Postsurgical Outpatient Use of Limb Compression Devices for Venous Thromboembolism Prophylaxis
Outpatient use of limb compression devices is considered not medically necessary for venous thromboembolism (VTE) prophylaxis for the following:
Outpatient use of limb compression devices to prevent VTE is considered not medically necessary for periods longer than 30 days following ANY surgery.
Current standard of care to prevent venous thromboembolism (VTE) after major orthopedic and some non-orthopedic surgeries includes the use of limb compression devices and pharmacological prophylaxis while the patient is hospitalized. After discharge, most patients are ambulatory, and ongoing use of limb compression devices has not been demonstrated to improve outcomes.
Medically necessary indications after Major Orthopedic surgery:
Patient had total hip arthroplasty total knee arthroplasty, or hip fracture surgery, AND one of the following:
Medically necessary indications after Major Non-orthopedic surgery:
One of the following:
Summarization of AACP guidelines on determining high risk for venous thromboembolism (VTE) (3)(5):
Summarization of ACOG guidelines to determine high risk for VTE in gynecologic surgery (2):
Guidance on Duration of Use
In patients with contraindications to pharmacologic prophylaxis who are undergoing major orthopedic surgery total hip arthroplasty (THA), total knee arthroplasty (TKA) or hip fracture surgery (HFS), the American College of Chest Physicians (ACCP) guidelines are consistent with use of intermittent pneumatic compression devices for 10-14 days after surgery. (3) The ACCP suggestion on extended prophylaxis (up to 35 days) was a weak recommendation that did not mention pneumatic compression devices as an option.
In the ACCP guideline on VTE prophylaxis in patients undergoing non-orthopedic surgery, the length of standard duration or “limited duration” prophylaxis was not defined. However, “extended duration” pharmacologic prophylaxis was defined as 4 weeks; this was recommended only for patients at high risk for VTE undergoing abdominal or pelvic surgery for cancer and not otherwise at high risk for major bleeding complications.
Definition of Terms
Major Orthopedic Surgery for this policy is a total hip arthroplasty (THA/THR), total knee arthroplasty (TKA/TKR), or hip fracture surgery (HFS).
Major Non-Orthopedic Surgery includes open abdominal or open pelvic surgery, especially for a cancer diagnosis, or surgery is considered major due to the patient’s age is greater than 60, or anesthesia time is 2 hours or more.
Nonmajor Orthopedic Surgery includes but is not limited to arthroscopic and fracture repair procedures.
Increased risk of bleeding (increasing the risk of VTE) includes a history of prior excessive bleeding, severe renal failure, advanced liver disease, on anti-platelet therapy (e.g. NSAID, ASA, Plavix, Ticlid), extensive surgical dissection, and revision surgery, disorders of coagulation (e.g. hemophilia, Von Willebrands, ITP and others). (For ACCP guidelines and AAO guidelines see Rationale section, Practice Guidelines and Position Statements)
Venous thromboembolism (VTE) is the combination of a deep vein thrombosis (DVT) and pulmonary embolism (PE).
Patients undergoing major orthopedic surgery are at increased risk for venous thromboembolism (VTE). Patients undergoing other types of surgery may also be at increased risk of VTE. Limb compression devices are one option for thromboprophylaxis and are commonly used in the hospital setting. Outpatient use of compression devices following hospitalization, with or without pharmacologic prophylaxis, has also been proposed.
Patients undergoing major surgery are at increased risk of developing deep vein thrombosis (DVT) and pulmonary embolism (PE), together known as venous thromboembolism (VTE). Patients who are having major orthopedic surgery (defined here as total hip arthroplasty [THA], total knee arthroplasty [TKA] and hip fracture surgery [HFS]) are at particularly high risk. Risk of DVT is increased due to venous stasis of the lower limbs as a consequence of immobility during and after surgery. In addition, direct venous wall damage associated with the surgical procedure may occur. DVTs are frequently asymptomatic and generally resolve when mobility is restored. However, some episodes of acute DVT can be associated with substantial morbidity and mortality. The most serious adverse consequence of an acute DVT is a PE, which can be fatal; this occurs when the DVT detaches and migrates to the lungs. In addition, DVT may produce long-term vascular damage that leads to chronic venous insufficiency. Without thromboprophylaxis, the incidence of venographically detected DVT is approximately 42-57% after total hip replacement, and the risk of pulmonary embolism is approximately 1-28%. (1) Other surgical patients may also be at increased risk of VTE during and after hospitalization. For example, it is estimated that rates of VTE without prophylaxis after gynecologic surgery is about 15-40%. (2)
Thus, antithrombotic prophylaxis is recommended for patients undergoing major orthopedic surgery and other surgical patients at increased risk of VTE. For patients undergoing major orthopedic surgery, clinical practice guidelines published in 2012 by the American College of Chest Physicians (ACCP) recommend that one of several pharmacologic agents or mechanical prophylaxis be provided rather than no thromboprophylaxis. (3) The guidelines further recommend the use of pharmacologic prophylaxis during hospitalization, whether or not patients are using a limb compression device. A minimum of 10-14 days of prophylaxis is recommended, a portion of which can be after discharge during outpatient use.
