Evaluating the Impact of Formal Dermatopathology Fellowship Training on Practice Patterns and Characteristics Among Dermatologists Submitting Claims for Medicare Pathology Examinations in 2017

Abstract

Objective

Dermatopathology training is incorporated into dermatology residency programs, yet some individuals choose to pursue additional fellowship training in dermatopathology. We sought to characterize the impact of dermatopathology training on practice patterns of dermatologists performing pathology exams in the Medicare population.

Methods

We conducted a cross-sectional review of the 2017 Medicare Public Use File to identify all residency-trained dermatologists performing microscopic pathology exams in 2017. Formal dermatopathology fellowship training was identified through the American Board of Medical Specialties website.

Results

There were 2,150 dermatologists performing pathology examinations in 2017; of these, 592 (27.5%) had formal dermatopathology fellowship training. Fellowship training was associated with a higher mean annual pathology exam volume. Those without fellowship training were more likely to practice in the West, in non-academic settings, and in non-metropolitan counties.

Limitations

Residency-trained pathologists are not included in the sample. Additionally, some dermatopathology labs may submit claims as an entire organization, rather than at the level of each physician.

Conclusion

Fellowship-trained dermatopathologists perform pathology exams at a higher rate than their non-fellowship-trained colleagues and are concentrated in academic practice settings. The findings support the important role of fellowship-trained dermatopathologists in evaluating a wide array of skin pathologies.

Although dermatopathology is taught during dermatology residency and this training is utilized by some dermatologists to interpret their own pathology specimens in practice,1 other dermatologists pursue advanced training through a dermatopathology fellowship. We sought to identify the impact of dermatopathology training on the practice patterns of dermatologists performing pathology exams, which has not been previously examined at the national level.

METHODS

We conducted a cross-sectional review of the 2017 Medicare Public Use File to identify all residency-trained dermatologists performing microscopic pathology exams in 2017.2 Dermatologists with greater than 10 microscopic pathology exams (Current Procedural Terminology codes 88304, 88305, 88307, or 88309) in 2017 were included in the analysis. For each dermatologist, formal dermatopathology fellowship training was determined through the American Board of Medical Specialties and professional websites.3 Dermatologist characteristics, practice patterns, and mean annual pathology exam volume were assessed. Each clinician’s billing address was used to identify geographic county, which was correlated with American Community Survey data to determine rurality and income. For all billed pathology exams, the published allowed Medicare payment amount was verified against that in the database to evaluate for potential errors, leading to the removal of eight entries.

RESULTS

Among 2,150 dermatologists performing pathology exams, we identified 592 (27.5%) with dermatopathology fellowship training. Compared to those with fellowship training, those without were more likely to have at least 20 or more years of dermatology experience (53.6% vs. 47.1%, p<0.0001) and practice in the West (46.1% vs. 34.1%, p<0.0001), in non-academic settings (99.7% vs 76.8%, p=0.007), and in non-metro counties (6.9% vs. 4.6%, p=0.047). Fellowship training was associated with a significantly higher mean annual pathology exam volume (2,493 vs. 793, p<0.0001) (Table 1).

TABLE 1.

Characteristics of Dermatologists Performing Microscopic Pathology Exams in the Medicare Population, Stratified by Fellowship Training

