Understanding Coding and Payment for Psychiatrists’ Services: How We Got Here and Where We’re Going

At the time of its inception in 1966, Medicare based physician payment on “customary, prevailing, and reasonable” charges (Figure 1). Physician payments were—and still are—covered under Medicare Part B, and hospital charges were and are covered under Part A, with separate rules for each.

Over the 25 years that followed, paying physicians on the basis of the amorphous idea of what was “customary, prevailing, and reasonable” created a number of problems, including inconsistent reimbursement patterns and policies, unsustainable growth in program costs, wage and price freezes put in place to stanch these costs, and a general compensation structure that overvalued new technology and procedures while devaluing the traditional care of increasingly complex patients.

1980s

In the mid-1980s, policy makers considered a variety of solutions to controlling the escalating cost of Part B payments, which eventually included the concept of a relative value scale (RVS). Although there was no absolute metric for all medical services, there was consensus among physicians that one service could be compared with another and found to be relatively more or less difficult. Policy makers had some experience with establishing the relative values of medical services by application of this concept, especially in California. Eventually, this scale came to include both the resources associated with delivering the service and the difficulty of the work, which allowed magnitude comparison among various services. Although this concept has reasonable face validity within services provided by a single specialty, there are doubts about validity when the service is provided by multiple specialties and even greater difficulties when comparisons are sought among different services provided by different specialties.

In 1985, a national study of Medicare reimbursement was undertaken at Harvard University, funded by the Health Care Financing Administration (HCFA; now known as the Centers for Medicare and Medicaid Services [CMS]). The study involved a resource-based relative value scale (RBRVS). Phase 1 of the study was completed in 1988, and a report was submitted to HCFA (1), with relative value units (RVUs) for the work of physicians in each medical service covered by Medicare. Psychiatry was one of the 18 specialties that participated in phase I to establish these values.

In 1989, Congress enacted the Omnibus Budget Reconciliation Act (OBRA) (2). It mandated that, beginning in 1992, Medicare payments would be based on an RBRVS. In 2000, the RBRVS was expanded to account for practice expenses (in practice expense RVUs) and the cost of malpractice and professional liability insurance (in professional liability RVUs), along with the level of physician work RVUs.

The sum of these three RVUs is the total RVUs for each code. This total represents a ratio or percentage, which is multiplied by the conversion factor for that year. In this manner, reimbursement rates are readjusted each year, even when the service RVUs have not changed. There are separate facility rates for services in hospitals and other facilities, and there are nonfacility rates for services in physician offices and other nonfacility settings. In addition, a geographic adjustment (the geographic practice cost index), applied to the total RVUs, accounts for regional variations in the costs of providing services. The OBRA required that each Current Procedural Terminology (CPT) code be reviewed at least every five years (in a five-year review) to correct any errors and to take into account any changes in technology or practice changes and any other events that would require adjustments.

1997

In 1997, as part of the Balanced Budget Act (3) of that year, Congress established something called the sustainable growth rate (SGR) in an attempt to control the growth in spending for physician-based services; no such law was passed for the hospital-based services. The SGR mandated that a Part B spending target be established each year on the basis of the gross domestic product. If more Part B services are being used, then the conversion factor for the next year’s fees (the number multiplied by the RVUs assigned to each service to determine the Medicare payment) must go down.

The SGR ignores all the drivers of demand for physician-based services that might account for an increased demand. This budgetary limit creates a “zero sum game,” whereby when the codes change in value or codes are added, the number of work RVUs in place changes, and the conversion factor, which establishes how much an RVU is worth, must also change to maintain neutrality. When the total of RVUs increases in number and value, the conversion factor goes down. This set up an annual intervention by Congress whereby every year since 2003 Congress passed a law to override what would have been a negative adjustment to the conversion factor. The “doc fix” ensured that physician income for Part B services would remain stable but did nothing to address the mounting debt associated with the increase in spending, which would need to be repaid at some unknown date in the future, by some unknown method.

