Outcome-Relevant Effects of Shared Decision Making
A systematic review
; ; ; ; ;
Background: Shared decision making (SDM) is considered a gold standard for the cooperation of doctor and patient. SDM improves patients’ overall satisfaction and their confidence in decisions that have been taken. The extent to which it might also positively affect patient-relevant, disease-related endpoints is a matter of debate.
Methods: We systematically searched the PubMed database and the Cochrane Library for publications on controlled intervention studies of SDM. The quality of the intervention and the risk of bias in each publication were assessed on the basis of pre-defined inclusion and exclusion criteria. The effects of SDM on patient-relevant, disease-related endpoints were compared, and effect sizes were calculated.
Results: We identified 22 trials that differed widely regarding the patient populations studied, the types of intervention performed, and the mode of implementation of SDM. In ten articles, 57% of the endpoints that were considered relevant were significantly improved by the SDM intervention compared to the control group. The median effect size (Cohen’s d) was 0.53 (0.14–1.49). In 12 trials, outcomes did not differ between the two groups. In all 22 studies identified, 39% of the relevant outcomes were significantly improved compared with the control groups.
Conclusion: The trials performed to date to addressing the effect of SDM on patient-relevant, disease-related endpoints are insufficient in both quantity and quality. Although just under half of the trials reviewed here indicated a positive effect, no final conclusion can be drawn. A consensus-based standardization of both SDM-promoting measures and appropriate clinical studies are needed.
The relationship between doctor and patient is continually changing under the influence of social and societal changes, among others (1). During the 20th century, improved access to information and patients’ desire for more autonomy contributed to a change in the relationship between doctor and patient from a mainly disease- and/or physician-orientated one towards dialogue and partnership. The bilateral exchange of information is now emphasized, and the patient plays a more active role in the treatment process. In Germany, this development is also reflected at the health political and legal levels—for example, in the so-called Patients’ Rights Act of 2013 (§ 630 of Germany’s civil code) and in the Medical Association's Professional Code for Physicians in Germany. The latter states that patients’ consent to treatment should be obtained after patients were first given information on the treatment’s nature, importance, and consequences, including alternative treatments and associated risks. Increasingly, patients’ participation in decisions is also finding its way into treatment guidelines, for example in terms of recommendations on shared decision making (SDM) (2).
Active participation of patients in their treatment process can enhance clinical and psychosocial patient-related outcomes. Among others, a willingness and readiness to undertake initial treatment steps, trust in the medical decision, perception of risk, and realistic expectations can be fostered (3). Further relevant effects include increased satisfaction of patients and better adherence to treatment (4).
SDM is a crucial method in implementing patients’ education and participation. On the background of a partnership between doctor and patient, this is an interaction between both parties with the objective to use communication tools to reach a decision on an appropriate approach—for example, relating to treatment. Imparting current scientific evidence is a crucial component in this setting. The aim is to reach the best possible decision according to the clinical requirements and the patient’s preferences (3, 5). It needs to be borne in mind that the approach of patient participation is continually evolving. Depending on the respective perceptions of shared decision making, but also on the context, the implementation may differ (6–8). Irrespective of this, SDM is a gold standard for the cooperation between doctor and patient (1). Studies have shown, however, that SDM has not yet comprehensively been adopted in clinical practice. Decisions about treatment are often still unilaterally made by doctors (9), although patients have a fundamental need to be involved in such decisions (10, 11).
Evidence for positive effects of increased patient participation in general has been delivered primarily for not directly disease-related outcomes (3, 4). The current state of research regarding the clinical relevance of SDM in particular is, however, not clear. A systematic review similar to our study investigated the effect on patient-relevant, disease-related outcomes, but did not focus on SDM as a form of patient participation and limited the number of studies included to merely seven, on the basis of very strict inclusion and exclusion criteria (12). A detailed study from the Cochrane Collaboration focused primarily on the implementation of SDM as the study endpoint, but did not consider many of the outcomes and studies addressed by our own review (13). In our systematic review, we included controlled studies that explicitly investigated the effect of patient participation in the form of SDM on patient-relevant, disease-related outcomes.
We conducted a literature search with a cut-off date of 30 June 2014 in PubMed and the Cochrane Library, using the search terms “patient participation” and “decision making” and limiting the search to “clinical trial” as the publication type. We did not search for “shared decision making” as a medical subject heading (MeSH) term, since in our experience, some interventions are consistent with shared decision making but are not explicitly named as such.
