Download Consultation in University-Based and Community

391
Consultation in University-Based and Community-Based Infectious Disease
Practices: A Prospective Study
Daniel J. Sexton, G. Ralph Corey, Christopher W. Ingram, Vickie
M. Morris, and Hubert B. Haywood III
Division of Infectious Diseases, Duke University Medical Center,
Durham; and Raleigh Infectious Diseases Associates PA,
Raleigh, North Carolina
Estimates on the current and future manpower requirements for infectious disease consultants vary widely [1-4].
Such approximations have been based more on opinion than
on fact because data that quantify and characterize the actual
work of infectious disease physicians are sparse. Indeed, even
the exact number of clinicians practicing in the subspecialty
of infectious diseases in the United States is unknown.
A questionnaire and telephone survey sponsored by the
Infectious Diseases Society of America in 1987 provided useful information on the numbers, type, and practice patterns
of infectious disease physicians but did not include information on the types of patients seen by these clinicians [5]. Such
data could be useful to persons involved in the training of
infectious disease physicians, to policy planners, and to the
social architects who may design future medical care systems
by bureaucratic or government edicts. Because of a lack of
such data, we undertook a study to examine the rates, timing,
and type of inpatient consultations by infectious disease specialists in two settings: a community-based consultative practice and the consult service of a large university medical
center.
Methods
Clinicians were asked to complete a standard data form for
each inpatient consultation during the period from 1 Sep-
Received 8 April 1994; revised 21 June 1994.
Reprints or correspondence: Dr. Daniel J. Sexton, Duke University Medical Center, Box 3605, Durham, North Carolina 27710.
Clinical Infectious Diseases 1995;20:391-3
© 1995 by The University of Chicago. All rights reserved.
1058-4838/95/2002-0025$02.00
tember 1992 to 31 March 1993. These clinicians included
16 infectious disease physicians at Duke University Medical
Center, which consists of a 1,000-bed university hospital and
a 350-bed university-affiliated Veterans Administration hospital, and three infectious disease physicians from Raleigh
Infectious Diseases Associates who practice at three community hospitals that range in size from 150 to 550 beds. Information was collected concerning the time a consultation was
requested, the type of physician or service requesting a consultation, and known or suspected pathogens, disease processes, and affected organ systems. The data form allowed
physicians to designate multiple pathogens, disease processes, and/or affected organ systems for individual patients.
In addition, data were collected on whether recommendations were made to adjust or initiate antimicrobial therapy as
a result of the consultation, whether the patient had a nosocomial infection, and whether the patient was in an intensive
care unit at the time of consultation.
Patients at the university hospitals were seen by infectious
disease fellows and faculty; patients at the three community
hospitals were seen by one of three infectious disease consultants in a private practice group that was established in 1991.
Differences between proportions were calculated with use of
the x 2 test, and P values of .05 were considered statistically
significant.
Results
Sixteen university-based faculty members completed a total of 859 consultations, whereas the three infectious disease
physicians in private practice completed a total of 507 con-
Downloaded from http://cid.oxfordjournals.org/ at Penn State University (Paterno Lib) on May 12, 2016
Infectious disease physicians in university and community practices completed a standard data
form following each of 1,366 inpatient consultations during a 7-month period. The rate of consultation was higher in the university practice than in the community practice (3.4 vs. 1.8 per 100
discharges, respectively). Known or suspected bacterial pathogens accounted for more than half
of all consultations in both practice groups. The three organ systems most commonly affected by
infection were pulmonary (20% in university practice vs. 19% in community practice), skin and
soft tissue (13% in university practice vs. 20% in community practice), and musculoskeletal (12%
in university practice vs. 16% in community practice). Bloodstream infection, pneumonia, unexplained fever, osteomyelitis, urinary tract infection, and cellulitis were the six most common
disease processes that led to consultation in both practice groups. The percentage of patients with
noninfectious diseases and the percentage for whom a change in antimicrobial therapy was advised was nearly identical in both practice settings. Physicians in private practice performed more
consultations on weekends (20% vs. 11% in university practices, P < .001) and between 6:00 P.M.
and 7:00 A.M. (15% vs. 6% in university practices, P < .001). The scope and diversity of the work
of consultants in community practices are nearly identical to those of their colleagues in university-based practices.
392
Sexton et al.
Table 1. Referring service for consultations in infectious diseases,
by type of practice: 1 September 1992 to 31 March 1993.
