Coverage Policy Manual
Policy #: 2015003
Category: DME
Initiated: January 2015
Last Review: February 2019
  Patient-actuated End Range Motion Stretching Devices

Loss of full ROM occurs in a significant proportion of patients following surgical procedures around the joint, such as total knee arthroplasty (TKA) or anterior cruciate ligament (ACL) reconstruction. The most common cause for severe postoperative motion loss is the development of intra-articular or extra-articular arthrofibrosis. Arthrofibrosis, characterized by periarticular fibrosis and bands of scar tissue, is described as a painful loss of end range motion compared with the normal contralateral side. Loss of ROM of the knee can lead to impairments in walking, sitting and rising from a chair, and navigating stairs. A 2010 publication estimated that given the annual rates of TKA and ACL reconstruction, the number of major knee surgery patients affected by arthrofibrosis in the United States would be at least 85,000 per year, and approximately 21,000 patients each year would be at risk of requiring additional surgery (Stephenson, 2010).
Treatment of arthrofibrosis may include physiotherapy, manipulation under anesthesia, arthroscopic or open lysis of adhesions, or revision surgery. Conservative treatment typically consists of postoperative physical therapy with pressure stretching techniques and home exercises. When traditional rehabilitation has failed, serial casting, static braces, or dynamic splints that provide low load prolonged stretch may be used. Dynamic splints use spring loading or elastic bands to provide low-intensity tension (less than that exerted by a physical therapist) and are designed to be worn over relatively long periods (i.e., 6-8 hrs, or overnight). Dynamic splinting devices include the Advance Dynamic ROM, DeROM, Dynasplint, EMPI advance, LMB Pro-glide, Pro-glide Dynamic ROM, SaeboFlex, SaeboReach, and Ultraflex.
This policy focuses on patient-actuated mechanical devices that provide either high-intensity stretch or static progressive stretch in the home. The efficacy of a stretching regimen to permanently remodel tissue is considered to be a function of the intensity, length of the session, number of times per day, and number of days per week that stretching is performed (Jacobs, 2011).  Devices that provide high-intensity stretching in the home are ERMI (ERMI Inc.) devices. Other devices, such as the JAS (Joint Active Systems, Inc.), provide a moderate-intensity force to hold a joint at its end range and gradually increase the stretch (static progressive stretch). The Static-Pro (DeRoyal) is another brace design that applies a static progressive stretch. In contrast to the long periods of low-intensity stretch provided by dynamic splinting devices, ERMI, JAS and Static-Pro devices are designed to be used for brief periods of 15 to 30 minutes, in up to 8 sessions per day.
Specific ERMI devices are the Shoulder Flexionater, Knee Flexionater, Knee Extensionater, Elbow Extensionater, and the MPJ Extensionater. These are intended primarily to address excessive scar tissue around the joint using progressive stretching alternating with periods of relaxation, with torque similar to that applied by physical therapists that is near or at the pain threshold. The patient uses a hydraulic pump to control the load, which can range from a few ounces to 500 lbs. For example, to use the ERMI Knee/Ankle Flexionater, patients pull a lever to increase knee flexion angle, as well as the amount of torque being applied to the joint. The hydraulic system amplifies the force of the lever into a greater torque applied to the knee for about 5 to 10 minutes. Periods of flexion are interspersed by 5- to 10-minute recovery intervals where the knee is released back into extension.
Joint Active Systems include the JAS Elbow, JAS Shoulder, JAS Ankle, JAS Knee, JAS Wrist, and JAS Pronation-Supination. Patients are instructed to use the JAS devices for 30 minutes, 3 times a day. During each 30-minute session, patients adjust their device by turning a ratchet to the maximum tolerated position of end range stretch. Each position is held for 5 minutes to allow for tissue relaxation to occur, and the device is then advanced to a new position of stretch (static progressive stretch). It is proposed that the JAS systems unload the joint to reduce joint surface pressures during the stretch. Other devices that provide static progressive stretch include Static-Pro Knee, Static-Pro Wrist, and Static-Pro Elbow. Static-Pro devices provide moderate torque by turning a knob and combine static stretching with stress relaxation.
Regulatory Status
The U.S. Food and Drug Administration (FDA) has determined that devices classified as “Exerciser, Non-Measuring” are class I devices and as such are exempt from premarket notification application and FDA clearance. FDA product code: ION
There are HCPCS codes for these types of devices:
E1801 Static progressive stretch elbow device, extension and/or flexion, with or without range of motion adjustment, includes all components and accessories
E1806 Static progressive stretch wrist device, flexion and/or extension, with or without range of motion adjustment, includes all components and accessories
E1811 Static progressive stretch knee device, extension and/or flexion, with or without range of motion adjustment, includes all components and accessories
E1816 Static progressive stretch ankle device, flexion and/or extension, with or without range of motion adjustment, includes all components and accessories
E1818 Static progressive stretch forearm pronation/supination device, with or without range of motion adjustment, includes all components and accessories

