Impact of Clinical Decision Support with Mandatory versus Voluntary Venous Thromboembolism Risk Assessment in Hospitalized Patients.

Background Venous thromboembolism (VTE) causes significant preventable morbidity and mortality in hospitalized patients. Assessing VTE risk is essential to initiating appropriate prophylaxis and reducing VTE outcomes. Studies show that computerized clinical decision support (CDS) can improve VTE risk assessment (RA), prophylaxis, and outcomes but few examined the effectiveness of specific design features. From 2008 to 2016, University of Michigan Health implemented CDS for VTE prevention in four stages, which alternated between voluntary and mandatory RA using the 2005 Caprini model and generated inpatient orders for risk-appropriate prophylaxis based on CHEST guidelines. This cross-sectional study evaluated the impact of mandatory versus voluntary RA on VTE prophylaxis and outcomes for adult medical and surgical patients admitted to the health system. Methods Interrupted time series analysis was conducted to evaluate the trend in smart order set-recommended VTE prophylaxis by CDS stage. Logistic regression with CDS stage as the primary independent variable was used in pairwise comparisons of VTE during hospitalization and within 90 days post-discharge for mandatory versus voluntary RA. Adjusted odd ratios (ORs) were calculated for total, in-hospital, and post-discharge VTE. Results In this study of 223,405 inpatients over 8 years, smart order set-recommended prophylaxis increased from 65 to 79%; it increased significantly when voluntary RA in Stage 1 became mandatory in Stage 2 (10.59%, p < 0.001) and decreased significantly when it returned to voluntary in Stage 3 (-11.24%, p < 0.001). The rate increased slightly when mandatory RA was reestablished in Stage 4 (0.23%, p = 0.935). Adjusted ORs for VTE were lower for mandatory RA versus adjacent stages with voluntary RA. The adjusted OR for Stage 2 versus Stage 1 was 14% lower ( p < 0.05) and versus Stage 3 was 11% lower ( p < 0.05). The adjusted OR for Stage 4 versus Stage 3 was 4% lower ( p = 0.60). These results were driven by changes in in-hospital VTE. By contrast, the incidence of post-discharge VTE increased in each successive stage. Conclusion Mandatory RA was more effective in improving smart order set-recommended prophylaxis and VTE outcomes, particularly in-hospital VTE. Post-discharge VTE increased despite high adherence to risk-appropriate prophylaxis, indicating that guidelines for extended, post-discharge prophylaxis are needed to further reduce VTE for hospitalized patients.

The relationship between diabetes mellitus and 30-day readmission rates.

BACKGROUND: It is estimated that 9.3% of the population in the United States have diabetes mellitus (DM), 28% of which are undiagnosed. The high prevalence of DM makes it a common comorbid condition in hospitalized patients. In recent years, government agencies and healthcare systems have increasingly focused on 30-day readmission rates to determine the complexity of their patient populations and to improve quality. Thirty-day readmission rates for hospitalized patients with DM are reported to be between 14.4 and 22.7%, much higher than the rate for all hospitalized patients (8.5-13.5%). The objectives of this study were to (1) determine the incidence and causes of 30-day readmission rates for patients with diabetes listed as either the primary reason for the index admission or with diabetes listed as a secondary diagnosis compared to those without DM and (2) evaluate the impact on readmission of two specialized inpatient DM services: the Hyperglycemic Intensive Insulin Program (HIIP) and Endocrine Consults (ENDO). METHODS: For this study, DM was defined as any ICD-9 discharge diagnosis (principal or secondary) of 250.xx. Readmissions were defined as any unscheduled inpatient admission, emergency department (ED) visit, or observation unit stay. We analyzed two separate sets of patient data. The first pilot study was a retrospective chart review of all patients with a principle or secondary admission diagnosis of diabetes admitted to any adult service within the University of Michigan Health System (UMHS) between October 1, 2013 and December 31, 2013. We then did further uncontrolled analysis of the patients with a principal admitting diagnosis of diabetes. The second larger retrospective study included all adults discharged from UMHS between October 1, 2013 and September 30, 2014 with principal or secondary discharge diagnosis of DM (ICD-9-CM: 250.xx). RESULTS: In the pilot study of 7763 admissions, the readmission rate was 26% for patients with DM and 22% for patients without DM. In patients with a primary diagnosis of DM on index admission, the most common cause for readmission was DM-related. In the larger study were 37,702 adult inpatient discharges between October 1, 2013 and September 30, 2014. Of these, 20.9% had DM listed as an encounter diagnosis. Rates for all encounters (inpatient, ED and Observation care) were 24.3% in patients with DM compared to 17.7% in those without DM (p < 0.001). The most common cause for readmission in patients with DM as a secondary diagnosis to the index admission was infection-related. During the index hospital stay, only a small proportion of patients with DM (approximately 12%) received any DM service consult. Those who received a DM consult had a higher case mix index compared to those who did not. Despite the higher acuity, there was a lower rate of ED /observation readmission in patients followed by the DM services (6.6% HIIP or ENDO vs. 9.6% no HIIP or ENDO, p = 0.0012), though no difference in the inpatient readmission rates (17.6% HIIP or ENDO vs. 17.4% no HIIP or ENDO, p = 0.89) was noted. CONCLUSIONS: Patients with both a primary or secondary diagnosis of DM have higher readmission rates. The reasons for readmission vary; patients with a principal diagnosis of DM have more DM related readmissions and those with secondary diagnosis having more infection-related readmissions. DM services were used in a small proportion of patients and may have contributed to lower DM related ED revisits. Further prospective studies evaluating the role of these services in terms of glucose management, patient education and outpatient follow up on readmission are needed to identify interventions important to reducing readmission rates.