2015

In April 2015, Congress passed the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) (4), which established a new way to determine payments to physicians. The MACRA repealed the flawed SGR formula that triggered deep cuts in payments for physician services, year after year. In its place, the law requires annual, across-the-board “updates” in Medicare Part B payments, which generally equate to a modest annual increase or at least a freeze of current rates. In addition, the MACRA replaced current quality-reporting programs with a new Merit-Based Incentive Payment System (MIPS), which provides upward or downward payment adjustments, depending on performance, as well as bonuses for participation in certain alternative payment models (APMs). The changes are designed to reward physicians for demonstrating a high level of quality of care or participating in new models of care that reward quality and efficiency.

The current codes and reimbursement rates for psychiatric and mental health services were adopted in 2013 and implemented in 2014, and they fundamentally transformed the business of psychiatric practice. Under Medicare Part B, physician work is codified (Figure 2) according to the American Medical Association’s (AMA’s) CPT (5), which provides the taxonomy used to document medical services provided in the United States. The CPT editorial panel develops codes that describe and delineate specific health services provided by individual clinicians (Figure 3). The codes for each year appear in the annual CPT Codebook, issued by the AMA.

The AMA issued the first CPT codes in 1966, and CMS adopted these in 1983 as the basis for reporting physician services under Medicare. Initially, these codes included only services by physicians. In 1997, however, these codes came to include the services of nonphysicians, including psychologists, social workers, occupational therapists, podiatrists, chiropractors, and others. The Health Insurance Portability and Accountability Act of 1996 (6) mandates that claims for all medical services must be submitted with CPT codes. The CPT editorial panel meets three times a year, taking input from specialty societies for new codes and revising the code structure to more accurately describe medical services as they are currently provided. More information about the CPT process and editorial panel is available on the AMA website at https://www.ama-assn.org/search/ama-assn/CPT.

New procedures often begin as Category III codes, temporary tracking codes to describe new work or new technology. Once practitioners have sufficient experience with the Category III code, it can be reviewed by the CPT Editorial Panel and designated a Category I, or a standard CPT code. The values of category III codes are set by the individual Medicare contractors (“carrier pricing”), whereas the values of Category I CPT codes are set by the CMS on the basis of multiple sources of information including recommendations from the AMA/Specialty Society Relative Value Scale (RVS) Update Committee (RUC) (see below).

The codes within CPT contain a large category of evaluation and management (E/M) codes. The E/M codes are further grouped by site of service, whether a patient is new or established, and the level of intensity or complexity of the service delivered. There are also more than 7,000 CPT codes for specific surgical and nonsurgical procedures, as well as codes for medical interventions. For example, these include the codes for electroconvulsive therapy (ECT) and transcranial magnetic stimulation (TMS). Many codes for surgical procedures bundle together the hospital days and the outpatient testing and follow-up appointments that occur within a particular “global period.” The global periods are generally 90 days, 30 days, or even 0 days (when all relevant services occur the same day).

Two specific observations are worth mentioning at this point. The first is that the CPT codebook directs the provider to “select the name of the procedure or service that accurately identifies the service performed. Do not select a CPT code that merely approximates the service provided. If no such specific code exists, then report the service using the appropriate unlisted procedure or service code” (5). The second is that the CPT codes are intentionally crafted not to be provider or specialty specific (e.g., the code for ECT does not specify that it should be administered by a psychiatrist).

The Process of Valuing Services

Codes enter the CPT or RUC process through several doors. First, the CPT editorial panel describes a new procedure (e.g., TMS), then eventually through the RUC process, the specialty society providers determine where the work of the new procedure fits compared with other, previously valued procedures. Many existing codes are reviewed at the request of CMS, according to several “screens” developed by the RUC or CMS or even required by Congress. The purpose of these is to identify codes that may be misvalued. One example of a screen is rapid growth in the total utilization of a code. A society may bring a code forward for review, but it must provide proof that the code was previously misvalued and must meet the “compelling evidence” standards of the RUC before a review of the code is even considered. Individuals can also request review of specific codes by sending a request to CMS.

