Journal of Adolescent Health 39 (2006) 887 892

Original article

Cervicitis: To Treat or Not To Treat? The Role of Patient Preferences

and Decision Analysis
Jeanelle Sheeder, M.S.P.H.a,*, Catherine Stevens-Simon, M.D.a, Dennis Lezotte, Ph.D.b,
Judith Glazner, M.S.b, and Stephen Scott, M.D.c
a
Department of Pediatrics, Denver, Colorado
Department of Preventive Medicine and Biometrics, Denver, Colorado
c
Department of Obstetrics and Gynecology, University of Colorado Health Sciences Center, Denver, Colorado
Manuscript received December 28, 2005; manuscript accepted June 9, 2006
b

Abstract

Purpose: Mucopurulent cervicitis is neither a sensitive nor a specific indicator of antibiotic

sensitive infection. This analysis examines the positive and negative ramifications of treating
cervicitis empirically as a Chlamydial (CT) infection. It begins where prior analyses leave off, with
the number of cases of pelvic inflammatory disease (PID) prevented.
Methods: Three treatments were compared: 1) treat empirically/refer partner; 2) test, treat, and
base partner treatment on results; 3) test, base treatment on results. The outcomes were the physical
sequelae of PID and the psychological sequelae of being diagnosed with CT in a hypothetical cohort
of 500 teenagers with cervicitis, among whom the prevalence of CT averaged 33%, but ranged
between 10% and 70%.
Results: At a CT prevalence of 33%, Treatments 1 and 2 prevented three times as many cases of
PID-related physical sequelae (n 14) as Treatment 3 (n 5). However, to prevent these 14 cases
of physical sequelae, with Treatment 1, 163 teens needlessly suffer the psychological sequelae of a
false CT diagnosis and with Treatment 2, 101 do so. The ratio of physical sequelae prevented to
psychological sequelae caused, changed in relationship to the prevalence of CT, but was always
numerically most favorable with Treatment 3. Moreover, it was the only therapeutic approach for
which overall morbidity never exceeded the PID-related physical morbidity incurred in the absence
of treatment.
Conclusions: By including the effects of over diagnosing and treating CT, we have demonstrated
how the risks and benefits of empiric and nonempiric cervicitis therapy vary in relationship to CT
prevalence. Failure to consider both the physical and the psychological aspects of patient well-being
may mean that well-intentioned policies to reduce physical morbidity do not result in an overall
improvement in health of teenagers. 2006 Society for Adolescent Medicine. All rights reserved.

Keywords:

Sexually transmitted diseases; Cervicitis; Pelvic inflammatory; Disease; Psychological stress

Mucopurulent cervicitis, a clinical syndrome characterized by erythema, edema, and friability of the ectocervix
and purulent endocervical exudate, is among the most common gynecologic problems in general practice [1]. Treat*Address correspondence to: Ms. Jeanelle Sheeder, Department of
Pediatrics, Division of Adolescent Medicine, University of Colorado
Health Sciences Center, The Childrens Hospital, 1056 East 19th Street,
Box B025, Denver, CO 80218.
E-mail address: jeanelle.sheeder@uchsc.edu

ment poses a perplexing dilemma of major public health

importance because Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (GC), the two most common bacterial
causes, are isolated from less than a third of women with
these clinical findings [2 4]. Moreover, mucopurulent cervicitis often persists for weeks after documented eradication
of these pathogens and can be caused by other infectious
and noninfectious agents [5,6]. The utility of empirically
treating cervicitis patients and their sexual partners for CT
and GC might be questioned more frequently if the conse-

1054-139X/06/$ see front matter 2006 Society for Adolescent Medicine. All rights reserved.
doi:10.1016/j.jadohealth.2006.06.005

