ajr%2E05%2E0615

7/24/2019 ajr%2E05%2E0615

1/11

12 AJR:186, January 2006

AJR2006; 186:1222

0361803X/06/186112

American Roentgen Ray Society

M E D IC A L I M A G I N G

A C E N T U R Y

O F

witzunshinetage oftric

ologists

T h e P r a ct i c e o f R a d io l og y O r i gi n a l R e s ea r ch

A Portrait of Pediatric Radiologistsin the United States

Leonard Merewitz1

Jonathan H. Sunshine2,3

Keywords:pediatric imaging, practice of radiology

DOI:10.2214/AJR.05.0615

Received April 15, 2005; accepted after revision

April 19, 2005.

1LAMA Consulting, Chevy Chase, MD 20815-3940

([email protected]).

2

Research Department, American College of Radiology,1891 Preston White Dr., Reston, VA 20191.

Address correspondence to J. H. Sunshine

([email protected]).

3Department of Diagnostic Radiology, Yale University

School of Medicine, New Haven, CT.

OBJECTIVE. In recognition of the importance of pediatric radiology and the apparent

shortage of radiologists in the field, the purpose of this study was to provide an extensive and

detailed portrait of pediatric radiologists, their professional activities, and the practices in

which they work.

MATERIALS AND METHODS. We tabulated data from the American College of Ra-

diologys 2003 Survey of Radiologists, a stratified random sample survey that achieved a 63%

response rate with a total of 1,924 responses. Responses were weighted to make them repre-

sentative of all radiologists in the United States. We compare information about pediatric radi-

ologists with that for other radiologists.

RESULTS. Approximately 3% of radiologists, some 800900 physicians, are pediatric radi-

ologists. Depending on how pediatric radiologist is defined, two thirds to three quarters of them

spend 70% or more of their clinical work time doing pediatric radiology. Unlike other radiolo-

gists, a greater percentage of pediatric radiologists desire a reduction in workload (with a corre-

sponding reduction in income) than desire an increase in workload. Pediatric radiologists who

spend 70% or more of their clinical work time in their field are older than radiologists in general

(average age, 55 vs 51 years), and the fraction of pediatric radiologists younger than 45 years is

lower than for other subspecialists ( 20% vs 37%). Pediatric radiologists are disproportionately

women (one third or more, depending on definition, are women, vs 19% for other subspecialists

and 15% for nonsubspecialists), hospital-based, in academic practices (approximately half vs one

fifth for other subspecialists), and in the main cities of large metropolitan areas.

CONCLUSION. A shortage of pediatric radiologists exists and is likely to intensify. Ac-cess to pediatric radiologists is probably a problem except for children in large metropolitan

areas who connect readily to academic hospitals. Means to overcome these problems need to

be actively sought.

ediatric radiology is an important

subspecialty within radiology and

one particularly concerned about

a shortage of well-qualified per-

sonnel. As well, pediatric radiology suffers a

financial handicap because the very low pay-

ment rates characteristic of Medicaid apply to

a greater proportion of children than to any

other age group.

The American College of Radiology (ACR),as part of its mission of providing important

and useful information to the professions it

serves, periodically conducts large-scale,

multitopic surveys of the members of these

professions and their practices [112].

In response to the concerns of pediatric ra-

diology, the ACRs 2003 Survey of Radiolo-

gists, the most recent of these periodic surveys,

was designed with a special focus on pediatric

radiologists. The surveys special focus on pe-

diatric radiologists consisted of having a de-

sign that included more ways of recognizing

and defining pediatric radiologists than almost

any other field within radiology. This article,

an in-depth portrait of pediatric radiologists, is

one of the first full-scale papers to be produced

from the 2003 survey.

This article first shows how many pediatric

radiologists there are under each of many pos-sible definitions of who is a pediatric radiolo-

gist, the relationship among these definitions

of pediatric radiologists, and the demographics

of pediatric radiologists (Tables 13 and

Table S1). It then details the work activities of

pediatric radiologists, such as types of proce-

dures performed and work hours (Tables 46

and S2S5), and the characteristics of the prac-

tices in which pediatric radiologists work

P

7/24/2019 ajr%2E05%2E0615

2/11

Shortage of Pediatric Radiologists

AJR:186, January 2006 13

(Tables 79 and S6S10). (Tables S1S10 ap-

pear in the electronic supplement to the AJRand provide more detail than the tables in this

hard-copy version.) Throughout, comparisons

with other categories of radiologists are made.

