Shawn J. Gagne

Adverse Drug Events After Hospital Discharge in Older Adults: Types, Severity, and Involvement of Beers Criteria Medications

To characterize adverse drug events (ADEs) occurring within the high‐risk 45‐day period after hospitalization in older adults.

Vaccine counseling: A content analysis of patient–physician discussions regarding human papilloma virus vaccine☆

OBJECTIVES (1) Describe content and character of patient-physician human papilloma virus (HPV) vaccine discussions; (2) explore the relationship between selected characteristics and vaccine uptake. METHODS Content analyses were conducted on 184 transcripts of audio-taped patient encounters with 11-26 year old female patients that occurred from August 2008 to March 2009 and contained mention of the HPV vaccine. Directed qualitative content analysis sought to identify key themes with a focus on elements related to communication. Quantitative content analysis included determination of associations between selected factors (e.g., physician specialty, communication variables, patient age) and vaccination rates. RESULTS Communication themes identified though qualitative content analysis demonstrated potential opportunities for improvement in vaccine communication were identified. Quantitative content analysis showed twenty-eight percent of eligible patients received HPV vaccine and on average these patients were younger (17.0 vs. 19.6 years). The youngest and oldest patients were vaccinated less frequently. CONCLUSIONS Targeting age groups with lower vaccination rates may increase overall vaccine uptake. Additional quantitative analyses of patient-physician discussions about vaccine may generate further recommendations regarding optimal communication strategies for HPV vaccine counseling.

An electronic health record-based intervention to increase follow-up office visits and decrease rehospitalization in older adults

To assess the effect of an electronic health record–based transitional care intervention involving automated alerts to primary care providers and staff when older adults were discharged from the hospital.

Baseline and Follow-up Laboratory Monitoring of Cardiovascular Medications

Background Laboratory monitoring of medications is typically used to establish safety prior to drug initiation and to detect drug-related injury following initiation. It is unclear whether black box warnings (BBWs) as well as evidence- and consensus-based clinical guidelines increase the likelihood of appropriate monitoring. Objective: To determine the proportion of patients newly initiated on selected cardiovascular medications with baseline assessment and follow-up laboratory monitoring and compare the prevalence of laboratory testing for drugs with and without BBWs and guidelines. Methods: This cross-sectional study Included patients aged 18 years or older from a large multispecialty group practice who were prescribed a cardiovascular medication (angiotensin converting enzyme inhibitors, angiotensin II receptor blockers, amlodarone, digoxin, lipid-lowering agents, diuretics, and potassium supplements) between January 1 and July 31, 2008. The primary outcome measure was laboratory test ordering for baseline assessment and follow-up monitoring of newly initiated cardiovascular medications. Results: The number of new users of each study drug ranged from 49 to 1757 during the study period. Baseline laboratory test ordering across study drugs ranged from 37.4% to 94.8%, and follow-up laboratory test ordering ranged from 20.0% to 77.2%. Laboratory tests for drugs with baseline laboratory assessment recommendations in BBWs were more commonly ordered than for drugs without BBWs (66.4% vs 78.0%, p < 0.001), Drugs with follow-up monitoring recommendations in clinical guidelines had a lower prevalence of monitoring (33.1% va 50.7%. p< 0.001). Conclusions: Baseline assessment of cardiovascular medication monitoring is variable. Quality measurement of adherence to BBW recommendations may improve monitoring.

PS2-6: Who Opens Alerts to Physicians? (And Who Doesn’t?)