888

J. Sheeder et al. / Journal of Adolescent Health 39 (2006) 887 892

quences of not treating these sexually transmitted diseases

(STDs) were less dire. However, with the annual treatment
costs of pelvic inflammatory disease (PID) approximately
2 billion dollars in the United States [7], concerns about the
consequences of under-diagnosis and under-treatment usually overshadow concerns about the consequences of overdiagnosis and over-treatment [1 4,8].
Teenagers with mucopurulent cervicitis are especially
apt to be treated empirically. In comparison with most other
groups, they have higher sexually transmitted disease (STD)
prevalence rates, are especially vulnerable to the adverse
reproductive consequences of untreated STDs, and are generally perceived as noncompliant with follow-up [1,8].
However, taken together, the low positive predictive value
of mucopurulent cervicitis for treatable infections [2 6], the
futility of treating index cases if their sexual networks are
not treated [9 11], the adverse psychological effects of
being diagnosed with an STD [1217], and the serious
public health problem created by the indiscriminate use of
antibiotics [18], raise serious concerns about the wisdom of
empirically treating this common gynecologic problem as
an STD.
Decision analysis has been used to evaluate the financial
costs of empirically treating STDs [19 21]. Pelvic inflammatory disease (PID), the most common sequelae of CT and
GC cervicitis, develops in 10 40% of untreated women, but
only 35% of treated women [8,22,23]. Almost half of
women who develop PID also develop costly complications
ranging from chronic pelvic pain to ectopic pregnancy and
infertility [7,2225]. Thus, most decision analyses favor
empiric treatment of cervicitis [19 21]. Approximately
70% of the male sexual partners of CT cervicitis patients are
infected with CT [26,27]. Although severe sequelae of CT
infections are encountered less commonly in males than
females, the risk of re-infection from untreated male partners is not insignificant [1,9 11,26,27]. Hence, some decision analyses favor empiric treatment of both cervicitis
patients and their partners [19,20]. We were unable to find
any models that included the human and environmental
costs of treating individuals for STDs they do not have. This
is not because researchers are unaware them. It has been
documented repeatedly that the diagnosis of an STD evokes
dysphoric feelings, causes anxiety, depression, and decreased self-esteem, disrupts romantic relationships, and
precipitates domestic violence [1217]. In one study, it was
found that after controlling for potential confounders, the
incidence of depressive symptoms among previously nondepressed participants in the National Longitudinal Study of
Adolescent Health averaged 30% during the following year
if there was an STD diagnosis and 11% if there was not
[13]. In another study, it was reported that approximately
25% of young adults felt that the diagnosis of an STD would
cause them to feel angry and depressed [14]. The results of
a third study demonstrated that teenagers tend to be as
psychologically risk-averse to STDs as they are physically

[17]. Similar documentation exists for the serious public

health problem posed by the indiscriminate use of antibiotics [18]. These known risks of over-diagnosis and overtreatment are occasionally considered. Yet, they rarely temper enthusiasm for treating cervicitis empirically as an STD
[1 4,8,19 21]. To illustrate the magnitude of the threat
empiric treatment of cervicitis as an STD poses to teenagers,
this analysis begins where prior decision analyses leave off;
with PID. We demonstrate how empiric and nonempiric
treatment affect the ratio of physical to psychological sequelae incurred and the ratio of physical sequelae prevented
to psychological sequelae caused. Because GC is isolated
from teenage cervicitis patients far less commonly than CT
[1,8], we limited our analysis to CT-related morbidity.
Methods
The results of prior decision analyses consistently demonstrate that PID antedates all the costly physical sequelae
of CT cervicitis [19 21]. Hence, it is invariably the pivotal
node in these algorithms [19 21]. These analyses are very
similar. Rather than reiterate the logic behind them, we refer
the reader to those publications [19 21]. In preparation for
our analysis we applied these previously published algorithms to a hypothetical population of 500 urban teen-clinic
patients with cervicitis among whom the prevalence of CT
is likely to be approximately 33% [2,8,10,11,19 21]. Because epidemiological studies show that the prevalence of
CT in teenage cervicitis patients can vary from a low of
10% in a suburban pediatric office [28] to a high of 70% in
an urban STD clinic [9], we examined these scenarios.
Taken together, the results of prior studies demonstrate that the risk of developing PID increases in relationship to the length of time that elapses between the
diagnosis and treatment of CT cervicitis [5,7,23]. For example, it is estimated that approximately 5% of women who
are treated for CT cervicitis within 5 days of diagnosis
develop PID, compared with 15% of those who are treated
5 to 10 days after diagnosis, 25% of those who are treated
between 10 and 20 days after diagnosis, and 35% who are
untreated [5,7,23]. Extrapolating from previously published
decision analysis algorithms [19 21] to our hypothetical
population, we estimated that with empiric treatment of
cervicitis patients and their partners, 20 teenagers would
develop PID. Due to re-infection by untreated partners
[26,27], we estimated that with empiric treatment of cervicitis patients and test-based treatment of their partners, 21
teenagers would develop PID. Based on 5% false-negative
test results [29,30], treatment delays, loss to follow-up, and
re-infection by untreated partners [8,9,22,23,26,27], we estimated that with nonempiric treatment, 46 teenagers would
develop PID.
Starting from these figures, we compared the physical
and psychological ramifications of three widely accepted
approaches to the treatment of cervicitis: Treatment 1) treat