The Society for Pediatric Radiology (SPR)

has conducted a number of in-depth, highly in-

formative surveys of its members and has pub-

lished data from one ([13]; Jennifer Boylan,

personal communication, February 2005). Re-

searchers have also published results of sur-

veys of SPR members [1416]. The advan-

tages of this article relative to those studies are

that it has more recent information, contains

systematic comparisons with other radiolo-

gists, and is based on a systematic sample of allradiologists in the United States, meaning that

it includes radiologists active in the pediatric

field who are not SPR members.

Materials and MethodsData Source and Survey Methods

Data are from the ACRs 2003 Survey of Radiol-

ogists (hereinafter referred to as the 2003 survey).

The survey and its methods have been described in

detail elsewhere [17, 18]. In brief, the 2003 survey

was similar to its predecessor, the ACRs 1995 Sur-

vey of Radiologists and Radiation Oncologists

[15], but it incorporated important improvements

throughout the survey process. These ranged frommore thorough canvassing of all ACR leadership to

identify issues of importance and ascertain priorities

among them, through use of a multifaceted tailored

design method [19] to maximize the response rate,

to use of an expanded and more intensive array of

steps to improve data quality.

The questionnaire for the 2003 survey consisted

of 36 items; many items in turn contained multiple

subitems. Questionnaire items and topics were elic-

ited from two rounds of canvassing ACR physician

and staff leaders, winnowed according to prioritiesindicated by top leadership, and pretested in two

large pretests conducted in autumn 2002, with re-

finements made after each pretest.

The survey sample, a stratified random sample

composed of four strata, was taken primarily from

the American Medical Associations (AMA) Phy-

sician Masterfile [20], a reasonably complete list-

ing of all allopathic physicians in the United

States, whether or not AMA members, but also in-

cluded a sample of osteopathic radiologists ob-

tained from the American Osteopathic College of

Radiology. The sample included residents, fel-

lows, and retirees, not merely posttraining, profes-

sionally active physicians.The survey was administered by our contractor,

the Center for Survey Reasearch, University of Vir-

ginia, by mail between March and August 2003,

with nonrespondents being sent up to four remail-

ings as necessary and other steps taken to boost the

response rate. The response rate achieved was 63%,

with 1,924 usable responses. Responses were

weighted so that the weighted statistics would be

representative of the answers that would have been

received if all physicians in the United States in the

four strata had been surveyed and had responded.

Our leading tool to minim ize data deficiencies

was the designation of the 12 items on the ques-

tionnaire judged most crucial as core questions.When questionnaires were returned, our contrac-

tor checked that these 12 items were indeed an-

swered and made three designated consistency

checks involving them. If any problems were

found with the core items, the Center telephoned

the respondent to obtain the missing response(s)

or to resolve the consistency problems. In addi-

tion, data used in this article have been cleaned

and edited to further minimize deficiencies.

TABLE 1: Number of Pediatric Radiologists

Definition of Pediatric RadiologistUnweighted

No. of Responses % of RadiologistsWeighted

No. of Radiologists

Pediatric radiologist in AMA Masterfile 31 2.3 650

Have pediatric CAQ 46 3.4 900

Did pediatric fellowship 44 3.4 900

Subspecialty society highly important 23 1.7 450

Pediatric is primary specialty 39 3.0 800

Pediatric is secondary specialty 6 0.5 150

Does any pediatric radiology 135 13.1 3,500

Pediatric is 30% of clinical time 33 3.3 900




NoteCAQ = certificate of added qualification.

Definition of Variables

The definition of most variables has been de-

tailed in a previous article [18] and also is apparent

from the Results section and the tables. We detail

here only the definitions of pediatric radiologists

that we used.

To provide the fullest picture of pediatric radiol-ogists, this article contains information on a large

number of definitions of who is a pediatric radiolo-

gist. The definitions are as follows:

Physicians who have listed their main specialty

as pediatric radiology in the AMA Masterfile.

This definition presumably underreports the

number of pediatric radiologists and particularly

omits older physicians in the field because radi-

ology subspecialty designations generally be-

came available in the Masterfile only relatively

recently. Most likely, many pediatric radiolo-

gists who started in the field earlier never

changed their self-designated specialty listing in

the Masterfile. Respondents who reported in the 2003 survey

having a certificate of added qualification

(CAQ) from the American Board of Radiology

in pediatric radiology.

Respondents who reported they did a fellowship

in pediatric radiology.

Respondents who reported that the SPR, the

most prominent subspecialty society in the field,

is one of the two most important professional or-

ganizations to which they belong.

