Adolescent Sexually Transmitted Infection Screening in the Emergency Department (STI)

MedStar Health, Nationwide Children's Hospital, Children's National Research Institute, Children's Hospital of Philadelphia, Ann & Robert H Lurie Children's Hospital of Chicago, Children's Hospital and Health System Foundation, Wisconsin, Baylor College of Medicine, Children's Hospital Colorado, University of Cincinnati, Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD)

Sexually transmitted infections (STIs) are highly prevalent among adolescents. Despite established principles for STI control, clinical practices related to screening and diagnosis, treatment, and prevention of STIs among adolescents are suboptimal. There is an urgent need to expand our screening programs to nontraditional healthcare settings such as emergency departments (ED) to reach those adolescents who would otherwise not receive preventive healthcare, and to determine the most efficient and cost-effective method for providing this screening. The goal of this study is to leverage our recent insights obtained from single center ED-based adolescent gonorrhea and chlamydia screening research and apply them across a national pediatric ED research network to determine the most clinically effective and cost-effective screening approach for adolescents when implemented into a real-world clinical setting through a pragmatic trial. This will be accomplished through a network of children's hospital EDs with a track record of robust research collaboration (Pediatric Emergency Care Applied Research Network or PECARN). This intervention will rely on an innovative approach that electronically integrates patient-reported data to guide clinical decision support. The investigators will apply human factors modeling methods to perform ED workflow evaluations at each participating pediatric ED to determine the most efficient way to integrate the screening process into everyday clinical care. Following these analyses, the investigators will conduct a comparative effectiveness pragmatic trial of targeted STI screening (screening only those disclosing high risk sexual behavior) versus universally-offered STI screening (offered to all, regardless of risk) through electronic integration of patient reported data for provision of clinical decision support. The investigators will then develop decision analytic models to evaluate the cost-effectiveness of targeted screening compared to universally offered screening.