Validation of the Caprini Venous Thromboembolism Risk Assessment Model in Critically Ill Surgical Patients.

IMPORTANCE: Appropriate risk stratification for venous thromboembolism (VTE) is essential to providing appropriate thromboprophylaxis and avoiding morbidity and mortality. OBJECTIVE: To validate the Caprini VTE risk assessment model in a previously unstudied high-risk cohort: critically ill surgical patients. DESIGN, SETTING, AND PARTICIPANTS: We performed a retrospective cohort study of 4844 adults (≥18 years old) admitted to a 20-bed surgical intensive care unit in a large tertiary care academic hospital during a 5-year period (July 1, 2007, through June 30, 2012). MAIN OUTCOMES AND MEASURES: The main study outcome was VTE (defined as patients with deep vein thrombosis or pulmonary embolism) that occurred during the patient's initial hospital admission. RESULTS: The study population was distributed among risk levels as follows: low, 5.3%; moderate, 19.9%; high, 31.6%; highest, 25.4%; and superhigh, 14.9%. The overall incidence of inpatient VTE was 7.5% and increased with risk level: 3.5% in low-risk patients, 5.5% in moderate-risk patients, 6.6% in high-risk patients, 8.6% in highest-risk patients, and 11.5% in superhigh-risk patients. Patients with Caprini scores greater than 8 were significantly more likely to develop inpatient VTE events when compared with patients with Caprini scores of 7 to 8 (odds ratio [OR], 1.37; 95% CI, 1.02-1.85; P = .04), 5 to 6 (OR, 1.35; 95% CI, 1.16-1.57; P < .001), 3 to 4 (OR, 1.30; 95% CI, 1.16-1.47; P < .001), or 0 to 2 (OR, 1.37; 95% CI, 1.16-1.64; P < .001). Similarly, patients with Caprini scores of 7 to 8 were significantly more likely to develop inpatient VTE when compared with patients with Caprini scores of 5 to 6 (OR, 1.33; 95% CI, 1.01-1.75; P = .04), 3 to 4 (OR, 1.27; 95% CI, 1.08-1.51; P = .005), or 0 to 2 (OR, 1.38; 95% CI, 1.10-1.74; P = .006). CONCLUSIONS AND RELEVANCE: The Caprini VTE risk assessment model is valid. This study supports the use of individual risk assessment in critically ill surgical patients.

Chemoprophylaxis for venous thromboembolism in otolaryngology.

IMPORTANCE: Venous thromboembolism (VTE) causes significant morbidity and mortality in surgical patients. Despite strong evidence that thromboprophylaxis reduces the incidence VTE, guidelines for prophylaxis in otolaryngology are not well established. Key to the development of VTE prophylaxis recommendations are effective VTE risk stratification and evaluation of the benefits and harms of prophylaxis. OBJECTIVE: To evaluate the effectiveness and safety of VTE chemoprophylaxis among a population of otolaryngology patients stratified by risk. DESIGN, SETTING, AND PARTICIPANTS: Retrospective cohort study of 3498 adult patients admitted for otolaryngologic surgery at a single-institution academic tertiary care medical center between September 1, 2003, and June 30, 2010. INTERVENTIONS: Patients were stratified into 2 groups based on whether they received VTE chemoprophylaxis. MAIN OUTCOMES AND MEASURES: Incidence of VTE and bleeding-related complications within 30 days after surgery. RESULTS: Of 1482 patients receiving VTE chemoprophylaxis, 18 (1.2%) developed a VTE compared with 27 of 2016 patients (1.3%) who did not receive prophylaxis (P = .75). Patients with Caprini VTE risk scores greater than 7 were less likely to have a VTE with perioperative chemoprophylaxis (5.3% vs 10.4%; P = .06). Of patients with VTE chemoprophylaxis, 3.5% developed a bleeding complication compared with 1.2% of patients without prophylaxis (P < .001). Bleeding complications were associated with concomitant use of antiplatelet medications and chemoprophylaxis. Among patients undergoing free tissue transfer, chemoprophylaxis significantly decreased the incidence of VTE (2.1% vs 7.7%; P = .002) and increased bleeding complications (11.9% vs 4.5%; P = .01). In all other patients, VTE chemoprophylaxis did not significantly influence the likelihood of VTE (1.0% vs 0.6%; P = .12) or bleeding (1.5% vs 0.9%; P = .15). CONCLUSIONS AND RELEVANCE: Effectiveness and safety of VTE chemoprophylaxis differed between patient subgroups, defined by Caprini risk score and by procedure. Effectiveness was most evident in patients with high Caprini risk scores and microvascular free tissue reconstruction. Bleeding complications were associated with VTE chemoprophylaxis administered in close proximity to potent antiplatelet therapy. The Caprini risk assessment model appears to be an effective tool to stratify otolaryngology patients by risk for VTE. Patients undergoing free tissue reconstruction merit further study before developing recommendations for VTE prophylaxis because of their higher risk of both VTE and bleeding.

