Health Care Cost Reductions with Machine Learning-Directed Evaluations during Radiation Therapy - An Economic Analysis of a Randomized Controlled Study.

BACKGROUND: Machine learning (ML) may cost-effectively direct health care by identifying patients most likely to benefit from preventative interventions to avoid negative and expensive outcomes. System for High-Intensity Evaluation During Radiation Therapy (SHIELD-RT; NCT04277650) was a single-institution, randomized controlled study in which electronic health record-based ML accurately identified patients at high risk for acute care (emergency visit or hospitalization) during radiotherapy (RT) and targeted them for supplemental clinical evaluations. This ML-directed intervention resulted in decreased acute care utilization. Given the limited prospective data showing the ability of ML to direct interventions cost-efficiently, an economic analysis was performed. METHODS: A post hoc economic analysis was conducted of SHIELD-RT that included RT courses from January 7, 2019, to June 30, 2019. ML-identified high-risk courses (≥10% risk of acute care during RT) were randomized to receive standard of care weekly clinical evaluations with ad hoc supplemental evaluations per clinician discretion versus mandatory twice-weekly evaluations. The primary outcome was difference in mean total medical costs during and 15 days after RT. Acute care costs were obtained via institutional cost accounting. Physician and intervention costs were estimated via Medicare and Medicaid data. Negative binomial regression was used to estimate cost outcomes after adjustment for patient and disease factors. RESULTS: A total of 311 high-risk RT courses among 305 patients were randomized to the standard (n=157) or the intervention (n=154) group. Unadjusted mean intervention group supplemental visit costs were $155 per course (95% confidence interval, $142 to $168). The intervention group had fewer acute care visits per course (standard, 0.47; intervention, 0.31; P=0.04). Total mean adjusted costs were $3110 per course for the standard group and $1494 for the intervention group (difference in means, $1616 [95% confidence interval, $1450 to $1783]; P=0.03). CONCLUSIONS: In this economic analysis of a randomized controlled, health care ML study, mandatory supplemental evaluations for ML-identified high-risk patients were associated with both reduced total medical costs and improved clinical outcomes. Further study is needed to determine whether economic results are generalizable. (Funded in part by The Duke Endowment, The Conquer Cancer Foundation, the Duke Department of Radiation Oncology, and the National Cancer Institute of the National Institutes of Health [R01CA277782]; ClinicalTrials.gov number, NCT04277650.).

Evaluating registry-based trial economics: Results from the STRESS clinical trial.

Background: Registry-based trials have the potential to reduce randomized clinical trial (RCT) costs. However, observed cost differences also may be achieved through pragmatic trial designs. A systematic comparison of trial costs across different designs has not been previously performed. Methods: We conducted a study to compare the current Steroids to Reduce Systemic inflammation after infant heart surgery (STRESS) registry-based RCT vs. two established designs: pragmatic RCT and explanatory RCT. The primary outcome was total RCT design costs. Secondary outcomes included: RCT duration and personnel hours. Costs were estimated using the Duke Clinical Research Institute's pricing model. Results: The Registry-Based RCT estimated duration was 31.9 weeks greater than the other designs (259.5 vs. 227.6 weeks). This delay was caused by the Registry-Based design's periodic data harvesting that delayed site closing and statistical reporting. Total personnel hours were greatest for the Explanatory design followed by the Pragmatic design and the Registry-Based design (52,488 vs 29,763 vs. 24,480 h, respectively). Total costs were greatest for the Explanatory design followed by the Pragmatic design and the Registry-Based design ($10,140,263 vs. $4,164,863 vs. $3,268,504, respectively). Thus, Registry-Based total costs were 32 % of the Explanatory and 78 % of the Pragmatic design. Conclusion: Total costs for the STRESS RCT with a registry-based design were less than those for a pragmatic design and much less than an explanatory design. Cost savings reflect design elements and leveraging of registry resources to improve cost efficiency, but delays to trial completion should be considered.

Economic analysis of a single institutional review board data exchange standard in multisite clinical studies.

