To examine whether applying case management in general practices reduces thromboembolic events requiring hospitalization and major bleeding events (combined primary outcome). Secondary endpoints were mortality, frequency and duration of hospitalization, severe treatment interactions, adverse events, quality of anticoagulation, health-related quality of life and intervention costs, patients’ assessment of chronic illness care, self-reported adherence to medication, GP and HCA knowledge, patient knowledge and satisfaction with shared decision-making.
Cluster-randomized controlled trial undertaken at 52 general practices in Germany with adult patients with a long-term indication for oral anticoagulation. The complex intervention included training for healthcare assistants, information and quality circles for general practitioners and 24 months of case management for patients. Assessment was after 12 and 24 months. The intention-to-treat population included all randomized practices and patients, while the per-protocol analysis included only those that received treatment without major protocol violations.
The mean (SD) age of the 736 patients was 73.5 (9.4) years and 597 (81.1%) had atrial fibrillation. After 24 months, the primary endpoint had occurred in 40 (11.0%) intervention and 48 (12.9%) control patients (hazard ratio 0.83, 95% CI 0.55 to 1.25; P = .37). Patients’ perceived quality of care, their knowledge, and HCAs’ knowledge, had improved significantly at 24 months. The other secondary endpoints did not differ between groups. In the intervention group, hospital admissions were significantly reduced in patients that received treatment without major protocol deviations.
Citation: Siebenhofer A, Ulrich L-R, Mergenthal K, Berghold A, Pregartner G, Kemperdick B, et al. (2019) Primary care management for patients receiving long-term antithrombotic treatment: A cluster-randomized controlled trial. PLoS ONE 14(1): e0209366. https://doi.org/10.1371/journal.pone.0209366
Editor: Hugo ten Cate, Maastricht University Medical Center, NETHERLANDS
Received: July 20, 2018; Accepted: November 30, 2018; Published: January 9, 2019
Copyright: © 2019 Siebenhofer et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: We uploaded an additional minimal data set to Data Archiving and Networked Services (DANS) which is available at https://doi.org/10.17026/dans-2z5-pbz6.
Funding: This study was funded by the German Federal Ministry of Education and Research (grant number 01GY1145). The funding body had no role in study design, data collection, data analysis, data interpretation, writing the report, or the decision to submit this paper for publication. None of the authors have been paid by a pharmaceutical company or another for-profit organization to write this article. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Competing interests: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Juliana J. Petersen has been a co-investigator in the PANORA study (‘Prevalence of anti-cyclic citrullinated peptide (CCP) positivity in patients with new non-specific onset of musculoskeletal symptoms, possibly related to early rheumatoid arthritis in general practices in Germany’), which is being conducted by the Fraunhofer Institute and financed by Bristol‐Meyer Squibb. She is employed by the Institute of General Practice of Goethe-University Frankfurt and has never personally received financial remuneration from a pharmaceutical company. Karola Mergenthal is employed as project manager in the PANORA study. She has also never personally received financial remuneration from a pharmaceutical company. Andrea Siebenhofer received funding from the Federation of Austrian Social Insurance Institutions (HVB) for the preparation of a systematic review on self-management of oral anticoagulation in 2014 and was financed by ROCHE Diagnostics to carry out a study on self-management of oral anticoagulation from 2002-2005. Sebastian Harder has received honoraria for scientific lectures from Boehringer Ingelheim GmbH, Pfizer GmbH, Daiichi Sankyo GmbH, and Bayer AG. The other authors declare no competing interests. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Abbreviations: Co-MoL, Coagulation-Monitoring-List; DOACs, direct oral anticoagulants; GP, general practitioner; HCA, healthcare assistant; INR, international normalized ratio; ITT, intention to treat; OAC, oral anticoagulation; OR, odds ratio; PP, per protocol; SD, standard deviation; VKA, vitamin k antagonists
Oral anticoagulation (OAC) has been shown to be highly effective in preventing thromboembolic complications in patients for whom it is indicated. In antithrombotic treatment, vitamin K antagonists (VKAs) have been the agent of choice for several decades. VKAs carry a considerable risk of adverse thromboembolic and bleeding events, particularly in the case of dose deviations when international normalized ratio (INR) values are outside the target range . However, when patients are able to perform self-management, thromboembolic events and all-cause mortality are less frequent, and treatment-related quality of life rises [2,3].