No. (%) of consultations
Referring service
177
101
97
81
65
43
32
31
36
23
27
24
51
71
(21)
(12)
(11)
(9)
(7)
(5)
(4)
(4)
(4)
(3)
(3)
(3)
(6)
(8)
Community practice
176 (35)
39 (8)
17 (3)
42 (8)
33 (7)
31 (6)
23 (5)
24 (3)
9 (2)
30 (1)
31 (4)
10 (<1)
28 (6)
50 (10)
sultations during the same 7-month period. The rate of consultations in university practice was 3.4/100 discharges; the
consultation rate in private practice was 1.8/100 discharges.
For the university-based specialists, consultations were requested by a total of 23 different services or subspecialties;
for the private practice physicians, consultations were requested by a total of 19 different services or subspecialties.
For the university-based group, 47% of their consultations
were by referral from general internists and medical subspecialists and 40% were by referral from general surgeons and
surgical subspecialists. The private group saw more patients
by referral from general internists and medical subspecialists
(63%) and fewer patients from general surgeons and surgical
subspecialists (28%) (table 1). A similar percentage of patients in both groups were in intensive care units at the time
of consultation ( 1 8% in university practice vs. 20% in private
practice).
Infectious disease physicians in private practice performed
more consultations on weekends than university-based specialists (20% vs. 1 1 %, respectively, P < .001). There was also
a significant difference between the two groups in the percentage of consultations requested between 6 P.M. and 7
A.M. ( 1 5% in private practice vs. 6% in university practice, P
< .001).
A microbiological diagnosis had been established at the
time of consultation for 41% of patients seen by the private
group and for 38% of patients seen by the university group. A
total of 369 pathogens were identified or suspected in patients seen by private practice specialists; 662 pathogens
were identified or suspected by university-based consultants
(table 2). Staphylococcus aureus was the most common pathogen identified or suspected in both patient groups. The frequency of other pathogens identified was similar among both
groups with only one relevant exception: more inpatients
seen by university-based consultants were infected with the
human immunodeficiency virus (HIV) (9% vs. 2% in private
practice, P < .05). However, during the study period, a total
of 28 patients with HIV infection were admitted by the infectious disease physicians in private practice.
Patients from both clinical settings had infectious diseases
that involved a wide array of organ systems. The most commonly involved organ systems in patients seen by university
vs. private practice physicians were pulmonary (20% vs.
19%), skin and soft tissue (13% vs. 20%), and musculoskeletal
(12% vs. 1 6%), respectively.
Bloodstream infection, pneumonia, unexplained fever, osteomyelitis, urinary tract infection, and cellulitis were the six
most common disease processes that led to consultation in
both practice groups (table 3). Slightly more patients seen by
university-based consultants had nosocomial infections (33%
vs. 26% in private practice).
Known or suspected bacterial pathogens accounted for
more than half of all consultations in both practice groups
(55% in university practice vs. 62% in private practice). Some-
Table 2. Categories of disease and known or suspected pathogens
found in patients seen in infectious disease consultations, by type
of practice: 1 September 1 992 to 31 March 1993.
No. (%) of consultations
Category of disease or pathogen
Category of disease
Bacterial
Viral
Fungal
Mycobacterial
Protozoan*
Noninfectious disease
Pathogen
Staphylococcus aureus
HIV
Candida
Enterococci
Coagulase-negative staphylococci
Streptococci
Anaerobes
Pseudomonas
Escherichia coli
Mycobacterium tuberculosis
Nontuberculous mycobacteria
Klebsiella
Enterobacter
Pneumocystis carinii
Other bacteria
CMV/HSV/VZV
Other fungi
University
practice
Community
practice
473 (55)
99 (12)
73 (9)
38 (4)
16 (2)
120 (14)
313 (62)
32 (6)
37 (7)
15 (3)
7 (1)
45 (9)
113 (15)
79 (9)
43 (5)
42 (5)
41 (5)
38 (4)
31 (4)
32 (4)
32 (4)
17 (2)
19 (2)
19 (2)
15 (2)
12(1)
78 (9)
23 (3)
28 (3)
75 (13)
8 (2)
23 (5)
31 (6)
34 (7)
45 (9)
24 (5)
30 (6)
15 (3)
12 (2)
1 (<1)
8 (2)
7 (1)
3 (<1)
2 (6)
12 (2)
9 (2)
NOTE. HIV = human immunodeficiency virus; CMV = cytomegalovirus; HSV = herpes simplex virus; VZV = varicella zoster virus.
* Inclues P. carinii.
Downloaded from http://cid.oxfordjournals.org/ at Penn State University (Paterno Lib) on May 12, 2016
General medicine
Orthopedics
General surgery
Cardiology
Cardiovascular surgery
Family medicine
Neurosurgery
Hematology-oncology
Pulmonary
Obstetrics-gynecology
Neurology
Psychiatry
Other surgical subspecialties
Other medical subspecialties
University practice
CID 1995;20 (February)
Consultation in Infectious Disease Practices CID 1995;20 (February)
Table 3. Known or suspected disease processes in patients seen in
infectious disease consultations, by type of practice: 1 September
1992 to 31 March 1993.
