Adverse Drug Event Prevention

Opening Remarks and Conference Agenda

Don Wright, MD, MPH (Deputy Assistant Secretary for Health and Director of ODPHP); Wanda Jones, DrPH (Principal Deputy Assistant Secretary for Health); and Dale Hu, MD, MPH (Acting Director, Division of Health Care Quality, ODPHP; Captain [CAPT], US Public Health Service) welcomed conference participants and gave an overview of the ADE Action Plan.

Morning plenary sessions provided an overview of the surveillance, evidence-based prevention, incentives and oversights, and additional research needs of 3 drug classes. Afternoon breakout sessions allowed for facilitated dialogue among conference participants in order to discuss measures for tracking national progress in ADE prevention and to identify opportunities to ensure safe and high-quality care for patients who are prescribed the high-priority medications outlined in the ADE Action Plan.

Morning Plenary Session 1: Preventing Anticoagulant ADEs and Monitoring Progress

Nadine Shehab, PharmD, MPH (Senior Service Fellow, Division of Healthcare Quality Promotion, Centers for Disease Control and Prevention)

Dr Shehab provided an overview of the tremendous national burden of anticoagulant-related ADEs and presented strategies for population-based prevention of these events. She outlined the capabilities and challenges of existing anticoagulant-related ADE surveillance systems. For example, at the national level, administrative claims data can be useful for capturing bleeding events, but the accuracy of diagnostic coding (International Classification of Diseases, Ninth Revision [ICD-9]) for identifying these events is not well-established. The impact of the transition to ICD-10 is unknown. Current systems developed for the monitoring and prevention of health care–associated infections, such as the Centers for Disease Control and Prevention’s (CDC) National Healthcare Safety Network, might serve as a good example for benchmarking progress and gaps in ADE prevention efforts.

In addition to highlighting the need for improved surveillance systems, Dr Shehab outlined health policy strategies that might be effective in preventing anticoagulant-related ADEs. She recommended that 2 key areas be addressed: minimizing payment or coverage barriers to delivering evidence-based anticoagulant management and expanding nationally recognized health care quality measures to address safe use of anticoagulants.

Dr Shehab noted that current nationally recognized health care quality measures generally focus on whether anticoagulants are being used, but do not include adequate metrics for whether anticoagulants are used safely. For example, the National Quality Forum (NQF) has endorsed a number of measures related to venous thromboembolism prophylaxis but far fewer that relate to bleeding risk mitigation.

Dr Shehab concluded by summarizing that the surveillance of anticoagulant-related ADEs is currently conducted through a variety of national and local data sources (Table 1). For example, the Medicare Patient Safety Monitoring System (MPSMS) collects national data on inpatient bleeding events within selected patient populations via retrospective chart review, and a number of local process metrics support appropriate monitoring of patients on warfarin. Moving forward, the ADE Action Plan suggests that existing quality measures should be expanded to address (1) safe use of anticoagulants, (2) use of new oral anticoagulants, (3) high-risk populations and settings (eg, long-term care), and (4) clinical outcomes versus surrogate indicators.

OPEN IN VIEWER

Table 1.

Inventory of Potential Measures for Identifying Anticoagulant ADEs.