BACKGROUND: Single Institutional Review Boards (sIRB) are not achieving the benefits envisioned by the National Institutes of Health. The recently published Health Level Seven (HL7®) Fast Healthcare Interoperability Resources (FHIR®) data exchange standard seeks to improve sIRB operational efficiency. METHODS AND RESULTS: We conducted a study to determine whether the use of this standard would be economically attractive for sIRB workflows collectively and for Reviewing and Relying institutions. We examined four sIRB-associated workflows at a single institution: (1) Initial Study Protocol Application, (2) Site Addition for an Approved sIRB study, (3) Continuing Review, and (4) Medical and Non-Medical Event Reporting. Task-level information identified personnel roles and their associated hour requirements for completion. Tasks that would be eliminated by the data exchange standard were identified. Personnel costs were estimated using annual salaries by role. No tasks would be eliminated in the Initial Study Protocol Application or Medical and Non-Medical Event Reporting workflows through use of the proposed data exchange standard. Site Addition workflow hours would be reduced by 2.50 h per site (from 15.50 to 13.00 h) and Continuing Review hours would be reduced by 9.00 h per site per study year (from 36.50 to 27.50 h). Associated costs savings were $251 for the Site Addition workflow (from $1609 to $1358) and $1033 for the Continuing Review workflow (from $4110 to $3076). CONCLUSION: Use of the proposed HL7 FHIR® data exchange standard would be economically attractive for sIRB workflows collectively and for each entity participating in the new workflows.

Ascertaining Nonfatal Endpoints in Clinical Trials: Central Adjudication Versus Patient Insurance Claims.

BACKGROUND: The 21st Century Cures Act allows the US Food and Drug Administration (FDA) to utilize real-world data (RWD) to create real-world evidence (RWE) for new indications or post approval study requirements. We compared central adjudication with two insurance claims data sources to understand how endpoint accuracy differences impact RWE results. METHODS: We developed a decision analytic model to compare differences in efficacy (all-cause death, stroke and myocardial infarction) and safety (bleeding requiring transfusion) results for a simulated acute coronary syndrome antiplatelet therapy clinical trial. Endpoint accuracy metrics were derived from previous studies that compared centrally-adjudicated and insurance claims-based clinical trial endpoints. RESULTS: Efficacy endpoint results per 100 patients were similar for the central adjudication model (intervention event rate, 11.3; control, 13.7; difference, 2.4) and the prospective claims data collection model (intervention event rate, 11.2; control 13.6; difference, 2.3). However, the retrospective claims linking model's efficacy results were larger (intervention event rate, 14.6; control, 18.0; difference, 3.4). True positive event rate results (intervention, control and difference) for both insurance claims-based models were less than the central adjudication model due to false negative events. Differences in false positive event rates were responsible for differences in efficacy results for the two insurance claims-based models. CONCLUSION: Efficacy endpoint results differed by data source. Investigators need guidance to determine which data sources produce regulatory-grade RWE.

eSource-Enabled vs. Traditional Clinical Trial Data Collection Methods: A Site-Level Economic Analysis.

Directly extracting data from site electronic health records for updating clinical trial databases (eSource) can reduce site data collection times and errors. We conducted a study to determine clinical trial characteristics that make eSource vs. traditional data collection methods more and less economically attractive. The number of patients a site enrolls, the number of study data elements, study coordinator data collection times, and the percent of study data elements that can be extracted via eSource software all impact eSource economic attractiveness. However, these factors may not impact all clinical trial designs in the same way.

Index and follow-up costs of endovascular abdominal aortic aneurysm repair from the Endurant Stent Graft System Post Approval Study (ENGAGE PAS).