Study selection and data extraction
KH and JM each applied the inclusion and exclusion criteria listed below to the search results. The inclusion criteria were:
- A controlled study design
- The explicit use of the terms “shared decision making” (SDM) or “participatory decision making” in the description of the intervention or information indicating that patients were involved in making a decision about treatment according to the first criterion of Charles et al. (Box) (14)
- Fulfillment of at least one additional SDM criterion according to Charles et al. (14)
- Assessment of patient-relevant, disease-related outcomes.
The Charles criteria were applied in accordance with Joosten et al., among others (4). As patient-relevant, disease-related outcomes we considered parameters that were directly related to disease-dependent wellbeing or to a patient’s prognosis (Tables 1 and 2, eTables 1 and 2). We did not consider outcomes that related explicitly to the communication between doctor and patient—for example, the degree of informedness or satisfaction with the information—the doctor–patient relationship in general—such as the satisfaction of a patient with his/her doctor—or patient participation itself—for example, the number of questions asked by patients.
Our exclusion criteria were:
- Use of a decision aid as the only difference between control group and intervention group.
- Intended implementation of SDM in the control group too.
- Lacking statistical data supporting the reported results.
After we had finished identifying suitable studies as described above, KH extracted the data. Only patient-relevant, disease-related outcomes were considered, as described in our inclusion criteria (15, e1).
Quality of publications and interventions
The methodological quality of the studies and the implementation of SDM interventions were assessed according to Joosten and colleagues (4). KH and JM individually evaluated the methodological quality of the included publications regarding the risk of bias. To this end, we adapted the Cochrane Back Review Group’s recommendations for systematic reviews (16) (eBox). Furthermore, we assessed the extent of SDM implementation on the basis of the number of identifiably applied key aspects according to Charles et al. as indicators of intervention quality (14) (Box). Since the literature does not provide a single definition of SDM, we adopted for our review the four Charles criteria—in accordance with Joosten et al.—which, in our view, can be considered a common denominator in interpretations of SDM (4, 8, 14).
Where the publications provided sufficient data, we calculated the corresponding effect sizes as Cohen’s d, a measure of the relevance of statistically significant effects on patient-relevant, disease-related outcomes (eTable 3) (e2–e5).
The search in PubMed yielded 422 hits and that in the Cochrane Library 222 hits. After removal of duplicates, we verified the remaining 430 publications on the basis of their title and abstract with regard to the research question and the defined inclusion and exclusion criteria. Study protocols and designs without reporting concrete results were excluded. Of the 59 remaining articles, 37 were excluded after a review of their full text; ultimately, 22 controlled intervention studies were included in our review article (eFigure, Tables 1 and 2, eTables 1 and 2). We determined the risk for bias and found that the relevant information was often not fully reported. On average, 69% of the criteria were met. The proportion of lacking or unclear data relating to the criteria under study in the publications was 12%. It has to be kept in mind that we did not assess the quality of particular studies but that of the respective publications.
In 10 of the 22 studies (45%), the intervention resulted in a significant improvement in at least one outcome compared with the control group (17–26) (Table 1, eTable 1). In 12 of the studies (55%) this was not the case (27–38) (Table 2, eTable 2). In the studies that showed benefits from the intervention, 57% of the relevant outcomes were significantly improved compared with the control group; this was the case for 39% when all 22 studies were considered. In studies that showed a benefit from SDM, the median effect size was 0.53 (range 0.14–1.49). Values were evenly distributed among ranges indicating small, moderate, and large effect sizes (eTable 1). We deemed individual outcomes in three studies not meaningful as significant differences existed between groups already at baseline. However, in one these three studies the positive effect on several other outcomes was greater than in the control group (23). In neither of the remaining two studies did any other relevant outcome show greater improvement in the intervention group compared to the control group (37, 38).