The ACCP guidelines noted that compliance as a major issue with pneumatic compression devices used for thromboprophylaxis and recommend that, if this prophylactic option is selected, use should be limited to portable, battery-operated devices. Moreover, it is recommended that devices be used for 18 hours per day. A 2009 non-randomized study found that there was better compliance with a portable battery-operated pneumatic compression device compared to a non-mobile device when used by patients in the hospital following hip or knee replacement surgery. (4)
The ACCP also issued guidelines on VTE prophylaxis in non-orthopedic surgery patients. (5) For patients undergoing general or abdominal-pelvic surgery who have a risk of VTE of 3% or higher, the ACCP recommends prophylaxis with pharmacologic agents or intermittent pneumatic compression rather than no prophylaxis. For patients at low risk for VTE (about 1.5%), the guidelines suggest mechanical prophylaxis. Unlike the guidelines on major orthopedic surgery, which recommend a minimum of 10-14 days of VTE prophylaxis, the guidelines on non-orthopedic surgery patients do not include a general timeframe for prophylaxis. They do, however, define “extended duration” pharmacologic prophylaxis as lasting 4 weeks; the latter is recommended only for patients at high risk for VTE, undergoing abdominal or pelvic surgery for cancer who are not otherwise at high risk for major bleeding complications.
National clinical guidelines have not specifically recommended use of limb compression devices in the outpatient setting. However, with the availability of portable, battery-operated devices, there is interest in use of outpatient limb compression devices for deep vein thrombosis (DVT) following discharge from the hospital for major orthopedic and non-orthopedic surgery.
Various pneumatic and peristaltic limb compression devices, with indications including prevention of DVT, have been cleared for marketing by the U.S. Food and Drug Administration (FDA) through the 510(k) process. Portable devices that have been cleared by the FDA include:
Venowave™ VW5 (Venowave Inc.; Stouffville, Ontario, Canada): The device is a peristaltic pump that is strapped to the leg below the knee. It is powered using a single NiMH (nickel metal hydride) AA size battery. Different models are available that have a different maximum number of wave plate cycles per minute.
ActiveCare+SFT® System (Medical Compression Systems LTD, Or Akiva, Israel): The device applies sequential pneumatic compression to the lower limb; it has the option of being battery-operated. Foot compression is achieved with use of a single-celled foot sleeve. Calf and thigh compression requires use of a 3-celled cuff sleeve.
Restep® DVT System (Stortford Medical LLC, West Windsor, NJ): This is a lightweight device that utilizes single chamber pressure cuffs attached to the patient’s lower legs.
Kendall SCD™ 700 Sequential Compression System (Covidien, Mansfield, MA): This pneumatic compression device can be used in the clinic or at-home. It has a two-pronged plug and is not battery-operated.
Medical policies are systematically developed guidelines that serve as a resource for Company staff when determining coverage for specific medical procedures, drugs or devices. Coverage for medical services is subject to the limits and conditions of the member benefit plan. Members and their providers should consult the member benefit booklet or contact a customer service representative to determine whether there are any benefit limitations applicable to this service or supply. This policy does not apply to Medicare Advantage.
