August 18, 2026
Strategic Insights
Continuous AI MSK Care: What the evidence shows
From session-based care to continuous, proactive recovery built around the member.
Musculoskeletal conditions are the number one source of healthcare spending in the United States and the leading cause of workplace disability. They affect one in two U.S. adults, more than 124 million people.¹ ² ³ They are also among the most treatable conditions in medicine. That gap is not a clinical problem. It is a delivery problem.
Traditional MSK care has always been episodic. A member sees a clinician, gets a plan, and feels better for a while. Then life gets busy, no one reaches out, and months later a claim arrives. The periods between sessions decide outcomes, and the old model leaves those periods empty.
This whitepaper shows what changes when care does not stop. It lays out the evidence behind continuous MSK care, the clinical and financial results it produces, and the questions benefits leaders should ask any vendor before renewal.
What this report helps you analyze
Untreated MSK conditions do not leave the balance sheet. They move to another line item as emergency visits, specialist referrals, imaging, avoidable surgery, and downstream mental health cost. The model most employers rely on was never built to prevent that. It engages members only when they show up, and it goes dark in between.
This report gives you the frameworks to diagnose where episodic care is costing you, to pressure-test vendor claims, and to build the finance case for a continuous model you can validate against your own claims data.
Key insights
- Why episodic MSK care keeps failing to bend the cost curve, and why continuous care replaces the old model rather than patches it.
- Where untreated MSK cost is already surfacing in your claims data.¹
- What a continuous care model returns: 3.2x return on investment and $3,177 in direct savings per member per year.⁷
- How continuous care drives a 68% increase in productivity through fewer sick days and less accommodation.⁶
- Evidence that this model matches, and in some head-to-head comparisons exceeds, high-intensity in-person physical therapy on engagement and clinical effectiveness.⁸
- How Phoenix, Sword's AI Care Specialist, works alongside a designated Doctor of Physical Therapy inside a single clinician-led care relationship.
Contributors to White Paper

Exploring the breakthroughs behind AI Care

Evidence-based healthcare insights
Footnotes
- 1
JAMA. 2020;323(9):863–884. doi:10.1001/jama.2020.0734.
- 2
United States Bone and Joint Initiative. The Burden of Musculoskeletal Diseases in the United States (BMUS), 4th edition, 2016.
- 3
United States Bone and Joint Initiative. BMUS, 4th edition. Rosemont, IL; 2016.
- 4
Sword member data, 2025. Members entering programs with moderate to severe surgery intent (30% or above) who end with no or only mild intent (below 30%). Sword member base, 2025 data, 127K engaged members.
- 5
Sword member data, 2025. Members scoring 5 or greater on the Patient Global Impression of Change (PGIC) scale. Sword member base, 2025 data, 127K engaged members. PGIC scale reference: Musculoskeletal Sci Pract. 2023 Feb;63:102709. doi:10.1016/j.msksp.2022.102709.
- 6
Sword Team. RSC study white paper, November 8, 2024. (68% increase in productivity.)
- 7
J Med Internet Res. 2023;25:e49236; JMIR Rehabil Assist Technol. 2019;6(1):e14523; npj Digit. Med. 6, 121 (2023).