- Email: [email protected]
What Is Palliative Care in Germany? Results from a Representative Survey
Vol. 23 No. 6 June 2002
Journal of Pain and Symptom Management
471
Original Article
What Is Palliative Care in Germany? Results from a Representative Survey Lukas Radbruch,* MD, Friedemann Nauck,* MD, Martin Fuchs, MD, Karl Neuwöhner, BA, Dieter Schulenberg, BSc, Gabriele Lindena, DVM, and the Working Group on the Core Documentation for Palliative Care Units in Germany Pain Clinic, Department of Anesthesiology (L.R.), University of Cologne, Cologne; Department of Anesthesiology and Palliative Care (F.N.), Malteser Hospital, Bonn; Palliative Care Unit, Department of Internal Medicine (M.F.), St. Elisabeth Hospital, Neuwied; Palliative Care Unit (K.N.), Dr. Hancken Hospital, Stade; German Cancer Society (D.S.), Frankfurt; and Mundipharma GmbH (G.L.), Limburg, Germany
Abstract The recent development of palliative care inpatient units in Germany has been impressive. As a first step for quality assurance, a core documentation form was developed in 1996. The core documentation form consisted of 4 pages with 35 items documenting physical and psychosocial symptoms at the time of admission, diagnostic and therapeutic procedures before and during inpatient treatment, and outcome of physical and psychosocial dimensions. Checklists were used for most items and free text entries could be added. Of the 65 palliative inpatient units in Germany, 44 participated in the second phase of the evaluation of the core documentation in 2000. Eight units were affiliated with anesthesiology departments, 31 with internal medicine, two with radiotherapy, 1 with a surgical department, and 2 units were not affiliated with a department of the hospital. A total of 1087 patients were assessed in the 44 units during a period of up to 3 months. There was a high variability between units in all checklist items of the core documentation. Compared to units affiliated with internal medicine departments, units affiliated with anesthesiology departments performed less chemotherapy, but more immunotherapy; gave fewer infusions and blood transfusions but more skin and wound care, and more lymphatic drainage and massage; and documented psychosocial interventions more frequently for patients as well as for relatives. In one-third of the patients, a consent for omission of therapeutic options was documented. Inpatient treatment ended with the death of the patient in 45.3% of patients and with discharge in 51.6% (not documented 3.1%). The efficacy of inpatient treatment was rated very high by the staff. In conclusion, we found large variation in the documentation pertaining to palliative care patients at the time of admission, as well as for inpatient treatment, among palliative care units in Germany. This was related to the affiliation of the units at least to some degree, but also to differences in interests and documentation discipline. We suggest that training procedures for
Address reprint requests to: Lukas Radbruch, MD, Klinik für Anästhesiologie und Operative Intensivmedizin, Universitätskliniken Köln, 50924 Köln, Germany. *Both authors contributed equally to the manuscript. Accepted for publication: September 4, 2001. © U.S. Cancer Pain Relief Committee, 2002 Published by Elsevier, New York, New York
0885-3924/02/$–see front matter PII S0885-3924(02)00408-6
472
Radbruch et al.
Vol. 23 No. 6 June 2002
documentation should be included in crossectional surveys, as the results may not be comparable otherwise. However, common documentation instruments may be the first step towards an interdisciplinary discussion on aims and methods in palliative care. J Pain Symptom Manage 2002;23:471–483. © U.S. Cancer Pain Relief Committee, 2002. Key Words Palliative care units, survey, outcome assessment, symptom prevalence, therapeutic procedures, clinical audit
Introduction The development of palliative care inpatient units in Germany has been impressive during the past 10 years. The number of inpatient units for palliative care has increased from 21 in 1993 to 65 in 2000, and the number of inpatient hospices has increased from 11 to 87.1 However, up to now no standards for inpatient palliative care wards have been established in Germany, and with the increasing number of units, differences in the aims and working procedures of the units have become visible. One example is the use of chemotherapy as a palliative treatment for alleviation of pain or other symptoms. Some palliative care physicians, mostly with training in anesthesiology, strongly reject chemotherapy for the patients in their units,2 while others, with an oncology background, will accept or even suggest it for their patients. In a recent survey among the 65 palliative care units in Germany, nearly all units reported that they assessed pain and other symptoms regularly in their units.3 However, personal communication with the physicians revealed that in most cases no standardized documentation systems were used, and that only a few of the items relevant for the assessment of pain and other symptoms were documented, and mostly in an inconsistent and incomplete way. Problems other than physical symptoms were not assessed at all. Staff in the palliative care units often finds assessment and documentation systems time consuming and a burden not only to themselves but also to the patients. However, most palliative care specialists agree that standardized documentation is necessary for quality assurance, not least to justify the expenses for palliative medicine in the continuous dispute with hospital administrations and health insurance companies. This is confirmed by the high percentage of participation in this
survey. We have shown recently that standardized self-assessment by the patients is possible in a German palliative care unit with high acceptance by the patients.4,5 In 1996, the German Ministry of Health initiated a meeting of physicians from palliative care units to define a core instrument to be recommended for use in the inpatient units, and to enable the evaluation of the concepts and procedures used in the different units. A working group from several palliative care units, the German Cancer Association, and the German Association of Palliative Medicine produced a draft for such an instrument, which was tested in a first series of patients from 33 units in 1999.6 The instrument was modified after the evaluation of that series. The discussion about standards in palliative care was carried on intensively among the different specialities and professions in the working group. Palliative medicine is not designated as a speciality in Germany, although the first chair in palliative medicine was established in Bonn in 1999. Palliative care units are affiliated with anesthesiology, oncology, radiotherapy, or surgical departments, and physicians in the units have training in one of these specialities. Data from the second assessment period in the year 2000 was evaluated to gain information about the variability in the documentation and the services among the palliative care units and on factors influencing this variability.
Methods In 1996, a core documentation form was developed by a working group from the German Cancer Association and the German Association for Palliative Medicine. Thirty-three units took part in a first phase of documentation in
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
1999, when 757 patients were assessed with this core documentation form. The form was modified following this evaluation6 and an instrument for self-assessment by the patient was added. In 1999, all palliative care units listed in the German Hospice Directory7 were informed of the survey and invited to participate in the second phase of the development of the instrument. A principal investigator was identified in each of the 44 units that agreed to participate. These investigators took part in the preparatory meeting and were briefed about the trial design and the documentation procedures. They informed their staff members about the project and ensured that patients were documented consecutively. Units confirmed information on staff as well as other structural data from the last survey for the German hospice directory in 1999.3 The units were also asked about the affiliation of their unit to the hospital. Patient documentation started on 1 April 2000. All patients admitted to the unit were entered into the survey. In each unit, up to 30 patients were documented consecutively. Documentation was stopped at the end of July, if 30 patients had not been recruited by then. The core documentation form consisted of 4 pages with 35 items. Checklists were used for most items, with the possibility to add free text entries. Multiple entries were possible wherever appropriate. At the time of admission, demographic data, social background, and previous caregivers were documented. Checklists were available for reasons for admission, symptoms, concomitant diseases, and for psychosocial and nursing problems. Diagnosis of the underlying disease was entered as free text as well as with the ICD-10 classification. Previous antineoplastic therapies, palliative therapeutic measures, and analgesic regimen were documented. Performance status was entered according to the classification of the Eastern Cooperative Oncology Group (ECOG).8,9 Pain, other symptoms, level of care needed, provision of care at home and patient satisfaction were scored on 5-point categorical scales. These assessments were repeated at the end of the inpatient treatment, either by discharge or death. At this time, antineoplastic and palliative therapies, including analgesic therapies, performed during inpatient treatment were documented. When omission or
473
withholding of therapies was discussed with patients or relatives, this was documented as well. Psychosocial interventions during inpatient treatment and the amount of support organized for the continuation of care after discharge were recorded. A manual was provided to give brief information on the background of the core documentation, and explain intent and options for each item. Items on the form were classified with a color code to show which profession was most likely to provide appropriate assessment of this item. It was suggested that the form should be completed at staff meetings, receiving input from all professions involved in palliative care. However, this was not controlled, and the investigators were encouraged to complete the forms themselves rather than skip items if specialized staff was not available. The minimal documentation system (MIDOS) for palliative care patients was offered to the participating units as an instrument for self-assessment by the patients. MIDOS was available as a paper form as well as an electronic data base (Microsoft Access). The items of the core documentation forms were also provided in this data base. However, only one unit sent in data collected with this database. Data from the survey were processed in an electronic database and evaluated with the statistical package SPSS. Student’s t -test was calculated to compare mean prevalences of items between patients from units affiliated with anesthesiology or internal medicine, using P 0.05 for significance level. Patients from other units (surgical, radiotherapy, others) were not included in this comparison, as it was felt that the wide variability between units and the small number of patients from these units would prevent meaningful evaluations. Symptom intensities in MIDOS were converted to sum scores for pain intensity (2 items) and symptom intensity (7 items).10 The correlation between self-assessment with MIDOS and the items of the core documentation were calculated as Pearson’s correlation coefficient. The trial design did not include any therapeutic interventions. The survey did not place any burden on the patient, and did not lead to documentation of data other than that recommended for routine assessment and quality assurance. Therefore, approval from ethics committees was not sought in this trial involving centers throughout Germany.
474
Radbruch et al.
Results Eight units were affiliated with anesthesiology departments, 31 with internal medicine, 2 with radiotherapy, 1 with a surgical department, and 2 units were not affiliated with any specific department of the hospital. A total of 381 beds were available in these 44 units (mean 8.7 beds per unit, range 2–25). In one unit, no rooms had been assigned to palliative care, but rather 2 (not always the same) of the 26 beds in this unit were declared as palliative care beds, with additional staff provided by the hospital for palliative care. Units were staffed with physicians (median 1 physician per unit, range 0.5–4), nurses (median 9.15, range 2–28), psychologists (median 0.4, range 0–1), social workers (median 0.5, range 0–3), chaplains (median 0.5, range 0–3), physiotherapists (median 0.5, range 0–3.5) and others such as art or music therapists, secretaries or out-patient palliative care service staff (range 0–6). Eight units were not staffed with social workers, 8 units were without chaplains, and 11 units were without physiotherapists. No unit-based psychologists were available in 16 units. These units had to rely on hospital staff for these disciplines. Nurse:bed ratios ranged from 0.2:1 to 2:1 (mean 1.08:1), with only 6 of the 44 units reaching ratios of 1.4:1 or more.
Vol. 23 No. 6 June 2002
A total of 1087 patients were evaluated in the 44 units. More than 25 patients were included in 31 units, and 13 to 24 in the others. For 812 patients, this was the first admission in the palliative care unit, and for the other 269 patients, it was the 2nd to 25th readmission (not documented in 6 patients). Mean age of the 499 men and 585 women (not documented in 3 patients) was 65.6 years (range 19–95 years). Diagnosis of the underlying disease is shown in Table 1 and performance status is depicted in Table 2. In units affiliated with anesthesiology departments, significantly more patients with cancer of the respiratory tract were admitted, though this was not correlated with differences in related symptoms such as dyspnea (Table 3). There was a high degree of variability among units in the reasons for admission, and the prevalence of symptoms and concomittant diseases (Tables 3 and 4). The same amount of variation occurred for diagnostic and therapeutic procedures in the palliative care units (Table 5), as well as for psychosocial and physiotherapeutic procedures (Table 6). The use of prosthetic or orthopedic devices, tube feeding, laser therapy, insertion of tracheal stents or of spinal catheters, as well as the use of bisphosphonates, were provided as items in the checklist, but were not documented in any patient.
