Comments and Responses Regarding Draft Local Coverage Determination:

Outpatient Physical and Occupational Therapy Services

_______________________________________________________________________

As an important part of Medicare Local Coverage Determination (LCD) development, National Government Services solicits comments from the provider community and from members of the public who may be affected by or interested in our LCDs. The purpose of the advice and comment process is to gain the expertise and experience of those commenting.

We would like to thank those who suggested changes to the draft Outpatient Physical and Occupational Therapy Services LCD. The official notice period for the final LCD begins on September 17, 2010, and the final determination will become effective on November 1, 2010.

________________________________________________________________________________

Comment:

The following comments were made by the president of the APTA:

Of particular concern, the draft LCD would require that the following list of V codes be used as the primary diagnosis.

• V57.1: Care Involving Other Physical Therapy

• V57.21: Care Involving Occupational Therapy

• V57.3: Care Involving Speech-Language Pathology

• V87.81: Care Involving Orthotic Training

• V57.89: Care Involving Other Specified Rehabilitation Procedure

We have serious concerns about requiring these V codes as the primary diagnosis. First, it appears that several of the V codes have been inadvertently misidentified in the LCD. For instance, V57.1 is for “care involving physical therapy,” not “not care involving other physical therapy.” Additionally, these codes do not provide sufficient information regarding the patient’s diagnosis to explain why the patient needs physical, occupational, or speech-language pathology services.

Another limitation of the V code as the primary diagnosis code is in the area of research. Researchers often use only the primary diagnosis code in their work. If the primary diagnosis code only indicates that physical or occupational therapy services were provided, that does not give researchers sufficient information to conduct important research in the areas of public health, quality 2 improvement, or patient outcomes. Finally, other mechanisms exist to help track what services are provided by whom. Specifically, whenever a physical therapist submits a claim for payment, he or she always places the GP modifier on the claim which indicates that the services were delivered as part of a physical therapy plan of care. As a result, a second mechanism, such as the V code, is not necessary to track utilization.

The LCD also requires that a secondary diagnosis code indicating the specific condition for which services are provided be included on the claim form. Failure to place a second diagnosis code on the claim appears to increase the chance that the claim will be denied.

Some billing software limits the number of diagnosis codes that can be placed on the claim form which could be problematic. Also, some billing clearinghouses split claims to facilitate the submission process.

If a claim is split and both the V code and secondary code are not kept on each claim, it could lead to unnecessary denials for the provider.

Most importantly, there are other diagnosis codes that more appropriately describe the patient’s condition which could be used by a medical reviewer to understand the rationale for the services delivered.

Response:

We believe that we have correctly stated the descriptors for the V57.1-V57.89 series of ICD-9 codes.  We will re-check to make sure that there are no typographical errors in the LCD and SIA related to this. 

The commenter stated that "...these codes do not provide sufficient information regarding the patient’s diagnosis to explain why the patient needs physical, occupational, or speech-language pathology services."  We agree.  However, that is specifically why the LCD requires the use of a secondary diagnosis code, (which he objects to further on in his comments).

The commenter also stated that the use of these codes would interfere with persons extracting claims data for research purposes.  We are not familiar with the methodology used by researchers for their studies.  We suggest these individuals familiarize themselves with the claims coding and adjust accordingly. Regardless, NGS is required to conform to correct coding standards under law (HIPAA).  While I am sympathetic to the needs of these researchers, these individuals must address these needs with CMS and/or Congress, who set these rules.  We find it curious that researchers considers only one diagnosis code out of many on a claim, and wonder whether this affects the accuracy of studies performed in such a manner.

As we stated at both the New York and Connecticut CAC meetings, we agree that the use of these ICD-9 codes is a potentially unnecessary redundancy in view of the mandatory use of the GO, GN and GP modifiers to identify the specific plan of care under which such services are being provided.  This is precisely why, during the discussion at the CAC, NGS requested that the Boston Regional Office discuss this diagnosis coding issue with Central Office.  The issue was brought to NGS's attention by a professional coder who accused NGS of fostering "incorrect coding" and requested this change, so that the LCD would be consistent with "correct coding."  We are seeking CMS relief from this coding requirement.

