To develop and pilot-test a Web-based implementation of a team resource management (TRM) intervention to improve medication safety in primary care.

 Eight practices were randomized to either the Web-based TRM or usual practice (4 practices in each group). Primary outcome was adverse drug events (ADEs) in older adults, ascertained using a trigger tool chart review at two 12-month periods (before and after the intervention). The prospective TRM approach, designed to inculcate ownership and empowerment, facilitates systematic appraisal of risk and error reduction. This approach uses the highly adaptable and transferable Safety Enhancement and Monitoring Instrument that is Patient Centered.

 The rate of ADEs decreased from 25.8 to 18.3 per 100 patients per year in the intervention group. The rate was virtually unchanged in the control group (24.3 vs 24.8). In an analysis of covariance at the practice level, being in the intervention group was associated with a lower rate of ADEs. The interaction between time (preintervention vs postintervention) and group (intervention vs control) was not signifi cant (

 = .104) but showed a trend toward a decrease in the intervention group compared with the control group over time.

 The Web-based TRM intervention proved feasible and demonstrated potential for effectiveness in various ambulatory settings. This pilot study was limited by small size and short follow-up period. Future studies should test the intervention on a larger scale over a longer period of time and should explore methods for overcoming common barriers to change.

This randomized controlled trial pilot-tested a Web-based implementation of a self-empowering team resource management (TRM) intervention aimed at improving medication safety in primary care settings.

		 The intervention sites showed a significant reduction in adverse drug events.

		 Interaction between time (pre- vs postintervention) and the 2 arms of the study was not statistically significant.

		 This study suggests that the Web-based TRM has the potential to improve medication safety in busy primary care offices.

Medication use is recognized to be a high-risk activity across all settings. A 2007 Institute of Medicine (IOM) report on this subject acknowledges that the rates and impact of medication errors are huge but are poorly understood.

 In ambulatory settings, medication errors and adverse drug events (ADEs) are important safety issues. Gurwitz and colleagues have estimated (by extrapolation) that Medicare enrollees alone suffer approximately 500,000 preventable ADEs per year.

Lack of awareness of the type, incidence, and consequences of errors in any setting is one of the most important barriers to reducing these errors as well as improving safety and quality of care. The most commonly used method for estimating vulnerabilities in healthcare is to 

 collect and count errors through voluntary reporting systems (often referred to as incident reports). These are fraught with difficulty due to various issues, including underreporting; according to IOM’s 2000

report, only 5% of known errors are typically reported.3 Error reporting often does not promote understanding of the organizational structure and processes of care. Instead it tends to be associated with blame and shame, and frequently results in antagonism between team members that undermines mutual respect, trust, and cooperation. Bates and colleagues have described diffi culties involved in defi ning and quantifying errors; they report that even direct observational studies, which are highly labor intensive, often miss errors.

 rather than retrospective, and encourages involvement of all team members in identifying and prioritizing safety and quality problems, invokes failure modes and effects analysis. This type of analysis has been widely used in other high-risk industries and has been advocated by the IOM as a means of analyzing a system to identify its weaknesses (failure modes) and possible consequences of failure (effects), and to prioritize areas for improvement.3 We adapted and tailored this methodology to allow for the levels of resources and expertise available in ambulatory settings and developed an instrument that has been well received by staff in a variety of settings. The details of the concepts, rationale, and processes behind this highly transferable instrument, termed Safety Enhancement and Monitoring Instrument that is Patient Centered (SEMI-P), are described elsewhere.

The objectives of this study were to develop and pilot-test an information technology (IT)—based team resource management (TRM) system, based on SEMI-P, aimed at improving medication safety in primary care. We examined (1) the ability of this intervention to reduce selected ADEs among geriatric patients, (2) the ability of this intervention to improve monitoring of geriatric patients taking selected chronic medications, and (3) how offi ce staff used and applied this IT-based TRM tool for improving geriatric medication safety.

This was a randomized controlled trial of a Web-based TRM intervention to reduce ADEs in primary care. Randomization was at the site level; 4 sites were assigned to the intervention and 4 to a control state (usual practice). In all 8 sites, ADEs were ascertained using a previously published trigger tool methodology.

All practices in the current IT-facilitated study were part of the Upstate New York Practice Based Research Network and had electronic medical records in place for at least 12 months prior to the start of the study. Both groups contained a variety of practice types including safety net practices. Urban, suburban, and rural practices of various sizes and with various ownership structures were represented. All staff at the above sites—including physicians, physician extenders, nurses, medical assistants, administrative staff (secretarial and management), and all others (eg, dieticians and social workers if present)&mdash;were invited to participate in surveys and team discussions.

The project addressed the IOM priority areas of frailty associated with old age and medication management. The study protocol was approved by the Social and Behavioral Sciences Institutional Review Board of the State University of New York at Buffalo.

The intervention involved implementation of the Webbased TRM system in the 4 practices that were randomly assigned to this group. The system uses a cyclical safety improvement process facilitated by use of 2 anonymous online staff surveys. The fi rst survey instrument is SEMI-P, which is a failure modes and effects analysis tool designed for the ambulatory setting that focuses on medication management processes.

 The second instrument is the safety attitudes questionnaire— ambulatory version (SAQ-A), which provides measures of the safety climate.

Among the published safety climate surveys available at the time of this study, the SAQ was determined by Colla and colleagues to have the best psychometric properties.

 For each survey, online instructional clips and automated analysis with visual presentation of results were developed and implemented as part of the IT system.

 is a screen shot of the first of 12 pages of the online SEMI-P survey. Each member of every intervention practice (including all job categories) was invited to complete this survey anonymously as part of a staff meeting. The same procedure was followed for the SAQ-A.

At each site, after administration of each survey, the practice teams immediately regrouped and reviewed the results and commenced discussion. 

 is a screen shot showing the compiled results of the SEMI-P, generated by the system in real time. 

 shows example results from the SAQA in a visual format that was developed especially for this project to highlight strengths and weaknesses in the safety culture.

These steps were followed by a series of staff meetings, held during times when no patient care was being delivered (generally at lunchtime), which all available staff attended. At these meetings, the survey results were reviewed and discussed, leading to consensus-based prioritization of medication safety issues. This consensus was also aided by visualized displays of individual (anonymous) team members’ opinions. Examples of prioritized areas included poor patient education about medications, high no-show rate, poor medication tracking, and poor coordination/teamwork with respect to handling of medication refi ll requests.

In subsequent staff meetings, teams worked together to address the chosen priorities by developing feasible system changes to improve medication safety. Examples include consistent use of patient education materials for high-risk medications, inclusion of diagnosis on prescriptions, patient reminders regarding follow-up, patient-carried medication lists, redesign of medication refi ll work fl ow, better training and follow-up for new personnel, and better employee performance feedback. The figures in the 

) illustrate the visual aid to consensus forming and the overall view of the whole methodology.