Table 1 Diagnosis Patients (%)
ICD-code
Diagnosis
All Patients (n 1087)
C00–C14 C15–C26 C30–C39 C40–C41 C43–C44 C45–C49 C50 C51–C58 C60–C63 C64–C68 C69–C72 C73–C75 C76–C80 C81–C96
Mouth and pharynx Gastrointestinal tract Respiratory tract Skeletal system Skin Connective tissue Breast Female genital tract Male genital tract Urinary tract Eye and central nervous system Endocrine glands Undefined primary cancer Lymphoma, leukemia Others Not documented
4.1 28.2 15.4 0.4 1.5 1.1 13.4 8.6 5.1 6.1 2.9 0.6 2.8 6.3 2.8 0.7
Anesthesia (n 193)
Internal Medicine (n 765)
5.7 29.0 10.4 0 2.1 3.1 13.5 6.7 6.2 4.7 4.1 0 4.1 5.7 4.2 0.5
3.0 29.4 16.6 0.5 1.3 0.8 12.4 8.5 5.2 6.3 2.9 0.8 2.7 7.1 2.5 0
Values are given as percentages of patients. Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
475
Table 2 Performance Status Patients (%)
ECOG 0 1
2
3 4
Performance Status Fully active, able to carry on all pre-disease performance without restriction Restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature (e.g. light house work, office work) Ambulatory and capable of all self-care but unable to carry out any work activities, up and about more than 50% of waking hours Capable of only limited self-care, confined to bed or chair more than 50% of waking hours Completely disabled, cannot carry on any self-care, totally confined to bed or chair
Not documented
All Patients (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
1.4
0.5
1.8
6.5
5.7
6.7
21.7
17.5
23.4
33.4
33.0
33.9
36.9
43.3
34.2
0.1
0
0
Values are given as percentages of patients.
In one-third of the patients, consent for omission of therapeutic options was documented (Table 7). Inpatient treatment ended with the death of the patient in 492 cases (45.3%) and with discharge in 561 patients (51.6%, not documented 3.1%). In the internal medicine units, 44% of the patients were treated until death, compared to 55% of patients in the anesthesiology units. Mean duration of inpatient treatment was 13.8 12.1 days (death: 12.1 11.6 days; discharge: 15.2 12.3), with a range of 0 to 111 days. There were no differences between anesthesiology and internal medicine units, though the mean duration of treatment was longer for the surgical and the two radiotherapy units. Patients were discharged home in most cases; only 11.8% of the patients were discharged to other inpatient units (hospices 2.9%, nursing home 4.0%, other hospital 4.9%). For patients discharged home, outpatient palliative care services were available for 6.5%, and outpatient hospice services were available for 3.1%. In most cases, the general practitioner (29.3% of the patients), the family (34.5%), or the patient themself (15.5%) was named as the primary care giver for the patient after discharge (multiple entries 34.5%). The efficacy of treatment in the palliative care units was rated very high by the staff (Ta-
ble 8). The percentage of patients with mild or no residual pain was reduced from 44.5% at the time of admission to 85.0%. Good or very good symptom control was documented for 23.7% at the time of admission versus 66.9% at the end of inpatient treatment. More than half (50.2%) of the patients were reported to be satisfied or very satisfied with their situation at the time of admission versus 83.2% at the end of inpatient treatment. There were no differences related to the affiliation of the units. For 159 patients, self-assessments by the patients with the minimal documentation systems for palliative care (MIDOS) were available. Assessment of pain intensity with MIDOS at the time of admission correlated highly with pain intensity in the core documentation (r 0.73). Symptom intensity sum values in MIDOS showed no correlation with symptom intensity in the core documentation, and only minimal correlation with intensity of the leading symptom (r 0.29), but some correlation with performance status (r 0.46) and nursing requirements (r 0.44). Anxiety in MIDOS was correlated to psychic stability in the core documentation (r 0.47) and well-being in MIDOS with the ECOG performance status (r 0.44). Correlation of the differences between admission and discharge of the sum values of MIDOS with the differences of the efficacy ratings
476
Radbruch et al.
Vol. 23 No. 6 June 2002
Table 3 Reasons for Admission and Symptoms at the Time of Admission Patients (%)
Reasons for admission Pain Physical symptoms Nutrition Terminal care Nursing problems Social problems Psychological problems Others Multiple responses Symptoms at time of admission Weakness Pain Appetite loss Nausea Cachexia Dyspnea Constipation Neurological symptoms Anemia Edema Psychiatric symptoms Sleeping disorders Cough Ascites Dysphagia Urinary disorders Dyspepsia Fever Diarrhea Tumor bleeding Pruritus Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
57.5 54.9 36.2 15.5 15.2 14.2 13.8 18.9 73.6
54.6 50.0 33.5 12.4 21.7 21.1 16.0 12.4 72.2
57.3 56.8 37.1 15.1 14.4 12.8 12.8 19.2 73.1
80 80 44 8 8 4 20 8 92
16 8 7 0 0 0 0 0 26
88 93 83 62 59 41 43 65 100
76.5 64.6 49.6 36.8 32.9 29.4 24.2 19.6 16.9 15.6 15.0 11.4 11.0 8.9 8.7 7.5 6.2 5.2 5.2 4.0 2.8 10.2 1.3 93.7
80.9 66.0 52.1 41.8 38.7 34.0 30.4 21.1 15.5 22.7 18.6 21.1 10.3 7.7 8.8 11.3 4.6 6.2 4.6 2.1 3.6 8.2 0 94.8
76.1 63.0 50.1 37.1 30.8 27.0 24.5 19.4 17.7 14.2 14.5 9.8 11.5 9.4 8.9 7.4 7.1 5.2 5.7 4.0 2.8 10.7 1.4 93.3
80 80 48 28 12 36 20 28 0 4 12 8 8 4 0 0 4 0 4 0 0 12 0 96
46 21 20 4 0 8 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 67
97 96 85 85 72 80 50 41 52 42 36 50 45 30 28 38 52 20 18 16 12 41 18 100
Surgical (n 25)
Minimum
Maximum
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
in the different dimensions were calculated. Pain relief in MIDOS correlated with differences of pain intensities (r 0.52) in the core documentation from these patients. Relief of symptoms other than pain in MIDOS correlated only marginally with performance status (r 0.26) and psychological stability of the patient (r 0.24). Documentation discipline was investigated with several indicators. In 15 checklist questions of the core documentation instrument, multiple entries were possible, for a total number of 153 items. The sum of items chosen ranged from 3 to 65 in all patients. The mean number of items used in the patients of the different units ranged from 19 to 48, with no differences between anesthesiology (mean 33.8 items) and internal medicine units (mean 32.3
items). The surgical unit showed no differences in the mean number of items (34 items), though an emphasis on the documentation of certain areas, such as pain and other physical symptoms, reasons for admission, drug therapy, or psychosocial interventions for patients and relatives can be seen in the tables. For 140 patients, no locations for metastases were given, but for only 66 of these patients (47%) the item “no metastases” was answered positively. On the other hand, for 5 patients this item was answered affirmatively, though at least one location for metastases had been confirmed.
Discussion Data on palliative care from cross-sectional and multinational surveys have been collected
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
477
Table 4 Concomitant Diseases and Metastases at the Time of Admission Patients (%)
Concomitant diseases Cardiovascular diseases Metabolic disorders Lung diseases Neurological disorders Movement-related disorders Gastrointestinal tract diseases Urogenital tract diseases Other tumors Infections Head and brain diseases Skin diseases Other diseases Multiple responses Metastases Liver Bone Lung Lymph nodes Peritoneal Central nervous system Pleura Skin Gastrointestinal tract Urogenital tract Adrenal glands Breast Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
31.9 14.7 13.6 11.9 10.4 9.8 7.9 8.0 5.3 4.9 2.7 6.8 37.0
32.5 12.9 20.1 13.4 15.5 13.4 9.3 6.2 7.7 4.6 1.5 8.2 44.3
32.6 15.1 11.5 11.8 7.8 3.2 7.8 8.8 5.4 4.9 3.2 6.6 36.0
28 16 8 20 20 4 20 16 0 16 4 12 48
4 0 0 0 0 0 0 0 0 0 0 0 4
54 36 52 39 40 44 30 24 29 21 14 26 72
32.6 30.8 25.7 24.3 18.1 13.5 9.3 6.7 6.1 4.9 2.4 1.8 9.9 6.5 55.0
35.1 34.5 23.7 23.7 11.3 15.0 5.2 6.7 7.7 4.1 1.5 2.1 10.3 4.1 52.6
30.9 29.2 24.6 25.0 19.4 13.6 10.2 6.9 5.8 5.0 2.2 1.8 9.4 7.5 53.7
28 36 28 12 16 20 4 0 8 0 8 0 12 16 60
4 0 7 0 0 0 0 0 0 0 0 0 0 0 13
63 65 52 63 43 35 44 24 24 23 15 16 37 39 88
Surgical (n 25)
Minimum
Maximum
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
before.11 However, even though these surveys included a large number of centers, they did not attempt to deliver representative data for the countries involved. National surveys have collected data on structural quality of palliative care service,12–14 or on views and attitudes of patients, relatives, or staff,15–18 but articles on the palliative care delivered to the patients and its outcome usually come from a single unit or service (see reviews19,20). As 44 of the 65 palliative care units in Germany participated in this survey, this is the first report of representative longitudinal data on the practice of inpatient palliative care in a European country. The participating units comprised 72.2% of the palliative care beds in Germany. Inpatient hospices were not included in the survey, as these units have a different function in palliative care in Germany. Whereas palliative care units are affiliated with a hospital, have physicians on their regular staff, and are
budgeted along with the other hospital beds, receiving full funding from the national health insurance system, hospices usually stand alone without hospital affiliation and their staff does not include physicians. Reimbursement from the health insurance system is much less for hospice beds than for palliative care beds, and patients must pay part of the costs in many hospices. Palliative care units try to treat symptoms and psychosocial problems and discharge patients at home or to a hospice if possible, whereas hospices often have an expected survival time of less than six months as an admission criterion, and care for their patients in the final phase of the disease until death. The need for standardized assessment tools and documentation systems as a prerequisite for effectiveness in palliative care is widely acknowledged. This is demonstrated by the large number of German palliative care units that participated in this survey. In preliminary dis-
478
Radbruch et al.