Although subject to a number of concerns , direct oral anticoagulants (DOACs) are considered an effective alternative to VKAs in the long-term treatment of anticoagulation, and prescriptions have risen strongly since they were approved in 2011 .
Patients taking oral anticoagulation often suffer from multiple chronic conditions and have complex health care needs. Understanding and managing complex patients is a quintessential feature of primary care . Organizing services to improve care for these patients has been identified as a priority for the health care system, and especially for primary care research .
In Germany, management of patients taking OAC is typically carried out by general practitioners (GPs) in their practices. They generally employ one or more healthcare assistants (HCAs), whose role is comparable to medical assistants in the United States. In small primary care settings, resources are often limited and extensive collaborative models may be difficult to implement. The effectiveness of programs that expand the role of healthcare assistants in primary care to include chronic care services, such as case management in patients with depression [8,9] and chronic heart failure  have shown positive effects.
The aim of this study was to improve antithrombotic management in primary health care by having a healthcare assistant perform major elements of case management, and testing its effectiveness in reducing thromboembolic events requiring hospital admission, and major bleeding events.
Study design and population
The primary care management for optimized antithrombotic treatment (PICANT) study was an open cluster-randomized controlled trial undertaken at 52 general practices in Germany. The trial was registered at ISRCTN41847489 and approved by the ethics committee (E 191/11) of Frankfurt University Hospital on June 26, 2012. The study protocol and the practice recruitment process is described in detail elsewhere [11,12]. In brief, we identified potentially eligible practices from a list provided by the Association of Statutory Health Insurance Physicians (mandatory registration of GP practices). As the list only contains the names and addresses of GPs, we mailed information on the trial to 568 randomly selected practices (6% of all registered practices in 2012) and invited them to participate. Inclusion criteria were only checked for those who were interested in participation. Practice recruitment was stopped when 52 practices had enrolled, even though further practices were interested in participating.
Each participating practice was visited after practice recruitment but before cluster randomization, and asked to generate a screening list of potentially eligible patients. The practice software was used to generate the lists, and each practice was advised by study team members based on predefined instructions and search terms . The GPs then checked the lists and deleted cases of patients that had only been seen occasionally, or had died in the meantime. Inclusion criteria were then assessed by the GP and study team for 30 randomly selected patients from the list, with the aim of recruiting 15 patients per practice for the study. A documentation sheet was filled in for each screened patient. To avoid selection bias, the order of the patients assessed for eligibility was chosen using the random number generator function in Microsoft Excel.
Patients were eligible for inclusion if they were ≥ 18 years of age, had a long-term indication for oral anticoagulation based on the guidelines valid at the time, and were prescribed coumarins, antiplatelet therapies, or the DOACs that were on the market when the study began (dabigatran, rivaroxaban). Exclusion criteria were dementia, diseases resulting in a life expectancy of < 6 months, psychosis, severe sight disorders or auditory defects, alcohol- or drug abuse, residence in institutions that did not allow study participation, and a lack of German language skills.
We obtained written informed consent from participants. The assessment occurred three times: at baseline and at follow-ups after 12 and 24 months.
Randomization and masking
After the baseline assessment had been completed, a member of the Institute of General Practice that had no further involvement in the study used the web-based randomization tool “Randomizer for Clinical Trials” (http://www.randomizer.at) to consecutively and randomly allocate practices to the intervention or routine care arm in a ratio of 1:1. Randomization was stratified according to the number of inhabitants in the postal area where the practice was located and using permuted blocks of size 8. The statisticians were blinded to group assignment during the analysis .
Before randomization, all practices were provided with the evidence-based “Anticoagulation” guideline for general practitioners prepared by the Guideline Group of the German state of Hesse, and a standardized information brochure for patients issued by the German College of General Practitioners and Family Physicians .
