Who’s Overworked and Who’s Underworked among Radiologists Academy Health Annual Meeting June 26-28, 2005 Boston, MA Cristian Meghea, PhD ([email protected])

Who’s Overworked and Who’s Underworked among Radiologists

Academy Health Annual MeetingJune 26-28, 2005

Boston, MA

Cristian Meghea, PhD ([email protected])

Jonathan Sunshine, PhD ([email protected] )

Research Dept., Am. College of Radiology, Reston, VA

Pre-publication information. Please do not cite

1. Context

• Physicians in general– 1980s and 1990s: predicted excess by 2000– At present, and in future: shortage, specialists in

particular

• Radiologists in particular– Past decades: relationship between supply and

demand fluctuated– A few years ago: severe shortage– Recently: situation changed, shortage eased (some

indicators even suggest no shortage)

2. Radiologist shortage: long-term trends

0

1

2

3

4

5

1990 1992 1994 1996 1998 2000 2002 2004

Year

Job

Lis

tin

gs

Per

Jo

b S

eeke

r

Number of job listings per job seeker at Professional Bureau Placement Service of the American College of Radiology during week of annual meeting of Radiological Society of North America, 1990-2004.

3. This paper…

• Provides an update on the radiologist labor market

• Uses innovative measure for shortage/surplus: the extent to which physicians wanted more/less work if income changes proportionately– real world, physician preference measure for

shortage/surplus

• Want more work = surplus; want less work = shortage

• Uses information on workload in addition to radiologist counts

4. Data: 2003 Survey of Radiologists

• American College of Radiology, 2003 Survey of

Radiologists

• Nationally representative random sample of all

radiologists in the U.S.

• Question of interest: how does your workload compare

with the desired one

1. Workload OK

2. Want __% less work and proportionately less income

3. Want __% more work and proportionately more

income

5. Method for analyzing shortage/surplus

• Overall (all radiologists): desired change in workload

• By subgroups (by subspecialty, gender, practice type, and practice location): desired change in workload

• What accounts for the different workload changes desired by radiologists: multivariate regression analysis

Want LESSwork

Work OK

Want MOREwork

ALL

Percentage of radiologists 17 67 16 100

Mean weekly hours54

(0.98)47

(0.48)49

(0.88)49

(0.40)

Mean annual hours 2283

(50.49)1964

(23.03)2064

(43.91)2034

(19.36)

Mean % change sought21

(0.71)NA

26 (1.28)

0.59 (0.46)

Mean annual hour change sought501

(24.69)NA

546 (33.67)

1.59 (10.90)

6. Overall shortage/surplus

• Close overall balance supply/demand?

• Same percentages want less (17%) and more (16%) work

• Overall desired percentage change not different than 0

• Accounting for the hours worked, overall desired workload change not different than 0

7. Overall shortage/surplus

8. Imbalances by subspecialty

Subspecialty Mean annual hoursChange sought

(hours)Change sought

(%)

Body 2150 (60)

49 (32)

2.3

Breast 1797 (53)

- 26 (26)

- 1.4

MRI 2048(84)

27(53)

1.3

Musculoskeletal 2102(82)

- 70(45)

- 3.3

Neuroradiology 2142(59)

35(33)

1.6

Nuclear 2107(78)

46(47)

2.2

Ultrasound 2033(90)

26(72)

1.3

• Imbalances within subspecialties?

• Each subspecialty: average desired workload change not different than 0

9. Imbalances by subspecialty

10. Imbalances, other subgroups

Annual hoursChange sought

(hours)Change

sought (%)

Academic 2229 (46)

83** (36)

3.7

Non-acad, government2013

(108)246** (88)

12

Non-acad, private 2094 (26)

- 52**(15)

- 2.4

Large metro 2038 (25)

22(17)

1.1

Small metro 2088 (30)

-4(17)

- 0.2

Non metro 2056 (53)

- 58**(30)

- 2.8

Women 1859 (42)

- 1(21)

- 0.05

Men 2102 (20)

2(12)

0.1

• Other factors associated with desire for workload change?

• Practice type:• Academic: want to work more (+ 4%)• Government: want to work more (+ 12%)• Private: want to work less (- 2%)

• Location:• Non-metro area: want to work less (- 3%)

11. Imbalances, other subgroups

12. Desired change in workload

Coefficient Standard error

Annual hours -0.09* 0.06

Age 34 and less (vs. 45-54) 168.91** 79.08

Age between 35 and 44 (vs. 45-54) 39.73 32.42

Age between 55 and 64 (vs. 45-54) -74.73** 37.83

Age 65 and over (vs. 45-54) -54.23 68.09

Gender: male (vs. female) 59.70* 36.21

Practice is in small metro area (vs. large) -16.97 30.21

Practice is in non-metro area (vs. large) -84.10* 45.10

Academic practice (vs. private) 151.51*** 51.73

Govt, non-acad. practice (vs. private) 466.82*** 159.01

*: p<0.10; **: p<0.05; ***: p<0.01. Full-time radiologists.

• What accounts for the different workload changes desired by radiologists

– Hours worked: more hour worked smaller increase desired, as expected

– Age: younger radiologists want larger workload increases

– Location: non-metro radiologists want smaller increase

– Practice type: academic and government radiologists want larger increase than private radiologists

13. Desired change in workload

• Overall balance between the demand and supply of radiologists in 2003

• Within this overall balance

– shortages of private radiologists and of radiologists in non-metropolitan areas

– surpluses of academic and government radiologists• Relevance

– Present and projected shortage of physicians, specialists in particular

– The shortage of radiologists eased in the recent years– Innovative measure of shortage/surplus broadly

applicable in the health workforce

14. Summary and conclusions