Thanks to the revolution in biologic therapy, the annual number of intravenous infusions at the Massachusetts General Hospital (MGH) rheumatology clinic’s small (two-chair) infusion center increased from 1,247 to 1,856 between 2009 and 2014. Related billings skyrocketed from nearly $16 million to more than $40 million. To understand this major shift, one must pause briefly to appreciate the medical history that led to it and then to examine how MGH is redesigning its care processes to bring these novel therapies to patients. Central to the plan is a collaboration with academic partners at the MIT Sloan School of Management.
The Medical Backstory
In 1980, a rheumatoid arthritis (RA) patient at the MGH rheumatology clinic would have received a weekly in-clinic infusion of gold salts and, occasionally, undergone laboratory monitoring.
In 1995, that same patient would get a handwritten prescription for 6 to 10 tablets of methotrexate and instructions to return for blood work several times a year. By 2010, the patient would probably have received a biologic agent, administered intravenously.
This sea change for patients with many types of rheumatologic disease (not just RA) has been astonishing: arthritic joints relieved, kidneys and eyesight saved, vasculitis flares prevented, all with substantially lower cumulative doses of glucocorticoids and fewer conventional immunosuppressive drugs than patients used to receive.
The Organizational “Front Story”
Medical breakthroughs are not so astonishing, though, if they are not smoothly incorporated into the care-delivery process. At the MGH rheumatology clinic, physicians, nurses, and other staff had to change the way they worked to get the right medication to the right patient. No longer did patients simply leave with a handwritten prescription for methotrexate or prednisone. Post-visit tasks for faculty, fellows, and staff now required prior-authorization paperwork for the use of biologic agents, “peer-to-peer” discussions with payers when applications were refused, and appeals to foundations to cover drug costs when primary payers would not foot the bill. The rheumatology faculty grew increasingly frustrated with delays in scheduling patients for the limited number of infusion chairs, even after patients’ treatments were approved.
The result: Clinical outcomes were vastly improving, but staff morale was sinking.
The MGH rheumatologists took action by tapping into an existing research collaboration between MGH and the MIT Sloan School of Management. This collaboration seeks to achieve systems-level changes in academic medical centers and other large-scale health care delivery systems by using operations research methodologies — an academic term for analyzing and informing decisions in complex organizational systems.
In January 2015, MGH identified the “infused biologics initiative” as a focus area for the collaboration. A special project team was assembled, comprising MGH physicians and staff, plus MIT Sloan graduate students from the Leaders for Global Operations program and postdoctoral fellows. The last three authors of this article led the working group.
Infused biologic therapies are now prescribed in many medical specialties, not just rheumatology. Indeed, nearly every MGH department employs some form of them. But delivery of the medications to patients was inefficient because rapid expansion in this therapeutic area was not paralleled by expansion in system-level capacity. In short, each clinic tackled the logistical burdens in a silo, essentially operating mom-and-pop shops of biologic therapy. Staff frequently took on administrative roles for which they were untrained, ill-equipped, and (consequently) unenthusiastic.
The result: potentially dangerous delays in linking patients to the treatments they needed
Consider the example of patients with granulomatosis with polyangiitis (formerly Wegener’s), who often experience rapid disease flares whereby kidney inflammation can progress quickly to end-stage renal disease or the lungs develop life-threatening hemorrhage. Preventing such events requires swift administration of an infused biologic therapy, rituximab. Although the drug is FDA-approved for this indication, its administration still requires prior authorization, a process frequently lasting days to weeks (and sometimes many staff hours).
Fitting the patient into an ever-tighter infusion schedule often leads to even more delays. Rightly, nervous rheumatologists, worried about delays in addressing serious disease flares, often admitted patients to the hospital or even sent them outside MGH for treatment.