The initiatives tool within the software was used by staff to defi ne their goals and objectives for safety improvement, identify and assign specific work steps to individual team members, track progress, coordinate meetings, and remind staff of their commitments. The indicators tool was used to define specific measurable outcomes related to each initiative and to graphically track these over time in order to determine whether the stated objectives were being met.

The primary outcome at the 4 intervention and 4 control sites was the rate of ADEs (measured using a trigger tool methodology). A secondary outcome was compliance with Healthcare Effectiveness Data and Information Set (HEDIS) guidelines for laboratory monitoring for patients who were prescribed certain medications chronically (meaning that they were prescribed the medication for 6 or more months out of a 12-month period). Both of these outcomes were for older adults (aged 

65 years) since these patients are known to be at higher risk of adverse events.

A Web-based data capture tool was developed and implemented as part of the IT system for both the trigger tool and the HEDIS measure.

These 2 outcomes were ascertained for a baseline period defi ned as the 12 months before the start of the intervention and an end point period defi ned as the 12 months after the start of the intervention. For each period (baseline and end point), research assistants reviewed an independent sample of 100 charts of patients aged 

65 years at each site. The HEDIS laboratory monitoring measurement was completed by these research assistants. For the trigger tool, the research assistants conducted the fi rst of 2 steps, known as the screening step. Charts identifi ed in this fi rst step as having triggers underwent secondary review (the review step) by a physician or pharmacist who reviewed each trigger to determine whether an adverse drug event occurred, and if so, its severity and preventability.

Rates of ADEs were calculated for each practice, normalized per 100 patient-years at both the preintervention and postintervention time periods. Chi-square and

 tests were used (where applicable) to determine baseline differences in patient characteristics between the intervention and comparison groups. Variables with significant baseline differences were included as factors in an analysis of covariance (ANCOVA). This analysis was at the site level. All analyses were performed using IBM SPSS version 19.0 (IBM Corporation, Armonk, NY).

Articles

METHODS

Self-empowering team resource management, when aided by information technology, appears to help reduce adverse drug events in primary care offices.

IT-Enabled Systems Engineering Approach to Monitoring and Reducing ADEs

To identify the factors and strategies that were associated with successful implementation of hospital-based information technology (IT) systems in US Department of Veterans Affairs (VA) hospitals, and how these might apply to other hospitals.

Qualitative analysis of 118 interviews conducted at 7 VA hospitals. The study focused on the inpatient setting, where nurses are the main patient-care providers; thus, the research emphasized the impact of Computerized Patient Record System and Bar Code Medication Administration on nurses. Hospitals were selected to represent a range of IT implementation dates, facility sizes, and geography. The subjects included nurses, pharmacists, physicians, IT staff, and managers. Interviews were guided by a semi-structured interview protocol, and a thematic analysis was conducted, with initial codes drawn from the content of the interview guides. Additional themes were proposed as the coding was conducted.

Five broad themes arose as factors which affected the process and success of implementation: (1) organizational stability and implementation team leadership, (2) implementation timelines, (3) equipment availability and reliability, (4) staff training, and (5) changes in work fl ow.

Overall IT implementation success in the VA depended on: (1) whether there was support for change from both leaders and staff, (2) development of a gradual and flexible implementation approach, (3) allocation of adequate resources for equipment and infrastructure, hands-on support, and deployment of additional staff, and (4) how the implementation team planned for setbacks, and continued the process to achieve success. Problems that developed in the early stages of implementation tended to become persistent, and poor implementation can lead to patient harm.

The information technology (IT) implementation experience of a large diverse system, like the US Department of Veterans Affairs (VA), which has a mix of rural, urban, teaching, and community hospitals, provides valuable information about the issues surrounding implementation for hospitals in the private sector. IT implementation success in the VA depended on:

		Whether there was support for change from both leaders and staff.

		 Development of a gradual and fl exible implementation approach.

		 Allocation of adequate resources for equipment and infrastructure, hands-on support, and deployment of additional staff.

		 How the implementation team planned for setbacks, and continued the process to achieve success.

Computerized patient records and bar-code medication systems continue to gain favor in healthcare.

 In the hospital setting, 2 of the most important technologies are computerized patient records and medication administration systems.

 These systems are expected to bring about improvements in patient safety, work processes, and staff morale,6-11 and reduce rates of medication errors. 

 Research to date has reached mixed conclusions as to whether such improvements have occurred.

 Several studies demonstrate that the implementation process for hospital health information technology (HIT) is important to determining overall success.

The US Department of Veterans Affairs (VA) has made one of the largest investments in HIT in the United States, implementing a fully integrated system across its 162 hospitals nationwide.

 The VA is the nation’s largest integrated healthcare system, with more than 7.2 million veterans enrolled for health services.

 Their development of an HIT system can be traced to the 1970s. The VA’s Computerized Patient Record System (CPRS) and Bar Code Medication Administration (BCMA) are central to inpatient care.

CPRS was phased in over a decade starting in the early 1990s, and consists of a comprehensive electronic patient medical record with computerized physician ordering, covering both outpatient and inpatient services. By 2002, all VA hospitals had implemented CPRS; the vast majority had implemented this system by 2000. BCMA, on the other hand, was implemented over a much shorter time period, with VA headquarters requiring implementation 1 year after the software became available.

 This system created a computerized pharmacy ordering, distribution, and administration system for use in the inpatient setting. Bedside scanning of patient identifi cation wristbands and medications was the key component of the system, providing validation that each medication matched the orders for each patient.

This paper reports the findings from the qualitative component of a national, retrospective, mixed-methods study of the implementation of CPRS and BCMA. There were no prospective systemwide or multisite evaluations of the implementation of CPRS or BCMA. The study focused on the inpatient setting, where nurses are the main patient care providers; thus, the research emphasized the impact of CPRS and BCMA on nurses. The qualitative portion of the study focused on 4 issues: (1) understanding the approaches used to implement the VA’s HIT systems, (2) identifying factors that affected the process of implementation, (3) ascertaining the issues that determined success of implementation, and (4) understanding what nursing staff and leaders believe are the current strengths and weaknesses of CPRS and BCMA. The experience of a large diverse system like the VA, which has a mix of rural, urban, teaching, and community hospitals, provides valuable information about the issues surrounding information system implementation for hospitals in the private sector.

We conducted site visits at 7 VA hospitals selected to represent a range of implementation timelines, geography, and staff characteristics. All site visits were conducted after receiving approval from each facility’s Institutional Review Board (IRB), as well as the IRBs of the University of California, San Francisco, Stanford University, and the Boston VA Healthcare System.