Dermatologist Subgroup	Reference – All Dermatologists, %	Dermatologists Performing Pathology Exams, n (%)			Mean Annual Aggregate Pathology Volume (SD)
		No DP Fellowship	Formal DP Fellowship	P-value	No DP Fellowship	Formal DP Fellowship	P-value
All Subgroups	100.0	1,558 (72.5%)	592 (27.5%)	–	793 (1,236)	2,493 (3,790)	<0.0001
Dermatologist Sex
Female	47.7	532 (34.2)	212 (35.8)	0.469	496 (846)	2,203 (3,874)	<0.0001
Male	52.3	1,026 (65.8)	380 (64.2)		947 (1,371)	2,656 (3,737)	<0.0001
Professional Degree
Allopathic (M.D.)	93.4	1,429 (93.7)	566 (97.7)	< 0.0001	775 (1,173)	2,515 (3,829)	<0.0001
Osteopathic (D.O.)	6.6	96 (6.3)	13 (2.3)		1,111 (2,026)	2,306 (3,421)	0.2400
Years of Practice Experience
<20	55.0	709 (46.4)	307 (52.9)	0.007	696 (1,210)	2,673 (3,856)	<0.0001
≥20	45.0	820 (53.6)	273 (47.1)		887 (1,254)	2,294 (3,725)	<0.0001
Practice Setting
Non–Academic Setting	85.0	1,531 (99.7)	447 (76.8)	< 0.0001	798 (1,236)	2,617 (4,178)	<0.0001
Academic Hospital Setting	15.0	5 (0.3)	135 (23.2)		73 (74)	2,102 (2,015)	<0.0001
Geographic Region
Northeast	22.6	92 (6.0)	92 (15.5)	< 0.0001	690 (1,300)	2,874 (3,666)	<0.0001
Midwest	19.2	296 (19.0)	121 (20.6)		804 (1,162)	2,541 (4,799)	0.0001
South	35.2	451 (28.9)	175 (29.8)		967 (1,408)	3,034 (4,141)	<0.0001
West	23.0	719 (46.1)	200 (34.1)		692 (1,126)	1,851 (2,610)	<0.0001
County Population Density
Metro	95.0	1,449 (93.1)	564 (95.4)	0.047	775 (1,238)	2,411 (3,352)	< 0.0001
Non–Metro	5.0	107 (6.9)	27 (4.6)		1,041 (1,185)	4,324 (8,903)	0.0668
Dermatologist Shortage* in County
No Shortage	68.3	1,052 (67.6)	406 (69.1)	0.524	702 (1,136)	2,353 (3,481)	< 0.0001
Relative Shortage	31.7	504 (32.4)	182 (30.9)		986 (1,404)	2,845 (4,419)	< 0.0001
Median County Income
Below National Median	40.6	585 (37.7)	224 (38.1)	0.856	1,001 (1,409)	3,139 (4,695)	< 0.0001
Above National Median	59.4	968 (62.3)	364 (61.9)		670 (1,103)	2,115 (3,066)	< 0.0001

The table denotes the personal and practice characteristics of dermatologists performing microscopic pathology exams in the Medicare population in 2017, stratified by presence of formal dermatopathology fellowship training. The table includes dermatologists who billed for pathology exams as individuals; those billing under group organizations were not captured in the dataset. For select variables, frequencies do not sum to the group total due to unavailable data.

^*Dermatologist shortage defined as <4.0 per 100,000 population in county. DP: Dermatopathology; HCPCS: Healthcare Common Procedure Coding System; Path.: Pathology; SD: Standard Deviation

DISCUSSION

The findings demonstrate that a significant number of non-fellowship-trained dermatologists interpret skin pathology specimens in clinical practice, although at lower volumes than their fellowship-trained peers. While the clinical context of this practice is unclear, it may indicate the comprehensiveness of dermatopathology training in dermatology residencies, many of which contain specific dermatopathology rotations.1 Survey data suggest that dermatology residency programs in the Southern United States may have a greater overall volume of dermatopathology training.1 Given that many dermatologists practice near their residency training site,4 this could partially explain the relatively greater mean volume of pathology exams in this region.

Fellowship-trained dermatopathologists have additional formal training and high accuracy in identifying complex non-melanocytic and melanocytic neoplasms.5 The substantially greater pathology exam volume among these individuals, likely through higher referrals, may reflect this extensive experience and demonstrate their important role in evaluating a wide array of skin pathologies.

Non-fellowship-trained dermatologists who submitted pathology claims were more frequently located in the West compared to the Northeast. The explanation for this is unclear but might be related to the relatively high concentration of dermatopathology fellowship programs in the Northeast (~36%) compared to other regions. Presence of fellowship training did not differ by dermatologist density or county income. However, mean pathology exam volumes were overall higher in rural, low-income counties and in those with dermatologist shortages, potentially signaling a need for dermatopathology services in these areas.

This study was unable to identify all physicians performing pathology exams, as residency-trained pathologists are not included in the dermatology sample and because some dermatopathology labs submit claims as an entire organization (see additional Table footnotes). However, the identified sample of fellowship-trained dermatopathologists is representative of the broader dermatopathology workforce in terms of practice setting and gender.6 Additionally, these pathology statistics represent Medicare-associated specimens only, and total volumes are likely higher for many individuals. Despite these shortcomings, the study provides greater characterization of dermatologists performing pathology exams at a national level stratified by fellowship training.