Aetna considers constraint-induced movement therapy (CIMT) medically necessary for treatment of upper‑limb hemiparesis following stroke only when all required eligibility elements are documented: the patient has a diagnosis of stroke with upper‑limb hemiparesis, is medically stable and cognitively capable of participating, and demonstrates at least 10 degrees of active wrist and finger extension. Providers should align reported procedure codes (for example, therapeutic/OT/PT codes such as 97110, 97112, 97530, 97140 and applicable HCPCS for home therapy G0151, S9131) and ICD‑10 selection with the clinical indication and the investigational exclusions listed in the policy (see codes for constraint‑induced aphasia 92507/92508 and neurostimulation-related codes with conditional exclusions).

Uses and specific combinations listed in the policy are classified as experimental and investigational — not covered. Examples include CIMT for brachial plexus palsy, cerebral palsy, congenital hemiplegia, hemiplegia from brain tumors, lower‑limb hemiparesis post‑stroke, multiple sclerosis, Parkinson’s disease, spinal cord injury, traumatic brain injury; and CIMT combined with peripheral nerve stimulation, transcranial direct current stimulation, transcranial magnetic stimulation, biofeedback, electroacupuncture, group‑based CIMT, and constraint‑induced aphasia/language therapy alone or combined with TMS. These indications or combinations should not be billed as covered CIMT interventions under the policy and are subject to denial.

Evidence summarized in systematic reviews and randomized trials shows important effect modifiers and comparative findings that should inform coverage decisions and authorization requests. For post‑stroke patients, timing and dose matter: pooled acute/subacute RCT data suggest a trend favoring CIMT but indicate that low‑intensity CIMT may show greater benefit than very high‑intensity protocols in some analyses; separate dose‑matched comparisons frequently find no superiority of CIMT over an equivalent amount of intensive bimanual or individualized upper‑limb therapy. For children with unilateral cerebral palsy, CIMT outperformed low‑dose therapy on some measures but not dose‑matched or high‑dose alternatives, and overall study quality was judged low to very low. Adjuncts such as tDCS show small improvements for activities of daily living in some meta‑analyses but effects on upper‑extremity function are inconsistent and attenuate in sensitivity analyses restricted to higher‑quality trials.

Protocol features and outcome domains reported across the evidence base that are relevant to documentation and prior authorization include: intensive, repetitive task practice emphasizing the affected limb combined with restraint of the unaffected limb for defined daily periods (examples reported range from brief daily sessions up to multi‑hour per‑day programs over 1–10 weeks with mean total hours around 137 in pediatric trials); commonly used outcome instruments include active range of motion, Fugl‑Meyer, Motor Activity Log (MAL/PMAL‑R), Assisting Hand Assessment (AHA), Box and Blocks, hand dynamometry and quality‑of‑life measures. For obstetric brachial plexus palsy, reviewers recommend a minimum program duration of about 2 weeks, but no single standardized mCIMT protocol is supported due to heterogeneity.

Provider action summary — prior authorization and documentation expectations: prior authorization requests for CIMT should document the clinical rationale and planned dose/intensity (frequency, session length, total planned hours), demonstrate that eligibility criteria are met (stroke diagnosis, medical stability, cognitive status, and documented motor threshold such as ability to extend wrist and 3 fingers ≥ 10°), and list targeted outcome measures to be used for monitoring. Where intensive, daily protocols are proposed (e.g., consecutive daily sessions for ~2 weeks), prior authorization should explicitly justify feasibility and setting. The policy does not prescribe a single required prior authorization code set in the clinical evidence sections, but it does enumerate relevant CPT/HCPCS/ICD‑10 codes and lists specific codes for constraint‑induced aphasia and neurostimulation exclusions — providers must code consistently with the documented indication and the policy’s investigational exclusions.

Coverage‑relevant operational notes and risks: incomplete reporting of drop‑outs, adverse events or inadequate outcome measurement in submitted documentation may affect coverage determinations; nonadherence risk (for example, inability to attend daily sessions over 2 consecutive weeks) may limit expected benefit and can be a reason to deny or modify authorization. Finally, the Clinical Policy Bulletin is an administrative guidance document and does not itself guarantee coverage; follow plan‑specific prior authorization procedures and links in the administrative information to ensure compliance.

Evidence limitations and clinical implications summarized in the policy affect which uses of CIMT are considered investigational or not medically necessary. The policy explicitly states that uses listed in the experimental and investigational section are considered not medically necessary due to insufficient evidence, and that uncertainty remains regarding optimal dosing, timing, and which specific CIMT models (including mCIMT variants) are superior.

Key evidence‑based findings that influence coverage stance: in stroke, small RCTs and meta‑analyses show improvements in some upper‑limb outcomes after CIMT but trials are heterogeneous and sample sizes are often small; Nijland et al’s acute/sub‑acute stroke meta‑analysis (5 RCTs, n=106) found favorable mean differences but suggested low‑intensity CIMT may be more beneficial than high‑intensity CIMT in that time window. In children with unilateral cerebral palsy, a Cochrane review (36 trials, 1,264 participants) reported that CIMT improved bi‑manual performance and uni‑manual capacity versus low‑dose comparators but showed no advantage over dose‑matched or high‑dose therapies; the overall quality of evidence was rated low to very low.

