example, relating to physical, functional, or social support issues. Therapists elicited patients’ personal, realistic goals for recovery and their perceived blocks to attaining these goals. By working with patients’ dilemmas and ambivalence and through supporting and reinforcing optimism and self-efficacy, therapists enabled patients to identify their own solutions.
Therapists received 4 days of training in motivational interviewing by a specialist followed by up to 10 practice sessions until competent and confident of the technique. The therapists were supervised by a clinical psychologist through team meetings and one-to-one clinical supervision sessions on a monthly basis with additional informal support throughout the study. Therapy sessions were audio recorded to allow the therapist to reflect on and prepare for the next session. The therapists had backgrounds in nursing and psychology (nonclinical).
The quality of the application of motivational interviewing was assessed by analyzing a purposive sample of 60 sessions from different patients. The sample was balanced across therapist, session number (ie, 1 to 4), and when, over the 31-month recruitment period, the patient was recruited. A clinical psychologist reviewed the content of 20 therapist utterances around the midpoint of each session using a structured evaluation tool, “Motivational Interviewing Skill Code (version 2).”18 Utterances rated motivational interviewing-consistent included: open questions, reflections, advise with permission, affirm, emphasize control, reflect, reframe, and support. Utterances rated motivational interviewing-inconsistent included: advise without permission, confront, direct, raise concern without permission, and warn. The percentage of motivational interviewing-consistent utterances was determined [total MI-consistent/(total MI-consistent plus MI-inconsistent)×100].
At 3 months poststroke, patients were sent a mailed questionnaire. In addition to the outcome data detailed subsequently, the questionnaire also recorded place of residence, availability of significant other, social support, further stroke events, or other new comorbidity. If the patient did not return the questionnaire within 2 weeks, they were telephoned by a second research nurse, blind to group allocation, and given the option of declining, having a further questionnaire mailed, completing the questionnaire over the telephone, or receiving a home visit to assist with completion.
The responses on returned questionnaires were checked for completeness and any sign of emotional distress (GHQ-28 >14 and/or responding positively to any items indicating suicidal tendency). For any missing items, the patient was contacted and asked for their response. Data entry was by a person blind to group allocation. For patients showing signs of emotional distress, the hospital clinical team or general practitioner was contacted.
Mood was assessed with the GHQ-28. Secondary outcomes included depression screen (Yale20); function (Barthel); and beliefs and expectations of recovery (SEQ).
Mood (GHQ-28) was dichotomized into “normal” (<5) or “low” (≥5). The Yale was treated as dichotomous (“yes” or “no”), the Barthel as categorical (18 to 20, good; 11 to 17, moderate; 0 to 10, poor; dead) and the SEQ as scale data. For the SEQ, we examined beliefs (SEQ Help subscale), expectations (SEQ Happen subscale), and the difference between expectations and beliefs.
We chose a sample size of 200 per group. In a previous study, 44.4% of patients suitable for entry into this trial had GHQ-28 scores of below 5 at 3 months.21 We deemed a 15% between-groups difference in percentage of patients with normal mood to be clinically relevant. We calculated that 187 patients per group were needed, which we inflated to 200 to allow for 5% to 10% dropout.
Data analysis was by intention-to-treat and was undertaken according to an analysis plan drawn up by the trial statistician in conjunction with the trial steering group. The plan was peer-reviewed by 2 stroke experts (PL and NL) and was posted on the Clinical Practice Research Unit web site before any analysis being performed. No interim analyses were undertaken. The analysis of the primary outcome was performed blind to group membership.