In advance of site visits, an Advisory Committee was assembled, consisting of VA medical, pharmacy, and nursing leaders, as well as representatives of the VA headquarters. The committee was asked to identify issues and themes that they anticipated to be important to understanding the impact of the implementation and use of the VA’s HIT systems on nurses. A semi-structured interview guide was developed after this meeting, based on a review of the literature on technology implementation and the effects of IT systems, and the committee’s suggestions. The committee members provided feedback regarding the content of the guide. This guide was used both to conduct the interviews and provide themes for the initial coding of the data. The guide is available in 

Site selection was based on a unique Web-based survey of VA facilities documenting when each major component of CPRS and BCMA was implemented, VA staff satisfaction survey data, facility-level staff turnover data, geography, and the level of care provided by each VA hospital. We initially identifi ed 27 facilities that represented a range of characteristics, and then consulted with the Advisory Committee and the VA Headquarters Nursing Offi ce to recruit a final set of 7 facilities.

A total of 118 interviews were conducted over a 15-month period (June 2006 through September 2007) with nurses, pharmacists, nurse managers, information technology staff, and senior management. At most sites, physicians were not included because the study’s focus was inpatient IT systems, which are primarily used by nurses. Interview subjects were selected by site coordinators, who were recommended by members of the Advisory Committee and/or the VA Headquarters Nursing Offi ce, and were employees of the VA facilities. The principal investigator (PI) provided the site coordinators with a detailed list of the job classifications of the people to be interviewed (see 

 for categorizations and counts of the interview subjects). Interviews lasted 30 to 60 minutes and were held in private meeting and conference rooms, using the semi-structured interview protocol. Notes were taken by the investigators and were entered into ATLAS.ti to facilitate analysis after the site visit was complete.

A thematic analysis was conducted with initial codes drawn from the content of the interview guides. Additional themes were proposed as the coding was conducted and were added if there was concurrence among members of the research team. All coding was completed by the PI, while other members of the research team reviewed codes of 1 to 2 interviews per site. Investigators who visited each site reviewed the themes identifi ed from that site’s interviews.

Staff and managers faced numerous challenges while CPRS and BCMA were being implemented. As one VA employee stated, these IT systems changed “how we organize, document, and communicate regarding patient care,” changes that touched all aspects of healthcare delivery. The all-encompassing scope of IT implementation led to what another staff member described as “a big culture change.” This process of change can be tumultuous. A number of staff and managers used terms such as “frightened,” “nervous,” and “scary” to describe how nurses felt about CPRS and BCMA at first; one manager said there was “reluctance by many staff members to truly embrace the system, see the opportunities.” Another manager said, “a lot of clinical staff thought if they didn’t use CPRS it would go away.” The most resistant nurses and physicians reportedly left the VA through retirement or turnover.

Managers and staff agreed that employees who were not “tech savvy” had more diffi culty adapting to HIT; one manager noted that “less computer-skilled nurses struggle more.” Another manager observed that “some staff didn’t know how to use computers or mice,” which hampered their training. Many managers also noted that clinicians “tend to have weak keyboarding skills” because their jobs did not normally require much typing. At some sites, managers reported that older nurses were less likely to be comfortable with computers.

Five themes arose as important infl uencers of the process and success of implementation: (1) organizational stability and implementation team leadership, (2) implementation timelines, (3) equipment availability and reliability, (4) staff training, and (5) changes in work fl ow.

Organizational Stability and Implementation Team Leadership

The broader culture of the facility fundamentally affected the success at each site. One VA employee said, in a “large organizational deployment, [the organization] needs [to be] very stable and fault-tolerant… In planning, [you] have to have good leadership to articulate the nursing position.” Sites with unsupportive management teams, or where staff did not respect the ability of management, faced more challenges both during and after implementation of BCMA and CPRS. Managers often set the tone for how their departments or staff accepted HIT. As one leader noted, “if nurse managers were in support, you could get a lot farther.”

The teams that led the CPRS and BCMA implementations were crucial to success. For CPRS, the medical staff was usually viewed as most important, at least in part because CPRS has a central role in the outpatient setting. At every site we visited, staff and leadership recognized that nursing involvement in the BCMA implementation was crucial, and that pharmacy and IT staff had to be partners in the process. As one manager said, “success is all about teamwork.”

At some sites, the responsibility of implementation was not accepted by all parts of the team. For example, at 2 sites, pharmacists thought BCMA was primarily a nursing program, and thus “meetings didn’t go anywhere.” These sites had more diffi culty, because problems were not addressed in a cohesive way. Several staff members and managers noted that implementation of both systems required involvement of end users; sites that had physicians and staff nurses in visible roles during implementation achieved buy-in from other care providers more easily.

Flexibility in implementation helped staff adapt to CPRS and BCMA, regardless of how the implementation was specifically planned. CPRS evolved over time, with its antecedents developed in the early 1980s. There was no implementation deadline for CPRS, and most VA facilities used a gradual unitby- unit rollout to implement CPRS. One nurse noted that, at her site, “not much prep work was done before implementation, but that was somewhat okay because it was phased in.”

In contrast, the implementation of BCMA was mandated by VA headquarters with a June 30, 2000, deadline, which was only 1 year after the software was made available. The short timeline for BCMA implementation made staff feel pressured to use a system that they perceived had fl aws. One staff member argued “the software wasn’t ready, and the hardware had not been researched,” making it a “big mistake” to require implementation at that time. Some sites introduced BCMA in departments with relatively stable patient populations, such as in the long-term care ward. Other sites selected psychiatric units for the initial implementation, because patients go to a specifi c desk to receive their medications. In these wards, mobile carts and wireless scanners were not needed to operate BCMA, thus reducing the complexity of the implementation. These approaches seemed to work well because they gave the implementation team the opportunity to gain experience with the system in a controlled environment.

For both CPRS and BCMA, VA facilities were allowed to select their own hardware. With CPRS, a variety of confi gurations could be used, with mobile computer carts, a centralized station, or bedside computers being acceptable options. Adequate access to and reliability of hardware and computer networks were essential during the implementation of CPRS. Many of the people we interviewed reported problems in this area, and staff spent much time and energy dealing with computer issues. As one interviewee said, “there were also system issues at fi rst; not enough laptops, problems with wireless access, inadequate server. CPRS itself wasn’t causing an increase in needed RN hours. But, the IT problems were causing problems that increased RN hours.” The question of how many computers would be adequate for use of CPRS does not have a clear answer. During the implementation period, when staff members were unfamiliar with the system and needed more time to become familiar with it, perceived shortages of computers were frustrating to them.

Network problems plagued some facilities. One person observed that, at their site, “when the system was fi rst rolled out, there wasn’t enough wireless bandwidth, and the network coverage was spotty in some patient-care areas.” This led to the system being perceived as “very slow.” Some interviewees attributed speed problems to the computer servers not having enough capacity or being defi cient. Network and server problems were reported that would disconnect users or “would lock up or kick people out, so notes were lost.” Over time, these problems were resolved because demand on the system declined as the staff became more effi cient with its use, as well as because hardware and networks were upgraded.

