A chart review of management of ischemic stroke patients in Germany

Background Ischemic stroke (IS) poses physical, emotional, and economic burdens on both patients and the healthcare system in Germany. However, the management of IS is not well described, especially after hospital discharge. In this study, we aim to describe the management of IS at onset, admission, and during follow-up. Methods German general practitioners (GPs) (n=40) extracted data on patient characteristics, hospitalizations, discharge, and ambulatory care from both GPs patient databases and hospital letters. Descriptive analyses were conducted. Results The sample included 185 patients with a mean age of 70 years [standard deviation (SD)=11.7]. Most patients (63%) contacted the Emergency Medical Services, while 36% contacted their GPs. The majority of patients were hospitalized within 1 h from onset, and the length of stay was on average 14 days. Half of the patients (50%) were admitted to the stroke unit, and 16% of patients received thrombolysis treatment with 2 h (SD=2.6) of time to treatment. Of the admitted patients, 32% were discharged to their homes, while the remaining patients were discharged to nursing homes (16.2%) and rehabilitation centers (47.6%). During the 12 months follow-up, 22% of patients were re-hospitalized and patients visited their GP (11.7 times), psychologist or psychiatrist (9.5 times), and neurologist (2.2 times). Death rate after stroke event was 13%. Conclusion The rate of patients who received thrombolysis is lower than the optimal rate in Germany. More research is needed to determine the factors that could predict the utilization of thrombolysis treatment.

I n Germany, stroke affects 200,000 people annually and claims the lives of 66,000 people, making it the third cause of death. It is also the cause of 64,000 disabled patients annually with 15% of them requiring external assistance. The incidence stroke rate in Germany has been estimated to be approximately 160Á240 per 100,000 people annually, with half of the patients being older than 70 years of age (1).
Beyond the physical and emotional burden, stroke inflicts a considerable economic burden on both payers and society. In 2004, the total financial burden on a major source of healthcare funding in Germany, the statutory health insurance (SHI), was estimated at t7.1 billion. This cost, in an incidence-based, bottom-up and direct-cost-ofischemic-stroke study from the third-party payer's perspective, was mostly driven by outpatient treatment (40%; t2.8 billion), inpatient treatment (22%; t1.6 billion), rehabilitation (21%; t1.5 billion), and nursing (17%; t1.2 billion) (2). Further, the average cost for 12-day hospitalization during the acute phase was estimated to be t4,650 per patient (49% of direct costs) while total direct cost over 12 months was estimated at t9,452. From a societal perspective, the mean indirect costs were t2,014, with t1,344 attributed to days off work and t670 attributed to early retirement. Total associated stroke costs are impacted by patient characteristics, severity, and type of stroke (3).
Through information campaigns directed towards the general German population, prevention guidelines are well established in terms of raising awareness about stroke threats. The German Neurological Society (DGN) and the German Stroke Society (DSG) issued two joint guidelines entitled Primary and Secondary Prevention of Cerebral Ischemia in 2008 and Acute Treatment of Ischemic Stroke in 2009. The primary objective was to prevent cerebral ischemia or transient ischemic attacks (TIAs) in patients without previous cerebrovascular diseases. The second aim was to prevent recurrent cerebral ischemia (TIA or stroke) after the initial event (4). These guidelines provided practical recommendations on different risk factors' prevention, treatment, and suggestions for medication, but they lacked timelines for treatment duration. In addition to DSG and DGN, Germany also adopted the European Stroke Organization's (ESO) Guidelines for Management of IS and TIA in 2008.
Several studies have examined stroke management throughout both pre-hospital and acute phases. These studies evaluated interventions such as dissemination of knowledge about stroke to the general population, described mode of arrival, arrival time, and delays in arrival to the hospital and transfers to units within hospitals (5Á8). Results from these studies highlighted the effectiveness of dissemination of stroke knowledge, so that patients, who received an educational letter describing stroke symptoms and the importance of calling the Emergency Medical Services (EMS) were admitted to facilities of care at a faster rate than those who did not receive the letter (5). Education, knowledge of the time window, direct contact with an emergency facility after stroke onset, number of known symptoms, living with someone, and having a stroke history were factors that improved arrival time (7).
The management of IS post-discharge is not well described in the literature, as there are only few studies describing patient pathways during the follow-up period. The CERISE project, which compared service delivery after discharge in four European rehabilitation units (SRU) at 2, 4, and 6 months, reported that Germany had the highest discharge-to-home rates compared to Switzerland, Belgium, and the UK (9). Yet another study, which evaluated the effectiveness of semi-intensive stroke unit (SI-SU) compared with conventional care (CC) for patients with acute IS or TIA over a 20-month period, estimated the average length of hospitalization to be 11.2 days in SI-SU and 13.3 in CC (10). Although these studies had a follow-up period, they did not report in detail the pathways IS patients had followed to manage the disease.