Effective, January 2015
Patient-actuated End Range Motion Stretching Devices, for treatment are mechanical devices that are used at home to increase range of motion (ROM), does not meet Primary Coverage Criteria that there be scientific evidence of effectiveness in improving health outcomes.
For members with contracts without Primary Coverage Criteria, Patient-actuated End Range Motion Stretching Devices, for treatment are mechanical devices that are used at home to increase range of motion (ROM) is considered investigational. Investigational services are exclusions in the member benefit certificate of coverage.

This policy was created in 2015 with a search of the MEDLINE database through December 3, 2014.
Assessment of efficacy for therapeutic interventions involves a determination of whether the intervention improves health outcomes. The optimal study design for a therapeutic intervention is a randomized controlled trial (RCT) that includes clinically relevant measures of health outcomes. Intermediate outcome measures, also known as surrogate outcome measures, may also be adequate if there is an established link between the intermediate outcome and true health outcomes. Nonrandomized comparative studies and uncontrolled studies can sometimes provide useful information on health outcomes but are prone to biases such as noncomparability of treatment groups, placebo effect, and variable natural history of the condition.
An improvement in functional outcomes, such as ability to perform ADLs, is the primary goal of this intervention. Joint ROM is an intermediate outcome, and there are some links between ROM and functional outcomes. For example, one small study correlated knee ROM with functional parameters and concluded that 110˚ is considered the functional ROM necessary to allow patients to perform common activities of daily living such as navigating stairs, rising up from a low chair or commode, entering or exiting from a car, or tying one’s shoes (Rowe, 2000). This threshold of ROM is therefore sometimes used as a measure of treatment success for individual patients. In addition, loss of knee end range motion of more than 15º, which occurs in about 1 to 2% of patients following anterior cruciate ligament reconstruction, has also been associated with loss of quadriceps muscle strength and the development of osteoarthritis (Shelbourne, 2009). According to the International Knee Documentation Committee (IKDC) criteria, an extension deficit of 6º to 10º or a flexion deficit of 16º to 25º when compared with the noninvolved knee is considered abnormal, while an extension deficit of more than 10º or a flexion deficit of more than 25º when compared with the noninvolved knee is considered severely abnormal (IKDC, 2000). ROM thresholds in joints other than the knee have been less clearly defined.
High-Intensity End Range Motion Devices  
In 2012, Papotto and Mills reported a small (n=20) RCT that compared high-intensity versus lower intensity stretch devices for home therapy in patients who had undergone total knee arthroplasty (TKA) (Papotto, 2012).  High-intensity stretch was performed with the End Range Motion Improvement (ERMI) knee/ankle flexionater. Patients were asked to stretch at an intensity that mimicked the intensity of the stretch being provided by their physical therapists during outpatient therapy sessions and to use the device in 20- to 30-minute sessions with a total of 60 minutes of end range stretching per day. The lower intensity stretch group used the StaticPro Knee, which consists of a brace that is secured to the upper and lower leg with cuffs and straps. The patient uses a hand crank to increase the torque on the knee, which results in lower applied force to the knee than the hydraulic system. Patients were asked to use the StaticPro Knee in 3 sessions of 30 minutes each day, increasing the force applied to the joint every 5 minutes. After an average of 7 weeks of treatment, patients treated with ERMI gained 29.9˚ of motion compared with 17.0˚ following treatment with the lower intensity stretching device (p<0.001). Knee flexion of 110˚ or more was obtained in 91% of the ERMI group compared with 22% of the StaticPro group (p<0.001). Improvement in the 100-point Western Ontario and McMaster (WOMAC) University Osteoarthritis Index that measures stiffness, pain, and physical function, was significantly greater in the ERMI group (25.6 vs 12.4, p=0.048).
