Article Text

PDF

Short report
Impact of expedited partner therapy (EPT) implementation on chlamydia incidence in the USA
  1. Okeoma Mmeje1,2,
  2. Sarah Wallett3,
  3. Giselle Kolenic1,
  4. Jason Bell1
  1. 1 Department of Obstetrics and Gynecology, University of Michigan Medical School, Ann Arbor, Michigan, USA
  2. 2 Department of Health Behavior and Health Education, University of Michigan School of Public Health, Ann Arbor, Michigan, USA
  3. 3 Planned Parenthood Greater Memphis Region, Memphis, Tennessee, USA
  1. Correspondence to Dr Okeoma Mmeje, Department of Obstetrics and Gynecology, University of Michigan, L4100 Women's Hospital, 1500 East Medical Center Drive, Ann Arbor, MI 48109-5276, USA; ommeje{at}umich.edu

Abstract

Objectives The diagnosis and treatment of Chlamydia trachomatis infection is important in preventing persistent or recurrent infection. Expedited partner therapy (EPT) is the favoured and supported method for STI treatment of the Centers for Disease Control and Prevention when the provider cannot be assured that all recent sexual partner(s) will seek therapy. EPT is legally permissible in 38 states and is endorsed by healthcare organisations to decrease the rates of chlamydia and gonorrhoea infection. Our study investigated the impact of EPT legal status (permissible, potentially allowable or prohibited) on C. trachomatis infection rates for each state.

Methods Our ecological study modelled the number of reported chlamydia cases from 2000 to 2013 as a function of year, legal status and the interaction between year and legality. We used a negative binomial regression model that included state fixed effects (including the District of Columbia) to account for both the repeated measures per state and state-specific characteristics that could not be measured for inclusion in this study. The lagged number of C. trachomatis cases was included as a covariate and each state's total population for a given year was included in the model as an exposure parameter. States were designated Y (EPT permissible), N (EPT prohibited) and M (EPT potentially allowable), and the legal status of each state could vary over time.

Results Each legal category saw an increase in the incidence rate of C. trachomatis infection, but on average, the incidence rate for states with prohibitive EPT legislation grew significantly faster over time compared with the rate for the states where EPT was permissible. The average increase in predicted incidence rates per year for states with Y, N and M legal status were 14.1 (95% CI (12.0 to 16.2)), 17.5 (95% CI (15.9 to 19.2)) and 16.8 (95% CI (15.0 to 18.6)) cases per 100 000 persons per year, respectively, when controlling for state-specific effects.

Conclusions Our model suggests that a lack of EPT legislation is associated with an increase in STI rates. States with potentially allowable EPT legislation as of 2013 (n=8) should consider permitting EPT as a component of a multipronged strategy for treatment of sexual partners to prevent C. trachomatis infection.

  • CHLAMYDIA INFECTION
  • SEXUAL HEALTH
  • REPRODUCTIVE HEALTH
  • TREATMENT

Statistics from Altmetric.com

Introduction

Expedited partner therapy (EPT) is intended to decrease repeat Chlamydia trachomatis and/or Neisseria gonorrhoeae infection by ensuring that sexual partner(s) are treated to prevent persistent or recurrent infection. Medications or prescriptions are provided to the index partner for treatment of their sexual partner(s) without examination of the sexual partner(s) by the healthcare provider. EPT is recommended by the Centers for Disease Control and Prevention (CDC) as a method of treatment when the provider cannot be assured that the recent sexual partner(s), or those in the 60 days prior to diagnosis, will seek therapy.1 ,2 Dispensing medications to the index partner for his or her sexual partner(s) is the preferred method given limited data on the efficacy of only providing prescriptions. In the USA, the legal status of EPT is inconsistent with sound public health practice to ensure treatment and prevent infection given the concern for liability among healthcare providers. EPT legislation or administrative regulation by medical or pharmacy boards highlights the impact of healthcare policy on healthcare provider behaviour and practice in the setting of potentially allowable or prohibitive legislation.3 ,4 Therefore, EPT legislation or administrative regulation should support the treatment of sexual partners in an effort to increase the receipt of EPT by individuals infected with chlamydia and/or gonorrhoea. The use of EPT has been shown to decrease the rates of recurrent chlamydia and gonorrhoea infection.5 ,6

In 2000, 2 states had permissible EPT legislation, 38 states had prohibitive EPT legislation and 11 states had potentially allowable EPT legislation. In 2013, 34 states had permissible EPT legislation, 8 states had prohibitive EPT legislation and 9 states had potentially allowable EPT legislation. As of August 2015, EPT was permissible in 38 states, prohibited in 4 states and potentially allowable in 8 states. In the states prohibiting EPT, the legal implications may hinder partner treatment, resulting in higher incidence rates of C. trachomatis infection. We investigated the association of EPT legal status (permissible, prohibited or potentially allowable) on the reported rate of chlamydia cases for each state. Historical data were used to model future incidence rates of C. trachomatis infection. Our analysis will assess the legal status of EPT and the rates of C. trachomatis infection for each state, with the intent of guiding healthcare policies in the future.