The RUC was established in 1991 to recommend values to CMS for new and revised codes (Figure 4). The RUC generally meets three times a year, usually in January, April, and October. Special meetings may be called as well. The membership of the RUC includes representatives appointed by major national specialty societies, including the American Psychiatric Association (APA). Of the 31 voting RUC members, 21 are appointed in this manner, and four are elected to rotating seats. The remaining six seats include the RUC chair; the cochair of the Health Care Professionals Advisory Council (HCPAC), which represents nonphysician practitioners; the chair of the Practice Expense Subcommittee; and representatives of the AMA, the American Osteopathic Association, and the CPT editorial panel. Representatives from CMS also attend the RUC meetings as nonvoting guests. Additional information about the RUC composition, process, and procedures is available on the AMA website at https://www.ama-assn.org/practice-management/rbrvs-resource-based-relative-value-scale.

The RUC develops recommended values for codes and submits these recommendations to CMS. CMS staff review the recommendations as they prepare the next year’s Medicare physician fee schedule; decide which to adopt and which to reject or tweak; and then finalize the RVUs as part of the Final Rule, which is released in late fall each year. This advisory role for the RUC is intended to avoid a situation in which physicians might be seen to be setting their own prices. Federal antitrust law specifically prohibits the provider of a good or service who has a monopoly from setting the price for that good or service—and physicians are defined as having such a monopoly.

The RUC Advisory Committee is made up of advisors from the 118 specialty societies that are part of the AMA House of Delegates. The role of the RUC advisor (often working along with an RVS committee within his or her specialty) is to advise and help develop recommendations for consideration by the RUC for the work values, times, and practice expenses of new and revised codes as well as codes that have been identified as being misvalued or in other screens applied by CMS or the RUC. As codes come up to be valued or revalued, the RUC queries each of the societies as to their level of interest in participating in the valuation process. Those societies whose members perform the services may participate in a survey process and present recommendations to the RUC, help inform that process, or opt not to participate at all.

The RUC members appointed by the specialty societies to sit at the table and deliberate valuation proposals are not allowed to advocate for their particular specialty or society. Rather, they serve as impartial participants who thoughtfully consider, as physicians, the recommendations being presented to ensure that the codes are fairly valued relative to existing services. The specialty societies’ advisors, conversely, are the advocates, presenting valuation proposals for formal review by the RUC and making the case for valuation in front of the voting RUC representatives.

The survey process, conducted by the specialty societies whose members will be using the codes, includes the work of developing a “vignette” (description of the service and a typical patient) for each code under review, identifying a random sample of the specialty to be surveyed on this code, developing a list of services (by CPT code) that are similar to the work under review (which helps in determining a relative value), and administering the survey. The AMA RUC staff have developed a standard RUC survey instrument. The RUC Research Subcommittee is charged with review and approval of changes to the standard RUC survey, so the subcommittee may be involved if problems are anticipated.

The purpose of the survey is to get data on the time and effort it takes to do the work described in treating the typical patient. This includes identifying a service that is similar to the one being surveyed and comparing the two. Once the survey results are back, the societies prepare recommendations for the RUC, typically including time spent on patient care before the provider sees the patient (preservice time), during the actual visit (intraservice time), and after the visit concludes (postservice time), as well as hospital visits, outpatient visits, and work RVU for the service in question. The societies also prepare recommendations for the practice expense, which includes expenses for staff and supplies. The AMA RUC staff require prior approval of education regarding the RUC survey process for potential survey participants.

The RUC is charged with making recommendations on valuation proposals for services provided by physicians. (The HCPAC is composed of representatives of specialty societies for nonphysician practitioners who may be enrolled in Medicare. One of the HCPAC cochairs is from the AMA. The HCPAC considers and makes recommendations regarding valuation proposals for services performed exclusively—or primarily—by nonphysician Medicare providers.)

Prior to each RUC meeting, the valuation proposals for each group of codes are organized into a “tab” that is assigned to two or three RUC members for review and comment. As the RUC meeting begins, the first presentation for that tab is made to the Practice Expense Subcommittee, which goes over a line-by-line accounting of clinical staff and supply expenses for each code under consideration. After that, presenters for each tab (often the RUC advisors of the interested societies) make their way to the table and present their recommendations to the RUC on physician work on each code. The RUC members assigned to review that tab lead the discussion of the code, including making comments or asking questions about the data that were presented. All 31 voting RUC members participate in this process, having previously received not only the recommendations of the societies but also the reviewer’s comments. Eventually the RUC votes regarding whether to accept the society’s recommended values.