Inadequate venous thromboembolism risk stratification predicts venous thromboembolic events in surgical intensive care unit patients.

BACKGROUND: Surgical intensive care unit (SICU) patients are known to be at high risk for venous thromboembolism (VTE). The 2005 Caprini Risk Assessment Model (RAM) predicts VTE risk in surgical patients. However, a physician's ability to accurately complete this RAM and the effect that inaccurate RAM completion might have on VTE risk remain unknown. STUDY DESIGN: Between 2009 and 2012, physicians completed a 2005 Caprini score for all SICU admissions at our institution. For comparison, we used a previously validated, computer-generated score. Regression-based techniques examined the effect of inadequate risk stratification on inpatient VTE risk, when controlling for other confounders. RESULTS: Among 3,338 consecutive SICU admissions, 55.2% had computer-generated scores that were higher than the physician-reported score, and 20.6% of scores were equal. Computer-generated scores were higher than physician-reported scores for both median (6 vs 5) and interquartile range (5 to 8 vs 3 to 7). Inter-rater reliability between the 2 scores was poor (kappa = 0.238). Risk score underestimation by ≥2 points was significantly associated with inpatient VTE (7.67% vs 4.59%, p = 0.002). Regression analysis demonstrated that each additional day's delay in chemoprophylaxis (odds ratio [OR] 1.05, 95% CI 1.01 to 1.08, p = 0.011) and under-risk stratification by ≥2 points (OR 2.46, 95% CI 1.53 to 3.96, p < 0.001) were independent predictors of inpatient VTE, as were higher admission APACHE score, personal history of VTE, recent pneumonia, and younger age. CONCLUSIONS: Physicians under-risk stratify SICU patients when using the 2005 Caprini RAM. As hospitals incorporate electronic medical records into daily practice, computer-calculated Caprini scores may result in more accurate VTE risk stratification. Inadequate VTE risk assessment and delay to chemoprophylaxis carry independent and significant increased risk for VTE.

Implementation of a small bowel obstruction guideline improves hospital efficiency.

BACKGROUND: We performed an internal review of triage decisions and outcomes for all patients admitted for small bowel obstruction (SBO). Concern for potential delays in operation led to formalization of an institution-wide SBO management guideline. We hypothesized that use of the guideline would improve initial triage and patient outcomes. METHODS: Members of the departments of surgery, medicine, and emergency medicine created a SBO service triage and initial management guideline that was instituted in 2011 after education and a multidisciplinary Grand Rounds on the subject. Administrative data from fiscal year 2010 (FY2010) was compared with the first 6 months of 2011. Time to computed tomography scan, the OR, general surgery (GS) consultation, and hospital duration of stay were collected and compared for those admitted to a medicine service before (Med2010) and after (Med2011) the guideline and those admitted to a general surgery service before (GS2010) and after (GS2011) the guideline. Groups were compared with Student t test and χ2 analysis. RESULTS: There were 490 SBO admissions in FY2010 and 240 in the first 6 months of 2011. After implementation of the guidelines, the percent of SBO patients admitted to GS2011 increased from 55 to 66% (P < .01). The percent of patients admitted to a medicine service requiring operation for SBO did not change from 14 to 7% for Med2011, but there was a shorter time to GS consultation (P < .001). Time from admission to operation decreased from 0.9 to 0.4 days (P < .05) with a mean decrease in hospital duration of stay of 2 days (8 ± 6 compared with 6 ± 4 days, P < .001) for those admitted during GS2011. CONCLUSION: Implementation of a hospital-wide SBO guideline that addressed initial management and triage shortened time to operative intervention and hospital duration of stay for patients requiring operative therapy for SBO.

