Evaluation of a Digital Decision Aid for Knee Replacement Surgery

A Stepped-Wedge, Cluster-Randomized Trial

Background: We studied whether an individualized digital decision aid can improve decision-making quality for or against knee arthroplasty.

Methods: An app-based decision aid (EKIT tool) was developed and studied in a stepped-wedge, cluster-randomized trial. Consecutive patients with knee osteoarthritis who were candidates for knee replacement were included in 10 centers in Germany. All subjects were asked via app on a tablet about their symptoms, prior treatments, and preferences and goals for treatment. For the subjects in the intervention group, the EKIT tool was used in the doctor-patient discussion to visualize the individual disease burden and degree of fulfillment of the indication criteria, and structured information on knee arthroplasty was provided. In the control group, the discussion was conducted without the EKIT tool in accordance with the local standard in each participating center. The primary endpoint was the quality of the patient’s decision on the basis of the discussion of indications, as measured with the Hip and Knee Quality Decision Instrument (HK-DQI). (Registration number: ClinicalTrials.gov:NCT04837053).

Results: 1092 patients were included, and data from 1055 patients were analyzed (616 in the intervention group and 439 in the control group). Good decision quality, as rated by the HK-DQI, was achieved by 86.0% of patients in the intervention group and 67.4% of patients in the control group (relative risk, 1.24; 95 % confidence interval, [1.15; 1.33]).

Conclusion: A digital decision aid significantly improved the quality of decision-making for or against knee replacement surgery. The widespread use of this instrument may have an even larger effect, as this trial was conducted mainly in hospitals with high case numbers.

Knee replacement is a successful and (cost-)effective treatment option for advanced osteoarthritis of the knee when conservative management has provided inadequate relief of symptoms (1, 2, 3, 4). In Germany about 180 000 knee replacements are implanted each year (5), with frequency varying greatly from region to region (6, 7). One possible reason for this variability is the absence of standardized decision criteria on the basis of which the indication for knee replacement can be reached in a transparent and uniform manner (6, 8). For this reason, the project “Evidence and consensus-based indication criteria for knee arthroplasty (EKIT-Knee)” was initiated (8) and established as an S2k Guideline (4, 9).

Although knee replacements have very good survival rates (no revision surgery after 15 years in over 90%) (10) and knee pain and function improve in 95% of those operated (11), not all patients benefit from the surgery. In a recent systematic literature review, the proportion of dissatisfied patients is reported to be ten percent (12). The reasons for this include, amongst others, indications, unrealistic expectations placed on the knee replacement, and poor patient information during the shared decision-making process.

A randomized controlled trial (RCT) showed that patients report better outcomes after knee replacement following shared decision-making (SDM) (13). In Germany, SDM is required by the Patients’ Rights Act and is enshrined in law (14) but is not actually applied to a sufficient degree for common elective procedures in trauma and orthopedic surgery (15).

Given the problems described above (heterogeneous indications, not enough consideration of treatment goals), a decision aid that addresses both guideline-based indication criteria and the patient’s treatment goals and preferences could improve this situation by providing additional support for the doctor-patient discussion.

The aim of the present study was to examine to what extent a digital decision-support tool developed specifically for this purpose (EKIT tool), which took into account not only the individual degree of fulfillment of the guideline-based indication criteria but also the personal expectations of the patients, can improve decision-making quality for or against knee arthroplasty.

Methods

Study registration, ethics, and data protection

The study was registered with ClinicalTrials.gov (NCT04837053), and the study protocol was published in advance (16). More details can be found in the eMethods section. The Ethics Committee of the Faculty of Medicine at the Technische Universität Dresden (EK-271062020) approved the study. The publication is in accordance with the CONSORT reporting standard for stepped-wedge cluster randomized studies (17).

Study design

The value-based TKR (total knee replacement) study is a prospective cluster randomized study using a stepped-wedge design (SW-RCT). The intervention was introduced in each cluster (study center) in a stepwise fashion according to a predefined schedule and in a randomized order (Table 1). The clusters therefore served as a control group before implementation of the intervention, and each cluster completed an intervention phase lasting at least four months and a control phase of at least two months.

