CurrentAetnaPolicy 0437

Speech Generating Devices

This Aetna clinical policy governs medical necessity, coverage limitations, and coding for speech generating devices and associated accessories for members with severe expressive speech disabilities; it affects providers ordering, evaluating, or supplying SGDs and speech-language pathologists performing evaluations.

Policy Summary

PayerAetna

PolicySpeech Generating Devices

Policy CodePolicy 0437

Change TypeNo material clinical or coverage changes indicated

Effective DateAug 8, 2000

Next Review DateMay 9, 2024

Key ActionObtain and document a formal written SLP evaluation with specific elements and forward it to the treating physician prior to ordering an SGD.

SourceLink

1SGD per member

4–6 monthsWait after stroke/trauma

92521–92609, E2500–E2512Representative codes

07/11/2023Last review

05/09/2024Next review

Coverage and Medical Necessity Criteria

Trek Health ingests and normalizes Transparency in Coverage data and payer policy updates to give provider organizations a clear view of how commercial reimbursement behaves across markets, payers, and services. Our platform transforms raw payer disclosures into structured intelligence that supports contract evaluation, payer negotiations, and service line strategy. By combining market benchmarks with ongoing policy visibility, Trek helps teams identify variability, risk, and opportunity in commercial reimbursement. The result is faster insight, stronger negotiating positions, and more informed financial decisions.

Evidence summary / background

ALL of the following

SGDs (speech generating devices) provide synthesized or digitized speech output and may include software that enables laptops, desktops, or PDAs to function as SGDs. Synthesized speech allows independent message formulation rather than playback of pre-recorded messages.
SGDs may permit multiple methods of message formulation (e.g., letters, words, pictures, symbols) and multiple access methods (e.g., direct touch, indirect selection, specialized access devices such as joysticks, head mice, optical pointers, scanning devices).
Accessories and upgrades include mounting systems, switches, access devices (optical head pointers, joysticks, scanning devices), memory or software updates, and interface components required to operate the SGD; these should be documented as medically necessary when indicated by the SLP evaluation.
Assessment of need should be performed by a qualified speech-language pathologist (SLP) with appropriate credentials (e.g., Certificate of Clinical Competence from ASHA). The SLP must evaluate cognitive and language abilities and document functional communication goals, training plan, alternative communication methods considered, and demonstration of the member's ability to use the device and accessories.
Evidence from small trials and single-subject studies in autism spectrum disorder (ASD) and developmental language delays suggests that SGDs can increase communication acts, requesting behaviors, reciprocity, and peer engagement in some children, particularly when combined with peer-mediated or structured instructional approaches; however, study limitations (small sample sizes, heterogeneous participants, varied baselines, and potential confounds) limit generalizability.
Randomized and comparative studies indicate both high-tech (SGD) and low-tech (PECS) augmentative and alternative communication approaches can produce significant improvements in functional communication, with no consistent advantage for one modality across all participants; individualized assessment is essential.
Use of SGDs is generally considered after other treatments and, for acquired disorders (eg, stroke, head trauma), is typically not initiated less than 4 to 6 months post-injury to allow for potential spontaneous recovery.
Non-covered items and limitations: Internet/phone services, home modifications to support SGD use, and hardware/software features not directly necessary for generation of audible/verbal speech, email, text, or phone messages (eg, document creation, games, music, video conferencing) are not covered. Desktop/laptop computers, tablets, smartphones, and other non-dedicated devices are not covered as DME because they have substantial non-medical utility.
SGDs with multiple interfaces or multiple access methods are generally not considered medically necessary unless the clinical record clearly documents the need for multiple interfaces or access methods for the individual member.
Multi-lingual modules for SGDs are considered not medically necessary. Only one SGD or speech-generating software program is considered medically necessary per member at a time.