OBJECTIVE: Trials for endovascular aneurysm repair (EVAR) report lower perioperative mortality and morbidity, but also higher costs compared with open repair. However, few studies have examined the subsequent cost of follow-up evaluations and interventions. Therefore, we present the index and 5-year follow-up costs of EVAR from the Endurant Stent Graft System Post Approval Study. METHODS: From August 2011 to June 2012, 178 patients were enrolled in the Endurant Stent Graft System Post Approval Study de novo cohort and treated with the Medtronic Endurant stent graft system (Medtronic Vascular, Santa Rosa, Calif), of whom 171 (96%) consented for inclusion in the economic analysis and 177 participated in the quality-of-life (QOL) assessment over a 5-year follow-up period. Cost data for the index and follow-up hospitalizations were tabulated directly from hospital bills and categorized by Uniform Billing codes. Surgeon costs were calculated by Current Procedural Terminology codes for each intervention. Current Procedural Terminology codes were also used to calculate imaging and clinic follow-up reimbursement as surrogate to cost based on year-specific Medicare payment rates. Additionally, we compared aneurysm-related versus nonaneurysm-related subsequent hospitalization costs and report EuroQol 5D QOL dimensions. RESULTS: The mean hospital cost per person for the index EVAR was $45,304 (interquartile range [IQR], $25,932-$44,784). The largest contributor to the overall cost was operating room supplies, which accounted for 50% of the total cost at a mean of $22,849 per person. One hundred patients had 233 additional post index admission inpatient admissions; however, only 32 readmissions (14%) were aneurysm related, with a median cost of $13,119 (IQR, $4570-$24,153) compared with a nonaneurysm-related median cost of $6609 (IQR, $1244-$26,466). Additionally, 32 patients were admitted a total of 37 times for additional procedures after index admission, of which 14 (38%) were aneurysm-related. The median cost of hospitalization for aneurysm-related subsequent intervention was $22,023 (IQR, $13,177-$47,752), compared with a median nonaneurysm-related subsequent intervention cost of $19,007 (IQR, $8708-$33,301). After the initial 30-day visit, outpatient follow-up imaging reimbursement averaged $550 per person per year ($475 for computed tomography scans, $75 for the abdomen), whereas annual office visits averaged $107 per person per year, for a total follow-up reimbursement of $657 per person per year. There were no significant differences in the five EuroQol 5D QOL dimensions at each follow-up compared with baseline. CONCLUSIONS: Costs associated with index EVAR are driven primarily by cost of operating room supplies, including graft components. Subsequent admissions are largely not aneurysm related; however, cost of aneurysm-related hospitalizations is higher than for nonaneurysm admissions. These data will serve as a baseline for comparison with open repair and other devices.

eSource for Standardized Health Information Exchange in Clinical Research: A Systematic Review.

The availability of research and outcomes data is the primary limitation to evidence-based practice. Today, only a fraction of clinical decisions are based upon evidence derived from randomized control trials (RCTs), the gold-standard of knowledge discovery. At the same time, clinical trial complexity has steadily increased as has the effort required at clinical investigational sites. Direct use of electronic health record (EHR) data for clinical trials has the potential to address some of these needs, improving data quality and reducing cost.

Assessing the Financial Value of Patient Engagement: A Quantitative Approach from CTTI's Patient Groups and Clinical Trials Project.

BACKGROUND: While patient groups, regulators, and sponsors are increasingly considering engaging with patients in the design and conduct of clinical development programs, sponsors are often reluctant to go beyond pilot programs because of uncertainty in the return on investment. We developed an approach to estimate the financial value of patient engagement. METHODS: Expected net present value (ENPV) is a common technique that integrates the key business drivers of cost, time, revenue, and risk into a summary metric for project strategy and portfolio decisions. We assessed the impact of patient engagement on ENPV for a typical oncology development program entering phase 2 or phase 3. RESULTS: For a pre-phase 2 project, the cumulative impact of a patient engagement activity that avoids one protocol amendment and improves enrollment, adherence, and retention is an increase in net present value (NPV) of $62MM ($65MM for pre-phase 3) and an increase in ENPV of $35MM ($75MM for pre-phase 3). Compared with an investment of $100,000 in patient engagement, the NPV and ENPV increases can exceed 500-fold the investment. This ENPV increase is the equivalent of accelerating a pre-phase 2 product launch by 2½ years (1½ years for pre-phase 3). CONCLUSIONS: Risk-adjusted financial models can assess the impact of patient engagement. A combination of empirical data and subjective parameter estimates shows that engagement activities with the potential to avoid protocol amendments and/or improve enrollment, adherence, and retention may add considerable financial value. This approach can help sponsors assess patient engagement investment decisions.