Materials and methods

Data source

Using an ecological study design, C. trachomatis cases for each state from 2000 to 2013 were collected, along with the state's population and EPT legal status. The incidence of C. trachomatis cases was determined through the use of the CDC's National Center for HIV/AIDS, Viral Hepatitis, STD and TB Prevention (NCHHSTP) Atlas.7 The legal status of EPT over time was initially determined through the use of the CDC's website8 and confirmed through review of each state's EPT legislation.

Statistical model

A negative binomial regression with state-specific fixed effects was used to model the number of reported C. trachomatis cases (rates per 100 000) from 2000 to 2013 from CDC surveillance data 7 as a function of year, legal status and the interaction between year and legal status. State fixed effects (including the District of Columbia) were included in the model to control for both the repeated measures per state and any stable (time-invariant) state-specific characteristics that could not be measured for inclusion in this study. The lagged number of C. trachomatis cases was included as a covariate and each state's total population for a given year was included in the model as an exposure parameter, allowing the results to be interpreted as rates. The equation for the fitted model is provided below:

Embedded Image

States were designated Y (EPT permissible), N (EPT prohibited) or M (EPT potentially allowable), and the legal status of each state could vary from year to year. Significant model interaction terms indicate that the trajectory of reported cases over time depends on the legal status of the state. Incidence rates per 100 000 persons are reported. Stata V.14 was used for all analyses.

Results

Statistically significant main effects for year (p<0.001) and legal status (overall p=0.002) as well as significant interactions between year and legal status (overall p=0.005) were observed. All 50 state-specific fixed effects are also statistically significant (p<0.001 to p=0.006). These results, detailed below, indicate that C. trachomatis infection rates are estimated to increase over time for all three legal status groups, but the average increase in incidence rates over time for N states is significantly larger than that of Y states when controlling for stable state-specific effects.

The interaction terms in the model indicate a significant change in the slope for year depending on the EPT legal status of the state. C. trachomatis incidence increased in each category of legality, but on average, the incidence rate for states with N legal status grew significantly faster over time compared with the rate for the Y states. The average increase in predicted incidence rates per year for states with Y, N and M legal status were 14.1 (95% CI (12.0 to 16.2)), 17.5 (95% CI (15.9 to 19.2)) and 16.8 (95% CI (15.0 to 18.6)) cases per 100 000 persons per year, respectively, when controlling for stable state-specific effects. The average predicted incidence rate per year for Y and N states is significantly different (difference=3.4 cases per 100 000 persons per year, SE=1.2, p=0.004). On average, significant differences in the average yearly increase in predicted incidence rates were not found between Y and M states (difference=2.7 cases per 100 000 persons per year, SE=1.4, p=0.055) nor between N and M states (difference=0.74 cases per 100 000 persons per year, SE=1.3, p=0.558).

Estimated incidence rates per year by legal status are displayed in figure 1. When controlling for stable state-specific effects, the average predicted incidence rate for states with legal status Y in 2000 was 286.8 cases per 100 000 persons per year (95% CI (268.9 to 304.6), P≤0.001) and 255.5 cases per 100 000 persons per year (95% CI (247.8 to 263.2), P≤0.001) for N legal status states; this difference is statistically significant (difference=31.2 cases per 100 000 persons per year, SE=9.2, p=0.001). By 2013, the average predicted incidence rate for states with legal status Y increased to 466.9 cases per 100 000 persons per year (95% CI (454.0 to 479.7), p≤0.001) and to 477.9 (95% CI (460.8, 494.9), p≤0.001) for states with legal status N; this difference is not statistically significant (difference=11.0 cases per 100 000 persons per year, SE=8.8, p=0.210).

Figure 1

Adjusted incidence rate predictions of per year by legal status with 95% prediction intervals.

Discussion

Our analysis models the effect of EPT legislation on the incidence of reported cases of C. trachomatis infection over time in the USA. This analysis is important to understanding the effects of healthcare legislation on clinical practice and health-related outcomes.