Vol. 23 No. 6 June 2002
Table 5 Diagnostic and Therapeutic Procedures During Inpatient Stay in Palliative Units Patients (%)
Diagnostic procedures Laboratory investigations Radiology Ultrasonography Electrocardiography CT scan Endoscopic procedures MR scan Others None Multiple responses Therapeutic procedures Chemotherapy Radiotherapy Immunotherapy Hormonal therapy Surgical procedure Drug treatment for symptom control Infusions Blood transfusions Skin and wound care Urinary catheter Inhalation therapy Central venous line Gastric tube, PEG Pleural aspiration, pleurodesis Port Ascites paracentesis Anus praeter Tracheostoma Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
80.6 42.0 38.0 31.0 13.7 10.6 2.1 8.7 12.1 61.3
76.3 30.9 21.1 9.8 8.2 10.3 1.0 3.6 17.0 42.3
83.1 45.7 45.3 40.8 13.3 11.5 1.9 10.9 9.6 68.0
92 12 0 0 4 0 0 0 8 16
9 3 0 0 0 0 0 0 0 4
100 100 100 96 68 60 14 56 72 100
8.7 5.1 4.6 4.9 1.1 80.0 46.7 14.3 23.6 21.1 9.5 7.0 4.8 4.4 4.2 2.7 1.9 1.0 11.5 1.9 72.3
1.5 4.1 11.3 5.7 1.0 79.9 40.7 10.8 37.1 22.2 7.7 5.7 3.1 3.1 4.1 2.1 1.5 5.2 9.3 3.1 65.5
10.6 4.1 3.5 4.8 1.2 80.1 48.6 16.6 20.1 20.1 11.1 7.5 4.8 4.5 4.3 2.7 1.8 1.0 11.6 1.6 74.1
0 8 0 0 0 96 56 0 40 56 0 20 24 4 16 0 12 4 12 0 88
0 0 0 0 0 22 4 0 0 0 0 0 0 0 0 0 0 0 0 0 25
44 48 88 24 8 100 93 40 68 63 36 47 24 16 31 21 12 9 43 10 100
Surgical (n 25)
Minimum
Maximum
Boldfaced numbers are items with differences between patients treated in anesthesiology and internal medicine units.
cussions, personal remarks from the participants underlined the importance they placed on assessment, quality assurance, and the need for good standardized documentation tools. Currently, no recommendations for standardized assessment and documentation are available in Germany, though a tool for patients’ self-assessment has been validated and computer-based documentation systems have been described.4,5,21 The core documentation group may also be regarded as the first step to establish clinical audit in German palliative care units. In the preliminary discussions in the working group, the main areas of concern were identified and agreed upon among the participating units. Documentation and assessment of the outcome was then performed with the core documentation and the results were fed back to the participants in subsequent
meetings of the working group, thus completing the audit cycle.22 During the recent past, the quality and the availability of palliative care beds has developed rapidly in Germany.1 However, with the growing number of units, concerns have been raised about the quality of the services.23 In this survey, only 6 units had a ratio of nursing staff to inpatient beds of 1.4:1 or more as recommended by the German Association for Palliative Medicine. The assessment at the time of admission showed symptoms related to reduced performance status in advanced cancer, such as weakness, appetite loss, and cachexia, to be predominant. High prevalences of symptoms, such as pain, nausea, dyspnea, and constipation, for which more specific treatment options are available, give an indication of undertreat-
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
479
Table 6 Psychosocial and Physiotherapeutic Interventions During Inpatient Palliative Therapy Patients (%)
Psychosocial interventions for the patient From physician From nurse From psychologist From social worker From chaplain From volunteers From others None Psychosocial interventions for relatives From physician From nurse From psychologist From social worker From clergy From volunteers From others None Multiple responses for psychosocial interventions Physiotherapeutic interventions Physiotherapy Change of position Mobilization Breathing therapy Lymphatic drainage, massage Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
Surgical (n 25)
Minimum
Maximum
57.6 52.6 28.3 28.8 34.7 17.4 1.5 7.8
75.7 71.5 24.4 26.4 41.5 18.7 0.5 7.3
56.0 50.5 27.6 29.5 33.1 16.3 1.2 8.5
100 100 4 8 8 44 0 0
0 0 0 0 0 0 0 0
100 100 92 89 100 88 16 43
46.9 37.8 13.7 18.6 13.6 5.5 4.7 8.0 78.2
69.4 58.6 15.5 20.2 16.6 9.8 0 6.2 84.0
43.9 34.4 12.4 18.6 12.4 4.3 0.5 9.2 77.5
84 84 4 0 0 28 0 0 100
0 0 0 0 0 0 0 0
92 92 71 74 50 56 4 43
33.9 32.1 30.4 19.5 16.8 4.9 21.0 41.1
29.9 32.5 25.8 15.5 27.8 8.8 21.1 42.8
33.0 34.7 32.2 21.1 14.0 3.8 20.3 41.5
84 24 28 0 4 0 0 32
0 0 0 0 0 0 0
88 78 88 64 59 47 76
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
ment. Symptom prevalences were in the range described in other trials.24–33 The range of symptom prevalences in these studies correlated well with the range of prevalences in the units described in our survey. Pain was documented in 38–84% of the patients in the literature, and in 21–96% of patients in the palliative care units in this survey. Likewise, nausea was found in 6–51% of patients in the literature and in 4–85% of patients in our units. In Germany, palliative medicine is not a recognized speciality. Physicians involved in palliative medicine mostly came from an anesthesiology background in the early years, and from training in internal medicine and oncology more recently. The great majority of palliative care units in Germany are affiliated with hospital departments, in most cases internal medicine or anesthesiology. The palliative care unit with the longest standing was established in 1983 at the University of Cologne and is the only one that is affiliated with the surgical department. Most units are staffed by physicians
from the department they are affiliated with, though in several cases, physicians with other speciality training were working in some palliative care units. Interdisciplinary cooperation is a major feature of palliative medicine and may be more difficult for units staffed only by physicians with the same speciality training. This also is evident in the data collected, as units affiliated with anesthesiology departments performed much less chemotherapy, but more immunotherapy; gave fewer infusions and blood transfusions but more skin and wound care, more lymphatic drainage and massage; and documented psychosocial interventions more frequently for patients as well as for relatives than those affiliated to internal medicine departments. This raises the question about the general aim of treatment in palliative care units. We did not calculate statistical differences for the other units, as the small number of these units and their patients would have prevented a meaningful evaluation, considering the large
480
Radbruch et al.
Vol. 23 No. 6 June 2002
Table 7 Consent to Omission of Therapeutic Measures Patients (%)
Procedures Resuscitation Chemotherapy Artificial ventilation Assisted feeding Surgical procedures Radiotherapy Dialysis, ultrafiltration Blood transfusions Hydration Drug therapy of concomitant diseases Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
Surgical (n 25)
Minimum
Maximum
17.3 13.9 13.9 13.6 11.9 10.4 7.3 5.0 4.8 2.9 2.1 63.3 21.7
18.6 4.6 16.5 9.8 11.3 4.1 4.6 4.1 4.1 4.1 2.1 62.9 22.7
14.4 12.2 10.7 12.6 8.8 8.4 4.1 2.3 3.9 1.3 2.2 65.5 18.9
4 4 4 4 4 4 4 4 0 4 0 88 8
0 0 0 0 0 0 0 0 0 0 0 12 0
76 88 64 67 84 72 66 59 40 36 13 100 88
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
variations among units. Differences among the specialities may have been related to differences in the patients treated. However, though the number of patients with malignant disease in the respiratory tracts showed significant differences between anesthesiology and internal medicine, this was not related to differences in the prevalence of specific respiratory symptoms.
Differences among the units and specialities in the prevalence of symptoms or therapies may be related to different attitudes towards documentation. Documentation discipline was investigated with several indicators. The number of items marked in those sections of the questionnaire where multiple entries were possible showed large differences, though the spe-
Table 8 Outcome of Inpatient Therapy Patients (%) All Patients (n 1087) Pain None Mild Moderate Severe As bas as can be Not documented Symptoms Very good symptom control Good symptom control Moderate symptom control Only little symptom control Symptoms not controlled Not documented Satisfaction of patients or relatives Very satisfied Satisfied Little satisfied Dissatisfied Very dissatisfied Not documented
Anesthesia (n 193)
Internal Medicine (n 765)
26.5 18.0 26.7 24.8 3.5 0.5
55.7 29.3 10.4 3.4 0.3 0.9
27.6 15.6 27.1 25.0 4.7
56.0 29.3 12.0 2.1 0.5
28.1 18.8 23.9 25.6 3.6
58.3 28.2 9.5 3.8 0.3
4.5 19.2 36.2 32.0 4.5 3.6
16.3 50.6 21.3 7.7 1.7 2.3
3.8 18.0 32.8 37.7 7.7
17.3 50.8 23.0 6.8 2.1
5.0 20.1 37.0 33.5 4.4
16.7 50.6 22.1 8.6 2.0
11.7 38.5 26.7 11.0 2.1 9.9
31.5 51.7 10.1 1.8 0.4 4.5
11.6 41.5 33.3 10.9 2.7
42.3 45.6 8.8 3.3
13.5 45.0 27.9 11.4 2.1
29.8 56.9 11.1 1.9 0.4
Values are given as percentages of patients. Boldfaced numbers are assessments at the time of admission; unboldfaced numbers are assessments at the end of inpatient treatment (discharge or death).
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
ciality of the units did not influence this number. Data from the surgical unit show that this unit may be taken as an example of specific emphasis on certain areas of the documentation. This could be due to different conceptions of the intent of the questions; it also could be related to areas of interest of the staff performing the documentation. Documentation discipline may be connected more to the attitudes of the persons completing the questionnaire than to specific traits of the unit or special patient populations. Omission of therapeutic options with the consent of the patient is one of the key features of palliative care. However, the intent of the question about consent for omission of therapy may not have been clear enough, and, consequently, wide variability again was documented. Rejection of resuscitation was documented for no patients in some units and in up to 78% of patients in others; consent on omission of further chemotherapy varied between 0% and 88% of patients in the different units. End-oflife decisions and advance directives are not common in Germany. Written advance directories were found only in 18% of the patients from German palliative care units, compared to 79% of patients in units in the United States.34 This may be due to a lack of information among patients and staff. More specific research is needed in this area in Germany, but the setting of the core documentation may not be suited for assessment of these questions. The outcome of inpatient palliative care was rated very high in the survey, as shown by the large percentages of patients with minimal or no residual pain, good symptom control, and high satisfaction with therapy. However, the documentation by the staff of the units may have been heavily biased, not only from overestimation of their own work, but also from patients reporting good results to please the staff of the unit. The correlation between self-assessment and assessment by staff were available only for few patients, and showed moderate correlations of outcome ratings at most, with the exception of a higher correlation for pain relief. Comparison of assessments by patients and staff at the time of admission showed higher correlations, at least for some items, but again demonstrated a lack of correlation between self-assessment and assessment by staff for symptom intensity. This underlines the im-
481
portance of the additional input from a selfassessment instrument in palliative care outcome assessment.22 The core documentation instrument used in this survey was comprehensive and extensive. Higginson and Hearn have suggested the Palliative Care Outcomes Scale (POS), with only 12 items.22 Several of these items (pain intensity, symptom intensity, anxiousness of patient or relatives, main problems in the last days, performance status) were integrated in the German core documentation. Other items from the POS, such as the ability to share feelings with friends or family or the time wasted on appointments related to health care, were not available in the core documentation, and might have given as much information as the more comprehensive sections of the core documentation. Use of shorter and more concise instruments such as the POS could have improved the correlation of staff ratings in the different units, as well as the correlation of self-assessment and assessment by staff. In conclusion, we found large variation in the documentation pertaining to palliative care patients at the time of admission, as well as for inpatient treatment among palliative care units in Germany. This was related to the affilitation of the units, at least to some degree, but also to differences in documentation discipline. We suggest the integration of training procedures for documentation in multicenter surveys, as the results may not be comparable otherwise. Training procedures could involve constructed case reports, detailed manuals, or training sessions. Questionnaires and documentation systems should be kept as short as possible, and clear rules for completion should be included in the questionnaires (e.g., “mark only items leading to significant impairment of function”). Following these recommendations, the core documentation for palliative care units may be developed into an instrument for audit, thereby improving the quality of palliative care in Germany considerably, and enabling standards for palliative inpatient care to be established.