Identifying patients at high risk for venous thromboembolism requiring treatment after outpatient surgery.

OBJECTIVE: To identify independent predictors of 30-day venous thromboembolism (VTE) events requiring treatment after outpatient surgery. BACKGROUND: An increasing proportion of surgical procedures are performed in the outpatient setting. The incidence of VTE requiring treatment after outpatient surgery is unknown. METHODS: Prospective observational cohort study using the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database from 2005 to 2009. Adult patients who had outpatient surgery or surgery with subsequent 23-hour observation were included. The main outcome measure was 30-day VTE requiring treatment. Patients were randomly assigned to derivation (N = 173,501) or validation (N = 85,730) cohorts. Logistic regression examined independent risk factors for 30-day VTE. A weighted risk index was created and applied to the validation cohort. Stratified analyses examined 30-day VTE by risk level. RESULTS: Thirty-day incidence of VTE for the overall cohort was 0.15%. Independent risk factors included current pregnancy (adjusted odds ratio [OR] = 7.80, P = 0.044), active cancer (OR = 3.66, P = 0.005), age 41 to 59 years (OR = 1.72, P = 0.008), age 60 years or more (OR = 2.48, P < 0.001), body mass index 40 kg/m or higher (OR = 1.81, P = 0.015), operative time 120 minutes or more (OR = 1.69, P = 0.027), arthroscopic surgery (OR = 5.16, P < 0.001), saphenofemoral junction surgery (OR = 13.20, P < 0.001), and venous surgery not involving the great saphenous vein (OR = 15.61, P < 0.001). The weighted risk index identified a 20-fold variation in 30-day VTE between low (0.06%) and highest risk (1.18%) patients. CONCLUSIONS: Thirty-day VTE risk after outpatient surgery can be quantified using a weighted risk index. The risk index identifies a high-risk subgroup of patients with 30-day VTE rates of 1.18%.

Stratifying the risk of venous thromboembolism in otolaryngology.

OBJECTIVE: The consequences of perioperative venous thromboembolism (VTE) are devastating; identifying patients at risk is an essential step in reducing morbidity and mortality. The utility of perioperative VTE risk assessment in otolaryngology is unknown. This study was designed to risk-stratify a diverse population of otolaryngology patients for VTE events. STUDY DESIGN: Retrospective cohort study. SETTING: Single-institution academic tertiary care medical center. SUBJECTS AND METHODS: Adult patients presenting for otolaryngologic surgery requiring hospital admission from 2003 to 2010 who did not receive VTE chemoprophylaxis were included. The Caprini risk assessment was retrospectively scored via a validated method of electronic chart abstraction. Primary study variables were Caprini risk scores and the incidence of perioperative venous thromboembolic outcomes. RESULTS: A total of 2016 patients were identified. The overall 30-day rate of VTE was 1.3%. The incidence of VTE in patients with a Caprini risk score of 6 or less was 0.5%. For patients with scores of 7 or 8, the incidence was 2.4%. Patients with a Caprini risk score greater than 8 had an 18.3% incidence of VTE and were significantly more likely to develop a VTE when compared to patients with a Caprini risk score less than 8 (P < .001). The mean risk score for patients with VTE (7.4) was significantly higher than the risk score for patients without VTE (4.8) (P < .001). CONCLUSION: The Caprini risk assessment model effectively risk-stratifies otolaryngology patients for 30-day VTE events and allows otolaryngologists to identify patient subgroups who have a higher risk of VTE in the absence of chemoprophylaxis.

PA-driven VTE risk assessment improves compliance with recommended prophylaxis.

OBJECTIVE: Venous thromboembolism (VTE) is a major cause of morbidity and mortality in hospitalized patients, particularly surgical patients. We hypothesize that PAs are well-positioned to assist health systems with implementation of efforts to reduce the rates of this in-hospital complication and increase adherence to published standards for VTE prophylaxis. METHODS: We conducted a retrospective cohort study of general surgical patients who underwent an operation at the University of Michigan between July 2005 and June 2007. The PAs in the Department of Surgery implemented a VTE assessment and prophylaxis intervention in June 2006. Preintervention VTE risk scores were calculated using patient demographic information, operating room data, and diagnosis codes from the International Statistical Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Those calculated scores were then tested on patients who had a VTE risk score documented by PAs. Postintervention VTE was determined using ICD-9-CM diagnosis codes for deep vein thrombosis (DVT) or pulmonary embolism (PE) and identified as "acquired in hospital" or readmitted with a principal diagnosis of DVT or PE within 30 days following surgery. We then compared the frequency with which patients in the preintervention and postintervention periods received recommended VTE prophylaxis. RESULTS: Overall, 2,046 patients underwent surgery during the study period. There were 1,079 patients in the preintervention group and 967 patients in the postintervention group, with no systematic differences in the case mix between the two groups. For all patients with a risk score of 3 or higher (indicating high and highest risk combined), orders for appropriate prophylaxis improved from an average of 23.1% in the preintervention group to an average of 63.7% in the postintervention group. Similarly, for all patients with a risk score of 5 or higher (indicating highest risk), orders for appropriate prophylaxis improved from an average of 29.4% in the preintervention group to an average of 69.5% in the postintervention group. CONCLUSIONS: Through a PA-driven VTE risk assessment process, we dramatically increased the number of patients within our health system who were prescribed appropriate orders for VTE prophylaxis according to published guidelines and according to individual patient risk.