Anticoagulant Outcome and Process Measures
Scope		Measure/Data Elements	Agency	Data System/Brief Description
Inpatient	National	1. Incidence of bleeding events and coagulation abnormalities	AHRQ	a. HCUP—Inpatient stays with ICD-9 codes and E-codes from administrative claims and/or EHR data
				b. MPSMS—Inpatient stays with coagulation laboratory abnormalities from medical record review
		2. Incidence of bleeding events and coagulation abnormalities in Federal Health System Populations	a. IHS	a. Resource and Patient Management System (RPMS) EHR—Adverse Reaction Tracking (ART) System
			b. VA	b. Integrated Databases—ADE identification by ICD-9 codes, primary hospitalizations, ED or clinic visits, and laboratory values
	Local/facility-level	3. Quality improvement process measures		Individual Facility Reports:
				a. TJC 2014 National Patient Safety Goal 03.05.01 element: Evaluate anticoagulation safety practices, take action to improve practices, and measure the effectiveness of those actions in a time frame determined by the organization
				b. Number of VTE patients receiving unfractionated heparin with platelet count monitoring by nomogram or protocol (former NQF 0374)
				c. Number of VTE patients discharged to home, home health, home hospice with written discharge instructions (former NQF 0375)
Inpatient/outpatient		4. Rates of clinician-diagnosed or patient-reported ADEs		Spontaneous Reports:
			a. DOD	a. Patient Safety Reporting System
			b. FDA	b. FAERS
			c. VA	c. VA-ADERS
		5. Proposed Meaningful Use Measures (Stage 3)	ONC	a. Inpatient Electronic Anticoagulation Management Flowsheet
				b. Clinical notification of INR testing with interacting anti-infective medication
				c. INR test performed 7 to 14 days following out-of-range INR
				d. INR retesting evaluation (>30 days since last INR test)
				e. Last INR test (patients’ lists stratified by 30-day intervals)
Outpatient	National	6. Incidence of bleeding events and coagulation abnormalities	a. FDA	a. Sentinel, Mini-Sentinel—ED visits, hospitalizations for ADEs from administrative claims/EHR
			b. CDC	b. NEISS-CADES—ED visits, emergent hospitalizations, diagnosed anticoagulant-related bleeding or coagulation abnormality from medical record review
			c. AHRQ	c. NEDS—ED visits, emergent hospitalizations, diagnosed anticoagulant-related bleeding or coagulation abnormality from administrative claims
		7. Incidence of bleeding events and coagulation abnormalities in Federal Health System Populations	a. DOD	a. Pharmacovigilance Defense Application System—Outpatient clinic visits, ED visits, ICD-9/CPT codes
			b. VA	b. Integrated Databases—ADE identification by ICD-9 codes, primary hospitalizations, ED or clinic visits, and laboratory values
			c. IHS	c. RPMS EHR
	Local/facility-level	8. Quality Improvement Process Measures	CMS	a. Average percentage of monthly intervals in which INR test not received (NQF 0555)
				b. Percentage of episodes of patient receiving concomitant antibiotic therapy with INR test performed (NQF 0556)
				c. Average percentage of time patients have INR test results within therapeutic range (ie, TTR) (CMS EHR Incentive Program)

Abbreviations: ADE, adverse drug event; AHRQ, Agency for Healthcare Research and Quality; CDC, Centers for Disease Control and Prevention; CMS, Centers for Medicare & Medicaid Services; CPT, Current Procedural Terminology; DOD, US Department of Defense; E-codes, External Cause-of-Injury Codes; ED, emergency department; EHR, electronic health record; FAERS, FDA Adverse Event Reporting System; FDA, US Food and Drug Administration; HCUP, Healthcare Cost and Utilization Project; IHS, Indian Health Service; ICD, International Classification of Diseases; INR, international normalized ratio; MPSMS, Medicare Patient Safety Monitoring System; NEDS, National Electronic Disease Surveillance; NEISS-CADES, National Electronic Injury Surveillance System-Cooperative Adverse Drug Event Surveillance Project; NQF, National Quality Forum; ONC, Office of the National Coordinator for Health Information Technology; TJC, The Joint Commission; TTR, time in therapeutic range; VA, US Department of Veterans Affairs; VA-ADERS, VA Adverse Drug Event Reporting System; VTE, venous thromboembolism.

Anita Thomas, PharmD (Center for Clinical Standards and Quality, Centers for Medicare & Medicaid Services [CMS])

Dr Thomas provided additional background on strategies for the prevention of anticoagulant-related ADEs. The ADE Action Plan highlights improved communication, patient engagement, and care coordination as a key strategy for anticoagulant-related ADE prevention. Medication reconciliation and appropriate simplification of medication regimens are also areas for improvement. She further emphasized that although evidence-based practices are optimal, it is also critical to engage communities to identify local best practices.