No. (%) of consultations
Disease process
140
131
129
83
81
80
76
66
58
59
47
94
(16)
(15)
(15)
(10)
(9)
(9)
(9)
(8)
(7)
(7)
(5)
(11)
Community practice
79 (16)
87 (17)
128 (25)
36 (7)
57 (11)
26 (5)
31 (7)
51 (10)
50 (10)
8 (2)
19 (4)
30 (6)
what surprisingly, 14% of university consultations and 9% of
private practice consultations were for noninfectious diseases (such as drug fever, rheumatologic disorders, noninfectious pulmonary diseases, or pretransplantation evaluations).
Infectious disease specialists in private practice recommended a change in antimicrobial therapy in 66% of all consultations; a change in therapy was advised for 59% of patients seen by university consultants.
Discussion
We were surprised to find so many similarities between the
inpatient consultation practices of infectious disease specialists in university-based practice and those of specialists in
private practice. The only remarkable differences found between the two groups were in the percentages of patients
with HIV infection and in the frequency of night and weekend requests for consultation.
In an earlier report, one of the authors noted that the rate
of infectious disease consultation was higher in a private
practice group in Oklahoma City (3.1 consults per 100 discharges) than in the university-based practice at Duke University Medical Center (2.3 consults per 100 discharges) [6].
Opposite findings were noted in our study. The private group
practice surveyed in the current report was established in
1991; thus, its rate of inpatient consultation may currently
be lower than that of older, more established private practices in infectious diseases. Furthermore, the rate of inpatient
infectious disease consultations at Duke University Medical
Center has increased by 48% since 1987.
Most patients seen in consultations by each group had
bacterial disease, particularly infections with gram-positive
cocci. A microbiological diagnosis had been established at
the time of consultation for approximately 40% of all patients. These findings suggest that infectious disease physicians are consulted almost as frequently for their expertise in
therapy design as for their diagnostic skill. Indeed, physicians
from both groups recommended a change in existing therapy
in almost two-thirds of consultations.
These data have implications for directors of training programs in infectious diseases. Fellowship programs in infectious diseases should prepare trainees to assist in the management of a broad array of subspecialty-related infections;
the programs should also provide in-depth education on the
problems most often seen by infectious disease consultants
such as skin, soft tissue, musculoskeletal, and pulmonary infections. As almost one-third of all consultations were directly or indirectly related to nosocomial infections, training
in the principles and management of hospital-acquired infections is important. Since surgical consultations represented
28%-40% of all consultations, thorough education about the
principles of surgical infections is necessary. Finally, as
—10% of consultations were for noninfectious diseases, a
wide knowledge of general internal medicine is mandatory
for infectious disease clinicians.
Our study did not examine the outcome of the consultation process. Further studies that examine how the specialized skills of infectious disease specialists impact patient care
and hospital outcomes are needed. Although such studies are
difficult to design and undertake, they could be important to
the continuing success of clinical infectious disease practices
in both university medical centers and the community.
Our data illustrate that the scope and diversity of the work
of infectious disease specialists in community-based practice
are nearly identical to those of their colleagues in universitybased practice. Although future changes in reimbursement
policies and managed care contracts may substantially alter
the way all subspecialists practice, infectious disease clinicians currently have diverse and challenging work that requires a broad knowledge of infectious diseases and frequent
interaction with numerous other subspecialists and clinicians.
References
1. Beeson PB. Too many specialists, too few generalists. Pharos 1991; 54:
2-6.
2. Ervin FR. The bell tolls for the infectious diseases clinician. J Infect Dis
1986; 153:183-5.
3. Eickhoff TC. Whither infectious diseases? Some data at last. J Infect Dis
1992; 165:201-4.
4. Petersdorf RG. Whither infectious diseases? Memories, manpower, and
money. J Infect Dis 1986; 153:189-95.
5. Hamory BH, Hicks LL, Manpower and Training Committee, Infectious
Diseases Society of America. Infectious disease manpower in the
United States-1986. 1. Description of infectious disease physicians.
J Infect Dis 1992; 165:205-17.
6. Sexton DJ. Rates of infectious disease consultations in hospitals of different sizes and types, 1986-1987 [letter]. Rev Infect Dis 1991; 13:527.
Downloaded from http://cid.oxfordjournals.org/ at Penn State University (Paterno Lib) on May 12, 2016
Bloodstream infection
Pneumonia
Unexplained fever
Osteomyelitis
Urinary tract infection
Cellulitis
Intraabdominal infection
Wound infection
CNS infection
HIV infection
Abscess
Other
University practice
393