A validation study of a retrospective venous thromboembolism risk scoring method.

OBJECTIVES: Validate a retrospective venous thromboembolism (VTE) risk scoring method, which was developed at the University of Michigan Health System and based on the Caprini risk assessment model, and assess the confounding effects of VTE prophylaxis. BACKGROUND: Assessing patients for risk of VTE is essential to initiating appropriate prophylaxis and reducing the mortality and morbidity associated with deep vein thrombosis and pulmonary embolism. METHODS: VTE risk factors were identified for 8216 inpatients from the National Surgical Quality Improvement Program using the retrospective scoring method. Logistic regression was used to calculate odds ratios (OR) for VTE within 30 days after surgery for risk factors and risk level. A bivariate probit model estimated the effects of risk while controlling for adherence to prophylaxis guidelines. RESULTS: Distribution of the study population by risk level was highest, 52.1%; high, 36.5%; moderate, 10.4%; and low, 0.9%. Incidence of VTE within 30 days was overall 1.4%; by risk level: highest, 1.94%; high, 0.97%; moderate, 0.70%; low, 0%. Controlling for length of hospitalization (>2 d) and fiscal year, pregnancy or postpartum (OR = 8.3; 1.0-68, P < 0.05), recent sepsis (4.0; 1.4-10.9, P < 0.01), malignancy (2.3; 1.5-3.3, P < 0.01), history of VTE (2.1; 1.1-4.1, P < 0.05), and central venous access (1.8; 1.1-3.0, P < 0.05) were significantly associated with VTE. Risk level was significantly associated with VTE (1.9; 1.3-2.6, P < 0.01). The bivariate probit demonstrated significant correlation between the probability of VTE and lack of adherence to prophylaxis guidelines (rho = 0.299, P = 0.013). CONCLUSION: The retrospective risk scoring method is valid and supports use of individual patient assessment of risk for VTE within 30 days after surgery.

Improved outcomes in patients with head and neck cancer using a standardized care protocol for postoperative alcohol withdrawal.

OBJECTIVE: To show clinical benefit in the main outcome measures by the use of a standardized protocol for identification, characterization, and treatment of alcohol withdrawal syndrome (AWS) in postoperative patients with head and neck cancer. DESIGN: Prospective cohort study with a retrospective cohort control. SETTING: Tertiary care university. PATIENTS: A total of 26 consecutive postoperative patients with AWS were selected from among 652 patients with head and neck cancer to be enrolled in the protocol from March 2003 through March 2005. Controls consisted of 14 of 981 consecutive patients with AWS from March 2000 through December 2002. INTERVENTION: Application of a standardized care protocol. MAIN OUTCOME MEASURES: Sensitivity and specificity of preoperative screening for AWS risk, predictability of outcomes, length of stay, transfers to the intensive care unit (ICU), AWS symptoms, postoperative morbidity and mortality, doses of pharmacotherapy required, and charges. RESULTS: Protocol patients demonstrated significantly fewer AWS-related ICU transfers and less delirium and violence than preprotocol patients. Mortality, wound complications, hospital charges, and doses of benzodiazepines, clonidine, and haloperidol were not significantly different between these 2 groups. Preoperative medical history correlated poorly with AWS outcomes. Screening was 87.5% sensitive and 99.7% specific. Late enrollees to the protocol (false-negative screening results) showed many significantly worse outcomes than immediate enrollees. CONCLUSION: Use of the standardized AWS symptom-triggered protocol decreased delirium, violence, and AWS-related ICU transfers without significantly increasing hospital charges.

Exploratory analyses of the "failure to rescue" measure: evaluation through medical record review.

Clinicians and hospital leaders are interested in assessing the degree to which the failure-to-rescue indicator identifies true problems in processes of care at the individual or system level. Failure-to-rescue complications are flagged through administrative data, and the clinical course of events is evaluated. The review suggests that many factors influence whether a case is included in the measure, such as existing health problems, the presence of complex comorbidities, and variation in clinical documentation and coding practices.

