Diagnostic Imaging
November 2002

Practice Management

Effective workforce strategy recruits, retains radiologists

University ties, Web site, and time spent with potential recruits all help in process of finding and keeping staff

Bay Imaging Consultants (BIC) serves 11 hospitals and seven outpatient centers in two counties on the eastern shore of the San Francisco Bay. This area, the birthplace of Kaiser, boasts three medical schools within 75 miles. The reimbursement rates in Oakland, the largest city we serve, are among the lowest in the nation.¹

Despite these challenges, we have historically had the luxury of top-notch trainees knocking on our door, thanks to the popularity of the Northern California region. The area's soaring cost of living, however, has forced many recent graduates to seek employment elsewhere, and the competition for those who wish to stay is intense.

In the 1990s, six preexisting groups gradually merged into BIC, a single physician-owned entity. This consolidation allows contracting over a wide area and sharing of administrative costs, which are provided by a separate physician-owned management company. We hoped our numbers would increase job security as well as market power. We pooled incomes, creating an "index fund" of the local radiology practices. The governance issues of such mega-groups have been discussed in these pages extensively and will not be revisited here.

BIC employs nearly 60 radiologists, most of whom are partners or on a partnership track. With a group this size, we must recruit almost every year just to keep up with retirements. One of the most critical parts of our workforce strategy, therefore, is finding and retaining first-rate radiologists.

As the supply of radiologists began to dwindle in the late '90s, we sought ways to use our size to make the group more attractive to trainees, not to mention current staff who were being tempted by more lucrative offers elsewhere. Early retirements also thinned our ranks, although stock market woes have slowed that trend.

Some of the more obvious steps to increase efficiency, and thereby income, had already been taken. Many of the hospitals had PACS and radiology information system projects under way, but each facility seemed to use a different vendor, making practice-wide pooling of daytime work a challenge. We have been much more successful in linking our outpatient facilities with a BIC-owned teleradiology/PACS. This system was introduced to all our affiliated hospitals as the backbone for night coverage and retrieval of images from offsite imaging centers.

We also established a Web-based system that has become popular among key referring physicians, particularly neurosurgeons, who can review images from office or home. While the image archive is password-protected, the Web site, which can be visited at www.bayimagingconsultants.com, has become an unexpected aid when recruiting. I routinely refer interested trainees to the site to learn more about our practice. All our radiologists are listed, and recruits often find someone they know directly or indirectly.

One potential attraction of a practice-wide PACS is the ability to concentrate unusual cases for subspecialists, but this benefit can be a double-edged sword in community hospitals. Aside from the technical challenges, clinicians still expect the local reading room staff to be available to review cases with them on the spot. Another goal has been to redirect work to the least busy radiologist, but this is difficult with so many hospitals involved. It is only a matter of time, however, until these technical problems are solved, as several excellent articles have suggested.^2-4

SURVIVING THE SHORTAGE

In the meantime, the radiologist shortage is likely to persist. Our practice will not win in a nationwide bidding war; we can only recruit trainees who have already decided that living in our area is worth its high cost. This is a personal choice based on lifestyle and family desires. In interviewing candidates, it has become apparent that focusing on these lifestyle issues might be the best way to attract people to our group. But we first have to get them to come and talk to us.

Local university links have been our best source of leads. The trainees already know the area, and many have made the decision to settle here. We encourage clinical faculty participation by our staff and provide moonlighting opportunities, both of which are great ways to meet trainees. As the pace of private practice becomes more frantic, attending and teaching on a volunteer basis at local training centers is declining, which is counterproductive in searching for new talent.

As a board member of the Margulis Society, the radiology alumni group of the University of California, San Francisco, I have been delighted by the university's response to collaboration with local practices. There is even discussion of having trainees visit nearby groups to see what private practice is like. Some residents may be seduced away from academia, but I suspect those individuals would have had short-lived university careers anyway. Others will not find the grass greener on the community side and will return with renewed commitment to academics. In either case, all are likely to benefit from finding the right fit sooner rather than later.