Dr Thomas then highlighted the role of the CMS Quality Improvement Organization (QIO) in decreasing ADEs. The QIO is an integral part of national quality strategies to improve care for individuals, improve health for populations, and reduce health care costs. Within the QIO, the Quality Innovation Network champions local-level, results-oriented change while facilitating learning and action networks as well as providing technical assistance for ADE prevention.

Scott Kaatz, DO, MSc, FACP (Chief Quality Officer, Hurly Medical Center, Michigan State University, College of Human Medicine)

Dr Kaatz provided a clinical perspective on opportunities for improvement in anticoagulation use. He recognized that fiscal constraints can limit the ability of health care systems to implement inpatient anticoagulation services, and that the lack of reimbursement for outpatient anticoagulation clinics limits their widespread utilization. Dr Kaatz emphasized, however, that large integrated health care systems can successfully implement systems-based strategies to manage anticoagulation, as exemplified by the US Department of Veterans Affairs (VA). Resources for a systems-based approach to improving anticoagulation safety are readily available through the Anticoagulation Forum’s Anticoagulation Centers of Excellence, the National Certification Board for Anticoagulation Providers, and the Indian Health Service’s 3-day training program, among other training programs.

Dr Kaatz concluded the anticoagulation session with an emphasis on the proposed incentives and oversight opportunities for preventing anticoagulant-related ADEs across health care settings. Within the inpatient environment, quality incentives robust enough to effect change are needed. Moving forward, the outpatient side will require modifications in the reimbursement structure for anticoagulation clinic care. Furthermore, there must be a mechanism for anticoagulation clinics to manage extended care.

Afternoon Breakout Session Summary: Preventing Anticoagulant ADEs and Monitoring Progress

Dan Budnitz, MD, MPH (Director, Medication Safety Program, Division of Healthcare Quality Promotion, CDC; Captain [CAPT], US Public Health Service) provided an overview of the key themes emerging in the anticoagulant-related ADE prevention breakout session where the potential measures for tracking national progress (Table 1) were discussed among conference participants. CAPT Budnitz noted that collecting information about ADEs through national measures is important to generate action on a national level, but comprehensive information on patient populations across health care settings probably cannot be based on a single system. Using administrative claims codes to measure anticoagulant-related ADEs can be problematic and may require additional examination and validation. Similarly, facility measures useful for local quality improvement may require modifications in order to scale to the national level. CAPT Budnitz emphasized that in considering ways to improve anticoagulation management and reduce these ADEs, the benefits of well-managed anticoagulation should not be overlooked.

Morning Plenary Session 2: Preventing Diabetes Agents ADEs and Monitoring Progress

Leonard Pogach, MD, MPH, FACP (National Director of Medicine, Office of Specialty Care/Office of Patient Care Services, VA)

Dr Pogach provided an overview of serious hypoglycemic events in outpatient settings and highlighted the contribution of hypoglycemia to the burden of ADEs. Insulin is the most common drug implicated in ADE-related ED visits, ⁶ and insulin and oral diabetes agents are implicated in 25% of emergent hospitalizations for ADEs in older adults. ⁷

Opportunities for advancing the prevention of hypoglycemia include safer care, patient and family engagement, effective communication and coordination of care, and science-driven prevention and treatment. Dr Pogach described the VA’s leadership role in preventing hypoglycemia and emphasized strategies such as individualized glycemic targets and shared decision making, as well as universal health literacy and health numeracy.

Dr Pogach concluded his portion of the presentation with a discussion of health information technology including recommendations for more robust electronic health records (EHRs) that link pharmacy and laboratory data, include data elements that record comorbid conditions, and are able to create patient lists stratified by specific lab values and certain risk factors.

Mary Andrawis, PharmD, MPH (Senior Advisor, Innovation Center, CMS)

Dr Andrawis gave an overview of the impact of diabetes agents–related ADEs and then focused on the prevention of serious hypoglycemic events in inpatient settings. She noted that challenges to inpatient glycemic control include the lack of a standard definition for serious, severe, or clinically significant hypoglycemia; conflicting evidence regarding ideal glycemic targets; a lack of systematic identification of patients at risk for hypoglycemia; and barriers to multidisciplinary coordination.