Do the AHRQ patient safety indicators flag conditions that are present at the time of hospital admission?

OBJECTIVE: The Agency for Healthcare Research and Quality (AHRQ) developed 20 patient safety indicators (PSIs) to identify potentially preventable complications of acute inpatient care based on administrative data. The objective of this patient safety performance study was to assess the impact of cases flagged by each PSI for diagnoses that were actually present on admission on unadjusted PSI rates. METHODS: The latest AHRQ PSI software, which allows users to produce 14 of the 20 PSIs for adult inpatients both without and with a "present on admission" (PoA) variable, was applied to administrative data for adult patients discharged from the University of Michigan Health System (UMHS) in 2006. The impact of the PoA values on unadjusted PSI rates was evaluated. Because of concerns about the accuracy of PoA values, results were compared with those of a prior analysis at UMHS that was similar but based on a review of medical records. FINDINGS: Thirteen PSIs had at least 1 case in the numerator. Rates for all but 1 of the 13 were lower using the PoA values and the reduction was statistically significant for 5 PSIs: decubitus ulcer (P < 0.001), foreign body left in (P = 0.033), selected infections due to medical care (P < 0.001), postoperative physiologic and metabolic derangement (P = 0.039), and postoperative pulmonary embolism or deep vein thrombosis (P < 0.001). Results were consistent with those of the analysis of medical records. CONCLUSIONS: Unadjusted PSI rates at UMHS are substantially overstated, because the PSIs do not differentiate preexisting conditions from complications and therefore include false positive cases. Because of these findings and the lack of a broader study of the validity of the indicators, PSIs should not be used to profile hospital performance.

Developing dashboards to measure and manage inpatient pharmacy costs.

PURPOSE: A pharmacy dashboard was developed and implemented to identify trends in drug use; normalize data for patient volumes; allow the user to determine whether changes were due to cost increases, actual use changes, or both; permit evaluation of drug use within key diagnoses for individual services; and allow multiple formats and layers of analysis. SUMMARY: At the University of Michigan, a drug cost and drug-use database had been developed, and an enterprise data warehouse had been implemented. The data warehouse served as the core data source for the pharmacy dashboard effort. The department of pharmacy services worked with the clinical information and decision support services (CIDSS) unit to develop and promote the use of the pharmacy dashboard. On the basis of the core data in the data warehouse, the dashboard was designed to harness these data for the useful and actionable presentation to service chiefs, individual house officers, and administration for a better understanding of drug costs and use within the institution. In addition to the drug cost and drug-use trend data, the dashboard included help features such as access to the Drugdex database for detailed clinical information about specific drugs, including indications, adverse effects, and drug interactions. CONCLUSION: The cost reports that were developed through the collaboration between the CIDSS and the department of pharmacy services have permitted a more robust understanding of drug-use trends by a wider audience of physician leaders and clinical pharmacists at the institution. These data are critical in formulating strategies to control drug expenses by each service.

Population-based medical and disease management: an evaluation of cost and quality.

Reports by the Institute of Medicine and the Health Care Financing Administration have emphasized that the integration of medical care delivery, evidence-based medicine, and chronic care disease management may play a significant role in improving the quality of care and reducing medical care costs. The specific aim of this project is to assess the impact of an integrated set of care coordination tools and chronic disease management interventions on utilization, cost, and quality of care for a population of beneficiaries who have complementary health coverage through a plan designed to apply proactive medical and disease management processes. The utilization of health care services by the study population was compared to another population from the same geographic service area and covered by a traditional fee-for-service indemnity insurance plan that provided few medical or disease management services. Evaluation of the difference in utilization was based on the difference in the cost per-member-per-month (PMPM) in a 1-year measurement period, after adjusting for differences in fee schedules, case-mix and healthcare benefit design. After adjustments for both case-mix and benefit differences, the study group is $63 PMPM less costly than the comparison population for all members. Cost differences are largest in the 55-64 and 65 and above age groups. The study group is $115 PMPM lower than the comparison population for the age category of 65 years and older, after adjustments for case-mix and benefits. Health Plan Employer and Data Information Set (HEDIS)-based quality outcomes are near the 90th percentile for most indications. The cost outcomes of a population served by proactive, population-based disease management and complex care management, compared to an unmanaged population, demonstrates the potential of coordinated medical and disease management programs. Further studies utilizing appropriate methodologies would be beneficial.

Linear programming to optimize performance in a department of surgery.