Using Medical Informatics to Improve Clinical Trial Operations.

The continued escalation of clinical trial costs is becoming a public health concern. During the past decade, medical research funding peaked and there is growing concern that there may be insufficient resources to test many promising medical products. Recent changes in the regulatory environment create opportunities for the use of medical informatics to improve clinical trial operations and reduce costs. We report on a Medical Informatics Europe 2016 workshop conducted during the Health - Exploring Complexity (HEC) 2016 conference. We review presentation given on Secondary Data Use, eSource, and Data Quality in Clinical Trials and report on the workshop's discussions.

Randomized Trial of Population-Based Clinical Decision Support to Facilitate Care Transitions.

Medicaid beneficiaries in 6 North Carolina counties were randomly assigned to 1 of 3 clinical decision support (CDS) care transition strategies: (1) usual care (Control), (2) CDS messaging to patients and their medical homes (Reports), or (3) CDS messaging to patients, their medical homes, and their care managers (Reports+). We included 7146 Medicaid patients and evaluated transitions from specialist visit, ER and hospital encounters back to the patient's medical home. Patients enrolled in Medicare and Medicaid were not eligible. The number of care manager contacts was greater for patients in the Reports+ Group than in the Control Group. However, there were no treatment-related differences in emergency department (ED) encounter rates, or in the secondary outcomes of outpatient and hospital encounter rates and medical costs. Study monitors found study intervention documentation in approximately 60% of patient charts. These results highlight the importance of effectively integrating information interventions into healthcare delivery workflow systems.

The Effects of Two Workplace Weight Management Programs and Weight Loss on Health Care Utilization and Costs.

OBJECTIVES: The aim of this study was to compare the impact of two worksite weight management (WM [education] and WM+ [education plus counseling]) programs, on health care utilization and costs. Secondarily, compare the intervention groups to an observational control group of obese workers. Finally, evaluate the impact of actual weight loss on these outcomes. METHODS: Estimate the change in the WM and WM+ intervention groups. Using propensity score adjustment compare the two intervention groups with the observational control group; and compare those who lost weight with those who did not. RESULTS: No significant differences between the two intervention groups, or between these intervention groups and the observational control group. Those who lost weight reduced their overall health care costs. CONCLUSIONS: To achieve weight loss and associated morbidity reductions, more extensive and intensive interventions, with more attention to motivation and compliance, are required.

Economic Outcomes of Extracorporeal Membrane Oxygenation With and Without Ambulation as a Bridge to Lung Transplantation.

BACKGROUND: An increasing number of centers are using active rehabilitation and ambulation for critically ill patients on extracorporeal membrane oxygenation (ECMO) as a bridge to lung transplantation. This investigation assessed the economic impact at a single center of ambulatory versus non-ambulatory ECMO strategies as a bridge to lung transplantation. METHODS: We conducted a single-center retrospective cohort analysis of all subjects supported with ECMO as a bridge to lung transplantation (N = 9) from 2007 to 2012. Subjects who were rehabilitated while supported with ECMO before lung transplantation were compared with those who were not rehabilitated during ECMO. Hospital cost data for the month before transplantation through 12 months after the initial post-transplant hospital discharge were compared. RESULTS: The median cost (interquartile range [IQR]) in the 30 d before transplant for the ambulatory cohort was $88,137 (IQR $38,589-$122,111) compared with $52,124 (IQR $23,824-$69,929) for the non-ambulatory cohort (P = .08). The median post-transplant ICU cost for the ambulatory cohort was $38,468 (IQR $23,611-$64,126) compared with $143,407 (IQR $112,199-$168,993) for the non-ambulatory cohort (P = .01). The median total hospital cost for subjects supported with ambulatory ECMO was $213,086 (IQR $166,767-$264,536) compared with $273,291 (IQR $237,299-$374,175) for non-ambulatory ECMO subjects (P = .05). The median total cost for the ambulatory cohort was $268,194 (IQR $219,972-$517,320) compared with $300,307 (IQR $274,262-$394,913) for the non-ambulatory cohort (P = .14). CONCLUSIONS: Subjects supported with ambulatory ECMO had a 22% ($60,204) reduction in total hospital cost, 73% ($104,939) reduction in post-transplant ICU cost, and 11% ($32,133) reduction in total cost compared with non-ambulatory ECMO subjects. This analysis demonstrates the potential economic benefit of rehabilitation and ambulation during ECMO compared with a traditional strategy.