Venous Thromboembolism (VTE) Prophylaxis in Major Orthopedic Surgery Patients
Patients without a contraindication to prophylaxis with pharmaceutical agents
Anticoagulation is the mainstay of deep vein thrombosis (DVT) prophylaxis after major surgery and is continued into the outpatient setting. Treatment with pneumatic compression devices may offer addition benefit when used in conjunction with anticoagulation in the inpatient setting but is not commonly used in the outpatient setting. The ideal study design to evaluate whether there is benefit in the outpatient setting would be a randomized controlled trial (RCT) comparing outpatient anticoagulation alone to anticoagulation plus pneumatic compression devices. Key health outcomes include incidence of DVT and pulmonary embolism (PE), as well as measures of functional status and/or quality of life associated with these outcomes.
Randomized controlled trials (RCT)
No RCTs with the above design were identified.
In 2012, Kakkos and colleagues published a meta-analysis of RCTs evaluating combined use of anticoagulation and mechanical DVT prophylaxis following joint replacement surgery; however, the study focused on inpatient thromboprophylaxis. (7) The authors identified 4 trials that compared anticoagulation alone to anticoagulation plus use of pneumatic compression devices. Three of the 4 studies used pneumatic compression devices only until discharge from the hospital. In the fourth study, it was not clearly stated that pump use was limited to the inpatient setting, but inpatient use was implied (e.g., the article stated that staff checked several times a day to ensure correct use of the pump system). Meta-analyses found statistically significantly lower incidences of DVT in the group that used compression pumps in addition to anticoagulation compared to anticoagulation-only. In a pooled analysis of 4 trials on hip replacement, the incidence of DVT was 9.7% in the anticoagulation-only group and 0.9% in the combined treatment group (risk ratio [RR]: 0.17; 95% confidence interval [CI]: 0.06 to 0.46). Similarly, when findings from 2 trials on knee replacement were pooled, the incidence of DVT was 18.7% in the anticoagulation-only group and 3.7% in the combined treatment group (RR: 0.27; 95% CI: 0.08 to 0.89).
A 2013 systematic review and meta-analysis by Sobieraj and colleagues was similar to the Kakkos et al. study, described above. It reviewed studies comparing combined pharmacologic and mechanical prophylaxis to either method alone in patients undergoing major orthopedic surgery. (8) Six RCTs were identified and these studies all focused on inpatient treatment. Most studies evaluating outcomes in the post-operative period. In 1 study that followed patients for 90 days, intermittent pneumatic compression continued only until hospital discharge.
There several reasons why the benefit of pneumatic compression devices in the hospital setting may not extrapolate to benefit in the outpatient setting. First, the level of mobility is necessarily less in the hospital than in the outpatient setting, indicating a different risk for DVT. Also, the use of pneumatic compression devices in the hospital can be more highly controlled and monitored. In the outpatient setting, questions about the degree of compliance with the devices, including the ability to correctly use them in the absence of professional supervision are considerations. No comparative studies were identified on compliance with pneumatic compression devices in the outpatient setting.
A 2006 case series by Giannoni and colleagues in Italy included both inpatient and outpatient DVT prophylaxis with pneumatic compression devices and anticoagulants. (9) The study included 34 patients who underwent total knee replacement (4 patients had bilateral replacements). All patients used a pneumatic compression device (A-V Impulse foot pump system) for 15 days. The mean hospital stay was 7 days, and the range was 5 to 12 days. The compression devices were worn for an average of 14 hours per day (range 8 to 18 hours). Patients were also treated with LMWH, beginning after surgery and continuing until the operated leg was completely weight bearing (15-30 days). Ultrasonography detected DVTs in 3 of 34 (8.8%) patients; all were distal DVTs. One symptomatic DVT developed on the fourth post-operative day, and there were 2 subclinical DVTs detected at the routine 1 month ultrasonography examination. This study did not include a comparison group of patients who did not use a pneumatic compression device. In addition, given the range of length of hospital stay, some patients received their entire course of prophylactic treatment as inpatients. Compliance with pneumatic compression devices was not reported.
Patients with a contraindication to prophylaxis with pharmaceutical agents
Patients with contraindications to anticoagulants need to be treated with non-pharmacologic measures. The ideal study design for this question would be an RCT comparing prophylaxis with pneumatic compression devices alone in the outpatient setting to no prophylaxis or to alternative methods of prophylaxis.
Randomized controlled trials
No RCTs with the above design were identified. However, one recent RCT provided data that might be useful for answering the question of whether outpatient use of pneumatic compression devices are beneficial in the absence of outpatient anticoagulant use.
