Shared medical appointment (SMA) census per hour vs mean provider census (MPC) per hour for regular clinic appointments (RCAs) by department. Data are presented as mean (SEM). *P < .001.
Relationship between mean shared medical appointment (SMA) census per hour and mean provider census (MPC) per hour for regular clinic appointments (RCAs). Data are presented as a trend line of the means.
Difference in profit per hour between mean provider census (MPC) for regular clinic appointments (RCAs) and mean shared medical appointment (SMA) census by department. Data are presented as means and 95% confidence intervals. *P < .001 (95% confidence interval >0.00).
Sidorsky T, Huang Z, Dinulos JGH. A Business Case for Shared Medical Appointments in DermatologyImproving Access and the Bottom Line. Arch Dermatol. 2010;146(4):374-381. doi:10.1001/archdermatol.2010.32
To evaluate the economic viability of shared medical appointments (SMAs) in dermatology. Secondary objectives include a comparison of the hourly adjusted census levels generated by SMAs compared with regular clinic appointments (RCAs), as well as a comparison between the economic viability of dermatology SMAs and SMAs in other fields of medicine.
Outpatient clinics within an academic medical center, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire.
No patient-identifying information was obtained or reported. The SMA census data included 301 SMAs (11 different programs and 5 separate departments), representing 2045 appointments over 16 months. Comparisons between patient groups were based on data from the SMA census and mean provider census (MPC) for RCAs, matched on reason for appointment.
Main Outcome Measures
Hourly adjusted census levels and profit differences (charges less costs) between SMAs and MPC for RCAs.
All individual and departmental SMAs generated significantly higher mean census levels and profits per hour than the respective non-SMA MPC of the health care provider leading the SMA (individual, P < .05; departmental, P < .001). All dermatology SMAs generated significantly greater differences in hourly adjusted census levels and profit in comparisons between SMAs and MPC for RCAs than the respective measures in all other departments (P < .001).
Taken together, the results of this study provide strong evidence to support a business case for SMAs in dermatology as a means of simultaneously improving access, productivity, and the bottom line.
Shared medical appointments (SMAs) are an emerging clinic model designed to simultaneously address issues related to health care access, patient education, and shared decision making, as well as clinical outcomes and increased health care provider (hereinafter, “provider”) productivity. Developed in 1996 by Edward Noffsinger, PhD,1 SMAs afford patients with similar medical illnesses the opportunity to interact with one another in a group setting, while also providing each individual a private physician consultation. Studies have demonstrated that SMAs may improve the quality of patient care; through increased efficiency, patients have easier access to care and, as a result of the model's improved educational design and structured visit schedule, require fewer emergency department and specialty clinic visits.2,3
Many medical specialties in a variety of health care settings throughout the country have begun to develop and implement SMAs. Regardless of the specialty or setting, the same general intentions underpin all SMAs. However, to address the diversity of diagnoses, patient needs, patient populations, and health care resources, 3 distinct (each with individual variations) SMA models have been commonly described: (1) drop-in group medical appointments for follow-up care of established patients, (2) physical SMAs for physical examinations in both the primary and specialty setting, and (3) cooperative health care clinics for primary care and chronic illness.4
Given the rising pressure on providers to bolster both clinical and economic productivity, SMAs are increasingly becoming an attractive option in a variety of specialties and settings. Purporting the clinical value of SMAs, Noffsinger and others have published numerous descriptive articles with suggestions on how to best develop and implement successful SMAs.1,4- 15 However, not all SMAs have unanimously demonstrated economic viability. Indeed, a case study of the Luther Midelfort Mayo System by Christianson and Warrick16 reports specific failure of drop-in group medical appointments to achieve economic viability within a mixed-payer system. Thus, while SMAs have been reported to increase provider productivity,17 there is a need for comprehensive discipline-specific economic evaluation of SMAs.
A few SMA programs have been developed in dermatology clinics throughout the country, but the model is by no means widespread in application. It has often been noted that a key issue in the success and economic viability of SMAs is the ability to consistently fill the SMA to an optimal census level. This in turn, would seem to depend on a number of factors including diagnosis prevalence, applicability of diagnosis to an SMA model, demand for clinical services, the level of accessibility and length of wait time to acquire an appointment, and the capacity of the SMA in terms of medical provider to patient ratios. There is a high prevalence of a number of dermatologic diseases that are well suited for SMAs, such as suspicious nevi or findings from full-body skin checks and acne. Furthermore, there is high demand for dermatologic services for these conditions. Access in dermatology is a growing problem. The mean overall patient wait time to see a dermatologist in the United States is 33 days, and it takes an average of 53 days to see a dermatologist in an academic clinic.18 Therefore, it would seem that outpatient dermatology clinics would be consistently capable of meeting optimal census levels with patients having conditions that fit nicely into the SMA model; outpatient dermatology is uniquely suited to implement SMAs in a way that would be both clinically and economically advantageous.