Dr Andrawis discussed 2 significant challenges that underscore the need to systematically identify inpatients at risk for hypoglycemia: (1) failure to adjust insulin/diabetes regimens in response to decreases in oral intake such as an unexpected interruption of tube feedings or other sources of nutrition and (2) failure to respond appropriately to an initial hypoglycemic event. To address these challenges and help prevent hypoglycemia in the inpatient setting, Dr Andrawis recommended adopting basal/bolus insulin protocols instead of sliding-scale insulin, instituting a nurse-driven protocol for hypoglycemia management, and coordinating blood glucose testing with meals and insulin administration.

Andrew I. Geller, MD (Medical Officer, Division of Healthcare Quality Promotion, CDC; Lieutenant Commander (LCDR), US Public Health Service)

LCDR Geller highlighted key considerations in surveillance, such as selecting data sources and metrics for monitoring ADEs, and the relative importance of quantifying known ADEs rather than signal detection of rare or unrecognized ADEs. Accurate quantification requires active surveillance as opposed to relying on passive or voluntary reporting systems, and when possible, actual patient harms should be measured rather than surrogate markers.

LCDR Geller focused on data sources that can be leveraged to monitor diabetes agents–related ADEs in both the inpatient and ambulatory settings. The discussion included an overview of sources such as the Agency for Healthcare Research and Quality’s MPSMS, which has been used to measure the percentage of patients administered hypoglycemic agents with an episode of hypoglycemia requiring intervention in the inpatient setting. For ambulatory patients, the National Health Interview Survey, a household survey that quantifies the number of Americans self-reporting a diabetes diagnosis and treatment with insulin or oral diabetes agents, can be leveraged to derive denominator data.

LCDR Geller then reviewed several of the existing measure concepts for hypoglycemia (Table 2). For ambulatory patients, Healthy People 2020’s Medical Product Safety (MPS) topic area currently has an objective (MPS-5.2) to reduce ED visits for overdoses from injectable antidiabetic agents. For hospitalized patients, NQF Endocrine Measure 2363 looks at the rate of severe hypoglycemic events following the administration of an antidiabetic agent. LCDR Geller’s discussion of data sources and metrics laid the groundwork for the afternoon breakout sessions.

OPEN IN VIEWER

Table 2.

Inventory of Potential Measures for Identifying Diabetes Agents ADEs.