The study, reported in 2 publications, one in 2010 and the other in 2011, was conducted at multiple centers in the United States and included 395 patients undergoing total hip replacement. (10, 11) Individuals with a previous history of thrombosis, known coagulation disorder, solid malignant tumor, peptic ulcer disease or mental disorder were excluded. Patients were randomized to 10 days of DVT prophylaxis using either low-molecular-weight heparin (LMWH) or a mobile pneumatic compression device (ActiveCare+SFT). Treatment continued for 10 days after surgery in both groups; patients received a variable portion of their treatment after hospital discharge. Patients in the compression device group could also receive aspirin if recommended by their doctor. Patients were examined with bilateral duplex ultrasound on days 10-12 following surgery. The mean length of hospital stay was 3.2 days in both groups. Length of hospital stay ranged from 2 days to 10 days; thus, patients had between 0 days and 8 days of outpatient use of their assigned method of prophylaxis. According to ultrasound findings, 8 of 196 (4.1%) in the pneumatic compression group and 8 of 190 (4.2%) in the LMWH group had a DVT. In addition, 2 pulmonary emboli were detected in each group. The incidence of venous thromboembolic events did not differ significantly between groups. However, the rate of major bleeding was significantly higher in the LMWH group. A total of 11 (6%) of patients in the LMWH group had a major bleeding event compared to no patients in the pneumatic compression group (p=0.0004). Rates of minor bleeding were similar in the 2 groups; 78 (40%) in the LMWH group and 74 (37%) in the pneumatic compression group. In addition, compliance with the mobile compression devices was monitored using internal timers in the device. According to these data, patients used the device for a mean of 11 days (range 1 to 15 days) and for a mean of 20 hours per day. Mean use of the device was 83% of possible usable time. Findings on compliance were not reported separately for inpatient and outpatient use of the devices.
There is very little published evidence on the efficacy of outpatient use of limb pneumatic compression devices for deep vein thrombosis prophylaxis after major orthopedic surgery. There are no RCTs that evaluate outpatient use of pneumatic compression as an adjunct to pharmacologic prophylaxis in patients without a contraindication to anticoagulants. Some RCTs have evaluated the inpatient use of pneumatic compression as an adjunct to pharmacologic agents, but the results of these trials might not be able to be extrapolated to the outpatient setting.
There is also a lack of evidence on compliance with limb pneumatic compression devices in the outpatient setting. National clinical guidelines support the use of pneumatic compression devices for DVT prophylaxis after major orthopedic surgery in patients who are not candidates for pharmacologic prophylaxis due to a high risk of bleeding. In addition, one RCT that reported similar rates of post-operative DVT in patients who received pneumatic compression devices or low-molecular-weight evidence provides some evidence in support of pneumatic compression devices as the sole intervention in the outpatient setting. This study was limited in that much of the treatment occurred in the hospital, and patients with a known coagulation disorder were excluded from participation.
Venous thromboembolism (VTE) prophylaxis in major non-orthopedic surgery patients
Patients with and without a contraindication to prophylaxis with pharmaceutical agents
Randomized controlled trials
No RCTs were identified that specifically addressed the comparison between inpatient-only and inpatient and outpatient use of pneumatic compression devices as an adjunct to anticoagulant use in patients undergoing major non-orthopedic surgery. Moreover, no RCTs were identified that compared intermittent pneumatic compression (IPC) in the outpatient setting to no prophylaxis beyond inpatient use in patients with contraindications to pharmaceutical agents. Two systematic reviews of RCTs on VTE prophylaxis in patients undergoing major non-orthopedic surgery were examined, one a Cochrane review on VTE prevention in high-risk patients and the other on VTE prevention after gynecologic surgery; neither meta-analysis included RCTs relevant to the research question being considered. (12,13)
However, an RCT by Sobieraj-Teague and colleagues may contribute some relevant data. (14) The non-blinded study, conducted in Canada, compared inpatient and outpatient use of Venowave, a portable battery-operated IPC device, to usual care only in 150 adult patients undergoing cranial or spinal neurosurgery. As part of usual care, all patients were prescribed graduated compression stockings and early mobilization. Patients could also receive pharmacologic treatment at the discretion of their physician. A total of 19 of 75 patients (25%) in the Venowave group and 26 of 75 patients (35%) in the control group received anticoagulants (unfractionated or LMW heparin) and an additional 4 (5%) in the Venogram group and 7 (9%) in the control group used aspirin. In the Venowave group, devices were worn until development of VTE, patient refusal, until undergoing a screening bilateral venogram at day 9 (+/- 2 days) or earlier if patients were discharged from the hospital earlier and were unwilling to return for a venogram. The median day of hospital discharge was day 4. Patients who continued using the Venowave device at home received home visits at least daily to optimize compliance. Eight patients did not undergo screening venography. Mean time to screening was 7.3 days in the Venowave group and 7.5 days in the control group. The primary efficacy outcome was a composite of asymptomatic DVTs and symptomatic PEs. VTE occurred in 3 patients (4%) in the Venowave group and 14 (19%) in the control group. The difference between groups was statistically significant (RR: 0.21, 95% CI: 0.05 to 0.75). Most of the VTEs were asymptomatic, and there were no PEs. Two patients in the control group and none in the Venowave group experienced a symptomatic DVT. Among the 75 evaluable patients in the Venogram group, 17 (23.3%) were continuous users, 39 (53.4%) were intermittent users, and 17 (23.3%) discontinued use of the device before their venogram assessment. Compliance might have been lower if patients had not received daily home visits.