If the time and work recommendations pass, then the chair of the Practice Expense Subcommittee presents his or her recommendations for practice expenses, which the full RUC also votes on. If the time and work recommendations fail, they are often assigned to a facilitation committee, and each member of the RUC is asked to submit written comments and recommendations to facilitate an appropriate valuation. Each facilitation committee is composed of a number of RUC members (and experienced RUC advisors) who meet with the presenters and take into account the RUC’s written comments to craft a recommendation for the RUC to reconsider. The facilitation committee chair presents a report to the entire RUC, detailing the committee’s recommendations, which the RUC then votes on. Following acceptance of the facilitation committee report, the RUC votes on the practice expense recommendation.

The RUC uses a variety of methods in developing its recommendations. The most important is the survey. The RBRVS at the heart of Medicare physician payment by its nature requires comparison of the relative value of new (and revised) codes for services, as compared with existing codes and services. The fundamental concept behind valuation is magnitude estimation, which basically compares the code under review and determines whether it is bigger or smaller or how much more or less it is worth than services that have been valued previously.

A method that the surgical societies have used frequently is the intensity of work per unit of time. This calculation creates a number as a surrogate for an intensity measure. It appears to be valid when one compares similar services within the same family but presents challenges when used across families, particularly because some services actually have a negative number. Comparing the values of different codes is particularly problematic if they have different (or no) global periods. In addition to the ongoing workload of valuing individual codes, the RUC members, subcommittees, and participating societies have also been involved in developing and discussing alternative valuation methodology, alternative physician payment models, and other emerging issues.

As the ultimate decision maker regarding the Medicare valuation of physician services, CMS has an integral, although not always transparent, role in this process (Figure 5). As a payer, CMS is accountable to ensure that the services billed for were provided. Although the CPT editorial panel established the framework for the E/M codes, CMS published specific E/M documentation guidelines in 1995 to enable auditors to determine whether the service for the particular charge had actually been delivered. The guidelines established eight elements for defining the depth of a history taken, specifically delineated the elements that constituted a comprehensive exam, and provided a method for quantifying the degree of the cognitive work (medical decision making) required for the code.

CMS defined the elements of the history, which are not unfamiliar to physicians but unfortunately do not always translate well into the work of psychiatry. They include location, duration, quality, severity, timing, context, modifying factors, and associated signs and symptoms. A high-level history requires four of the above elements or comments on three chronic conditions the patient has.

In the 1995 CMS guidelines for a comprehensive examination, the highest level only acknowledged a general physical exam, which was problematic for many specialties (including psychiatry) because it included elements (breast, pelvic, and rectal) not appropriate to the specialty. CMS attempted to correct for this problem by publishing new guidelines in 1997 that defined single-system examinations for specialties that could be used to fulfill the requirements for a comprehensive examination. The structure of these single-system exams is similar, with all including vital signs and general appearance, after which they narrow slightly. The single-system psychiatry exam includes elements of the neurologic and musculoskeletal exam as well as elements specific to psychiatry (e.g., hallucinations and delusions).

Unfortunately, the structure of the psychiatric specialty examination, as described in the CMS documentation guidelines, is not consistent with the way most psychiatrists conceptualize psychiatric function. The guidelines specify how many of the elements listed must be included for the exam to be defined as limited, detailed, or comprehensive, whether or not those specific elements are pertinent to the problem at hand. Despite this, CMS, as well as the AMA and other specialty societies, has been resistant to restructuring the psychiatric specialty examination, out of concern that to do so would precipitate a wholesale revision of the E/M codes.