Coding and Timing

Covered CPT CodesCPTCovered

92521	Evaluation of speech fluency (eg, stuttering, cluttering)
92522	Evaluation of speech sound production (eg, articulation, phonological process, apraxia, dysarthria)
92523	Evaluation of speech sound production (eg, articulation, phonological process, apraxia, dysarthria); with evaluation of language comprehension and expression (eg, receptive and expressive language)
92524	Behavioral and qualitative analysis of voice and resonance
92607	Evaluation for prescription for speech-generating augmentative and alternative communication device, face-to-face with the patient; first hour
92608	Each additional 30 minutes
92609	Therapeutic services for the use of speech-generating device, including programming and modification

Covered HCPCS Codes and related codesHCPCSCovered

E2500	Speech generating device, digitized speech, using pre-recorded messages, ≤ 8 minutes recording time
E2502	Speech generating device, digitized speech, >8 minutes to ≤20 minutes recording time
E2504	Speech generating device, digitized speech, >20 minutes to ≤40 minutes recording time
E2506	Speech generating device, digitized speech, >40 minutes recording time
E2508	Speech generating device, synthesized speech, requiring message formulation by spelling and access by physical contact with the device
E2510	Speech generating device, synthesized speech, permitting multiple methods of message formulation and multiple methods of device access
E2511	Speech generating software program, for personal computer or personal digital assistant
E2512	Accessory for speech generating device, mounting system
E2599	Accessory for speech generating device, not otherwise classified
G0153	Services performed by a qualified speech-language pathologist in the home health or hospice setting, each 15 minutes

1–10 of 13

1/2

Other related HCPCS codesHCPCS

E1902	Communication board, nonelectronic augmentative or alternative communication device
L8500	Artificial larynx, any type
L8505	Artificial larynx replacement battery/accessory, any type

Representative ICD-10 CodesICD-10

F80.0 - F80.9	Specific developmental disorders of speech and language
I69.020 - I69.028, I69.120 - I69.128, I69.220 - I69.228, I69.320 - I69.328, I69.820 - I69.828, I69.920 - I69.928	Speech and language deficits following cerebrovascular disease
R47.01 - R47.9	Speech disturbances, not elsewhere classified

Waiting period after acquired disorder — 4-6 months before initiation of SGD

Waiting periodFor acquired disorders (e.g., stroke, head trauma), initiation of an SGD is not usually started less than 4 to 6 months after the event.

RationaleSpeech may gradually improve after head trauma or stroke, so SGDs are used as a last option.

ApplicabilityDelay applies to initiation of SGD following trauma or stroke; clinical judgment and documented recovery trajectory should guide timing.

Provider Responsibilities, Prior Authorization, and Documentation

Prior Authorization

Prior Authorization Determined by Plan

Prior authorization is determined by the member's plan. Clinical Policy Bulletins (CPBs) such as this one are intended to assist in administering plan benefits and do not replace plan-specific authorization rules; providers must check the member’s benefit plan for any prior authorization requirements before ordering an SGD.

Providers should verify prior authorization requirements with the payer and document authorization numbers when required.

Documentation Required

Required SLP Evaluation and Documentation

A formal, written SLP evaluation and recommendation must be completed and forwarded to the member’s treating physician prior to ordering the device. The evaluation must document: functional communication goals and treatment options; a device-specific training schedule; assessment of whether daily communication needs can be met with natural methods; demonstration of the member’s cognitive and physical ability to use the device; evaluation of type/severity/language/cognitive status and expected course; rationale for device and accessory selection; and for upgrades, the functional benefit compared to the prior device. The SLP performing the evaluation must not be employed by or have a financial relationship with the SGD supplier.

Documentation Required

Emphasize Individualized Clinical Documentation

Clinical studies and case reports emphasize the need to document individual cognitive, speech and language skills, device programming and vocabulary adjustments, and other individual characteristics that may predict communication outcomes. Detailed documentation supports medical necessity determinations and helps identify which patients are likely to benefit from SGDs.

Note

Comparative Trial Findings

A randomized pilot trial comparing high-tech SGDs to low-tech AAC (PECS) reported significant improvements with both approaches and found no significant difference between them. This highlights that high-tech devices may be appropriate for some patients, but outcomes can be similar to low-tech options; device selection should be individualized and justified in the SLP evaluation.

Denial Risk

Non-Covered Services and Nonessential Features

Certain services and features are not covered. Internet or phone services and modifications to a member’s home to accommodate SGD use are not covered. Features or hardware/software not necessary for generation of audible/verbal speech, email, text, or phone messages (for example, document/spreadsheet creation, games, music, or general-purpose computing, and video conferencing) are not considered medically necessary.