Acknowledgments Part of the preparatory work and the data evaluation, as well as the development of an electronic data base for the core instrument,
482
Radbruch et al.
was supported by a financial grant from the Mundipharma Company, Limburg, Germany. We would like to thank the physicians, nurses, social workers, psychologists, and other staff members who spent much time and enthusiasm with the data documentation. Additional thanks are directed to the principal investigators of the participating centers, who participated in the development of the core instrument: Jan Hinnerk Stange, Palliativstation des Marienkrankenhaus, Bad Lippspringe; Imke Strohscheer, Palliativstation der Charité, Campus Virchow Klinikum, Berlin; Sabine Bothe, Palliativstation am Krankenhaus Spandau, Berlin; Annemarie Keller, Palliativstation am Malteser Krankenhaus, Berlin; Thomas Jehser, Interdisziplinäre Palliativstation am Gemeinschaftskrankenhaus Havelhöhe, Berlin; Christoph Ostgathe, Palliativstation am Malteser Krankenhaus, Bonn; Bernd Sittig, Palliativstation am St. Joseph Hospital, Bremerhaven; Christiane Glasmeyer, Palliativstation am St. -Johannes Krankenhaus, Dortmund; Barbara Schubert, Palliativstation am St. Joseph Stift, Dresden; Wolfgang Etspüler, Palliativstation der Vita Natura Klinik, Eppenbrunn; Thomas Montag, Palliativstation am Katholischen Krankenhaus, Erfurt; Bernhard Greiling, Palliativstation am St. Elisabeth-Krankenhaus, Eutin; Halves, Katharinen Hospiz am Park, Flensburg; Axel Elsner, Evangelisches Hospital für palliative Medizin, Frankfurt am Main; Dietmar Beck, Palliativstation der Universität Göttingen, Göttingen; Herbert Kaiser, Palliativstation am Städtischen Krankenhaus, Gütersloh; M. Kreiling, Palliativstation am St. Elisabeth Krankenhaus, Halle; Heike Schieder, Palliativstation des Allgemeinen Krankenhauses Barmbek, Hamburg; Hans-Joachim Lehmann, Palliativstation des Krankenhauses Rissen, Hamburg; Reiner Franke, Palliativstation des Kreiskrankenhaus, Hameln; Birgitt van Oorschot, Interdisziplinäre Onkologische Palliativabteilung an der Friedrich-Schiller-Universität, Jena; Wolfgang Spuck, Palliativstation des Rotes-Kreuz-Krankenhaus, Kassel; Hermann Ewald, Palliativstation der Universitätsklinik, Kiel; Ute Welling, Palliativstation des Universitätsklinikums, Köln; Waltraud Richter, Palliativstation des Bezirkskrankenhaus St. Georg, Leipzig; Joachim Steuber, Palliativstation am St. Elisabeth-Krankenhaus, Leipzig; G. Prillwitz, Palliativstation der Pfeiffersche Stiftun-
Vol. 23 No. 6 June 2002
gen, Magdeburg; Weiß, Palliativstation des Klinikum, Mannheim; Ulrike Münz, Johannes Hospiz am Krankenhaus, München; Claudia Bausewein, Palliativstation des Krankenhauses, München-Harlaching; Wolf-Diethard Pietruschka, Palliativstation des Klinikums, Neubrandenburg; H.-J. Wetzel, Palliativstation des FriedrichEbert-Krankenhauses, Neumünster; Martin Fuchs, Palliativeinrichtung am St. Elisabeth Krankenhaus, Neuwied; Otto Kloke, Palliativbereich am Elisabeth-Krankenhaus, Recklinghausen; Peter Ervens, Palliativstation des Krankenhauses Maria Stern, Remagen; Ulrike Thierbach, Palliativstation des Klinikum Südstadt, Rostock; G. Lenke, Palliativstation des Evangelischen-Freikirchlichen Krankenhauses, Rüdersdorf; Frank Förster, Palliativstation des Fachkrankenhauses Marienstift, Schwarzenberg; Karl Neuwöhner, Palliativstation der Klinik Dr. Hancken, Stade; Martin Surfleet, Palliativstation des Marienhospital, Stuttgart; Werner Richarz, Palliativstation des Herz Jesu Krankenhauses, Trier; Kertz, Palliativstation des St. Josef-Hospital, Troisdorf; Boris Hait, Palliativstation im Katharinen-Hospital, Unna; Andreas Geilen-Meerbach, Palliativstation am St. Michael Krankenhaus, Völklingen.
References 1. Sabatowski R, Radbruch L, Nauck F, et al. Über die Entwicklung palliativmedizinischer Einrichtungen in Deutschland. Zeitschrift für Palliativmedizin 2000;1:40–46. 2. Klaschik E, Nauck F, Radbruch L, Sabatowski R. Palliativmedizin—Definitionen und Grundzüge. Internist 2000;41:606–611. 3. Sabatowski R, Radbruch L, Nauck F, et al. Entwicklung und Stand der stationären palliativmedizinischen Einrichtungen in Deutschland. Schmerz, 2001;15:312–319. 4. Radbruch L, Sabatowski R, Loick G, et al. MIDOS—Validierung eines minimalen Dokumentationssystems für die Palliativmedizin. Schmerz 2000; 14:231–239. 5. Radbruch L, Sabatowski R, Loick G, et al. Cognitive impairment and its influence on pain and symptom assessment in a palliative care unit: development of a minimal documentation system. Palliat Med 2000;14:266–276. 6. Determann MM, Rzehak P, Küchler T, et al. Qualitätssicherung in der Palliativmedizin. Forum Deutsche Krebsgesellschaft 2000;15(2):10–13. 7. Sabatowski R, Radbruch L, Loick G, et al. Pallia-
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
tiv 2000—Stationäre und ambulanten Palliativ— und Hospizeinrichtungen in Deutschland. Limburg: Mundipharma GmbH, 1999. 8. Buccheri G, Ferrigno D, Tamburini M. Karnofsky and ECOG performance status scoring in lung cancer: a prospective, longitudinal study of 536 patients from a single institution. Eur J Cancer 1996; 32A(7):1135–1141. 9. Conill C, Verger E, Salamero M. Performance status assessment in cancer patients. Cancer 1990; 65(8):1864–1866. 10. Radbruch L, Sabatowski R, Loick G, et al. Midos— Validation of a minimal documentation system for palliative medicine. Schmerz 2000;14(4):231–239.
483
MIDOS—An electronic database for the palliative care unit. Schmerz 2000;14(4):257–263. 22. Higginson IJ, Hearn J. Palliative care audit: tools, objectives, and models for training in assessment, monitoring, and review. In: Portenoy RK, Bruera E, editors. Topics in palliative care. Oxford: Oxford University Press, 2000:95–118. 23. Nauck F, Ostgathe C, Lund S, et al. Qualitätssicherung in der Palliativmedizin. Ermittlung der Struktur- und Prozessqualität auf den Palliativstationen in NRW. Zeitschrift für ärztliche Fortbildung und Qualitätssicherung 2000;94:587-594. 24. Lichter I, Hunt E. The last 48 hours of life. J Palliat Care 1990;6(4):7–15.
11. Caraceni A, Portenoy RK, Ashby MA, et al. An international survey of cancer pain characteristics and syndromes. Pain 1999;82(3):263–274.
25. Reuben DB, Mor V, Hiris J. Clinical symptoms and length of survival in patients with terminal cancer. Arch Intern Med 1988;148(7):1586–1591.
12. Eve A, Smith AM, Tebbit P. Hospice and palliative care in the UK 1994–5, including a summary of trends 1990–5. Palliat Med 1997;11(1):31–43.
26. Brescia FJ, Adler D, Gray G, et al. Hospitalized advanced cancer patients: a profile. J Pain Symptom Manage 1990;5(4):221–227.
13. Sabatowski R, Radbruch L, Loick G, et al. Palliative care in Germany—14 years on. Eur J Palliat Care 1998;5:52–55.
27. Curtis EB, Krech R, Walsh TD. Common symptoms in patients with advanced cancer. J Palliat Care 1991;7(2):25–29.
14. Sabatowski R, Radbruch L, Müller M, et al. Ambulante Hospiz- und Palliativdienste in Deutschland. Zeitschrift für Palliativmedizin 2000;1:17–24.
28. Coyle N, Adelhardt J, Foley KM, Portenoy RK. Character of terminal illness in the advanced cancer patient: pain and other symptoms during the last four weeks of life. J Pain Symptom Manage 1990;5(2):83–93.
15. Jarrett N, Payne S, Turner P, Hillier R. ‘Someone to talk to’ and ‘pain control’: what people expect from a specialist palliative care team. Palliat Med 1999;13(2):139–144. 16. Morize V, Nguyen DT, Lorente C, Desfosses G. Descriptive epidemiological survey on a given day in all palliative care patients hospitalized in a French university hospital. Palliat Med 1999;13(2):105–117. 17. Peruselli C, Di Giulio P, Toscani F, et al. Home palliative care for terminal cancer patients: a survey on the final week of life. Palliat Med 1999;13(3):233–241. 18. Sebastian P, Varghese C, Sankaranarayanan R, et al. Evaluation of symptomatology in planning palliative care. Palliat Med 1993;7(1):27–34. 19. Rinck GC, van den Bos GA, Kleijnen J, et al. Methodologic issues in effectiveness research on palliative cancer care: a systematic review. J Clin Oncol 1997;15(4):1697–1707. 20. Hearn J, Higginson IJ. Outcome measures in palliative care for advanced cancer patients: a review. J Public Health Med 1997;19(2):193–199. 21. Radbruch L, Loick G, Sabatowski R, Elsner F.
29. Ventafridda V, Ripamonti C, de Conno F, et al. Symptom Prevalence and control during cancer patients’ last days of life. J Palliative Care 1990;6(3):7–11. 30. Seale C. Death from cancer and death from other causes: the relevance of the hospice approach. Palliat Med1991;5:12–19. 31. Grond S, Zech D, Diefenbach C, Bischoff A. Prevalence and pattern of symptoms in patients with cancer pain: a prospective evaluation of 1635 cancer patients referred to a pain clinic. J Pain Symptom Manage 1994;9:372–382. 32. Twycross RG, Lack S. Control of alimentary symptoms in far advanced cancer. Edinburgh: Churchill Livingstone, 1986. 33. Walsh D, Donnelly S, Rybicki L. The symptoms of advanced cancer: relationship to age, gender, and performance status in 1,000 patients. Support Care Cancer 2000;8(3):175–179. 34. Voltz R, Akabayashi A, Reese C, et al. End-of-life decisions and advance directives in palliative care: a cross-cultural survey of patients and health-care professionals. J Pain Symptom Manage 1998;16(3):153–162.