J. Sheeder et al. / Journal of Adolescent Health 39 (2006) 887 892

empirically and recommend that the patient refer her partner

for STD treatment; Treatment 2) test for infection, treat
empirically, and base partner treatment on the test results;
Treatment 3) test for infection and base patient and partner
treatment on the results. This analytic approach underestimates the adverse effects of over-treating STDs as it does
not include the detrimental effects associated with the indiscriminant use of antibiotics [18].
Based on our review of the literature, we estimated that:
1) approximately 40% of teenage girls with PID would
develop physical sequelae [8,22,23], and 2) approximately
30% of teenagers who are told they have an STD would
develop significant psychological sequelae, ranging from
dysphoric feelings, anxiety, and depression to decreased
self-esteem, disrupted romantic relationships, and domestic
violence [1217]. Such psychological angst is an unavoidable morbidity for teenagers who have STDs, but an unnecessary morbidity for those who do not. We could not actually model the physical sequelae of PID and the psychological
sequelae of an STD diagnosis because we had no patient
utilities for them and no empiric data from which to create
such measures. Hence, all calculations were performed in a
Microsoft Excel XP spreadsheet, rather than decision analysis software.
Results
The data presented in Table 1 show that if all 500
cervicitis patients are treated empirically with a gram of
azithromycin and instructed to have their sexual partners
treated (Treatment 1), approximately eight individuals will
suffer physical sequelae. However, on average, only 33% of
teenagers with cervicitis are infected with CT [2 4,8] and
only 70% of the male sexual partners of CT cervicitis
patients have CT [26,27]. Thus, in our hypothetical population of 500 teenagers, 335 females would be over-treated.
Assuming that 65% of these females actually refer their
male partners for treatment [9,3133], 209 males would be

Table 1
Treatment outcomes in a hypothetical population of 500 teenagers
with cervicitis
Sequelae

Treatment 1

Treatment 2

Treatment 3

Physical
Psychologicala
Ratio of physical to
psychological sequelae
Physical sequelae preventedb
Ratio of physical sequelae
prevented to psychological
sequelae caused

8
163

9
101

18
1

1:20
15

1:11
14

18:1
5

1:11

1:7

5:1

Psychological sequelae number of teenagers (males and females)

diagnosed with an STD they do not have .30.
b
If no one were treated, 23 teens would suffer physical sequelae, and no
one would suffer psychological sequelae.
a

