In 2006, the Centers for Disease Control and Prevention (CDC) released practice guidelines for the administration of expedited partner therapy (EPT) to sex partners of heterosexual patients diagnosed as having chlamydia or gonorrhea.1 Expedited partner therapy refers to the practice of providing treatment to partners of patients with sexually transmitted infections (STIs) without a direct assessment by a clinician. The usual mode of delivery is patient-delivered partner therapy in which the patient delivers a prescription or actual medications to the current sex partner(s). Since the release of the CDC guidelines, this practice has been endorsed by the American Medical Association,2 the American Congress of Obstetricians and Gynecologists,3 and the Society for Adolescent Medicine,4 and many states have initiated EPT programs5–7 or have moved legislatively to facilitate implementation.8
Evidence to date indicates that the EPT model is superior to the traditional means of partner notification (by the patient or provider) in terms of increased rate of partner notification and treatment.9,10 In addition, results from randomized controlled trials indicate that EPT reduces reinfection among patients infected with Chlamydia trachomatis (Ct) by approximately 15% to 20% and Neisseria gonorrhoeae (GC) by as much as 73%.9,11 Although the efficacy of EPT in these trials was consistent, it is questionable whether interventions shown to be effective in clinical trials can be replicated and brought to scale in the “real world” of STI clinical practice. For example, brief counseling demonstrated in randomized clinical trials to reduce new STIs by 30% after 6 months12 but has not been widely adopted in clinical settings presumably because even a short behavioral intervention was deemed not to be feasible in a busy practice setting.1 Although the efficacy of EPT is dependent on the partner taking the medications and that the drugs will effectively treat the disease, the effectiveness of EPT in clinical settings is by no means a forgone conclusion, and studies are needed to investigate factors associated with EPT uptake and evaluate outcomes in real-world circumstances. Expedited partner therapy was introduced in the DHMC in November, 2006. In this article, we described efforts to establish EPT as a routine service in the clinic, evaluated patient uptake, and compared reinfection rates between patients who accepted EPT and those who did not.
MATERIALS AND METHODS
The DMHC is the largest STI clinic and HIV testing facility in the Rocky Mountain region with nearly 16,000 visits annually. The clinic offers an array of services including free and confidential STI and HIV testing, counseling, and treatment, as well as family planning services for men and women.
The clinic initiated the EPT program in November 2006. Consistent with CDC guidelines, eligible patients included heterosexual men and women diagnosed as having chlamydia or gonorrhea (by nucleic acid amplification test) who returned to the clinic for treatment within 60 days of diagnosis and men diagnosed as having gonorrhea by Gram stain on the day of the visit. Men who have sex with men (MSM) were excluded per CDC guidelines because this creates missed opportunities for HIV testing and counseling.1 Eligible patients were offered ”partner packs” for up to 3 partners, by either a nurse or a health care assistant. Health care assistants are active licensed practical nurses or have graduated from an accredited program in Medical Office Assisting and are authorized to dispense medications to patients who do not require a physical examination. Medications included azithromycin (1 g of oral suspension or 4 250 mg tablets) for chlamydia exposure and a combination of azithromycin and cefpodoxime (400 mg tablet) for exposure to gonorrhea. Medications were dispensed from the clinic pharmacy and provided at no cost to the patient. The cost of all clinic medications is paid by the Colorado Department of Public Health and Environment. Packets were labeled with the index patient’s name, followed by the word “partner” and a unique identifier to track medications. Each packet contained an information sheet (in English and Spanish) encouraging the partner to seek medical care and included a set of instructions on how to take medication(s), possible adverse effects, and the clinic telephone number in case of questions. In February, 2007 (i.e., 3 months after the start of the program), the Colorado State Board of Pharmacy requested that the program be halted until the board amended its policy to explicitly allow the Colorado clinic pharmacies to dispense EPT and to align its policy with the Colorado State Board of Medical Examiners, which had already adopted an EPT policy in 2001.13 In July 2007, the board endorsed the use of EPT14 and the clinic resumed the program. At this time, documentation of EPT was checked sporadically by physician chart review. In May 2008, we developed a quality assurance protocol in which monthly reports systematically identified patient records if EPT information was incomplete or contradictory and the responsible clinician was notified of all charts. Providers were asked to correct the patient’s chart if irrefutable documentation existed to support the correction (e.g., a separate log of EPT treatment dispensations is kept in the clinic). In January 2009, we amended the clinic’s electronic medical record (EMR) to capture detailed information about EPT refusals, and we applied rules that required completion of EPT information before exiting the form. To supplement regular staff trainings, we reported on program progress at staff meetings on a biannual basis.