Journal of Pain and Symptom Management
471
Original Article
What Is Palliative Care in Germany? Results from a Representative Survey Lukas Radbruch,* MD, Friedemann Nauck,* MD, Martin Fuchs, MD, Karl Neuwöhner, BA, Dieter Schulenberg, BSc, Gabriele Lindena, DVM, and the Working Group on the Core Documentation for Palliative Care Units in Germany Pain Clinic, Department of Anesthesiology (L.R.), University of Cologne, Cologne; Department of Anesthesiology and Palliative Care (F.N.), Malteser Hospital, Bonn; Palliative Care Unit, Department of Internal Medicine (M.F.), St. Elisabeth Hospital, Neuwied; Palliative Care Unit (K.N.), Dr. Hancken Hospital, Stade; German Cancer Society (D.S.), Frankfurt; and Mundipharma GmbH (G.L.), Limburg, Germany
Abstract The recent development of palliative care inpatient units in Germany has been impressive. As a first step for quality assurance, a core documentation form was developed in 1996. The core documentation form consisted of 4 pages with 35 items documenting physical and psychosocial symptoms at the time of admission, diagnostic and therapeutic procedures before and during inpatient treatment, and outcome of physical and psychosocial dimensions. Checklists were used for most items and free text entries could be added. Of the 65 palliative inpatient units in Germany, 44 participated in the second phase of the evaluation of the core documentation in 2000. Eight units were affiliated with anesthesiology departments, 31 with internal medicine, two with radiotherapy, 1 with a surgical department, and 2 units were not affiliated with a department of the hospital. A total of 1087 patients were assessed in the 44 units during a period of up to 3 months. There was a high variability between units in all checklist items of the core documentation. Compared to units affiliated with internal medicine departments, units affiliated with anesthesiology departments performed less chemotherapy, but more immunotherapy; gave fewer infusions and blood transfusions but more skin and wound care, and more lymphatic drainage and massage; and documented psychosocial interventions more frequently for patients as well as for relatives. In one-third of the patients, a consent for omission of therapeutic options was documented. Inpatient treatment ended with the death of the patient in 45.3% of patients and with discharge in 51.6% (not documented 3.1%). The efficacy of inpatient treatment was rated very high by the staff. In conclusion, we found large variation in the documentation pertaining to palliative care patients at the time of admission, as well as for inpatient treatment, among palliative care units in Germany. This was related to the affiliation of the units at least to some degree, but also to differences in interests and documentation discipline. We suggest that training procedures for
Address reprint requests to: Lukas Radbruch, MD, Klinik für Anästhesiologie und Operative Intensivmedizin, Universitätskliniken Köln, 50924 Köln, Germany. *Both authors contributed equally to the manuscript. Accepted for publication: September 4, 2001. © U.S. Cancer Pain Relief Committee, 2002 Published by Elsevier, New York, New York
0885-3924/02/$–see front matter PII S0885-3924(02)00408-6
472
Radbruch et al.
Vol. 23 No. 6 June 2002
documentation should be included in crossectional surveys, as the results may not be comparable otherwise. However, common documentation instruments may be the first step towards an interdisciplinary discussion on aims and methods in palliative care. J Pain Symptom Manage 2002;23:471–483. © U.S. Cancer Pain Relief Committee, 2002. Key Words Palliative care units, survey, outcome assessment, symptom prevalence, therapeutic procedures, clinical audit
Introduction The development of palliative care inpatient units in Germany has been impressive during the past 10 years. The number of inpatient units for palliative care has increased from 21 in 1993 to 65 in 2000, and the number of inpatient hospices has increased from 11 to 87.1 However, up to now no standards for inpatient palliative care wards have been established in Germany, and with the increasing number of units, differences in the aims and working procedures of the units have become visible. One example is the use of chemotherapy as a palliative treatment for alleviation of pain or other symptoms. Some palliative care physicians, mostly with training in anesthesiology, strongly reject chemotherapy for the patients in their units,2 while others, with an oncology background, will accept or even suggest it for their patients. In a recent survey among the 65 palliative care units in Germany, nearly all units reported that they assessed pain and other symptoms regularly in their units.3 However, personal communication with the physicians revealed that in most cases no standardized documentation systems were used, and that only a few of the items relevant for the assessment of pain and other symptoms were documented, and mostly in an inconsistent and incomplete way. Problems other than physical symptoms were not assessed at all. Staff in the palliative care units often finds assessment and documentation systems time consuming and a burden not only to themselves but also to the patients. However, most palliative care specialists agree that standardized documentation is necessary for quality assurance, not least to justify the expenses for palliative medicine in the continuous dispute with hospital administrations and health insurance companies. This is confirmed by the high percentage of participation in this
survey. We have shown recently that standardized self-assessment by the patients is possible in a German palliative care unit with high acceptance by the patients.4,5 In 1996, the German Ministry of Health initiated a meeting of physicians from palliative care units to define a core instrument to be recommended for use in the inpatient units, and to enable the evaluation of the concepts and procedures used in the different units. A working group from several palliative care units, the German Cancer Association, and the German Association of Palliative Medicine produced a draft for such an instrument, which was tested in a first series of patients from 33 units in 1999.6 The instrument was modified after the evaluation of that series. The discussion about standards in palliative care was carried on intensively among the different specialities and professions in the working group. Palliative medicine is not designated as a speciality in Germany, although the first chair in palliative medicine was established in Bonn in 1999. Palliative care units are affiliated with anesthesiology, oncology, radiotherapy, or surgical departments, and physicians in the units have training in one of these specialities. Data from the second assessment period in the year 2000 was evaluated to gain information about the variability in the documentation and the services among the palliative care units and on factors influencing this variability.
Methods In 1996, a core documentation form was developed by a working group from the German Cancer Association and the German Association for Palliative Medicine. Thirty-three units took part in a first phase of documentation in
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
1999, when 757 patients were assessed with this core documentation form. The form was modified following this evaluation6 and an instrument for self-assessment by the patient was added. In 1999, all palliative care units listed in the German Hospice Directory7 were informed of the survey and invited to participate in the second phase of the development of the instrument. A principal investigator was identified in each of the 44 units that agreed to participate. These investigators took part in the preparatory meeting and were briefed about the trial design and the documentation procedures. They informed their staff members about the project and ensured that patients were documented consecutively. Units confirmed information on staff as well as other structural data from the last survey for the German hospice directory in 1999.3 The units were also asked about the affiliation of their unit to the hospital. Patient documentation started on 1 April 2000. All patients admitted to the unit were entered into the survey. In each unit, up to 30 patients were documented consecutively. Documentation was stopped at the end of July, if 30 patients had not been recruited by then. The core documentation form consisted of 4 pages with 35 items. Checklists were used for most items, with the possibility to add free text entries. Multiple entries were possible wherever appropriate. At the time of admission, demographic data, social background, and previous caregivers were documented. Checklists were available for reasons for admission, symptoms, concomitant diseases, and for psychosocial and nursing problems. Diagnosis of the underlying disease was entered as free text as well as with the ICD-10 classification. Previous antineoplastic therapies, palliative therapeutic measures, and analgesic regimen were documented. Performance status was entered according to the classification of the Eastern Cooperative Oncology Group (ECOG).8,9 Pain, other symptoms, level of care needed, provision of care at home and patient satisfaction were scored on 5-point categorical scales. These assessments were repeated at the end of the inpatient treatment, either by discharge or death. At this time, antineoplastic and palliative therapies, including analgesic therapies, performed during inpatient treatment were documented. When omission or
473
withholding of therapies was discussed with patients or relatives, this was documented as well. Psychosocial interventions during inpatient treatment and the amount of support organized for the continuation of care after discharge were recorded. A manual was provided to give brief information on the background of the core documentation, and explain intent and options for each item. Items on the form were classified with a color code to show which profession was most likely to provide appropriate assessment of this item. It was suggested that the form should be completed at staff meetings, receiving input from all professions involved in palliative care. However, this was not controlled, and the investigators were encouraged to complete the forms themselves rather than skip items if specialized staff was not available. The minimal documentation system (MIDOS) for palliative care patients was offered to the participating units as an instrument for self-assessment by the patients. MIDOS was available as a paper form as well as an electronic data base (Microsoft Access). The items of the core documentation forms were also provided in this data base. However, only one unit sent in data collected with this database. Data from the survey were processed in an electronic database and evaluated with the statistical package SPSS. Student’s t -test was calculated to compare mean prevalences of items between patients from units affiliated with anesthesiology or internal medicine, using P 0.05 for significance level. Patients from other units (surgical, radiotherapy, others) were not included in this comparison, as it was felt that the wide variability between units and the small number of patients from these units would prevent meaningful evaluations. Symptom intensities in MIDOS were converted to sum scores for pain intensity (2 items) and symptom intensity (7 items).10 The correlation between self-assessment with MIDOS and the items of the core documentation were calculated as Pearson’s correlation coefficient. The trial design did not include any therapeutic interventions. The survey did not place any burden on the patient, and did not lead to documentation of data other than that recommended for routine assessment and quality assurance. Therefore, approval from ethics committees was not sought in this trial involving centers throughout Germany.
474
Radbruch et al.
Results Eight units were affiliated with anesthesiology departments, 31 with internal medicine, 2 with radiotherapy, 1 with a surgical department, and 2 units were not affiliated with any specific department of the hospital. A total of 381 beds were available in these 44 units (mean 8.7 beds per unit, range 2–25). In one unit, no rooms had been assigned to palliative care, but rather 2 (not always the same) of the 26 beds in this unit were declared as palliative care beds, with additional staff provided by the hospital for palliative care. Units were staffed with physicians (median 1 physician per unit, range 0.5–4), nurses (median 9.15, range 2–28), psychologists (median 0.4, range 0–1), social workers (median 0.5, range 0–3), chaplains (median 0.5, range 0–3), physiotherapists (median 0.5, range 0–3.5) and others such as art or music therapists, secretaries or out-patient palliative care service staff (range 0–6). Eight units were not staffed with social workers, 8 units were without chaplains, and 11 units were without physiotherapists. No unit-based psychologists were available in 16 units. These units had to rely on hospital staff for these disciplines. Nurse:bed ratios ranged from 0.2:1 to 2:1 (mean 1.08:1), with only 6 of the 44 units reaching ratios of 1.4:1 or more.
Vol. 23 No. 6 June 2002
A total of 1087 patients were evaluated in the 44 units. More than 25 patients were included in 31 units, and 13 to 24 in the others. For 812 patients, this was the first admission in the palliative care unit, and for the other 269 patients, it was the 2nd to 25th readmission (not documented in 6 patients). Mean age of the 499 men and 585 women (not documented in 3 patients) was 65.6 years (range 19–95 years). Diagnosis of the underlying disease is shown in Table 1 and performance status is depicted in Table 2. In units affiliated with anesthesiology departments, significantly more patients with cancer of the respiratory tract were admitted, though this was not correlated with differences in related symptoms such as dyspnea (Table 3). There was a high degree of variability among units in the reasons for admission, and the prevalence of symptoms and concomittant diseases (Tables 3 and 4). The same amount of variation occurred for diagnostic and therapeutic procedures in the palliative care units (Table 5), as well as for psychosocial and physiotherapeutic procedures (Table 6). The use of prosthetic or orthopedic devices, tube feeding, laser therapy, insertion of tracheal stents or of spinal catheters, as well as the use of bisphosphonates, were provided as items in the checklist, but were not documented in any patient.