Table 1

Stepped-wedge design of the value-based TKR study – scheduling of the various intervention phases in the sequences

Study setting

Fifteen centers were asked to participate in the study. Of these, ten centers of different levels of care finally participated. They had performed at least 200 primary knee arthroplasties per year and had the necessary infrastructure to conduct such a study. In detail, these were two university hospitals, one hospital of the German statutory accident insurance, four major orthopedic specialist clinics, and three outpatient centers with inpatient beds.

Recruitment, inclusion and exclusion criteria, study schedule

The recruitment strategy was based on a weekly consecutive inclusion of the first two suitable patients after verification of the inclusion and exclusion criteria.

Patients were included in the study if they had osteoarthritis of the knee and had been referred to one of the study centers for knee arthroplasty. At this point in time, a confirmed indication for surgery had not yet been established. Other inclusion criteria were:

• Age of at least 18 years
• Understanding the course of study and the German language
• Written informed consent to participate in the study.

Recruitment of the study participants and obtaining the informed consent were carried out during the intervention and control phases using the same procedure. At the time of inclusion to the study, patient data were masked with respect to the study arm allocation.

Once informed consent for study participation had been obtained, the patients completed the web-based questionnaire using a tablet. Data acquisition was performed according to the same principle for both the intervention and the control group (eTable 1). These data formed the basis for application of the EKIT tool in the intervention group (Figure).

Figure

Study process

eTable 1

Data acquisition

Study arms

The control group comprised those patients who had been included in the study centers during the control phase prior to implementation of the intervention (Figure). The doctor-patient discussion was conducted in accordance with the local standard in each participating center. The intervention group comprised patients who had been included in the study by the centers during the intervention phase. Here, the doctor-patient discussion was conducted using the EKIT tool.

Intervention

The intervention to be tested was the application of an app-based decision aid developed specifically for the study (EKIT tool). The aim was to improve decision quality based on consideration of the patient’s details (prior treatment, symptoms, expectations) and guideline-based indication criteria in comparison with routine care without their use.

The EKIT tool was developed in cooperation with clinical experts, health-service researchers, and medical IT specialists. Comprehensibility and data entry were tested in advance with the target group (patients and orthopedic surgeons) by conducting a feedback discussion after use, based on the think-aloud method. Any necessary adjustments were then made.

The EKIT tool includes three important features to promote SDM and to support consultation:

• It provides a systematic and structured presentation of the individual patient-specific disease burden.
• It enables visualization of the degree of fulfillment of the guideline-based indication criteria (4).
• It provides prepared health information about knee arthroplasty.

The developed decision-making tool was evaluated using the International Patient Decision Aid Standards instrument (IPDASi) (18) (eTable 2).

eTable 2

International Patient Decision Aid Standards instrument (IPDASi) (18)

Patient-specific and disease-specific information

The following patient data was obtained using the tablet to record the patient’s disease burden (eSupplement Figure 1):

• Prior treatments
• Knee function (Oxford Knee Score, OKS) (19)
• Health-related quality of life questionnaire (20)
• The patient’s own goals and preferences for treatment.

The study doctors added the following information on completion of the clinical and radiographical examination:

• The Kellgren and Lawrence osteoarthritis grade
• Leg axis
• Range of motion and stability of the knee joint
• Any comorbidities.

This information was subsequently available for the doctor-patient discussion. Functional impairment and quality of life (QoL) of the individual patient were compared with a large cohort of National Health Service patients before planned knee arthroplasty surgery (11) (eSupplement Figures 2–4).

The degree of fulfillment of the guideline recommendations

The degree of fulfillment was visualized for the indication criteria for knee arthroplasty (main criteria, secondary criteria, contraindications) (eSupplement Figure 2). The following five main criteria had to be met to establish the indication:

• Presence of moderate or severe knee osteoarthritis
• Knee pain for at least three months
• Guideline-based conservative treatment for at least three months with no satisfactory outcome
• Knee joint restricting QoL
• Patient-reported level of suffering.