Hardware and network problems were even more signifi - cant for BCMA. Sites needed mobile carts that could carry medications, scanners, and computers. The computers had to connect to the hospital’s server, thus hospitals had to use wireless networks. They also needed to purchase bar-coded wristbands for patients and develop a method to ensure that all medications in the pharmacy were bar-coded. There was no guidance from headquarters about equipment, so every site scrambled to do its own research and decision making.

MANAGERIAL

This paper reports findings from a qualitative analysis of US Department of Veterans Affairs hospitals on factors affecting success in implementing 2 information technology systems.

What Determines Successful Implementation of Inpatient Information Technology Systems?

 To compare adherence between oncedaily (QD) and twice-daily (BID) dosing with chronic-use prescription medications used by patients with cardiovascular disease.

Analysis consisted of 1,077,474 patients aged 

18 years with a prescription index date from January 1 to December 31, 2007, for an antidiabetic, antihyperlipidemic, antiplatelet, or cardiac agent with QD or BID dosing. Adherence (medication possession ratio [MPR]) was the number of days of medication supplied between the first prescription fi ll date and the subsequent 365 days divided by 365 days. Overall mean MPR and comparisons between dosing frequency groups were assessed with a generalized estimating equation. Covariates included age at index date, gender, Charlson comorbidity index, therapeutic class, dosing frequency, and the interaction between therapeutic class and dosing frequency group.

Overall, the adjusted mean MPR ± standard error (SE) value for QD agents was 13.6% greater than BID agents (0.66 ± 0.0006 vs 0.57 ± 0.0016; 

 <.01). The adjusted mean MPR value for QD agents was 2.9%, 17.5%, and 29.4% greater than BID agents in the antidiabetic, antihyperlipidemic, and antiplatelet therapeutic classes, respectively. For cardiac agents, the adjusted mean MPR value was similar between QD and BID agents. Carvedilol represented approximately 80% of the cardiac agents in the BID group. The adjusted mean MPR ± SE for carvedilol phosphate QD was 0.73 ± 0.0024 and 0.65 ± 0.0027 for carvedilol BID (11% difference; 

 In this large analysis, the QD dosing regimen was related to greater adherence versus a BID regimen.

The current study is the first large-scale effort to validate the often-held assumption that the complexity of a cardiovascular dosing regimen impacts medication adherence.

		 Across the 4 therapeutic classes studied (antidiabetic, antihyperlipidemic, antiplatelet, and cardiac), adherence for once-a-day regimens was 14% higher than for twice-daily agents.

		 Regardless of dosing regimen, adherence for many patients remains suboptimal.

		 Providing patients with access to simplifi ed dosing regimens may be an important factor in maximizing therapeutic success.

Cardiovascular disease (CVD) continues to be the leading cause of mortality in the western world. In the United States alone, more than 830,000 people die each year from CVD, which accounted for 34.3% of all deaths in 2006.

 The prevalence of CVD in the US population is staggering, with 17.6 million, 5.8 million, and 6.4 million individuals living with coronary heart disease, heart failure, or a history of stroke, respectively.

 These numbers are likely to rise with the increasing elderly population, which is anticipated to be 20% of the US population by 2030.

Over the last decade, numerous clinical trials of medications have demonstrated signifi cant benefi t in treating patients with CVD and associated risk factors, such as diabetes mellitus, and are now highly recommended in clinical guidelines.

For these medications to provide their benefi ts, patients need to be adherent to the prescribed regimen. Adherence, as defi ned by the International Society for Pharmacoeconomics and Outcomes Research, is “the degree or extent of conformity to the recommendations about day-to-day treatment by the provider with respect to the timing, dosage, and frequency. It may be defi ned as ‘the extent to which a patient acts in accordance with the prescribed interval and dose of a dosing regimen.’”

 While this is one defi nition of adherence, it can basically be explained as “Does the patient take the medication as prescribed?”

Despite evidence of signifi cant clinical benefit, for patients with CVD, adherence is often suboptimal.

 It has been estimated that 40% to 50% of patients with CVD may be nonadherent, and this reduction in adherence may be seen at approximately 3 to 6 months of therapy.

Medication nonadherence is a commonly recognized source of adverse patient outcomes and use of healthcare and associated costs.

 Studies have demonstrated that nonadherence to therapies for hypertension, dyslipidemia, or chronic stable angina results in greater morbidity, increased risk for hospitalization, or greater mortality. 

For example, data from the PREMIER (Prospective Registry Evaluating Myocardial Infarction: Events and Recovery) registry have demonstrated that lack of adherence with the antiplatelet agent clopidogrel in patients with acute myocardial infarction was associated with a 10-fold increase in mortality (0.7% vs 7.5%; 

 <.0001) and increase in cardiac hospitalizations (14% vs 23%; 

A number of factors may infl uence adherence, one being the complexity of the medication regimen.

 Complexity of a prescribed medication regimen has been shown to be inversely related to patient adherence.

 The complexity consists of 3 major domains: the number of medications prescribed, the complexity of administration, and daily dosing frequency.

 For patients with CVD, who often take multiple medications, including the use of combination products, it can be diffi cult to signifi cantly reduce the number of medications needed for optimal patient outcomes. The complexity of the administrations is not typically an issue in the management of chronic CVD since most medications are taken orally. Subsequently, the simplifi cation of daily dosing frequency may have the most potential to improve drug adherence in patients with CVD.

While it may seem intuitive that a simplifi ed dosing schedule would improve adherence, the literature investigating the relationship between adherence and daily dosing frequency is limited to smaller studies that are fairly old and meta-analyses of these previous studies.

 The research has suggested that patients are more adherent to drugs with a once-daily (QD) dosing schedule compared with drugs requiring more frequent dosing, but there are concerns about the generalizability of the results. Another major limitation of the current data is that much of the data focus on a single drug or drug class. Additionally, the data are limited to the assessment of adherence of a single therapeutic class, such as only hypertension, dyslipidemia, or diabetes. There have been studies that have compared medication adherence across a variety of different disease states, but these analyses often included disease states that are unrelated, such as epilepsy, hypertension, and osteoporosis. 

 Recognizing the scarcity of broad-based multitherapeutic evidence related to adherence and the need to generate relevant data for the patients with CVD, this study attempts to estimate adherence from the broader multi-therapeutic perspective of classes of medications that are commonly used by patients with CVD. The objective of this study is to compare patient adherence between QD and twice-daily (BID) dosing with chronic-use prescription medications commonly used by patients across the spectrum of CVD with a large, recent, commercial database.