The complex intervention included the provision of additional tools and guidelines to GPs and their practice teams (for intervention details see S1 Table). During an interactive one-day workshop, HCAs were trained to carry out case management and educate patients , assess adherence to medication and symptoms in patients, and to regularly monitor them using the Coagulation-Monitoring-List (Co-MoL) . Furthermore, HCAs were taught to encourage patients to perform self-management whenever feasible. GPs and HCAs were both provided with information materials and fact sheets on phenprocoumon, dabigatran, rivaroxaban, acetylsalicylic acid, and clopidogrel. In addition, we telephoned GPs immediately after randomization and provided them with further information on case management. Quality circles to discuss the practical problems involved in anticoagulation and preparing individual case reports took place three times during the course of the trial (for further details on the intervention please see S1 Table and Siebenhofer et al. ). To help cover increased staff costs, intervention practice teams received €50 per enrolled patient and assessment.
For the duration of the trial, the patients in the control group continued to receive treatment-as-usual from their GPs, meaning that except the fact that GPs received the “Anticoagulation” guideline for general practitioners before randomization, no additional advice was given and no regular monitoring visits took place. After the trial had ended, a similar HCA workshop to that provided in the intervention group was offered to participating practices. A financial incentive of €25 per enrolled patient and assessment was granted.
The primary patient-relevant endpoint was the combination of all thromboembolic events requiring hospitalization and major bleeding complications (if more than one event occurred in a patient, only the earliest event was considered; for the exact definition of major bleeding and thromboembolic events see study protocol ). Two external, independent, and blinded reviewers cross-checked primary endpoints by assessing hospital discharge letters and case report forms (TG and MS).
The following key secondary endpoints were evaluated: all-cause and cause-related mortality rates, frequency and duration of hospitalization, number of recurrent strokes (ischemic and hemorrhagic stroke), major bleeding and thromboembolic complications (counting all events), number of patients with at least one potentially severe treatment interaction, total number of potentially severe treatment interactions involving oral anticoagulants, number of adverse events, quality of anticoagulation (i.e. time within therapeutic range) , health-related quality of life (EQ-5D) , and costs from the payer’s perspective (German statutory health insurance).
Further secondary outcomes were investigated to explain factors that may have influenced the intervention’s effectiveness: patients’ assessment of chronic illness care (PACIC short version) , self-reported adherence to medication (questionnaire by Morisky; sum score 0–4, with lower scores indicating lower adherence) , GP and HCA knowledge (self-developed knowledge questionnaire, sum score 0–12), patient knowledge (questionnaire developed by Hua, sum score 0–13) , and satisfaction with shared decision-making (Man-Song Hing test) .
Sample size was calculated using the primary combined endpoint. We anticipated an event rate of 15% in the routine care and 7.5% in the intervention group, an intra-cluster correlation coefficient (ICC) of 0.01, and 15 patients per practice. Using a chi-square test, 317 patients per group were required to detect the difference in event rates at a 5% significance level and a power of 80%. To allow for patient withdrawals and practice loss, we aimed to include 23 general practices and 345 patients per group. The sample size calculation was performed using nQuery Advisor 7.0. For the primary endpoint, the time from randomization to first thromboembolic event requiring hospitalization, or major bleeding complication, was analyzed using a Cox proportional hazards model with robust sandwich estimates to account for clustering. Secondary survival endpoints were likewise analyzed. Mixed-effects regression models with practices as random effects were used to analyze all remaining outcomes—linear models for continuous data, logistic models for binary data and Poisson models for count data. Accordingly, results are either hazard ratios (HR), mean differences (MD), odds ratios (OR) or risk ratios (RR), each reported with a 95% confidence interval (CI). We also present ICCs.
The primary analysis was of the intention-to-treat (ITT) population, including all randomized practices and their patients. We also performed a per-protocol (PP) analysis that included only those practices and patients that received treatment without major protocol violations. Furthermore, in a modified intention-to-treat analysis (mITT), patients switching to the new antithrombotic treatment were censored at the time of switching. We performed sensitivity analyses for survival outcomes using the date of the first HCA training session as an alternative starting point in the intervention group, and adding the median time between randomization and the beginning of intervention group training to the start date in the control group. We also performed subgroup analyses for gender.