The studies were quite heterogeneous regarding disorders, outcomes, and interventions (Tables 1 and 2, eTables 1 and 2). The studies showing favorable results for SDM, as well as those where the effects in the control and intervention groups did not differ, included, among others, patients with cardiovascular disorders, respiratory disorders, mental disorders, or tumor disease. Examples of patient-relevant, disease-related outcomes addressed in the studies are disease-related surrogate parameters—for example, blood pressure—or psychological constructs (data obtained using patient questionnaires), as in depression or schizophrenia. The studied interventions ranged from brief conversations with patients that immediately preceded doctor–patient contact to several hours of staff training delivered over several weeks (eTables 1 and 2). In the studies showing benefit from SDM, patients were more often directly targeted by the intervention.
The extent to which SDM was implemented differed widely (Box). Twelve studies showed only one additional characteristic in addition to the participation of doctor and patient (17, 22, 23, 25, 27, 29, 31, 33, 35–38), five studies showed two further characteristics (19, 20, 24, 32, 34). All four SDM criteria were realized in five studies (23%) (18, 21, 26, 28, 30). No association was seen between outcome effects and the degree of SDM implementation (data not shown). The degree of SDM implementation might have been underestimated, as not all criteria were verifiable on the basis of the publications.
In 12 of the 22 studies, the outcomes in patients who had received an SDM intervention did not show any greater improvement than the outcomes in patients of the respective control groups. Ten studies showed an advantage for SDM for patient-relevant, disease-related outcomes compared with the control groups. The observed effect sizes were high to very high in some cases.
The heterogeneity of diseases in the included studies does not allow for identification of disorders with a higher probability of SDM being beneficial. Similarly, we did not observe any association between the extent of SDM implementation and the outcome effect (data not shown). In the 10 studies that showed favorable results for SDM it is of note that most of the interventions (70%) directly aimed at the patient. In only three studies the intervention focused on doctors or nursing staff. Only 4 (33%) of the 12 studies in which the intervention showed no additional outcome benefit patients had been the primary target of the intervention. This finding suggests that an SDM intervention might improve patient-relevant, disease-related outcomes mainly if the intervention directly aims at the patient. Regarding SDM implementation as the outcome, Légaré et al. found that measures that involve doctors as well as patients are most effective (13).
Two studies in which treatment adherence was improved by the intervention to a greater extent than in the control groups showed an advantage for all outcomes that was deemed relevant for the present review (Table 1, eTable 1b). In 8 studies in which adherence was not improved as a result of the intervention, no significant outcome differences were seen (Table 2, eTable 2b). This finding suggests an association between adherence and improvements in patient-relevant, disease-related outcomes, similar to that we have recently found for antihypertensive treatment (5).
The fact that about half of the analyzed studies did not show any positive effects as a result of an SDM intervention may have different reasons. As the methodological quality of many studies was limited, effects may have remained hidden. Two other reviews that differed from our approach in terms of their search strategies also stated that conclusive results regarding the effects of improved patient participation in general (12) and SDM in particular (8) on patient-relevant, disease-related outcomes are lacking.
The quality of the implementation of SDM interventions was low to moderate in many studies. Twelve of the 22 studies included in our systematic review did not meet more than two Charles criteria. We did not find an association between the degree of SDM implementation and effects on the investigated outcomes (data not shown). However, considering the often insufficient information given in the publications, we cannot exclude that the aspects of SDM were not implemented intensely enough to reveal a corresponding effect. The problem of fitting the context—that is, the difficulty in finding the degree of SDM that meets the complex, context-specific needs of the individual patient—may also play a role (39).
It should also be considered that communication with and inclusion of patients in the control and intervention groups may not have differed sufficiently from one another. Thus, in studies where patients in the control group received so-called standard care, certain aspects of SDM may have been implemented more or less unintentionally. Of note, the literature does not provide an accepted consensus on the definition of SDM. Makoul and colleagues identified 31 different conceptual definitions (40). We and Josten et al. used SDM criteria originally defined by Charles et al. (4, 14), representing only one SDM concept that has been modified by the same group later (6).
As suggested above, it is likely that perceptions of SDM and thus the tested interventions may have differed in the studies included in our review. Furthermore, the inclusion criteria we applied probably did not cover every concept of SDM. This may mean that potentially relevant studies might have been missed.
In sum, we wish to remind readers that we assumed a study to show benefits for SDM if at least one outcome in the intervention group was improved to a greater extent than in the control group. The number of outcomes addressed in a study increases the risk of a cumulative α error, and thus the proportion of SDM interventions considered outcome-relevant may have been subject to overestimation.