The study data are from the Thomson Reuters MarketScan Research Databases (Ann Arbor, Michigan). The MarketScan Commercial Claims and Encounters plus the Medicare Supplemental databases are large insurance claims databases with both working age employer-sponsored insurance and patients with Medicare Supplemental Insurance. Pharmacy claims contained quantities of drug dispensed, days supply, and costs.

Data were from July 1, 2006, to December 31, 2008. The first prescription fi ll date, or index date, for 1 or more of the drugs of interest, as described below, occurred between January 1 and December 31, 2007, and prescription data needed to be available for the entire period of time between 6 months prior to and 12 months after the index date. The unit of analysis was drug therapy for 1 year; therefore, patients were permitted to have prescription claims for more than 1 drug in this analysis.

Drugs were selected for inclusion in the study if they were in capsule or tablet form and could be grouped into 1 of the following therapeutic classes: antidiabetic (eg, metformin, thiazolidinediones, sulfonylureas), antihyperlipidemic (eg, statins, fi brates), antiplatelet (eg, clopidogrel, aspirin/dipyridamole, cilostazol), or cardiac agents (eg, ACE inhibitors, antiarrhythmic agents, alpha-beta blockers, calcium channel blockers). For each prescription claim, the quantity of pills per day was calculated. A drug was designated as having either QD or BID dosing if 

80% of the claims had a quantity per day of either 1 or 2, respectively.

Patients in the analysis were 18 years or older and were required to be continuously enrolled in a health plan for inclusion in the study. Patient baseline comorbidities were quantified with a simple unweighted Charlson index score.

Patients with catastrophic healthcare costs (eg, index drug costs greater than $1000, pre-index prescription costs greater than $10,000, or pre-index medical costs greater than $100,000) were excluded from the analysis in order to capture a typical healthcare experience. Patients with prescription claims for less than a 30-day supply were excluded.

The adherence measure used in this analysis was the medication possession ratio (MPR). The MPR was defi ned as the number of days of medication supplied from the fi rst prescription fill date up to 365 days divided by 365 days. Calculations of MPR greater than 1.0 were set to 1.0.

The primary objective was to compare the mean MPR values between the QD and BID groups. A generalized estimating equation (GEE) was used to estimate the mean MPR and to compare the dosing frequency groups. The GEE accounted for the correlations resulting from a patient presenting data for multiple drugs either within a therapeutic class or in different therapeutic classes. An exchangeable correlation structure was used in the GEE. Alternative to mean, MPR greater than 80% has been suggested as a measure of adherence and was also calculated. To account for differences in patient demographic and clinical characteristics, the multivariate model adjusted for the potential confounding variables (age at index date, gender, and Charlson comorbidity index) and included therapeutic class, dosing frequency (QD or BID), and the interaction between therapeutic class and dosing frequency group.

The mean MPR adjusted for the model covariates and the standard error (SE) are presented, unless noted otherwise. P values less than or equal to 0.05 were considered statistically significant. No adjustments were made for multiplicity. All analyses were conducted with SAS (SAS Institute, Inc, Cary, North Carolina) version 9.2.

The sample sizes for the analysis, after applying the study inclusion and exclusion criteria, are presented in the 

. The final analysis database contained 1,440,254 (QD: 1,384,226 and BID: 56,028) therapy observations. This represented 1,077,474 patients, and their baseline characteristics at the index event for a therapy are presented in 

. The average age of patients was 59 years for patients taking QD medications and 61 years for patients with BID medications. Approximately 45% and 30% of the patients were in the Southern and North Central regions of the United States, respectively. Across the therapeutic classes, the average Charlson index score ranged from an average of 0.5 to 1.6 and the number of concomitant medications ranged from an average of 3 to 6. Approximately 80% of patients who took antidiabetic agents had a diagnosis of diabetes. Of patients in the antiplatelet agent class with a QD dosing frequency, 65% had a diagnosis of ischemic heart disease, and approximately 50% of patients in the cardiac agent class had a diagnosis of hypertension. For all patients, one-third had a diagnosis of hyperlipidemia. Approximately 50% of patients participated in a preferred provider organization insurance plan. For healthcare costs, the average cost of the index drug ranged from $30 for the antihyperlipidemic agent class with a BID dosing frequency to $255 for the antidiabetic agent class with the QD dosing frequency. Inpatient and outpatient pre-index medical costs ranged from an average of $3041 for patients in the antidiabetic agent class with a BID dosing frequency to $18,522 for patients in the antiplatelet agent class with a QD dosing frequency.

Overall, the adjusted mean MPR ± SE value for QD agents was 15.8% greater than BID agents (0.66 ± 0.0006 vs 0.57 ± 0.0016; 

). The adjusted mean MPR values for the QD agents were 3.0%, 21.2%, and 41.7% greater than for BID agents in the antidiabetic, antihyperlipidemic, and antiplatelet therapeutic classes, respectively (Table 2). In the antidiabetic therapeutic class, most of the QD agents were composed of pioglitazone and sitagliptin, which had MPRs of about 0.70. The BID agents in the antidiabetic therapeutic class were largely metformin combination products with an MPR ranging from 0.54 to 0.72. In the antihyperlipidemic therapeutic class, the QD agents were chiefl y statins with a wide range of MPRs (0.50 to 0.67), while the BID agents were predominantly composed of gemfi brozil (MPR 0.50). The antiplatelet therapeutic class demonstrated the largest difference between QD and BID agents. This therapeutic class had a larger percentage of BID agents compared with QD agents versus the other therapeutic classes. The principal QD agent in the antiplatelet class was clopidogrel with an MPR of 0.71. The BID agents were primarily composed of aspirin/dipyridamole, cilostazol, and ticlopidine, which had MPRs of 0.56, 0.43, and 0.50, respectively.

Conversely, for the cardiac agent class, the adjusted mean MPR value was similar between QD and BID agents (0% difference). Further analysis of this therapeutic class identified that carvedilol represented approximately 80% of the total BID group of cardiac agents. Since carvedilol is also available in a QD formulation we conducted a comparison between the 2 carvedilol formulations. A GEE was also used to model carvedilol data. The model contained age, gender, and Charlson comorbidity index. There were a total of 36,081 therapy observations of carvedilol in the database (18,619 for carvedilol phosphate QD and 17,462 for carvedilol BID) among 34,339 patients. The adjusted mean MPR ± SE for carvedilol phosphate QD was 0.73 ± 0.0024 and for carvedilol BID it was 0.65 ± 0.0027, representing a 12.3% (

Comparisons were also completed using an MPR of greater than 80% adherence defi nition for the overall group and each subgroup (

). These results are consistent with the overall findings of a signifi cantly higher rate of adherence with the QD group compared with the BID group. The subgroup fi ndings were also consistent, with a significantly higher rate of adherence with the QD group compared with the BID group for antidiabetic, antihyperlipidemic, and antiplatelet agents, and no difference in the general group of cardiac agents.

