Background
Corresponding to the Federal Statistical Office about 1% of the population in Germany dies within a year [
1] and studies as well as the German Association for Palliative Medicine (DGP) estimate that about 90% of these require some level of palliative care [
2‐
5]. To facilitate understanding of palliative care provision in the German health care system in the international community, we use the terminology defined by the European Association for Palliative Care (EAPC) in the “White Paper on standards and norms for hospice and palliative care in Europe” [
6,
7]. According to the EAPC, palliative care can be made available on three different levels: palliative care approach, general palliative care and specialist palliative care – whilst palliative care approach and general palliative are often dealt as a joint category when focusing on the contrast to specialist palliative care, as we will do during this study protocol. While most patients with a life-limiting disease may only require general palliative care as provided by primary care professionals or specialists, a substantial amount of patients suffer from a high burden of symptoms and therefore require specialist palliative care. In case of outpatient care in Germany, general palliative care can be provided as General Outpatient Palliative Care (AAPV) and patients with a need for specialist palliative care can be referred to Specialized Outpatient Palliative Care (SAPV). SAPV is regularly provided by multi-professional home palliative care teams, comprising at least physicians and nurses, delivering specialized palliative care at the homes of patients with complex needs and suffering from incurable, progressive and life-limiting diseases.
Since the German Act to Strengthen Competition in Statutory Health Insurance (GKV-WSG) was introduced in 2007, SAPV has been fully covered by the Statutory Health Insurance [
8] and can be prescribed by outpatient physicians as general practitioners (GPs),or by hospital physicians. Eligibility criteria as well as type and extent of SAPV services are determined by the SAPV directive of the Federal Joint Committee (G-BA, the highest decision-making body of the joint self-government of physicians, dentists, hospitals and health insurance funds in Germany) [
9]. The SAPV directive defines objectives of SAPV (§1), eligibility requirements (§2, §3), particularly complex care (§4), content and scope of SAPV (§5), collaboration with other care providers (§6), prescription of SAPV (§7) and the assessment of entitlement for benefits by health insurance funds (§8). This directive has been worded in general terms to enable health insurance funds and care providers to make individual agreements (selective contracting), based on regional specifications and already existing structures. The resulting regulations are highly variable regarding care provision, cooperation and quality assurance as well as financial reimbursement and has led to different models [
10]. This is reflected in over 290 different SAPV contracts nationwide [
11], so that team structures, cooperation with other care givers, patient inclusion and exclusion criteria, processes of care, treatment intensity and financing models may differ substantially between the SAPV teams.
In 2016 about 55.000 patients (6% of the deceased in Germany) had received SAPV [
11] before their death. This proportion is well below the estimated 10% of dying patients with a need for specialist palliative care at home [
2]. However, recent analyses in Germany focused mainly on the description of (regional) SAPV-structures and utilization [
12‐
16] as, for instance, the study “Faktencheck Gesundheit” investigated regional disparities of SAPV prescriptions [
17]. Until now, no nationwide assessment of SAPV from different perspectives has been conducted [
18]. In our view, there is a special need for an evaluation of the SAPV directive taking into account:
1)
The heterogeneity of organizational structures and processes and their effects on patient care: Heterogeneity is particularly represented by different forms of ownership, team structure, integration of different professions, reimbursement systems and regional factors like urban or rural settings [
14]. This also leads to different levels of involvement of the SAPV teams (consultation, coordination of care, partial or full provision of care). The potential consequences of these different contractual elements on patients´ care are mostly unknown.
2)
As palliative care is a multi-professional, patient-centered care model, different perspectives of affected persons and involved professionals should be considered [
19].
a.
Although the individual needs of patients and their relatives are key aspects of the SAPV directive, there are no formal requirements for a corresponding quality assurance. Available research does not focus on the patient perspective at all [
17] or only does so in specific scenarios [
14,
20]. Assessing quality of care without consideration of the patients perspective will remain incomplete [
21]. With the help of the questionnaire QUAPS (Quality Assurance in Outpatient Palliative Care) we will be able to investigate quality of SAPV from patients’ perspective [
22].
b.