Respondents who reported that they subspecial-

ize to at least a small extent and report that pedi-

atirc radiology is their main or secondary sub-

specialty. Respondents who reported spending any of their

clinical work time performing pediatric radiol-

ogy. Although this is obviously an overly broad

definition of pediatric radiologist, it is a category

of interest.

Respondents who reported that they spend at least

a specified fraction of their clinical work time per-

forming pediatric radiology. Thresholds used for

this criterion were 30%, 40%, 50%, and 70%.

Analysis MethodsAll information presented in the remainder of

this article, unless otherwise noted, is based on

weighted data and refers to physicians from allstrata combined. Where numbers of physicians are

given (in Table 1), an adjustment was made for item

nonresponse that is, for the failure of a limited

percentage of respondents to answer each ques-

tion so that the numbers are, as always, represen-

tative of what the answers would have been if all

physicians of interest in the country had responded.

Reported standard errors (SEs) and tests of

statistical significance are calculated taking into

7/24/2019 ajr%2E05%2E0615

3/11

Merewitz and Sunshine


account not only the weighted nature of the data but

also the complex survey design that is, the fact

that responses come from distinct strata. The SEs

are calculated with the Statistical Analysis System

software procedure surveymeans (SAS Institute).

All data analysis was conducted with SAS soft-

ware, release 9.0 (SAS Institute). Because mostcomparisons are made between pediatric radiolo-

gists and five comparison groups of radiologists

(Tables 3 and S1), we use a two-tailed z-test with a

p value of 0.01 or less as the criterion of statistical

significance, in keeping with the Bonferroni

inequality [21]. In two instances in which visual in-

spection of the data showed relatively large differ-

ences between pediatric radiologists and other radi-

ologists but the differences were not significant at

p 0.01, we applied the more common criterion of

p 0.05; these instances are explicitly noted in the

text. SEs for percentages in the tables can be ap-

proximated by the formula:

where p = percentage / 100 and n = unweighted

number of observations (from Table 1 or 3). The

factor of 1.2 approximates the effects of the com-

plex survey design and item nonresponse.

ResultsNumber, Characteristics, andDemographics of Pediatric Radiologists

Among useful definitions of who is a pe-

diatric radiologist, the highest estimate of

the number of pediatric radiologists, approx-

imately 950 or 3.5% of all radiologists, re-sults from defining as pediatric radiologists

those who reported that pediatric radiology

is their primary or secondary subspecialty

(Table 1). The lowest estimate, approxi-

mately 700 or 2.6% of radiologists, results

from defining pediatric radiologists as those

who spend 70% or more of their clinical

work time in the field. Those who report pe-

diatric radiology is their secondary specialty

are fewapproximately 150 persons, or

0.5% of all radiologists.

Table 2 shows the relationship between

pairs of characteristics of pediatric radiolo-

gists. Each row of the table refers to those ra-diologists with the characteristic listed at the

left of the row, and each cell tells what per-

centage of these radiologists also have the

characteristic listed at the top of the column

in which the cell appears. For example, the

table shows, in its third row, that of those

who did a fellowship in pediatric radiology,

approximately 70% have the CAQ, a similar

percentage spend 30% or more of their clin-

ical work time performing pediatric proce-

dures, and approximately half spend 70% or

more of their clinical work time in the field.

But one sixth are not currently doing any pe-

diatric radiology. Other highlights of the ta-

ble include the following: Of those who re-

ported on the 2003 survey that pediatricradiology was their primary specialty, five

sixths have a CAQ, four fifths did a fellow-

ship in the field, four fifths spend 30% or

more of their clinical work time performing

pediatric procedures, and two thirds spend

70% or more of their clinical work time in

the field. Of those who spend 30% or more

of their clinical work time in pediatric radi-

ology, five sixths have the CAQ, 70% did a

pediatric fellowship, and four fifths spend

70% or more of their clinical work time in

the field. Of the half dozen respondents who

reported pediatric radiology as their second-

ary specialty, only one third reported spend-ing as much as 30% of their clinical work

time in the field and none reported having

done a fellowship, having the CAQ, or

spending 40% or more of their clinical work

time in the field.

Tables 3 and S1 present the demographic

characteristics of pediatric radiologists and

the corresponding demographics of five com-

parison groups of radiologists. Pediatric radi-

ologists in general are, on average, about

5153 years old, not significantly different

from other specialists or radiologists overall.

But those practicing 70% or more of their

time in the field are significantly older (aver-aging 55 years) than other subspecialists.

