Quality of Life During and After Inpatient Stroke Rehabilitation
To the Editor:
We read with great interest the recent Stroke article by Hopman and Werner.1
Recently we performed a similar study on an Italian sample of stroke survivors. To evaluate the effects of the rehabilitation, we performed a multiperspective study through traditional parameters and validated Health Related Quality of Life (HRQoL) measurement. We prospectively evaluated all outpatients, who were admitted for physical rehabilitation at the Don C. Gnocchi Foundation of Rome, between June and December 2002.
Thirty-four stroke survivors (mean age, 66; SD, 11.6; range, 38 to 88; 18 males and 16 females; duration of disease: mean, 4 years; SD, 3.4) were evaluated by means of (1) clinical assessment, (2) self-administered questionnaire for general health (SF-36),2–4 and (3) standardized disability measurements (Functional Independence Measure, Barthel Index, Deambulation Index). The HRQoL evaluation was performed at baseline (T0) and at the end of rehabilitation (T1). Furthermore, patients were contacted after 2 to 6 months (mean, 3.5 months; SD, 1.6) of treatment, and a telephone re-evaluation on quality of life was performed through telephone administration of the SF-36 (T2).
The Table summarizes SF-36 scores (mean and standard deviation) at baseline (T0), changes in SF-36 from baseline to the end of rehabilitation (gT1−T0: evolution gradient=end rehabilitation score−initial score), and from the end of rehabilitation to follow-up (gT2−T1: evolution gradient=follow-up score−end rehabilitation score).
After rehabilitation (T1), we observed an improvement in all but 1 (bodily-pain) of the SF-36 subscores, although only 2 were statistically significant (Table). In the same way, the SF-36 main scores (mental composite [MCS] and physical composite score [PCS]) improved, but only PCS was statistically significant.
At the follow-up evaluation (T2), 2 subscores (Role Physical and Vitality) and 1 main score (MCS) showed statistically significant improvement, and 4 subscores showed nonsignificant improvement (General Health, Social Functioning, Role Emotional, Mental Health). However, no significant worsening of Physical Functioning and Physical Composite score was observed. The Bodily Pain continued to worsen.
Overall results are similar in the Italian and Canadian studies. The main difference was observed at follow-up (T2), where the Canadian sample worsened more than the Italian sample, but this difference could be because our sample included only outpatients.
One of the most important advantages of using standardized validated measurements is the possibility to compare the outcomes of samples enrolled in different centers or in different countries.
It would be interesting to know the SF-36 data (mean and standard deviation) at the baseline. In fact, we cannot compare the Canadian sample with the Italian sample because of the absence of rough SF-36 data at baseline (only a figure is reported). Moreover, it would be interesting to know the duration of disease of the Canadian sample. In fact, these data could be one reason for the different behavior of the 2 samples after rehabilitation and mainly at follow-up.
We believe that these kinds of studies and comparisons between samples are needed to better understand the diseases we deal with and to improve therapeutic approaches.
I would like to respond to the letter by Drs Padua et al, which discusses a recently published article assessing quality of life during and after inpatient stroke rehabilitation.1 It is interesting to note that the rehabilitation process had a comparable, positive impact in both the Italian and the Canadian samples, but the Canadian patients worsened more than the Italian patients during the months following rehabilitation.
The authors are correct in suggesting that one reason for this difference was that fact that the Italian sample included only outpatients, while the Canadian sample consisted of inpatients. The other reason for this difference, which is probably the more important of the two, is the time of disease duration. In the Italian sample, the mean disease duration was 4 years (SD 3.4 years) while in the Canadian sample, disease duration was measured in days. The time between the onset of stroke and admission to rehabilitation was 25.1 days (SD 22.6 days).
The Table contains the SF-362 scores at baseline, which the authors may find useful for comparison purposes. The differences between the groups at baseline are evident. For example, Canadian Physical Function domain scores were much lower than the Italian ones at baseline (11.3 versus 33.7), which would explain why the Canadian sample had a larger improvement (17.3 versus 5.0) during rehabilitation. The Role Emotional domain was much higher in the Canadian sample at baseline (64.7 versus 20.5), but that is likely due to the fact that the patients had not returned home to resume their normal role. It is also of interest to note that the Canadian Role Emotional scores drop significantly in the 6 months following discharge (23.8 points). In fact, if one does the calculations to determine the Role Emotional score at the time of the follow-up, the 2 groups are similar (41.6 for the Italian group and 47.3 for the Canadian group).
The authors point out that one of the advantages of using standardized and validated instruments is the opportunity to compare outcomes in different samples and even in different countries. I agree, and I believe that this is an excellent example of such an opportunity.