Table 1 Diagnosis Patients (%)
ICD-code
Diagnosis
All Patients (n 1087)
C00–C14 C15–C26 C30–C39 C40–C41 C43–C44 C45–C49 C50 C51–C58 C60–C63 C64–C68 C69–C72 C73–C75 C76–C80 C81–C96
Mouth and pharynx Gastrointestinal tract Respiratory tract Skeletal system Skin Connective tissue Breast Female genital tract Male genital tract Urinary tract Eye and central nervous system Endocrine glands Undefined primary cancer Lymphoma, leukemia Others Not documented
4.1 28.2 15.4 0.4 1.5 1.1 13.4 8.6 5.1 6.1 2.9 0.6 2.8 6.3 2.8 0.7
Anesthesia (n 193)
Internal Medicine (n 765)
5.7 29.0 10.4 0 2.1 3.1 13.5 6.7 6.2 4.7 4.1 0 4.1 5.7 4.2 0.5
3.0 29.4 16.6 0.5 1.3 0.8 12.4 8.5 5.2 6.3 2.9 0.8 2.7 7.1 2.5 0
Values are given as percentages of patients. Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
475
Table 2 Performance Status Patients (%)
ECOG 0 1
2
3 4
Performance Status Fully active, able to carry on all pre-disease performance without restriction Restricted in physically strenuous activity but ambulatory and able to carry out work of a light or sedentary nature (e.g. light house work, office work) Ambulatory and capable of all self-care but unable to carry out any work activities, up and about more than 50% of waking hours Capable of only limited self-care, confined to bed or chair more than 50% of waking hours Completely disabled, cannot carry on any self-care, totally confined to bed or chair
Not documented
All Patients (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
1.4
0.5
1.8
6.5
5.7
6.7
21.7
17.5
23.4
33.4
33.0
33.9
36.9
43.3
34.2
0.1
0
0
Values are given as percentages of patients.
In one-third of the patients, consent for omission of therapeutic options was documented (Table 7). Inpatient treatment ended with the death of the patient in 492 cases (45.3%) and with discharge in 561 patients (51.6%, not documented 3.1%). In the internal medicine units, 44% of the patients were treated until death, compared to 55% of patients in the anesthesiology units. Mean duration of inpatient treatment was 13.8 12.1 days (death: 12.1 11.6 days; discharge: 15.2 12.3), with a range of 0 to 111 days. There were no differences between anesthesiology and internal medicine units, though the mean duration of treatment was longer for the surgical and the two radiotherapy units. Patients were discharged home in most cases; only 11.8% of the patients were discharged to other inpatient units (hospices 2.9%, nursing home 4.0%, other hospital 4.9%). For patients discharged home, outpatient palliative care services were available for 6.5%, and outpatient hospice services were available for 3.1%. In most cases, the general practitioner (29.3% of the patients), the family (34.5%), or the patient themself (15.5%) was named as the primary care giver for the patient after discharge (multiple entries 34.5%). The efficacy of treatment in the palliative care units was rated very high by the staff (Ta-
ble 8). The percentage of patients with mild or no residual pain was reduced from 44.5% at the time of admission to 85.0%. Good or very good symptom control was documented for 23.7% at the time of admission versus 66.9% at the end of inpatient treatment. More than half (50.2%) of the patients were reported to be satisfied or very satisfied with their situation at the time of admission versus 83.2% at the end of inpatient treatment. There were no differences related to the affiliation of the units. For 159 patients, self-assessments by the patients with the minimal documentation systems for palliative care (MIDOS) were available. Assessment of pain intensity with MIDOS at the time of admission correlated highly with pain intensity in the core documentation (r 0.73). Symptom intensity sum values in MIDOS showed no correlation with symptom intensity in the core documentation, and only minimal correlation with intensity of the leading symptom (r 0.29), but some correlation with performance status (r 0.46) and nursing requirements (r 0.44). Anxiety in MIDOS was correlated to psychic stability in the core documentation (r 0.47) and well-being in MIDOS with the ECOG performance status (r 0.44). Correlation of the differences between admission and discharge of the sum values of MIDOS with the differences of the efficacy ratings
476
Radbruch et al.
Vol. 23 No. 6 June 2002
Table 3 Reasons for Admission and Symptoms at the Time of Admission Patients (%)
Reasons for admission Pain Physical symptoms Nutrition Terminal care Nursing problems Social problems Psychological problems Others Multiple responses Symptoms at time of admission Weakness Pain Appetite loss Nausea Cachexia Dyspnea Constipation Neurological symptoms Anemia Edema Psychiatric symptoms Sleeping disorders Cough Ascites Dysphagia Urinary disorders Dyspepsia Fever Diarrhea Tumor bleeding Pruritus Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
57.5 54.9 36.2 15.5 15.2 14.2 13.8 18.9 73.6
54.6 50.0 33.5 12.4 21.7 21.1 16.0 12.4 72.2
57.3 56.8 37.1 15.1 14.4 12.8 12.8 19.2 73.1
80 80 44 8 8 4 20 8 92
16 8 7 0 0 0 0 0 26
88 93 83 62 59 41 43 65 100
76.5 64.6 49.6 36.8 32.9 29.4 24.2 19.6 16.9 15.6 15.0 11.4 11.0 8.9 8.7 7.5 6.2 5.2 5.2 4.0 2.8 10.2 1.3 93.7
80.9 66.0 52.1 41.8 38.7 34.0 30.4 21.1 15.5 22.7 18.6 21.1 10.3 7.7 8.8 11.3 4.6 6.2 4.6 2.1 3.6 8.2 0 94.8
76.1 63.0 50.1 37.1 30.8 27.0 24.5 19.4 17.7 14.2 14.5 9.8 11.5 9.4 8.9 7.4 7.1 5.2 5.7 4.0 2.8 10.7 1.4 93.3
80 80 48 28 12 36 20 28 0 4 12 8 8 4 0 0 4 0 4 0 0 12 0 96
46 21 20 4 0 8 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 67
97 96 85 85 72 80 50 41 52 42 36 50 45 30 28 38 52 20 18 16 12 41 18 100
Surgical (n 25)
Minimum
Maximum
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
in the different dimensions were calculated. Pain relief in MIDOS correlated with differences of pain intensities (r 0.52) in the core documentation from these patients. Relief of symptoms other than pain in MIDOS correlated only marginally with performance status (r 0.26) and psychological stability of the patient (r 0.24). Documentation discipline was investigated with several indicators. In 15 checklist questions of the core documentation instrument, multiple entries were possible, for a total number of 153 items. The sum of items chosen ranged from 3 to 65 in all patients. The mean number of items used in the patients of the different units ranged from 19 to 48, with no differences between anesthesiology (mean 33.8 items) and internal medicine units (mean 32.3
items). The surgical unit showed no differences in the mean number of items (34 items), though an emphasis on the documentation of certain areas, such as pain and other physical symptoms, reasons for admission, drug therapy, or psychosocial interventions for patients and relatives can be seen in the tables. For 140 patients, no locations for metastases were given, but for only 66 of these patients (47%) the item “no metastases” was answered positively. On the other hand, for 5 patients this item was answered affirmatively, though at least one location for metastases had been confirmed.
Discussion Data on palliative care from cross-sectional and multinational surveys have been collected
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
477
Table 4 Concomitant Diseases and Metastases at the Time of Admission Patients (%)
Concomitant diseases Cardiovascular diseases Metabolic disorders Lung diseases Neurological disorders Movement-related disorders Gastrointestinal tract diseases Urogenital tract diseases Other tumors Infections Head and brain diseases Skin diseases Other diseases Multiple responses Metastases Liver Bone Lung Lymph nodes Peritoneal Central nervous system Pleura Skin Gastrointestinal tract Urogenital tract Adrenal glands Breast Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
31.9 14.7 13.6 11.9 10.4 9.8 7.9 8.0 5.3 4.9 2.7 6.8 37.0
32.5 12.9 20.1 13.4 15.5 13.4 9.3 6.2 7.7 4.6 1.5 8.2 44.3
32.6 15.1 11.5 11.8 7.8 3.2 7.8 8.8 5.4 4.9 3.2 6.6 36.0
28 16 8 20 20 4 20 16 0 16 4 12 48
4 0 0 0 0 0 0 0 0 0 0 0 4
54 36 52 39 40 44 30 24 29 21 14 26 72
32.6 30.8 25.7 24.3 18.1 13.5 9.3 6.7 6.1 4.9 2.4 1.8 9.9 6.5 55.0
35.1 34.5 23.7 23.7 11.3 15.0 5.2 6.7 7.7 4.1 1.5 2.1 10.3 4.1 52.6
30.9 29.2 24.6 25.0 19.4 13.6 10.2 6.9 5.8 5.0 2.2 1.8 9.4 7.5 53.7
28 36 28 12 16 20 4 0 8 0 8 0 12 16 60
4 0 7 0 0 0 0 0 0 0 0 0 0 0 13
63 65 52 63 43 35 44 24 24 23 15 16 37 39 88
Surgical (n 25)
Minimum
Maximum
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
before.11 However, even though these surveys included a large number of centers, they did not attempt to deliver representative data for the countries involved. National surveys have collected data on structural quality of palliative care service,12–14 or on views and attitudes of patients, relatives, or staff,15–18 but articles on the palliative care delivered to the patients and its outcome usually come from a single unit or service (see reviews19,20). As 44 of the 65 palliative care units in Germany participated in this survey, this is the first report of representative longitudinal data on the practice of inpatient palliative care in a European country. The participating units comprised 72.2% of the palliative care beds in Germany. Inpatient hospices were not included in the survey, as these units have a different function in palliative care in Germany. Whereas palliative care units are affiliated with a hospital, have physicians on their regular staff, and are
budgeted along with the other hospital beds, receiving full funding from the national health insurance system, hospices usually stand alone without hospital affiliation and their staff does not include physicians. Reimbursement from the health insurance system is much less for hospice beds than for palliative care beds, and patients must pay part of the costs in many hospices. Palliative care units try to treat symptoms and psychosocial problems and discharge patients at home or to a hospice if possible, whereas hospices often have an expected survival time of less than six months as an admission criterion, and care for their patients in the final phase of the disease until death. The need for standardized assessment tools and documentation systems as a prerequisite for effectiveness in palliative care is widely acknowledged. This is demonstrated by the large number of German palliative care units that participated in this survey. In preliminary dis-
478
Radbruch et al.