Attempting to quantify the medical decision making has been even more problematic. The current model identifies three dimensions that determine the difficulty of the medical decision making. These are the nature and number of problems or management options, the quantity and complexity of data, and the risk of the intervention. The difficulty of the medical decision making is determined by the highest two of the three dimensions (an online supplement accompanying the online version of this article contains a chart developed by the APA and American Academy of Child and Adolescent Psychiatry [AACAP] to aid psychiatrists in determining how to select the appropriate E/M codes).

Although the E/M codes have, in concept, been available for psychiatrists to use since their inception, they generally were not used in outpatient care before 2013, in part because many states and payers did not include them in the fee schedules for psychiatrists. Historically, psychiatric services were bundled together, which resulted in a small number of CPT codes to describe psychiatric services (7). For example, one of the earliest CPT codes, 90841 (individual medical psychotherapy by a physician, with continuing medical diagnostic evaluation, and drug management when indicated, including insight oriented, behavior modifying or supportive psychotherapy; time unspecified) included four different services (psychotherapy, medical diagnostic evaluation, psychiatric evaluation, and drug management). This provided a limited number of options for psychiatrists billing their services.

As a result of this structure, psychiatrists were not allowed to bill for an E/M service and a psychotherapy visit on the same day. Instead, they had the option of billing for “medical psychotherapy” or a medication management service, depending on the nature of the work (8). Early revisions included adding timeframes to the codes. These codes were quite basic: 90801 for initial evaluations, 90843 for services of 20–30 minutes, and 90844 for services of 45–50 minutes.

In 1995, as part of the regular RUC review process at the time, APA requested a review of the codes in the psychiatry section on the basis of the vast changes in work that had occurred since the codes were initially valued. HCFA was moving to incorporate all professional services, including those of psychologists and social workers, in the Medicare physician fee schedule and “asked” APA to include those groups in any survey process. A joint survey was completed, (although not without APA voicing concerns), and the recommendations, which included an increase in RVUs for five services, were put forward to the RUC. HCFA, however, dismissed the compelling evidence arguments and declined to adopt the changes.

Following significant advocacy by the APA, CMS introduced a series of Healthcare Common Procedure Coding System (HCPCS) codes beginning with the letter G (G-codes), to take effect in January 1997, that provided greater granularity. These codes divided up the work by site (inpatient versus outpatient), duration of service, and whether E/M services were delivered (Figure 6). Psychologists and social workers would no longer be able to bill the medical psychotherapy codes but rather would bill a psychotherapy code that did not include any medical evaluation or medical management services. The CPT codes (90804–90829) went to the RUC, and the standard procedure was launched to determine appropriate recommendations for the valuation of the new codes.

Unfortunately for psychiatry, the results of the surveys undertaken to determine value were not particularly helpful. Survey respondents not only were unfamiliar with the survey tool, they also were unfamiliar with the new coding structure and the codes they used for comparison. Also, the surveying groups (psychiatrists, psychologists, and social workers) did very different kinds of work and saw very different patients from one another. Nonetheless, the expectation was that a recommendation be made, so, by a methodology of regression analysis, values were determined, recommended to, and accepted (with some reductions) by CMS, and they became the standard with the 1999 Medicare physician fee schedule.

As time passed, it became apparent to psychiatrists that the value of the codes used by psychiatrists and other mental health providers was slowly drifting down. This was occurring due to the fixed sum for physician services under Medicare, such that for every new procedure valued or increase in the value of an existing service, the conversion factor applied to all other codes (including the extant psychiatric codes) decreased. Furthermore, increases in the value of the E/M codes distorted the original crosswalk that had existed between the E/M codes and the codes most frequently used by psychiatrists. Another development was increasing concern that the code most frequently used by psychiatrists (90862, medication management) was overvalued and should be resurveyed.

In anticipation of the 2010 five-year review of RVUs and the opportunity presented to address the misvaluation, the board of trustees of the APA approved efforts by the Committee on RBRVS, Codes, and Reimbursement to seek change. Representatives of APA, the AACAP, the American Psychological Association, the National Association of Social Workers, the American Nurses Association, and the American Psychiatric Nurses Association met and determined to have those procedure codes representing their work included in that review. They drafted a group letter to CMS laying forth their case of compelling evidence that the codes had been inappropriately valued.