While university connections may seem irrelevant to groups without nearby training programs, in some remote practices a radiologist flies considerable distances a few times a year to serve as clinical faculty. These connections have provided a pipeline that keeps those groups stocked with excellent radiologists, at a far lower cost than recruiters.

Few groups subsidize clinical faculty work by their radiologists, which is unfortunate, but those that do have been pleasantly surprised. Another advantage of such contact is the chance to learn firsthand about candidates' work habits. Full-time faculty are more likely to give candid recommendations to active clinical faculty than to unknown radiologists who cold-call them. In addition, when trainees ask their professors about employment opportunities, practices with active clinical faculty are most likely to come to mind.

Even if group members cannot spare time to participate, it pays to have them stay in touch with the programs where they trained. There are few centers where at least one of my 60 colleagues does not have a contact. We have made recruiting a group-wide activity by posting on our internal Web site all the good CVs we receive. I often ask my colleague with the fewest degrees of separation from a candidate's training program to check references.

GROUP EFFORT

Even with such tools, mega-group recruiting becomes fragmented if each subdivision works separately. Divisions compete and confusion reigns, as everyone has a different, less-than-up-to-date explanation of the partnership track, buy-in, and other particulars. To counteract this confusion, I was appointed BIC recruitment coordinator, largely because I was out of town when it came up on the agenda.

We posted our positions on the Web sites of the American College of Radiology, RSNA, and other radiology organizations, as well as proprietary sites such as Auntminnie.com and RadWorking.com. All these have a common problem, however. We get a lot of hits by candidates who are only mildly interested in our area. Some trainees use Web sites to electronically send their CV to many practices. One whom I contacted by phone said, "I'm filling out a spread-sheet, so I need to know your average partner's income and years to partnership. And where is Oakland in relation to San Francisco?"

We spent a lot of money giving people free trips to the Bay Area, but the yield was low. Although we no longer pay travel expenses for first interviews, that has not dissuaded serious candidates. Print ads seem passe, as they are high cost, low yield, and often out of date, though they apparently remain de rigueur for filling academic posts.

I was given the equivalent of a paid day off each month during the recruiting season to work on filling our three open positions. This expense to the group represents a fraction of what professional search firms would charge for such an assignment. Headhunters are a costly option that we have considered but not pursued. Their forte seems to be to expose candidates to lesser known areas without local training programs, which often have underappreciated opportunities. Even if a recruiter brings a candidate to a practice, that individual should be researched by someone in the group.

Be nice to your candidates: Wine and dine them, but most important, free people up to interview them. We staff an extra radiologist whenever a candidate comes to interview. A frazzled radiologist who is interrupted by wet reads gives a negative (if accurate) impression of your practice. Let your younger radiologists do the talking; they understand the needs of the applicants. Insist that all relevant staff meet the candidate in person or by phone. Meetings to rank candidates are hopeless if apathetic colleagues claim, "I never really met that person."

The process should not end when the contract is mailed. If an applicant demands a higher income, it often can be offered in return for decreased time off-a senior partner would likely enjoy the extra vacation. Generous moving expenses are a nice tax-free bonus, as are welcoming events. Once the new recruits arrive, recognize that it takes time to learn new PACS, dictation, and coding systems.

We shortened our partnership track and developed a home loan program, which is critical in expensive areas like ours. Early home ownership is key to retention, and I have seen too many excellent young radiologists leave the Bay Area just before making partner (or associate professor), following a discouraging outing with a realtor. While we cannot compete nationally in terms of the income/housing cost ratio, our goal is to be among the top in index for our area, so we can attract and retain those who are eager to live in this region.

WHAT DO RADIOLOGISTS WANT?

What many trainees-and veteran radiologists-value, even more than income, is a practice that is fair and flexible in work assignments, days and evenings off, and vacations. Our way to provide these requirements is to let the marketplace work its magic.

All our radiologists are cross-privileged at our hospitals to allow full market participation. Small groups of radiologists who insist that no others can perform their functions risk becoming cartels. We increasing provide cross-divisional assignments to make better use of subspecialists. Walking occasionally in a partner's shoes is the perfect antidote to the divisional squabbles that are the Achilles heel of mega-groups.