Adjunctive interventions (for example, tDCS) and combined protocols have mixed or preliminary results: meta‑analyses found small improvements in activities of daily living with tDCS added to therapy at end of intervention (moderate‑quality evidence) but effects on upper‑extremity function and muscle strength were inconsistent and often non‑significant when restricting to higher quality trials; reporting of adverse events and drop‑outs was incomplete in many studies and should be tracked in clinical documentation.

Implications for practice and authorization: because dose and comparator matter, payers may reasonably require documentation that the proposed CIMT program specifies frequency, duration, total planned hours, setting (clinic vs home), and measurable baseline and follow‑up outcome instruments (e.g., Fugl‑Meyer, MAL, AHA, Box and Blocks). For pediatric OBPP, available reviews recommend at least a 2‑week program, but heterogeneity prevents endorsing a single protocol. Studies that excluded patients with severe cognitive impairment or severe spasticity limit generalizability; such patients typically do not meet the documented eligibility criteria and may be excluded from coverage.

Alternatives and step‑therapy considerations: evidence consistently shows that CIMT is superior to low‑dose comparators but not to dose‑matched intensive alternatives; therefore, step‑therapy approaches that consider equivalent‑intensity upper‑limb programs (for example, bimanual intensive therapy or individualized occupational therapy) prior to or instead of CIMT can be reasonable. The policy does not impose explicit step‑therapy mandates but documents the comparative findings and suggests sequencing and justification should be included when requesting coverage.

Administrative reminder: providers must follow plan‑specific prior authorization procedures and use the policy’s listed codes appropriately. The Clinical Policy Bulletin provides administrative links and notes that it does not constitute a contract or guarantee of coverage — supporting documentation and adherence to the insurer’s authorization processes remain necessary for coverage determinations.

Prior authorization and documentation guidance in the policy is oriented around demonstrating eligibility, clinical rationale, and the planned dose/intensity of CIMT when coverage is requested. Required clinical documentation should include the diagnosis (stroke with upper‑limb hemiparesis when seeking coverage), evidence of medical stability and intact cognition, and objective baseline motor function demonstrating at least 10 degrees of active wrist and finger (or wrist and 3 fingers) extension as the minimum threshold cited in the policy.

Authorization requests should specify the proposed protocol: session frequency, session length, total planned hours, restraint method and hours per day, setting (clinic, home, or hybrid), and targeted outcome measures for monitoring (examples cited in trials include Fugl‑Meyer, Motor Activity Log/PMAL‑R, Assisting Hand Assessment, Box and Blocks, grip dynamometry and active range of motion). For intensive CIMT protocols that require daily sessions over consecutive weeks (for example, programs with daily sessions for approximately 2–3 weeks), prior authorization should explicitly justify feasibility and patient adherence.

Coding and billing alignment: the policy enumerates relevant CPT/HCPCS and ICD‑10 codes that should be used consistent with the indication and investigational exclusions. For constraint‑induced aphasia, use 92507 (individual) or 92508 (group) as appropriate. Therapy procedure codes commonly used for CIMT interventions include 97110, 97112, 97530, 97140; home‑based or per‑diem codes include G0151 and S9131. Neurostimulation and programming codes (e.g., 64555–64595, 90867–90869, 95970–95975) are listed with conditional exclusions when combined with CIMT in the policy and should not be billed as covered when the policy lists the combination as investigational.

Reporting expectations and outcome documentation: trials and the policy recommend routine reporting of drop‑outs and adverse events; submitted clinical records should document adverse events, adherence, and pre/post outcome measures. Evidence quality and incomplete reporting in the literature are noted as factors that may affect coverage determinations — lack of robust outcome data or failure to document planned intensity/duration and outcomes may increase the risk of denial.

Operational and adherence considerations: the policy highlights that some intensive or group protocols (for example, daily attendance over 2 consecutive weeks) may be infeasible for some patients due to transportation or other barriers; inability to adhere to the planned intensive schedule may limit expected benefit and is relevant to coverage decisions. Finally, providers are reminded that the Clinical Policy Bulletin is administrative guidance and not a guarantee of payment; follow plan‑specific prior authorization processes and include links and review history where required.

Background: Constraint‑induced movement therapy (CIMT) is an intensive rehabilitation approach that aims to increase use of the more‑affected upper extremity by restraining the less‑affected limb during periods of focused, repetitive practice. Typical study protocols constrain the unaffected limb for defined periods and deliver concentrated therapy over a short course (commonly about 2–3 weeks). Evidence demonstrates small to moderate improvements in selected post‑stroke patients, while findings in other conditions (e.g., pediatric cerebral palsy, MS) are mixed and limited by heterogeneity and small sample sizes.