889

over-diagnosed. Thus, a total of 544 individuals would be

told they have an STD that they do not have. As a result of
this over-diagnosis, 163 teenagers would needlessly suffer
the psychological sequelae of being told they have an STD.
If all 500 cervicitis patients are tested for CT and treated
empirically with a gram of azithromycin, but partner treatment is based on these test results (Treatment 2), approximately nine individuals will suffer physical sequelae.
Again, because only 33% of cervicitis patients have CT, 335
females would be over-diagnosed. However, because falsepositive test results are extremely uncommon (i.e., 1%)
[29,30], almost no males would be over-diagnosed. Thus,
only 101 individuals would needlessly suffer the psychological sequelae of being told they have an STD. As shown
in Table 1, basing male treatment on female test results
reduces the ratio of physical to psychological sequelae from
approximately 1:20 to 1:11, and the ratio of physical sequelae prevented to psychological sequelae caused is also
reduced. If no one is treated, 23 teens would suffer physical
sequelae and no one would suffer psychological sequelae.
Thus, with Treatment 1, 15 cases of PID sequelae would
be prevented and with Treatment 2, 14 cases of PID sequelae would be prevented and the ratios of physical sequelae prevented to psychological sequelae caused would
be 1:11 and 1:7, respectively. Finally, if all 500 cervicitis
patients are tested for CT, and treatment is based on these
test results (Treatment 3), false negative test results, treatment delays, loss to follow-up, and re-infection by untreated
partners would create nine additional cases of physical sequelae. As a result, approximately 18 individuals will suffer
physical sequelae and only one case is prevented. In addition, with Treatment 3, one individual will suffer psychological sequelae due to false-positive test results.
Hence, the ratio of physical to psychological sequelae
would be approximately 18:1 and the ratio of physical
sequelae prevented to psychological sequelae caused would
be 5:1 (Table 1).
Figure 1 shows how the ratio of physical to psychological sequelae changes in relationship to the prevalence of
CT among cervicitis patients. When the prevalence is low
(10%), adopting either empiric treatment approach results in
a very low physical-morbidity-prevented to psychologicalmorbidity-caused ratio. Indeed, depending on how partners
are treated, 136 222 teenagers would needlessly suffer psychological sequelae to spare three of the seven teenagers
who would develop physical sequelae if untreated. By contrast, at the high end of the prevalence range (70%), adopting Treatments 1 or 2 would mean that 69 and 45 teenagers,
respectively, would suffer unnecessary psychological sequelae. Thus, with empiric treatment, the ratio of physical
morbidity prevented to psychological morbidity caused approaches one. Treatment 3 becomes less appealing as the
prevalence of CT increases. However, overall morbidity is
always lowest with Treatment 3. Indeed, Figure 1 shows
that it is the only therapeutic approach for which the com-

890

J. Sheeder et al. / Journal of Adolescent Health 39 (2006) 887 892

Figure 1. Change in the prevalence of physical and psychological sequelae in relationship to the prevalence of CT in a hypothetical population of
500 teenagers with cervicitis.

bination of psychological and physical sequelae related to

over- and under-treatment never exceeds the PID-related
physical morbidity incurred if no one is treated.
Discussion
The risks of under-diagnosis and under-treatment are
common themes in the CT literature [8], but the risks of
over-diagnosis and over-treatment are rarely discussed. Our
analysis makes a novel contribution, as it graphically illustrates that there are significant costs as well as benefits to
empirically treating mucopurulent cervicitis as an STD.
This is because most teenagers with this common physical
finding do not have a treatable lower genital tract infection
[2 6], and being diagnosed with an STD can have damaging psychological sequelae [1217]. Because most STDs

are asymptomatic and other factors can cause uninfected

cervices to appear inflamed and cervical exudates to become
purulent, routine periodic screening with highly sensitive
nucleic acid amplification tests is the mainstay of disease
detection [1,8].
Mucopurulent cervicitis is a widely accepted trigger for
STD testing [1,8]. However, there is far less consensus
concerning how best to operationalize the Center for Disease Controls vague recommendation to treat cervicitis
empirically if the likelihood of infection with CT or GC is
high or the patient is unlikely to return for treatment [1].
Our analysis suggests that this is because there is no single
optimal therapeutic approach. We have demonstrated that
the magnitude of the risk of developing physical and psychological sequelae depends upon the prevalence of CT.
Yet, we were unable to select an optimal approach. This