If not all criteria were completely fulfilled, secondary criteria such as pronounced instability, for example, could still justify the indication (4).

Health information

The EKIT tool was used to provide patients with standardized information on the benefits and risks of knee replacement surgery compared with conservative treatment adopted from the evidence-based health information of the German Institute for Quality and Efficiency in Health Care (IQWIG) (5) (eSupplement Figure 4). This included information on complication rate, healing process, the likelihood of a revision surgery, and general satisfaction after knee replacement surgery.

Shared decision-making

Finally, the patient’s treatment goals were presented. Their likelihood of fulfillment after knee replacement surgery was assessed by the study doctors and discussed with the patients. The decision for or against knee arthroplasty was ultimately made with appropriate justification.

Endpoints

The primary endpoint was decision quality after the doctor-patient discussion as measured with the Hip and Knee Quality Decision Instrument (HK-DQI) (21). This binary endpoint comprised the following:

• A knowledge score regarding the benefits and risks of the operation (informed choice)
• A concordance score summarizing the match between the patient’s treatment preference and the doctor’s treatment recommendation.

The knowledge score was modified for the German health system and consisted of five questions regarding knee arthroplasty (eTable 3). The primary endpoint was recorded immediately after the discussion.

eTable 3

HK-DQI knowledge score

According to the HK-DQI, good decision quality (19) is achieved if 60% or more of the knowledge questions were answered correctly and the treatment decision for or against knee arthroplasty was made in line with the treatment preference expressed by the patient. As a change to the study protocol, the actual therapy carried out after twelve months was not used as the basis for assessing the quality of the decision, as this was also affected by other factors (in particular postponed surgery due to the COVID-19 pandemic).

Secondary endpoints were satisfaction with the informed consent discussion and user-friendliness of the EKIT tool (eTable 1). Other endpoints will be evaluated and published separately.

Sample size calculation

A total of 1080 patients (an average of 108 patients per study center) had to be recruited in order to demonstrate a ten percent increase in the “good decision quality” parameter in the intervention group (with an alpha error of 0.05%, a power ≥ 80%, ten participating study centers, and an assumed drop-out rate of 11%) (eMethods section).

Statistical analysis

The proportions of the included patients who achieved good decision quality, as rated by the binary endpoint HK-DQI, were compared between the intervention and control group using Poisson regression model with a log link function and robust standard error estimation (22) (eMethods section). The intervention effect of the primary endpoint is expressed as the relative risk (RR) with a 95% confidence interval (CI). The statistical analysis was performed using the R statistics software version 4.2.2 (23).

Results

Descriptive results

Between 06/2021 and 03/2023, a total 1092 patients from ten study centers were included in the study. It was not possible to evaluate the data of 37 patients at the start of the study due to withdrawal of consent to participate (n = 32) or technical problems (n = 5). The data of 1055 patients (96.6%; control group: n = 439; intervention group: n = 616) are included in the analysis. The CONSORT flow diagram (eSupplement Figure 5) provides an overview of the patients included in each group at the different time points of the study. Age, sex, body mass index (BMI), knee function (OKS), health-related quality of life (EQ-5D score), and prior treatment as well as the musculoskeletal symptoms at baseline were comparable between the two groups (Table 2).

Table 2

Patient characteristics (absolute and relative frequencies or mean value [MV] and standard deviation)

The consultations revealed that 92.3% of participants presented Kellgren and Lawrence grade 3 osteoarthritis of the knee (moderate osteoarthritis) or grade 4 (severe osteoarthritis). Almost all the patients reported pain (51.8% constantly, 39.9% daily, 6.9% several times a week, and 1.4% infrequently). The subjective level of suffering was very high with 7.3 (SD = 2.1) on a scale of ten, and presentation at the clinic had been preceded by a sometimes very long period of treatment of the knee complaints (74.4% >1 year, 14.9% three to 12 months, 10.7% <3 months). A number of different forms of treatment had been provided during this time (including oral analgesics 77.3%, topical ointments 72.4%, intraarticular injections 52.0%, physiotherapy 56.5%).