The study period spanned November 2006 through April 2011, excluding the 6-month hiatus in 2007. During this period, 2679 non-MSM patients were diagnosed as having gonorrhea on the day of visit or diagnosed as having chlamydial infection and/or gonorrhea by nucleic acid amplification test and returned to the clinic for treatment within 6 to 60 days of the test date. We omitted 37 visits because of incomplete information about treatment and an additional 64 because we could not confirm that EPT was offered by the provider. The final data set used to examine EPT acceptance included 2578 visits among 2345 patients. We examined reasons for EPT refusal among 681 patients who were eligible to receive EPT but declined. To investigate reinfection, we identified 351 patients who returned to the clinic within 21 to 90 days of treatment initiation and were retested for the original infection. These patients returned either because of symptoms (207) or for an unknown reason (144).
Analyses addressed 5 distinct objectives. First, we described the overall study group by the diagnosis that rendered them eligible for EPT: chlamydia, gonorrhea, or both. We compared differences in demographic, clinical, and risk variables across diagnoses using the χ2 statistic. Second, we graphed monthly rates of EPT acceptance over the observation period and compared rates over the startup, the pilot, and the two process improvement phases. Third, we used binomial regression to estimate prevalence ratios and their associated 95% confidence intervals to examine the association between EPT acceptance and several patient characteristics. Fourth, for those who refused EPT, we generated descriptive statistics to examine reasons for refusal. Finally, to describe the clinical impact of the program, we compared reinfection rates (with the original organism) among 351 patients who returned to the clinic with 21 to 90 days of treatment of the original infection. We compared reinfection rates among those who initially accepted EPT to those who refused, stratified by sex and type of diagnosis.
We conducted all analyses using SAS 9.3 (SAS Institute, Cary, NC). This study was considered program evaluation: data used for analyses were retrospectively gathered from clinic patients; however, results were aggregated and therefore anonymous. Accordingly, the study was deemed exempt from review by the Colorado Multiple Institutional Review Board.
EPT Eligible Study Population
When comparing the demographic, behavioral, and clinical characteristics of the clinic population eligible for EPT by type of diagnosis (Ct+, GC+, or Ct+/GC+), significant differences emerged (Table 1). In the eligible population, a higher proportion of women than men were diagnosed as having chlamydia (65.2% vs. 34.8%; P < 0.01), whereas conversely, a higher proportion of men were diagnosed as having gonorrhea (86.6% vs. 13.5%; P < 0.01). Most of those diagnosed as having gonorrhea (60.2%) were black, whereas most patients diagnosed as having chlamydia were Hispanic (44.4%) or white (24.6%; P < 0.01). Although there were no differences by prior diagnosis nor the number of partners reported in the last 3 months, there was a difference by type of provider: those diagnosed as having gonorrhea were more likely to have been treated by a nurse (80.5%), compared to 64.1% of those diagnosed as having chlamydia (P < 0.01).
EPT Acceptance over Time
Expedited partner therapy acceptance rates over the four observation periods are shown in Figure 1. During the brief startup phase (November 2006–February 2007), an average of 19.9% of eligible patients accepted EPT. During the pharmacy review (March 2007–August 2007), no EPT was dispensed; however, when the program resumed, an average of 16.7% patients accepted EPT over the pilot phase (September 2007–April 2008). After initiation of the quality assurance protocol in May 2008, rates increased significantly to 24.1% over the next 8 months. Finally, the requirement of complete documentation of EPT in the EMR elicited significantly higher rates of patient acceptance, averaging 48.0% from January 2009 to April 2011, a 100% improvement over the previous phase.
Factors Associated With EPT Acceptance
Owing to the differences in the populations infected with gonorrhea or chlamydia, we analyzed factors associated with EPT acceptance by the diagnosis that determined eligibility for EPT (Table 2). Overall, women were significantly more likely to accept EPT than men when diagnosed as having chlamydia (prevalence ratio, 1.6; 95% confidence interval [CI], 1.3–1.8) or both chlamydia and gonorrhea (prevalence ratio, 4.1; 95% CI, 1.1–15.6); however, there was no difference by gender among those diagnosed as having gonorrhea alone. The only difference in acceptance by race was among those diagnosed as having chlamydia: black patients were less likely to accept EPT than whites (prevalence ratio, 0.8; 95% CI, 0.7–0.9). Those with 1 or more prior clinic visits were more likely to accept EPT when diagnosed as having either chlamydia or gonorrhea (prevalence ratio, 1.2; 95% CI, 1.0–1.4). Compared with patients treated by a health care assistant, patients treated by a nurse were more likely to accept EPT when diagnosed as having chlamydia (prevalence ratio, 1.7; 95% CI, 1.4–2.0) and gonorrhea (prevalence ratio, 3.3, 95% CI, 2.3–5.0).
