Commercial risk will be a critical catalyst of progress – it’s complicated, but is it possible? We think so.

Blog Post

5 keys to a high-performing hospital-employed medical group that won't require a subsidy

October 17, 2017

    Hospitals around the country face unprecedented pressures to conserve resources while still growing market share. We sat down with Dr. Tim Crowley—a key member of the team leading network development efforts for the newly expanded Steward Health Care and former president of Pinnacle Health Medical Group in Harrisburg, Pennsylvania—to discuss operational changes medical groups can make to improve performance.

    Advisory Board: In your experience, what are some of the changes medical group leaders must implement to boost performance, especially financially?

    TC: These are the five most important changes which can give you a medical group that no longer requires a subsidy from the hospital.

    1. Improve access

    This starts with taking control of the provider schedules. The goal should be that 95% of a provider's contracted hours be occupied with direct face-to-face patient encounters. Unfortunately, some groups have literally thousands of distinct scheduling rules that restrict patient access. As a medical group leader you have to implement scheduling standards: a set of rules for appointment types and how they should be used. This is not easy and it's important to secure physician buy-in when it comes to the design. The goal should be to optimize patient access and to fill clinician schedules. Getting this done requires thorough execution and continuous monitoring.

    2. Implement performance-based compensation

    Too often, there's a disconnect between medical group compensation and reimbursement. Linking provider compensation to quality, revenue, and patient experience aligns the hospital and the medical providers. The process of reaching an agreement on a comp plan can be painful, but it doesn't need to be. At our first all provider meeting, I just asked: "Raise your hand if you think you should be paid less for doing the same work as your colleagues." Of course no one raised their hand. Then I said "Raise your hand if you think you should be paid MORE for the same work." When no one raised their hand, I told them, "Great, we just agreed on a comp plan. Market-based pay for market-based work." We created a compensation model that was 80% based on productivity and 20% based on quality measures and patient experience—but the quality metrics came first; they were the "gateway" to earning the productivity bonuses. Without achieving the quality goals you couldn't get the productivity money.

    9 ways to engaging physicians in compensation redesign

    3. Leverage Advanced Clinical Practitioners (ACPs)

    We started with 70 physicians and 30 ACPs, and then we added an army of highly qualified ACPs to our provider ranks. By 2016, we had 75 physicians and 75 ACPs. We gave the ACPs equal status and value as our physician providers which made a huge difference. By changing how we treated the ACPs, we not only improved our performance, we had a conga line of folks eager to join our ranks. It made recruiting much easier.

    4. Invest in improving coding and documentation accuracy

    It's important to give intensive coding and documentation training to all providers. Sometimes that means having trainers come in and do doc-to-doc training. The trainer sits down with a group of four or five physicians, goes through a sample of their charts and their codes, and shows them what they did and what they coded and how that impacted payment. I had the trainers come back two or three times to make sure this got done right and didn't fall off.

    This wasn't cheap—I think the service cost around $500,000—but the ROI on it turned out to be five to one. Training helps avoid under coding, prevents over coding, and improves both collections and compliance.

    5. Utilize available revenue sources

    Finally, there were three critically impactful sources of revenue, which had been neglected before we came in: Adult Wellness Visits (AWVs), Transitional Care Management codes, and Chronic Care Management codes. Each requires a different approach, but they all provided significant revenue improvement.

    We were at about the national average of 6% of the eligible Medicare patients getting an AWV. The payment at the time was $164 per AWV and now with an "end-of-life discussion" the visit pays over $200 per AWV. The downstream revenue for each AWV is estimated between $360 and $500 as a result of tests ordered at the visit, like mammograms, colonoscopies, etc. Groups need to focus their attention on these activities—beyond the obvious revenue enhancements, the outcomes improvement for the panel are tremendous. The Transitional Care Management codes pay $200 for every patient discharged from the hospital who is contacted within two days and seen within seven. The Chronic Care Management codes pay over $500 per patient per year for those patients who are eligible. CMS estimates that around two-thirds of Medicare patients should qualify. Do the math on your Medicare population and you will see that this opportunity alone would be transformative to the economics of the groups.

    Advisory Board: Many medical groups face challenges when it comes to scheduling and access—especially monitoring and tracking patient access. What strategies have you found to be successful on these fronts?

    TC: At my past organization, we looked at scheduling from the perspective of both the provider and patient. On the provider side, because we'd taken control of schedules, we had a team monitoring the percentage of appointments providers attended every day. Every day a provider missed the benchmark, we reached out to determine why. At the same time, we educated the providers on how revenue and collections impacted their pay.

    From the patient viewpoint, we hired a group of professional patient impersonators—our "secret shoppers"—to call the offices and pretend to have various ailments and to ask for various types of appointments. For example, one group would call and say, "I'm new to the area and would like to establish myself with one of your PCPs. When can I get in?" Then we'd catalogue the answers (usually the answer was months). Another type of call went, "I'm a patient of Dr. X, and I've a bad cough and fever and I would really like to be seen today or as soon as possible. When can you squeeze me in?" We were shocked to learn that the wait time for an appointment was often 3 weeks—even though we could see numerous openings in the schedule before then! We fixed that in a hurry.

    Advisory Board: These days, many medical groups are working to centralize back office operations, but it can be costly, and if not done well, may not prove beneficial. What are your views on centralized services, given all the financial pressures medical groups face?

    TC: Many practices have the same amount of staff working in little nooks and crannies in old offices regardless of the number of physicians at that practice. With 25 practices, you get 25 groups of the same people doing the same thing. If you centralize administrative functions like scanning, scheduling, prescriptions refills, and managed care referrals, you can reduce the number of FTE's without negatively impacting patient service. However, we found that although centralizing requests for scanning and referrals can be fairly straightforward, managing a centralized call center can be challenging.

    We learned that an operator could handle about 80 calls a day, but since we didn't have any actual data on the number and types of calls we were getting per day from the old analog system, we dramatically understaffed operators and ended up having a "call dropped rate" of around 80% initially. Eventually, we got the number right and provided the right training. We painfully learned that we couldn't just drop untrained personnel into the call center. But this cost us a lot of credibility with both the physicians and our patients. There were times I wanted to break the whole thing and send all the calls back to the offices, but we stuck with it.

    We learned that calls fall into four basic types and that each type needed a different level of training and experience. The four types were: 

    • "I need an appointment": These went directly to a scheduler
    • "I just got a note from my physician about a clinical result": Calls like these were routed to an MA or nurse
    • "I need a refill of my prescription": We brought pharmacy technicians into the call center to handle these
    • "I have a big red swollen leg and need to know what to do": These went to a nurse triage group in the call center, but that meant we had to hire nurses and equip them with clinical triage algorithms to deal with these calls

    None of this was cheap. The call center cost $800,000 and we ended up with over 50 FTE's there and a half dozen pharmacy techs for the refills. The good news was that ALL of these calls were going somewhere other than the offices–so the providers and staff there could focus on the patients.

    Eventually, through the centralization-generated efficiencies, despite all the challenges, we were able to reduce amount of staff we needed across all our sites. That generated FTE and cost savings and helped us reduce the hospital subsidy from $100,000/provider to zero.


    Register for the national meeting

    Join us for the 2017-18 Medical Group Strategy Council national meeting

    Register Now

    Have a Question?


    Ask our experts a question on any topic in health care by visiting our member portal, AskAdvisory.