The preferred treatment reported by the participants was mainly knee replacement 73.1% and conservative treatment 9.1%, while 17.8% were unsure (eTable 4).

eTable 4

Treatment preference before the doctor-patient discussion

Primary endpoint

Good decision quality was demonstrated in 86.0% of patients in the intervention group as opposed to 67.4% of patients in the control group (Table 3, eSupplement Figure 6).

Table 3

HK-DQI (absolute and relative frequencies)

The model-based comparison revealed significantly better decision quality in the intervention group with a relative risk RR of 1.24 (95% CI [1.15; 1.33]) in the HK-DQI total knowledge score. The adjusted sensitivity analysis carried out for the OKS and the EQ-5D scores demonstrated no intervention-related change (RRsens = 1.23 [1.14; 1.32]).

The number needed to treat (NNT) based on the absolute risk reduction (ARR = 0.19) was 6.

Secondary endpoints

As part of the shared decision-making process, 86.4% of participants agreed upon having knee arthroplasty, in 1.1% joint-preserving surgery, and in 12.5% to continue conservative treatment (eTables 5 and 6).

eTable 5

Treatment recommendation

eTable 6

Subgroup analysis according to treatment preference prior to the doctor-patient discussion

Data entry in the EKIT tool via tablet in the intervention group was achieved without assistance by 47.2% of participants, 24.4% required help, and 28.4% were performed by the study assistant in consultation with the patients (eTable 7).

eTable 7

Reasons why the tablet was not used independently for data entry (multiple entries)

Satisfaction with the informed consent discussion was very high both in the intervention and the control group among patients (98.7%) as well as doctors (97.1%), with no relevant difference between the groups (eTable 8).

eTable 8

Satisfaction with the informed consent discussion (doctor/patient)

Discussion

The present study demonstrated significantly better decision quality resulting from use of the EKIT tool during the decision-making process for or against knee replacement surgery. Patient data was gathered, evidence-based health information was conveyed, and individual treatment objectives were agreed upon with the aid of the EKIT tool. Based on this, a standardized and quality-assured decision aid was provided by visualizing individual fulfilment of the guideline-based indication criteria. To our knowledge, this is the first digital decision-support tool which gathers such data for reaching an indication for surgery, combines this with patient expectations, while at the same time providing structured support for shared decision-making.

Decision aids are standardized instruments that help patients to participate in the decision-making process (24). The overriding goal is to reach a values-congruent and informed medical decision, together with the clinical team (25). Patients are more satisfied, and it is more likely that they will participate actively in the treatment if they are actively involved in the therapy decision-making (26). A systematic review of ten randomized studies (case figures ranging from n = 100 to 1911) involving patients with osteoarthritis of the hip or knee who were considering joint arthroplasty showed that decision-making aids improve knowledge about the planned treatment and decision quality while lowering decisional conflict (27). However, there was only “low to moderate certainty of evidence” according to the GRADE system with regard to decision quality. A systematic review of six randomized controlled studies (RCTs) (n between 120 and 336) on decision aids found that, with respect to knee arthroplasty, patients were better informed about treatment options for knee osteoarthritis (28).

An RCT involving a comparatively small number of patients (n = 155) showed an improvement in decision quality (HK-DQI, Odds Ratio [OR] 2.08; 95% CI [1.08; 4.02]) (29). The HK-DQI indicated good decision quality for 41.3% of the participants in the control group and 60.0% in the intervention group. This indicated that, despite using the decision aid, the proportion of patients with good decision-making quality was lower than that of the control group of our study. Consequently, the actual effect of the decision aid was less pronounced in our study.