Impact of pediatric exclusivity on drug labeling and demonstrations of efficacy.

BACKGROUND: Besides vaccines and otitis media medicines, most products prescribed for children have not been studied in the pediatric population. To remedy this, Congress enacted legislation in 1997, known as pediatric exclusivity (PE), which provides 6 months of additional market protection to drug sponsors in exchange for studying their products in children. METHODS: We reviewed requests for pediatric studies and subsequent labeling for drugs granted PE from 1998 through 2012. Regression analysis estimates the probability of demonstrating efficacy in PE trials. Variables include therapeutic group, year of exclusivity, product sales, initiation process, and small disease population. RESULTS: From 1998 through 2012, the US Food and Drug Administration issued 401 pediatric study requests. For 189 drugs, studies were completed and granted exclusivity. A total of 173 drugs (92%) received new pediatric labeling, with 108 (57%) receiving a new or expanded pediatric indication. Three drugs had non-efficacy trials. Efficacy was not established for 78 drugs. Oncology, cardiovascular, and endocrine drugs were less likely to demonstrate efficacy (P < .01) compared with gastrointestinal and pain/anesthesia drugs. Drugs studied later in the program were less likely to demonstrate efficacy (P < .05). Sales, initiation process, and small disease population were not significant predictors. CONCLUSIONS: Most drugs (173; 92%) granted exclusivity added pediatric information to their labeling as a result of PE, with 108 (57%) receiving a new or expanded pediatric indication. Therapeutic area and year of exclusivity influenced the likelihood of obtaining a pediatric indication. Positive and negative outcomes continue to inform the construct of future pediatric trials.

Reduced length of hospital stay in colorectal surgery after implementation of an enhanced recovery protocol.

BACKGROUND: Enhanced recovery after surgery (ERAS) is a multimodal approach to perioperative care that combines a range of interventions to enable early mobilization and feeding after surgery. We investigated the feasibility, clinical effectiveness, and cost savings of an ERAS program at a major U. S. teaching hospital. METHODS: Data were collected from consecutive patients undergoing open or laparoscopic colorectal surgery during 2 time periods, before and after implementation of an ERAS protocol. Data collected included patient demographics, operative, and perioperative surgical and anesthesia data, need for analgesics, complications, inpatient medical costs, and 30-day readmission rates. RESULTS: There were 99 patients in the traditional care group, and 142 in the ERAS group. The median length of stay (LOS) was 5 days in the ERAS group compared with 7 days in the traditional group (P < 0.001). The reduction in LOS was significant for both open procedures (median 6 vs 7 days, P = 0.01), and laparoscopic procedures (4 vs 6 days, P < 0.0001). ERAS patients had fewer urinary tract infections (13% vs 24%, P = 0.03). Readmission rates were lower in ERAS patients (9.8% vs 20.2%, P = 0.02). DISCUSSION: Implementation of an enhanced recovery protocol for colorectal surgery at a tertiary medical center was associated with a significantly reduced LOS and incidence of urinary tract infection. This is consistent with that of other studies in the literature and suggests that enhanced recovery programs could be implemented successfully and should be considered in U.S. hospitals.

Is overweight and class I obesity associated with increased health claims costs?

