Overestimation of HCV prevalence by assessing positive anti-HCV results only.

Full text

Page 1

availability of tobacco cessation programs may not be
the only barrier to quitting. For example, studies show
that minority underserved populations are less likely
to engage in formal tobacco cessation interventions.2
Despite racial and socioeconomic differences in smok-
ing topography and outcome, the majority of pub-
lished research in this area is descriptive in nature,
with far less research focused on clinical trials within
underserved minority smokers.3For instance, at the
University of Chicago, after implementing a fax-to-
quit referral system to a smoking cessation clinic for
predominantly low-income minority hospitalized
smokers, we observed that few (?5% of those
referred) attended the cessation program after dis-
charge. Patients who did not attend cited a multitude
of other barriers for not taking advantage of this
opportunity, mostly related to access and social sup-
port. Therefore, it is imperative that future efforts to
improve smoking cessation in disadvantaged and low-
income populations address these barriers.
Correspondence: Dr Shah, Section of Hospital Medi-
cine,UniversityofChicago,5841SMarylandAve,Room
W306, MC 5000, Chicago, IL 60637 (lshah@medicine
.bsd.uchicago.edu).
1. DawoodN,VaccarinoV,ReidKJ,SpertusJA,HamidN,ParasharS;PREMIER
Registry Investigators. Predictors of smoking cessation after a myocardial in-
farction:theroleofinstitutionalsmokingcessationprogramsinimprovingsuccess.
Arch Intern Med. 2008;168(18):1961-1967.
2. Fiore MC, Bailey WC, Cohen SJ, et al. Treating Tobacco Use and Dependence:
Clinical Practice Guidelines. Washington, DC: US Dept of Health and Human
Services, Public Health Service; 2008.
3. King A, Sa ´nchez-Johnson L, Van Orman S, Cao D, Matthews A. A pilot com-
munity-based intensive smoking cessation intervention in African Ameri-
cans:feasibility,acceptabilityandearlyoutcomeindicators.JNatlMedAssoc.
2008;100(2):208-217.
In reply
We appreciate the feedback from Shah et al on our
article.1They raise concern on the use of “economic bur-
den” instead of annual income as a potential predictor of
smoking cessation after a myocardial infarction (MI) in
our study. To assess economic burden, at the time of MI
hospitalization, we asked patients whether they avoided
health care services in the prior year due to costs.
Although our bivariate analysis showed that annual
income less than $10000 was a significant predictor of
not quitting smoking, we chose to use “economic burden”
for the following reasons: (1) We focused specifically on
patients’ self-reported financial burden rather than infer-
ring it from their reported income because patients with
higher income may avoid health care services because of,
for example, personal debts, and similarly, persons with
less income may receive external funds for health care
and may not avoid care because of costs. Economic bur-
den thus ascertained directly from the patients is a strong
predictor of adverse outcomes after acute MI.2(2) In the
preliminary multivariable analysis including various
demographic, clinical, and quality of care factors in our
study, income did not predict smoking cessation after MI,
whereas perceived economic burden was a stronger pre-
dictor. And (3) since we had some missing data regarding
the self-reported income, we decided to remove the
income variable and add economic burden because the
results were practically unchanged.
We suggested that smoking cessation programs be
offered in hospitals caring for underprivileged and indi-
gent populations because we found that hospitals with
smoking cessation programs admitted patients with
higher socioeconomic status than did hospitals without
these programs. We, however, agree with Shah et al
that lack of availability of tobacco cessation programs
may not be the only barrier to quitting. As they point
out, patients with lower socioeconomic status may have
several barriers to nonparticipation such as poor access
to such programs and low social support. That is
why these programs should be made available in hospi-
tals caring for patients with low socioeconomic status.
We concur with Shah et al that future efforts and
trials should focus attention on barriers to access smok-
ing cessation interventions, especially in underserved
populations.
Correspondence: Dr Parashar, Division of Cardiology,
Department of Medicine, Emory University School of
Medicine, EPICORE, 1256 Briarcliff Rd, Building A, Ste
1 N, Atlanta, GA 30307 (susmita.parashar@emory
.edu).
Financial Disclosure: Dr Spertus has received consult-
ingfeesandhonorariafromCVTherapeuticsandUnited
HealthcareaswellasresearchgrantsfromCVTherapeu-
tics.
1. Dawood N, Vaccarino V, Reid KJ, Spertus JA, Hamid N, Parashar S; PREMIER
Registry Investigators. Predictors of smoking cessation after a myocardial in-
farction:theroleofinstitutionalsmokingcessationprogramsinimprovingsuccess.
Arch Intern Med. 2008;168(18):1961-1967.
2. RahimiAR,SpertusJA,ReidKJ,BernheimSM,KrumholzHM.Financialbar-
riers to health care and outcomes after acute myocardial infarction. JAMA.
2007;297(10):1063-1072.
Overestimation of HCV Prevalence by
Assessing Positive Anti-HCV Results Only
W
fection in primary care by McGinn et al,1and we note
that HCV prevalence was established only on the basis
of the antibody (anti-HCV) results. It did not include
supplemental testing to confirm the positive anti-HCV
results.Wethinkthisisagoldenopportunityforahealthy
e have carefully read the report about the
hepatitisCvirus(HCV)infectionscreening
tool to identify patients at risk for HCV in-
Lisa M. Shah, MD, MA
Vineet Arora, MD, MA
Andrea King, PhD
Jerry Krishnan, MD, PhD
Susmita Parashar, MD, MPH, MS
Nazeera Dawood, MD, MPH
Kimberly J. Reid, MPH
John Spertus, MD, MPH
Viola Vaccarino, MD, PhD
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 9), MAY 11, 2009WWW.ARCHINTERNMED.COM
903
©2009 American Medical Association. All rights reserved.
on May 11, 2009 www.archinternmed.comDownloaded from

