This is the first study in Germany to investigate the impact of intensive pharmaceutical care (PC) versus standard care in a larger patient population with ischemic stroke. This study demonstrates that PC is able to prevent deterioration of most HRQoL parameters over a 12 months period. Our findings indicate that PC as performed in the hospital and in the community setting is feasible and has a clear benefit and positive impact on patient's HRQoL. Moreover, this study is a step forward towards monitoring the implementation of PC within hospitals and within the community setting and towards evaluating the role of pharmacists within a specific therapeutic team. We found that a PC concept stabilizes the HRQoL in several scales as a result of an intensified involvement by pharmacists.
In a comparison of the subscales and the summary measures in both groups upon study entry there were no significant differences, thus providing an optimal starting point for measuring the impact of PC on HRQoL. After 12 months the comparison of the HRQoL of the patients who received intensified PC with those who received standard care showed that the vitality of the patients in CG was significantly lower than of patients in IG. The other subscales and the summary measures showed lower, but not significant, HRQoL scores in CG indicating inferior HRQoL as compared with the patients in IG. Due to ethical considerations community pharmacies were not restricted from delivering PC as it did not seem appropriate to keep the control pharmacists from providing PC in the case that DRP became apparent. For this reason the comparison may not to be statistically significant.
Most importantly, there was a significant deterioration of 7 of 8 subscales as well as both summary measures of the HRQoL over the 12 months period in the CG. That means that patients who do not receive intensified PC have a higher chance of deteriorating HRQoL. For the IG, there was only a significant deterioration in bodily pain at follow up. The deterioration of the HRQoL in patients with cerebrovascular diseases has been reported in previous trials in several parameters [3, 13, 14]. The quality of the patient's recovery can be evaluated on the basis of the activity of daily living (ADL), social activities, and return to work. Independently of ADL significant deleterious effects in HRQoL result from continuing loss of function due to the stroke. The deterioration in the HRQoL can be caused by the presence of anomalous perception and dissatisfaction in patients with minor disability levels that were incompletely restored after the stroke. Thus, the consequences of mild to moderate stroke can affect all dimensions of HRQoL despite the patient achieving full independence as measured in ADL .
By participating in this study the HRQoL of the patients in the IG was stable in several subscales and both summary measures indicating that the HRQoL can be maintained by intensified PC. Several studies have also demonstrated that PC may have a positive impact on the patients' HRQoL for example in patients with asthma or hypertension [8, 15, 16]. Graesel et al. investigated the impact of an intensified transition concept between inpatient neurological rehabilitation and home care of patients with stroke. The intensified transition concept included four additional elements: a psycho-educational seminar for family carers; an individual training course for carers; therapeutic weekend care at home before discharge; and finally, telephone counselling three months after discharge. The results show that this concept did not lead to any significant benefit, neither to functional status nor to HRQoL . This finding may indicate that the standard transition already contains effective elements of preparation for home care and that therefore no further decisive advantage can be attained. Otherwise the number of patients (n = 62) as well as the observation period (n = 6 months) in the study was limited, thus excluding long-term effects.
The HRQoL of the patients in the CG decreased significantly in the subscales and in both summary measures. The deterioration of the physical role was significant while physical functioning did not significantly deteriorate, which may indicate that patients perceived a decline in their physical conditions.
There is a large difference in the frequency of diabetes mellitus in both groups (IG: n = 12 (= 13.3%); CG: n = 39 (= 23.6%)). Diabetes mellitus has detrimental effects on a range of health outcomes including HRQoL; for example people with diabetes have lower scores at the SF-36 . Thus, the higher proportion of patients with diabetes in the CG may possibly cause the deterioration of HRQoL.
It should be pointed out that a considerably higher proportion of subjects in the IG than in the CG refused to participate. Those patients who refused to participate had comparable demographic data, but may have had more medical, social or psychological problems which could have affected the HRQoL. The difference in HRQoL after one year could be due to selection bias.
The mortality rate in both groups seems to be remarkably low (IG: n = 1; CG: n = 5); which is due to the fact that patients with a Barthel index of less than 30 points at discharge from hospital and/or who live in a nursing home were excluded from this study. These exclusion criteria were defined because only patients mobile enough to go the pharmacy can be given PC.
An important difficulty in the analysis of HRQoL in stroke patients is that few studies use the same measurement tools and the same period of follow up. Furthermore, there are no data in the literature from the patients' HRQoL from before the cerebrovascular event.
The limitations of the present study include the fact that no blinding or randomization could be performed. A target number of 116 participants were calculated for each group. However, only 90 patients could be recruited for the IG, as many patients refused to participate. The time of enrolment was extended from 12 to 20 months and then stopped. Moreover, patients who were lost to follow-up were not evaluated for HRQoL, because no follow-up data were available. It should be pointed out that, due to ethical considerations, it did not seem to be appropriate to keep the control pharmacists from caring in the case that DRP became apparent. Furthermore, we did not have any information on socioeconomic status, financial resources or caregiver support in patients of either group, which could have influenced the outcome of the patients.