The commenter criticizes the requirement that a secondary diagnosis indicating the specific condition for which the services were provided be entered on the claim, despite his earlier statement that "...these codes do not provide sufficient information regarding the patient’s diagnosis to explain why the patient needs physical, occupational, or speech-language pathology services" (vide supra).   He is correct that failure to include a secondary diagnosis would likely result in the claim being denied. 

He also states that "Some billing software limits the number of diagnosis codes that can be placed on the claim form which could be problematic."  The Medicare Part B processing system (MCS) currently allows for at least four diagnoses to be included in the header if submitted on paper and eight if submitted electronically.  Thus the number of diagnoses that can be submitted on a single claim by a single group would appear to be a non-issue.   The fact that claims are split "to facilitate submission" as he suggests is of some concern.  However, if submitted this way, the claim lines are divided among different claims but the diagnosis codes should still be retained on each claim submitted.

The commenter continues by stating, "Most importantly, there are other diagnosis codes that more appropriately describe the patient’s condition which could be used by a medical reviewer to understand the rationale for the services delivered."  We agree with him, which is precisely why we require the secondary diagnoses, and allow for additional diagnoses to also be included.

*****

Comment:

A commenter stated that as a physical therapist and compliance officer for

PTPP, she wishes to disagree with the proposed draft for ICD-9 codes that support medical necessity. She believes that the V-codes are non specific and will not accurately portray the reason/s that a beneficiary would be receiving care.

Response:

See previous Comment and Response.

*****

Comment:

A representative of the CT PT Association commented that the coding requirement (use of the V57 diagnosis) is duplicative. She also noted that claims are sometimes denied because a secondary diagnosis is not on a claim. Does this mean that three diagnoses will now be required on some claims?

Response:

See previous Comment and Response.

*****

Comment:

A NY Physical Therapy Association representative addressed the ICD-9 issue and the use of the V codes. She asked for the language that the professional coder was referring to so that the national association can better address their rebuttal. They consider that the use of the GO, GN and GP modifiers clearly indicate who is delivering the service and to include the primary diagnosis by using the V codes would be more confusing and less accurate. She would like CMS to review and eliminate this requirement so there is more accurate coding of the actual disease or disability that the patient is being treated for. Dr. Deutsch indicated that he will send her the reference that the professional coder is quoting. (The reference can be found at http://www.cdc.gov/nchs/data/icd9/icdguide09.pdf, pages 1, 12, 74-76, and 98.)

Response:

See previous Comment and Response.

*****

Comment:

An officer of the APTA had these comments:

ICD-9 Codes that Support Medical Necessity

The HPA Section of the APTA urges NGS to reconsider to use of the “Primary Diagnoses” in the draft LCD to be the V codes as the reason for the encounter for therapy services. We do not believe reporting these codes provide useful information. The use of these codes as the primary diagnosis has the unintended consequence of complicating analyses of physical therapy utilization. While use of secondary and tertiary diagnoses can help describe the provided service, it does not allow for a definitive determination of the diagnosis that is the primary reason for treatment. Thus the data for physical therapy claims is contaminated by the use of the “V” codes as the primary diagnosis. Please consider either not using the “V” codes at all or making the “V” codes subordinate to the primary ICD-9 code that is the reason for treatment.

The Centers for Medicare and Medicaid (CMS) would also agree with HPA in that the primary diagnosis code on the claim form should be the code that best represents why the Medicare beneficiary requires physical therapy services as evidenced in CMS Publication 100-04, Medicare Claims Processing Manual, Chapter 5, Part B Outpatient Rehabilitation and CORF/OPT Services, Section 10.2C2b. On page 14 in this manual, CMS states the following: “Bill the most relevant diagnosis. As always, when billing for therapy services, the ICD-9 code that best relates to the reason for the treatment shall be on the claim, unless there is a compelling reason. For example, when a patient with diabetes is being treated for gait training due to amputation, the preferred diagnosis is abnormality of gait (which characterizes the treatment). Where it is possible in accordance with State and local laws and the contractors local coverage determinations, avoid using vague or general diagnoses”. HPA feels the use of V codes, as the primary diagnosis on the claim form is vague and a general diagnosis and does not best represent why the patient requires physical therapy services.