A large randomized multicenter study (n = 854) showed that improved patient-centered decision-making (HK-DQI) led to greater gains in quality of life and better disease-specific outcome (Knee Osteoarthritis Outcome Score [KOOS]) in osteoarthritis of the knee (13). However, only 68% of participants reached a qualitatively good and informed choice. Applying the same definition, this corresponds to the rate in our control group (routine care). By contrast, use of the EKIT tool resulted in good decision quality in 86 percent of participants, which is significantly more often. The manner of application is a possible explanation for this. While the majority of decision aids reported in publications to date involve entirely online use (28), the EKIT tool was developed explicitly for use during a direct doctor-patient discussion. This means that issues arising during direct interaction can be better dealt with than in a purely online-based process. Those patients in particular who were uncertain about their treatment preference prior to the doctor-patient discussion could have benefited from this.

The strengths of the present study are the large sample size, its multicenter randomized setting, and the involvement of clinics of various care levels to represent the health care structures in Germany. Other strengths include consideration and inclusion of indication criteria (subjective level of suffering, amongst others) and the incorporation of expectations in the individual therapy decision.

Limitations of the study are the different recruitment rates of the study centers due to the restrictions placed on elective surgery during the Covid-19 pandemic. Nevertheless, the evaluation proved to be robust, thus allowing the assumption that the results are valid. Only study centers undertaking at least 200 knee arthroplasty procedures per year were included in order to achieve the calculated case numbers, so smaller, less specialized clinics were excluded. This may have led to an underestimation of the true effect of the EKIT tool, as the participating centers were already highly standardized, which in turn could represent a potential reason for the high decision quality in the control group as compared with the literature (13, 29). An additional interview study to evaluate the EKIT tool involving doctors and patients will help provide further insights in this respect.

Before the EKIT tool is adopted for use in routine health care, adjustments will be necessary that are directed toward reducing the amount of data to be recorded and simplifying its application. Data input was also a technical challenge for the more elderly study participants; only half of them were able to enter the data independently and without assistance. The average age (67 years) was just below the average of 69 years for knee arthroplasty surgery in Germany (30). Despite the expected increase in digital media use in the future, further adaptations for the short and medium-term use of the EKIT tool in routine care are required, such as simplifying data input by introducing voice support, for example.

Conclusion

The newly developed EKIT tool showed a significantly positive effect on decision quality and, once adapted, has the potential of being integrated in routine health care. Focusing on clinics with high case numbers could underestimate the effect if used nationwide.

Acknowledgments

We would like to thank the Federal Innovation Fund for supporting the “Value-based Total Knee Replacement” project (Value-based TKR; 01VSF19036) as well as all the clinics, practices, and patients involved in the project.

Data-sharing statement

Anonymized patient data on which the results of the present article are based, including the Supplementary Material, may be made available three months to two years after publication to those researchers who submit a methodologically meaningful analysis proposal. Any inquiries should be addressed to the corresponding author at [e-mail].

Funding

The Value-based Total Knee Replacement project was supported by the Innovation Fund of the Joint Federal Committee (Value-based TKR, 01VSF19036).

Conflict of interest statement

JL has received institutional funding from Aesculap, Arthrosehilfe, Link, Mathys, Smith & Nephew, und Zimmer Biomet. He has received payments for lectures from Aesculap und Mathys and is member of the Advisory Board of Aesculap.

KPG has received institutional funding from Zimmer, Link, and Aesculap. He has received payments for lectures from the working group Endoprostheses and the Zimmer company. KPG is a member of the Board of Directors or Executive Committee of the German Society for Orthopedics and Orthopedic Surgery (DGOOC), the German Society for Orthopedics and Trauma (DGOU), EndoCert, and Endoprosthesis Register Germany (EPRD).

