This week’s review brings together five articles spanning shared decision making (SDM), digital decision aids, and emerging human–AI collaboration models in rehabilitation. Each paper offers a different vantage point on how clinicians and patients can make more informed, values-aligned decisions in modern care environments.

Monday: Hoffmann, Bakhit, & Michaleff (2022)

Shared decision making and physical therapy: What, when, how, and why?

Hoffmann and colleagues provide a clear, comprehensive masterclass on SDM within physical therapy. Their discussion centers on SDM as a collaborative process that blends evidence with patient preferences, helping clinicians move away from paternalistic care. The article addresses the mechanics of SDM — identifying decisions, explaining conditions, weighing options, eliciting values, and arriving at a joint plan — and highlights where SDM is most useful, particularly when multiple treatment paths offer similar effectiveness but different benefit-harm profiles.

Despite strong support for SDM across the profession, real-world uptake remains limited. Observational research reveals low engagement scores, suggesting that even experienced clinicians tend to underestimate how much patients wish to be actively involved. The authors highlight organizational barriers but also point to the promise of patient decision aids, structured teaching, and integration of SDM competencies into PT curricula.

Main findings

• SDM improves communication, satisfaction, and expectation accuracy.

• PTs report favorable attitudes toward SDM but often practice it incompletely.

• Patients consistently desire more involvement than clinicians assume.

• SDM is particularly valuable when multiple acceptable treatment paths exist.

• Training and embedded workflow support are essential for sustainable use.

APA Citation

Hoffmann, T., Bakhit, M., & Michaleff, Z. (2022). Shared decision making and physical therapy: What, when, how, and why? Brazilian Journal of Physical Therapy, 26(1), 100382. https://doi.org/10.1016/j.bjpt.2021.100382

Tuesday: Moore & Kaplan (2018)

A framework and resources for shared decision making: Opportunities for improved physical therapy outcomes

Moore and Kaplan outline a structured, three-stage framework for SDM: preparing for collaboration, exchanging information, and affirming and implementing a care plan. The authors connect SDM to a wide range of rehabilitation contexts, illustrating how SDM improves adherence, communication, and outcomes.

By drawing on the Theoretical Domains Framework, the article shows why SDM adoption lags even among clinicians who value collaboration. Knowledge gaps, ingrained habits, workflow challenges, and inconsistent organizational support all play a role. To bridge these gaps, the authors recommend practical tools such as teach-back methods, motivational interviewing, decision aids, and the consistent use of patient-reported outcomes.

Main findings

• SDM improves patient engagement, adherence, and satisfaction.

• Actual use remains low, despite strong clinician endorsement.

• The three-stage SDM model offers a clear path for implementation.

• TDF analysis helps identify barriers that can be addressed through training and system design.

• Practical, communication-focused tools make SDM more actionable.

APA Citation

Moore, C. L., & Kaplan, S. L. (2018). A framework and resources for shared decision making: Opportunities for improved physical therapy outcomes. Physical Therapy, 98(12), 1022–1036. https://doi.org/10.1093/ptj/pzy102

Wednesday: Pel-Littel et al. (2021)

Barriers and facilitators for shared decision making in older patients with multiple chronic conditions

Pel-Littel and colleagues examine SDM through the lens of aging, multimorbidity, and the added complexity of caregiver involvement. Their systematic review identifies 149 barriers and 67 facilitators across patient, caregiver, clinician, organizational, and policy domains.

Older adults with multiple chronic conditions often experience functional or cognitive limitations that make SDM difficult, yet their lived experience and contextual knowledge are deeply valuable inputs. The review highlights the need for triadic SDM — patient, caregiver, clinician — and identifies explicit invitation, adapted communication, and coordination across providers as essential features of effective decision processes in this population.

Main findings

• SDM is underused in older adults with MCCs, despite high potential benefit.

• Barriers include cognitive impairment, time pressure, fragmented care, and communication challenges.

• Facilitators include caregiver support, structured information, and explicit invitations to participate.

• Effective SDM often requires a triadic care model.

• Organizational and policy environments greatly influence uptake.