BACKGROUND: Linear programming is an analytic method that can be used to develop models for health care that optimize distribution of resources through mathematical means. STUDY DESIGN: The linear programming model contained objective, decision, and constraint elements. The objective was to optimize financial outcomes for both the hospital and physicians in the Department of Surgery. The decision concerns procedure mix or the number of each type of surgical procedure. Constraints apply to resources that are consumed during the course of the patient's surgical encounter. RESULTS: The optimal solution produced an increase in professional payments of 3.6% and an increase in hospital total margin of 16.1%. This solution favored surgical procedures that require inpatient care; these patients had greater comorbidity, reflected in a higher case-mix index of 3.74 compared to 2.97. Substantial differences were noted in use of general care and ICU days, and in consumption of preoperative, intraoperative, and recovery room time. CONCLUSIONS: Aligning quality surgical care with optimal financial performance may be assisted by mathematical models such as linear programming.

Relevance of the Agency for Healthcare Research and Quality Patient Safety Indicators for children's hospitals.

OBJECTIVES: Patient safety indicators (PSIs) were developed by the Agency for Healthcare Research and Quality. Our objectives were (1) to apply these algorithms to the National Association of Children's Hospitals and Related Institutions (NACHRI) Aggregate Case Mix Comparative Database for 1999-2002, (2) to establish mean rates for each of the PSI events in children's hospitals, (3) to investigate the inadequacies of PSIs in relation to pediatric diagnoses, and (4) to express the data in such a way that children's hospitals could use the PSIs determined to be appropriate for pediatric use for comparison with their own data. In addition, we wanted to use the data to set priorities for ongoing clinical investigations and to propose interventions if the indicators demonstrated preventable errors. METHODS: The Agency for Healthcare Research and Quality PSI algorithms (version 2.1, revision 1) were applied to children's hospital administrative data (1.92 million discharges) from the NACHRI Aggregate Case Mix Comparative Database for 1999-2002. Rates were measured for the following events: complications of anesthesia, death in low-mortality diagnosis-related groups (DRGs), decubitus ulcer, failure to rescue (ie, death resulting from a complication, rather than the primary diagnosis), foreign body left in during a procedure, iatrogenic pneumothorax, infection attributable to medical care (ie, infections related to surgery or device placement), postoperative hemorrhage or hematoma, postoperative pulmonary embolism or venous thrombosis, postoperative wound dehiscence, and accidental puncture/laceration. RESULTS: Across the 4 years of data, the mean risk-adjusted rates of PSI events ranged from 0.01% (0.1 event per 1000 discharges) for a foreign body left in during a procedure to 14.0% (140 events per 1000 discharges) for failure to rescue. Review of International Classification of Diseases, Ninth Revision, Clinical Modification codes associated with each PSI category showed that the failure to rescue and death in low-mortality DRG indicators involved very complex cases and did not predict preventable events in the majority of cases. The PSI for infection attributable to medical care appeared to be accurate the majority of the time. Incident risk-adjusted rates of infections attributable to medical care averaged 0.35% (3.5 events per 1000 discharges) and varied up to fivefold from the lowest rate to the highest rate. The highest rates were up to 1.8 times the average. CONCLUSIONS: PSIs derived from administrative data are indicators of patient safety concerns and can be relevant as screening tools for children's hospitals; however, cases identified by these indicators do not always represent preventable events. Some, such as a foreign body left in during a procedure, iatrogenic pneumothorax, infection attributable to medical care, decubitus ulcer, and venous thrombosis, seem to be appropriate for pediatric care and may be directly amenable to system changes. Evidence-based practices regarding those particular indicators that have been reported in the adult literature need to be investigated in the pediatric population. In their present form, 2 of the indicators, namely, failure to rescue and death in low-mortality DRGs, are inaccurate for the pediatric population, do not represent preventable errors in the majority of pediatric cases, and should not be used to estimate quality of care or preventable deaths in children's hospitals. The PSIs can assist institutions in prioritizing chart review-based investigations; if clusters of validated events emerge in reviews, then improvement activities can be initiated. Large aggregate databases, such as the NACHRI Case Mix Database, can help establish mean rates of potential pediatric events, giving children's hospitals a context within which to examine their own data.

Clinical redesign using all patient refined diagnosis related groups.

