Our study is entitled The Economic Value of an On-Call Neurosurgical Resident Physician. Our disclosures, no financial or material support is accepted as part of the study and none of the authors have any financial relationships to disclose. The objective of our study was to identify the economic value of on-call services provided by neurosurgical residents. Some background, in 2013 the estimated median GME cost per resident across teaching hospitals annually was approximately $135,000 with annual payouts of approximately $15 billion. Overall, the average resident physician's salary is $61,000 with an average neurosurgery resident's salary of $64,000 annually. There are direct and indirect costs involved in resident training. The direct cost of training a neurosurgical resident is estimated to be $172,000 per year or $1.2 million over the course of a 7-year residency. Indirect costs, known as the teaching effect, include increases in OR time and a slight increase in hospital length of stay and post-operative infections. Our institution, the time spent on primary calls taken during the PGY-2 through 4 years in a de-escalated scale with a tertiary and secondary backup system filled by the PGY-4 through PGY-7. In our study, the term on-call refers to either an overnight weekday 14-hour shift or a 24-hour weekend or holiday shift. While on primary call, the neurosurgical resident is responsible for seeing consults and admissions, operating and providing critical care for neurosurgical patients at each of our four downtown hospitals. The study period was July 1, 2014 through June 30, 2016. The primary author prospectively kept a call log each time he was on call and recorded the following data. In our study, intervention was defined as a patient undergoing a major neurosurgical operation or placement of an EVD or ICP monitor, but we did not include minor procedures such as lumbar punctures or VP shunt taps. We used standard statistical analysis in our study. Billable activity by the residents while on call was then matched with the CPT codes and then we used the 2014 MGMA data to determine the nationwide median reimbursement for neurosurgeons. We used the median reimbursement of $84.37 per work RVU in our data analysis. We used the following CPT codes for describing indirectly supervised procedures and bedside procedures were typically done independently by the on-call resident. For directly supervised operations, we only used the primary CPT code in our calculations. Emergent operations were performed under direct supervision, meaning that the attending neurosurgeon on call was physically present with the junior resident functioning at least as a first assist if not doing majority of the case. Given that the attending surgeon must be present for the critical portions of the operation, we elected to qualify the activity of the on-call resident neurosurgeon during these operations as an assistant surgeon. By qualifying this, we used the CPT-80 modifier. By using this modifier, an assistant surgeon is able to bill for 16% of the total cost of the operation. It's important to remember that these data only represent activities performed while on call and do not reflect any potential billable activities performed during regular working non-pager hours. You can see here in table one, the call volume for the primary author during the two-year period, and the primary author evaluated 1,929 new patients while on call during the study period. And of these 1,929 consults and admissions, 17.1% resulted in emergent or urgent neurosurgical intervention. You can see here in table two, we summarize the economic productivity of a single on-call neurosurgical resident, and this describes the indirect supervision activity. You see here over the two-year study period, we found the productivity to be 7,000 work RVUs or 3,500 work RVUs annually. This equates to approximately $300,000 annually from a single neurosurgical resident on call. We then extrapolated the data to estimate the economic productivity of our entire resident cohort. You can see the results here. Over the two-year study period, we estimated from indirectly supervised procedures, the entire resident cohort could produce 34,000 work RVUs or 17,000 work RVUs annually. This equated to approximately $1.4 million in annual reimbursement for our entire resident cohort while on call from indirectly supervised procedures. This table shows the single resident direct supervision on-call activities. As you can see there on the left, we describe the procedure or operation performed. Over the two-year study period, this resulted in approximately 7,000 work RVUs or using the 16% assistant rate, approximately 1,100 work RVUs. This results in an annual 3,500 work RVUs or using the 16% assistant rate, 560 work RVUs which equates to approximately $47,000. This slide shows the direct supervision activities for our entire resident cohort with our extrapolated data. This shows over the two-year period, the total work RVUs from the directly supervised procedures results in 34,000 work RVUs or at the 16% assistant rate, 5,400 work RVUs. Annually this amounts to approximately 17,000 work RVUs or using the 16% assistant rate, approximately 2,700 work RVUs. Annually this accounts for approximately $231,000. So the total on-call economic value calculations from our single neurosurgical resident and also extrapolating our data to our entire resident cohort is seen here. To