JS has received institutional funding from the Federal Ministry of Health (BMG), the Federal Ministry of Education and Research (BMBF), the Free State of Saxony, Novartis, Sanofi, ALK, and Pfizer. He has received consultancy fees from (Advisory Boards) of Sanofi, Lilly, and ALK. JS is a member of the Expert Advisory Board Health and Nursing Care of the BMG and member of the Traffic Light Coalition’s Government Commission for a Modern and Needs-based Hospital Care.

The other authors declare that no conflict of interest exists.

eBox

Value-based TKR Study Group (collaborators)

Manuscript received on 10 April 2024, revised version accepted on 16 July 2024.

Translated from the original German by Dr. Grahame Larkin

Corresponding author
Prof. Dr. med. Jörg Lützner

UniversitätsCentrum für Orthopädie, Unfall- & Plastische Chirurgie

Universitätsklinikum Carl Gustav Carus Dresden an der Technischen

Universität Dresden

Fetscherstr. 74, 01307 Dresden

Cite this as:
Lützner J, Deckert S, Beyer F, Hahn W, Malzahn J, Sedlmayr M, Günther KP, Schmitt J, Lange T, and the Value-based TKR-Studygroup: Evaluation of a digital decision aid for knee replacement surgery—a stepped-wedge, cluster-randomized trial. Dtsch Arztebl Int 2024; 121: 566–72. DOI: 10.3238/arztebl.m2024.0152

References

1. Bannuru RR, Osani MC, Vaysbrot EE, et al.: OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage 2019; 27: 1578–89

2. Kamaruzaman H, Kinghorn P, Oppong R: Cost-effectiveness of surgical interventions for the management of osteoarthritis: a systematic review of the literature. BMC Musculoskelet Disord 2017; 18: 183

3. Nakano N, Shoman H, Olavarria F, Matsumoto T, Kuroda R, Khanduja V: Why are patients dissatisfied following a total knee replacement? A systematic review. Int Orthop 2020; 44: 1971–2007

4. Lützner J, Lange T, Schmitt J, et al.: [The S2k guideline: indications for knee endoprosthesis: evidence and consent-based indications for total knee arthroplasty]. Orthopäde 2018; 47: 777–81

5. Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen (IQWiG): Gesundheitsinformation: Kniearthrose (Gonarthrose). IQWiG 2018.

6. Schäfer T, Pritzkuleit R, Jeszenszky C, et al.: Trends and geographical variation of primary hip and knee joint replacement in Germany. Osteoarthritis Cartilage 2013; 21: 279–88

7. Grote-Westrick M, Zich K, Klemperer D, et al.: Faktencheck Gesundheit 2015: Regionale Unterschiede in der Gesundheitsversorgung im Zeitvergleich. Gütersloh: Bertelsmann Stiftung 2015.

8. Schmitt J, Lange T, Günther KP, et al.: Indication criteria for total knee arthroplasty in patients with osteoarthritis—a multi-perspective consensus study. Z Orthop Unfall 2017; 155: 539–48

9. Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften e.V. (AWMF): S2k-Leitlinie: Indikation Knieendoprothese. AWMF: 2023; 43.

10. Evans JT, Walker RW, Evans JP, Blom AW, Sayers A, Whitehouse MR: How long does a knee replacement last? A systematic review and meta-analysis of case series and national registry reports with more than 15 years of follow-up. Lancet 2019; 393: 655–63

11. National Health Service (NHS): Digital—finalised patient reported outcome measures (PROMs) in England for hip and knee replacement procedures (April 2021 to March 2022) https://digital.nhs.uk/data-and-information/publications/statistical/patient-reported-outcome-measures-proms/finalised-hip-and-knee-replacement-procedures-april-2021-to-march-2022 (last accessed on 14 February 2024).