As such, the purpose of this study was to examine the economic viability of dermatology SMAs both independently and in comparison with SMAs in other fields of medicine. The principles guiding hypotheses are that (1) dermatology SMAs addressing appropriate diagnoses are more economically productive than regularly scheduled appointments for the same diagnoses and (2) on average, SMAs in dermatology are more economically productive than SMAs in other specialties. By gaining a better appreciation for the economic viability of SMAs in dermatology, this study hopes to help contribute to the development of a best-practice business model for SMAs in dermatology, which simultaneously optimizes patient access, satisfaction, outcomes, and the bottom line.
This study was performed at Dartmouth-Hitchcock Medical Center (DHMC) in Lebanon, New Hampshire. DHMC was an early adopter of SMAs; currently 6 different departments run at least 1 SMA. This study only evaluated SMA programs in the dermatology, plastic surgery, gastroenterology, orthopedic surgery, and dental departments, which had been enrolling patients for at least 1 full year. As such, the number of programs reviewed in each of the departments varied. Specifically, while 4 plastic surgery and dermatology programs were evaluated, only 1 program from the gastroenterology, orthopedic surgery, and dental departments were reviewed. Thus, 11 programs were reviewed overall (Table 1). While the majority of SMAs at DHMC are subject to a mixed-payer reimbursement system, both SMA and non-SMA visits for a given diagnosis faced the same payer mix.
Census data for 11 SMA programs from September 2007 through December 2008 were collected from DHMC's electronic medical record system. Collected data included the number of patients who attended each SMA, the dollar amount charged per patient, and the staffing mix and staffing time required for all SMAs. For comparison, mean provider census per hour (MPC per hour) for regular (non-SMA) clinic appointments and the staffing mix required for RCAs were collected for the providers who conducted the SMAs. To avoid a potential selection bias, all intradepartmental comparisons were matched by diagnosis. For example, the mean SMA census per hour for the “Acne” SMA (evaluation of teenagers with acne) was compared with the MPC per hour for patients with acne seen by the same provider in RCAs. Thus, MPC per hour represents a theoretically homogeneous patient census seen in a sequentially scheduled regular clinic structure based on the amount of time a provider allocates to a patient with a specific diagnosis. No patient health identifying information was reviewed in this study.
Patients who call to arrange an appointment for a condition addressed by one of the SMA programs are offered a choice between an SMA and an individual appointment. Secretaries describe the SMA structure and what to expect from the visit. It is assumed that patients who choose to schedule an SMA visit understand the format and consent to such a visit. Shared medical appointments are grouped by diagnosis, and SMA census targets are predetermined by the managing provider. Shared medical appointments begin in a conference room, where patients are shown a film or PowerPoint presentation related to their condition. A nurse then provides education related to the manifestations of their condition and discusses common management strategies. Depending on the specific SMA, the group portion of the visit generally lasts approximately 1 hour. Patients are then seen individually by the physician in a private examination room. The physician may perform any necessary diagnostics, prescribe medication, and arrange for follow-up during this portion of the visit. Generally, more time-consuming procedures are scheduled for a later visit. The total time that a patient spends with the provider and the total time that the provider spends seeing all of the patients vary by SMA (Table 2). For dermatology SMAs, providers generally spend 1 hour seeing patients.
Providers may bill on their time or for the amount of history, examination, or medical decision making they have performed during the visit. To be consistent with Medicare guidelines and avoid any limiting reimbursement policy variations between private payers, billing is only issued for time spent individually with the provider. Patient time spent with auxiliary staff is not billed. If another provider has asked the patient to see the SMA physician, the visit may be billed as a consult (Current Procedural Terminology [CPT] code 99241, 99242, or rarely 99243). Patients who either have never been seen in the department or have not been seen within the last 3 years are billed as a new patient (CPT code 99201, 99202, or rarely 99203). If they have been seen within the last 3 years, the visit is billed as an established patient (CPT code 99212, 99213, or rarely 99214). Although modifiers exist for rural locations, they are not applied to DHMC SMAs.