Diabetes Agents Outcome and Process Measures
Scope		Measure/Data Elements	Agency	Data System/Brief Description
Inpatient	National	1. Incidence of hypoglycemic events	AHRQ	a. HCUP—Inpatient stays with ICD-9 codes and E-codes from administrative claims and/or EHR data
				b. MPSMS—Inpatient stays with combination of laboratory triggers and clinical triggers from medical record review
		2. Incidence of hypoglycemic events in Federal Health System Populations	a. IHS	a. Resource and Patient Management System (RPMS-EHR)—EHR entry in the Problem List of “hypoglycemia”
			b. VA	b. Integrated Databases—ADE identification by ICD-9 codes, primary hospitalizations, ED or clinic visits, and laboratory values
	Local/facility-level	3. a Rate of hypoglycemic events following administration of an antidiabetic agent (NQF 2363)	ONC	a. Proposed Meaningful Use Measure Concept
				Severe hypoglycemia (<40 mg/dL)
				Proposed data source: Electronic clinical data-EHR, laboratory, pharmacy
		4. Rate of all-cause unplanned readmission (NQF 1789)	CMS	a. MedPAR—Medicare Part A Claims for all-cause readmissions after admission for any eligible condition (including diabetes) within 30 days of hospital discharge
		5. Admission rates for diabetes short-term complications (NQF 0272)	AHRQ	a. HCUP State Inpatient Database—Discharges for short-term diabetes complications (ketoacidosis, hyperosmolarity, coma)
		6. a Mild hypoglycemia	ONC	a. Proposed Meaningful Use Measure Concept
		7. a Recurrent hypoglycemia	ONC	a. Proposed Meaningful Use Measure Concept
		8. a Documentation of etiology of hypoglycemic event	ONC	Proposed Meaningful Use Measure Concept—Clinical decision support
Inpatients and outpatients		Spontaneous reports
		9. Rates of clinician-diagnosed or patient-reported hypoglycemia	a. DOD	a. Patient Safety Reporting System—Passive reporting
			b. FDA	b. FAERS—Passive reporting
			c. VA	c. VA-ADERS—Passive reporting
Outpatients	National	10. Incidence of hypoglycemic events	a. FDA	a. Sentinel, Mini-Sentinel—ED visits, hospitalizations for hypoglycemic events from administrative claims/EHR
			b. CDC	b. NEISS-CADES—ED visits, emergent hospitalizations, diagnosed hypoglycemic events from medical record review
			c. AHRQ	c. NEDS—ED visits with hypoglycemia as first listed diagnosis
		11. Incidence of hypoglycemic events in Federal Health System Populations	a. DOD	a. Pharmacovigilance Defense Application System—Outpatient clinic visits, ED visits, ICD-9/CPT codes
			b. VA	b. Integrated Databases—ADE identification by ICD-9 codes, primary hospitalizations, ED or clinic visits, and laboratory values
			c. IHS	c. RPMS—EHR entry in the Problem List of “hypoglycemia”
	Local/facility-level	12. Rates of patients with HbA1c poor control	NCQA	a. HEDIS—Data from HMOs and PPOs. Poor control defined as HbA1c >9%
		13. Rates of patients with HbA1c control	NCQA	a. HEDIS—Data from HMOs and PPOs. Control defined as HbA1c <8% or <7% for subset younger than 65 years old
		14. a Percentage of patients ≥65 years with diabetes, out-of-range HbA1c, and chronic comorbidities	ONC	a. Proposed Meaningful Use Measure Concept (Stage 3)
				b. Overtreatment Concept—Comorbidities include current substance abuse, advanced microvascular complications, cognitive impairment, and limited life expectancy

Abbreviations: ADE, adverse drug event; AHRQ, Agency for Healthcare Research and Quality; CDC, Centers for Disease Control and Prevention; CMS, Centers for Medicare & Medicaid Services; CPT, Current Procedural Terminology; DOD, Department of Defense; E-Codes, External Cause-of-Injury Codes; ED, emergency department; EHR, electronic health record; FAERS, FDA Adverse Event Reporting System; FDA, Food and Drug Administration; HCUP, Healthcare Cost and Utilization Project; HEDIS, Healthcare Effectiveness Data and Information Set; HMO, health maintenance organization; ICD, International Classification of Diseases; IHS, Indian Health Service; MedPAR, Medicare Provider Analysis and Review; MPSMS, Medicare Patient Safety Monitoring System; NCQA, National Committee for Quality Assurance; NEDS, Nationwide Emergency Department Sample; NEISS-CADES, National Electronic Injury Surveillance System-Cooperative Adverse Drug Event Surveillance Project; NQF, National Quality Forum; ONC, Office of the National Coordinator for Health Information Technology; PPO, preferred provider organization; VA, US Department of Veterans Affairs; VA-ADERS, VA Adverse Drug Event Reporting System.

a

Not currently in use.

Afternoon Breakout Session Summary: Preventing Diabetes Agents ADEs and Monitoring Progress

Dr Pogach summarized key themes discussed in the diabetes agents–related ADE measures breakout session where the potential measures for tracking national progress in diabetes agents ADE prevention were discussed (Table 2). First, the breakout participants agreed that it was important to meet meaningful use criteria in order to improve and monitor patient care; however, it is essential to carefully consider unintended consequences of these systems. Dr Pogach also emphasized the need for a consistent definition of hypoglycemia and consistent glucose thresholds within the hospital systems. Participants also discussed general issues associated with patient reporting of hypoglycemia. For example, when reporting about a hypoglycemic event, additional qualitative details about the impact of the event on the patient’s quality of life can provide a wealth of information. There are not easy mechanisms for incorporating this narrative into the EHR, however. Furthermore, it may be necessary to stratify populations in order to address the specific needs of patients based on their age or comorbidities.