Home internet or phone service modifications — not covered
Device features not used to meet functional speaking needs (e.g., gaming, general computing) — not covered

Billing Rule

Authorization, Coding, and Step-Therapy Notes

This document does not specify a step therapy protocol or required ordering/coding authorization workflow. CPT/HCPCS codes listed are covered only if selection criteria are met; however, individual plans may impose prior authorization, coding or coverage limits. Providers must check member benefits for any additional coding, prior authorization, or utilization management requirements.

CPT/HCPCS codes in the CPB are covered only when the policy’s selection criteria are met
No step therapy protocol is defined in this CPB; follow plan-specific processes

Key Definitions

Digitized SGD — Device that delivers prerecorded whole-message speech output recorded by someone other than the user

DefinitionDigitized SGD: device that delivers 'whole message' speech output consisting of prerecorded words or phrases recorded by someone other than the user.

Output typePre-recorded audible/verbal speech messages that can be played back on demand.

Example featuresMay include capability to generate email, text, or phone messages and to download manufacturer updates.

Synthesized SGD — Device that translates user input into device-generated speech (text-to-speech)

DefinitionSynthesized SGD: device that translates user input into device-generated speech (text-to-speech) using algorithms representing linguistic rules, allowing independent message creation.

Message formulationMay require spelling or selection via keyboard, touch screen, or other display; not limited to prerecorded messages.

Software optionSpeech generating software programs can enable a laptop, desktop, or PDA to function as a synthesized SGD.

SGD with multiple methods — SGD with multiple message formulation or access methods (≥2 methods)

DefinitionSGD with multiple methods: an SGD that provides multiple methods of message formulation (≥2 of letters, words, pictures, symbols) or multiple methods of device access (≥2 of direct physical contact, indirect selection techniques, specialized access devices).

Message formulation methods (examples)Letters, words, pictures, symbols.

Access methods (examples)Direct physical contact (keyboard/touch screen), indirect selection/scanning, specialized devices (joystick, head mouse, optical pointer, infrared pointer).

SGD — Speech-generating devices definition (aided AAC producing digitized and/or synthesized speech)

Broad definitionSpeech-generating devices (SGDs) are aided augmentative and alternative communication devices or software that generate audible/verbal speech (digitized and/or synthesized) used solely by an individual with a severe speech impairment to meet functional speaking needs.

Included technologiesDigitized speech (prerecorded), synthesized speech (text-to-speech), software that enables computers or PDAs to function as SGDs.

PurposeProvide individuals with severe impairments (e.g., aphasia, apraxia, dysarthria) the ability to meet functional speaking needs when natural speech is insufficient.

Covered featuresMay include capabilities to send email/text/phone messages and to receive manufacturer software updates for covered features.

Speech generating device (SGD) — Definition implied by referenced sources

Functional descriptionDevice or technology used to produce spoken output to assist individuals with severe speech impairment; definition is consistent with referenced literature and Medicare DME guidance.

Clinical contextUsed for patients with severe speech impairments to meet functional communication needs; assessment should be by a qualified SLP.

Literature basisPolicy and definition reference Medicare DME MAC criteria and peer-reviewed studies on SGDs and AAC interventions.

Background and Clinical Evidence

Speech-generating devices (SGDs), also called augmentative and alternative communication devices, produce audible/verbal speech for individuals with severe expressive speech impairments (for example, anarthria, aphasia, apraxia, dysarthria). SGDs may use digitized (prerecorded whole-message output) or synthesized (text-to-speech) output and can be dedicated hardware or speech-generating software that enables a computer or PDA to function as an SGD. SGDs commonly include access devices (for example, optical head pointers, joysticks, scanning devices, head control mice) and mounting systems to enable use by persons with neurologic or motor limitations; assessment and device selection should be performed by a qualified speech-language pathologist.

References, Review Dates, and Administrative Notes

The policy references multiple clinical studies, reviews and guidance (including single-subject designs and small randomized or pilot studies) regarding use of SGDs across populations. However, the document does not specify a formal step therapy protocol. Decisions about sequencing of low- versus high-technology interventions and timing of device initiation (for acquired disorders the policy notes a typical delay of 4–6 months before initiating an SGD) should be informed by the cited literature and by individualized clinical assessment.

Administrative notes: this Clinical Policy Bulletin provides a summary of coverage criteria and referenced evidence to assist in benefit administration. It is not a contract and does not replace plan-specific provisions; prior authorization requirements and coverage determinations are made according to the member’s plan and program rules.