Also, although more than one third of all ra-

diology subspecialists are younger than 45

years, the fraction is considerably smaller for

pediatric radiologists. Specifically, the per-

centage of pediatric radiologists who are

3544 years old is generally smaller than for

comparison categories, but the difference is

significant only at the level ofp0.05.

More than 40% of radiologists who spend

more than 30% of their clinical work time in

pediatric radiology are women, compared

with half that percentage or less for other

subspecialists and for nonsubspecialized ra-diologists. As is true for radiologists in gen-

eral, more than 95% of pediatric radiologists

are board-certified. The proportion of pedi-

atric radiologists in the Northeast is some-

what lower than the proportion of nonpedi-

atric radiologists, and the proportion of

pediatric radiologists in the Midwest is cor-

respondingly elevated (results significant

only atp0.05).

The Work of Pediatric Radiologists

Pediatric radiologists average performing

approximately four and one half of the seven

major types of procedures (mostly technique

categories) into which we have divided radi-

ology (Tables 4 and S2). This is about the

same as for other radiologists. Pediatric radi-ologists are more likely to perform CT than

other radiologists, more likely to perform ra-

diography or fluoroscopy and (to some ex-

tent) sonography than other subspecialists,

and much less likely to do breast imaging.

Pediatric radiologists work at an average of

two distinct locations versus an average of al-

most three for other subspecialists and for

nonsubspecialists (Tables 5 and S3). On aver-

age, full-time radiologists who are pediatric

specialists report working approximately

5657 hours in a typical full week, not signif-

icantly different from other radiologists. Sim-

ilarly, the percentage of pediatric radiologistswho are part-timers is not significantly differ-

ent from the percentage for other radiologists.

With an average of 25 or fewer vacation days

by a number of definitions of who is a pediat-

ric radiologist, full-time pediatric radiologists

have less vacation than other full-time radiol-

ogists, who average approximately 35 vaca-

tion days. In contrast, both full-time pediatric

radiologists and other full-time radiologists

average approximately 10 days annually for

professional education and society meetings.

By most definitions of who is a pediatric

radiologist, approximately 15% of pediatric

radiologists would like less work, eventhough that means their income would de-

crease proportionately, and a smaller percent-

age would like more work with a correspond-

ing increase in income (Tables 5 and S4). In

contrast, although the difference is not signif-

icant, largely because of the limited number

of pediatric radiologists in our survey, for

other radiologists, the percentages desiring

more and less work are more even. Pediatric

radiologists are like other radiologists in re-

porting, on average, a level of enjoyment of

their work approximately halfway between

enjoy very much and enjoy somewhat.

Virtually all pediatric radiologists reportspending part of their work time in clinical

practice at hospitals, although only about

90% of subspecialists who are not pediatric

radiologists and 86% of nonsubspecialists do

so (Tables 6 and S5). Conversely, only ap-

proximately one fourth of pediatric radiolo-

gists do clinical work at nonhospital sites

compared with half of other subspecialists

and half of nonsubspecialists.

SE 1.2p 1 p( )

n----------------------=

7/24/2019 ajr%2E05%2E0615

4/11

7/24/2019 ajr%2E05%2E0615

5/11

16

AJR:186,January2006

TABLE 4: Major Categories of Procedures Performed by Radiologists

Category

MeanNo. of MajorCategoriesPerformed

% Who Perform

NuclearMedicine

InterventionalRadiology CT MRI Sono

Comparison category

All radiologists 4.8 54.4 44.4 82.5 67.6 78

Subspecialists I: those who subspecialize even slightly 4.7 50.6 47.2 82.2 69.0 74

Subspecialists II: those who spend 50% clinical time in one field 4.0 34.8 45.4 73.0 62.8 6

Subspecialists I except pediatric radiologists 4.7 50.3 47.8 81.5 69.0 74

Nonsubspecialists 5.0 59.9 39.7 83.4 66.6 83

Definition of pediatric radiologist

Has pediatric radiology CAQ 4.8c 52.2 35.1 98.0a,b,c,d,e 69.6 89

Did pediatric radiology fellowship 4.7 52.3 42.4 88.5c 63.5 88

Subspecialty society highly important 4.4 48.6 35.3 95.3a,b,c,d 52.1 9

Pediatric radiology is primary specialty 4.8c 56.4 39.3 96.4a,b,c,d,e 70.1 90

Pediatric radiology is 30% of clinical time 4.3 41.1 33.4 95.8a,b,c,d,e 63.7 84

Pediatric radiology is 50% of clinical time 4.3 42.3 29.3 95.3a,b,c,d 62.8 87

Pediatric radiology is 70% of clinical time 4.5 49.4 34.3 100.0a,b,c,d,e 65.6 90

NoteCAQ = certificate of added qualification.aStatistically significantly different from all radiologists, p< 0.01.b

Statistically significantly different from subspecialists I, p < 0.01.cStatistically significantly different from subspecialists II, p 1mil-

lion),amuch

higherpercentagethanistrue

forothersubsp

ecialists(Tables8andS7).