Background/Aims With the increasing adoption of electronic medical records (EMR), there are many opportunities to implement systems of information flow, like alerts to providers. Tests of such systems have varied results. Methods Within a multispecialty group practice that uses Epic EMR, we conducted a trial of automated alerts to primary care physicians and their staff when older patients were discharged from hospital to home. We generated alerts for new medications, drug interactions, dose problems, or monitoring needs. Provider staff received alerts to schedule an office visit. Alerts appeared in the recipient’s Epic in-basket. Using EMR “digital crumbs”, we tracked the length of time before alerts were viewed. We analyzed the impact of physician age, gender, department, and employment status (full-time, part-time). We also analyzed the impact of patient conditions, including age, gender, comorbidity, and number of office visits in the previous year. Results Of 763 alerts to physicians, 616 (81%) were opened within one day. Characteristics associated with timely opening were age <50 (OR 1.7, 95% CI 1.1, 2.6) and full-time employment (OR 2.9, 95% CI 1.6, 5.2). Of 1928 alerts to staff, 1173 (61%) were opened within one day. Physician characteristics associated with staff opening of alerts were gender (staff of male physicians were more likely to open the alerts within one day [OR 1.8, 95% CI 1.4, 2.4]), and being a member of the Family Medicine department (OR 1.9, 95% CI 1.3, 2.6) or sub-specialty department (OR 16.6, 95% CI 2.3, 122.3). Staff of full-time physicians were less likely to open alerts within one day (OR 0.64, 95% CI 0.47, 0.87). Controlling for patient characteristics had no impact on results. Conclusions Special efforts may be required to reach physicians working part-time and older physicians. The characteristics related to staff opening of alerts are specific to this group practice, but the high level of variability across physician types and departments is likely to be an issue in many settings. Design of a system directed at reaching staff quickly may require in-depth assessment of work flow and communication patterns in clinical departments.

PS1-40: Laboratory Monitoring of HEDIS Medications in the Ambulatory Setting: The Relative Contribution of Physician and Patient Behavior to Undermonitoring

Background/Aims Initiative to improve the quality and safety of pharmaceutical care have lead to the development of quality of care measures including standards for the appropriate monitoring of high-risk medications (defined as medications commonly implicated in adverse drug events or with narrow therapeutic window). While low rates of laboratory monitoring of high-risk medications might indicate poor physician performance, it is unclear how much patient non-adherence to physician-ordered tests contribute to undermeasruement of physician quality. The study aim is to determine, for Health Employer Data Information Set (HEDIS) quality of care, high-risk medication laboratory monitoring measures: the prevalence of completion of recommended monitoring tests; and the proportion of incomplete testing attributable to lack of clinician test ordering relative to patient non-adherence to ordered tests. Methods and Results We measured the ordering and completion of laboratory tests for HEDIS-based high-risk medications (ACE inhibitors, ARBs, digoxin and diuretics, and anticonvulants) in a large multispecialty group practice between January 1, 2008 and July 31, 2008. Laboratory test completion for HEDIS cardiovascular drugs was higher than for anticonvulsants. Each cardiovascular drug had a completion rate of > 85%, while the completion rate for anticonvulsants ranged from 30–75%. The lowest test completion rate was for phenobarbital levels to monitor phenobarbital (30%). For all cardiovascular and anticonvulsant drugs, the proportion of recommended tests not ordered by the clinician ranged from 5% to 60%. The lowest test order prevalence was for phenobarbital level for phenobarbital use (35%), followed by valproic acid level for valproate use(48%), and carbamazepine level for carbamazepine use (60%). Rates of patient noncompletion of ordered tests for all drugs was generally <10%. Conclusion Completion of laboratory monitoring tests for high-risk cardiovascular medications was higher than for high-risk anticonvulsants according to HEDIS guidelines. Clinician ordering behavior exhibited more variation than patient adherence to test orders. Underestimation of HEDIS quality of care monitoring due to patient non-adherence is minimal for cardiovascular medications, but higher for anticonvulsants.

B3-4: Identification of Adverse Drug Events Post-Hospital Discharge in a Geriatric Population.

Background/Aims Adverse drug events (ADEs), especially those that may be preventable, are among the most serious concerns regarding medication use in older persons. The purpose of this study was to describe the incidence, severity and preventability of ADEs occurring within 45 days post-hospitalization in an ambulatory geriatric population. Methods We studied 1000 consecutive discharges of patients aged 65 and older who received medical care from a large multispecialty medical group in Central Massachusetts. Discharges were excluded if the discharge diagnosis was psychiatric or if discharges were not to home. Three clinical pharmacist investigators reviewed the ambulatory records of each discharged patient to identify drug-related incidents occurring during the 45-day period post hospital discharge. Drug-related incidents were presented to a pair of physician-reviewers who independently classified incidents as to whether an adverse drug event was present, the severity of the event, and whether the event was preventable. When the physician-reviewers disagreed on the classification of an incident, they met and reached consensus; consensus was reached in all instances where there was initial disagreement. Results There were 242 ADEs identified, of which 35% (n = 84) were considered preventable. Of the preventable ADEs, 63% (n = 53) were categorized as significant, 32% (n = 27) were serious, and 5% (n = 4) were life-threatening. Nearly half of all ADEs occurred within the first two weeks following discharge. Conclusions Adverse drug events are common and often preventable among older persons in the ambulatory setting. The substantial portion of serious events that were considered preventable suggests opportunities for improving care during the post-hospital discharge period.