The effects of intervention on mood and depression screen were analyzed using logistic regression and its effect on the Barthel was analyzed using general ordered regression. The effects of the intervention on the SEQ beliefs, expectations, and the difference between SEQ beliefs and expectations scores was analyzed using general linear modeling. In each case, the analysis was adjusted using the baseline value of the outcome variable and the factors used for minimization. For any missing data required for statistical adjustment, hot-deck imputation was applied. This involved imputing missing item or score total values for a patient from another comparable, randomly selected patient. This selection was done, with replacement, from those with an observed value of the item or score and within the same stratum formed by the combination of the design minimization factors as the patient with the missing value. For outcome mood, imputation was based on the “assumptions” that those who are dead had psychologic distress, ie, using worst case imputation, and that the mood status of those who had data missing for any other reason had not changed since baseline, ie, using last observation carried forward.
All analyses were performed using SPSS (version 13) for Windows or Stata (version 9). All inferential analyses used a 5% significance level; 95% CIs were also obtained using bootstrapping when any necessary parametric assumptions appeared not to hold.
During the recruitment period, 1388 patients were entered onto the stroke register and screened for the trial. Of these, 696 (50.1%) met the inclusion criteria. We consented 411 patients to our study (age: median 70, interquartile range: 61 to 77 years; 58.4% male).
Of the patients entering the trial, 207 were randomized to control and 204 to intervention. The groups were similar on most characteristics (Table 1). Patient flow through the trial is shown in the Figure.
Of the patients allocated to intervention, 146 (71.6%) received 4 motivational interviewing sessions, 16 (7.8%) received no sessions, and of the remainder, similar numbers received 1, 2, or 3 sessions (Table 2). Patients typically started the intervention between 2 and 4 weeks poststroke and many patients (80.1%) had at least one session at home (Table 2).
The Motivational Interviewing (MI) Skill Code ratings demonstrated high-quality motivational interviewing technique across all 60 interview extracts (mean % MI-consistent: 98.1, range: 80% to 100% with 51 being 100% MI-consistent and 9 being 80% to 93.3% MI-consistent).
The median length of stay was 13 days (interquartile range: 7–35 days) for control and 16 days (interquartile range: 7–43 days) for the intervention. By 3 months, 190 (91.8%) and 187 (91.7%) patients in the control and intervention groups, respectively, had been discharged from the hospital. At the 3-month follow-up assessment, 16 patients had died. Of the remainder, 339 completed the questionnaire (response rate 85.8%), 46 declined to complete the questionnaire, and 10 had been lost to follow up (5 administrative error, 5 diagnosis was not stroke).
Outcome statistics are detailed in Table 3. We detected a significant benefit of motivational interviewing over usual care on mood at 3 months (P=0.03; OR [normal mood]: 1.60, 95% CI: 1.04 to 2.46). A protective effect of motivational interviewing against depression screen (P=0.03; OR: 1.65, 95% CI: 1.06 to 2.58) was also found. No significant effect of motivational interviewing over usual care was found on function, including death (OR [mild dependence relative to worse outcome]: 0.97, 95% CI: 0.62 to 1.52, P=0.88; OR [mild or moderate dependence relative to worse outcome]: 1.13, 95% CI: 0.59 to 2.17, P=0.71; OR [alive relative to dead]: 3.70, 95% CI: 0.68 to 20.22, P=0.13).
No significant difference in effect was found between intervention and control on mean SEQ beliefs subscale score (P=0.64: 95% CI: −1.9 to 1.2) or mean SEQ expectations subscale score (P=0.75; 95% CI: −1.5 to 2.1), or on their difference (P=0.11; 95% CI: −0.2 to 1.8).
This study has shown that in a representative sample of hospitalized patients with stroke, motivational interviewing has a beneficial effect on patients’ mood using the GHQ-28 and on self-reported depression using the Yale screening tool. There was no demonstrable effect of motivational interviewing on either function or expectations of recovery. The number of deaths was not significantly greater in the control group than in the intervention group.