OBJECTIVES: Evaluate the relationship between body mass index (BMI) and health claims costs over the last decade, assess the strength and nature of the relationship between BMI and costs, and identify comorbidities that may drive any increased costs. METHODS: Using 2001-2011 claims data for employees participating in annual health appraisals, annual paid claims costs were calculated. One-part negative binomial models were fit to evaluate the relationship between BMI and costs, controlling for age, gender, race/ethnicity, and calendar year period. RESULTS: The relationship between increasing BMI and increasing health claims costs is gradual and starts already at a BMI of 19. The nature of the relationship did not change notably over time. The most important obesity-related comorbidities, expressed as percent increase in cost per BMI unit, was cardiovascular disease (males 10.53, 95% CI [6.46, 14.77], females 4.27, 95% CI [1.25, 7.38), while cardiovascular agents (7.23, 95% CI [6.08, 8.39]) were the most important driver of pharmacy costs. CONCLUSION: In contrast to recent evidence relating to effects on mortality, we observed a gradual increase in health claims costs starting at the low end of the recommended BMI range.

The steps to health employee weight management randomized control trial: rationale, design and baseline characteristics.

BACKGROUND: The workplace can be an important setting for addressing obesity. An increasing number of employers offer weight management programs. PURPOSE: Present the design, rationale and baseline characteristics of the Steps to Health study (STH), a randomized trial to evaluate the effectiveness of two preexisting employee weight management programs offered at Duke University and Medical Center. METHODS: 550 obese (BMI ≥30) employee volunteers were randomized 1:1 to two programs. Baseline data, collected between January 2011 and July 2012, included height/weight, accelerometry, workplace injuries, health care utilization, and questionnaires querying socio-cognitive factors, perceptions of health climate, physical activity, and dietary intake. In secondary analyses participants in the two programs will also be compared to a non-randomized observational control group of obese employees. RESULTS: At baseline, the mean age was 45 years, 83% were female, 41% white, and 53% black. Mean BMI was 37.2. Participants consumed a mean of 2.37 servings of fruits and vegetables per day (in the past week), participated in 11.5 min of moderate-to-vigorous physical activity, and spent 620 min being sedentary. CONCLUSION: STH addresses the need for evaluation of worksite interventions to promote healthy weight. In addition to having direct positive effects on workers' health, worksite programs have the potential to increase productivity and reduce health care costs.

Economic analysis of endovascular repair versus surveillance for patients with small abdominal aortic aneurysms.

BACKGROUND: The Positive Impact of EndoVascular Options for Treating Aneurysms Early (PIVOTAL) trial enrolled individuals with small (4.0- to 5.0-cm diameter) abdominal aortic aneurysms (AAA) and reported no difference in rupture or aneurysm-related death for patients who received early endovascular repair (EVAR) vs surveillance with serial imaging studies. We evaluated resource use, medical cost, and quality of life outcomes associated with the PIVOTAL treatment strategies. METHODS: This prospective economic and quality of life study was conducted within a randomized trial, with PIVOTAL sites participating in the quality of life (n = 67) and economic (n = 63) studies. The PIVOTAL trial randomized 728 patients (366 early EVAR and 362 surveillance). We used information from 701 quality of life (351 early EVAR and 350 surveillance) and 614 economic (314 early EVAR and 300 surveillance) study participants enrolled in the PIVOTAL trial. The main outcome measures were total medical costs and the aneurysm repair rate at 48 months. RESULTS: After 6 months, the rate of aneurysm repair was 96 vs 10 per 100 patients in the early EVAR and surveillance groups, respectively (difference, 86; 95% confidence interval [CI], 82-90; P < .0001), and total medical costs were greater in the early EVAR group ($33,471 vs $5520; difference, $27,951; 95% CI, $25,156-$30,746; P < .0001). In months 7 through 48, however, the rate of aneurysm repair was 54 per 100 patients in the surveillance group, and total medical costs were higher for patients in the surveillance vs the early EVAR group ($40,592 vs $15,197; difference, $25,394; 95% CI, $15,184-$35,605; P < .0001). At 48 months' follow-up, early EVAR patients had greater cumulative use of AAA repair (97 vs 64 per 100 patients; difference, 34; 95% CI, 21-46; P < .0001), but there was no difference in total medical costs ($48,669 vs $46,112; difference, $2557; 95% CI, -$8043 to $13,156; P = .64). After discounting at 3% per annum, total medical costs for early EVAR and surveillance patients remained similar ($47,765 vs $43,532; difference, 4232; 95% CI, -$5561 to $14,025; P = .40). There were no treatment-related differences in quality of life at 24 months. CONCLUSIONS: A treatment strategy involving early repair of smaller AAA with EVAR is associated with no difference in total medical costs at 48 months vs surveillance with serial imaging studies. Longer follow-up is required to determine whether the late medical cost increases observed for surveillance will persist beyond 48 months.

