Results of and Comments on the 2000 Survey ofthe American Association of Academic Chief

Residents in Radiology1

Leo P. Lawler, MD, Jon Fromke, MD, R. Gilbert Jost, MD, Ronald G. Evens, MD

Rationale and Objectives. The American Association of Academic Chief Residents in Radiology (A3CR2) annually sur-veys radiology residency programs on issues related to training. The objective is to highlight national similarities, differ-ences, and trends to help programs establish standards and improve residency training.

Materials and Methods. Questionnaires were mailed to 180 accredited diagnostic radiology residency training programsin the United States. The survey covered the usual general topics and more specific topics considered every 4 years; for2000 the latter were on-call issues and the chief residency year.

Results. Completed surveys were returned from 63 programs (35%). Important findings included increased caseload andcall commitments, especially for smaller programs. Resident salaries appear to have increased more than the consumerprice index. Nonemergent after-hour coverage and teleradiology are now a large part of the resident work practice.Women continue to be underrepresented, with a trend downward. Chief residents are more involved in organizing prepara-tion for board examinations and have greater office facilities and more administrative duties.

Conclusion. This survey provided useful insights. All levels of residency face increased workloads. On-call hours havenot changed, but the work has intensified and the use of teleradiology has increased. Many programs have adopted a“night-float” system, and nonemergent after-hours coverage should be considered in any program evaluation. Continuedvigilance and sustained efforts are required to ensure that radiology is considered as a specialty by both men and women.With increased demands on attending physicians’ time, chief residents may need to take on more administrative responsi-bilities.

Key Words. Education; radiology and radiologists, departmental management.

The American Association of Academic Chief Residentsin Radiology (A3CR2) meets annually with the Associa-tion of University Radiologists (AUR) to discuss a widevariety of issues that pertain to radiologists in training.The A3CR2 includes chief residents from all accreditedprograms, including university, community, and military-

based hospitals. One key contribution of this organizationis the A3CR2 survey, which has been conducted annuallysince 1971 (1), with the database maintained at theMallinckrodt Institute of Radiology in St Louis, Mo.

MATERIALS AND METHODS

Survey Design

A questionnaire was formulated to include general top-ics that are reviewed annually (eg, staffing information,training program size, number of women), specific topicsconsidered every 4 years, and topics requested at theA3CR2 steering committee meeting in late 1999. The spe-cific topics for the 2000 survey were on-call issues and

Acad Radiol 2001; 8:777–781

1 From the Mallinckrodt Institute of Radiology, Washington UniversitySchool of Medicine, St Louis, Mo. Received December 21, 2000; revisionrequested February 14, 2001; revision received March 1; accepted March2. Address correspondence to L.P.L., Russell H. Morgan Department ofRadiology and Radiological Science, Johns Hopkins University, 601 NCaroline St, Rm 3253, Baltimore, MD 21287.

© AUR, 2001

777

Radiologic Education

the chief residency year. To optimize response, a mail-based survey was used, and the questionnaire was simpli-fied to a series of common stem questions leading to clearsets of choices, with minimal requirements for writtenword responses (2).

Data Collection and AnalysisSurveys were mailed to 180 accredited programs in the

United States for the 1999–2000 academic year, includingprograms in university, community, and military hospi-tals. The incumbent chief residents of these programscompleted the forms. Data were tabulated and analyzedwith commercial database software (Filemaker Pro 3.0;Claris Software, Cupertino, Calif). Previous survey infor-mation for 1991–1999 was reviewed and incorporated.For analysis, programs were arbitrarily divided by sizeinto small (�13 residents), medium (13–24 residents),large (25–40 residents), and jumbo (�40 residents).

RESULTS

Survey Response Rates

Sixty-three completed surveys were received from 180programs (35%). The response rate was similar to that of1999 (36%) (Sanjeev Bhalla, oral communication, January2000) and 1998 (32%) (3). The 4-year trends were com-pared (4) with 1996 results from a similar number of pro-grams (65 programs).

Residency Size, Profile, and WorkloadOur pool of respondents represented three jumbo, 17

large, 29 medium-sized, and 14 small programs. This sizedistribution closely reflects the national breakdown byprogram size.

The percentage change in matched positions offered in2000 by surveyed programs had decreased more than 6%for the jumbo programs, with slight decreases elsewhere.The number of fellows per resident had increased byabout 0.1 for all but the medium-sized programs. Thejumbo and large programs had the largest ratio of fellowsto residents (ie, 0.4), with the small and medium-sizedprograms having smaller ratios (ie, 0.2).