Since SAPV is acting in a complex social environment of patients, relatives and health care professionals, the evaluation of overall quality of palliative care requires an adapted and differentiated analysis considering cooperation and perspective of these stakeholders. Accordingly, SAPV is a process of interactions of these participants, located in the individual patient’s environment. Therefore it is necessary to describe the local cooperation of health care professionals.
c.
General practitioners (GP) are substantially involved in the end of life care of their patients [
23]. They play an important role in AAPV, can introduce SAPV to patients’ care by medical prescription and often remain subsequently involved [
24]. This key role strengthens the need of assessing the impact of GPs on the quality of SAPV in connection to patients, other health professionals and the SAPV directive.
3)
The interface between general and specialist outpatient palliative care appears to be under-explored [
11,
25]. Furthermore, available claims data studies focused on a small selection of specific aspects of palliative care [
26]. Investigation of utilization and costs of different types of palliative care using claims data is a useful addition. On that basis it might be possible to find hints for over-, under- and missupply of end-of-life-care considering health economic perspectives.
4)
Against the background of the requirements of the national ethics council [
27], it is an important goal to highlight the patients’ welfare as the leading principle in medical care. Taking this into account, it is necessary to focus on potential shortage of outpatient palliative care in specific patient groups, for example patients with non-malignant disorders.
So far, a systematic and nationwide data assessment in SAPV is lacking. In order to support the further development of SAPV and the guiding SAPV directive, it is necessary to assess the process, extent and quality of SAPV and their determining factors. For this purpose, SAVOIR (Evaluation of the SAPV directive, Outcomes, Interactions and regional variations) was funded by the German Innovations Fund (launched by the Federal Joint Committee) to evaluate the execution of the SAPV directive and to present suggestions for a revision of the SAPV directive.
The project is performed by the Department of Palliative Care (coordinator), Institute of General Practice and Family Medicine (both Jena University Hospital), Clinic for Palliative Medicine (University Medical Center Göttingen), Center for Interdisciplinary Health Research (ZIG, University of Augsburg), German federal association for SAPV (BAG-SAPV), German Association for Palliative Medicine (DGP) and the health insurance fund BARMER.
Methods/design
Aim of the study
The project will study care provision in SAPV from the perspective of patients, SAPV teams, other care givers, GPs and payers. The influence of different contracts, team and network structures, regional and geographic settings on processes and results including patient-reported outcomes will be analyzed. This will be carried out in five subprojects:
Subproject 1: The aim is to investigate the structural characteristics of SAPV (determined by contracts and organizational team structures) and assess if and how these aspects influence patient care.
Subproject 2: The aim is to investigate the quality of care from the perspective of patients.
Subproject 3: The aim is to estimate the quality of care from the perspective of SAPV teams and external network partners such as hospices, ambulatory nursing services, nursing homes and other care givers.
Subproject 4: The aim is to investigate the type and extent of GPs palliative care with and without SAPV and the quality of palliative care from their perspective.
Subproject 5: The aim is to describe and distinct different types of palliative care, the patients receiving these types of care, their health care utilization and quality of care in its regional variations by means of health insurance fund claims data.
Primarily, mixed-methods data analysis will allow the identification of variables that influence care provision and may be associated with quality of SAPV. Based on these findings, the SAVOIR study group will provide recommendations for the revision of the SAPV directive for the Federal Joint Committee.
Design and setting
The project will be based on different data sources: subproject 1a) registry data on team and organizational structures from the German Guide for hospice and palliative care services as well as model contracts in all German regions and 1b) clinical routine patient data from electronic medical record (EMR) of a sample of SAPV teams, subproject 2) patient-reported outcomes in a sample of about 50 SAPV teams, each collecting data from 25 patients, subproject 3) qualitative interviews with members of SAPV teams and linked providers involved in SAPV, like nursing and hospice services, subproject 4) survey data from GPs and subproject 5) health insurance fund claims data from patients who have died in 2016 and were covered by the health insurance fund BARMER.
This study is conducted in Germany from July 2017 to June 2019 and includes a transdisciplinary and multi-method design. The overall project is characterized by a cross-project use of data and a combination of quantitative and qualitative research designs. The focus of the quantitative research is obtaining a representative cross-section of the German SAPV, for instance by mapping the 17 Regional Associations of Statutory Health Insurance Physicians (KV). These regions are equivalent to the 16 Federal German States, but North Rhine-Westphalia is divided into North Rhine and Westphalia-Lippe.