Our model suggests that the effect of EPT legislation or regulation is associated with sharper average increases in C. trachomatis incidence over time. On average, C. trachomatis infection incidence rates were significantly higher in states where EPT is legally permissible compared with states where EPT was prohibited in 2000, but by 2013, there were no significant differences between incidence rates in states with permissible and prohibited EPT legislation. The increase in the permissibility of EPT has enhanced our ability to treat the sexual partner(s) of those infected with C. trachomatis. Most importantly, EPT legislation has helped to curb the incidence of persistent or recurrent C. trachomatis infection.5 Healthcare policies that support treatment of C. trachomatis with EPT may impact incidence rates. States with potentially allowable EPT legislation as of 2013 (n=8) should consider revising their state laws to permit the use of EPT.9 Five of the eight states with potentially allowable EPT legislation in 2015 had a chlamydia incidence rate >400/100 000 in 2013.7 States with a potentially allowable EPT legislation could foster ambiguity, which may contribute to inadequate treatment of sexual partners and increased incidence of C. trachomatis infection.9 These policies may influence the comfort of healthcare providers in offering treatment to partners who may not be under their care, ultimately impacting the quality of care patients receive.

Limitations

Our analysis is limited to reported cases of C. trachomatis infection because the use of EPT is limited to C. trachomatis in some states. Some states may have under-reporting of C. trachomatis infection due to changes in diagnostic screening and reporting practices rather than actual trends of decreased incidence. Morbidity trends are also likely to be influenced by changes in testing technology as laboratories expand their use of more sensitive tests (eg, nucleic acid amplification tests). Similarly, C. trachomatis screening efforts have increased nationally in high-risk populations.10 The statistical model presented is not robust to this potential measurement error, and as such, may not be based on the actual number of C. trachomatis infections in the USA from 2000 to 2013. Furthermore, the collected data are inclusive of all C. trachomatis infections reported in heterosexual and/or homosexual partnerships, which may affect our predicted cases, because the use of EPT is limited to heterosexual couples as recommended by the CDC.10

Unmeasurable stable state-specific covariates are controlled for in this model as long as the effects of these variables are also time invariant (eg, we assume state per capita income has the same effect on C. trachomatis infections in 2000 as in 2013). Failing to measure and appropriately model stable state-specific covariates that are not time invariant could lead to omitted variable bias. Additional time-varying state characteristics were not included in the statistical model, which could confound the relationship between C. trachomatis infections and EPT legal status and potentially result in omitted variable bias.

Our state-specific fixed effect statistical model may also have higher SEs than alternative modelling frameworks if the predictors of C. trachomatis infections have little variation over time for each state but vary greatly across states. As more states adopt EPT, the sample sizes for N and M states decrease, which widens CIs and makes statistical differences more difficult to detect. Finally, each state's EPT designation (ie, permissible, prohibited and potentially allowable) by the CDC is subject to legal judgement and interpretation, which underlies the CDC's classification scheme and may have influenced the CDC legal classification system we adopted for our analysis.

Public health implications

We believe that legislation permitting EPT will enhance support of healthcare providers offering EPT to their patients thereby, enhancing our ability to treat and prevent new cases of chlamydia infection. Although our model is not exhaustive, it provides evidence to support EPT as a STD treatment and prevention strategy. Future studies incorporating additional years of data and time-varying state variables in the model will help to further assess the impact of health policy legislation on STD incidence. Evaluating these trends in the coming years will be important as healthcare systems and public health departments develop policies and implement EPT in their healthcare facilities.

Acknowledgments

The authors would like to thank the faculty and research staff of the Program on Women's Healthcare Effectiveness Research of the University of Michigan's Department of Obstetrics and Gynecology. The authors acknowledge Dr Vanessa Dalton for her research mentorship and advice; Ms Sarah Block for her editorial and graphic design expertise.

References

View Abstract

Footnotes

  • The abstract was presented at the 2016 STD Prevention Conference of the Centers for Disease Control and Prevention.

  • Handling editor Jackie A Cassell

  • Contributors OM and GK led the writing and revision of the manuscript. SW, GK and JB contributed to the vision and design of the project. GK conducted the data analysis. All authors read and approved the final manuscript. ‘The corresponding author has the right to grant on behalf of all authors and does grant on behalf of all authors, an exclusive license (or non-exclusive for government employees) on a worldwide basis to the BMJ Publishing Group Ltd to permit this article (if accepted) to be published in STI and any other BMJPGL products and sub-licenses such use and exploit all subsidiary rights, as set in our license http://group.bmj.com/products/journals/ instructions-for-authors/license-forms’.

  • Funding This project was supported in part by the Society of Family Planning Research Fund Trainee Grant programme.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.