Likeradiologistsgenerally,abouttwo

thirdsofpedia

tricradiologistsworkinprac-

ticesthatserve

bothhospitalandnonhospital

sites(Tables7

andS8).Therestofpediatric

radiologistsworkinhospital-onlypractices,

whereas,incontrast,some10%ofotherradi-

ologistsworkinnonhospital-onlypractices.

Thepracticesinwhichpediatricradiolo-

gistsworkare

significantlylesslikelytoper-

formbreastimagingthanthepracticesin

whichotherra

diologistswork(Tables9and

S9).Thesetablesrefertothecategoriesof

procedurespe

rformedbythepracticesin

whichradiologistswork,nottotheworkof

theindividualradiologiststhemselves,which

isshowninTables4andS2.

Pediatricrad

iologistsworkinpracticeswith

anaveragesize

of2128,largerthanthesizeof

12fornonsubs

pecialistsbutsimilartothatfor

othersubspecialists(Tables9andS10).De-

pendingonthe

definitionofwhoisapediatric

radiologist,abo

ut7590%ofpediatricradiolo-

gistshavecoveragefromamemberoftheir

groupwhoisinthesamesubspecialtywhen

theyareaway,which,attheupperend,ishigher

thanthepercen

tageforothersubspecialists.

Bymostdefinitions,3040%ofpediatric

radiologistsareinpracticesentirelyownedby

membersofthepractice.Thisislessthanthe

65%forothersubspecialistsandthe69%

characteristicofnonsubspecialists.

Discussion

ComparisonwithOtherInformationSources

Forapproximately25years,surveysof

SPRmembers

havebeenprovidinginforma-

tionaboutpediatricradiologists.Because

thesesurveysi

ncludedonlypediatricradiolo-

gists,theycould,unlikeoursurvey,easilyask

numerousped

iatric-radiology-specificques-

7/24/2019 ajr%2E05%2E0615

6/11

AJR:186,January2006

17

TABLE 5: Work Arrangements and Work Satisfaction of Radiologists

Category

MeanWeeklyHours of

Full-Timers% Part-Timers

MeanVacationDays of

Full-Timers

MeanProfessional

Days ofFull-Timers

MeanNo.of

PracticeLocations

% WhoWant Less

Work

MeLess

Des

Comparison category

All radiologists 52.7 19.7 34.1 9.4 2.9 16.5 2

Subspecialists I: those who subspecialize even slightly 53.0 16.5 33.5 9.8 2.9 14.9 19

Subspecialists II: those who spend 50% clinical time in one field 53.4 18.4 30.3 9.9 2.6 11.7 2Subspecialists I except pediatric radiologists 52.8 15.9 33.9 9.8 2.9 14.8 19

Nonsubspecialists 52.4 23.3 35.3 8.7 2.8 19.1 23


Has pediatric radiology CAQ 56.2 24.4 27.8 10.3 2.0a,b,c,d,e 11.1 18

Did pediatric radiology fellowship 56.3 31.3 24.9a,b,d,e 9.8 2.1a,b,d 15.3 22

Subspecialty society highly important 56.6 17.2 22.9a,b,c,d,e 10.7 2.2 11.5 16

Pediatric radiology is primary specialty 57.8a,e 27.6 26.0e 10.2 2.2d 17.6 20

Pediatric radiology is 30% of clinical time 57.0 21.0 25.8 9.1 2.0a,b,d,e 14.7 20

Pediatric radiology is 50% of clinical time 57.2 23.1 24.2a,b,d,e 9.4 1.8a,b,c,d,e 16.2 20

Pediatric radiology is 70% of clinical time 57.5 23.5 24.0a,b,d,e 9.9 1.8a,b,c,d,e 18.8 20

NoteEnjoyment scores: 2 = enjoy very much, 1 = enjoy somewhat, 0 = neither enjoy nor dislike, 1 = dislike somewhat, 2 = dislike very much. CAQ = certificate of ada

Statistically significantly different from all radiologists, p< 0.01.bStatistically significantly different from subspecialists I, p < 0.01.cStatistically significantly different from subspecialists II, p

