Following passage of the Patient Protection and Affordable Care Act, emphasis has been placed on two health care delivery redesign initiatives in the United States. One of these is the patient-centered medical home, which stresses strong primary care infrastructure. The other is the accountable care organization (ACO), which mandates coordinated care delivery across a full spectrum of health care. Transitions to these new models provide opportunities for innovation by both general and specialty medical services and are guided by two fundamental principles: (1) enhancing cost-effectiveness of clinical resources and (2) increasing quality of care delivery.
At academic health centers, faculty and administrators find themselves juggling the simultaneous conception, implementation, and evaluation of such redesign efforts while training the next generation of physicians.1,2 The ACO model, with its emphasis on cost-effectiveness, rewards health care systems that provide the highest-quality care in each component location—including primary care and specialty outpatient clinics—delivered by the most cost-efficient providers.
Skin diseases constitute large numbers of primary care and specialist visits with significant health care costs. Dermatology is a specialty with an ongoing workforce shortage3 and limited comparative effectiveness data. Addressing cost-efficiency in the management of skin diseases under the ACO model is made challenging by this lack of data and the number of clinical settings in which skin diseases are currently managed. As a result, opportunities exist to implement best practices that will have a significant impact on overall health care costs.
Following adoption of an alternative quality contract with a large insurer4 and transitioning to a Pioneer ACO5 at Massachusetts General Hospital (MGH), we sought to address the problem of identifying and implementing specialist-driven strategies for cost-effective, patient-centered care delivery in dermatology. Our method may represent a framework that can translate to other specialist workforces facing similar challenges.
The Department of Dermatology at MGH comprises 35 dermatologists who see patients at a main clinic adjacent to MGH, as well as at five satellite clinical locations in Boston and the surrounding cities. The annual clinical volume is more than 50,000 visits per year. To develop rational strategies for cost reduction and quality improvement within our dermatology department, we identified a committee of residents and faculty with backgrounds in clinical dermatology, dermatologic surgery, phototherapeutics, and clinical administration. This breadth of backgrounds was meant to capture wide expertise, as we hypothesized that multiple subspecialties of medical and surgical dermatology likely had an impact on overall costs (i.e., high-cost pharmaceuticals and high-cost procedures). Involvement of residents provided trainees with firsthand exposure to analytical metrics used in systems-based practice, an Accreditation Council for Graduate Medical Education core competency.
The committee first convened in 2012. Our initial tasks included identifying existing cost-containment plans within the department, with a focus on measures targeting the rate of rise in per-patient medical expense. We selected this focus to align with the administrative goals of our institution and physicians organization; the leadership of both were interested in innovative approaches to cost-effectiveness in an ACO. Hospital and physicians-organization administrators provided us with fiscal year (FY) 2011 data regarding clinical appointment volume, visit billing and coding, and provider prescribing behavior for the patients seen in our main clinic who were covered by the insurer with which we have an alternative quality contract. Hospital pharmacy services data identified the costs of all medications prescribed to these patients during FY 2011 as well as the costs of all dermatology-relevant medications available for common dermatologic conditions. The committee reviewed aggregated billing and administrative data from 9,721 office visits, 8,362 ambulatory procedures, and 3,676 prescriptions.
We found that although practitioners had adopted individual best practices, there were no department-wide initiatives focused on improving cost-effectiveness. We then looked for opportunities to implement three key strategies in the department:
- Substitute less expensive but therapeutically similar interventions
- Decrease utilization of some services
- Eliminate unnecessary tests and prescriptions
Using the FY 2011 data, we calculated summary and descriptive statistics for metrics related to clinical care including patient visit coding, prescription frequency by agent and by class, surgical intervention frequency, and evidence-based use of procedural therapeutics. Through collaborative data review and iterative discussions concerning cost-efficiency, we classified potential action areas aligned with fiscal incentives. During the committee’s discussions, we identified targets for action and developed department-wide practices for unified approaches toward reaching these goals. Additionally, we highlighted opportunities for interdepartmental collaboration that emphasized mutual alignment of patient, provider, and ACO cost-efficiency goals. Figure 1 summarizes our methodology.
Our committee identified four action areas for addressing specialist cost challenges within ACOs: (1) rational, cost-conscious prescribing within therapeutic classes; (2) enhanced management of urgent access (UA) and follow-up clinical appointment scheduling; (3) standardization of procedures; and (4) assessment of interpractitioner variability. Below, we discuss examples of the practices we have adopted to target each of these areas and offer recommendations for other specialty departments.
Aim for rational, cost-conscious prescribing within therapeutic classes
Our initial cost analyses demonstrated that nearly 12% of all costs related to prescription medications. Wide prescribing variation existed among the department’s physicians, even within single medication classes. Limited comparative effectiveness research on dermatologic therapeutics and the availability of multiple agents for the same indications enable such variation.