J. Sheeder et al. / Journal of Adolescent Health 39 (2006) 887 892

is because the definition of optimal depends upon the

relative importance each patient and provider place on
avoiding these complications. Some may wish to avoid the
potential sequalae of PID and others the psychological distress that the diagnosis of an STD may cause. Typically,
utilities are created to model individual preferences [34
38]. However, we were unable to do so as we had no data
upon which to base the calculations. Previous work demonstrates that patients take diverse approaches to risk and
that the decisions they make often seem counterintuitive to
providers [36 38]. For example, one study found that parents were willing to risk a 1-in-200 chance of death to spare
their children from a venipucture [36]. Our model demonstrates the importance of taking both the physical and psychological dimensions of well-being into account. Thus, we
contend that the lack of patient utilities for the physical
sequelae of PID and the psychological sequelae of an STD
diagnosis make it impossible to apply decision analyses in
solving the therapeutic dilemma cervicitis poses. Nonetheless, the figure may help health care providers select the
treatment option that best considers each patients circumstances, values, and risk aversion preferences.
More broadly, this study demonstrates the complexity of
subjecting clinical problems to appraisal with decision analysis. The basic premise in decision analysis is that all
available information on likelihood, costs, and effects of the
various outcomes of particular choices can be quantified and
synthesized to help decision-makers make rational choices
[34,35]. Our analysis shows how three widely accepted
approaches to treating cervicitis affect the risk of developing physical and psychological sequelae. However, we were
unable to take individual variations in risk aversion and
preference into account. Thus, this should be viewed as only
one component of the decision-making process. There is
little doubt that delaying treatment of CT cervicitis increases the likelihood of developing PID-related physical
morbidity [25,7,8,22,23]. Conversely, empiric treatment of
nonCT cervicitis increases the likelihood of developing psychological morbidity [1217]. However, there is no evidence that the benefits of averting physical morbidities at
the cost of incurring psychological ones improve the wellbeing of teenagers with cervicitis. Given the possible psychological risks of empiric treatment, the risks and benefits
of testing and empirically treating cervicitis should be carefully considered. A balanced approach that simultaneously
weighs the need to prevent serious physical morbidity
against the potential risk of equally serious psychological
morbidity should be taken with the individual patient. Although not included in this analysis, the detrimental effects
of indiscriminately using antibiotics must also be considered. The magnitude of the implicit problem caused by the
over-treatment of cervicitis patients and their partners
makes it clear that efforts to reduce the threat of antibiotic
resistance by improving the prescribing practices of health
care providers must reach beyond respiratory infections.

891

Despite the intuitive appeal of empiric therapy, our analysis demonstrates that health care providers who fail to
consider the negative ramifications of this therapeutic approach are apt to become victims of inadvertent consequences. Because our analysis includes an estimate of the
number of patients who are apt to suffer psychological
trauma as a result of being told they have an STD they do
not have, it can aid providers in taking into account the
possibility that some patients may wish to avoid the risk for
psychological trauma over the risk of developing the physical sequelae of PID. Consumer-driven medical care has
costs as well as benefits [39,40]. It may be difficult to
understand and accept the decisions patients make [36].
However, Figure 1 illustrates that, in most scenarios, empiric treatment increases overall morbidity. To minimize the
likelihood of such paradoxical outcomes, research and policy addressing the treatment of cervicitis should include net
reduction in overall morbidity. For example, our model
demonstrates that if empiric therapy is selected, only the
partners of patients with CT should be treated. Failure to
consider the consequences of over-treatment may mean that
well-intentioned policies to reduce physical morbidity do
not lead to overall improvements in patient well-being.