We used SAS 9.4 and R, version 3.3.3, for the statistical analyses. A p-value of less than 5% was considered significant.
The final study sample comprised 736 patients (Fig 1). A comparison of baseline characteristics showed that the groups were similar in terms of practice type (42.3% single-handed practices in each group, see S2 Table). However, a lower proportion of intervention than control practices had third-party certification in quality management procedures (46.2% vs. 65.4%), and fewer of them provided structured courses for patients (42.3% vs. 61.5%).
We enrolled patients between July 2, 2012 and Dec 4, 2012. The mean (SD) number of patients recruited per practice was 14.0 (1.6) in the intervention and 14.3 (1.5) in the control group.
A baseline comparison of enrolled patients and those that were eligible but did not participate (‘non-participants’; n = 733) showed that the groups were similar with regard to age (mean (SD) age was 73.5 (9.4) years in the study population vs. 75.0 (10.9) among non-participants), sex (55.0% of participants were male vs. 52.9% of non-participants), and migration background (6.9% of participants vs. 8.2% of non-participants). More participants (n = 85, 12.3%) than non-participants (n = 52, 8.3%) performed OAC self-management previous to study recruitment.
The ITT and mITT analyses included all patients, compared with 313 intervention and 360 control recipients in the PP analysis.
Sociodemographic and clinical characteristics of patients at baseline were similar in intervention and control groups (Table 1).
During the 24-month study period, the primary endpoint occurred in 40 (11.0%) patients in the intervention and 48 (12.9%) patients in the control group (HR 0.83, 95% CI 0.55 to 1.25; P = .37) (Table 2). The median (IQR) time-to-event was 355.5 (170–575.5) days in the intervention and 293.5 (174–525.5) days in the control group (Fig 2).
After 24 months, the ITT analyses of further secondary outcomes showed that intervention patients rated their quality of chronic illness care more highly than control patients, with a mean (SD) PACIC score of 6.7 (2.8) vs. 5.9 (2.9)(Table 3). The change from baseline was significantly better (MD 0.87, 95% CI 0.37 to 1.37; P <0.001) in the intervention group. Mean (SD) patient knowledge of oral anticoagulation after 24 months was 6.4 (2.9) in the intervention group and 5.5 (2.5) in the control group. Again, changes from baseline were significantly better in the intervention group (MD 0.90, 95% CI 0.44 to 1.36; P <0.001). There were no differences in changes from baseline for either patient self-reported adherence to medication, or satisfaction with shared decision-making. The improvement in HCA knowledge of OAC was significantly greater in the intervention group, whereas no difference was found in GP knowledge of OAC (Table 3).
No major differences between the two groups regarding the baseline characteristics were observed for the PP. Therefore the same analyses were applied to this population. For the primary endpoint, the PP (Table 4) and modified ITT (S3 Table), as well as the sensitivity (S4 Table) and the subgroup analysis (S5 Table) resulted in similar findings to the ITT analysis. For the key secondary endpoints, the PP analyses showed that statistically significantly fewer intervention (n = 150, 47.9%) than control patients (n = 202, 56.1%) were hospitalized within 24 months (OR 0.72; 95% CI 0.53 to 0.97; P = 0.031; Table 4). The number of hospitalizations per patient was also lower among intervention patients (RR 0.85, 95% CI 0.72 to 1.00; P = 0.047), whereas the PP analysis yielded similar results to the ITT analysis with regard to the other key secondary outcomes (Table 4).
The estimated mean cost of the intervention, including the training course for the HCAs and telephone calls for GPs, as well as all HCA and GP contacts (assessments and monitoring), was €215 per patient in the first year and €175 per patient in the second year (S6 Table).
