Supervision of psychiatry residents in outpatient clinics brings unique challenges relative to other specialties. There are at least three main issues to consider in setting up an outpatient clinic supervision model: quality patient care, resident education, and financial reimbursement for the clinic.

Methods of supervision of psychiatry resident clinic outpatients include the traditional “supervisory session,” involving a resident and supervisor meeting for an hour weekly to discuss patients after they have been seen. Alternatively, real-time models of supervision, including in-person, one-way mirror, or video feed supervision, may be used. An advantage of the former model is the unobtrusive nature of it. However, concerns have arisen about the quality of patient care and education it affords. For example, supervisors may offer formulations without direct evaluation of primary data [1]. Moreover, reimbursement may suffer, as clinics may not be able to bill all insurance types.

Consequently, per anecdotal reports, real-time supervision models have increasingly been used in the recent years. Patient care may benefit as a result. For example, one study showed more than twice as many patients remained in active treatment or terminated after successful short-term care with in-person compared to traditional supervision for the intake appointment [2]. Conversely, real-time supervision may interfere with rapport building and may keep residents from asking uncomfortable questions in front of supervisors. Real-time supervision could also be detrimental to education if it results in less sense of resident accountability.

Somewhere in between real-time supervision and the classic supervisory session is a post hoc review of audio/video recordings, which may offer some of the benefits from each of the other two models. The authors of a recent review concluded that objective methods of supervision, including recordings or real-time supervision, enhance competence [3]. Additionally, another recent paper described how webcam-facilitated video recordings may be a feasible and affordable way to provide objective methods of supervision [4].

Specific data concerning methods of supervision currently being used in training programs are sparse [3]. Awareness of what other training directors are doing may be illuminating for design of programs’ own supervision models. Accordingly, the aim of this study was to survey US psychiatry residency training directors to determine methods they are using for psychiatry resident clinic supervision.

Methods

The authors anonymously surveyed all 183 directors of US adult psychiatric residency training programs in the 2012 American Association of Directors of Psychiatric Residency Training Membership Directory. Directors were sent an email invitation in 2012 to complete an anonymous web-based survey on psychiatry resident clinic supervision practices. A survey reminder was sent 2 and 4 weeks later.

The survey consisted of 17 multiple-choice and free-response style questions. The Institutional Review Board of Samaritan Health Services, Corvallis, Oregon, granted approval as an exempted study.

Results

Seventy of 183 (38 %) program directors completed the survey. See Table 1 for questions asked and specific responses.

Table 1 Survey results from 70 directors of US adult psychiatry residency training programs
Full size table

Role of Attendings

When considering both medication management and psychotherapy visits, the majority of programs use some degree of real-time supervision of residents. Many respondents indicated that they supervise resident outpatients in a number of different ways (question 1). For example, some programs utilize both live supervision and review of cases with individual supervisors on a weekly basis. In the optional comments section, some noted that they utilize live supervision only for new intakes and post hoc supervision for follow-ups. Others commented that whether live supervision is used or not depends on insurance and acuity. For example, some programs only use live supervision for Medicare patients, citing that Medicare only reimburses if live supervision is utilized.

Use of Audio/Video Recording

When asked if they record any resident outpatient clinic medication management or psychotherapy sessions, respondents were divided (question 5). Once recorded, resident sessions are reviewed in a variety of ways, with the majority reviewing them within 1 week of the session (question 6).

Billing

When asked if their primary resident outpatient clinic bills for resident medication management appointments (question 7), most respondents reported they do. About half reported they billed based on real-time supervision methods (question 8).

Respondents were also asked if they bill for resident psychotherapy sessions (question 9). Fewer programs, but still a substantial minority, bill for these visits as compared to medication management visits. Of those who do bill, it usually is not based on real-time supervision. Almost half of programs who bill for resident psychotherapy visits are able to bill for the entire time the resident spends with the patient (question 11). Programs that do not bill for all resident psychotherapy visits were asked the reasons (question 12). Over half of respondents said it is not feasible due to faculty time necessary to observe the required portion of sessions and/or the required real-time faculty supervision was not cost effective.

General optional comments from the survey included the theme of many that their clinics bill private insurance companies because they do not require live supervision. Medicare and Medicaid patients often either are not billed or are not seen by residents.