References
[1] Centers for Disease Control and Prevention. Sexually transmitted
disease treatment guidelines. MMWR Recomm Rep 2002;51:32.
[2] Moscicki B, Shafer MA, Millstein SG, et al. The use and limitations
of endocervical Gram stains and mucopurulent cervicitis as predictors
for Chlamydia trachomatis in female adolescents. Am J Obstet Gynecol 1987;157:6571.
[3] Moore SG, Miller WC, Hoffman IF, et al. Clinical utility of measuring white blood cells on vaginal wet mount and endocervical gram
stain for prediction of chlamydial and gonococcal infections. Sex
Transm Dis 2000;27:530 8.
[4] Marrazzo JM, Handsfield HH, Whittington WLH. Predicting chlamydial and gonococcal cervical infection: implications for management of cervicitis. Obstet Gynecol 2002;100:579 84.
[5] Paavonen J, Roberts PL, Stevens CE, et al. Randomized treatment of
mucopurulent cervicitis with doxycycline or amoxicillin. Am J Obstet
Gynecol 1989;161:128 35.
[6] Critchlow CW, Wolner-Hanssen P, Eschenbach DA, et al. Determinants of cervical ectopia and cervicitis: age, oral contraception, specific cervical infection, smoking, and douching. Am J Obstet Gynecol
1995;173:534 43.
[7] Rein DB, Kassler WJ, Irwin KL, et al. Direct medical cost of pelvic
inflammatory disease and its sequelae: decreasing, but still substantial. Obstet Gynecol 2000;95:397 402.
[8] Biro F, Workowski K, Blythe MJ, et al. NASPAG/JPAG roundtable
discussion annual clinical meeting 2003Philadelphia, PA: Sexually
transmitted diseases (STD) treatment guidelines 2002. J Pediatr Adolesc Gynecol 2004;17:143 6.
[9] Rietmeijer CA, Van Bemmelen R, Judson FN, et al. Incidence and
repeat infection rates of Chlamydia trachomatis among male and
female patients in an STD clinic. Sex Transm Dis 2002;29:6572.
[10] Orr DP, Johnston K, Brizendine E, et al. Subsequent sexually transmitted infection in urban adolescents and young adults. Arch Pediatr
Adolesc Med 2001;155:94753.

892

J. Sheeder et al. / Journal of Adolescent Health 39 (2006) 887 892

[11] Xu F, Schillinger JA, Markowitz LE, et al. Repeat Chlamydia trachomatis infection on women: analysis through a surveillance case
registry in Washington State, 19931998. Am J Epidemiol 2000;152:
1164 70.
[12] Rosenthal SL, Biro FM. A preliminary investigation of the psychological impact of sexually transmitted diseases in adolescent females.
Adolesc Pediatr Gynecol 1991;4:198 201.
[13] Shrier LA, Harris SK, Beardslee WR. Temporal associations between
depressive symptoms and self-reported sexually transmitted disease
among adolescents. Arch Pediatr Adolesc Med 2002;156:599 606.
[14] Ford CA, Jaccard J, Millstein SG, et al. Young adults attitudes,
beliefs, and feelings about testing for curable STDs outside of clinic
settings. J Adolesc Health 2004;34:266 9.
[15] Rainey DY, Stevens-Simon C, Kaplan D. Self-perception of infertility among adolescents. Am J Dis Child 1993;147;1053 6.
[16] Duncan B, Hart G, Scoular A, et al. Qualitative analysis of psychosocial impact of diagnosis of Chlamydia trachomatis: implications for
screening. BMJ 2001;322:1959.
[17] Cunningham SD, Kerrigan D, Pillay KB, Ellen JM. Understanding
the role of perceived severity in STD-related care-seeking delays. J
Adolesc Health 2005;37:69 74.
[18] The National Foundation for Infectious Diseases. Biological terrorist
attacks, antimicrobial resistance are growing global threats. Double
Helix 1998;23:4.
[19] Genc M, Mardh PA. A cost-effectiveness analysis of screening and
treatment for Chlamydia trachomatis infection in asymptomatic
women. Ann Intern Med 1996;124:17.
[20] Randolph AG, Washington AE. Screening for Chlamydia trachomatis in adolescent males: a cost-based decision analysis. Am J Public
Health 1990;80:54550.
[21] Trachtenberg AI, Washington E, Halldorson S. A cost-based decision
analysis for chlamydia screening in California family planning clinics. Obstet Gynecol 1988;71:101 8.
[22] Scholes D, Stergachis A, Heidrich FE, et al. Prevention of pelvic
inflammatory disease by screening for cervical chlamydial infections.
N Engl J Med 1996;334:1362 6.
[23] Rees E. The treatment of pelvic inflammatory disease. Am J Obstet
Gynecol 1980;138:10427.
[24] Hillis SD, Owens BS, Marchbanks PA, et al. Recurrent chlamydial
infections increase the risks of hospitalization for ectopic pregnancy
and pelvic inflammatory disease. Am J Obstet Gynecol 1997;176:
1037.
[25] Ness RB, Soper DE, Holley RL, et al. Effectiveness of inpatient and
outpatient treatment strategies for women with pelvic inflammatory
disease: results form the Pelvic Inflammatory disease Evaluation and
Clinical Health (PEACH) Randomized Trial. Am J Obstet Gynecol
2002;186:929 37.