There was a moderate correlation between knee ROM and post-treatment WOMAC scores (r =0.53, p=0.02).
Non-randomized Comparative Studies
Stephenson et al. (2010) reported an industry (ERMI)-funded retrospective comparative study of high- versus low-intensity stretch devices based on claims data in 60,359 patients who had a diagnosis of arthrofibrosis following knee injury or surgery (Stephenson, 2010).  There were 143 patients who used a high-intensity stretch device, 607 who used a low-intensity stretch device, and 59,609 who did not use any stretching device. To make the groups more comparable in terms of severity, the lower intensity stretch and no device patients were required to have a diagnosis relating to osteoarthrosis, ankyloses, contracture/fracture, or stiffness in the lower leg. After controlling for baseline differences in the type of knee surgery and musculoskeletal disease, the high-intensity stretch group had significantly lower rates of re-hospitalization than low-intensity stretch and no device patients. Significantly more patients with no device (47.4%) had a subsequent knee event within 6 months from the index surgery than either high-intensity (24.5%) or low-intensity (22.2%) stretch patients.
Uncontrolled trials
A frequently cited study from Branch (Medical Director of ERMI, Inc) et al. was reported in 2003 (Branch, 2003).  Patients (n=34) in this prospective series who did not have full knee ROM after 6 weeks of physical therapy were prescribed the ERMI Knee/Ankle Flexionater. The 2 patients in the study who had ROM greater than 115˚ at the start of therapy regained full ROM. Of the 6 patients with between 90˚ and 115˚ ROM at the start of therapy, 83% gained full ROM, and of the 16 patients with between 60˚ and 90˚ ROM at the start of therapy, 81% regained full ROM. For the 10 patients who began mechanical therapy with 0˚ to 60˚ ROM, only 40% regained full ROM, but this group regained the most ROM (mean, 79˚) of the 4 groups. With functional ROM defined as 115˚ or more, 31 of the 34 patients met this goal, and the improvement in ROM for the entire group was highly significant. A retrospective review from this group found that passive knee extension deficits that had reached a plateau with physical therapy decreased from 10.5˚ to 2.0˚ with the ERMI Knee Extensionater (Dempsey, 2010).
Uncontrolled trials
One industry-funded retrospective series with 36 patients was identified on the ERMI Shoulder Flexionater (Dempsey, 2011).  Patients with frozen shoulder who had failed 6 weeks of physical therapy (glenohumeral abduction and external rotation not equal to the opposite uninvolved limb) were treated with the Flexionater in combination with continued physical therapy. Patients were instructed to perform six 10-minute bouts of end-range stretching per day at home, using an intensity that was uncomfortable but not painful. Blinded evaluation at the end of treatment found that the ROM of the involved limb equaled that of the opposite limb. Scores on the American Shoulder and Elbow Society Standardized Shoulder Assessment Form (ASES) showed significant improvement (p<0.05), and patients with greater pain at baseline had the greatest improvement in ASES scores (gain of 50 points of 100 total).
Static Progressive Stretch Devices
Several reports on JAS devices have been published by a group of investigators that include Bonutti (stockholder in Joint Active Systems), McGrath, Ulrich, and Mont.
The small RCT by Popotto and Mills (2012) previously discussed, compared low-intensity stretch with a progressive stretch device (StaticPro Knee) with high-intensity stretch with an ERMI device and reported greater improvement in ROM for the group using the ERMI device.
Uncontrolled trials
A 2008 publication by Bonutti et al. (2008) reported on a series of 41 patients who used the JAS device after failing conventional physical therapy (Bonutti, 2008).   The patients in this study had a total ROM of less than 90º or a flexion contracture that impaired quality of life. Twenty-five of the patients had previously undergone manipulation under anesthesia. After a mean of 9 weeks of use (range, 3-27 weeks), ROM increased by a mean of 33º (range, 0º to 85º), with mean final extension of -6º and flexion of 108º. Outcomes were found to be comparable with other nonoperative treatments reported in the literature, but the improvements in ROM occurred in a shorter treatment time.