7/24/2019 ajr%2E05%2E0615

7/11

18

AJR:186,January2006

TABLE 6: Distribution of Work Time of Radiologists

Category

% of Time in Clinical Practice of Radiology at % of Time in PracticeManagement,

Professional SocietyWork, and so forth

Hospitals Nonhospital Sites

% WhoReport > 0

Mean %for Those

WhoReport > 0

% WhoReport > 0

Mean %for Those

WhoReport > 0

% WhoReport > 0

Mean %for Those

WhoReport > 0

% WRepor

Comparison category

All radiologists 88.8 76.6 49.5 45.9 32.6 12.1 27.5

Subspecialists I: those who subspecialize even slightly 91.0 73.8 47.9 44.1 36.9 12.6 36.6

Subspecialists II: those who spend 50% clinical time in one field 89.1 73.2 39.8 51.6 34.9 13.1 44.9

Subspecialists I except pediatric radiologists 90.5 73.8 49.3 44.8 36.7 12.4 34.9

Nonsubspecialists 86.4 80.6 51.8 48.1 26.1 10.9 14.2


Has pediatric radiology CAQ 97.2a,e 67.0a,e 35.3 36.9 46.4 13.8 68.7a,

Did pediatric radiology fellowship 98.1a,b,c,d,e 73.4 26.4a,b,d,e 35.1 44.5 13.8 59.8a,

Subspecialty society highly important 100.0a,b,c,d,e 67.4e 29.6 30.9 55.0e 11.5 86.4a,

Pediatric radiology is primary specialty 100.0a,b,c,d,e 73.2 26.3a,b,d,e 23.0a,b,c,d,e 39.4 15.3 60.8a,

Pediatric radiology is 30% of clinical time 100.0a,b,c,d,e 70.3 19.2a,b,c,d,e 36.4 40.1 16.2 80.8a,

Pediatric radiology is 50% of clinical time 100.0a,b,c,d,e 69.7 18.1a,b,c,d,e 30.5 44.4 16.2 81.9a,

Pediatric radiology is 70% of clinical time 100.0a,b,c,d,e 66.5e 21.0a,b,d,e 30.5 51.4 16.2 84.2a,

NoteCAQ = certificate of added qualification.aStatistically significantly different from all radiologists, p< 0.01.bStatistically significantly different from subspecialists I, p < 0.01.cStatistically significantly different from subspecialists II, p

7/24/2019 ajr%2E05%2E0615

8/11

AJR:186,January2006

19

TABLE 7: Type and Location of Practices in Which Radiologists Work

Category

Practice Type (%) T

Academic Private Locum Tenens Solo MultispecialtyH

Comparison category

All radiologists 15.5 41.0 3.6 5.0 27.8

Subspecialists I: those who subspecialize even slightly 21.8 39.4 1.4 2.2 28.4

Subspecialists II: those who spend 50% clinical time in one field 33.5 28.6 1.2 2.2 26.7

Subspecialists I except pediatric radiologists 20.4 40.6 1.5 2.4 28.5

Nonsubspecialists 6.3 43.9 6.3 9.1 26.7


Has pediatric radiology CAQ 53.0a,b,c,d,e 21.0a,b,d,e 0.0a,b,d,e 0.0a,b,c,d,e 20.1

Did pediatric radiology fellowship 43.5a,b,d,e 20.6a,b,d,e 2.1e 1.9e 22.6

Subspecialty society highly important 68.4a,b,c,d,e 11.9a,b,c,d,e 0.0a,b,d,e 0.0a,b,c,d,e 19.7

Pediatric radiology is primary specialty 45.5a,b,d,e 18.0a,b,d,e 0.0a,b,d,e 0.0a,b,c,d,e 26.0

Pediatric radiology is 30% of clinical time 60.5a,b,c,d,e 17.9a,b,d,e 0.0a,b,d,e 0.0a,b,c,d,e 12.1a,b,c,d,e

Pediatric radiology is 50% of clinical time 62.5 16.6 0.0 0.0 10.4

Pediatric radiology is 70% of clinical time 68.7a,b,c,d,e 15.5a,b,d,e 0.0a,b,d,e 0.0a,b,c,d,e 12.1a,b,c,d,e

NoteData are based on numbers of radiologists, not practices. CAQ = certificate of added qualification.aStatistically significantly different from all radiologists, p< 0.01.

bStatistically significantly different from subspecialists I, p < 0.01.cStatistically significantly different from subspecialists II, p