Objectives
We sought to describe the management of IS at the pre-hospitalization phase, the acute phase (during hospitalization) as well as during 1 year of follow-up, relying on general practitioners (GPs) to extract all information about their IS patients.

Methods
This is a retrospective chart review study, in which 40 GPs were randomly recruited from a representative panel of German GPs. The GPs were responsible for completing a questionnaire about their patients, relying on medical files. GPs were asked to include patients, who met the inclusion criteria, i.e., stroke patients, who did not suffer from transitional stroke, hemorrhagic stroke, or stroke with undetermined reasons. Patients were selected in a chronologic order based on stroke onset date (starting with most recent case) during 2007Á2008. All data extraction took place between November 2009 and January 2010.
The questionnaire was designed to collect information by the GPs on IS patients with regards to: 1) patient characteristics (age, sex, occupation, living condition, medical history including stroke history, cardiovascular history, and previous ischemic event), 2) pre-hospitalization period (first health care contacts, mode of arrival), 3) hospitalization (type of hospital, first unit admitted to, units transferred to within hospital, imaging procedures, time between admission and treatment, type of treatment and length of stay in the different hospital units, severity scales), and 4) one-year follow-up after discharge (first place post-discharge, re-hospitalization, outpatient care, sick leave, and changing occupation). Information with regards to arrival method, type of hospital, first unit admitted to, units transferred to within hospital, imaging procedures (CT scans, MRI, ultrasound, or cerebral angiogram) and time between admission and treatment, type of treatment, and length of stay were documented in the hospital letter that was sent to the GPs from each treating hospital. Questionnaires were completed by GPs based on their electronic and paper medical records including hospital letters reporting the different units where patients stayed at hospital and a summary of the results of imaging tests. Means standard deviations, frequencies, and percentages are reported.
In an effort to validate data extractions, a random sample of five GPs was selected to confront study outcomes to GPs' experience, to identify GPs' level of insight on patient management in acute phase at hospital and post-discharge in rehabilitation centers, and to assess GPs' opinions when confronting our study results to the available evidence in literature.

Results
A sample of 40 GPs was randomly recruited, who extracted information for the 204 patients included in this study. Of these patients, 19 were excluded because complete 12 months of follow-up was not available.
The analyzed sample consisted of 185 patients, who were mostly retirees (74%) with a mean age of 70911.7 years. Based on the interviews, GPs saw at least 10 stroke cases per year with average age ranging between 70 and 75 years. The sample included slightly more male patients (53%), and the majority was living in an urban setting (72%). Although the vast majority of this population did not have a prior stroke event, 73.5% of them had cardiovascular diseases or stroke risk factors (Table 1). Most patients (63%) contacted the EMS while a nonnegligible number of patients first contacted their GP (36%). Most patients were transported to health care facilities via emergency vehicles ( Table 2).
Almost all patients (95%) received at least one imaging test, mainly CT scan without contrast injection (61.5%) and 37% of patients had cervical Doppler ultrasound examination. Patients admitted to hospitals were treated within an average of 292.6 h and 16.3% received thrombolysis treatment. Length of stay was on average 13.699.8 days (Table 3).
According to the GPs, admission to the hospital occurs within 1 h of the event. Length of stay is up to 2 weeks. In the random sample, two of the five GPs were not familiar with rt-PA treatment and could not differentiate it from heparinotherapy. All five agreed that imaging and transfer pathways in hospital units were well described in the hospital letter while clinical scales of severity were not used at hospitals or not reported in the hospital letter. However, scales are used in the rehabilitation centers.
Follow-up phase At discharge, 47.6% of patients were transferred to rehabilitation centers, 16.2% to nursing homes, and 31.9% were discharged to their homes (Table 4). Of the latter, 23.7% received specific home medical care. All five GPs estimated that 50% of patients end up at the rehabilitation centers after stroke and estimate that the rehospitalization rate is 10Á15%.
During the 12 months follow-up period, re-hospitalization was confirmed for 21.6% of the patients with average length of stay 15.3914.2 days. Further, IS patients were assisted by professional carers (53%), personal carers (28.9%), or both (18.1%). Of the employed patients, 59% had at least one reported sick leave and 5.9% changed their occupation.
Additionally, patients sought outpatient medical care by visiting their GPs (11.7 visits) psychologist or psychiatrist (9.5 visits), and neurologists (2.2 visits). Patients received on average 21.1 weeks of nursing care, 17.1 weeks  of physiotherapy, 11.0 weeks of occupational therapy, and 12.7 weeks of speech therapy.