BIC has always welcomed radiologists who prefer to work part-time because of special commitments, such as young children, and semiretired partners, who are a great asset as "plug and play" workers. The latter are already accepted by the medical staff and provide an invaluable bullpen for emergencies. This part-time staff willingly accepts salary adjustments in return for the luxury of choosing when they work.

We have totally transparent work accounting, in which every shift and every radiologist is tracked on a Web-accessible spreadsheet from which our payroll is derived. Our compensation scheme is simple, based on one weekday shift or "X" as the unit of labor. Weekend days have a premium, and each hospital's interventional beeper has an X tracking value, varying with how much work its bearer logged in the final three months of the prior year. X is also given for administrative tasks. We considered relative value unit-based compensation schemes an alternative, but they are fraught by the inequities of reimbursement for high- versus low-tech examinations.

An individual's compensation is linked to the number of X worked. Under this system, partners can elect to work at 80% of the mean, or perhaps 50% in their later years. Those with greater income needs can trade in days off and vacation for extra assignments. Most practices have a "black market" where senior radiologists sell weekends, nights, and other less desirable shifts to junior staff. We created an open "white market" that allows individuals to exercise their preferences for income, free time, and day versus night work.

Allowing the compensation for these shifts to reach the level where volunteers emerge will solve a host of problems. BIC embraced the marketplace model in which each assignment finds its market price. In general, volunteer armies work better, especially for night work.

We abandoned forced work assignments, such as late night shifts, and fixed partnership draws. When we recently experienced a critical shortage of volunteers for the night shift, we increased the pay for the shift by about 15%, rather than drafting partners-some of whom would rather have fled to Canada. Suddenly, a queue of workers formed. We chose to ration these precious shifts instead of ratcheting down the wage.

Under this system, a candidate (with the exception of an interventionalist) can opt out of night work entirely and need not even own a pager. Few new recruits make this choice because after-hours work nets such a premium, but it is an invaluable option for couples with young families. As 45% of entering medical students are women,⁵ practices with inflexible work arrangements will have difficulty recruiting from the pool of young talent. I myself have applied to work 80% with BIC when our first child is born. My wife is a general surgeon and has no intention of slowing down.

My own mother, a cardiologist, trained in Stockholm in the 1940s. During her professional lifetime, a base of 8000 male physicians in Sweden, working 80 hours a week, evolved into a half-female force of triple that number, working 40-hour weeks.^6,7 And unlike the U.S., the population of Sweden has remained relatively stable. Most recent social trends evolved a generation earlier in Scandinavia than in the U.S., and those affecting the medical workforce are unlikely to be an exception.

Luckily, PACS and broadband links are improving. These tools, coupled with our creativity and willingness to give up traditional models will produce effective new workforce strategies for future radiologists.

Dr. Gaensler is a staff radiologist for Bay Imaging Consultants in Berkeley, CA.

References
1. Arenson RL, Burnside ES. Surviving managed care. AJR 1997;169:3-10.
2. Andriole KP, Avrin DE, Yin L, et al. PACS databases and enrichment of the folder manager concept. J Digit Imaging 2000;13(1):3-12.
3. Andriole KP, Avrin DE, Yin L, et al. Relevant priors prefetching algorithm performance for a picture archiving and communications system. J Digit Imaging 2000;13(2suppl 1):73-75.
4. Arenson RL, Andriole KP, Avrin DE, Gould RG. Computers in imaging and health care: now and in the future. J Digit Imaging 2000;13(4)145-156.
5. Bickel J, Clark C, Lawson RM. Women in U.S. academic medicine statistics 2000-2001. American Association of Medical Colleges, Washington, DC.
6. Gaensler EHL, Jonsson E, Neuhauser D. Controlling medical technology in Sweden. In: Banta D, Kemp K, eds. A cross comparison of medical technology assessment in ten countries. Office of Technology Assessment, 1980.
7. Swedish Medical Association, http://www.slf.se/, Publications.