We conclude that to date, research on the effect of SDM on patient-relevant, disease-related outcomes is insufficient regarding the number of studies and the quality of relevant publications. Although nearly half of the identified studies indicated outcome-relevant effectiveness of SDM, the results so far do not allow for any conclusive assessment of the outcome relevance of SDM. A consensus-based standardization of SDM promoting measures and appropriate clinical studies are needed and desirable.
Conflict of interest statement
Professor Albus has received consultancy fees as an advisory board member from UCB Pharma and lecture honoraria from Berlin-Chemie and Actelion.
The other authors declare that no conflict of interest exists.
Manuscript received on 4 February 2015, revised version accepted on
14 April 2015.
Translated from the original German by Birte Twisselmann, PhD.
PD Dr. Jan Matthes
Institut für Pharmakologie, Universität zu Köln
Gleueler Straße 24, 50931 Köln
For eReferences please refer to:
eBox, eFigure, eTables:
PD Dr. med. Jan Matthes
Department of Psychosomatics and Psychotherapy, University Hospital of Cologne: Dr. phil. Koerfer,
Prof. Dr. med. Albus
Institute of Medical Statistics, Informatics and Epidemiology, University of Cologne: Dipl.-Stat. Kuhr
|1.||Laidsaar-Powell RC, Bu S, McCaffery KJ: Partnering with and involving patients. In: Martin LR, DiMatteo RM (eds.): The oxford handbook of health communication, behavior change, and treatment adherence. New York: Oxford University Press 2014; 84–108.|
|2.||Perk J, De Backer G, Gohlke H, et al.: European guidelines on cardiovascular disease prevention in clinical practice (version 2012). The Fifth Joint Task Force of the European Society of Cardiology and Other Societies on Cardiovascular Disease Prevention in Clinical Practice (constituted by representatives of nine societies and by invited experts). Eur Heart J 2012; 33: 1635–701 CrossRef MEDLINE|
|3.||Loh A SD, Kriston L, Härter M: Shared decision making in medicine. Dtsch Arztebl 2007; 104: A 1483–8 VOLLTEXT|
|4.||Joosten EA, DeFuentes-Merillas L, de Weert GH, Sensky T, van der Staak CP, de Jong CA: Systematic review of the effects of shared decision-making on patient satisfaction, treatment adherence and health status. Psychother Psychosom 2008; 77: 219–26 CrossRef MEDLINE|
|5.||Matthes J, Albus C: Improving adherence with medication: a selective literature review based on the example of hypertension treatment. Dtsch Arztebl Int 2014; 111: 41–7 VOLLTEXT|
|6.||Charles C, Gafni A, Whelan T: Decision-making in the physician-patient encounter: revisiting the shared treatment decision-making model. Soc Sci Med 1999; 49: 651–61 CrossRef|
|7.||Elwyn G, Frosch D, Thomson R, et al.: Shared decision making: |
a model for clinical practice. J Gen Intern Med 2012; 27: 1361–7 CrossRef MEDLINE PubMed Central
|8.||Shay LA, Lafata JE. Where is the evidence? A systematic review of shared decision making and patient outcomes. Med Decis Making 2015; 35: 114–31 CrossRef MEDLINE|
|9.||Karnieli-Miller O, Eisikovits Z: Physician as partner or salesman? Shared decision-making in real-time encounters. Soc Sci Med 2009; 69: 1–8 CrossRef MEDLINE|
|10.||Cullati S, Courvoisier DS, Charvet-Bérard AI, Perneger TV: Desire for autonomy in health care decisions: a general population survey. Patient Educ Couns 2011; 83: 134–8 CrossRef MEDLINE|
|11.||Guadagnoli E, Ward P: Patient participation in decision-making. Soc Sci Med 1998; 47: 329–39 CrossRef|
|12.||Sanders AR, van Weeghel I, Vogelaar M, et al: Effects of improved patient participation in primary care on health-related outcomes: a systematic review. Fam Pract 2013; 30: 365–78 CrossRef MEDLINE PubMed Central|
|13.||Légaré F, Stacey D, Turcotte S, et al.: Interventions for improving the adoption of shared decision making by healthcare professionals. Cochrane Database Syst Rev 2014; 9: CD006732 CrossRef|