As an example, our department generated a total of 385 prescriptions for topical antifungal agents. Our pharmacy data demonstrated an order of magnitude difference in the cost per prescription within topical antifungals, ranging from $22 to $277 per agent. The most expensive antifungal accounted for 5% of the prescriptions but nearly 38% of the prescription costs within this class. On this realization, we reviewed all antifungal agents for both cost and efficacy, and then delineated department-wide preferences for first-choice prescription and nonprescription antifungals. We introduced an incentive program promoting effective generic medication substitutions for common brand name agents with slightly different formulations to avoid automatic substitution at outpatient pharmacies. Subsequently, we implemented monthly “cost pearl” presentations in which faculty identify areas where medication or procedural substitution is clinically appropriate and saves costs.
The practice of evidence-based medicine has significant implications for cost-efficiency and resource utilization, particularly under the ACO model. Although more comparative efficacy data are needed to better inform treatment decisions, specialty workforces should review their prescribing practices to identify and limit use of medications with significant costs but without added quality toward clinical end points. Specialist physicians who frequently prescribe particular medication classes should champion such measures, given evidence suggesting that specialists have opportunities for greater cost savings than do primary care providers via cost-effective therapeutic substitutions.6,7
Enhance management of UA and follow-up clinical appointment scheduling
The spectrum of care for dermatologic disease ranges from a single appointment encompassing evaluation, diagnosis, and removal to long-term disease management requiring frequent medication adjustment and treatment. Wait times for new patient appointments can be long. Analysis of our department’s clinical volume revealed that 23% of 2011 visits comprised new patient encounters, up from 19% in 2008.
Before our institution’s transition to a Pioneer ACO, our department had restructured its clinical triage approach. In 2008, we instituted a UA clinic model, where several clinic sessions each week were dedicated to focused evaluations of single, urgent dermatologic issues. Primary care providers within our institution are empowered to request urgent evaluations of patients by dermatologists, and the resulting UA clinic appointments often occur within 48 hours. Now, 38% of all new patients are seen in less than two weeks. Over the program’s first four years (2008–2011), we witnessed consistent growth in UA clinic referrals. In response to demand, we have doubled the number of UA clinic sessions. Our dermatology residents rate UA clinics highly for educational content as they have chances to make new diagnoses on first disease presentations.
Using UA clinics within specialty fields helps address multiple issues related to cost-efficiency and quality. Expedited specialist access results in earlier diagnosis for patients with subtle or atypical clinical presentations and decreases both therapeutic misuse and overuse. Empowering primary care providers to request urgent specialist input underscores the bidirectional accountability in both appropriate referral practice and accurate diagnosis, fostering interdepartmental professional interaction and feedback. Finally, incorporating residents and students into such clinics presents a strategy that leverages cost-effectiveness while providing unique learning opportunities.
Modulating the timing of return visits is another consideration for enhancing quality of care. Although follow-up visit determination depends on patient, provider, and system factors, developing department-wide, rational approaches to follow-up scheduling based on duration of therapeutic courses can result in more effective care delivery.8,9 In our department, this issue represents an ongoing focus.
Standardize indications for procedural interventions
Dermatology is a field with a range of diagnostic and therapeutic procedures. Review of our department’s data showed that procedures accounted for 47% of our costs. We determined that skin biopsies constituted 18% of procedures performed and the destruction of premalignant skin lesions accounted for nearly 17%. Both types of procedure are routinely done during the course of an office visit; neither requires the patient to schedule a separate procedural appointment. After a skin biopsy is performed, additional tissue removal may be required based on the histological diagnosis and the physician’s clinical assessment.10 Our committee’s discussion of postbiopsy clinical management practices demonstrated mixed perceptions of appropriate strategies and encouraged us to collaborate with our dermatopathologists to develop standardized approaches that allow for individual clinicians’ discretion. Discussion of two potential approaches—encouraging full removal of lesions to obviate reexcision versus reaching clinical consensus that some partially removed lesions do not need reexcision—is ongoing.
Our review of procedural data attributed 8% of procedural visits and 2% of overall clinical costs to ultraviolet phototherapy treatments (i.e., therapeutic light exposure for specific inflammatory skin disorders). Our department has a long-standing history of pioneering work using phototherapy. However, as part of our efforts to enhance value, we reconsidered current indications for phototherapy in our clinic. This has resulted in increased efforts to track patients undergoing treatments for diagnoses with partial but incomplete evidence of efficacy by closely monitoring therapeutic indications, interval response to treatment, and clinical outcomes.
Curbing rising procedural costs is a primary concern addressed by ACOs employing global payment systems. Specialists performing procedures will face increasing constraints around procedural volume if clear cost-effectiveness cannot be demonstrated. Several societies have joined forces to create appropriate use criteria for procedures,11 providing guidance for both their members and the broader medical community—including patients, providers, and payers—regarding rational utilization. On a smaller scale, as individual institutions transition to ACOs they can examine local procedural use and identify best practices for cost-effective care while preserving procedural volume needed for trainee education.
Identify and assess interpractitioner variability
Our final approach emerged from discussions of evidence-based best practices within the clinic. In reviewing the procedural charges generated from our department, we found that over 15% of procedures were destruction of benign skin lesions (which may be done for either medical or cosmetic purposes, but the latter is not covered by health insurance). A range of practices existed among physicians, from never performing the service to always doing so at the patient’s request. There was confusion regarding signage in selected clinic rooms stating that specific insurers would not cover benign skin lesion removal; this led some patients and providers to conclude that the process was not covered at all, whereas others inferred that all other insurers would pay for it. We therefore eliminated the signs, which enabled physicians to provide clear information to patients about coverage for the procedure.