APA Citation

Pel-Littel, R. E., Snaterse, M., Teppich, N. M., Buurman, B. M., van Etten-Jamaludin, F. S., van Weert, J. C. M., Minkman, M. M., & Scholte op Reimer, W. J. M. (2021). Barriers and facilitators for shared decision making in older patients with multiple chronic conditions: A systematic review. BMC Geriatrics, 21(112). https://doi.org/10.1186/s12877-021-02050-y

Thursday: Jayakumar et al. (2021)

AI-enabled patient decision aid vs educational material for knee osteoarthritis

In this randomized trial, Jayakumar and colleagues evaluated whether an AI-powered patient decision aid could improve SDM and decision quality among individuals with knee osteoarthritis considering total knee replacement. The tool integrates patient-reported outcomes with machine learning to generate personalized predictions, offering an individualized counterpoint to generic education materials.

The AI tool improved decision quality by twenty percent, increased shared decision-making scores, enhanced patient satisfaction, and led to better functional outcomes — without extending consultation time. Importantly, the intervention did not lead to higher rates of surgery, suggesting that improved SDM does not necessarily drive more invasive care.

Main findings

• AI-enabled tool significantly improved decision quality and SDM.

• Patient satisfaction and KOOS JR scores improved.

• No difference in surgery rates or visit duration.

• Personalized PROM-based predictions can meaningfully enhance SDM.

APA Citation

Jayakumar, P., Moore, M. G., Furlough, K. A., Uhler, L. M., Andrawis, J. P., Koenig, K. M., Aksan, N., Rathouz, P. J., & Bozic, K. J. (2021). Comparison of an artificial intelligence-enabled patient decision aid vs educational material on decision quality, shared decision-making, patient experience, and functional outcomes in adults with knee osteoarthritis: A randomized clinical trial. JAMA Network Open, 4(2), e2037107. https://doi.org/10.1001/jamanetworkopen.2020.37107

Friday: Lee et al. (2021)

A human–AI collaborative approach for rehabilitation assessment

Lee and colleagues explored how explainable, collaborative AI can complement therapist expertise in stroke rehabilitation assessment. Their system blends machine learning with clinician-defined rules and provides a transparent visualization interface. Therapists can refine the algorithm by giving feedback on feature relevance, aligning model behavior with clinical judgment.

The study demonstrates that human–AI collaboration improves both assessment accuracy and therapist agreement. The iterative feedback loop also enhances model performance, supporting the idea that AI systems designed for co-learning — rather than one-way automation — may be more trustworthy and useful in clinical practice.

Main findings

• Human-AI collaboration improved assessment accuracy and inter-rater agreement.

• Therapist feedback significantly enhanced model performance.

• Hybrid rules-plus-ML design supports transparency and clinician trust.

• Visualizations make complex motion data clinically interpretable.

• Findings illustrate how AI can augment, not replace, therapist expertise.

APA Citation

Lee, M. H., Siewiorek, D. P., Smailagic, A., Bernardino, A., & Bermúdez i Badia, S. (2021). A human-AI collaborative approach for clinical decision making on rehabilitation assessment. In Proceedings of the CHI Conference on Human Factors in Computing Systems (pp. 1–14). ACM. https://doi.org/10.1145/3411764.3445472

Final Summary of the Week

Taken together, these articles paint a clear picture: meaningful decision making in rehabilitation sits at the intersection of communication, shared values, evidence, and increasingly, intelligent digital tools.

• SDM remains a professional ideal that is not yet consistently realized.

• Older adults and individuals with multimorbidity require adapted, triadic SDM approaches.

• AI-enabled tools can enhance SDM, personalize information, and improve patient experience.

• Collaborative, explainable AI models show promise for complex tasks such as movement assessment.

• Training, workflow integration, and organizational support remain essential across all applications.

The week as a whole highlights a profession striving to blend long-standing therapeutic principles with new technological tools to create more collaborative, personalized, and effective care.

Topics to Discuss This Week

• How to embed structured SDM processes into everyday PT workflows.

• Training models that support SDM skill development in students and clinicians.

• The role of caregivers in SDM for older adults with multimorbidity.

• Ethical considerations surrounding AI-supported decision making.

• How AI tools should be designed to complement, not replace, clinician judgment.

• Strategies for aligning value-based care with SDM and patient-reported outcomes.

• What “shared” decision making should look like in hybrid digital-clinical models.