The Sobieraj-Teague study did not specifically exclude patients with a contraindication to pharmaceutical agents. Moreover, only about 30% of participants were prescribed heparin or aspirin. This suggests that study findings might be applicable to patients who are not taking pharmaceutical agents (i.e., including those with a contraindication). Generalizability of study findings is not clear, however, as the authors did not report VTE prevalence among patients who did or did not take anticoagulants or aspirin.
There is very little published evidence on the efficacy of outpatient use of limb pneumatic compression devices for deep vein thrombosis prophylaxis after major non-orthopedic surgery. There are no RCTs that evaluate outpatient use of pneumatic compression as an adjunct to pharmacologic prophylaxis in patients without a contraindication to anticoagulants. There is also a lack of evidence on compliance with limb pneumatic compression devices in the outpatient setting. National clinical guidelines support the use of pneumatic compression devices DVT prophylaxis after major non-orthopedic surgery in individuals at moderate- and high-risk of DVT. Moreover, one RCT, which found significantly fewer VTEs in patients undergoing cranial or neurosurgery who used a portable IPC than those receiving usual care; treatment occurred in both the inpatient and outpatient settings.
Patients undergoing major surgery, particularly orthopedic surgery, are at high-risk for venous thromboembolism (VTE), and VTE prophylaxis for high-risk patients may be indicated beyond the period of hospitalization. Pharmacologic prophylaxis is the mainstay of treatment, but some patients have contraindications to anticoagulation, such as a high bleeding risk. For these patients who are undergoing major orthopedic surgery or other high-risk surgeries, pneumatic compression devices are a reasonable alternative when prophylaxis is indicated in the outpatient setting. This is based on support in clinical practice guidelines, evidence from RCTs on different populations, and the lack of other good alternatives. Therefore, the use of pneumatic compression devices for outpatient VTE prophylaxis may be considered medically necessary when prophylaxis is indicated but there are contraindications to anticoagulation.
For patients who do not have contraindications to anticoagulation, the evidence is not sufficient to determine whether pneumatic compression devices offer additional benefit. There is a lack of studies that evaluate the added benefit of pneumatic compression devices in addition to anticoagulants and a lack of evidence on outpatient compliance. Therefore, outpatient use of limb pneumatic compression devices for VTE prophylaxis after major orthopedic surgery in patients who do not have a contraindication to pharmacologic prophylaxis is considered not medically necessary.
Practice Guidelines and Position Statements
In 2012, the American College of Chest Physicians (ACCP) published updated evidence-based guidelines on prevention of VTE in orthopedic surgery and non-orthopedic surgical patients.
ACCP recommendations on use of limb compression devices in orthopedic surgical patients (3):
For all of the above recommendations related to pneumatic compression pumps, the ACCP recommended only portable, battery-powered devices be used and stated that efforts should be made to ensure devices are worn for 18 hours per day. The authors noted that compliance is the biggest challenge associated with use of pneumatic compression devices.