CMS agreed to a review by the RUC. That review started at the research committee of the RUC, which recommended to the RUC that instead of just a revaluation of the existing codes, the entire family of psychiatric procedure codes should be referred back to CPT to be restructured. The RUC agreed to that recommendation and moved the family of codes to CPT for restructuring.

Within weeks, a CPT work group had been formed to begin this work, with representatives from the CPT editorial panel and of the groups noted above. The group struggled to define parameters measuring the specifics of work done by mental health practitioners that would not create onerous documentation requirements. When the chair of the original work group was given a new assignment, the work group was dissolved, and a new work group formed with new leadership. This new group struggled with the same issues as the previous one but eventually was able to craft a compromise framework that was accepted by CPT.

This compromise saw the elimination of 90862 and the separation of the diagnostic evaluation into two new initial evaluation codes, including one that included the assessment of the patient’s medical (physical health) status, differentiating between physician and nonphysician work. In addition, different methodology was now used for defining the E/M work and psychotherapy work in a psychiatric encounter, with the use of the standard physician E/M codes to record the E/M work and an “add-on” psychotherapy code. (An add-on code is a service that is always performed with another service and must never be reported as a stand-alone code. In this case, the primary service is an E/M service, and the psychotherapy is considered secondary for coding purposes.) Other additions included an interactive add-on code, codes for crisis interventions, and a code that those few psychologists licensed to prescribe could use when prescribing (in states with scope-of-practice laws permitting this).

The new codes were sent from the CPT editorial panel to the RUC to be valued, and the APA and other mental health provider societies began their work, as described above. Dr. Jeremy Musher was the spokesperson for the APA at CPT and led the presentations at the RUC on both practice expense and work and time recommendations. The family of codes went through prefacilitation, and, in the end, the values for the codes were approved as the societies had recommended.

The RUC recommendations were forwarded to CMS for consideration as they reviewed valuations for the 2013 Medicare physician fee schedule. CMS chose not to accept the value recommendations for 2013, citing that the entire family had not yet been surveyed. Instead, for 2013 CMS created interim values for the new code structure.

Unfortunately, when they did so, they did not apply consistent logic, especially regarding practice expense. The components included in the practice expense values had recently been revisited for all societies through an AMA survey. The survey results showed that psychiatrists did not “typically” (defined as more than 50% of the respondents) have clinical staff in their offices. Given the nature of the services provided, psychiatrists have limited medical equipment and medical supplies in their offices. Costs attributed to work performed by clinical staff were the largest component of the practice expense values up to this point.

Unable to combat the AMA survey results, the RUC stripped that category from the values, and substantially lower practice expense values were attributed to psychiatric services. CMS chose to add those new, lower practice expense values to the new psychiatric initial evaluation with medical services (90792). However, the old, higher practice expense values continued to be attached to the new nonmedical initial evaluation, which resulted in a greater value for the nonmedical code. All attempts by the APA to have this corrected failed, and this anomaly ended up in the Final Rule for 2013.

The remaining codes were surveyed in 2013, which enabled CMS to go forward with final values for all of the codes. Late in 2013, the Final Rule for 2014 was released, and the values recommended by the RUC that the APA and its partners had championed were now applied by CMS, with one exception. CMS chose not to value or cover the CPT code for pharmacologic management when done by a psychologist.

The timeframe of these changes was no different than for other coding changes that occur every year. However, the private payers struggled to implement the new fee schedule in 2013 because of the extent and nature of the change in the coding framework. The APA’s board of trustees continues to direct current activities in this area, including the continued efforts of the APA Committee on RBRVS, Codes, and Reimbursement regarding new codes, ongoing efforts by the committee to educate members about coding changes and correct coding and documentation, work done directly with payers regarding payment policy, and the consideration of potential legal action to prevent payer policies that violate federal laws requiring parity between payment for mental health and medical services.

The Future

For a number of years, we have been hearing about the paradigm shift in the way physician and other medical services are provided to patients as well as the way such services will be reimbursed. When physicians try to redesign the ways they deliver services to provide higher quality patient care at a lower cost, they find that barriers in current payment systems prevent them from doing so. Barriers include the lack of payment or inadequate payment for high-valued services and financial penalties for delivering a different mix of services.