Discussion

Psychiatry residency programs appear to differ in the extent to which patient care, educational needs, and reimbursement are the primary drivers in the structuring of clinic supervision. Programs vary more in the arrangement of their supervision of psychotherapy sessions compared to medication management visits. The majority of programs utilize live supervision for medication management but not for psychotherapy. Reasons that programs use live supervision less frequently for psychotherapy include concerns about expense and time requirements, and to a lesser degree, perceived intrusiveness of it. Review of audio or video recordings during weekly supervisory hours may represent one way for programs to navigate the former concern; though post hoc review may not provide all of the benefits of real-time supervision, it would afford relatively objective review of actual patient material. Additionally, if live supervisors are not present for an entire session, they may miss important elements of it, and thus listening to recordings post hoc may in some ways provide a better opportunity to listen to those parts of the session retrospectively determined to be most critical.

Several programs report their residents are not seeing Medicare and Medicaid patients because of live supervision requirements for billing. Consequently, residents may be missing out on caring for and learning about important segments of the patient population. Importantly, there are differences in requirements between states and between insurance companies.

In an ideal world, this respondent may have it right: “We try to ensure supervision based on situation and need [determined by] patient issues and resident skill level first, then layer on payment, rather than allow payment to be the only driver of supervision needs. There is a disconnect between what an insurer will pay and what makes sense for a given trainee to be responsible for.” Obviously, clinics need to remain financially solvent, and training directors are merely one stakeholder group that cannot assume education of trainees trumps reimbursement issues. However, training directors should continuously be asking if there are ways they can align the different objectives of outpatient clinics so that multiple goals are met. A theme of what many respondents appear to find most desirable is flexibility about when and who receives the greatest degree of supervision and when supervision is real-time versus post hoc. Such flexibility could include: more consistent and extensive real-time supervision for underclass compared to upper class residents; more consistent and extensive real-time supervision for new patient evaluations compared to follow-up appointments; and supervisor availability in clinic if deemed necessary and appropriate by upper class residents as they are seeing patients (i.e., underclass residents would automatically have supervision during every visit, but upper class residents could have more flexibility in being able to call for supervision for their more complicated patient encounters). Indeed, an “as needed” element to supervision for upper class residents could be an exercise in learning how and when to seek guidance. However, residents are still learning and cannot always be relied upon to know what they do not know. As a safety net, there should be a mechanism of ensuring that all patients in treatment have some form of ongoing supervision, regardless of whether the supervision occurs in real-time.

Our study has limitations. Responder bias is a possibility, with those with the motivation to respond to the issue of psychiatry resident supervision perhaps more likely to participate. The relatively low response rate (38 %) may limit the generalizability of findings. We have no way of knowing if those directors who responded represent a unique subset in any way, as we did not assess locations or affiliation of respondents’ programs; this could also limit generalizablity. Finally, our survey was conducted prior to the implementation of Current Procedural Terminology (CPT) changes based on Evaluation and Management (E/M) coding for psychiatrists that went into effect in January 2013. Programs may have revised their supervision and billing procedures subsequent to those changes, though for the many programs utilizing self-pay and sliding scale mechanisms for resident psychotherapy visits, there would have been less need to change. General E/M requirements state that teaching physicians, in order to bill for E/M services, are required to document (1) that they performed the service or were physically present during the key or critical portions of the service when performed by the resident and (2) participation in the management of the patient [5]. Psychiatry-specific E/M requirements note “for certain psychiatric services, the requirement for the presence of the teaching physician during the service may be met by concurrent observation of the service through the use of a one-way mirror or video equipment. Audio-only equipment does not satisfy the physical presence requirement” [6]. Anecdotally, hospital legal departments, faculty members, and state psychiatric associations differ on exact interpretation of the E/M requirements for psychiatry resident supervision.

Future studies should examine objective patient outcomes with different types of resident supervision, including comparison of the traditional psychotherapy hour with various methods of real-time supervision and audio/video recordings to be able to recommend best practices. Additionally, studies should look at the types of experiences residents may be missing if precluded from treating Medicare or Medicaid patients. Third, studies should examine the impact of new CPT requirements on patient care, reimbursement, and education. Finally, research should delineate supervision models that best allow programs to evaluate residents based on the forthcoming milestones.

Implications for Educators

• Psychiatry residency programs utilize a variety of methods for managing psychotherapy supervision in outpatient clinics.

• Most psychiatry residency programs utilize live supervision for medication management visits in outpatient clinics.

• Supervision methods in some psychiatry residency programs vary depending on patient insurance.