CC1-02: Technological Resources and Personnel Costs Required to Implement an Automated Alert System for Primary Care Physicians When Patients Transition from Hospitals to Home

Background/Aims With the adoption of electronic medical records (EMR) by medical group practices, there are opportunities to improve the quality of care for patients who are discharged from hospitals and intermediate care facilities. While transitions within vertically-integrated healthcare systems are amenable to EMR-based transition interventions, there is little guidance for medical groups without integrated hospital-EMR access who wish to automate the flow of patient information during critical transitions in care. Our aim was to describe the technological resources, expertise and time needed to develop and implement an automated system providing critical information and alerts to primary care physicians when their patients transition from hospitals or skilled nursing facilities to home. Methods Within a large medical group practice with an EMR, we developed and implemented an automated alert system that provides notification of discharges, reminders of the need for follow-up visits, new drugs added during the in-patient stay, warnings about drug-drug interactions, and recommendations for dosing changes and laboratory monitoring of high risk drugs. We tracked components of the information system required to accomplish this as well as the time spent by team members. We used US national averages of relevant hourly wages to estimate personnel costs. Results Critical components of the information system are notifications of hospital discharges through an admission, discharge and transfer registration (ADT) interface, linkage to the group practice scheduling system, timely access to information on pharmacy dispensing and lab tests, and an interface engine to direct messages to specific physicians and staff. Total personnel cost was

PS2-13: Why Patients Fail to Complete Laboratory Monitoring Requests

76,314. Nearly half (47%) was for 614 hours by physicians who developed content, provided overall project management, and reviewed alerts during a test period to ensure that only “actionable” alerts would be sent. Discussion Implementing a system to provide a flow of critical information about patient transitions requires strong internal informatics expertise, cooperation between facilities and ambulatory providers, development of a number of electronic linkages, and extensive commitment of physician time.

Technological resources and personnel costs required to implement an automated alert system for ambulatory physicians when patients are discharged from hospitals to home

Background/Aims Inadequate laboratory monitoring of high-risk medications is a medical error associated with preventable adverse drug events. Patient failure to complete ordered laboratory tests is not uncommon. We conducted in-depth interviews with patients who did not complete a recently ordered laboratory test to identify barriers to test completion. Methods The target population was patients in a large multispecialty group practice who failed to complete an ordered laboratory test. We conducted semi-structured interviews with 18 patients who did not complete a monitoring test order related to 1 of 7 high-risk medications (angiotensin converting enzyme inhibitors (ACE) inhibitors, angiotensin II receptor blockers (ARBs), digoxin, phenytoin, statins, thyroid replacement hormone, or valproic acid) between July 2008 and October 2010. For comparison, we also interviewed 5 patients who completed all ordered laboratory tests. The interviews occurred in person when possible (n=17) as well as by phone (n=6). All interviews were recorded and transcribed and analyzed using a grounded theory approach. Results The laboratory monitoring discussions included the following themes: physician communication about lab tests, patient comprehension of lab test purpose, patient attitude toward lab tests, and barriers to test completion. The most prominent barrier to test completion was forgetting. Patients often did not realize they had missed an ordered test. When patients did acknowledge missing an ordered test, they often noted that they did not encounter any barriers besides forgetting. Regarding potential interventions, patients reported that understanding the indication for the tests would not help reduce non-completion. Several patients understood the reason for the test, but still failed to complete it. While some thought it was important to understand the reason for the test, others trusted their physician and did not believe it was essential to understand. Some patients were using an online personal health portal system offered through the practice that included information about upcoming tests. Conclusions Forgetting was the most prominent theme for non-completion of ordered test. Patients felt that providing timely (near desired test completion date) patient reminders, either by phone or by e-mail, would likely improve completion rates of laboratory monitoring for high-risk medications, while providing more information about the test indication would not.