Our trial compared a best-practice model for optimized antithrombotic treatment in patients with a long-term indication for oral anticoagulation to routine care. Even though the main outcomes in the intervention and control groups did not differ significantly, the intervention appears to have positively influenced process parameters such as patients’ perceived quality of care and patient knowledge , and HCA knowledge about OAC. For patients obtaining treatment without major protocol deviations (per-protocol analysis), hospital admissions were significantly reduced in the intervention group. Some of our specific intervention elements, such as symptom monitoring and follow up, may have contributed to reduced hospital admissions.
Intervention costs were reasonable and similar to a recent study on HCA-based case management for high-risk patients .
To the best of our knowledge, PICANT is the largest trial to date to investigate the effects of a complex intervention involving primary care-based case management, self-management of OAC, and additional patient education. The intervention is feasible in a ‘real world’ setting and does not require additional personnel but rather relies on HCAs as a valuable resource to provide team-based care to chronically ill patients. The professionalization of non-physician health professionals, such as nurses and HCAs, has been identified as a cost-efficient way to improve healthcare . Team-based care approaches and delegation of “transactional tasks” (e.g., documentation of care) to other clinical professionals and staff who have less training means physicians have more time for “personalized” aspects of patient care (e.g., customizing care for individual patients) .
Healthcare assistants are a promising resource for delivery of care management to high-risk patients in small primary care practices. Our study aimed to limit the major risks associated with poor anticoagulation control (e.g., long intervals between measurements). Healthcare assistants (supervised by GPs) were able to assume a new role in chronic care management of patients with oral anticoagulation.
In our paper, we restricted our intervention to optimizing antithrombotic management. However, recent studies that examined multidisciplinary integrated care approaches in a patient group with atrial fibrillation, and took multiple co-morbidities into account, showed clear superiority in terms of reducing cardiovascular hospitalization and mortality . Furthermore, an ongoing cluster-RCT is currently seeking to demonstrate the feasibility of integrated atrial fibrillation care in 1000 elderly primary care patients from around 18 to 30 general practices in terms of its potential effectiveness on patient relevant outcomes . Although interesting, this approach goes beyond the scope of our study.
We acknowledge a potential selection bias since a slightly higher percentage of participants than non-participants performed self-management. Participants may therefore have been more highly motivated than the eligible population from which we drew the sample. In addition, both groups already showed fair to good OAC quality at baseline, and in both, the number of patients with INR values within their therapeutic ranges increased further. More of the control practices had third-party certification in quality management than intervention practices. It is therefore probable that oral anticoagulation management in our control group was particularly good, which would have made it more difficult to demonstrate statistically significant differences between the groups. We designed our study in accordance with the recommendations of the extended CONSORT statement , so even though we had to deal with certain external constraints, such as limited available funding and the limited duration of the funding period, we consider our results reliable. The intervention intensity was limited so that the additional tasks associated with PICANT would fit into healthcare assistants' daily workflow. However, we over-estimated the anticipated effect of our intervention, and the calculated patient numbers of 317 in each group turned out to be too low. This is a recurrent problem in cluster RCTs, as recently described by Siebenhofer at al. in a methodological systematic review . As our actual ICC was lower than the assumed value, the underestimation of intra-cluster similarities was not a limitation in our study.
The study took place against a background of increasing prescriptions of DOACs. Since receiving approval in 2011, prescriptions of DOACs have risen strongly, with 38 million (m) defined daily doses (DDDs) prescribed in Germany in 2012 (vs. 389m DDDs of VKAs), and 253m DDDs prescribed in 2015 (vs. 346m DDDs of VKAs) .
The percentage of VKA patients that switched to DOACs was 7.6% in our trial, lower than the 15% observed in a study by Bleckwenn et al., which was also conducted in German primary care practices . One should bear in mind that the study took place in 2012, when, for example, the enthusiasm of cardiologists for DOACs was not fully shared by GPs. One reason for this was an unfamiliar inability to monitor patients to ensure adherence, while another was cost, since GPs are held responsible for the lion’s share of overall drug costs in Germany. Education and monitoring remain necessary when VKAs are replaced with DOACs. Amara et al. have recently shown that DOAC patients have serious knowledge gaps with respect to their medication , with only 21% aware that regular monitoring of renal function is recommended.