This was a methodologically robust study, meeting the criteria set by Knapp and colleagues in a review of strategies to resolve psychosocial difficulties after stroke.11 Our study was a trial of sufficient size to detect a difference between groups, and the sample size was informed through an appropriate power calculation. The study did not have an attention control, rather a usual care arm, and it has been suggested that the use of such a “care as usual” arm makes interpretation of the results easier.11 Trained therapists, under close supervision, provided a theoretically driven intervention that was delivered in a high-quality way consistent with motivational interviewing principles. Few published studies have examined the effects of talk-based interventions on patients’ mood after stroke, and none has shown a benefit on patient mood.11 There are a number of methodological issues that may explain the difference between the results reported here and other studies, including when the intervention is started, the intensity of the intervention, and the training of the therapists.
In not having an attention control group, there is the possibility that the positive finding on mood may be attributable to bias. That is, a placebo effect or a kind of social desirability could have elicited a more positive response from the intervention group than the usual care group. Two factors militate against this. First, it would make the assumption that the patients “knew” the primary outcome was the GHQ-28 and altered their response on this; there were no group differences for the other scales. Second, given the time between the last contact with the patient and the mailed questionnaire, we believe it is unlikely that the patients in the intervention group would exhibit social desirability. An alternative approach would have been to consent patients to follow up rather than randomization as has been done previously8; however, such a procedure poses an ethical dilemma.8 Perhaps the most methodologically sound approach would be to have both a usual care and attention control group; however, such an approach leads to a resource issue and furthermore, begs the question as to what an appropriate attention would be. There is a danger that an inappropriate attention control group, without a usual care group, could have a negative impact on the outcome resulting in the intervention appearing to be effective.
We started our intervention within the first 4 weeks poststroke, and no published studies have examined interventions started within the first 4 weeks after stroke. The studies tend to start on discharge,6 around 30 days,8 6 weeks9 or even as late as 11 months22 or 2 years after stroke.23 None have shown a benefit of talk-based therapy on their primary outcome measures, suggesting that the timing of the intervention may be an important factor. The timing of the intervention could be about preventing the development of mood problems, but this cannot be the only explanation.
Two studies have provided counseling, but the frequency of visits has been approximately 1 month apart. This has meant that, depending on the period of follow up, patients have received, on average, between 38 and 12 sessions.9 In the latter study, patients received a substantial number of sessions, but they were spread over 12 months. In the Life After Stroke study, patients received 10 sessions over a 10-week period, which did not affect the primary outcome of misery, but there was a small benefit on the Dartmouth COOP scale in favor of the intervention.6 Thus, the frequency of sessions may contribute to the success of the intervention.
The therapists in our study received 4 days of training in motivational interviewing by a consultant followed by 10 practice sessions and were closely supervised by a clinical psychologist. In the Life After Stroke study, the psychologic input was provided by psychology assistants who were closely supervised by a clinical psychologist.6 Although training has been given in other studies, the level of intervention-specific, psychology-based training does not appear to have been as high.8,9
This application of motivational interviewing to patients with stroke appears to have been appropriate, because it had a beneficial effect on patients’ mood. The lack of effect on expectations may be attributable to the inability of the SEQ to detect change and may be more a reflection of the person’s attitude to achieving desired outcomes. Given the positive effect of motivational interviewing on mood, its lack of effect on function was surprising given that mood is suggested to be related to function poststroke.24 However, because no other study had demonstrated the effectiveness of a talk-based therapy on mood 3 months after stroke, it is difficult to make suggestions about why a change in mood was not matched by a change in function. In the study reported here, there were fewer deaths in the intervention group, which may reflect previous research that has demonstrated the detrimental effect of depression on survival.4 Mortality was not a defined outcome of the study, and the total number of deaths was quite low, but this is worth exploring in the future. Further exploration is also required to identify the characteristics of patients benefiting from the intervention.