Vol. 23 No. 6 June 2002
Table 5 Diagnostic and Therapeutic Procedures During Inpatient Stay in Palliative Units Patients (%)
Diagnostic procedures Laboratory investigations Radiology Ultrasonography Electrocardiography CT scan Endoscopic procedures MR scan Others None Multiple responses Therapeutic procedures Chemotherapy Radiotherapy Immunotherapy Hormonal therapy Surgical procedure Drug treatment for symptom control Infusions Blood transfusions Skin and wound care Urinary catheter Inhalation therapy Central venous line Gastric tube, PEG Pleural aspiration, pleurodesis Port Ascites paracentesis Anus praeter Tracheostoma Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
80.6 42.0 38.0 31.0 13.7 10.6 2.1 8.7 12.1 61.3
76.3 30.9 21.1 9.8 8.2 10.3 1.0 3.6 17.0 42.3
83.1 45.7 45.3 40.8 13.3 11.5 1.9 10.9 9.6 68.0
92 12 0 0 4 0 0 0 8 16
9 3 0 0 0 0 0 0 0 4
100 100 100 96 68 60 14 56 72 100
8.7 5.1 4.6 4.9 1.1 80.0 46.7 14.3 23.6 21.1 9.5 7.0 4.8 4.4 4.2 2.7 1.9 1.0 11.5 1.9 72.3
1.5 4.1 11.3 5.7 1.0 79.9 40.7 10.8 37.1 22.2 7.7 5.7 3.1 3.1 4.1 2.1 1.5 5.2 9.3 3.1 65.5
10.6 4.1 3.5 4.8 1.2 80.1 48.6 16.6 20.1 20.1 11.1 7.5 4.8 4.5 4.3 2.7 1.8 1.0 11.6 1.6 74.1
0 8 0 0 0 96 56 0 40 56 0 20 24 4 16 0 12 4 12 0 88
0 0 0 0 0 22 4 0 0 0 0 0 0 0 0 0 0 0 0 0 25
44 48 88 24 8 100 93 40 68 63 36 47 24 16 31 21 12 9 43 10 100
Surgical (n 25)
Minimum
Maximum
Boldfaced numbers are items with differences between patients treated in anesthesiology and internal medicine units.
cussions, personal remarks from the participants underlined the importance they placed on assessment, quality assurance, and the need for good standardized documentation tools. Currently, no recommendations for standardized assessment and documentation are available in Germany, though a tool for patients’ self-assessment has been validated and computer-based documentation systems have been described.4,5,21 The core documentation group may also be regarded as the first step to establish clinical audit in German palliative care units. In the preliminary discussions in the working group, the main areas of concern were identified and agreed upon among the participating units. Documentation and assessment of the outcome was then performed with the core documentation and the results were fed back to the participants in subsequent
meetings of the working group, thus completing the audit cycle.22 During the recent past, the quality and the availability of palliative care beds has developed rapidly in Germany.1 However, with the growing number of units, concerns have been raised about the quality of the services.23 In this survey, only 6 units had a ratio of nursing staff to inpatient beds of 1.4:1 or more as recommended by the German Association for Palliative Medicine. The assessment at the time of admission showed symptoms related to reduced performance status in advanced cancer, such as weakness, appetite loss, and cachexia, to be predominant. High prevalences of symptoms, such as pain, nausea, dyspnea, and constipation, for which more specific treatment options are available, give an indication of undertreat-
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
479
Table 6 Psychosocial and Physiotherapeutic Interventions During Inpatient Palliative Therapy Patients (%)
Psychosocial interventions for the patient From physician From nurse From psychologist From social worker From chaplain From volunteers From others None Psychosocial interventions for relatives From physician From nurse From psychologist From social worker From clergy From volunteers From others None Multiple responses for psychosocial interventions Physiotherapeutic interventions Physiotherapy Change of position Mobilization Breathing therapy Lymphatic drainage, massage Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
Surgical (n 25)
Minimum
Maximum
57.6 52.6 28.3 28.8 34.7 17.4 1.5 7.8
75.7 71.5 24.4 26.4 41.5 18.7 0.5 7.3
56.0 50.5 27.6 29.5 33.1 16.3 1.2 8.5
100 100 4 8 8 44 0 0
0 0 0 0 0 0 0 0
100 100 92 89 100 88 16 43
46.9 37.8 13.7 18.6 13.6 5.5 4.7 8.0 78.2
69.4 58.6 15.5 20.2 16.6 9.8 0 6.2 84.0
43.9 34.4 12.4 18.6 12.4 4.3 0.5 9.2 77.5
84 84 4 0 0 28 0 0 100
0 0 0 0 0 0 0 0
92 92 71 74 50 56 4 43
33.9 32.1 30.4 19.5 16.8 4.9 21.0 41.1
29.9 32.5 25.8 15.5 27.8 8.8 21.1 42.8
33.0 34.7 32.2 21.1 14.0 3.8 20.3 41.5
84 24 28 0 4 0 0 32
0 0 0 0 0 0 0
88 78 88 64 59 47 76
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
ment. Symptom prevalences were in the range described in other trials.24–33 The range of symptom prevalences in these studies correlated well with the range of prevalences in the units described in our survey. Pain was documented in 38–84% of the patients in the literature, and in 21–96% of patients in the palliative care units in this survey. Likewise, nausea was found in 6–51% of patients in the literature and in 4–85% of patients in our units. In Germany, palliative medicine is not a recognized speciality. Physicians involved in palliative medicine mostly came from an anesthesiology background in the early years, and from training in internal medicine and oncology more recently. The great majority of palliative care units in Germany are affiliated with hospital departments, in most cases internal medicine or anesthesiology. The palliative care unit with the longest standing was established in 1983 at the University of Cologne and is the only one that is affiliated with the surgical department. Most units are staffed by physicians
from the department they are affiliated with, though in several cases, physicians with other speciality training were working in some palliative care units. Interdisciplinary cooperation is a major feature of palliative medicine and may be more difficult for units staffed only by physicians with the same speciality training. This also is evident in the data collected, as units affiliated with anesthesiology departments performed much less chemotherapy, but more immunotherapy; gave fewer infusions and blood transfusions but more skin and wound care, more lymphatic drainage and massage; and documented psychosocial interventions more frequently for patients as well as for relatives than those affiliated to internal medicine departments. This raises the question about the general aim of treatment in palliative care units. We did not calculate statistical differences for the other units, as the small number of these units and their patients would have prevented a meaningful evaluation, considering the large
480
Radbruch et al.
Vol. 23 No. 6 June 2002
Table 7 Consent to Omission of Therapeutic Measures Patients (%)
Procedures Resuscitation Chemotherapy Artificial ventilation Assisted feeding Surgical procedures Radiotherapy Dialysis, ultrafiltration Blood transfusions Hydration Drug therapy of concomitant diseases Others None Multiple responses
Range of Units (% patients)
All (n 1087)
Anesthesia (n 193)
Internal Medicine (n 765)
Surgical (n 25)
Minimum
Maximum
17.3 13.9 13.9 13.6 11.9 10.4 7.3 5.0 4.8 2.9 2.1 63.3 21.7
18.6 4.6 16.5 9.8 11.3 4.1 4.6 4.1 4.1 4.1 2.1 62.9 22.7
14.4 12.2 10.7 12.6 8.8 8.4 4.1 2.3 3.9 1.3 2.2 65.5 18.9
4 4 4 4 4 4 4 4 0 4 0 88 8
0 0 0 0 0 0 0 0 0 0 0 12 0
76 88 64 67 84 72 66 59 40 36 13 100 88
Boldfaced numbers are items with significant differences between patients treated in anesthesiology and internal medicine units.
variations among units. Differences among the specialities may have been related to differences in the patients treated. However, though the number of patients with malignant disease in the respiratory tracts showed significant differences between anesthesiology and internal medicine, this was not related to differences in the prevalence of specific respiratory symptoms.
Differences among the units and specialities in the prevalence of symptoms or therapies may be related to different attitudes towards documentation. Documentation discipline was investigated with several indicators. The number of items marked in those sections of the questionnaire where multiple entries were possible showed large differences, though the spe-
Table 8 Outcome of Inpatient Therapy Patients (%) All Patients (n 1087) Pain None Mild Moderate Severe As bas as can be Not documented Symptoms Very good symptom control Good symptom control Moderate symptom control Only little symptom control Symptoms not controlled Not documented Satisfaction of patients or relatives Very satisfied Satisfied Little satisfied Dissatisfied Very dissatisfied Not documented
Anesthesia (n 193)
Internal Medicine (n 765)
26.5 18.0 26.7 24.8 3.5 0.5
55.7 29.3 10.4 3.4 0.3 0.9
27.6 15.6 27.1 25.0 4.7
56.0 29.3 12.0 2.1 0.5
28.1 18.8 23.9 25.6 3.6
58.3 28.2 9.5 3.8 0.3
4.5 19.2 36.2 32.0 4.5 3.6
16.3 50.6 21.3 7.7 1.7 2.3
3.8 18.0 32.8 37.7 7.7
17.3 50.8 23.0 6.8 2.1
5.0 20.1 37.0 33.5 4.4
16.7 50.6 22.1 8.6 2.0
11.7 38.5 26.7 11.0 2.1 9.9
31.5 51.7 10.1 1.8 0.4 4.5
11.6 41.5 33.3 10.9 2.7
42.3 45.6 8.8 3.3
13.5 45.0 27.9 11.4 2.1
29.8 56.9 11.1 1.9 0.4
Values are given as percentages of patients. Boldfaced numbers are assessments at the time of admission; unboldfaced numbers are assessments at the end of inpatient treatment (discharge or death).
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
ciality of the units did not influence this number. Data from the surgical unit show that this unit may be taken as an example of specific emphasis on certain areas of the documentation. This could be due to different conceptions of the intent of the questions; it also could be related to areas of interest of the staff performing the documentation. Documentation discipline may be connected more to the attitudes of the persons completing the questionnaire than to specific traits of the unit or special patient populations. Omission of therapeutic options with the consent of the patient is one of the key features of palliative care. However, the intent of the question about consent for omission of therapy may not have been clear enough, and, consequently, wide variability again was documented. Rejection of resuscitation was documented for no patients in some units and in up to 78% of patients in others; consent on omission of further chemotherapy varied between 0% and 88% of patients in the different units. End-oflife decisions and advance directives are not common in Germany. Written advance directories were found only in 18% of the patients from German palliative care units, compared to 79% of patients in units in the United States.34 This may be due to a lack of information among patients and staff. More specific research is needed in this area in Germany, but the setting of the core documentation may not be suited for assessment of these questions. The outcome of inpatient palliative care was rated very high in the survey, as shown by the large percentages of patients with minimal or no residual pain, good symptom control, and high satisfaction with therapy. However, the documentation by the staff of the units may have been heavily biased, not only from overestimation of their own work, but also from patients reporting good results to please the staff of the unit. The correlation between self-assessment and assessment by staff were available only for few patients, and showed moderate correlations of outcome ratings at most, with the exception of a higher correlation for pain relief. Comparison of assessments by patients and staff at the time of admission showed higher correlations, at least for some items, but again demonstrated a lack of correlation between self-assessment and assessment by staff for symptom intensity. This underlines the im-
481
portance of the additional input from a selfassessment instrument in palliative care outcome assessment.22 The core documentation instrument used in this survey was comprehensive and extensive. Higginson and Hearn have suggested the Palliative Care Outcomes Scale (POS), with only 12 items.22 Several of these items (pain intensity, symptom intensity, anxiousness of patient or relatives, main problems in the last days, performance status) were integrated in the German core documentation. Other items from the POS, such as the ability to share feelings with friends or family or the time wasted on appointments related to health care, were not available in the core documentation, and might have given as much information as the more comprehensive sections of the core documentation. Use of shorter and more concise instruments such as the POS could have improved the correlation of staff ratings in the different units, as well as the correlation of self-assessment and assessment by staff. In conclusion, we found large variation in the documentation pertaining to palliative care patients at the time of admission, as well as for inpatient treatment among palliative care units in Germany. This was related to the affilitation of the units, at least to some degree, but also to differences in documentation discipline. We suggest the integration of training procedures for documentation in multicenter surveys, as the results may not be comparable otherwise. Training procedures could involve constructed case reports, detailed manuals, or training sessions. Questionnaires and documentation systems should be kept as short as possible, and clear rules for completion should be included in the questionnaires (e.g., “mark only items leading to significant impairment of function”). Following these recommendations, the core documentation for palliative care units may be developed into an instrument for audit, thereby improving the quality of palliative care in Germany considerably, and enabling standards for palliative inpatient care to be established.