7/24/2019 ajr%2E05%2E0615

9/11

20

AJR:186,January2006

TABLE 8: Degree of Urbanness of Practice Location of Radiologists

Category

Large Metropolitan Area Small Metropolitan A

% in Main City % in Suburb % in Main City %

Comparison category

All radiologists 28.9 19.0 29.4

Subspecialists I: those who subspecialize even slightly 35.6 18.8 32.0

Subspecialists II: those who spend 50% clinical time in one field 42.8 17.9 29.9

Subspecialists I except pediatric radiologists 34.7 18.9 31.8

Nonsubspecialists 17.8 18.8 26.3


Has pediatric radiology CAQ 45.1e 16.4 35.8

Did pediatric radiology fellowship 41.6e 16.3 38.1

Subspecialty society highly important 59.6a,e 7.8 32.6

Pediatric radiology is primary specialty 47.0e 19.1 33.8

Pediatric radiology is 30% of clinical time 63.8a,b,d,e 2.9a,b,c,d,e 33.4

Pediatric radiology is 50% of clinical time 63.1a,b,d,e 0.0a,b,c,d,e 36.9

Pediatric radiology is 70% of clinical time 67.9a,b,c,d,e 0.0a,b,c,d,e 32.1

NoteData are based on numbers of radiologists, not practices. CAQ = certificate of added qualification.aStatistically significantly different from all radiologists, p< 0.01.bStatistically significantly different from subspecialists I, p < 0.01.cStatistically significantly different from subspecialists II, p

7/24/2019 ajr%2E05%2E0615

10/11

AJR:186,January2006

21

TABLE 9: Other Characteristics of Practices of Radiologists

Category

Mean No. of MajorCategories of

Procedures Provided bythe Practice

% of Radiologists inPractices Providing

Breast Imaging Mean Practice Size

% of RCov

RadiSub

With

Comparison category

All radiologists 6.3 90.5 18.2

Subspecialists I: those who subspecialize even slightly 6.4 90.0 21.9

Subspecialists II: those who spend 50% clinical time in one field 6.1 84.1 25.2

Subspecialists I except pediatric radiologists 6.4 91.4 21.8

Nonsubspecialists 6.2 91.8 12.5


Has pediatric radiology CAQ 6.2 64.1a,b,c,d,e 21.5e

Did pediatric radiology fellowship 6.3 67.8a,b,d,e 22.2e

Subspecialty society highly important 6.3 52.4a,b,c,d,e 23.1

Pediatric radiology is primary specialty 6.6c,e 65.8a,b,d,e 22.3e

Pediatric radiology is 30% of clinical time 6.4 59.2a,b,c,d,e 28.1e



NoteData are based on numbers of radiologists, not practices. CAQ = certificate of added qualification.aStatistically significantly different from all radiologists, p< 0.01.bStatistically significantly different from subspecialists I, p< 0.01.cStatistically significantly different from subspecialists II, p

7/24/2019 ajr%2E05%2E0615

11/11

Merewitz and Sunshine


gists reporting time spent in any field and the

reported amounts of time spent would have

been larger. This is unlikely to have produced

substantial bias with respect to radiologists

who spend large amounts of time in a field, and

these are the radiologists on whom this study

concentrates. However, the number of radiolo-gists who, in fact, perform small amounts of

pediatric radiology may be twice or more the

number reported [18].

The involvement of representatives of indi-

vidual subspecialties in the design of the sur-

vey was limited and, because our survey was

a survey of all types of radiologists, it neces-

sarily had fewer questions relevant only to pe-

diatric radiology than a survey of only pediat-

ric radiologists would have had.

Conclusions

Our detailed portrait of pediatric radiolo-

gists shows that theirs is a small subspecialtyand that most practitioners of this subspecialty

spend most of their clinical time doing pediat-

ric work. Something of a shortage of radiolo-

gists exists, the shortage is likely to grow worse

over time, and ready remedies to the shortage

are not at hand. Thus, the problem of a shortage

needs concentrated attention. Similarly, the

problem of access to pediatric radiology ser-

vices needs thoughtful action to resolve.

Acknowledgments

Stuart A. Royal, president of the SPR, and

David C. Kushner, chair of the board of the SPR,

provided valuable insights and suggestions andmade SPR information available to us. Jennifer

Boylan, executive director of the SPR, assem-

bled and transmitted to us the studies the SPR

has conducted. We thank all those who re-

sponded to the 2003 Survey of Radiologists. By

contributing the time needed to complete the

questionnaire, they have helped make important

information available to the entire profession.