Here is the link to CMS Pub 100-04, Chapter 5. The quoted section can be found on page 14.

The use of V codes as the primary diagnosis will not assist CMS in gathering correct and accurate diagnoses as to why the patient required physical therapy services. In the CY 2006 Outpatient Therapy Services Utilization Report conducted by Computer Sciences Corporation and released on February 1, 2008, the report concluded, “In prior utilization reports, CSC included the generic V57 series ICD-9 codes (Care involving use of rehabilitation procedures). However, this series of codes, which do not describe a particular medical or functional condition, is commonly listed first in facility claims. In an effort to better identify the underlying condition creating the need for the outpatient therapy services, we varied the ICD-9 analytic approach to seek the first available claim ICD-9 code on the claim. In the event that there were no diagnosis codes other than the V57 series, our analysis defaulted to using the available V57 series code. The net result of this new approach was that the V57 diagnosis codes, which were the most common used principal claim diagnosis for PT and OT claims, and third most common for SLP claims in CY 2004, were rendered insignificant in this analysis.

Here is the link to that report and the diagnosis information can be found on page 21.

http://www.cms.gov/TherapyServices/downloads/2_CY2006OutpatientTherapyUtilizationReport_PDF_Final.pdf

As the profession moves towards the implementation of ICD-10 on October 1, 2013, those diagnosis codes are more specific to the patient’s functional deficits and the reason why they require physical therapy services. HPA feels that if the proposed language was finalized into the final LCD, that this would be a step backwards for the profession of physical therapy in accurately describing why a patient requires physical therapy services and a direct contradiction to ICD-10 and CMS.

In conclusion, HPA urges National Government Services to remove the proposed language listed under “ICD-9 Codes that Support Medical Necessity” in the draft LCD and to maintain that section as it currently appears in the active LCD that states:

ICD-9 Codes that Support Medical Necessity

It is the responsibility of the provider to code to the highest level specified in the ICD-9-CM (e.g., to the fourth or fifth digit). The correct use of an ICD-9-CM code does not assure coverage of a service. The service must be reasonable and necessary in the specific case and must meet the criteria specified in this determination.

See the “Indications and Limitations of Coverage and/or Medical Necessity” section.

XX000 Not Applicable

Response:

See previous Comment and Response.

NGS gratefully acknowledges the commenter’s assurance that CMS agrees with HPA since, as we have publicly indicated, we believe that the use of these encounter codes is redundant and provides no additional information that is not already identified on the claim by the concurrent use of the GN, GO and GP modifiers. We have requested CMS clarification and relief from this correct coding requirement (via the Boston Regional Office) and await follow-up.

*****

Comment:

A Senior Regulatory Analyst, Reimbursement and Regulatory Policy, from the AOTA submitted these comments:

Use of V Codes

The draft LCD includes the following language under “ICD-9 Codes that Support Medical Necessity”:

When coding for therapy services, the primary diagnosis codes should indicate the reason for the encounter and the specific condition for which therapy services are provided MUST also be included as secondary and subsequent diagnoses. Claims without secondary diagnoses may be denied.

Primary Diagnoses:

V57.1 CARE INVOLVING OTHER PHYSICAL THERAPY

V57.21 CARE INVOLVING OCCUPATIONAL THERAPY

V57.3 CARE INVOLVING SPEECH-LANGUAGE THERAPY

V57.81 CARE INVOLVING ORTHOTIC TRAINING

V57.89 CARE INVOLVING OTHER SPECIFIED REHABILITATION PROCEDURE

We believe this requirement is inappropriate, unnecessary, and counterproductive. The ICD-9-CM manual indicates that V codes may be used in any healthcare setting “to deal with occasions when circumstances other than a disease or injury classifiable to categories 001-999 are recorded as ‘diagnoses’ or ‘problems.’” However, the ICD-9-CM Manual does not require the use of V codes to identify rehabilitation services.