From a sample of about four SAPV teams per KV, aggregated data of approximately 70 patients per SAPV team are associated to structural characteristics of the selected teams (subproject 1). Following this, a patient survey of a sub-group of about 25 patients per SAPV team will be conducted, focusing on the quality of care from the perspective of patients and on linking these findings to individual demographic and clinical data (subproject 2). A postal survey of about 1200 GPs using a standardized, quantitative study design in eight KV domains is conducted to explore their involvement in and needs regarding SAPV (subproject 4). An analysis of palliative care based on claims data [
17,
28], provided by the health insurance fund BARMER, is performed to represent the utilization of the different types of in- and outpatient palliative care services in Germany (subproject 5). Data which cannot be obtained via quantitative research designs – such as interpretations and evaluation categories of quality of care by SAPV teams and their network of care givers and service providers – will be explored by a qualitative design. This is carried out as an on-site field study involving comparative case studies. Ten SAPV teams, each of which represents a minimal or maximal type within a set of contrasting structural characteristics, will be explored by means of field trips of several days duration (subproject 3). Wherever possible, synergies between the SAVOIR subprojects will be used. This involves the exchange of data (e.g. regarding the empirically grounded selection criteria for SAPV teams) and cooperation regarding content and concept of the subprojects.
Data management
The data management and all required steps of data preparation (for example check for plausibility and integrity) will be located at the Center for Clinical Trials at University Hospital Jena. All participant data will be pseudonymized. In addition, data collected from the subprojects will be managed at the corresponding institution: in subproject 4, data from the survey will be anonymously evaluated. The claims data analysis in subproject 5 will be based on the recommendations of the “Good Practice Secondary Data Analysis” [
29]. The data management of subproject 3 will be located at the Center for Interdisciplinary Health Research (ZIG) at the University of Augsburg.
Ethics
Votes from the local ethics committees of the respective institutions were obtained from Jena University Hospital (No. 5312–10/17, 5316–10/17, 5317–10/17), University Medical Center Göttingen (No. 31/8/17) and University of Augsburg (06.09.2018).
Sample and procedure
Subproject 1
Structural characteristics of SAPV and their impact on quality of care.
Subproject 2
Quality assurance in SAPV – a quantitative collection of data from patients´ perspective.
Subproject 3
Beneficial and inhibiting factors for a successful performance of SAPV teams from the perspective of SAPV team members and network partners.
Subproject 4
General Practitioners‘care for palliative patients within and outside SAPV (GPs‘perspective).
Hypotheses: The self-perception and tasks of GPs in palliative care in general and the prescription of SAPV by GPs in particular are determined by patient-, physician-, qualification-, payment-related and other environmental (i.e. palliative infrastructure) factors.
Subproject 5
SAPV in comparison to other types of palliative care: patient characteristics and resource utilization (payer’s perspective).
Hypotheses: The type, extent and quality of palliative care a patient receives during the last months before death differs according to patient characteristics as well as regional structures of palliative care provision.
Discussion
Recent analyses in Germany focused mainly on the description of (regional) SAPV structures and utilization [
12‐
15,
17]. A nationwide assessment of SAPV from different and relevant perspectives has not yet been conducted. To provide this, SAVOIR addresses: [
1] the structural characteristics of SAPV and their impact on patient care, [
2] the quality of care from the perspective of patients, [
3] the quality of care from the perspective of SAPV teams, hospices, ambulatory nursing services, nursing homes and other care givers, [
4] the content and extent of care from the perspective of GPs and [
5] the health insurance fund as a payer. The comprehensive, nationwide data assessment in the SAVOIR study uses different perspectives and data sources as well as both quantitative and qualitative methods. This allows us to identify the current conditions and special needs in palliative care from patients, relatives, care providers and payers in Germany. Subsequently we will compare different care models and provide an overview of SAPV provision in Germany and moreover we will provide an in-depth analysis of differences - even if we will not be able to take the entirety of differences into account. Based on these findings, factors related with high quality of care in SAPV will be identified.
Although SAVOIR uses multi-method and transdisciplinary approach with unique advantages, clinical research in patients with an advanced and life-limiting disease remains challenging, i.e. with regard to suitable research methods as well as the participation of healthcare professionals and the recruitment of patients [
31‐
34].