Motivational interviewing was effective with up to just 4 sessions, which would facilitate its introduction as part of usual stroke care. The technique of motivational interviewing is relatively straightforward to learn, so it may be possible to include training in motivational interviewing as part of the curriculum for healthcare professionals. From a service point of view, the lack of availability of staff, skilled in providing psychologic input after stroke, could be addressed in part by enabling healthcare professionals with the skills of motivational interviewing. The techniques involved in motivational interviewing can be used in the context of “normal” conversation. Therefore, further research is needed then on the impact of, training, for example, nurses, and therapists, to provide motivational interviewing within the context of their usual interactions with the patient. However, it should be emphasized that the therapists in this study received regular clinical psychologic supervision throughout.
Similarly, although healthcare professionals may use motivational interviewing as an intervention in its own right, further research may examine the use of motivational interviewing as an adjunct to other talk-based or pharmacological therapies.
We thank Helen Gardner and Irene McClelland for data collection; Denise Forshaw and Joanna McAdam for data input; Jenny Marsden for data cleaning and assistance with the analysis; members of the Steering Committee for their advice and suggestions; Peter Langhorne and Nadina Lincoln who reviewed and commented on the analysis plan; the staff of the Stroke Wards at the hospital; and the patients and caregivers who agreed to take part in the study. We also thank the reviewers for their comments and suggestions on the first submission of this article.
Sources of Funding
Funding for this study was provided by the NHS Executive, Department of Health North West, and Royal Liverpool and Broadgreen University Hospitals NHS Trust. The funders of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
- Received August 10, 2006.
- Revision received October 9, 2006.
- Accepted October 19, 2006.
Diller L, Bishop DS. Depression and stroke. Top Stroke Rehabil. 1995; 2: 44–55.
House A, Knapp P, Bamford J, Vail A. Mortality at 12 and 24 months after stroke may be associated with depressive symptoms at 1 month. Stroke. 2001; 32: 696–701.
Leathley M, Fall S, Sharma AK, Watkins C, Barer D. Life after stroke: multi-centre trial of psychosocial interventions after stroke [Abstract]. Cerebrovasc Dis. 2003; 16 (suppl 4): 70.
Lincoln NB, Flannaghan T. Cognitive behavioural psychotherapy for depression following stroke. Stroke. 2003; 34: 111–115.
Dennis M, O’Rourke S, Slattery J, Staniforth T, Warlow C. Evaluation of a stroke family care worker: results of a randomised controlled trial. BMJ. 1997; 314: 1071–1076.
Forster A, Young J. Specialist nurse support for patients with stroke in the community: a randomised controlled trial. BMJ. 1997; 312: 1642–1646.
Hackett ML Anderson CS, House AO. Management of depression after stroke. A systematic review of pharmacological therapies. Stroke. 2005; 36: 1098–1103.
Knapp P, Young J, House A, Forster A. Non-drug strategies to resolve psychosocial difficulties after stroke. Age Ageing. 2000; 29: 23–30.
Miller WR, Rollnick S. Motivational Interviewing: Preparing People to Change Addictive Behaviour. New York: Guilford Press; 1991.
Miller WR, and Rollnick S. Motivational Interviewing: Preparing People for Change, 2nd ed. New York: Guilford Press; 2002.
Watkins CL, Leathley MJ, Davies A. The Stroke Expectations Questionnaire (SEQ): identification of patients’ ideas about recovery. Clin Rehabil. 1998; 12: 173–174.
Miller WR, Moyers TB, Ernst D, Amrhein P. Manual for the Motivational Interviewing Skill Code (MISC) Version 2. New Mexico: Centre on Alcoholism, Substance Abuse and Addictions, The University of New Mexico; 2003.
Mahoney J, Drinka TJK, Abler R, Gunter-Hunt G, Matthews C, Gravenstein S, Carnes M. Screening for depression: single question versus GDS. J Am Geriatr Soc. 1994; 9: 1006–1008.
Watkins CL. The Effects of Patient’s Expectations on the Rehabilitation Process [PhD thesis]. Liverpool, UK: University of Liverpool; 1999.
Chemerinski E, Robinson RG, Kosier JT. Improved recovery in activities of daily living associated with remission of poststroke depression. Stroke. 2001; 32: 113–117.