Using a large pharmacy claims database, we identifi ed a 15.8% higher rate of adherence to medications with a QD dosing frequency compared with those given BID. To our knowledge, this unique analysis represents the largest single evaluation of adherence to medications commonly used by patients with CVD, while also providing data on specific classes of CVD medications. Most previous studies have either focused on a single disease state or have reported on multiple medical conditions that are typically unrelated (eg, CVDrelated conditions and osteoporosis).

CLINICAL

In this database analysis, greater adherence was observed for once-daily dosing compared with twice-daily dosing with chronic-use prescription medications used by patients with cardiovascular disease.

Adherence and Dosing Frequency of Common Medications for Cardiovascular Patients

 To estimate cost savings associated with ProvenHealth Navigator (PHN), which is an advanced model of patient-centered medical homes (PCMHs) developed by Geisinger Health System, and determine whether those savings increase over time.

A retrospective claims data analysis of 43 primary care clinics that were converted into PHN sites between 2006 and 2010. The study population included Geisinger Health Plan’s Medicare Advantage plan enrollees who were 65 years or older treated in these clinics (26,303 unique members).

Two patient-level multivariate regression models (with and without interaction effects between prescription drug coverage and PHN exposure) with member fixed effects were used to estimate the effect of members’ exposure to PHN on per-member per-month total cost, controlling for member risk, seasonality, yearly trend, and a set of baseline clinic characteristics.

 In both models, a longer period of PHN exposure was signifi cantly associated with a lower total cost. The total cumulative cost savings over the study period was 7.1% (95% confi dence interval [CI] 2.6-11.6) using the model with the prescription drug coverage interaction effects and 4.3% (95% CI 0.4-8.3) using the model without the interaction effects. Corresponding return on investment was 1.7 (95% CI 0.3-3.0) and 1.0 (95% Cl —0.1 to 2.0), respectively.

 Our fi nding suggests that PCMHs can lead to significant and sustainable cost savings over time.

 Geisinger Health Systems has implemented its own version of patient-centered medical homes called ProvenHealth Navigator (PHN) since 2006. Our results demonstrate that the longer the member has been exposed to PHN sites, the lower the cost of care. The return on investment (ROI) has not yet signifi cantly exceeded the break-even point. However, because the magnitude of savings from PHN depends on the length of members’ exposure to PHN, it remains to be seen whether the cumulative ROI can eventually exceed the break-even point.

 Cost savings are achievable by redesigning primary care, but it takes time to reap the benefits.

At this time of national focus on the need to simultaneously improve quality and reduce cost, novel solutions are needed. The fragmented healthcare delivery system further adds to the challenge.

 The key stakeholders—physicians, patients, health insurers, and hospital administrators, as well as the federal government&mdash;now recognize the need for healthcare redesign to increase the quality of care while containing the cost of care.

In particular, reorienting the current healthcare delivery system to one in which greater emphasis is given to more effective primary care is likely to be an important solution to the problem.

Recently, patient-centered medical homes (PCMHs) have gained attention as a way to re-engineer the care-delivery process. In the broadest sense of the term, PCMHs refer to “provision of comprehensive primary care services that facilitates communication and shared decision-making between the patient, his/her primary care providers, other providers, and the patient’s family.”

 There is early but growing evidence that PCMHs have the potential to improve care for individuals, improve health of populations, and slow the growth in costs of healthcare by reviving primary care and bringing the patient and family to the center of the care-delivery system

The ultimate goal of PCMHs is to improve patient outcomes via redesigned primary care. The preventive aspect of primary care enhances the value proposition of PCMHs as a long-term intervention. Such an emphasis on primary care, therefore, implies that PCMHs may have the desirable effect of reducing cost of care. The early experiences of the PCMH demonstrations strongly suggest this.

 In this study, we explicitly explore this possibility by evaluating the effect of ProvenHealth Navigator (PHN), an advanced model of PCMHs developed and implemented by Geisinger Health System (Geisinger), since 2006 on cost of care.

PHN consists of the following 5 core program components: (1) patientcentered primary care, (2) integrated population management, (3) medical “neighborhood”

 (ie, alignment of key community partners, such as home health agencies, skilled nursing facilities, outpatient/ancillary services, hospital facilities, and community pharmacies), (4) comprehensive quality improvement, and (5) value-based reimbursement redesign that includes a quality, outcome-based pay-for-performance program. More detailed descriptions of how the PHN intervention was designed and implemented are published elsewhere.

PHN is designed to move resources further “upstream” in the primary care setting to reduce “downstream” costs from the highest acuity settings. In this system, primary care is considered upstream in the sense that it serves as the starting point in the chain of care delivery. In general, care becomes more expensive as patients move downstream to specialty and inpatient care. Therefore, successful upstream efforts are expected to result in reductions of inpatient care—related costs due to fewer uncontrolled exacerbations of chronic diseases and more effective care transitions that prevent hospital readmissions and unnecessary duplication of services.

A PHN site refers to one of the primary care clinics that has been designated as such by Geisinger and has undergone significant changes in its management and operations in accordance with the PHN practice redesign. PHN was implemented in several phases over the 5-year period from late 2006 through 2010. At its core, the PHN intervention can potentially reduce cost over time while improving quality via:

• Prevention: redesigned primary care using automation to leverage resources and enhance reliability for early detection and interventions prior to exacerbations and complications

• Chronic disease optimization: redesigned primary care using a high-touch, high-technology approach to manage emerging exacerbations and to minimize complications

• Comprehensive care management: health-plan— trained, practice-embedded “concentrated care–RNs” who serve as nurse case managers focused on proactive identification and management of the more medically complex patients

This analysis focuses on individual members’ exposure to PHN by measuring how long a member has been exposed to a PHN site and examining whether a prolonged exposure to PHN is associated with a lower cost of care over time. It is expected that the longer the patient has been exposed to a PHN site, the greater the cost savings.

, a number of primary care clinics in the Geisinger Health Plan (GHP) provider network were selected and converted to PHN sites during each phase of the intervention. As a result, primary care clinics became PHN sites at different times. We exploit this variation in time of PHN conversion to estimate the effect of PHN exposure on total cost. The fi gure also shows the trends in average permember per-month (PMPM) total cost among clinics in each PHN implementation phase. For baseline clinic characteristics in each phase, see 

Our data originated from GHP’s claims database covering the period between January 1, 2006, and December 31, 2010. The unit of observation was member-month (ie, there was a unique record for each member for each month during which the patient was a GHP member). If the member did not have any claims for that month, claim amounts were recorded as $0. Our sample was restricted to GHP’s Medicare Advantage plan members who were at least 65 years of age during the study period (because PHN had focused primarily on the Medicare population during this time) and enrolled in 43 primary care clinics that eventually became PHN sites by the end of 2010 (37 with physicians employed by Geisinger and 6 non- Geisinger primary care practices). About 30% of all patients treated by practices with Geisinger-employed physicians are covered by GHP.