Given the large percentage of costs (47%) attributable to procedures, we are currently collecting additional data on individual physician-level trends in procedure frequency. Specialists may be uniquely positioned to both determine the necessity of diagnostic procedures and perform them. By analyzing biopsy rates in our department and offering clinicians information regarding their standing relative to their colleagues, we seek to understand the motivations driving high diagnostic test utilization rates. We posit that applying similar approaches in other specialty settings will help inform clinical decision making, contain costs, and frame local guidelines for appropriate use.
Relationships with referring providers for patient follow-up were another area of high interpractitioner variability. Some clinicians preferred that patients follow up with their primary care providers once their dermatologic issues resolved, but others continued to see such patients for follow-up at regular but lengthy intervals. Repeated follow-up for stable patients likely provides marginal additional quality, but it stymies access for new patients. In ACOs, specialists are incentivized to develop strong plans for interdepartmental communication with referring providers regarding follow-up care. To this end, we have begun to plan more teaching and outreach sessions for primary care providers at our institution. These sessions will focus on prereferral diagnosis and triage as well as ongoing management of stable chronic disease and will be designed to enhance the patient-centeredness of dermatologic care delivery both within and outside our department. Additionally, we maintain active roles in dermatology training for both residents in internal medicine and pediatrics and medical students, helping prepare the next generation of primary care providers for practice in ACOs.12
Involving specialists in health care delivery redesign efforts provides additional perspectives in the critical appraisal of quality and cost-effectiveness in a fixed-resource setting. The action areas we identified in our dermatology department include changes in both clinical decision making and operational practices aimed at enhancing overall quality and value. Examining the long-term cost implications of these changes is a primary focus of continuing analysis.
Our initiatives, combined with enhanced outreach to referring providers and departments, emphasize the coordination necessary in an ACO. These strategies for optimizing specialist care in an ACO can be adapted and used by other specialties. Engaging specialists in such delivery reform efforts targeted at cost-efficiency warrants continued attention given increasing emphasis on system-wide accountability. We continue to evaluate our practices and aim for further innovation to address challenges. Importantly, our efforts have been streamlined with the administrative goals of our institution and physicians organization, critically bolstering buy-in from members of our department. More observational data are needed and will likely emerge in real-time regarding strategies for optimal care delivery in an ACO, particularly as academic health centers attempt to balance clinical innovation with medical education.
Acknowledgments: The authors thank Dr. Timothy Ferris, medical director, and Dr. Sandhya Rao, associate medical director, of the Massachusetts General Physicians Organization for their assistance in data acquisition and project support.
1. Colla CH, Wennberg DE, Meara E, et al. Spending differences associated with the Medicare Physician Group Practice Demonstration. JAMA. 2012;308:1015–1023
2. Tallia AF, Howard J. An academic health center sees both challenges and enabling forces as it creates an accountable care organization. Health Aff (Millwood). 2012;31:2388–2394
3. Kimball AB, Resneck JS Jr. The US dermatology workforce: A specialty remains in shortage. J Am Acad Dermatol. 2008;59:741–745
4. Chernew ME, Mechanic RE, Landon BE, Safran DG. Private–payer innovation in Massachusetts: The “alternative quality contract.” Health Aff (Millwood). 2011;30:51–61
6. Bijl D, Van Sonderen E, Haaijer-Ruskamp FM. Prescription changes and drug costs at the interface between primary and specialist care. Eur J Clin Pharmacol. 1998;54:333–336
7. Beovic B, Kreft S, Seme K, Cizman M. The impact of total control of antibiotic prescribing by infectious disease specialist on antibiotic consumption and cost. J Chemother. 2009;21:46–51
8. Heaton E, Levender MM, Feldman SR. Timing of office visits can be a powerful tool to improve adherence in the treatment of dermatologic conditions. J Dermatolog Treat. 2013;24:82–88
9. Shah A, Yentzer BA, Feldman SR. Timing of return office visit affects adherence to topical treatment in patients with atopic dermatitis: An analysis of 5 studies. Cutis. 2013;91:105–107
10. Duffy KL, Mann DJ, Petronic-Rosic V, Shea CR. Clinical decision making based on histopathologic grading and margin status of dysplastic nevi. Arch Dermatol. 2012;148:259–260
11. Connolly SM, Baker DR, Coldiron BM, et al.Ad Hoc Task Force. AAD/ACMS/ASDSA/ASMS 2012 appropriate use criteria for Mohs micrographic surgery: A report of the American Academy of Dermatology, American College of Mohs Surgery, American Society for Dermatologic Surgery Association, and the American Society for Mohs Surgery. J Am Acad Dermatol. 2012;67:531–550
12. Garr DR, Margalit R, Jameton A, Cerra FB. Commentary: Educating the present and future health care workforce to provide care to populations. Acad Med. 2012;87:1159–1160