Reasons for EPT Refusal
The most common reason for refusing EPT reported by the index case was that the partner had been or would be notified (41.3%), with slightly fewer reporting that the partner had already been or was currently being treated (35.2%). The least common reason reported was that the patient had not had contact with the partner or another reason not defined in the EMR (23.5%). The distribution of refusal reason differed significantly by sex, age, and number of partners reported in the last months (Table 3).
Reinfection Rates Among Those Accepting or Refusing EPT
Clinical impact, measured by reinfection rates among those who returned to the clinic for retesting, indicated that those patients who accepted treatment for their partner(s) were less likely to test positive when retested for the original infection (Table 4). Of the 1478 persons originally diagnosed as having chlamydia, 240 returned to the clinic for retesting. Among 147 who refused EPT at the initial visit, 23 (15.7%) retested positive, whereas among the 93 who had accepted EPT at the original visit, 11 (11.8%) retested positive, a nonsignificant reduction of 24.8%. The effect was greater among those who were originally diagnosed as having gonorrhea: among the 1208 persons originally diagnosed, 129 returned for retesting, with 25.6% of EPT refusers and 14.9% of acceptors retesting positive, a nonsignificant risk reduction of 41.8%. When examining reinfection rates, stratified by sex, results suggest that the odds of retesting positive appear to be less for women than for men for both types of diagnoses (Ct and/or GC: 9.1% vs. 21.4%).
Easy access to patient health care information has been shown to improve health outcomes and reduce costs.15,16 Besides access, an EMR provides many other benefits including simplified electronic reporting, disease surveillance, and relatively quick evaluation of clinic process changes. HealthDoc, the EMR used by the DMHC, provides additional flexibility, allowing changes in what data are collected, installing skip patterns, and applying rules that require complete documentation before the record can be closed. Our study demonstrated that although clinician chart review slightly improved EPT uptake, structural changes to the EMR implemented in the last phase of the project had the greatest effect. During this time, the proportion of patients accepting EPT rose significantly and immediately to an average of approximately 50%, a rate that is fairly consistent with recent reports from the San Francisco City Clinic (43%),17 Baltimore City Health Department clinics (61%),18 and the Family Planning Clinic at the Columbia University Medical Center in New York City (69%).19 Although we do not know whether this proportion is the maximum achievable, it is important to note that many patients refuse EPT because their partners either have already been treated or are being treated concurrently, as was shown in our study.
There are some data in the literature describing demographic factors associated with EPT acceptance. Golden et al.11 randomly selected 1757 heterosexuals diagnosed as having chlamydia or gonorrhea and found that women were significantly more likely to receive EPT. There was also a suggestion that those diagnosed as having chlamydia were more likely to accept, results that are consistent with those found in our population. Stephens et al.17 reported on the provision of EPT to all patients, regardless of sexual orientation, seen in the San Francisco City Clinic between 2005 and 2008. Most patients included in this study were identified as MSM and thus represented a different population from our clinic, which excluded MSM. Nonetheless, interesting contrasts exist when considering demographic and sexual behaviors. In the Denver population, EPT acceptance was significantly associated with a chlamydia diagnosis (vs. gonorrhea), being female, and being white (among those diagnosed as having chlamydia), with no association with number of sex partners reported in the last 3 months. Conversely, San Francisco patients were more likely to receive EPT if black or if reporting multiple partners in the last 3 months (among those diagnosed as having chlamydia), but there was no association with type of diagnosis (chlamydia vs. gonorrhea) or sex.
Two additional associations emerged when examining EPT acceptance in the Denver population. Acceptance increased with the number of previous clinic visits, suggesting 3 possibilities. As the number of visits increases, patients may feel increasingly comfortable with the clinic providers and the increasing level of trust encourages acceptance. Conversely, an increasing number of visits may represent a higher-risk patient who simply appreciates the convenience of EPT. Finally, repeat visits may indicate repeat infections caused by untreated partners, prompting patients to accept EPT. However, in our data set, we were not able to address these hypotheses. Clearly, additional studies are needed to understand who does or does not accept EPT, why they do so, and target efforts among appropriate populations. A recent study of key informant interviews with EPT acceptors at Baltimore City Health Department clinics revealed the following factors to be important for EPT acceptance: a concern for partners, a recognition that the chain of transmission needed to be broken, and, among women, understanding an asymptomatic male partner’s reluctance to seek health care.18
Provider characteristics, also, are associated with EPT acceptance. In our study, nurses were more successful in dispensing EPT than non-nursing assistants. Whether this is related to patient perceptions of provider expertise, lack of time and training on the part of the non-nursing assistants or confidence of nursing staff is unclear and requires additional investigation. If the latter is the major factor in this relationship, changes in non-nursing staff training and clinic protocols could increase EPT acceptance among the refusers.