Such scenarios prompted the MGH–MIT working team to seek a data-driven approach to analyzing all adult non-oncology infusion systems across the 10 MGH sites that used them. Data came from interviews with physicians, nurses, and administrators, plus analysis of each clinic’s scheduling and billing data and the hospital’s financial data. This work yielded the following key findings:
Increasing demand. Between 2009 and 2014, MGH realized 22% growth in the number of scheduled infusion services hours and a 34% increase in the number of infusion patients, primarily thanks to more than 50% growth in the numbers of rheumatology and gastroenterology patients receiving infused biologic treatments. (Notably, only a minority of patients in those two specialties receive biologic medications via infusion. Most of the patients self-administer injectable biologics, which showed similar growth in use during this time period.)
Poor patient access and scheduling. Across all clinics, patients’ median wait time to begin infusion of a new medication was 38 days, 28 of which were typically spent waiting for prior authorization from a payer. Surprisingly, although most infusion clinics found it difficult to provide infusion services in a timely manner (waits sometimes lasted 2 months), others operated far below capacity. Indeed, the average usage of infusion chairs across MGH was only 32%.
In addition, a 30% patient-cancellation rate, coupled with lack of a centralized scheduling and referral system, meant that a late cancellation in one clinic could not be filled with a patient from a clinic that had a long waiting list. As a result, patients needing urgent infusion were often hospitalized, accounting for roughly 1,000 inpatient bed-days per year. Many patients with less-urgent infusion needs sought care outside MGH.
Variation in staffing levels. Individual MGH infusion clinics had no economies of scale in nursing and lacked the cross-disciplinary competencies required for cross-coverage. Therefore, nursing absences for any reason — lunch breaks, sick leaves, vacations — crippled clinics’ abilities to schedule and administer infusions. In addition, infusion services siphoned nurses away from their usual duties, leaving clinics bereft of other essential nursing services.
Inefficient prior authorization process. The chief complaint in every clinic, however, was the burden of obtaining prior authorizations, thanks to the complex, time-intensive nature of approvals processes and payers’ ever-changing rules.
Solutions Derived from Operations Research
Using operations research principles, the project team focused on three areas — centralization of resources to achieve economies of scale, smart scheduling algorithms, and robust modeling for predicting resource and performance requirements. The team has now proposed the creation of the Medical Infusion Center (MIC), a centralized infusion service for MGH slated to open in May 2017. The MIC will have three central design elements:
- A single expert group will handle prior authorizations for all clinics. By comparison, a similar centralized prior-authorization unit within the MGH Cancer Center reduced the time to process prior authorizations from 4 weeks to only 3 to 4 days.
- A single, centralized pool of infusion nurses will staff the center. This change will make nurses’ schedules more flexible and accommodate cross-training of nurses on the broad array of medications infused in the MIC, thereby improving clinical expertise.
- Centralized scheduling, using a sophisticated algorithm, will reduce patient wait times and increase access. The algorithm, a modification of one developed for the MGH Cancer Center, seeks to optimize the resources required to deliver infusion therapy, including chairs and staff.
To provide easier access to suburban patients, MGH will split the MIC into two campuses, one at the main Boston hospital and another at MGH West, an existing outpatient satellite 15 miles away. Expanded hours at both sites (12 hours on weekdays plus some weekend hours) are expected to reduce the aggregate number of infusion chairs while accommodating new growth without unnecessary capital investments.
The MIC project at MGH illustrates the potential effectiveness of collaborations between hospitals and academia, combining clinical insights with operations research and analytics expertise to tackle an organizational challenge affecting a broad swath of hospital services.
Care-redesign efforts at large medical centers sometimes threaten stakeholders who fear relinquishing local control. But to most stakeholders affected by the MIC, the upsides for all have been clear: Physicians will be less frustrated with delays in getting treatment to patients. Nurses and clinic staff will be able to focus on their core duties, not prior authorizations and scheduling, which now will be handled by a professional, full-time group. The hospital will no longer lose patients to other medical systems for their infusions and hospitalize others unnecessarily. Most important, patients will be more likely to receive the right treatment when and where they need it.
Disclosure: Dr. Stone has received grant funding and served as a consultant to Roche Group / Genentech, the manufacturers of rituximab. Marshall’s research was performed at MGH, an MIT Leaders for Global Operations partner company.
This article originally appeared in NEJM Catalyst on October 20, 2016.