Page 2

debateandtorememberthatahighrateoffalse-positive
results occurs in low-prevalence populations (range,
15%-60%).2In the study by McGinn et al,1subjects
who were found to be reactive for HCV antibodies
were considered positive, and then the analyses erro-
neously included a proportion of false-positive anti-
body results. This explained the elevated prevalence
for positive HCV antibodies (8.3%; 95% confidence
interval, 6.7%-10.2%) reported by the authors. Despite
the accuracy of third-generation immunoassays in
detecting antibodies and the high reliability of the
automated equipment,3studies that screen for HCV
infection only on the basis of a positive antibody result
overestimate the rate of HCV infection.
Interestingly, the anti-HCV assay is a semiquantita-
tive test.2The results are expressed as the signal to
cutoff (S/CO) ratio and the sensitivity and specificity
of the immunoassays depend on the cutoff point that
is chosen to define the positivity of the antibody.2,4,5
To facilitate the practice of reflex supplemental testing
in positive anti-HCV samples, Alter et al2proposed an
algorithm that included an option in which low values
for the S/CO ratio are used to identify those samples
requiring further testing, specifically with the recom-
binant immunoblot assay recombinant immunobinant
assay) 3.0 test, to define false-positive results. The
value of the S/CO ratio is directly related to the anti-
body concentration; lower levels have been associated
with false-positive results and higher levels with true-
positive results for the antibody.2,4,5In a recent report,
we demonstrate a direct relationship between positive
supplemental testing and increased antibody levels; in
contrast, very low antibody levels predict a false-
positive result.5
The recommended supplemental testing are the re-
combinant immunoblot assay and HCV RNA testing.2
Validation of every positive anti-HCV result with reflex
supplemental testing or an option using the value of the
S/COratioisrequiredtoidentifyfalse-positiveresultsin
people without overt clinical picture of liver disease.2,4,5
Then, the accuracy of the prediction tool proposed by
McGinn et al1to identify patients with risk for HCV-
infection is limited by the inclusion of a proportion of
false-positive anti-HCV results.
Correspondence:DrContreras,PedrodeAlarco ´nNo.45,
casa61,ResidencialPortaMagna,JardinesVallarta,Zapo-
pan,Jalisco45120,Mexico(acontreras53@hotmail.com).
1. McGinn T, O’Connor-Moore N, Alfandre D, Gardenier D, Wisnivesky J.
Validation of a hepatitis C screening tool in primary care. Arch Intern Med.
2008;168(18):2009-2013.
2. Alter MJ, Kuhnert WL, Finelli L; Centers for Disease Control and Preven-
tion.Guidelinesforlaboratorytestingandresultreportingofantibodytohepa-
titis C virus. MMWR Recomm Rep. 2003;52(RR-3):1-13, 15.
3. Contreras AM, Tinoco E, Celis A, et al; donHCVir Mexican Study Group.
Hepatitis C antibody intraassay correlation: is retest in duplicate necessary?
Transfusion. 2007;47(9):1686-1690.
4. Contreras AM, Tornero-Romo C, Orozco-Herna ´ndez A, et al. Rediscovering
hepatitis C antibody: new screening and diagnosis strategies. Gac Med Mex.
2007;143(suppl 2):3-12.
5. ContrerasAM,Tornero-RomoC,ToribioJ,etal.VerylowhepatitisCantibody