In line with the recommendations of evidence-based medicine [30–32], various sources of information have been made available to allow this study to be reproduced. These include the study protocol , publications describing the development of the CoMol monitoring list , and the screening process , as well as information on patient education for self-management .
The complex intervention in our study was designed to be provided in addition to routine care of orally anti-coagulated patients with both vitamin K antagonists and newer direct oral anticoagulants. As the anticoagulation of most patients was already of high quality, the assumption that the intervention would improve the long-term outcomes of anticoagulated patients could not be proven. Nevertheless, it should be noted that quality is and remains optimal when primary care professionals have received adequate training. The intervention even led to improved process parameters such as patient knowledge and perceived quality of care, both of which are known to positively influence clinical outcomes and reduce hospital admissions. This was indeed the case in the per-protocol analysis, even though the combined primary outcome of all thromboembolic events requiring hospitalization and all major bleeding complications did not differ significantly between the groups. In addition, supportive team-based care can be provided to chronically ill patients using the publicly available tools we developed (teaching materials, monitoring lists, fact sheets and patient information).
S1 Table. Description of the intervention and control elements.
S2 Table. Characteristics of practices, GPs and healthcare assistants.
S3 Table. Modified intention-to-treat analysis of the primary and key secondary outcomes after 24 months.
S4 Table. Sensitivity analysis for the primary outcome after 24 months.
S5 Table. Subgroup analysis by gender for primary and key secondary outcomes after 24 months.
We would like to thank all practices and patients that participated in the pilot phase and the main trial.
We are most grateful to Antje Erler, MD, Ina Roehl, MD, Marion Torge, MD, and Julia Hirschfeld, MSc, for their support during data collection and analysis. In addition, Thomas Gary, MD, and Matthias Sunnus cross-checked primary outcomes. We would like to thank Corina Guethlin, PhD, and Martin Beyer for their support during protocol development and Prof. V. Hach-Wunderle, MD, and T. Gary, MD, as experts in hemostasis for critically reviewing the study materials. We would like to thank the members of the scientific board (Professor Jack E. Ansell, MD, Professor Herbert Watzke, MD, Professor Meinhard Kieser, Dr.sc.hum, Justine Rochon, MSc) and practice advisory board (Armin Wunder, MD, Joachim Fessler, MD, Wolfgang Blank, MD). We gratefully acknowledge the support of Christina Conrad, and Julia Ruland in the preparation of study materials.
We would like to express our gratitude to Prof. Chenot, MD, who provided us with the questionnaire developed by Duc Hua, MD, and to Malcolm Man-Son-Hing for permitting us to use the Man-Son-Hing questionnaire. We would also like to thank the Department of General Practice and Health Services Research, University Hospital Heidelberg, Heidelberg, Germany for providing us with the PACIC short form.
We are most grateful to Birgit Schorsch for performing the financial calculations and to Phillip Elliott for editing the final manuscript.
Prior presentation: A. Siebenhofer-Kroitzsch, J. J. Petersen, L. R. Ulrich, K. Mergenthal, B. Kemperdick, S. Rauck, S. Schulz-Rothe, F. M. Gerlach, G. Pregartner, A. Berghold. Wirksamkeit eines hausarztpraxisbasierten Case Managements zur Optimierung der oralen Antikoagulation–Ergebnisse der Cluster randomisierten PICANT-Studie [Effectiveness of a primary care based case management intervention to optimize oral anticoagulation- results of the cluster randomized controlled PICANT trial]. 51th congress of the German College of General Practitioners and Family Physicians, 21.-23.9.2017, Düsseldorf, Germany.
- 1. Oake N, Fergusson DA, Forster AJ, van Walraven C. Frequency of adverse events in patients with poor anticoagulation. A meta-analysis. Can Med Assoc J. 2007;176(11): 1589–1594.