According to the different research strategies in each subproject, we are conscious of the following limitations:
Subproject 1: In step 1a, teams may not provide their real personnel structures, but rather a socially desirable version. The interviews were conducted via phone, which may increase this effect. The regulatory contracts may not cover the “real care provision”, as there may be regional differences that were not captured in the interviews and that we are unaware of. Some regions have model contracts, but specific characteristics like financial reimbursement may still vary between teams. To assess patient care, we use clinical routine data. While our prospective assessment enables us to give the teams information about the extracted data in advance and thus provide a better data quality, clinical routine data still is difficult to interpret. The variables used may be differently understood by those who document them and thus result in various interpretations within and between collecting SAPV teams.
Subproject 2: Questioning palliative patients and their relatives can only be pursued under several limitations: Patients are often in a poor general condition, participation in research is an additional burden for them and palliative care teams are often not experienced in clinical research and are reluctant to address participations. A specific challenge is given by the fact that life expectancy in SAPV patients is often limited to days or few weeks and treatment ends in most cases with the death of the patient. Therefore, in contrast to most other types of care, it is not possible to approach patients at the end of SAPV treatment. On the other hand, patients have to have experienced a certain time of SAPV care before they can report on its quality. Therefore, there is only a small time window to obtain patient-reported information on quality of SAPV care [
35]. This may reduce the sample size and those participating may be a selected group of relatively “healthy” patients. However, pilot testing has revealed that the chosen approach will result in a sufficient number of data sets. Therefore we conduct a multi-method design, collecting treatment data of all patients agreeing to participate in subproject 1 over the time of recruitment for subproject 2.
Subproject 3: The field survey may lead to self-selection of participating SAPV teams: We expect that SAPV teams interested in research and improvement of palliative care will be more likely to participate. A measure to prevent self-selection of SAPV teams is the purposeful selection of participants via a set of various criteria such as the size of the team, economic prosperity of region, population density, organizational contexts, etc. Furthermore SAPV teams themselves select external network partners for the interviews. We expect that only external network partners are chosen for participation who are highly motivated, interested in palliative care and have positive experiences in working together with SAPV teams. Especially GPs with a more reserved attitude towards SAPV are unlikely to participate. This aspect is methodologically reflected during the data analysis.
Subproject 4: In the GP survey we have to deal with the fact that GPs more interested, qualified and/or experienced in end-of-life care might be more willing to participate in the survey than others. Therefore, we request information on the extent, experience and qualification in terms of palliative care from the GPs to be able to determine the impact of these factors on the results.
Subproject 5: Though the BARMER made great efforts to collect all relevant data on SAPV and to transfer it into the Scientific Data Warehouse, the claims data base on SAPV will not be fully complete by the end of the project and may not cover all existing health services within SAPV in Germany. The data do not contain full information on the qualification levels of care providers necessary to finer differentiate between levels of palliative care such as palliative care approach and general patient care. Apart from many advantages in analyzing claims data, their information value for the measurement of quality of care is restricted due to the lack of clinical and patient reported data.
To sum up, SAVOIR – consisting of five subprojects – will provide important information on the complex and regionally varying structures, processes, extent and quality of SAPV in Germany and the factors which determine the quality of SAPV from the different perspectives of stake-holders involved. We aim at deriving recommendations to further develop the SAPV directive. The results will add valuable knowledge on starting points to improve outpatient palliative care, mainly in Germany, that might be applicable for other countries, too.
Acknowledgements
SAVOIR Study Group: Anna Bauer, Lia Bergmann, Bianka Ditscheid, Cornelia Eichhorn, Antje Freytag, Elke Gaser, Michaela Hach, Ulrike Hammer, Aicko Helbig, Beata Hennig, Maximiliane Jansky, Michelle Kaufmann, Markus Krause, Sabine Krauss, Thomas Lehmann, Helmut L’hoest, Srikanth Maddela, Ursula Marschall, Martial Mboulla, Winfried Meißner, Heiner Melching, Cornelia Nageler, Friedemann Nauck, Sara Parhizkari, Judith Rothaug, Joachim Saam, Werner Schneider, Sven Schulz, Kathleen Stichling, Horst C. Vollmar, Julia von Hayek, Ulrich Wedding.