Morning Plenary Session 3: Preventing Opioid ADEs and Monitoring Progress

Robert Kerns, PhD (Director, Pain Research, Informatics, Multimorbidities and Education (PRIME) Center; Special Advisor for Pain Research, Veterans Health Administration)

Dr Kerns introduced and highlighted several important themes present in the ADE Action Plan regarding both the problem posed by the need for opioids to treat patient pain and subsequent possible ADEs and their prevention. For example, the available literature examining opioids versus placebo or nonopioids for chronic low-back pain is scarce and does not provide clear support for opioid use. Opioids can be essential tools to manage pain in both the acute and palliative care settings, but the widespread and growing controversy regarding opioid therapy for chronic pain needs to be addressed. The ADE Action Plan should be considered within the context of published guidelines for the management of acute postoperative pain, pain in the palliative care setting, and management of chronic noncancer pain—calling for judicious use of opioids in appropriately selected and monitored patients. Dr Kerns reiterated that the ADE Action Plan focuses only on the therapeutic use of opioids, not illicit and/or recreational use, medication withdrawal, intentional harm, and medication nonadherence.

Christopher M. Jones, PharmD, MPH (Senior Advisor, Office of Public Health Strategy and Analysis, Food and Drug Administration; Commander [CDR] US Public Health Service)

CDR Jones reviewed surveillance methods outlined in the opioids section of the ADE Action Plan, reemphasizing that disentangling use and abuse remains a challenge. There are a variety of opioid-related adverse outcomes to track, including clinical (primary) outcomes such as ED visits or deaths and intermediate (surrogate) outcomes such as clinical or laboratory values, (eg, toxicity screens) that precede or lead to clinical outcomes. The section also provides process measures that are indicators of actions aimed at mitigating risk for primary or surrogate outcomes. Recently, there have been advances in surveillance capabilities including more specific diagnostic and procedural coding, improved process measures across inpatient and outpatient settings, EHR integration, and Prescription Drug Monitoring Programs (PDMPs).

CDR Jones emphasized surveillance as the cornerstone of understanding the problem of opioid-related ADEs and of tracking progress to improve opioid safety. Current systems are not optimal for opioid-related ADE surveillance, but opportunities exist to develop clinical outcome and process measures, standardize definitions for opioid-related ADEs, create new agency collaborations, and conduct research to validate candidate metrics. First, federal agencies should explore ways to improve uptake of evidence-based strategies for safe opioid prescribing through the increased use of opioid-prescribing guidelines and through the development of appropriate provider training programs. Second, federal agencies should promote patient-centered, multimodal, team-based care to personalize pain management; properly manage patients with high-risk medical and mental comorbidities; and intensively manage patients at high risk for opioid overdose. Third, federal agencies should develop and encourage the use of patient education materials and tools. Such materials should be developed using health literacy principles and engage the patient to use opioids safely. Finally, federal agencies can leverage and learn from other federal partners’ individual initiatives related to patient care.

Trinka Coster, MD (DA Director, Pharmacovigilance Program, Department of Defense [DoD]; Colonel [COL]), US Army)

COL Coster gave an overview of the incentives and oversight involved in preventing opioid-related ADEs. She emphasized drug utilization reviews with administrative and/or point-of-sale data for better oversight with flags on the accounts of possible at-risk patients. The sharing of such data among pharmacies and physician practices also can assist in preventing unsafe opioid prescribing and dispensing.

In the inpatient setting, health care quality reporting measures can be expanded to include multidisciplinary models of pain management and metrics can be validated for the monitoring of patient-controlled analgesia. Meanwhile, outpatient strategies for improved opioid safety include defining metrics for process measures that identify high-risk patients; addressing financial barriers to using evidence-based prevention strategies; using multidisciplinary teams and multimodal treatments for pain management, especially in transitions of care; and developing metrics for identifying high-risk patients and high-risk prescribers using administrative data and PDMPs that lead to abuse or misuse.