|14.||Charles C, Gafni A, Whelan T: Shared decision-making in the medical encounter: what does it mean? (or it takes at least two to tango). Soc Sci Med 1997; 44: 681–92 CrossRef|
|15.||Dwamena F, Holmes-Rovner M, Gaulden CM, et al.: Interventions for providers to promote a patient-centred approach in clinical consultations. Cochrane Database Syst Rev 2012; 12: CD003267 CrossRef|
|16.||Furlan AD, Pennick V, Bombardier C, van Tulder M: 2009 updated method guidelines for systematic reviews in the Cochrane Back Review Group. Spine 2009; 34: 1929–41 CrossRef MEDLINE|
|17.||Deadman JM, Leinster SJ, Owens RG, Dewey ME, Slade PD: Taking responsibility for cancer treatment. Soc Sci Med 2001; 53: 669–77 CrossRef|
|18.||Deinzer A, Babel H, Veelken R, Kohnen R, Schmieder RE: Shared decision-making with hypertensive patients. Results of an implementation in Germany. Dtsch Med Wochenschr 2006; 131: 2592–6 CrossRef MEDLINE|
|19.||Greenfield S, Kaplan S, Ware JE, Jr: Expanding patient involvement in care. Effects on patient outcomes. Ann Intern Med 1985; 102: 520–8 CrossRef MEDLINE|
|20.||Greenfield S, Kaplan SH, Ware JE Jr, Yano EM, Frank HJ: Patients' participation in medical care: effects on blood sugar control and quality of life in diabetes. J Gen Intern Med 1988; 3: 448–57 CrossRef MEDLINE|
|21.||Joosten EA, de Jong CA, de Weert-van Oene GH, Sensky T, van der Staak CP: Shared decision-making reduces drug use and psychiatric severity in substance-dependent patients. Psychother Psychosom 2009; 78: 245–53 CrossRef MEDLINE|
|22.||Malm U, Ivarsson B, Allebeck P, Falloon IR: Integrated care in schizophrenia: a 2-year randomized controlled study of two community-based treatment programs. Acta Psychiatr Scand 2003; 107: 415–23 CrossRef|
|23.||Redfern J, Briffa T, Ellis E, Freedman SB: Choice of secondary prevention improves risk factors after acute coronary syndrome: 1-year follow-up of the CHOICE (Choice of Health Options In prevention of Cardiovascular Events) randomised controlled trial. Heart 2009; 95: 468–75 CrossRef MEDLINE|
|24.||Rost KM, Flavin KS, Cole K, McGill JB: Change in metabolic control and functional status after hospitalization. Impact of patient activation intervention in diabetic patients. Diabetes Care 1991; 14: 881–9 CrossRef MEDLINE|
|25.||van Roosmalen MS, Stalmeier PF, Verhoef LC, et al.: Randomized trial of a shared decision-making intervention consisting of trade-offs and individualized treatment information for BRCA1/2 mutation carriers. J Clin Oncol 2004; 22(16): 3293–301 CrossRef MEDLINE|
|26.||Wilson SR, Strub P, Buist AS, et al.: Shared treatment decision making improves adherence and outcomes in poorly controlled asthma. Am J Respir Crit Care Med 2010; 181: 566–77 CrossRef MEDLINE PubMed Central|
|27.||Holzel LP, Vollmer M, Kriston L, Siegel A, Harter M: [Patient participation in medical decision making within an integrated health care system in Germany: results of a controlled cohort study]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz 2012; 55: 1524–33 CrossRef MEDLINE|
|28.||Krones T, Keller H, Sonnichsen A, et al.: Absolute cardiovascular disease risk and shared decision making in primary care: a randomized controlled trial. Ann Fam Med 2008; 6: 218–27 CrossRef MEDLINE PubMed Central|
|29.||Edwards A, Elwyn G: Involving patients in decision making and communicating risk: a longitudinal evaluation of doctors' attitudes and confidence during a randomized trial. J Eval Clin Pract 2004; 10: 431–7 CrossRef MEDLINE|
|30.||Hamann J, Cohen R, Leucht S, Busch R, Kissling W: Shared decision making and long-term outcome in schizophrenia treatment. J Clin Psychiatry 2007; 68: 992–7 CrossRef|
|31.||Hamann J, Mendel R, Meier A, et al.: “How to speak to your psychiatrist”: shared decision-making training for inpatients with schizophrenia. Psychiatr Serv 2011; 62: 1218–21 CrossRef CrossRef MEDLINE|