Furthermore, this policy change to require use of V codes creates an unnecessary burden on providers and does nothing to improve claims processing. Since the policy applies to facilities that submit CMS 1450 (UB-04) claim forms or electronic equivalent, either a Revenue Center (43X for occupational therapy, 42X for physical therapy, and 44X for speech therapy) or a discipline-specific therapy modifier (GO for occupational therapy, GP for physical therapy, and GN for speech therapy) are currently used to identify the billed services as occupational therapy, physical therapy, or speech therapy. These modifiers are more effective at identifying which procedures are attributed to which discipline, because the modifiers are assigned at the procedure claim line detail level rather than the diagnosis code level. Since ICD-9 codes are only assigned at the claim level for facility claims, it would be impossible to attribute a specific claim ICD-9 code to an individual line, when there is more than one Revenue Center identified on the claim. When there is more than one Revenue Center identified on a claim, especially if there is more than one therapy discipline on the claim, only one principal claim diagnosis is possible.

Therefore, a facility could not identify two different principal claim diagnoses on a claim if occupational therapy and physical therapy are on the same claim. This rule would create unnecessary confusion for both coders and claims reviewers, when all relevant information is already available in the Revenue Center and therapy discipline modifier data.

Moreover, requiring the use of V codes for therapy services in this manner will have negative consequences in regard to use of Medicare claims data for analyzing therapy utilization trends and other research projects. Commonly, only the principle diagnosis is used in aggregating claims data. If all therapy claims reflect only a general V code, these analyses become meaningless. The use of some V codes (i.e., V-15.88, History of Falls) would be useful as additional codes to track particular issues. But the use of V codes as contemplated in this draft LCD offers no benefit.

AOTA requests that NGS delete the language in the draft LCD which requires that V codes always be used as principle diagnoses for therapy services.

Response:

See previous Comment and Response.

*****

Comment:

Commenting on coverage of Low Frequency Ultrasound, a Connecticut therapist said, I strongly believe that reimbursement, especially in the outpatient setting, will provide patients with access to a valuable tool for wound healing. I have contacted Dr Haug, in a different jurisdiction, and they have recently approved coverage for physical therapists in an outpatient setting.

Anecdotally, I have seen many patients with recalcitrant wounds healed well using this modality, after multiple other therapies failed. I personally know of MANY patients out there who do not have access to this modality, as wound clinics, staffed by nurses and MDs, do not have the time, and are not set up to see, patients three times per week. In order for this modality to be effective, case series have shown that three times per week is the most effective frequency. Physical therapists in outpatient settings are accustomed to seeing their patients at this frequency, and are therefore more likely to use the modality appropriately, getting more positive results than at wound clinics, where the modality may have been tried once or twice, but not used consistently for an appropriate length of time to allow effective healing. Personally, were reimbursement allowed, it would not only allow me to offer this non-invasive, painless and effective modality to my patients, but it would also afford many research opportunities, to allow evidence based practice patterns, rather than anecdotal evidence.

In my clinical practice I work with patients with spinal cord injury who are living in the community. My colleague is a seating and positioning expert who has published research on this topic. Many of her patients have chronic wounds due to inappropriate seating and positioning systems in their wheelchairs. Once she corrects these issues, this modality in particular would be an ideal tool to offer these patients, who often are not willing to lie still long enough for treatment with electrical stimulation, another tool in a physical therapist's wound healing arsenal. The short duration and effectiveness of low-frequency ultrasound would allow me to offer them a way to speed the healing of their wounds, and reduce the "ticking time-bomb" feeling many of them have waiting for the next painful, immobilizing and expensive infection.

I could go on to describe many other outpatient applications for this valuable tool, but I want to be respectful of your time. I would like to thank you for taking the time to consider my opinion, and I truly hope you will allow reimbursement in the outpatient setting for this modality.

If you have any further questions, or you would like any clarification regarding my experience, please do not hesitate to contact me.