Alternative payment models (APMs) can provide a way of overcoming the barriers in current payment systems so that physicians can deliver higher quality care for patients at lower costs for purchasers in ways that are financially feasible for physician practices. Characteristics of successful APMs include flexibility in care delivery, adequacy of payment, and accountability for costs and quality. MACRA encourages the creation of APMs and provides incentives for physicians to participate in them. MACRA explicitly encourages the development of “physician-focused payment models,” with no single approach that will work for all physicians and for their patients.

APMs span condition-based payment, multispecialty bundled payment, physician-facility bundled payment, and episode payment for a procedure. The hope is that APMs will allow recognized savings through reduction of emergency department visits and hospital admissions, reduction of unnecessary tests and treatments, more efficient service delivery, use of lower cost sites of service, and the prevention of serious conditions. If APMs are constructed well, we should see savings for the payer and increased physician practice revenue.

With MACRA legislation, CMS has enacted bonus reward systems and penalties. Physicians who are qualifying participants in accountable care organizations (ACOs) and certain other qualifying APMs are rewarded with bonuses and higher annual updates. It should be noted that not all ACOs have been effective, and many have ceased to function. MACRA offers physicians a choice between two payment pathways: a new quality program known as the MIPS that offers substantial bonuses for high performance, or involvement in an “advanced” APM, such as an ACO. The majority of practicing psychiatrists likely fall under the MIPS provision.

Most psychiatrists who participate in Medicare and cross the threshold, in terms of either absolute number of Medicare-insured patients or annual income from Medicare, need to meet the components of MIPS to avoid penalties, unless they meet the qualifications for being in an acceptable APM. MIPS has four components: quality reporting (formerly the Physician Quality Reporting System), cost (formerly the value-based payment modifier), advancing care information (formerly meaningful use), and the new category of improvement activities. Through a combination of weighting of these categories and addition of scores based on physician performance, there will be future increases or decreases in Medicare payments to physicians. Physicians who participate in advanced APMs do not have to participate with MIPS and will receive a 5% bonus from Medicare from 2019 through 2024, as well as higher annual updates in later years. The APA has been actively involved in creating a national mental health registry, known as PsychPRO, that will allow psychiatrists to participate. By doing so, they will meet many of the MIPS requirements to avoid penalties and to realize bonuses. PsychPRO has been approved by CMS for MIPS reporting as a qualified clinical data registry. Participation in PsychPRO is free to APA members.

APMs are already shaping psychiatric practice and the role of psychiatrists. There are now new billing codes (99492-99494) for collaborative care model (CoCM) services. These are for reimbursement of a team consisting of a primary care provider who submits the code, a behavioral health care manager employed by the primary care provider, and a consulting psychiatrist. The consulting psychiatrist contracts directly with the primary care provider. The psychiatrist provides regular case management through discussions with the care manager and, on occasion, directly with the primary care provider. There is the option for some direct face-to-face contact between the patient and the psychiatrist, and this service is billed separately. CMS approved codes for these services, as well as Medicare coverage and payment, starting in January 2017. For more information on these service, visit https://www.psychiatry.org/psychiatrists/practice/professional-interests/integrated-care/learn.

More than 80 randomized, clinical trials have demonstrated that care provided by this model improves access, improves quality of care, and decreases cost. Use of technology such as telemedicine may further expand this model. Certainly, other innovative models can be expected in the future. We are certainly entering a phase during which physicians and health systems will be challenged to provide cost-effective, evidence-based, high-quality patient- and population-based care.

Whatever the future holds, one trend that will likely continue is psychiatrists who select employment contracts, as opposed to a private practice model. With all of the implications of MIPS as well as the increasing complexity of navigating the system of payment for services, this may be a prudent choice. A challenge, however, is the dearth of psychiatrists and other physicians in leadership in larger organizations. It seems imperative to help guide those psychiatrists interested in administrative positions to engage in such leadership roles.