Adolescents are disproportionately affected by sexually transmitted infections (STIs). The STI epidemic among youth is a national public health priority as noted in the Healthy People 2020 objectives, which specifically recommend enhanced STI diagnosis, treatment, and reduction in adolescents. Failure to diagnose and treat STIs in a timely manner results in serious reproductive morbidity, mortality, and increased transmission of disease. Adolescents frequently access the emergency department (ED) for care, with the ED functioning as the primary source of health care for over 1.5 million adolescents. EDs are potentially high-yield sites for the delivery of sexual health services as part of a larger public health prevention strategy. Although the Centers for Disease Control and Prevention (CDC) recommend universal HIV screening in EDs, no recommendations currently exist for gonorrhea and chlamydia (GC/CT) screening. Addressing the effectiveness and integration of ED-based STI screening is critically needed. Insufficient knowledge of the ideal structure for delivery (i.e. universally-offered or targeted to high risk groups) is a barrier to the implementation of ED-based GC/CT screening. While universally-offered screening (offered to all, regardless of risk) may detect a larger number of cases than targeted screening (screening only those disclosing high risk sexual behavior), it is more resource-intensive and may result in more false positive cases due to a lower pretest probability of infection. The investigators each study targeted (Goyal) and universally-offered (Reed) ED-based GC/CT screening via electronically entered patient-reported data providing real-time clinical decision support (CDS). They have shown that both strategies are acceptable, feasible, and result in increased STI screening rates at their respective pediatric EDs; but it is unknown which method is most efficient and cost effective when instituted across a national sample of pediatric EDs with varying community STI prevalence. The scientific premise of this application is to leverage recent insights obtained from single center ED-based adolescent GC/CT screening research and apply them across a national pediatric ED research network to determine the most clinically effective and cost-effective screening approach for adolescents when implemented in a real-world clinical setting through a pragmatic trial. To be successful and sustainable, the ideal screening strategy must be easily incorporated into the clinical workflow. Electronic integration of patient-reported data with CDS offers one such solution with additional benefits: overcomes privacy concerns, circumvents provider biases about STI risk, and enables early detection of infection before development of adverse sequelae. The objective of this study is to compare targeted and universally-offered STI screening approaches by seamlessly integrating real-time CDS based on electronically-obtained patient-reported data into the normal ED workflow through human factors analysis followed by cost-effectiveness analyses. The investigators will execute a multicenter comparative effectiveness pragmatic trial within a national sample of pediatric EDs through the Pediatric Emergency Care Applied Research Network (PECARN). Specific Aim 1: To compare the effectiveness of usual care, targeted screening and universally-offered screening in EDs through a pragmatic trial that applies a human factors systems approach to implement GC/CT screening into routine clinical care. Specific Aim 2: To determine the most cost-effective approach for GC/CT detection (i.e. usual care, targeted screening, and universally-offered screening) in an adolescent ED population. Study Design: Workflow Evaluations: The investigators will first perform ED workflow evaluations to determine how to best integrate new processes into current workflow by applying a human factors workflow process mapping approach at each participating pediatric ED. Data from these methods will enable the development of site-specific adaptations for the implementation of electronically-enhanced STI screening processes that use patient reported outcomes (PROs) to generate CDS. This process is necessary to optimize the experimental environment by ensuring usability, which will be essential for adoption of this comparative effectiveness trial. Pragmatic Trial: The investigators will then conduct a comparative effectiveness pragmatic trial using a stepped wedge crossover design. The design also allows the investigators to control for the effects of secular trends (e.g. changes that are outside the interventions of the study). This was designed as a pragmatic trial to evaluate the performance of these screening strategies in routine clinical care, rather than under strict experimental conditions. Sites will be randomized to one of six positions. After collecting baseline data, each site will subsequently be randomized to either start with implementation of a targeted screening intervention (T) or a universally offered screening intervention (U). Each site will then cross-over and implement the opposite strategy. The final intervention will continue until the conclusion of the study. Cost-effectiveness analysis: Even if clinical strategies, such as universally offered screening for STIs, are proven to be effective, the costs of those strategies compared with usual care or more targeted approaches may make them impractical and unsustainable in a cost-constrained healthcare system. The result of a cost-effectiveness analysis is the incremental cost-effectiveness ratio (ICER), which describes the additional cost of the more expensive but more effective strategy being examined, in this case universally offered STI screening, compared with the lesser cost of the less effective strategy (we have hypothesized this will be targeted screening) divided by the measure of effectiveness being considered. For this project, the investigators plan to look at additional cost per STI detected, as well as cost per QALY gained. The investigators will develop decision analytic models (using Markov state transition structure) to evaluate the cost-effectiveness of universally offered screening compared with targeted screening. To be complete, "usual care" will be included as a third strategy. A simple model will only examine effectiveness in terms of STIs detected. A more complicated model will include short- and long-term consequences of treated and untreated disease. Short-term outcomes will include the likelihood of follow-up and treatment once a diagnosis has been made, along with the efficacy of antibiotic therapy. Based on test characteristics from the literature, small numbers of patients may have false positive or false negative test results. The investigators will model the consequences of both, in terms of additional costs and consequences of unnecessary treatment for false positives, and the costs and consequences of undetected/untreated STI in the case of false negatives. Outcomes also will be modeled for unscreened patients who later return for treatment of symptomatic disease (e.g., acute pelvic inflammatory disease). Long-term outcomes addressing the sequelae of untreated/ undetected STIs (tubal infertility, ectopic pregnancy, and chronic pelvic pain) and outcomes following successful detection and treatment will be modeled. Data generated from Aim 2 regarding the effectiveness of the two screening strategies in detecting GC/CT, the proportion of identified patients who ultimately receive treatment, and the prevalence ranges of GC and CT in cohorts presenting for ED care at the 6 sites in this study, will be used to inform probabilities in the model. Literature-based estimates will inform other probabilistic events and costs of long-term STI sequelae in the model. Incremental cost-effectiveness ratios will be calculated to determine cost-effectiveness. The analysis will use a health care sector perspective that includes medical costs borne by third-party payers and paid out-of-pocket by patients. These will include current and future costs related to both the intervention and downstream events related to treated or missed STIs. The investigators will perform a series of deterministic sensitivity analyses to explore the impact of parameter uncertainty on results, and known population variations. A key parameter of interest is the background rate of STI. This likely varies from location-to-location, and will be a key determinant to the cost-effectiveness of the universally offered screening strategy. The threshold of background STI prevalence above which universally offered screening becomes "very cost-effective" (i.e., an ICER < $50,000/QALY) will be calculated. As a result, the cost-effectiveness of the universally offered screening strategy may vary from location to location in the study, and in real-world use of the screening tool. Understanding this threshold may help shape policy for optimal screening strategies in different locations and settings. Study Procedures: Human Factors Workflow Analyses: Human factors workflow evaluations will occur at each of the 6 participating sites. These observations will be conducted to understand site-specific ED flow differences that may occur with respect to the care of adolescents. Observational data will be collected regarding ED workflow and information exchange using the tool TaskTracker to record clinical workflow and identify common workflow paths. These observations will be conducted during different times of the day to account for variation in ED volume and staffing fluctuations. Pragmatic Trial: Once the ideal workflow strategy is identified at each individual site, all participants in both the targeted and universally offered screening phases will use a handheld tablet device to provide electronic informed consent for participation. All participants will complete the previously developed and validated ACASI SHS containing questions regarding their personal sexual health history. The PROs tool will risk-stratify patients for STIs based on their reported sexual experience and/or presence of STI-related symptoms. Using the tablet device, patients will also provide consent for clinician-ordered STI testing (urine-based GC/CT). This will provide clinicians real-time EHR-integrated decision support for GC/CT testing. Since this is a pragmatic trial, the exact process of distributing tablets may differ by site and will be further clarified during the workflow analyses. The electronic informed consent document will contain information about the study. Reasons for not offering the tablet will be recorded by the ED staff. Furthermore, if adolescents decline study participation, a reason for refusal will also be recorded on the tablet. Targeted screening: During the targeted screening intervention, data from the SHS will be integrated into the EHR and will provide CDS for GC/CT testing based on SHS-calculated STI risk. Patients will be classified as at high risk for STIs if they disclose being sexually active and have either the presence of STI-related symptoms or any of the following high risk behaviors: more than 1 sexual partner in the last 3 months, no condom use during last sex, prior history of STI. Patients will be classified as at risk if they disclose being sexually active but do not disclose having any STI-related symptoms or any high-risk sexual behaviors. Patients will be classified as at low risk if they deny any history of sexual activity. When patients classify as at high risk, clinicians will receive CDS that STI testing is "highly recommended"; when they care for patients who classify as at risk, they will receive CDS that STI testing is "recommended"; when caring for patients who classify as at low risk, they will receive CDS that STI testing "is not necessary at this time." If the clinician chooses to follow the recommendation for screening based on patient's risk assessment, and the patient consents to testing on the tablet device, urine GC/CT testing will be performed. Universally offered screening: During the universally offered screening intervention, STI screening will be offered to all eligible adolescents, regardless of risk. All eligible patients will also complete the SHS, will be informed of the CDC GC/CT testing recommendations and then be given the option to decline STI testing using the tablet device. During this phase, the SHS results will not be available to the clinician. STI testing recommendations will be based only on the patient's decision to undergo GC/CT testing. Like the process followed in the targeted screening phase, if the clinician follows the CDS that informs the clinician that the patient agreed to GC/CT screening and consequently orders testing, urine GC/CT testing will be performed. Both approaches: If the patient is at risk for STIs and declines GC/CT testing, or the patient simply declines GC/CT testing on the tablet but the clinician believes the patient should be tested, the clinician will have the opportunity to engage in a shared decision-making process with the patient as per routine clinical care. Alternatively, if the clinician declines the CDS GC/CT testing recommendations, he/she will be asked to electronically document the reason the CDS is not being followed. All patients who test positive for an STI will be notified of their results and provided treatment based on each site's standard clinical processes for result notification and treatment.