Response:

The issue of non-contact low frequency ultrasound (MIST therapy) was independently reviewed by one of the NGS Medical Directors (CC) for a repeat reconsideration of coverage related to the NGS CPT Category III LCD in April 2010. At that time coverage was denied due to insufficient adequately controlled and designed studies reported. The commenter does report anecdotal experience with this technique, but fails to provide scientific literature supporting the medical necessity of this technique in the Medicare population.

*****

Comment:

The CT Gerontology CAC representative asked if FES devices are covered when used to assist in walking.

Response:

FES devices may be covered when used to assist patients with spinal cord injuries in walking. It is not covered when used for the purpose of preventing/treating muscle atrophy (Medicare National Coverage Manual 100-3, Part II, Section 160.12 (Neuromuscular Electrical Stimulation)):

Use for Walking in Patients with Spinal Cord Injury (SCI)

The type of NMES that is use to enhance the ability to walk of SCI patients is commonly referred to as functional electrical stimulation (FES). These devices are surface units that use electrical impulses to activate paralyzed or weak muscles in precise sequence. Coverage for the use of NMES/FES is limited to SCI patients for walking, who have completed a training program which consists of at least 32 physical therapy sessions with the device over a period of three months. The trial period of physical therapy will enable the physician treating the patient for his or her spinal cord injury to properly evaluate the person’s ability to use these devices frequently and for the long term. Physical therapy necessary to perform this training must be directly performed by the physical therapist as part of a one-on-one training program.

The goal of physical therapy must be to train SCI patients on the use of NMES/FES devices to achieve walking, not to reverse or retard muscle atrophy.

*****

Comment:

The Connecticut PM&R CAC representative asked if training the patient in the use of the FES device is covered when it is used to assist in walking.

He also asked if there is any coverage of electrical stimulation for arms, by occupational therapists and noted that some devices are used to promote neuroplasticity.

Response:

FES is a covered service when used to assist patients with spinal cord injuries in walking, per NCD 160.12. See previous response.

Neuromuscular electrical stimulation (NMES) is a covered service when used to treat muscle atrophy in patients where nerve supply to the muscle is intact, including brain, spinal cord and peripheral nerves with non-neurological reasons for disuse atrophy (e.g., burns, prolonged immobilization, hip replacement surgery), per NCD 160.12:

Treatment of Muscle Atrophy

Coverage of NMES to treat muscle atrophy is limited to the treatment of disuse atrophy where nerve supply to the muscle is intact, including brain, spinal cord and peripheral nerves, and other non-neurological reasons for disuse atrophy. Some examples would be casting or splinting of a limb, contracture due to scarring of soft tissue as in burn lesions, and hip replacement surgery (until orthotic training begins).

Coverage of neuromuscular electrical stimulation is specifically excluded from coverage for purposes other than (1) treatment of muscle atrophy due to non-neurological causes and (2) for walking in spinal cord injury patients, per NCD 160.12. See also the discussion in the final decision memo in the Federal Register July 22, 2002 at: http://www.cms.gov/mcd/viewdecisionmemo.asp?from2=viewdecisionmemo.asp&id=55&. Therefore, its use for treatment of arms and upper extremities remains non-covered, as would any services related to such treatments or home use.

*****

Comment:

Additional Comments from the president of the APTA were:

The LCD also contains a provision related to functional electrical stimulation (FES). Specifically, in the section entitled “Miscellaneous Services (Non-Covered), it is stated:

“Functional Electrical Stimulating (FES) devices other than those that assist in walking are not covered under Medicare [NCD 160.12]. Consequently, any services related to the evaluation for or training of patients to use such a device is not covered. Such devices may include, but is not limited, to the Ergys® system.”

As stated, it appears that the NCD 160.12 limits the use of FES to assisting patients with walking. However, this NCD does allow for the use of FES for patients with muscle atrophy. We have provided this language below for your reference.