- 2. Heneghan CJ, Garcia-Alamino JM, Spencer EA, Ward AM, Perera R, Bankhead C, et al. Self-monitoring and self-management of oral anticoagulation. Cochrane Database Syst Rev. 2016;7: CD003839. pmid:27378324
- 3. Siebenhofer A, Jeitler K, Horvath K, Habacher W, Schmidt L, Semlitsch T. Self-management of oral anticoagulation. Dtsch Arztebl Int. 2014;111(6): 83. pmid:24622604
- 4. Arzneimittelkommission der deutschen Ärzteschaft [Drug Commission of the German Medical Association] Orale Antikoagulation bei nicht valvulärem Vorhofflimmern. Empfehlungen zum Einsatz der direkten oralen Antikoagulanzien Dabigatran (Pradaxa®), Apixaban (Eliquis®), Edoxaban (Lixiana®) und Rivaroxaban (Xarelto®) [Recommendations for the use of direct oral anticoagulants dabigatran (Pradaxa®), apixaban (Eliquis®), edoxaban (Lixiana®) and rivaroxaban (Xarelto®)]. 2nd ed. Available from: http://www.akdae.de/Arzneimitteltherapie/LF/PDF/OAKVHF.pdf. Accessed 11 December 2018.
- 5. Hein L. Antithrombotika und Antihämorrhagika. In: Schwabe U, Paffrath D, editors. Arzneiverordnungs-Report 2016. Berlin, Heidelberg: Springer; 2016. pp. 351–368.
- 6. Bolen SD, Stange KC. Investing in relationships and teams to support managing complexity. J Gen Intern Med. 2017;32(3): 241–242. pmid:28004233
- 7. Hudon C, Chouinard M-C, Bayliss E, Nothelle S, Senn N, Shadmi E. Challenges and next steps for primary care research. Ann Fam Med. 2018;16(1): 85–86.
- 8. Gensichen J, von Korff M, Peitz M, Muth C, Beyer M, Güthlin C, et al. Case management for depression by health care assistants in small primary care practices—a cluster randomized trial. Ann Intern Med. 2009;151: 369–378. pmid:19755362
- 9. Gensichen J, Jaeger C, Peitz M, Torge M, Güthlin C, Mergenthal K, et al. Health care assistants in primary care depression management: role perception, burdening factors, and disease conception. Ann Fam Med. 2009;7(6): 513–519. pmid:19901310
- 10. Peters-Klimm F, Campbell S, Hermann K, Kunz CU, Muller-Tasch T, Szecsenyi J. Case management for patients with chronic systolic heart failure in primary care. The HICMan exploratory randomised controlled trial. Trials. 2010;11: 56. pmid:20478035
- 11. Siebenhofer A, Ulrich LR, Mergenthal K, Roehl I, Rauck S, Berghold A, et al. Primary care management for optimized antithrombotic treatment [PICANT]: study protocol for a cluster-randomized controlled trial. Implement Sci. 2012;7: 79. pmid:22929015
- 12. Ulrich L-R, Mergenthal K, Petersen JJ, Roehl I, Rauck S, et al. Anticoagulant treatment in German family practices—screening results from a cluster randomized controlled trial. BMC Fam Pract. 2014;15: 170. pmid:25344288
- 13. Ulrich L-R, Petersen JJ, Mergenthal K, Roehl I, Rauck S, Kemperdick B, et al. Eine Monitoring-Liste für ein hausärztliches Case Management bei oraler Antikoagulation [A monitoring list for oral anticoagulation case management in primary care]. Z Allg Med. 2013;89: 165–171.
- 14. Rosendaal FR, Cannegieter SC, van der Meer FJ, Briet E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost. 1993;69(3): 236–239. pmid:8470047
- 15. EuroQol group. EuroQol-a new facility for the measurement of health-related quality of life. Health Policy. 1990;16(3): 199–208. pmid:10109801
- 16. Goetz K, Freund T, Gensichen J, Miksch A, Szecsenyi J, Steinhaeuser J. Adaptation and psychometric properties of the PACIC short form. Am J Manag Care. 2012;18(2): 60.
- 17. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care. 1986;24(1): 67–74. pmid:3945130
- 18. Hua TD, Vormfelde SV, Abed MA, Schneider-Rudt H, Sobotta P, Friede T, et al. Practice nursed-based, individual and video-assisted patient education in oral anticoagulation-protocol of a cluster-randomized controlled trial. BMC Fam Pract. 2011;12(1): 17.