A randomized trial of population-based clinical decision support to manage health and resource use for Medicaid beneficiaries.

To determine whether a clinical decision support system can favorably impact the delivery of emergency department and hospital services. Randomized clinical trial of three clinical decision support delivery modalities: email messages to care managers (email), printed reports to clinic administrators (report) and letters to patients (letter) conducted among 20,180 Medicaid beneficiaries in Durham County, North Carolina with follow-up through 9 months. Patients in the email group had fewer low-severity emergency department encounters vs. controls (8.1 vs. 10.6/100 enrollees, p < 0.001) with no increase in outpatient encounters or medical costs. Patients in the letter group had more outpatient encounters and greater outpatient and total medical costs. There were no treatment-related differences for patients in the reports group. Among patients <18 years, those in the email group had fewer low severity (7.6 vs. 10.6/100 enrollees, p < 0.001) and total emergency department encounters (18.3 vs. 23.5/100 enrollees, p < 0.001), and lower emergency department ($63 vs. $89, p = 0.002) and total medical costs ($1,736 vs. $2,207, p = 0.009). Patients who were ≥18 years in the letter group had greater outpatient medical costs. There were no intervention-related differences in patient-reported assessments of quality of life and medical care received. The effectiveness of clinical decision support messaging depended upon the delivery modality and patient age. Health IT interventions must be carefully evaluated to ensure that the resultant outcomes are aligned with expectations as interventions can have differing effects on clinical and economic outcomes.

Population-based clinical decision support: a clinical and economic evaluation.

Governments are investing in health information technologies (HIT) to improve care quality and reduce medical costs. However, evidence of these benefits is limited. We conducted a randomized trial of three clinical decision support (CDS) interventions in 20,180 patients: email to care managers (n=3329), reports to primary care administrators (n=3368), letters to patients (n=3401), and controls (10,082). At 7-month follow-up, the letters to patients group had greater use of outpatient services and higher outpatient and total medical costs; whereas, the other groups had no change in clinical events or medical costs. As our CDS interventions were associated with no change or an increase in medical costs, it appears that investments in HIT without consideration for organizational context may not be sufficient to achieve improvements in clinical and economic outcomes.

Linking the National Cardiovascular Data Registry CathPCI Registry with Medicare claims data: validation of a longitudinal cohort of elderly patients undergoing cardiac catheterization.

BACKGROUND: The National Cardiovascular Data Registry CathPCI Registry was recently linked with longitudinal Centers for Medicare & Medicaid (CMS) claims data. The degree to which this linked cohort is representative of the overall CathPCI Registry and CMS PCI populations is unknown. METHODS AND RESULTS: CathPCI Registry records were linked to CMS inpatient claims using indirect identifiers. We examined the degree to which hospitals and patients in the linked cohort are representative of the elderly (≥65 years) CathPCI Registry and CMS populations. From 2004 to 2006, 1492 hospitals filed CMS PCI claims and 663 contributed CathPCI Registry data. Of these hospitals, 643 (97%) were linked across data sources. Compared with all CMS PCI hospitals, the linked data set contained fewer governmental, northeastern, southern, and low-volume (<200 beds) sites. Among CMS beneficiaries, 993,351 PCI procedures were performed, including 398,508 (40.1%) at centers in the linked database. Of these, 341,916 (86%) were linked to CathPCI Registry records. Linked and unlinked CMS patients had similar demographic and clinical features. In the CathPCI Registry database, 477,456 elderly patients underwent PCI, with 359,077 (75%) linked to CMS claims. Linked and unlinked National Cardiovascular Data Registry patients were similar, except for less commercial or health maintenance organization insurance in the linked cohort. CONCLUSIONS: By using deterministic matching strategies, a large and representative cohort with detailed clinical data from the CathPCI Registry and longitudinal follow-up from CMS claims has been created.