“Coverage of NMES to treat muscle atrophy is limited to the treatment of disuse atrophy where nerve supply to the muscle is intact, including brain, spinal cord and peripheral nerves, and other non-neurological reasons for disuse atrophy. Some examples would be casting or splinting of a limb, contracture due to scarring of soft tissue as in burn lesions, and hip replacement surgery (until orthotic training begins). (See §160.13 of the NCD Manual for an explanation of coverage of medically necessary supplies for the effective use of NMES.)”

For example, FES has been used for the treatment of individuals post stroke who are not able to use their upper extremity due to the effects of stroke. Research has shown that by using FES in conjunction with other rehabilitation interventions, such as therapeutic exercise, produced improved functional outcomes related to grasp, reach, and other essential upper extremity functions which further enabled increased patient independence in ADLs. (Neurorehabil Neural Repair. 2009 Jul-Aug;23(6):559-68. Epub 2009 Feb 3. Feasibility of iterative learning control mediated by functional electrical stimulation for reaching after stroke. Hughes AM, Freeman CT, Burridge JH, Chappell PH, Lewin PL, Rogers E.)

As a result, we encourage NGS to provide a more comprehensive description of when FES is covered and cite NCD 160.12. Devices, such as the Ergys system, can be used to care for patients with muscle atrophy and should be considered for coverage when justification is provided in the patient’s medical record.

Response:

We have again reviewed the NCD 160.12 regarding the coverage of FES. It clearly limits its use for walking in patients with spinal cord injury (SCI) to those patients meeting the specified criteria described in the NCD, including but not limited to an intact lower motor unite (L1 and below; both muscle and peripheral nerve); persons with muscle and joint stability for weight bearing at upper and lower extremities that can demonstrate balance and control to maintain an upright support posture independently; persons that demonstrate brisk muscle contraction to NMES and have sensory perception electrical stimulation sufficient for muscle contraction; persons that can transfer independently and can demonstrate independent standing tolerance for at least 3 minutes; and

persons with at least 6-month post recovery spinal cord injury and restorative surgery. It also states that “The goal of physical therapy must be to train SCI patients on the use of NMES/FES devices to achieve walking, not to reverse or retard muscle atrophy.”

The NCD does, as cited by the commenter, cover the use of NMES to treat muscle atrophy, BUT only in those cases of disuse atrophy related to non-neurological conditions, thus excluding patients following strokes.

Although literature cited by the commenter and submitted by others does reflect on-going research on the use of NMES/FES to improve functional outcomes in the treatment of upper extremity weakness or paresis (e.g., grasp, reach, etc), including following stroke or other neurological conditions, the National Coverage Determination 160.12 specifically precludes such coverage. NGS has re-reviewed this NCD with CMS Coverage and Analysis Group, to confirm the limitation of coverage. Contractors must conform to National Coverage Determinations, and may not override them or substitute Local Coverage Determinations.

We also note that this limitation extends to the evaluation and training for use of the Ergys® system as treatment of muscle atrophy (rather than assist walking), and are also noncovered by this NCD.

*****

Comment:

A NY Occupational Therapy Association representative indicated that the other change in the LCD was the non-coverage of self-administered ultrasound, and asked if the language for self administered ultrasound could be tightened up.

He also indicated that he had some concern with the language used for the functional electrical stimulation devices. He realizes this is part of the NCD and these devices would not be covered but they have some growing research experience with the use of devices other than the ERGYS for the upper extremity especially for distal muscle extension and flexion of the fingers. He indicated that they have had a lot of success with this. FES research articles were provided to the contractor.

Response:

See previous Comment and Responses (re: NMES/FES).

NGS appreciates these comments and the relevant literature submitted by the commenter for the contractor to review.

*****

Comment:

The NY Physical Therapy Association representative reiterated comments from the OT representative on the self administered ultrasound indicating that it can damage a person and should not be self administered. She indicated that ultrasound can cause subcutaneous burns and they would not like to have therapists training patients in the use of ultrasound.

Response:

See previous Comment and Response.

Comment:

The NY Pulmonary CAC representative asked if the time coverage had changed. He stated that a year or two ago there were some restrictions on how many therapies a provider could get in a year.

Response:

There were no changes of frequency parameters included in this LCD revision.

*****