Deborah Perfetto, PharmD, MS (Health Science Administrator, Center for Quality Improvement and Patient Safety, Agency for Healthcare Research and Quality)

Dr Perfetto concluded the plenary session on opioids by focusing on some of the specific measures and metrics that CDR Jones summarized earlier. Currently, data are collected on national opioid-related ADE outcomes measures by a number of systems that capture inpatient and outpatient events, ED visits, and prescribing patterns (Table 3).

OPEN IN VIEWER

Table 3.

Inventory of Potential Measures for Identifying Opioid ADEs.

Opioid Outcome and Process Measures
Measure or Data Element and Agency	Data System and Description
1. National Opioid ADE Incidence/Rate (AHRQ)	a. HCUP—Inpatient stays with ICD-9 codes (Inpatient)
	b. MPSMS—Opioids are not currently captured by MPSMS system, but will be included after conversion to QSRS (Inpatient)
	c. NEDS—ED visits with ICD-9 codes (Outpatient)
2. National Opioid ADE Incidence/Rate (CDC)	a. NEISS-CADES—ED visits for opioid overdose, and other ADEs, not related to misuse/abuse (Outpatient)
	b. NVSS-Mortality—Deaths due to overdose—ICD 10 (Outpatient)
3. National Opioid ADE Incidence/Rate (SAMSHA)	a. DAWN—ED visits for opioid overdose related to misuse/abuse—transitioning to new system (Outpatient)
4. FDA Reports on Opioid ADE	a. Spontaneous Reports: Clinician-diagnosed or patient-reported adverse drug event (Outpatient)
	b. FDA Adverse Event Reporting System (FAERS) (Outpatient)
	c. Mini-Sentinel—Health-related electronic data to monitor for safety signals (Outpatient)
5. National Opioid ADE Incidence/Rate(CMS)	a. Part D Claims Data—Prescribing to detect fraud and abuse (Outpatient)
6. National Opioid ADE Incidence/Rate (DEA)	a. ARCOS—Total number/amount of opioids sold by manufacturers/distributors to retail registrants as an estimate of exposure/supply (Outpatient)
7. Regional/Facility-level Opioid ADE Incidence/Rates (DOD)	a. Outpatient clinic visits, ED visits, hospitalizations with ICD-9 codes and/or CPT codes (Outpatient)
8. Regional/Facility-level Opioid ADE Incidence/Rates (VA)	a. ATHENA—Opioid process measures (Outpatient)
	b. Outpatient clinic visits, ED visits, hospitalizations for opioid overdoses, other relevant ADEs (Outpatient)
9. Regional/Facility-level Opioid ADE Incidence/Rates (State)	a. Prescription Drug Monitoring Program (PDMP)—Number of opioids prescribed linked with patient and prescriber; number of patients with multiple opioid prescribers; number of patients on high daily dose of opioids (Outpatient)
	b. Prescription Behavioral Surveillance System(PBSS): Collects de-identified data from multiple state PDMPs (CDC, FDA, BJA, Brandeis CoE)
10. Proposed Meaningful Use Criteria—Outpatient	a. High daily dose long-term opioids
	b. Co-prescribed long-term opioids and depressants
	c. Toxicology screen for long-term therapy
	d. PDMP check prior to and yearly for long-term therapy
	e. Written opioid care management plan
	f. Mental health assessment for long-term therapy
	g. Number of patients prescribed opioids in facility or practice
11. Proposed Meaningful Use Criteria—Inpatient and Clinical Decision Support	a. Monitoring of patients on patient-controlled analgesia opioid therapy
	b. Prescribing high dose opioids in opioid-naïve patients
	c. Clinical decision support rules to support measures concepts

Abbreviations: ADE, adverse drug event; AHRQ, Agency for Healthcare Research and Quality; ARCOS, Automation of Reports and Consolidated Orders System; BJA, Bureau of Justice Assistance; CDC, Centers for Disease Control and Prevention; CMS, Centers for Medicare & Medicaid Services; CPT, Current Procedural Terminology; DAWN, Drug Abuse Warning Network; DEA, Drug Enforcement Administration; DOD, Department of Defense; ED, emergency department; FDA, Food and Drug Administration; HCUP, Healthcare Cost and Utilization Project; ICD, International Classification of Diseases; MPSMS, Medicare Patient Safety Monitoring System; NEDS, Nationwide Emergency Department Sample; NEISS-CADES, National Electronic Injury Surveillance System-Cooperative Adverse Drug Event Surveillance Project; NCHS, National Center for Health Statistics; NVSS, National Vital Statistics System; QSRS, Quality and Safety Review System; SAMSHA, Substance Abuse and Mental Health Services Administration; VA, US Department of Veterans Affairs.