get these numbers, we combine the total number of work RVUs from admissions, consultations and procedures performed under direct and indirect supervision. This gives us the theoretical economic productivity of a neurosurgical resident in our program during the on-call shifts over the two-year study period. As you can see here, the single resident indirect and direct on-call activity results in the annual work RVU output of approximately 4,100 work RVUs with an estimated economic value of approximately $345,000. Extrapolating this data to our entire resident cohort results in an annual work RVU output of approximately 20,000 work RVUs or approximately $1.67 million. Previous studies have examined the potential economic value of residents and other specialties, but none have appraised neurosurgical residents. A study of plastic surgery residents over a one-year period looked an entire resident consultation and procedure service at a large academic medical center. They found that collectively 10,287 work RVUs could be produced. An orthopedic study looked at 33 residents at four academic medical centers and their consultations and procedure services over a three-month period. They found that over the three-month period, 9,142 work RVUs could be produced. A study of general surgery residents at an academic medical center, 24 residents assisting at surgery, performing minor procedures and consultations, and patient management over a three-month period, they estimated that approximately $71,000 to $95,000 per resident productivity could be produced annually. A pediatric study looked at a three-month study period of pediatric residents in outpatient clinics and they extrapolated their data to estimate each pediatric resident could produce approximately 725 work RVUs annually. The value of resident call coverage is different depending on the perspective. From attending physicians, it's the convenience of not being in-house during non-working hours, as well as increasing operational efficiency by allowing him or her to be involved in higher yield activities. From a hospital's perspective, it's the difference in the compensation the hospital must pay for the resident physician versus the amount the hospital must pay for another attending physician or a physician extender. It's important to note that non-physician providers typically earn a salary that's double that of resident physicians and they work about half the number of hours. Thus, hospitals may need to hire four non-physician providers to replace a single resident. Another beneficiary are insurance companies. Attending physicians are prohibited from billing CMS or private insurance companies for evaluations and procedures that were performed by residents if the attending physician was not physically present. Therefore, privately insured patients receive the benefit of the care delivered to them by resident physicians at no expense to the insurance company. The number of GME resident physicians supported by public funding has been capped by the Balanced Budget Act in 1997. The 2013 RAND research report found that over half of hospitals are now over their limit of Medicare-allocated resident positions. And as the need of the public for quality health care continues to grow, this cap puts hospitals and training programs under increasing pressure to find funding for these additional positions. So, some possible adjuvants to the limited GME funding are allowing resident physicians to bill the 16% of the surgical fee as an assistant surgeon using the AD modifier that we showed in our study. This could possibly cover the gap in GME funding. Or allowing resident physicians to bill the 13.6% of the surgical fee as an assistant at surgery, such as NPs and PAs are currently allowed to do. Limitations of our study include that this study results from a single neurosurgical resident qualifying his on-call experience, and thus the data is limited to one resident, although it's reasonable to assume that residents at the same institution would have similar experiences. However, our results might not reflect the call composition and schedule of residents at other institutions, therefore, they may not apply to other programs. These results also describe the economic value of a neurosurgical resident physician and are not journalizable to other specialties. The study also takes into account services performed while on-call and does not include the day-to-day activities, such as rounding on patients, outpatient clinics, assisting in scheduled operations during the resident's regular business hours. Our use of the AD modifier as a proxy for the theoretical work overview calculations probably underestimates the actual work value of the resident physician in these surgical cases, because in many cases, the on-call resident does much of the operation with the limited involvement from the attending. Thus, our data significantly underestimates the true overall work and economic value of a resident physician. So, in conclusion, we've shown that neurosurgical resident physicians are very valuable in producing economic value while on-call to far exceed the estimated cost of their postgraduate education. This is even using conservative estimates. This information should be considered when determining future GME funding and in estimations of the cost of resident education.

economic value

on-call neurosurgical residents

indirect cost

economic productivity

GME funding