Identification of Patients with Hepatitis C Virus
Infection in New Haven County Primary
Victor J. Navarro, M.D., Thomas E. St. Louis, M.S.P.H., and
Beth P. Bell, M.D., M.P.H.
Background: Primary care providers (PCPs) must identify per-
sons at risk for hepatitis C virus (HCV) infection, test them cor-
rectly, refer to subspecialists, and use published guidelines. The
objectives of this study were to describe HCV practices of New
Haven County PCPs. Study: All 652 PCPs in New Haven County,
Connecticut, were surveyed to determine practices related to hepa-
titis C, including risk factor ascertainment, testing routines, use of
published guidelines, and referral practices. Results: Of 181 eli-
gible respondents, 143 (79%) were internal medicine physicians
and 38 (21%) were family practitioners. Eighty-four PCPs (46%)
routinely asked about a history of blood transfusion, and 112
(62%) routinely asked about a history of injection drug use (IDU).
Most PCPs would test current or past IDU (91% versus 83%,
respectively), persons transfused prior to 1992 (79%), health care
workers with a history of a needle stick accident (88%), and a child
born to an HCV-infected mother (76%). PCPs frequently referred
patients with hepatitis C to gastroenterologists. Most PCPs (76%)
were familiar with available hepatitis C testing guidelines. Con-
clusions: Most PCPs test for HCV infection appropriately, but
many do not elicit risk factor histories that could identify such
persons. More effective training with emphasis on eliciting a his-
tory of pertinent risk factors is needed.
Key Words: Hepatitis C detection—Liver disease—Primary care.
most strongly associated with HCV infection include illegal
drug use and high-risk sexual behavior.1Because most of
people infected are younger than 50 years of age, the burden
of disease associated with HCV may increase over the next
several decades as these individuals reach an age at which
n estimated 2.7 million persons in the United States are
infected with hepatitis C virus (HCV).1The factors
the complications of chronic liver disease are more likely to
occur.2Improvements in the efficacy of therapies that may
prevent these complications increase the potential benefit of
early identification of infected individuals.
Because HCV infection is frequently asymptomatic, rec-
ognition of patients who might benefit from treatment will
require primary care providers (PCPs) to identify persons at
risk for infection, test them correctly, and make appropriate
referrals to subspecialists. Guidelines for the prevention and
control of HCV infection and HCV-related chronic liver
disease and for the management of patients with hepatitis C
were formulated by the Centers for Disease Control and
Prevention (CDC) and the National Institutes of Health, the
latter having been recently updated.3,4These publications
represent an important consensus approach to providing
practitioners with guidance on how to identify and manage
patients with hepatitis C. However, PCP use of these re-
sources and their practice routines with respect to these
factors have not been well characterized.
In light of the need to better understand how PCPs (i.e.,
family practitioners and internists) approach identification
of patients with HCV infection, we conducted a mailed
survey of PCPs in New Haven County, Connecticut. The
goal of this survey was to examine practice routines with
respect to ascertainment of selected exposures associated
with HCV infection, testing, resource utilization, and sub-
MATERIAL AND METHODS
The 652 physicians listed in insurance and state Medicaid pro-
vider directories as practicing family practice (FP) or adult internal
medicine (IM) in New Haven County (1998 adult population,
608,802) at the inception of the study in May 1999 were sent three
mailings, ≈1 month apart. Physicians who reported practicing pri-
mary care for at least 8 hours per week were considered eligible.
Using the CDC recommendations as a guide,3we designed
brief patient profiles to determine the frequency with which PCPs
elicit histories of blood transfusion and injection drug use (IDU),
known HCV exposures. The patient profiles included hypothetical
history of hospitalization, trauma, surgery, or childbirth. Respon-
dents were also given the option of reporting that a history of
transfusion was elicited from all patients or under no circumstance.
The elicitation of a history of IDU was also assessed using hypo-
thetical patient profiles. These profiles included history of incar-
ceration, body tattoo or piercing, or sexually transmitted disease.
Submitted August 4, 2002. Accepted January 7, 2003.
From the Jefferson Medical College (V.J.N.), Philadelphia, Pennsyl-
vania, the Connecticut Emerging Infections Program (T.E.S.), Yale Uni-
versity School of Medicine, New Haven, Connecticut, and the Hepatitis
Branch (B.P.B.), Centers for Disease Control and Prevention, Atlanta,
Address correspondence and reprint requests to Dr. Victor J. Navarro,
Division of Gastroenterology and Hepatology, Thomas Jefferson Univer-
sity, Suite 480, 132 South 10th Street, Philadelphia, PA 19127. E-mail:
This work was performed at the Connecticut Emerging Infections Pro-
gram and the Yale University School of Medicine (New Haven, CT) and
funded by the Centers for Disease Control and Prevention (Atlanta GA).
J Clin Gastroenterol 2003;36(5):431–435. © 2003 Lippincott Williams & Wilkins, Inc.