Based on this coding structure, the dollar amount charged and reimbursed per SMA patient and RCA patient with the same diagnosis were assumed equivalent. Unfortunately, specific staffing wage rates were not available. Medical support staffing hourly wage rates were estimated using national median values.19 While the staffing mix was often different for SMAs compared with RCAs, the wage rates for each staff member were the same regardless of the clinic format (Table 2). Because there was only 1 physician per SMA, and all physicians are paid on a salary basis, physician compensation was considered an identical fixed cost in both SMA and non-SMA clinics and thus did not affect the economic analysis. This was also true for the nurse practitioner who ran the gastroenterology (Hep-C [evaluation of patients with hepatitis C virus]) SMA. Similarly, property, plant, and equipment costs were considered to be identical sunk costs for both the SMA and non-SMA clinics. All other SMA operational components (census, staffing mix, and hours worked) were adjusted on a per-hour basis (SMA census per hour) for consistent comparison with MPC per hour for RCAs. Support staff and physicians often worked differing amounts of time in the SMAs (Table 2).
Descriptive statistical methods were used to evaluate the per-hour differences in patient census levels and economic effectiveness between SMAs and MPC for RCAs for individual programs and departments. Specifically, the P values were calculated for the difference between the 2 means using the t test distribution for groups fewer than 30 and the z statistic for groups of 30 or greater. A multiple regression analysis was also performed to assess the potential explanatory power of certain independent variables for SMA census.
The mean SMA census per hour for all individual and departmental SMAs were significantly greater than the non-SMA MPC per hour of the corresponding provider (P < .01 for all individual and departmental SMAs except Acne, Fresh [evaluation for facial rejuvenation options], and Shake [evaluation of hand pain and carpal tunnel syndrome] [P < .05]) (Table 3 and Figure 1). For dermatology, the mean census-per-hour difference between MPC for RCAs and SMAs was significantly greater than the respective mean census-per-hour difference for all other departments (P < .001) (Table 3). The mean SMA census per hour and non-SMA MPC per hour for dermatology was significantly greater than the mean SMA census per hour and non-SMA MPC per hour for all other departments (P < .001) (Table 4).
A regression model built to assess the relationship between various independent variables and mean SMA census per hour had significant explanatory power, with 49.7% of the variation in mean SMA census per hour explained by the variation in MPC per hour for RCAs (R2 = 0.497; P < .001) (Figure 2). By increasing non-SMA MPC per hour by 1 patient, mean SMA census per hour would be predicted to increase by a mean of 1.86 patients (P < .001; 95% confidence interval [CI], 1.65-2.07) (Table 5). While the model had significant explanatory power, the coefficient for dermatology MPC per hour was not significantly greater than zero (Table 6). As such, the difference in the effect of MPC per hour on mean SMA census per hour between dermatology and the other departments was not significant (Table 5).
All individual and departmental SMAs, as well as all SMAs combined, were significantly more profitable per hour compared with their respective non-SMA MPC per hour (P < .001 for all SMAs except Fresh and Shake [P < .05]) (Table 7 and Figure 3). The mean difference in profit per hour between MPC for RCAs and SMAs for dermatology was significantly greater than the respective mean difference in profit per hour for all other departments (P < .001) (Table 7). Lastly, breakeven analysis demonstrated that the mean reimbursement rate would have to be 10% or less for all SMAs for the difference in profit to equal zero (Table 8).
The results of this study suggest that SMAs may be an economically viable way to improve access and productivity in dermatology. The analysis findings illustrate the ability of individual and departmental SMAs to generate significantly greater mean census levels and profit per hour than that generated by the same provider during RCAs for matched diagnoses. Taking a step further, this study compared the performance of dermatology SMAs with those of other departments, demonstrating that SMAs in dermatology generate significantly greater census levels, census level differences from non-SMA MPC, and differences in profit per hour compared with SMAs in other departments.