ACCP recommendations on use of limb compression devices in non-orthopedic general and abdominal-pelvic surgical patients, stratified by patient risk of VTE and risk of bleeding (5):
Note: A recommended standard duration of prophylaxis was not defined. However, “extended duration” prophylaxis was defined as lasting 4 weeks.
Patient risk of VTE and risk of bleeding stratification
In 2011, the American Academy of Orthopaedic Surgeons (AAOS) published an updated guideline on prevention of venous thromboembolism in patients undergoing elective hip and knee arthroplasty. (6) The guideline included the following recommendations relevant to this policy:
In 2007, the American College of Obstetricians-Gynecologists (ACOG) published a practice bulletin on prevention of DVT and PE after gynecologic surgery. (2) As with the ACCP recommendations, described above, prophylaxis recommendations varied according to patient risk level. For patients at moderate and high risk of DVT, intermittent pneumatic compression was one of the recommended options for DVT prophylaxis. For patients at highest risk (i.e., older than 60 years plus prior VTE, cancer or molecular hypercoaguable state), IPC or graduated compression stockings plus LDUH or LMWH was recommended as a prophylaxis option. For all but the highest risk patients, the practice bulletin stated that, when IPC devices were used, “the devices should be used continuously until ambulation and discontinued only at the time of hospital discharge.” For the highest risk patients, the document stated that continuing prophylaxis for 2-4 weeks after discharge should be considered.
Medicare National Coverage
No national coverage determination was found for limb compression devices used to prevent DVT.
Water circulating cold pad with pump
Durable medical equipment, miscellaneous
Pneumatic compression, nonsegmental home model
Pneumatic compressor, segmental home model without calibrated gradient pressure
Pneumatic compressor, segmental home model with calibrated gradient pressure
Nonsegmental pneumatic appliance for use with pneumatic compressor, half arm
Segmental pneumatic appliance for use with pneumatic compressor, trunk
Segmental pneumatic appliance for use with pneumatic compressor, chest
Nonsegmental pneumatic appliance for use with pneumatic compressor, full leg
Nonsegmental pneumatic appliance for use with pneumatic compressor, full arm
Nonsegmental pneumatic appliance for use with pneumatic compressor, half leg
Segmental pneumatic appliance for use with pneumatic compressor, full leg
Segmental pneumatic appliance for use with pneumatic compressor, full arm
Segmental pneumatic appliance for use with pneumatic compressor, half leg
Segmental pneumatic appliance for use with pneumatic compressor, integrated, 2 full legs and trunk
Segmental gradient pressure pneumatic appliance, full leg
Segmental gradient pressure pneumatic appliance, full arm
Segmental gradient pressure pneumatic appliance, half leg
Pneumatic compression device, high pressure, rapid inflation/deflation cycle, for arterial insufficiency (unilateral or bilateral system)
Intermittent limb compression device (includes all accessories), not otherwise specified
New policy. New policy created with literature search through November 2012. Outpatient use of limb pneumatic compression devices after major orthopedic surgery is considered medically necessary in patients with a contraindication to pharmacological agents i.e., at high-risk for bleeding. Outpatient use is considered medically necessary after major non-orthopedic surgery in patients who are at moderate or high risk of venous thromboembolism with a contraindication to pharmacological agents. Other outpatient uses are investigational and outpatient use beyond 30 days post-surgery is not medically necessary.
Replace policy. Title changed to include the word “Postsurgical” and delete the word “Pneumatic” in the title and policy statements. Policy statement for investigational indications changed to not medically necessary indications for outpatient use of limb compression devices to prevent VTE. Policy and policy guidelines reformatted for usability. Added definition of nonmajor orthopedic surgery to Policy Guidelines. Policy updated with literature search through November 2013. Kendall SCD device added to Regulatory Status. Reference 8 added; others renumbered. Policy statements changed as noted. ICD-9 and ICD-10 codes removed; they were provided for informational purposes only.
Annual Review. No change to policy statement.
Disclaimer: This medical policy is a guide in evaluating the medical necessity of a particular service or treatment. The Company adopts policies after careful review of published peer-reviewed scientific literature, national guidelines and local standards of practice. Since medical technology is constantly changing, the Company reserves the right to review and update policies as appropriate. Member contracts differ in their benefits. Always consult the member benefit booklet or contact a member service representative to determine coverage for a specific medical service or supply. CPT codes, descriptions and materials are copyrighted by the American Medical Association (AMA).