- 19. Man-Son-Hing M, Laupacis A, O’Connor AM, Biggs J, Drake E, Yetisir E, et al. A patient decision aid regarding antithrombotic therapy for stroke prevention in atrial fibrillation: a randomized controlled trial. JAMA. 1999;282(8): 737–743. pmid:10463708
- 20. Maikranz V, Siebenhofer A, Ulrich L-R, Mergenthal K, Schulz-Rothe S, Kemperdick B, et al. Does a complex intervention increase patient knowledge about oral anticoagulation?—a cluster-randomised controlled trial. BMC Fam Pract. 2017;18(1): 15. pmid:28166725
- 21. Freund T, Peters-Klimm F, Boyd CM, Mahler C, Gensichen J, Erler A, et al. Medical assistant-based care management for high-risk patients in small primary care practices: a cluster randomized clinical trial. Ann Intern Med. 2016;164(5): 323–330. pmid:26833209
- 22. Busse R, Blümel M, Knieps F, Bärnighausen T. Statutory health insurance in Germany. A health system shaped by 135 years of solidarity, self-governance, and competition. Lancet. 2017;390(10097): 882–897. pmid:28684025
- 23. Reuben DB, Sinsky CA. From transactional tasks to personalized care—a new vision of physicians’ roles. Ann Fam Med. 2018;16(2): 168–169. pmid:29531111
- 24. Gallagher C, Elliott AD, Wong CX, Rangnekar G, Middeldorp ME, Mahajan R, et al. Integrated care in atrial fibrillation: a systematic review and meta-analysis. Heart. 2017;103(24): 1947–1953. pmid:28490616
- 25. van den Dries CJ, Oudega R, Elvan A, Rutten FH, van de Leur SJCM, Bilo HJG, et al. Integrated management of atrial fibrillation including tailoring of anticoagulation in primary care: study design of the ALL-IN cluster randomised trial. BMJ Open. 2017;7(9): e015510. pmid:28928175
- 26. Campbell MK, Piaggio G, Elbourne DR, Altman DG. Consort 2010 statement. Extension to cluster randomised trials. BMJ. 2012;345: e5661. pmid:22951546
- 27. Siebenhofer A, Paulitsch MA, Pregartner G, Berghold A, Jeitler K, Muth C, et al. Cluster-randomised controlled trials evaluating complex interventions in general practices are mostly ineffective. A systematic review. J Clin Epidemiol. 2017;94: 85–96. pmid:29111470
- 28. Bleckwenn M, Dinkel K, Weckbecker K, Mücke M. Einsatz der neuen oralen Antikoagulanzien in Hausarztpraxen [Use of new oral anticoagulants in primary care practices]. Z Allg Med. 2016;92(1): 28–32.
- 29. Amara W, Larsen TB, Sciaraffia E, Hernández Madrid A, Chen J, Estner H, et al. Patients’ attitude and knowledge about oral anticoagulation therapy. Results of a self-assessment survey in patients with atrial fibrillation conducted by the European Heart Rhythm Association. Europace. 2016;18(1): 151–155. pmid:26462697
- 30. Glasziou P, Chalmers I, Altman DG, Bastian H, Boutron I, Brice A, et al. Taking healthcare interventions from trial to practice. BMJ. 2010;341: c3852. pmid:20709714
- 31. Hoffmann T, Straus S. Sharing knowledge for health care. JAMA Intern Med. 2017;177(9): 1243–1244. pmid:28738133
- 32. Lehman R. Sharing as the future of medicine. JAMA Intern Med. 2017;177(9): 1237–1238. pmid:28672293
- 33. Hirschfeld J, Mergenthal K, Petersen JJ, Rauck S, Roehl I, Ulrich LR, et al. Patientenschulungen zum Gerinnungsselbstmanagement—Angebotssituation in Hessen [Patient education for self-management of oral anticoagulation—situation in Hesse]. Gesundheitswesen. 2014;76(10): 628–632. pmid:24165916