levelspredictfalsepositiveresultsandavoidsupplementaltesting. Transfusion.
2008;48(12):2540-2548.
Cost-effectiveness of Strategies for
Monitoring the Response to Antiretroviral
Therapy in Resource-Limited Settings
B
with human immunodeficiency virus in resource-
limited settings, concluding that monitoring with CD4
countswasbeneficialcomparedwithsymptom-basedap-
proaches.TheCD4countmonitoringstrategiesthatwere
considered involved the use of CD4 count monitoring
todecide(1)whentostartantiretroviraltreatment(ART)
and (2) when to switch to second-line regimens for pa-
tients already receiving ART. It would be helpful if the
authorsareabletorunmodelsthatconsiderthese2com-
ponents in isolation from each other. The use of CD4
counts rather than the presence of symptoms to decide
when to start treatment (the first component) results in
starting ART at a higher CD4 count, which is known to
be associated with lower death rates,2so the benefits of
CD4 count monitoring in this context would be ex-
pected.Incontrast,foragivengroupofpatientsstarting
ART, the benefits (and cost-effectiveness) of using CD4
countmonitoringoversymptom-basedmonitoringtode-
cideonwhentoswitchfromfirst-tosecond-linetherapy
are less clear and cannot be discerned from the results
presented.Considerationofthissecondcomponentwould
also better allow comparison with results from a simu-
lationmodelthatwedevelopedtocomparestrategiesfor
monitoring ART.3
ased on a computer simulation model, Ben-
davidandcolleagues1comparedthebenefitsand
costsofdifferentapproachestomonitoringpeople
Correspondence:DrPhillips,DepartmentofPrimaryCare
& Population Sciences, Royal Free and University Col-
lege Medical School, UCL, Rowland Hill Street, London
NW3 2 PF, England (a.phillips@pcps.ucl.ac.uk).
1. Bendavid E, Young SD, Katzenstein DA, Bayoumi AM, Sanders GD, Owens
DK. Cost-effectiveness of HIV monitoring strategies in resource-limited
settings: a Southern African Analysis. Arch Intern Med. 2008;168(17):1910-
1918.
2. Braitstein P, Brinkhof MW, Dabis F, et al; Antiretroviral Therapy in Lower
Income Countries (ART-LINC) Collaboration; ART Cohort Collaboration
(ART-CC)groups.MortalityofHIV-1-infectedpatientsinthefirstyearofan-
tiretroviral therapy: comparison between low-income and high-income
countries. Lancet. 2006;367(9513):817-824.
3. Phillips AN, Pillay D, Miners A, Bennett DE, Gilks CF, Lundgren JD. Out-
comes from monitoring of patients on antiretroviral therapy in resource-
limited settings with viral load, CD4 cell count, or clinical observation alone:
a computer simulation model. Lancet. 2008;371(9622):1443-1451.
In reply
Phillipsandcolleaguesrequestaclarificationofouranaly-
sis into the effectiveness of human immunodeficiency virus
monitoringinresource-limitedsettings.Theycorrectlypoint
outthatCD4countmonitoringtodeterminetimingoftreat-
ment initiation can be analyzed separately from monitor-
ing to determine when to switch therapy. In their analysis
based on a computer simulation model, Phillips et al find
Ana M. Contreras, MSc
Rodolfo J. Ochoa-Jime ´nez, MD
Andrew N. Phillips, PhD
Charles Gilks, DPhil
Jens D. Lundgren, MD, DMSc
(REPRINTED) ARCH INTERN MED/VOL 169 (NO. 9), MAY 11, 2009WWW.ARCHINTERNMED.COM
904
©2009 American Medical Association. All rights reserved.
on May 11, 2009 www.archinternmed.com Downloaded from