Reinfection rates among those who returned to the clinic for a retest suggest a benefit among those who accepted EPT. Although different from a recent study in San Francisco that found no protective effect with EPT use,17 our results are consistent with the results of the randomized control trials that showed a lower risk of reinfection with EPT use.9–11 Differences in reinfection rates among those accepting vs. not accepting EPT were not statistically significant in our study; however, this may be caused by a lack of power. It is important to point out that the randomized controlled trials did not consistently show a significant reduction for chlamydial infections and that the Golden study showed a much stronger effect for a reduction in gonorrheal infections,11 similar to our findings. In our clinic population, most of the gonorrhea diagnoses were among men (81.5%) who were less inclined to accept EPT than women. Staff and patient education could potentially increase EPT uptake rates among those who seem to receive the greatest benefit from this program.
The study is retrospective and therefore has limitations. Nine percent of the patient visits (233/2578) were among patients with more than 1 visit resulting in a positive test result for gonorrhea and/or chlamydia. Although these visits were treated as independent in the analyses, they are not, and presumably, this could have affected the standard errors and tests of significance. However, analyses using unduplicated patients (the first one of the period September 2007–April 2011) yielded similar results to those presented (data not shown). The number of previous sex partners was self-reported, and we were unable to validate this information. Record of previous clinic visits and previous and subsequent chlamydia or gonorrhea diagnoses were only those that we have recorded in our EMR. Similarly, we had no information on patients who did not return to the clinic for retesting; these patients may have had no need for retesting or may have sought care elsewhere. Finally, this was not a randomized controlled trial, and the reduction in reinfections among those accepting EPT may be caused by factors other than partner treatment.
The results of our study are in line with those from the randomized trials and, combined with our data on EPT uptake in response to various clinic interventions, should provide additional support for the effectiveness of this intervention.
1. Centers for Disease Control and Prevention. Expedited Partner Therapy in the Management of Sexually Transmitted Diseases. Atlanta, GA: US Department of Health and Human Services, 2006.
2. American Medical Association. Expedited Partner Therapy (Patient Delivered Partner Therapy): Report 7 on the Council on Science and Public Health (A-06). Chicago, IL: American Medical Association, 2006.
4. Position paper: Expedited partner therapy for adolescents diagnosed with chlamydia or gonorrhea: A position paper of the Society for Adolescent Medicine. J Adolesc Health 2009; 45: 303–309.
9. Schillinger JA, Kissinger P, Calvet H, et al.. Patient-delivered partner treatment with azithromycin to prevent repeated Chlamydia trachomatis infection among women. Sex Transm Dis 2003; 30: 49–56.
10. Kissinger P, Mohammed H, Richardson-Alston G, et al.. Patient-delivered partner treatment for male urethritis; A randomized controlled trial. Clin Infect Dis 2005; 41: 623–629.
11. Golden MR, Whittington WLH, Handsfield HH, et al.. Effect of expedited treatment of sex partners on recurrent or persistent gonorrhea or chlamydial infection. N Engl J Med 2005; 352: 676–685.
12. Kamb ML, Fishbein M, Douglas JM Jr, et al.. Efficacy of risk-reduction counseling to prevent human immunodeficiency virus and sexually transmitted diseases: A randomized controlled trial. Project RESPECT Study Group. JAMA 1998; 280: 1161–1167.
13. Colorado Medical Board, Policy Number 40-10: Appropriateness of Treating Partners of Patients with Sexually Transmitted Infections: [State of Colorado Department of Regulatory Affairs Web site]. May 10, 2001 (revised July 1, 2010). Available at: http://www.dora.state.co.us/Medical/policies/40-10.pdf
. Accessed November 19, 2011.
14. Colorado State Board of Pharmacy, Policy Number 40-4: Appropriateness of Labeling Prescriptions to Partners of Patients with Sexually Transmitted Infections: [State of Colorado Department of Regulatory Affairs Web site]. July 19, 2007. Available at: http://www.dora.state.co.us/pharmacy/policies/40-4.pdf
. Accessed November 19, 2011.
16. Thompson DI, Osheroff J, Classen D, et al.. Estimate the benefits of electronic medical records in hospitals and the need for a national benefits database. J Healthc Inf Manag 2007; 21: 62–68.
17. Stephens SC, Bernstein KT, Katz MH, et al.. The effectiveness of patient-delivered partner therapy and chlamydial and gonococcal reinfection in San Francisco. Sex Transm Dis 2010; 37: 525–529.
18. Temkin E, Klassen AC, Mmari K, et al.. A qualitative study of patients’ use of expedited partner therapy. Sex Transm Dis 2011; 38: 651–656.
19. Kerns J, Jones HE, Pressman EJ, et al.. Implementation of expedited partner therapy among women with chlamydia infection at an urban family planning clinic. Sex Transm Dis 2011; 38: 722–726.