OBJECTIVE: Clinical redesign of processes in hospitals that care for children has been limited by a paucity of severity-adjusted indicators that are sensitive enough to identify areas of concern. This is especially true of hospitals that analyze pediatric patient care using standard Centers for Medicare and Medicaid Services (CMS) diagnosis-related groups (DRGs). The objectives of this study were to determine whether 1) utilization of all-patient refined (APR)-DRG severity-adjusted indicators (length of stay, cost per case, readmission rate) from the National Association of Children's Hospitals and Related Institutions (NACHRI) database could identify areas for improvement at University of Michigan Mott Children's Hospital (UMMCH) and 2) hospital staff could use the information to implement successful clinical redesign. METHODS: The APR-DRG Classification System (version 20) was used with the NACHRI Case Mix Comparative Database by severity level comparison from 1999 to 2002. Indicators include average length of stay (ALOS), case mix index, cost per case, and readmission rate for low acuity asthma (APR-DRG 141.1). UMMCH cases of 141.1 (n = 511) were compared with NACHRI 141.1 (n = 64,312). Although not part of the standard report, mortality rates were calculated by NACHRI for UMMCH and an aggregate of NACHRI member children's hospitals. RESULTS: Data from 1999 revealed that in noncomplicated asthma cases (level 1 severity), the UMMCH ALOS versus NACHRI ALOS was slightly longer (UMMCH 2.16 days vs NACHRI 2.14 days), and the cost per case was higher (UMMCH $2824 vs NACHRI 2738 dollars), whereas levels 2, 3, and 4 cases (moderate, major, and extreme severity) indicated the ALOS and cost per case were lower than the national aggregate. This showed that the APR-DRG system was sensitive enough to distinguish variances of care within a diagnosis according to severity level. After analysis of internal data and meeting with clinicians to review the indicators, 3 separate clinical processes were targeted: 1) correct documentation of comorbidities and complications, 2) standardized preprinted orders were created with the involvement of the pediatric pulmonologists, and 3) standardized automatic education for parents was started on the first day of admission. Yearly data were reviewed and appropriate adjustments made in the education of both residents and staff. In 2002, the UMMCH ALOS dropped to 1.75 +/- .08 days from 2.16 +/- .09. In 2002, the NACHRI ALOS was 2.00 days +/- 0.01 versus the UMMCH ALOS of 1.75 days +/- 0.0845, indicating that the UMMCH ALOS dropped significantly lower than the NACHRI aggregate database over the 3-year period. Cost per case of UMMCH compared with NACHRI after the 3 years indicated that UMMCH increased 12%, whereas the NACHRI aggregate increased 18%. These data show that length of stay and cost per case relative to the national database improved after clinical redesign. Improvements have been sustained throughout the 3-year period. Readmission rates ranged from 2.97% to 0.80% and were less than the national cohort by the third year. There were no mortalities in the UMMCH inpatient asthma program. This demonstrates that clinicians believed that the data from the APR-DRG acuity-adjusted system was useful and that they were then able to apply classical clinical redesign strategies to improve cost-effectiveness and quality that was sustained over 3 years. CONCLUSIONS: Severity-adjusted indicators were useful for identifying areas appropriate for clinical redesign and contributed to the improvement in cost-effective patient care without a detriment in quality indicators. This method of using a large comparative database, having measures of severity, and using internal analysis is generalizable for pediatric hospitals and can contribute to ongoing attempts to improve cost-effectiveness and quality in medical care.

Reducing costs and improving outcomes of percutaneous coronary interventions.

OBJECTIVE: To describe cost reduction and quality improvement efforts in our percutaneous coronary intervention (PCI) program and how risk adjustment was used to assess the effects of these changes. STUDY DESIGN: Single center registry analysis. PATIENTS AND METHODS: Data were collected on 2158 PCIs performed between July 1, 1994, and June 30, 1997. Of these, 1126 PCIs reflected care provided after implementation of competitive bidding for catheterization lab supplies, and efforts to reduce the use of postprocedure heparin and to implement early arterial sheaths removal (postbidding period). Hospital costs were estimated using a microcost accounting method. In-hospital mortality rates during the 2 time periods were compared using standardized mortality ratio estimated with a previously validated risk adjustment model for in-hospital mortality. RESULTS: Compared with the prebidding period, the postbidding period was characterized by a significantly higher utilization of new technology (coronary stents and atherectomy devices 46% vs 25%; abciximab 19.1% vs 3.7, P<.01), and an overall increase in case complexity. Despite these changes, the average and median postbidding cost per case was dollars 1223 and dollars 1444 lower, respectively, than in the prebidding period. After adjustment for comorbidities, procedure variables, complications, and length of hospital stay, multivariate regression modeling identified the postbidding period as an independent predictor of lower hospital costs (P<.001) with an estimated adjusted cost savings of dollars 460. These cost savings were associated with trends toward a lower observed mortality rate, a higher predicted mortality rate, and a significantly lower standardized mortality ratio (SMR .71; 95% CI 0.48-0.9; P<.05). CONCLUSION: Despite an increase in case complexity and utilization of new technology, cost reductions can be achieved through competitive bidding for supplies and modifications of periprocedure care. Risk adjustment appears to be a valid tool for assessing the effectiveness of these efforts independently from changes in case mix.