All programs demonstrated an increase in the numberof examinations performed per resident in a year, with a53% increase from 7,500 examinations per resident in1996 to 11,500 in 2000. As in 1996, the smallest pro-grams had the highest rate of examinations per resident(n � 15,895), with the remaining programs reportingrates ranging from 10,282 to 12,037 examinations (Fig 1).

Women in Diagnostic Radiology TrainingIn the 2000 survey, women made up one-quarter of

residents, a proportion unchanged from 1996. When indi-vidual years were evaluated, the percentage of women inthe 1st and 2nd years was smaller than that in the moresenior years.

Resident RemunerationThe average salary for a 1st-year resident was $36,900,

increased by $5,900 from 1996 and by $8,500 from 1992,and the average salary for a 4th-year resident was$43,700, increased by $7,700 from 1996 and by $10,700from 1992 (Fig 2). These increases are greater than therise in consumer price index in the same period, from156.9% to 167.8% (l). As in 1996, 71% of programs sur-veyed offered a book and travel allowance, with an aver-age value of $850 in 2000.

Figure 1. Number of examinations performed per resident in1996 (gray bars) and 2000 (black bars), according to programsize; a 53% increase was noted overall.

Figure 2. Salaries for 1st- and 4th-year residents in 1992, 1996,and 2000; we noted a 19% increase in salary for 1st-year resi-dents (▫) and a 21% increase for 4th-year residents (‚).

LAWLER ET AL Academic Radiology, Vol 8, No 8, August 2001

778

On-Call CoverageWe found that most residents (66%) started taking call

6–12 months into residency training, an increase from28% in 1996. This increase was largely at the expense ofcall in the first 6-month block, which had decreased from44% in 1996 to 10% in 2000. There was little change inthe proportion of residents taking call after 1 year (22%in 2000 vs 25% in 1996). The proportion taking no callthroughout residency decreased from 3% in 1996 to 2%in 2000 (Fig 3a).

The average total call commitment for all programswas 161 nights over 4 years of residency, decreased from168 nights in 1996. This was broken down into an in-creased average in-house call (104 vs 95 nights in 1996)and a decreased average beeper call (ie, call from home)(57 vs 73 nights). The average total number of call nightsover 4 years of residency, based on program size, hadincreased for programs classified as large, from 158 to170 nights. Total call coverage for the remaining pro-grams had decreased from 111 to 100 nights for jumboprograms, from 177 to 149 nights for medium-sized pro-grams, and from 184 to 167 nights for small programs.Regarding the day after call, we found that, although thepractice of early dismissal had decreased from 34% to14% of programs, there was a large incremental increasein programs granting the following day off, from 29% to38%, and a decrease in programs requiring a full day ofwork after call, from 15% to 8%. There was a large in-crease in the number of residents for whom the postcallday was not an issue (from 22% to 40%), as their pro-grams had made “call” a separate rotation (Fig 3b). Thepercentage of programs using a “night-float” system had

increased from 42% to 44% since 1996. (Night float wasa dedicated residency rotation whose goal was to provideonly after-hours coverage. In most cases this wouldamount to 2–4 weeks of 12-hour shifts after hours as aservice distinct from elective work.)

We note that 53% of programs perform after-hour non-emergent coverage, particularly for cross-sectional imag-ing. In 33% of programs residents received remunerationfor after-hours nonemergent work distinct from call.

Teleradiology is defined as “the electronic transfer ofradiological images from a radiology site to a distantviewing station where interpretations are made,” accord-ing to the Society for Computed Applications in Radiol-ogy (6; 7, p 32). We found that 73% of programs provideteleradiology on call. In past surveys, programs using tele-radiology in general had increased from 30% in 1990, to42% in 1994, and to 58% in 1998 (3,8).