12. DeFrance MJ, Scuderi GR: Are 20% of patients actually dissatisfied following total knee arthroplasty? A systematic review of the literature. J Arthroplasty 2023; 38: 594–9

13. Sepucha KR, Vo H, Chang Y, et al.: Shared decision-making is associated with better outcomes in patients with knee but not hip osteoarthritis: the DECIDE-OA randomized study. J Bone Joint Surg Am 2022; 104: 62–9

14. Bundesministerium für Gesundheit (BMG): Patientenrechte (2024) https://www.bundesgesundheitsministerium.de/themen/praevention/patientenrechte/patientenrechte (last accessed on 23 July 2024)

15. Valentine KD, Cha T, Giardina JC, et al.: Assessing the quality of shared decision making for elective orthopedic surgery across a large healthcare system: cross-sectional survey study. BMC Musculoskelet Disord 2021; 22: 967

16. Lange T, Deckert S, Beyer F, et al.: An individualized decision aid for physicians and patients for total knee replacement in osteoarthritis (Value-based TKR study): study protocol for a multi-center, stepped wedge, cluster randomized controlled trial. BMC Musculoskel Dis 2021; 22: 783

17. Hemming K, Taljaard M, McKenzie JE, et al.: Reporting of stepped wedge cluster randomised trials: extension of the CONSORT 2010 statement with explanation and elaboration. BMJ 2018; 363: 1614

18. Elwyn G, O‘Connor AM, Bennett C, et al.: Assessing the quality of decision support technologies using the International Patient Decision Aid Standards instrument (IPDASi). PloS one 2009; 4: e4705

19. Naal FD, Impellizzeri FM, Sieverding M, et al.: The 12-item Oxford Knee Score: cross-cultural adaptation into German and assessment of its psychometric properties in patients with osteoarthritis of the knee. Osteoarthritis Cartilage 2009; 17: 49–52

20. Herdman M, Gudex C, Lloyd A, et al.: Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-5L). Qual Life Res 2011; 20: 1727–36

21. Sepucha KR, Stacey D, Clay CF, et al.: Decision quality instrument for treatment of hip and knee osteoarthritis: a psychometric evaluation. BMC Musculoskel Dis 2011; 12: 149

22. Zou GY, Donner A: Extension of the modified Poisson regression model to prospective studies with correlated binary data. Stat Methods Med Res 2013; 22: 661–70

23. R Core Team: R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2021.

24. Stacey D, Legare F, Lewis K, et al.: Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2017; 4: CD001431

25. National Institute for Health and Care Excellence (NICE): Shared decision making: NICE guideline. www.nice.org.uk/guidance/ng197/resources/shared-decision-making-pdf-66142087186885 (last accessed on 14 February 2024).

26. Hibbard JH, Greene J: What the evidence shows about patient activation: better health outcomes and care experiences; fewer data on costs. Health Aff 2013; 32: 207–14

27. Pacheco-Brousseau L, Charette M, Poitras S, Stacey D: Effectiveness of patient decision aids for total hip and knee arthroplasty decision-making: a systematic review. Osteoarthritis Cartilage 2021; 29: 1399–411

28. Riddle DL, Sando T, Tarver T, et al.: Shared decision-making applied to knee arthroplasty: a systematic review of randomized trials. Arthritis Care Res 2021; 73: 1125–33

29. Bansback N, Trenaman L, MacDonald KV, et al.: An online individualised patient decision aid improves the quality of decisions in patients considering total knee arthroplasty in routine care: a randomized controlled trial. Osteoarthr Cartil Open 2022; 4: 100286

30. Endoprothesenregister Deutschland (EPRD): Jahresbericht 2023: Mit Sicherheit mehr Qualität https://www.eprd.de/fileadmin/user_upload/Dateien/Publikationen/Berichte/Jahresbericht2023-Status5_2023–10–24_F.pdf (last accessed on 14 February 2024).

31. Stacey D, Taljaard M, Dervin G, et al.: Impact of patient decision aids on appropriate and timely access to hip or knee arthroplasty for osteoarthritis: a randomized controlled trial. Osteoarthritis Cartilage 2016; 24: 99–107

32. Lange T, Schmitt J, Kopkow C, et al.: What do patients expect from total knee arthroplasty? A Delphi Consensus Study on patient treatment goals. J Arthroplasty 2017; 32: 2093–9.e1