Screening, Emergency Department, Adolescents, Sexually Transmitted Infections, Sexually Transmitted Diseases, Cost Effectiveness, Pragmatic Trial

A comparative effectiveness pragmatic trial using a stepped wedge crossover design will be conducted. A stepped wedge design involves the sequential roll-out of an intervention or a sequence of interventions to individuals or sites over multiple time periods. At the end of the study, all sites will have received all interventions, but the order in which sites are enrolled is randomly assigned. Once the final intervention is implemented, that intervention will continue until the end of the study.

Data from the Sexual Health Screen (SHS) will be integrated into the Electronic Health Record (EHR) and will provide Clinical Decision Support (CDS) for GC/CT testing based on SHS-calculated STI risk. Patients will be classified as at high risk for STIs, at risk or low risk if they deny any history of sexual activity. When patients classify as at high risk, clinicians will receive CDS that STI testing is "highly recommended"; when they care for patients who classify as at risk, they will receive CDS that STI testing is "recommended"; when caring for patients who classify as at low risk, they will receive CDS that STI testing "is not necessary at this time." If the clinician chooses to follow the recommendation for screening based on patient's risk assessment, and the patient consents to testing on the tablet device, urine GC/CT testing will be performed.

During the universally offered screening intervention, STI screening will be offered to all eligible adolescents, regardless of risk. All eligible patients will also complete the SHS, will be informed of the CDC GC/CT testing recommendations and then be given the option to decline STI testing using the tablet device. During this phase, the SHS results will not be available to the clinician. STI testing recommendations will be based only on the patient's decision to undergo GC/CT testing. Like the process followed in the targeted screening phase, if the clinician follows the CDS that informs the clinician that the patient agreed to GC/CT screening and consequently orders testing, urine GC/CT testing will be performed.

Comparison of GC/CT positive rates between patients in the high and low risk sexual strata (determined by the patient entered SHS)

Inclusion Criteria 15-21 years of age Exclusion Criteria unable to understand English critically ill cognitive impairment or altered mental status unable to provide consent for completion of the sexual health screen and STI screening.