Two measures of PHN exposure were developed: First, a member-level PHN exposure measure was calculated as the number of months a member had been enrolled in a PHN clinic as of a given time. This exposure measure was then broken into 5 categories (0, 1-6 months, 7-12 months, 13-24 months, and >24 months) in order to capture any non-linear relationship between the cost of care and PHN exposure. Second, a clinic-level PHN exposure measure was also calculated as the number of months in which the clinic had been a PHN site as of a given period since its PHN conversion on the phase start date as shown in Figure 1.

The clinic-level PHN exposure variable is intended to capture the degree to which there might have been learning at clinics over time in accordance with PHN’s rapid-cycle innovation principle.

This is in recognition of PHN’s being a dynamic program that has undergone continuous modifi cations over time while maintaining its core components. As such, it is diffi cult to determine exactly at what point the PHN sites transitioned from practice transformation and redesign to fully functioning PHN sites. Nevertheless, we hypothesize that the longer a clinic has been designated a PHN site, the greater its impact on cost of care.

In addition to the member-level and clinic-level PHN exposure variables as described above, the covariates included age and hierarchical condition categories (HCC) risk score to capture each member’s case mix along with year-indicator variables to capture the temporal trend in total cost over time as well as month-indicator variables to capture seasonality. Other covariates included a set of baseline clinic characteristics as of 2005 (ie, percentage of male patients and average patient age in 2005, along with average total cost, HCC score, and number of inpatient admissions and readmissions in 2005). These variables were intended to capture any underlying differences among different clinics which may be correlated with each member’s total cost during the study period.

The main outcome measure was the total cost of care, defined as the total allowed amount (plan payment to providers plus copayments) for a given member in a given month (per-member-per-month [PMPM]) which included payments for inpatient and outpatient facility and professional services, as well as prescription drugs. Less than 2% of the total member- month observations in our sample had 0-value total cost, because capitation payments for certain services, such as behavioral health, were refl ected in the total cost calculations.

Because there are multiple observations for each member, a member fi xed-effects model was used to account for these repeated measures as well as to remove any unobserved and time-invariant member characteristics that could have biased our estimates. In this model, we exploit the within-member variation in the total cost and the PHN exposure variable over time. Thus, each member acts as his or her comparison. In other words, the question that our model seeks to answer may be stated as the following: relative to his or her own cost at 0 PHN exposure, what is each member’s expected total cost at 1 to 6 months, 7 to 12 months, 13 to 24 months, and greater than 24 months, of exposure?

Moreover, because members’ total costs vary depending on whether they have prescription drug coverage or not, a binary indicator variable for prescription drug coverage status during each month was also included as a covariate. Because of the fixed effects model used in this analysis, our estimate of the drug coverage effect relied on variation in the prescription drug coverage variable due to the change in each member’s drug coverage status over time. About 32% of the members picked up the drug coverage and dropped it at some point, or vice versa.

We hypothesized that the effect of PHN on total cost may differ depending on whether a member has prescription drug coverage in each period (ie, an interaction effect between prescription coverage and PHN exposure). For instance, a case manager might fi nd out whether his or her members have drug coverage or not, and if they have coverage, the case manager might encourage them to become more adherent to their drug therapy and therefore incur greater total cost. As such, we estimated 2 models: one with and another without the interaction effects. These 2 models might yield different estimates of total cost savings attributable to PHN, because in the model with interaction effects, the estimates depend both on the level of each member’s PHN exposure and the proportion of members who have drug coverage in each period. In the model without the interaction effects, the estimates depend only on the PHN exposure level.

Using the parameter estimates obtained from the fi xedeffects model, we calculated the expected total cost if none of the members had any PHN exposure (ie, member as well as clinic-level PHN exposure of 0 months) and compared them against the total costs given the actual levels of PHN exposure for each member in each given period (ie, the “observed” cost). This allowed us to estimate the effect of the PHN exposure on cost in terms of percent savings (see eAppendices A and B for the full regression model parameter estimates). The mean percent savings and the standard errors used to construct the 95% confi dence intervals (CIs) were obtained via bootstrap method with 200 replications.

As a sensitivity check, we also obtained a number of alternative estimates using different sample defi nitions. In particular, we obtained estimates with a restricted sample consisting of only those who did not switch their primary care clinics during the study period (about 80% of the full sample size). This allowed us to control for any clustering effects due to similarities among members enrolled in same clinics

 by including member-clinic fi xed effects in our regression model. The resulting estimates (available upon request) were similar to the reported estimates shown below.

, the fi nal sample consisted of 26,303 members over the 5-year period, corresponding to 1,053,445 member-month observations in 43 primary care clinics that were designated as PHN sites at different points in time between 2006 and 2010. The median age of the members was 76 years, and 42% were male. Slightly less than half of the member-month observations fell in the 0 member-level PHN exposure category; about 12% fell in the highest category of greater than 24 months of PHN exposure. Table 1 also indicates that greater PHN exposure is associated with higher total cost and age, as well as greater proportion of membermonths with prescription coverage. This suggests that the effect of PHN exposure on total cost is confounded by member age and presence of prescription drug coverage benefit.

On average, GHP members in our sample maintained their membership for about 40 months out of the maximum possible 60 months, suggesting a stable enrollment pattern. About 20% of the members in our sample switched from a non-PHN site to a PHN site during our study period, while only about 1% switched from a PHN site to a non-PHN site.

 shows the regression coeffi cients and their corresponding 95% CIs for the key covariates in the 2 models (with and without the interaction effects between drug coverage and PHN exposure). In both models, the coeffi cients on the PHN exposure variables are consistently negative and get increasingly larger as the length of exposure increases, suggesting that longer PHN exposure is consistently associated with lower total cost. However, in contrast, the coeffi cient estimates on the interaction terms between drug coverage and PHN exposure are consistently positive, suggesting that there is a significant interaction between them.

shows the estimated percent savings and the corresponding bootstrapped 95% CIs as obtained from the regression model parameter estimates shown in Table 2. Table 3 clearly supports the hypothesis that a longer exposure to PHN is associated with lower total cost and therefore greater savings, regardless of whether the interaction effects were included in the model. The largest and statistically most signifi cant percent saving was achieved in the highest category of PHN exposure (>24 months) in both models.

ProvenHealth Navigator, Geisinger's version of advanced patient-centered medical homes, is associated with significant savings in total cost of care over time.

Reducing Long-Term Cost by Transforming Primary Care: Evidence From Geisinger's Medical Home Model

To identify the optimal target of a future intervention to improve physician decision making in trauma triage.