Chief Residency IssuesAs in 1996, most chief residents are in their final year

of residency, although the percentage of 4th-year chiefresidents decreased from 62% to 58%, with a concomitantincrease for 3rd-year chief residents from 38% to 41%,and an unprecedented 1% representation for 2nd-yearchief residents. We saw little change in the wide range ofresponsibilities for chief residents (Fig 4), although wedid find that 15% more now were responsible to help res-idents review for the board examinations. Of the chiefresidents, 70%–90% were responsible for organization ofcurriculum and schedules for rotations, vacations, andcall, and 50%–65% participated in resident selection, resi-dent teaching, and student teaching. For the first time, we

Figure 3. (a) The schedule for beginning call commitments for residents in 1996 and 2000. (b) The postcall schedule for residents in1996 and 2000. The options for the day after being on call were to work a normal day, to leave early, or to have the day off (excused).N/A indicates that these categories do not apply, as call is a separate rotation.

Academic Radiology, Vol 8, No 8, August 2001 2000 A3CR2 SURVEY

779

surveyed the role of chief residents in organizing socialevents; 55% performed this function (Fig 4).

As to benefits, 73% of chief residents received a bonus(average, $1,475; range, $200–$6,000). Eighty-five per-cent went to an additional meeting, usually the AUR/A3CR2 meeting. There were large increases in the numberof chief residents with their own offices (48% vs 28% in1996) and with administrative duties (79% vs 20%).

DISCUSSION

The programs surveyed in 2000 are slightly differentfrom those in 1996. The size profile and numbers, how-ever, do reflect the national distribution of programs ascited by the Accreditation Council for Graduate MedicalEducation (ACGME) and the National Residency Match-ing Program (NRMP). The ACGME cites 4,009 accred-ited radiology residency training positions for 2000, com-pared with 4,206 in 1996. The NRMP cites a decrease inmatched positions offered for 2000 compared with 1996.The American College of Radiology (ACR) reports 1,2501st-year radiology residents registered in 2000, comparedwith 1,150 in 1995 and 1,400 in 1996. Although the totalnumber of programs has remained similar, the decrease in1st-year residents may represent some redistribution ofresident trainees or simply the differing respondent pool.The ACR study in 1999 revealed smaller planned reduc-tions in program size (9). Although positions filled in dil-agnostic radiology decreased from 1995 to 1998, therewas a 3.3% increase from 1999 to 2000, and 96% of 1st-year residency positions were filled in 1999 (9). If thetrickle-down effect of the robust job market continues,future surveys should demonstrate increased demand forradiology training positions both from medical studentsand postgraduates (10). The ACR reported three residentsapplying per job at the 1996 meeting of the RadiologicalSociety of North America and three jobs per resident atthe 2000 meeting. The large increase in the workload per-formed by residents in training must of course be re-flected at faculty and fellow levels, and we note that resi-dents in smaller programs do more.

The representation of women continues to fall short ofthe 34%–39% representation in medical schools in the1990s (11) and the overall 36% in residencies in 1999(12). Although there are fewer women in diagnostic radi-ology than in internal medicine, obstetrics and gynecol-ogy, and pediatrics, the percentage is still much largerthan the 10% in some of the surgical specialties, such asurology and orthopedics (12). The reasons for the discrep-

ancy are complex and beyond the scope of this article. Inrecent U.S. surveys of academic institutions, no sex dif-ference has been demonstrated at the assistant professorrank, although women are still underrepresented in pro-fessorial and tenured positions (13). Although women inradiology are likely to make more money than those inother specialties, they are also reported to be less likely tohave career satisfaction and more likely to suffer sex-based harassment (l). The ACR notes a narrowing ofthe sex gap among younger radiologists (15), but it re-mains to be seen whether the decreasing number of fe-male residents seen in our study will continue as a trend.

Over a 4-year residency, the number of call hours hasnot greatly changed, and we found more residents goinghome the next day. An increased workload has beennoted for radiologists in general, primarily due to cross-sectional imaging and increases in interventional proce-dures (16). With increased numbers of examinations, theworkload on call seems more intense, even though hourson call have changed little. In smaller programs, eachresident performs more examinations. Although largerprograms perform more examinations, they also havemore faculty and fellows. We saw a slight increase in thenight-float system, which many have found to improvethe quality of emergency coverage (17).

Our study has highlighted many blind spots regardingcurrent on-call issues. A large proportion of programssurveyed are performing nonemergent after-hours cover-age in areas such as magnetic resonance imaging. This

Figure 4. Responsibilities for chief residents in 1996 (white bars)and 2000 (gray bars).