Uncontrolled trials
 Ulrich et al. (2010) reported use of the JAS Elbow device in 37 patients (Ulrich, 2010).   Patients with deficits in flexion or extension had undergone a minimum of 6 weeks of exercise with a minimum of 2 weeks of minimal motion gain (<5º). After a mean treatment time of 10 weeks (range, 2-23 weeks), the ROM increased by a mean of 26º (range, 2-60) to a final ROM of 107º (range, 70-140). Results were compared with those of other upper extremity stretch devices (e.g., splints), which achieved similar success rates (81%-88%) with 6 to 10 hours of daily wear over 6 to 10 months.
Forearm Rotation
Uncontrolled trials
In 2009, McGrath et al reported on a series of 38 consecutive patients with limitations of forearm rotation that had plateaued with standard physical therapy (McGrath, 2009).  Treatment with a JAS pronation/supination device was started at an average of 21 weeks (range, 6-75) after the upper extremity injury. At the start of treatment, mean ROM was 96º (range, 20-150). After an average of 12 weeks of treatment (range, 357), mean ROM increased to 138º (range, 70-180).
Uncontrolled trials
McGrath et al. (2008) reported use of the JAS Wrist device in 47 consecutive patients with post-traumatic or post-surgical wrist stiffness (McGrath, 2013).   All patients had experienced a plateau in ROM (67º, range 18-114) after a mean of 12 weeks of physical therapy (range, 6-28) and were not expected to improve with standard therapeutic modalities. After a mean of 10 weeks of treatment (range, 4-26), ROM increased to 101º (range, 60-156).
In 2008, Lucado et al. reported a retrospective review of 25 patients with distal radius fractures who had been treated with a JAS Flexion/Extension device or JAS forearm Pronation/Supination device at their institutions (Lucado, 2008).   The mean time from injury to the initiation of static progressive stretch was 94 days (range, 48-188), and the duration of use was 75 days (range, 14-160). There were significant improvements in ROM and in scores on the Disability of the Arm, Shoulder, and Hand (DASH). The median DASH score improved from 43 to 19/100 after JAS therapy (p>0.000).
Ongoing and Unpublished Clinical Trials
A search of online site in January 2015 identified the following trials:
        • NCT01618227 A Prospective Randomized Trial of Rehabilitation With or Without Static Progressive Splinting for Wrist Stiffness will compare the JAS device versus rehabilitation without splinting. The study has an estimated enrollment of 60 patients with completion expected January 2015.
        • NCT02105857 The Effects of End-of-range Grade A+ Mobilization Following Acute Primary TKA is a randomized trial from Europe that evaluates use of the JAS Knee device after knee replacement compared with standard rehabilitation. The study has an enrollment of 32 patients. It is listed as completed as of April 2014.
Summary of evidence indicates there is a small body of evidence on patient-actuated end range motion stretching devices. The best evidence consists of end range motion improvement (ERMI) devices to treat the knee. One small randomized controlled trial and one larger retrospective comparative study report that high-intensity  stretching with ERMI devices improves range of motion (ROM) more than lower intensity stretching devices in patients who are post-knee injury or surgery. Other available data consists of retrospective case series that demonstrate improvement in end range motion in patients whose ROM has plateaued with conventional physical therapy. The clinical significance of gains in this surrogate outcome measure is unclear. Further high-quality comparative trials are needed to determine whether these patient-actuated devices improve functional outcomes compared with alternative treatments and to better define the patient population that might benefit.
Practice Guidelines and Position Statements
No guidelines or statements on patient-actuated ERMI devices were identified.
2017 Update
A literature search conducted through January 2017 did not reveal any new information that would prompt a change in the coverage statement. The key identified literature is summarized below.
Randomized Controlled Trials