Acknowledgments Part of the preparatory work and the data evaluation, as well as the development of an electronic data base for the core instrument,
482
Radbruch et al.
was supported by a financial grant from the Mundipharma Company, Limburg, Germany. We would like to thank the physicians, nurses, social workers, psychologists, and other staff members who spent much time and enthusiasm with the data documentation. Additional thanks are directed to the principal investigators of the participating centers, who participated in the development of the core instrument: Jan Hinnerk Stange, Palliativstation des Marienkrankenhaus, Bad Lippspringe; Imke Strohscheer, Palliativstation der Charité, Campus Virchow Klinikum, Berlin; Sabine Bothe, Palliativstation am Krankenhaus Spandau, Berlin; Annemarie Keller, Palliativstation am Malteser Krankenhaus, Berlin; Thomas Jehser, Interdisziplinäre Palliativstation am Gemeinschaftskrankenhaus Havelhöhe, Berlin; Christoph Ostgathe, Palliativstation am Malteser Krankenhaus, Bonn; Bernd Sittig, Palliativstation am St. Joseph Hospital, Bremerhaven; Christiane Glasmeyer, Palliativstation am St. -Johannes Krankenhaus, Dortmund; Barbara Schubert, Palliativstation am St. Joseph Stift, Dresden; Wolfgang Etspüler, Palliativstation der Vita Natura Klinik, Eppenbrunn; Thomas Montag, Palliativstation am Katholischen Krankenhaus, Erfurt; Bernhard Greiling, Palliativstation am St. Elisabeth-Krankenhaus, Eutin; Halves, Katharinen Hospiz am Park, Flensburg; Axel Elsner, Evangelisches Hospital für palliative Medizin, Frankfurt am Main; Dietmar Beck, Palliativstation der Universität Göttingen, Göttingen; Herbert Kaiser, Palliativstation am Städtischen Krankenhaus, Gütersloh; M. Kreiling, Palliativstation am St. Elisabeth Krankenhaus, Halle; Heike Schieder, Palliativstation des Allgemeinen Krankenhauses Barmbek, Hamburg; Hans-Joachim Lehmann, Palliativstation des Krankenhauses Rissen, Hamburg; Reiner Franke, Palliativstation des Kreiskrankenhaus, Hameln; Birgitt van Oorschot, Interdisziplinäre Onkologische Palliativabteilung an der Friedrich-Schiller-Universität, Jena; Wolfgang Spuck, Palliativstation des Rotes-Kreuz-Krankenhaus, Kassel; Hermann Ewald, Palliativstation der Universitätsklinik, Kiel; Ute Welling, Palliativstation des Universitätsklinikums, Köln; Waltraud Richter, Palliativstation des Bezirkskrankenhaus St. Georg, Leipzig; Joachim Steuber, Palliativstation am St. Elisabeth-Krankenhaus, Leipzig; G. Prillwitz, Palliativstation der Pfeiffersche Stiftun-
Vol. 23 No. 6 June 2002
gen, Magdeburg; Weiß, Palliativstation des Klinikum, Mannheim; Ulrike Münz, Johannes Hospiz am Krankenhaus, München; Claudia Bausewein, Palliativstation des Krankenhauses, München-Harlaching; Wolf-Diethard Pietruschka, Palliativstation des Klinikums, Neubrandenburg; H.-J. Wetzel, Palliativstation des FriedrichEbert-Krankenhauses, Neumünster; Martin Fuchs, Palliativeinrichtung am St. Elisabeth Krankenhaus, Neuwied; Otto Kloke, Palliativbereich am Elisabeth-Krankenhaus, Recklinghausen; Peter Ervens, Palliativstation des Krankenhauses Maria Stern, Remagen; Ulrike Thierbach, Palliativstation des Klinikum Südstadt, Rostock; G. Lenke, Palliativstation des Evangelischen-Freikirchlichen Krankenhauses, Rüdersdorf; Frank Förster, Palliativstation des Fachkrankenhauses Marienstift, Schwarzenberg; Karl Neuwöhner, Palliativstation der Klinik Dr. Hancken, Stade; Martin Surfleet, Palliativstation des Marienhospital, Stuttgart; Werner Richarz, Palliativstation des Herz Jesu Krankenhauses, Trier; Kertz, Palliativstation des St. Josef-Hospital, Troisdorf; Boris Hait, Palliativstation im Katharinen-Hospital, Unna; Andreas Geilen-Meerbach, Palliativstation am St. Michael Krankenhaus, Völklingen.
References 1. Sabatowski R, Radbruch L, Nauck F, et al. Über die Entwicklung palliativmedizinischer Einrichtungen in Deutschland. Zeitschrift für Palliativmedizin 2000;1:40–46. 2. Klaschik E, Nauck F, Radbruch L, Sabatowski R. Palliativmedizin—Definitionen und Grundzüge. Internist 2000;41:606–611. 3. Sabatowski R, Radbruch L, Nauck F, et al. Entwicklung und Stand der stationären palliativmedizinischen Einrichtungen in Deutschland. Schmerz, 2001;15:312–319. 4. Radbruch L, Sabatowski R, Loick G, et al. MIDOS—Validierung eines minimalen Dokumentationssystems für die Palliativmedizin. Schmerz 2000; 14:231–239. 5. Radbruch L, Sabatowski R, Loick G, et al. Cognitive impairment and its influence on pain and symptom assessment in a palliative care unit: development of a minimal documentation system. Palliat Med 2000;14:266–276. 6. Determann MM, Rzehak P, Küchler T, et al. Qualitätssicherung in der Palliativmedizin. Forum Deutsche Krebsgesellschaft 2000;15(2):10–13. 7. Sabatowski R, Radbruch L, Loick G, et al. Pallia-
Vol. 23 No. 6 June 2002
What Is Palliative Care in Germany?
tiv 2000—Stationäre und ambulanten Palliativ— und Hospizeinrichtungen in Deutschland. Limburg: Mundipharma GmbH, 1999. 8. Buccheri G, Ferrigno D, Tamburini M. Karnofsky and ECOG performance status scoring in lung cancer: a prospective, longitudinal study of 536 patients from a single institution. Eur J Cancer 1996; 32A(7):1135–1141. 9. Conill C, Verger E, Salamero M. Performance status assessment in cancer patients. Cancer 1990; 65(8):1864–1866. 10. Radbruch L, Sabatowski R, Loick G, et al. Midos— Validation of a minimal documentation system for palliative medicine. Schmerz 2000;14(4):231–239.
483
MIDOS—An electronic database for the palliative care unit. Schmerz 2000;14(4):257–263. 22. Higginson IJ, Hearn J. Palliative care audit: tools, objectives, and models for training in assessment, monitoring, and review. In: Portenoy RK, Bruera E, editors. Topics in palliative care. Oxford: Oxford University Press, 2000:95–118. 23. Nauck F, Ostgathe C, Lund S, et al. Qualitätssicherung in der Palliativmedizin. Ermittlung der Struktur- und Prozessqualität auf den Palliativstationen in NRW. Zeitschrift für ärztliche Fortbildung und Qualitätssicherung 2000;94:587-594. 24. Lichter I, Hunt E. The last 48 hours of life. J Palliat Care 1990;6(4):7–15.
11. Caraceni A, Portenoy RK, Ashby MA, et al. An international survey of cancer pain characteristics and syndromes. Pain 1999;82(3):263–274.
25. Reuben DB, Mor V, Hiris J. Clinical symptoms and length of survival in patients with terminal cancer. Arch Intern Med 1988;148(7):1586–1591.
12. Eve A, Smith AM, Tebbit P. Hospice and palliative care in the UK 1994–5, including a summary of trends 1990–5. Palliat Med 1997;11(1):31–43.
26. Brescia FJ, Adler D, Gray G, et al. Hospitalized advanced cancer patients: a profile. J Pain Symptom Manage 1990;5(4):221–227.
13. Sabatowski R, Radbruch L, Loick G, et al. Palliative care in Germany—14 years on. Eur J Palliat Care 1998;5:52–55.
27. Curtis EB, Krech R, Walsh TD. Common symptoms in patients with advanced cancer. J Palliat Care 1991;7(2):25–29.
14. Sabatowski R, Radbruch L, Müller M, et al. Ambulante Hospiz- und Palliativdienste in Deutschland. Zeitschrift für Palliativmedizin 2000;1:17–24.
28. Coyle N, Adelhardt J, Foley KM, Portenoy RK. Character of terminal illness in the advanced cancer patient: pain and other symptoms during the last four weeks of life. J Pain Symptom Manage 1990;5(2):83–93.
15. Jarrett N, Payne S, Turner P, Hillier R. ‘Someone to talk to’ and ‘pain control’: what people expect from a specialist palliative care team. Palliat Med 1999;13(2):139–144. 16. Morize V, Nguyen DT, Lorente C, Desfosses G. Descriptive epidemiological survey on a given day in all palliative care patients hospitalized in a French university hospital. Palliat Med 1999;13(2):105–117. 17. Peruselli C, Di Giulio P, Toscani F, et al. Home palliative care for terminal cancer patients: a survey on the final week of life. Palliat Med 1999;13(3):233–241. 18. Sebastian P, Varghese C, Sankaranarayanan R, et al. Evaluation of symptomatology in planning palliative care. Palliat Med 1993;7(1):27–34. 19. Rinck GC, van den Bos GA, Kleijnen J, et al. Methodologic issues in effectiveness research on palliative cancer care: a systematic review. J Clin Oncol 1997;15(4):1697–1707. 20. Hearn J, Higginson IJ. Outcome measures in palliative care for advanced cancer patients: a review. J Public Health Med 1997;19(2):193–199. 21. Radbruch L, Loick G, Sabatowski R, Elsner F.
29. Ventafridda V, Ripamonti C, de Conno F, et al. Symptom Prevalence and control during cancer patients’ last days of life. J Palliative Care 1990;6(3):7–11. 30. Seale C. Death from cancer and death from other causes: the relevance of the hospice approach. Palliat Med1991;5:12–19. 31. Grond S, Zech D, Diefenbach C, Bischoff A. Prevalence and pattern of symptoms in patients with cancer pain: a prospective evaluation of 1635 cancer patients referred to a pain clinic. J Pain Symptom Manage 1994;9:372–382. 32. Twycross RG, Lack S. Control of alimentary symptoms in far advanced cancer. Edinburgh: Churchill Livingstone, 1986. 33. Walsh D, Donnelly S, Rybicki L. The symptoms of advanced cancer: relationship to age, gender, and performance status in 1,000 patients. Support Care Cancer 2000;8(3):175–179. 34. Voltz R, Akabayashi A, Reese C, et al. End-of-life decisions and advance directives in palliative care: a cross-cultural survey of patients and health-care professionals. J Pain Symptom Manage 1998;16(3):153–162.