Comparison of incremental costeffectiveness ratios (ICERs) of current practice with hypothetical interventions targeting either physicians’ decisional thresholds (attitudes toward transferring patients to trauma centers) or perceptual sensitivity (ability to identify patients who meet transfer guidelines).

Taking the societal perspective, we constructed a Markov decision model, drawing estimates of triage patterns, mortality, utilities, and costs from the literature. We assumed that an intervention to change the decisional threshold would reduce undertriage but also increase overtriage more than an intervention to change perceptual sensitivity. We performed a series of 1-way sensitivity analyses and studied the most influential variables in a Monte Carlo simulation.

 The ICER of an intervention to change perceptual sensitivity was $62,799 per qualityadjusted life-year (QALY) gained compared with current practice. The ICER of an intervention to change the decisional threshold was $104,975/ QALY gained compared with an intervention to change perceptual sensitivity. These findings were most sensitive to the relative cost of hospitalizing patients with moderate to severe injuries and their relative risk of dying at non—trauma centers. In probabilistic sensitivity analyses, at a willingness-to-pay threshold of $100,000/QALY gained, there was a 62% likelihood that an intervention to change perceptual sensitivity was the most cost-effective alternative.

 Even a minor investment in changing decision making in trauma triage could greatly improve quality of care. The optimal intervention depends on the characteristics of the individual trauma systems.

Investing in an intervention to improve the decision making of individual physicians in trauma triage would yield significant value even if the intervention were expensive.

		 States such as Pennsylvania could invest up to $20 million to ensure triage according to the guidelines, and the combination of the intervention and care at a trauma center would still cost less than $100,000 per quality-adjusted life-year gained.

		 Quality improvement interventions should be tailored to the specific characteristics of each trauma system instead of using a one-size-fits-all strategy.

More than 30 million people receive trauma care every year in the United States,

 consuming approximately $400 billion annually.

 To optimize the allocation of resources, the American College of Surgeons advocates the regionalization of patients. 

 Regionalization requires that physicians at non—trauma centers transfer patients with moderate to severe injuries to specialty trauma centers while admitting those with minor injuries.

 Despite an unprecedented effort at the federal, state, and local levels to increase adherence to clinical practice guidelines supporting regionalization, only one-third of patients with moderate to severe injuries taken initially to non—trauma centers are actually transferred to trauma centers.

 Additional quality improvement interventions are needed.

Physician adherence to clinical practice triage guidelines can be thought of as the product of both decisional thresholds and perceptual sensitivity.

Decisional thresholds reflect physicians’ preferences for transferring or keeping patients based on norms, attitudes, and incentives. Perceptual sensitivity reflects physicians’ ability to discriminate between patients who do and do not meet clinical practice guidelines for transfer based on their knowledge of the guidelines and their intuitive judgments (heuristics).

 Decisional thresholds and perceptual sensitivity are distinct cognitive processes. Therefore, modification requires targeted interventions, which have different costs and benefits. Current efforts have included both top-down approaches aimed at shifting physicians’ decisional thresholds through a mixture of regulatory and organizational incentives

 and a bottom-up approach aimed at increasing physician perceptual sensitivity through Advanced Trauma Life Support (ATLS) training and certification.

To set priorities for quality improvement in trauma triage, we need to know whether it is more cost-effective to invest additional resources in changing physicians’ decisional thresholds or their perceptual sensitivity. The objective of our study was to compare the cost-effectiveness of current practice with that of 2 distinct hypothetical interventions to improve physician adherence to clinical practice guidelines: one that shifted decisional thresholds versus one that increased perceptual sensitivity.

Signal detection theory, a well-established behavioral science method, identifies 2 separate aspects of every dichotomous decision: decisional threshold and perceptual sensitivity (

). Interventions designed to impact these distinct cognitive processes require different designs and have different anticipated costs and benefits. An intervention to change decisional thresholds, such as a pay-for-performance initiative, would increase sensitivity. Although relatively inexpensive and easy to implement, shifting physicians’ decisional thresholds to make them more willing to transfer patients with moderate to severe injuries would reduce system efficiency, by increasing the number of patients with minor injuries transferred to trauma centers. On the other hand, an intervention to change perceptual sensitivity, such as a program to modify physician heuristics, would increase specificity. This kind of intervention would be relatively more expensive and difficult to implement, but would increase system efficiency.

) to compare the clinical and economic outcomes of a 40-year-old trauma patient taken initially to a non—trauma center under current conditions of national compliance with trauma triage guidelines compared with compliance resulting from 2 alternative hypothetical interventions—one targeting physicians’ decisional threshold and one targeting their perceptual sensitivity. As recommended by the Panel on Cost-Effectiveness in Health and Medicine, we took the societal perspective when we considered direct medical and nonmedical costs of obtaining care, and used a 1-month cycle length and a lifetime time horizon in the analysis. We discounted future costs and benefits at an annual rate of 3%.

 We constructed our model using TreeAge Pro 2009 software (TreeAge, Williamstown, Massachusetts).

All clinical parameters, utilities, and costs used in the base case and sensitivity analyses are summarized in 

 and described below. We drew probabilities of minor and moderate to severe trauma, of undertriage and overtriage, and of mortality conditional upon triage from the literature.

 In a retrospective analysis of Pennsylvania discharge data from 2001 to 2005, we found that 25% of patients presenting to non—trauma centers after trauma had a moderate to severe injury, defined as either an Injury Severity Score greater than 15 or as an injury categorized by the American College of Surgeons Committee on Trauma as “life-threatening” or “critical” (eg, open long bone fracture). Of these patients, 30% were transferred to a trauma center. Of patients with minor injuries, defined as the absence of a moderate to severe injury, 15% were transferred to a trauma center.

We drew age-adjusted probabilities of all-cause mortality from national health statistics.

To set priorities for quality improvement in trauma triage, we compared the cost-effectiveness of current practice with interventions to improve physician adherence to clinical guidelines.

March 2012

IT-Enabled Systems Engineering Approach to Monitoring and Reducing ADEs

What Determines Successful Implementation of Inpatient Information Technology Systems?

Adherence and Dosing Frequency of Common Medications for Cardiovascular Patients

Reducing Long-Term Cost by Transforming Primary Care: Evidence From Geisinger's Medical Home Model

Optimal Approach to Improving Trauma Triage Decisions: A Cost-Effectiveness Analysis

Healthcare Continuity From Hospital to Territory in Lombardy: TELEMACO Project

Off-Label Use of Antipsychotic Medications in Medicaid

Clinical Pathways for Oncology: More Rigor Needed When Evaluating Models

The Promise and Peril of Healthcare Forecasting

Impact of an Online Prescription Management Account on Medication Adherence