The Automation of Reports and Consolidated Orders System (ARCOS) provides current process measures on the national ADE incidence rate. The ARCOS system is an automated, comprehensive drug reporting system that monitors the flow of US Drug Enforcement Administration controlled substances from their point of manufacture through commercial distribution channels to point of sale or distribution at the dispensing/retail level.

In addition to outcome and process measures, there are a number of quality improvement measures in place to identify regional- or facility-level opioid-related ADE incidence rates. Organizations with key quality improvement measures include, but are not limited to, the DOD, the VA, PDMPs, and patient safety organizations. For example, PDMPs serve as networks to track patients with high daily doses of opioids, multiple prescribers, and connecting patients and their appropriate prescriber. One other group of measures discussed in detail included the Proposed Meaningful Use Criteria for both inpatient and outpatient safe opioid use. The purpose of these measures is to also act as a safety checkpoint with actions such as identifying opioid naïve patients or patients prescribed both opioids and central nervous system depressants such as benzodiazepines. These measures served as various focal points for discussion during the opioid breakout sessions in the afternoon.

Afternoon Breakout Session Summary: Preventing Opioid ADEs and Monitoring Progress

Dr Kerns provided highlights of key themes from the afternoon opioid-related ADE prevention breakout session where the potential measures for tracking national progress in opioid ADE prevention were discussed (Table 3). He noted that it would be essential to consider the ADE Action Plan within the context of the national imperative to improve pain care. The prevention of opioid-related ADEs will require patient-centered care in order to avoid the magnification of existing pain management stigma and disparities. The breakout participants agreed that attention to improved operationalization of existing strategies is necessary. For example, the urine drug screen depends on setting, patient population, context, and sensitivity of assay. Additionally, participants suggested the potential benefit of shortening the 90-day surveillance period in order to analyze the optimal follow-up threshold. The group expressed a similar concern about dosing thresholds and recommended the need for further research.

Closing Remarks

As part of closing remarks, CAPT Hu thanked participants for their attendance and thoughtful contributions to the important work of ADE prevention. He suggested that the diversity of backgrounds and expertise of participants greatly benefited the conference sessions. He commented that although a tremendous amount of work has been done to date, there is still a lot to be done. Over the coming year, the Federal Interagency Steering Committee will reconvene in order to discuss the potential measures for tracking national progress in ADE prevention. Ongoing federal efforts will target the 4 focus areas of the ADE Action Plan in order to improve surveillance, evidence-based prevention strategies, incentives and oversight, and further research.

Conclusions

In summary, the ADE Prevention: 2014 Action Plan Conference highlighted important ADE prevention strategies for anticoagulants, diabetes agents, and opioids as a national patient safety priority. The expert panel of plenary speakers illustrated broad-based federal support for this initiative. Conference participation from health care providers, advocacy groups, policy analysts, and industry representatives, among others, demonstrated substantial private sector interest in engaging with the challenge of ADE prevention. Echoed throughout the morning plenaries and afternoon breakout sessions was the sentiment that ongoing collaboration and innovative solutions will be needed in order to protect patients from ADE-related harms. The discussions held in the afternoon breakout sessions will serve as a basis for the reconvening of the Federal Interagency Workgroup for ADE Prevention. Under their guidance, and with input from public stakeholders and subject matter experts, national benchmarks will be established to measure progress in ADE prevention. Although a multitude of efforts, including those highlighted in this conference, continue toward this end on a day-to-day basis, the establishment of national measures and targets for reducing ADEs will help align efforts across the public and private sectors and coordinate targeted initiatives toward ADE prevention.