Previous studies have noted that one key factor in determining the success of SMAs is the ability of the SMA leader to consistently reach a threshold census target.12 This study found that all of the individual SMAs at DHMC generated significantly higher mean census levels per hour than the respective non-SMA MPC of the provider leading the SMA. This trend was consistent when assessed on the department level, as well as for all SMAs combined. Dermatology SMAs generated significantly higher mean census levels per hour than did SMAs in other departments. It was also found that the difference between mean SMA census per hour and MPC per hour in dermatology was significantly greater than that difference in other departments. The additional finding that MPC per hour in dermatology was significantly greater than MPC per hour in other specialties suggested the possibility that differences in MPC levels between departments may be responsible for the departmental differences between mean SMA per hour and MPC per hour. Regression analysis suggests that MPC per hour explains 49.7% of the variability in mean SMA census per hour and that increasing MPC per hour by 1 patient can be expected to increase the mean SMA census per hour by 1.86 patients (P < .001; 95% CI, 1.65-2.07). When the dermatology MPC per hour was added to the regression model, it was found that the difference in the effect of MPC per hour on mean SMA census per hour between dermatology and the other departments is not significant. These findings suggest that the significantly greater mean SMA census per hour seen in dermatology SMAs was not necessarily due to a significant difference in efficacy of dermatology SMAs but rather due to significantly greater MPC per hour compared with other departments. However, given the regression R2 of 0.497, it is likely that there are other important factors that would help explain the variation in mean SMA census per hour. As such, the hypothesis that diagnosis prevalence, applicability of diagnosis to an SMA model, demand for clinical services, and the level of accessibility and length of wait time to acquire an appointment may be key determinants in explaining why dermatology is able to improve use of SMAs. Further research is needed to explore the precise influence of these variables on SMA census levels in both dermatology and other specialties.
A key objective of this study was to determine the economic viability of SMAs in dermatology and how dermatology SMAs economically compared with SMAs in other departments. It should be noted, however, that SMAs generating significantly greater census levels per hour compared with non-SMA clinic appointments may offer significant benefits in terms of productivity, as defined by relative value units, and improvements in patient access even if they do not generate significantly greater profit.
Overall, the combined difference in profit for all SMAs was significantly greater than MPC for RCAs. All 11 SMA programs generated significantly greater profit per hour than their respective MPC, and all 5 departments demonstrated this capability. In fact, the breakeven analysis findings demonstrate that the mean reimbursement rate would have to be 10% or less for all SMA programs for the difference in profit to equal zero. Similar to departmental comparisons in census levels and differences in census levels, SMAs in dermatology generated significantly more profit than MPC for RCAs, and this profit difference was significantly greater in dermatology than in other departments. It is necessary to discuss the charge and cost structure in SMAs and non-SMA clinic appointments to understand the possible reasons for the differences in added profitability between SMAs and MPC for RCAs.
At DHMC, the cost structure in both SMAs and non-SMAs for all of the departments is fixed. The labor, property, and equipment costs are essentially the same regardless of the patient census, since these requirements are predetermined prior to the clinic appointment and are not significantly adjusted on a variable basis. Furthermore, the specific cost structure in each department SMA and non-SMA were not significantly different in terms of the degree to which the aforementioned resources were needed. However, charges in both SMAs and non-SMA clinic appointments are generated marginally on a per-patient basis. The charge amount is determined by the appointment code and is the same for an SMA patient and non-SMA patient with the same diagnosis and relative appointment complexity. Thus, any differences in the charge per patient between SMAs does not account for the differences in added profitability (SMA profit less MPC profit per hour) between departments. This principle is anecdotally illustrated by the fact that even though the orthopedic surgery SMA, Knee (evaluation of patients for knee replacement), charged significantly more per patient than all dermatology SMAs, the added profitability of the dermatology SMAs was significantly greater than the added profitability of Knee. Given this dynamic, to generate significantly greater profit in an SMA, the difference in SMA census must be large enough to offset the difference in fixed costs—primarily those attributed to support staff labor. Based on the regression results, MPC per hour is a key determinant in increasing mean SMA census per hour, where providers with higher MPC levels are expected to have incrementally larger SMA census levels.