References

1. Deitch C, Chan W, Sunshine J, Shaffer K. Profile of

U.S. radiologists at mid-decade: overview of find-

ings from the 1995 survey of radiologists.Radiol-

ogy1997; 202:6977

2. Deitch C, Sunshine J, Chan W, Shaffer K. Women

in the radiology profession: data from a 1995 na-tional survey.AJR1998; 170:263270

3. Chan W, Sunshine J, Kunkle L, Shaffer K. Charac-

teristics of radiology groups and of diagnostic radi-

ologists and radiation oncologists in different types

of practices.Radiology1998; 207:443453

4. Crewson P, Sunshine J. Professional satisfaction of

U.S. radiologists during a period of uncertainty.Ra-

diology1999; 213:2:589597

5. Crewson P, Sunshine J. Diagnostic radiologists

subspecialization and fields of practice.AJR2000;

174:5:12031209

6. Hogan C, Sunshine J, Schepps B. Hiring of diag-

nostic radiologists in 1998.AJR2001; 176:307312

7. Sunshine JH, Cypel YS, Schepps B. Diagnostic ra-diologists in 2000: basic characteristics, practices,

and issues related to the radiologist shortage.AJR

2002; 178:291301

8. Cypel Y, Sunshine JH, Schepps B. Radiation oncol-

ogists in 2000: demographic, professional and prac-

tice characteristics. Int J Radiat Oncol Biol Phys

2002; 53:720728

9. Bhargavan M, Sunshine JH. Workload of radiolo-

gists in the United States in 19981999 and trends

since 19951996.AJR2002; 179:11231128

10. Cypel YS, Sunshine JH. Basic characteristics of

radiology practices: results from the American

College of Radiologys 1999 survey.AJR2003;

181:34134911. Cypel Y, Sunshine JH. Diagnostic medical physi-

cists and their clinical activities. JACR 2004;

1:120126

12. Hogan C, Sunshine JH. Financial ratios in diagnos-

tic radiology practices: variability and trends.Radi-

ology 2004; 230:774782

13. Goske MJ, Lebowitz RL, Lieber M, Ablin D, Royal

S. Pediatric radiologists: who we are and what we

do. Pediatr Radiol2000; 30:581586

14. Leonidas JC, McCauley RGC, Faerber BN. Pediat-

ric radiologists in the United States and Canada: in-

volvement with newer imaging modalities.Radiol-

ogy 1981; 138:235237

15. Forman HP, Leonidas JC, Kirks DR. Clinical activ-

ities of pediatric radiologists in the United Statesand Canada: 10-year follow-up. Radiology 1990;

175:127129

16. Forman HP, Traubici J, Covey AM, Kamin DS, Le-

onidas JC, Sunshine JH. Pediatric radiology at the

millennium.Radiology2001; 220:109114

17. Bhargavan M, Sunshine JH. Workload of radi-

ologists in the United States in 2002-2003 and

trends since 1991-1992. Radio logy 2005;

236:920931

18. Sunshine JH, Lewis RS, Bhargavan M. A portrait of

interventional radiologists in the United States.AJR

2006(in press)

19. Dillman DA.Mail and Internet surveys: the tai-

lored design method, 2nd ed. New York, NY:Wiley, 2000:150153

20. American Medical Association, Physician Master-

file, 2003. Wood Dale, IL: Medical Marketing Ser-

vice, 2003

21. Pagano M, Gauvreau K. Principles of biostatistics.

Belmont, CA: Wadsworth Publishing Company,

1993:263264

22. Saketkhoo DD, Sunshine JH, Covey AM, Forman

HP. Findings in 2002 from a help wanted index of

job advertisements.AJR2003; 181:351357

23. Meghea C, Sunshine JH. Whos overworked and

whos underworked among radiologists? an update

on the radiologist shortage. Radiology 2005;

236:93293824. Arenson RL, Lu Y, Elliott SC, Jovais C, Avrin DE.

Measuring the academic radiologists clinical pro-

ductivity: survey results for subspecialty sections.

Acad Radiol2001; 8:524532

25. Arenson RL, Lu Y, Elliott SC, Jovais C, Avrin DE.

Measuring the academic radiologists clinical pro-

ductivity: applying RVU adjustment factors.Acad

Radiol 2001; 8:451453

A data supplement containing Tables S1S10 can be viewed in the online version of the article at

www.ajronline.org. These more detailed tables have been designed to augment the tables in the article.

F O R Y O U R I N F O R M A T I O N

Documents

ajr%2E05%2E0615