In 2012, Ibrahim and colleagues published an evaluator-blinded RCT of 60 patients with shoulder adhesive capsulitis randomized to 4 weeks with a JAS device in conjunction with physical therapy (including home exercises) or physical therapy alone (Ibrahim, 2012). The study was independently funded, although the devices were provided by Joint Active Systems. Patients were evaluated for ROM, functional outcomes with the Disabilities of the Arm Should and Hand (DASH) questionnaire, and a visual analog scale (VAS) for pain. The increase in ROM was significantly greater with SPS (by about 30°) for active and passive abduction and external rotation (p<0.001 for all measures). DASH scores decreased by 68 points compared to 58 in the control group (p<0.001). VAS pain scores improved from 4 to 1 (73% improvement) in the SPS group and from 5 to 1 in the control group (p=0.05), with a possible floor effect limiting detection of differences. Over 24 weeks of follow-up, these improvements were maintained in the JAS group, but declined in the group treated with physical therapy alone for 4 weeks. A 2014 study by Ibrahim and colleagues appears to report the same data, with additional follow-up at 1 year, and 2-year results added in a 2015 (Ibrahim, 2014; Hussein, 2015). At 1- and 2-year follow-ups, ROM was substantially higher in the JAS group (p<0.001), due primarily to an increase in DASH scores for the group treated with physical therapy over 2 years. The difference between groups in ROM at 2 years was 47.6° for passive external rotation, 29.1° for passive abductions, and 75.3° for active abduction. Final DASH scores were 2.5 in the JAS group compared to 36.2 in the control group (p<0.001). VAS scores for pain were low in both the SPS (1.2) and control (1.7; p=NS) groups. The study suggests that there was no loss to follow-up over 2 years, and no references to the prior publications were noted, raising questions about reporting. It is also unclear why functional limitations would increase in the control group, when adhesive capsulitis is generally a self-limiting condition.
Systematic Reviews
A 2013 systematic review by Muller and colleagues compared the effectiveness of dynamic, static, or static progressive stretching in patients with posttraumatic or postoperative elbow stiffness (Muller, 2013). They included 13 case reports and case series with a total of 247 patients (range, 1-37 patients). Mean time from the incident to the start of treatment was 6.9 months. The greatest increase in ROM was obtained with dynamic splints (46°), followed by SPS devices (40°) and static splints (34°). These differences were statistically significant (p<0.001), but may not be clinically significant. None of the included studies assessed patient compliance, which are potentially affected by the duration of wear and comfort of the device. This systematic review is limited by the inclusion of low-quality studies, including case reports.
Randomized Controlled Trials
Lindenhovious and colleagues reported an RCT that compared SPS with a JAS device to dynamic splinting in 66 patients with posttraumatic elbow stiffness (Lindenhovious, 2012). Patients included had a loss of more than 30º in flexion or extension after an elbow injury or surgery and had failed to improve for at least 4 weeks with regular stretching exercises. Use of splints was discontinued at the discretion of the patients or if progress in ROM had plateaued. Evaluation was conducted by an investigator not involved in the care of the patients, but did not appear to have been blinded. Ten percent of patients in the dynamic splinting cohort asked for a change in treatment due to discomfort with the dynamic splint. Follow-up at 12 months was available for 80% of patients in the SPS group and 68% of patients in the dynamic splinting group, potentially reflecting lower patient satisfaction with dynamic splinting. Mean change in elbow function, measured with the DASH questionnaire, differed significantly between the SPS (eg, 25 points) and dynamic splinting (32 points; p<0.05) groups at 6 months but not 12 months. Statistical analysis was intention-to-treat, but did not account for repeated measures or baseline covariates. ROM was similar in the 2 groups.
2018 Update
Annual policy review completed with a literature search using the MEDLINE database through January 2018. No new literature was identified that would prompt a change in the coverage statement.   
2019 Update
Annual policy review completed with a literature search using the MEDLINE database through January 2019. No new literature was identified that would prompt a change in the coverage statement.