LAWLER ET AL Academic Radiology, Vol 8, No 8, August 2001

780

practice is blurring the margins between on-call and non-call commitments. The use of teleradiology continues toincrease. Teleradiology is efficacious in off-hours cover-age (18), although the accuracy of interpretation varieswith the level of training (19). When teleradiology is usedin house, it may not imply an increase in workload andindeed may facilitate readings, but when it involves cov-erage of new remote sites, it can increase the workload.There is continued increase in the demand for facultycoverage 24 hours a day, 7 days a week, both on site andby means of teleradiology (20), though on-duty obliga-tions remain primarily a resident responsibility (21).

The 2000 survey demonstrated a trend toward perform-ing the chief residency year earlier in residency training.This survey also demonstrated more chief residents in-volved in organizing preparation for board examinationsand social events. More had their own offices and admin-istrative duties. We have the impression that chief resi-dents are becoming more integral in running the residencyprograms and suspect that increased clinical demands onfaculty may be partly responsible for this change in role.

REFERENCES

1. Evens RG. Report on a survey of chief residents in academic depart-ments of radiology. Invest Radiol 1972; 7:61–62.

2. Hall MF. Conducting physician surveys. J Healthc Mater Manage1994; 12:28–31.

3. Ruhs SA, Fromke MF, Evens RG. Results of and comments on the1998 survey of the American Association of Academic Chief Residentsin Radiology. Acad Radiol 1999; 6:706–709.

4. VanWagenen FK, Weidman ER, Duncan JR, Evens RG. Results of the1996 survey of the American Association of Academic Chief Residentsin Radiology. Acad Radiol 1997; 4:132–137.

5. The Consumer Price Index (CPI). Washington, DC: U.S. Department ofLabor, Bureau of Labor Statistics, 2000.

6. Hendee WR, Youker JE. Teleradiology: the maturation of a technology.Appl Radiol 1992; 21:13–16.

7. Gitlin J, Greberman M, Rowberg AH, et al. Understanding PACS: pic-ture archiving and communications systems. Harrisburg, Pa: Societyfor Computer Applications in Radiology, 1992.

8. Friedland JA, Shile PE, Wilson AJ. Radiologists’ use of teleradiology:responses to surveys in 1990 and 1994 (abstr). Radiology 1995;197(P):376.

9. Bushee GR, Sunshine JH, Schepps B. The status of diagnostic radiol-ogy training programs and their graduates 1999. AJR Am J Roentge-nol 2000; 175:963–966.

10. Rogers LF. Feeling the pinch: the radiology job market (editorial). AJRAm J Roentgenol 2000; 175:943.

11. U.S. National Center for Education Statistics. Digest of education sta-tistics. Available at: http://nces.ed.gov/. Accessed November 2000.

12. American Medical Association. Physician characteristics and distribu-tion in the U.S.: 1990-present. Available at: http://www.ama-assn.org/.Accessed November 2000.

13. Vydareny KH, Waldrop SM, Jackson VP, et al. Career advancement ofmen and women in academic radiology: is the playing field level?Acad Radiol 2000; 7:493–501.

14. Frank E, Vydareny K. Characteristics of women radiologists in theUnited States. AJR Am J Roentgenol 1999; 173:531–536.

15. Deitch CH, Sunshine JH, Chan WC, et al. Women in the radiologyprofession: data from a 1995 national survey. AJR Am J Roentgenol1998; 170:263–270.

16. Conoley PM. Productivity of radiologists in 1997: estimates based onanalysis of resource-based relative value units. AJR Am J Roentgenol2000; 175:591–595.

17. Mann FA, Danz PL. The night stalker effect: quality improvements witha dedicated night-call rotation. Invest Radiol 1993; 28:92–96.

18. DeCorato DR, Kagetsu NJ, Ablow RC. Off-hours interpretation of ra-diologic images of patients admitted to the emergency department:efficacy of teleradiology. AJR Am J Roentgenol 1995; 165:1293–1296.

19. Eng J, Mysko WK, Weller GE. Interpretation of emergency departmentradiographs: a comparison of emergency medicine physicians withradiologists, residents with faculty, and film with digital display. AJRAm J Roentgenol 2000; 175:1233–1238.

20. Rogers LF. Heeding the call: radiologists in the ED (emergency de-partment) (editorial). AJR Am J Roentgenol 2000; 175:1213.

21. Budler MW, McGuire CW, Holtz JK. Call obligations in radiologyresidency: a survey. Acad Radiol 1999; 6:535–538.

Academic Radiology, Vol 8, No 8, August 2001 2000 A3CR2 SURVEY

781