|32.||Legare F, Labrecque M, Cauchon M, Castel J, Turcotte S, Grimshaw J: Training family physicians in shared decision-making to reduce the overuse of antibiotics in acute respiratory infections: a cluster randomized trial. Can Med Assoc J 2012; 184: E726–34 CrossRef MEDLINE PubMed Central|
|33.||Loh A, Simon D, Wills CE, Kriston L, Niebling W, Harter M: The effects of a shared decision-making intervention in primary care of depression: a cluster-randomized controlled trial. Patient Educ Couns 2007; 67: 324–32 CrossRef MEDLINE|
|34.||Thomson RG, Eccles MP, Steen IN, et al.: A patient decision aid to support shared decision-making on anti-thrombotic treatment of patients with atrial fibrillation: randomised controlled trial. Qual Saf Health Care 2007; 16: 216–23 CrossRef MEDLINE PubMed Central|
|35.||Tinsel I, Buchholz A, Vach W, et al.: Shared decision-making in antihypertensive therapy: a cluster randomised controlled trial. BMC Fam Pract 2013; 14: 135 CrossRef MEDLINE PubMed Central|
|36.||Bieber C, Muller KG, Blumenstiel K, et al.: Long-term effects of a shared decision-making intervention on physician-patient interaction and outcome in fibromyalgia. A qualitative and quantitative 1 year follow-up of a randomized controlled trial. Patient Educ Couns 2006; 63: 357–66 CrossRef MEDLINE|
|37.||Davison BJ, Degner LF: Empowerment of men newly diagnosed with prostate cancer. Cancer Nurs 1997; 20: 187–96 CrossRef|
|38.||Koelewijn-van Loon MS, van der Weijden T, van Steenkiste B, et al.: Involving patients in cardiovascular risk management with nurse-led clinics: a cluster randomized controlled trial. Can Med Assoc J 2009; 181: E267–E74 CrossRef MEDLINE PubMed Central|
|39.||Murray E, Charles C, Gafni A: Shared decision-making in primary care: tailoring the Charles et al. model to fit the context of general practice. Patient Educ Couns 2006; 62: 205–11 CrossRef MEDLINE|
|40.||Makoul G, Clayman ML: An integrative model of shared decision making in medical encounters. Patient Educ Couns 2006; 60: 301–12 CrossRef MEDLINE|
|e1.||Ebell MH, Siwek J, Weiss BD, et al.: Simplifying the language of evidence to improve patient care. J Fam Pract 2004; 53: 111–20.|
|e2.||Rosenthal R: Parametric measures of effect size. In: Cooper H, Hedges L (eds.): The handbook of research synthesis. New York: Russel Sage Foundation 1994; 231–44.|
|e3.||Higgins J, Green S: Selecting studies and collecting data. In: Green S, Higgins J (eds.): Cochrane handbook for systematic reviews of interventions Version 5.1.0 (updated March 2011): The Cochrane Collaboration 2012.|
|e4.||Rosenthal R, DiMatteo MR: Meta-analysis: recent developments in quantitative methods for literature reviews. Annu Rev Psychol 2001; 52: 59–82 CrossRef MEDLINE|
|e5.||Lenhard W, Lenhard A: Psychometrica – Institut für psychologische Diagnostik. www.psychometrica.de/effekstaerke.html (last accessed on 20 January 2015).|
Preferences of Information Dissemination on Treatment for Bipolar Disorder: Patient-Centered Focus Group StudyJMIR Mental Health, 201910.2196/12848
Health Literacy as Communicative Action—A Qualitative Study among Persons at Risk in the Context of Predictive and Preventive MedicineInternational Journal of Environmental Research and Public Health, 202010.3390/ijerph17051718
Deutsches Aerzteblatt Online, 201610.3238/arztebl.2016.0299b
Experiencing Positive Health, as a Family, While Living With a Rare Complex Disease: Bringing Participatory Medicine Through Collaborative Decision Making Into the Real WorldJournal of Participatory Medicine, 202010.2196/17602
Deutsches Aerzteblatt Online, 201510.3238/arztebl.2015.0663
Deutsches Aerzteblatt Online, 201610.3238/arztebl.2016.0299a
Deutsches Aerzteblatt Online, 201610.3238/arztebl.2016.0299c