E1801Static progressive stretch elbow device, extension and/or flexion, with or without range of motion adjustment, includes all components and accessories
E1806Static progressive stretch wrist device, flexion and/or extension, with or without range of motion adjustment, includes all components and accessories
E1811Static progressive stretch knee device, extension and/or flexion, with or without range of motion adjustment, includes all components and accessories
E1816Static progressive stretch ankle device, flexion and/or extension, with or without range of motion adjustment, includes all components and accessories
E1818Static progressive stretch forearm pronation/supination device, with or without range of motion adjustment, includes all components and accessories
E1831Static progressive stretch toe device, extension and/or flexion, with or without range of motion adjustment, includes all components and accessories
E1841Static progressive stretch shoulder device, with or without range of motion adjustment, includes all components and accessories

References: Bonutti PM, McGrath MS, Ulrich SD, et al.(2008) Static progressive stretch for the treatment of knee stiffness. Knee. Aug 2008;15(4):272-276. PMID 18538574

Branch TP, Karsch RE, Mills TJ, et al.(2003) Mechanical therapy for loss of knee flexion. Am J Orthop (Belle Mead NJ). Apr 2003;32(4):195-200. PMID 12723771

Dempsey AL, Branch TP, Mills T, et al.(2010) High-intensity mechanical therapy for loss of knee extension for worker's compensation and non-compensation patients. Sports Med Arthrosc Rehabil Ther Technol. 2010;2:26. PMID 20939921

Dempsey AL, Mills T, Karsch RM, et al.(2011) Maximizing total end range time is safe and effective for the conservative treatment of frozen shoulder patients. Am J Phys Med Rehabil. Sep 2011;90(9):738-745. PMID 21430510

Hussein AZ, Ibrahim MI, Hellman MA, et al.(2015) Static progressive stretch is effective in treating shoulder adhesive capsulitis: Prospective, randomized, controlled study with a two-year follow-up. Eur J Physiother. 2015;17(3):138-147.

Ibrahim M, Donatelli R, Hellman M, et al.(2014) Efficacy of a static progressive stretch device as an adjunct to physical therapy in treating adhesive capsulitis of the shoulder: a prospective, randomised study. Physiotherapy. Sep 2014;100(3):228-234. PMID 24211154

Ibrahim MI, Johnson AJ, Pivec R, et al.(2012) Treatment of adhesive capsulitis of the shoulder with a static progressive stretch device: a prospective, randomized study. J Long Term Eff Med Implants. 2012;22(4):281-291. PMID 23662659

International Knee Documentation Committee.(2015) International Knee Documentation Committee criteria. 2000; Accessed January 2, 2015.

Jacobs CA, Sciascia AD.(2011) Factors that influence the efficacy of stretching programs for patients with hypomobility. Sports Health. Nov 2011;3(6):520-523. PMID 23016052

Lindenhovius AL, Doornberg JN, Brouwer KM, et al.(2012) A prospective randomized controlled trial of dynamic versus static progressive elbow splinting for posttraumatic elbow stiffness. J Bone Joint Surg Am. Apr 18 2012;94(8):694-700. PMID 22517385

McGrath MS, Ulrich SD, Bonutti PM, et al.(2008) Evaluation of static progressive stretch for the treatment of wrist stiffness. J Hand Surg Am. Nov 2008;33(9):1498-1504. PMID 18984330

McGrath MS, Ulrich SD, Bonutti PM, et al.(2009) Static progressive splinting for restoration of rotational motion of the forearm. J Hand Ther. Jan-Mar 2009;22(1):3-8; quiz 9. PMID 18950990

Muller AM, Sadoghi P, Lucas R, et al.(2013) Effectiveness of bracing in the treatment of nonosseous restriction of elbow mobility: a systematic review and meta-analysis of 13 studies. J Shoulder Elbow Surg. Aug 2013;22(8):1146-1152. PMID 23796383

Papotto BA, Mills T.(2012) Treatment of severe flexion deficits following total knee arthroplasty: a randomized clinical trial. Orthop Nurs. Jan-Feb 2012;31(1):29-34. PMID 22278649

Rowe PJ, Myles CM, Walker C, et al.(2000) knee joint kinematics in gait and other functional activities measured using flexible electrogoniometry: how much knee motion is sufficient for normal daily life? Oct 2000;12(2):143-155. PMID 10998612

Shelbourne KD, Gray T.(2009) Minimum 10-year results after anterior cruciate ligament reconstruction: how the loss of normal knee motion compounds other factors related to the development of osteoarthritis after surgery. Am J Sports Med. Mar 2009;37(3):471-480. PMID 19059893

Stephenson JJ, Quimbo RA, Gu T.(2010) Knee-attributable medical costs and risk of re-surgery among patients utilizing non-surgical treatment options for knee arthrofibrosis in a managed care population. Curr Med Res Opin. May 2010;26(5):1109-1118. PMID 20225995

Ulrich SD, Bonutti PM, Seyler TM, et al.(2010) Restoring range of motion via stress relaxation and static progressive stretch in posttraumatic elbow contractures. J Shoulder Elbow Surg. Mar 2010;19(2):196-201. PMID 19959379

Group specific policy will supersede this policy when applicable. This policy does not apply to the Wal-Mart Associates Group Health Plan participants or to the Tyson Group Health Plan participants.
CPT Codes Copyright © 2019 American Medical Association.