Taken together, the results of this study provide strong evidence to support a business case for SMAs in dermatology. All dermatology SMAs generated significantly greater census levels and were significantly more profitable per hour than their respective non-SMA MPC. Moreover, collectively, dermatology mean SMA census per hour, mean difference between SMA and MPC census per hour, and mean difference between SMA profitability and MPC profitability per hour were significantly greater than the respective measures in all other departments and all hospital SMAs combined. The mean incremental profit per hour for a dermatology SMA was $1005 (95% CI, $829-$1182). During the period for which data was collected for this study, 175 dermatology SMAs were held. As such, the dermatology department realized an estimated mean of $175 953 in additional profit from SMAs over this 16-month period. If the department schedules each of their 4 SMAs once per week, the department would benefit from a mean projected added yearly profit of $209 132. Outside of an academic department or group practice, a dermatologist in a solo private practice, who may only be able to hold a single SMA per week, could earn a mean added yearly profit of $52 283. Again, these values likely depend, in part, on a number of other variables. However, it seems quite possible that many dermatologists—most notably those in private practice who see an average of 142 patients per week18—actually have higher MPC levels than those demonstrated in this study (academic dermatologists see an average of 102 patients per week18) and would thus be expected to realize even larger hourly SMA census numbers and added profitability. In addition, for dermatologists practicing in areas with longer wait times, SMAs may be of even greater added value to patients and would likely reach census targets more consistently.
While the results of this study suggest significant economic benefits of SMAs, this is only 1 component of the total potential added value of SMAs in dermatology and in other fields of medicine. The mean difference in census per hour between SMAs and MPC for RCAs was 7.68 patients. Over the time frame of the data for this study, this amounts to an additional 1344 patient appointments for patients to gain access to dermatologic care. The problem of exceedingly long wait times to schedule an appointment, stemming from limited availability of dermatologists nationwide, requires that access to dermatologic care be improved. Indeed, the study of SMAs in dermatology demonstrates that access, patient care, and the economic bottom line are by no means antithetical; SMAs may be a powerful tool to simultaneously improve each of these important measures.
The study was limited by relatively few data points for some of the SMAs, since it was necessary to compare programs over a similar 1-year period. In the absence of raw data representing true variability, the study was constrained to allocating zero variability to MPC per hour for each provider. In practice, however, because appointments are scheduled for discrete times, any variability not captured in this study is likely minimal. In regard to the economic analysis, because precise wage rates were not available, national median values were used. In addition, while not all of the support staff were necessarily paid on an hourly wage, it was necessary to apply an hourly framework to adjust census and profit differences based on variable labor inputs and costs between the 2 models. Another limitation was that revenue was calculated based on charges rather than the actual reimbursement amount realized from each charge. By doing so, it is possible that some of the SMAs with marginal economic viability may actually not be significantly more profitable in actual dollar amounts received compared with MPC for RCAs. A detailed analysis of the potential influence of payer mix on reimbursement and economic viability was not performed. However, breakeven analysis illustrates that while reimbursement variability may have disparate consequences for various SMAs, reimbursement trends where even the lowest-rate payer, Medicaid, averages more than 10% of charges,20 suggest that these differences would be unlikely to completely eliminate the programs' economic viability. It should also be noted that there may be variability in how federal, state, and private payers comply with visits coded as consults when patients are seen in an SMA. If a provider were to consider implementing an SMA model and wishes to code visits requested by other providers as consults, it is strongly recommend that they first check with local regional payers to confirm their policies related to this issue. Lastly, it is important to note that SMAs may not be an ideal model to manage more complex dermatologic disorders and are likely better suited for less complicated, less time-consuming conditions. Again, MPC is a theoretical, diagnostically homogeneous census, which may not represent the typical mix of a provider's practice. However, the structure of sequentially scheduled RCAs constrains visits for a given diagnosis into the time allotted. Thus, the data based on the theoretical census used in this analysis would likely closely resemble data collected from a diagnostically homogeneous, sequentially scheduled clinic in practice. Importantly, the results of this study imply that by seeing diagnostically appropriate patients in SMAs, dermatologists may be able to not only increase access to care for these conditions but also free up regular clinic time to see additional patients with more complex and time-consuming dermatologic diseases.
Correspondence: James G. H. Dinulos, MD, Section of Dermatology, Dartmouth-Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756 (firstname.lastname@example.org).
Accepted for Publication: September 9, 2009.
Author Contributions: All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Sidorsky and Dinulos. Acquisition of data: Sidorsky and Huang. Analysis and interpretation of data: Sidorsky and Huang. Drafting of the manuscript: Sidorsky and Huang. Critical revision of the manuscript for important intellectual content: Sidorsky and Dinulos. Administrative, technical, and material support: Huang and Dinulos. Study supervision: Sidorsky and Dinulos.
Financial Disclosure: None reported
Additional Contributions: Praveen Kopalle provided assistance with the statistical analysis.