Mortality
In our sample, 24 (13.0%) patients died in the first year following the stroke event for all causes. Of them, 10 occurred in the acute phase and 14 in the chronic phase. Stroke was attributed to 16 of the cases: four were due to the actual stroke event, five were due to stroke recurrence, and seven were due to stroke sequel. According to the GPs, death rate could range between 13 and 20%.

Discussion
This study identified and recruited a sample of IS patients through GPs in Germany. We described the characteristics of patients and their management during the acute phase, as well as the management and healthcare resource utilization post-discharge during a 12-month follow-up period. GPs usually manage IS patients after hospital discharge, and are well informed about their patients' acute phase via hospital letters, which describe the hospital stay, procedures performed, and the treatment administered. These attributes qualified GPs to identify and recruit patients and further to collect and report information about the pre-admission phase, during hospital stay, and post-discharge of IS patients. Patients were recruited in a systematic way and were representative of the general IS patients population and consistent with the literature; our sample had similar demographic distribution in terms of age, sex, and medical history, specifically atrial fibrillation and previous history of stroke (8,11).
The current DGN and DSG guidelines recommend using CT scan as the main apparatus for diagnosis of stroke and must be performed as soon as possible. Consistent with the DGN and DSG guidelines, almost all patients (94%) received imaging procedures; 85% received CT scans as a diagnostic procedure. In addition, our data are Table 3. Hospitalization and Healthcare resource utilization at acute phase  Length of stay at hospital was 13.6; 95% CI: 12.2Á15.0, slightly higher than in the Diagnosis Related Group (11 days) and in published studies (11.1Á13.3 days) (10,12).
Thrombolysis with alteplase is a highly effective evidence-based treatment recommended for IS and should be administered within a 3 h time window after onset of symptoms in patients with acute IS (13). In Germany, alteplase is the only approved thrombolytic for patients with acute IS and the current knowledge about this treatment is still lacking. Generally, not all patients who qualify for this treatment receive it, resulting in lower rates of utilization. A survey by the European Stroke Facilities estimated alteplase rates to be as low as 4.76% of all IS patients in 178 participating hospitals in Germany and Austria (14). This rate was much lower than the estimated optimal rates of alteplase estimated in a community-based study. This study included patients with IS in the Netherlands, and estimated the number of patients eligible for rt-PA therapy if delays between onset and admission to hospital are to be avoided. Delays were defined as when the duration from onset to hospital admission exceeded 1.5 h. Doctor delays were defined as doctors taking more than 20 min to reach the patient. The study estimated that if delays can be avoided, the optimal proportion of acute IS patients to be treated with alteplase, can be as high as 24% in a given area (15). In the present study, the rate of thrombolysis was 16%. This estimation was possibly overly optimistic. Post-survey, GPs received more information about alteplase and were asked to redefine who of their patients received alteplase treatment. The alteplase rate decreased from 16 to 8% but remained substantially higher than the reported rates in the German population (4.7%). The high alteplase rate in this study can be explained by lack of knowledge about the treatment amongst the GPs, who extracted the information, even after having received more educational material about the treatment. Alternatively, the GPs did not differentiate between alteplase and other treatments such as heparinotherapy.
This method had several limitations. First, some GPs may have relied on their memory to recruit patients rather than using their electronic medical records or database to include patients in chronological order, based on the occurrence of stroke as was stated in the study protocol. This may have excluded deceased patients or patients transferred to nursing homes, thus resulting in lower mortality rate (13%) than that in the literature (14.7%) (16). Second, GPs may not have had access to all data elements such as sick leave, which was underreported. This might be because the first sick leave is usually prescribed by the doctors at hospital. Third, although we intended to have a representative sample of the GPs, practices in this study were self-selected and may not be typical of all GP practices in Germany. Physicians may have responded in a specific way out of interest in the management of IS stroke (17) and this may have also inflated the reported results.

Conclusions
This study describes the management of IS at onset of symptoms, at the acute phase in the hospital and during a 12-month follow-up. We described the healthcare resource utilization of IS patients post-discharge, which is a major driver of economic burden in long-term care. In addition, we documented that GPs receive a hospital letter which details the pathway in the hospital (unit stays) and imaging tests. GPs appear to provide reliable data about patient management. Also, we demonstrated the lack of knowledge among GPs about thrombolysis treatment and the fact that the rate of thrombolyzed patients is still low in Germany.