[26] Quinn TC, Gaydos C, Shepherd M, et al. Epidemiologic and microbiologic correlates of Chlamydia trachomatis infection in sexual
partnerships. JAMA 1996;276:1737 42.
[27] Lin JS, Donegan SP, Heeren TC, et al. Transmission of Chlamydia
trachomatis and Neisseria gonorrhoeae among men with urethritis
and their female sex partners. J Infect Dis 1998;178:170712. Available from: http://www.journals.uchicago.edu/JID/journal/issues/
v178n6/971603/ - fn1.
[28] Best D, Ford CA, Miller WC. Prevalence of Chlamydia trachomatis
and Neisseria gonorrhea infection in pediatric private practice. Pediatrics 2001;108:e103.
[29] Johnson RE, Newshall WJ, Papp JP. Screening tests to detect Chlamydia trachomatis and Neisserae gonorrheae infections2002.
MMWR Recomm Rep 2002;51:138.
[30] Beck-Sague CM, Farshy CE, Jackson TK, et al. Detection of Chlamydia trachomatis cervical infection by urine tests among adolescents clinics. J Adolesc Health 1998;22:197204.
[31] Schwebke JR, Sadler R, Sutton JM, et al. Positive screening tests for
gonorrhea and chlamydial infection fail to lead consistently to treatment of patients attending a sexually transmitted disease clinic. Sex
Transm Dis 1997;24:181 4.
[32] Schillinger JA, Kissinger P, Calvet H, et al. Patient-delivered partner
treatment with azithromycin to prevent repeated Chlamydia trachomatis infection among women: a randomized, controlled trial. Sex
Transm Dis 2003;30:49 56.
[33] St Lawrence JS, Montano DE, Kasprzyk D, et al. STD screening,
testing, case reporting, and clinical and partner notification practices:
a national survey of US physicians. Am J Public Health 2002;92:
1784 8.
[34] Taylor TR. Clinical decision analysis. Methods Inf Med 1976;15:
216 24.
[35] Fishburn PC. Utility Theory for Decision Making. Huntington, NY:
Robert E. Krieger Publishing Co, Inc., 1979.
[36] Madsen KA, Bennett JE, Downs SM. The role of parental preferences
in the management of fever without source among 3- to 36-month-old
children: a decision analysis. Pediatrics 2006;117:106775.
[37] Kramer MS, Etezadiamoli J, Ciampi A, et al. Parents versus physicians values for clinical outcomes in young febrile children. Pediatrics 1994;93:697702.
[38] Kramer MS, MacLellan A-M, Ciampi A, et al. Parents vs physicians
utilities (values) for clinical outcomes in potentially bacteremic children. J Clin Epidemiol 1990;43:1319 25.
[39] Kravitz RL, Bell RA, Azari R, et al. Direct observation of requests for
clinical services in office practice: what do patients want and do they
get it? Arch Intern Med 2003;163:1673 81.
[40] Kravitz RL, Epstein RM, Feldman MD, et al. Influence of patients
requests for direct-to-consumer advertised antidepressants: a randomized controlled trial. JAMA 2005;293:19952002.