Summary

Deaths associated with the co-prescribing of opioid and benzodiazepine pharmacotherapy have raised concerns among health care providers, federal agencies, and the public. The U.S. Congress has expressed concerns particularly about this issue in the veteran population. Veterans are more likely than non-veterans to experience pain, trauma, and mental health conditions from training and combat-related service and be prescribed pharmacotherapies for treatment. The U.S. Congress mandated that the Department of Veterans Affairs (VA) request an ad hoc committee of the National Academies of Sciences, Engineering, and Medicine (the National Academies) be assembled to evaluate the effects of prescribed opioid and benzodiazepine pharmacotherapy on all-cause mortality of U.S. veterans, including suicide, focusing on practices between 2007–2019.

Congress’s concerns drew attention to two concurrent public health issues: (1) the high prevalence of pain with evidence of less-than-optimal pain management, and (2) high rates of excessive prescription and nonprescription opioid use, misuse, and opioid use disorder (OUD) and related harms, including overdose and mortality. The opioid epidemic in the United States developed over many decades as a result of complex, interconnected factors and therapeutic decisions made by clinicians, health systems, regulators, and professional organizations in an effort to address growing concerns about unrelieved pain. This report does not explore the full history as to how, and why, the opioid epidemic, including concomitant prescribing with benzodiazepine pharmacotherapy, came to be, which has been discussed at length in prior National Academies reports. This report provides a brief summary of the opioid epidemic in the context of the committee’s study period of interest (2007–2019) and then focuses on studies (target trial emulations [TTEs]) the committee conducted in response to the statement of task (risks emerging from co-prescribing opioid and benzodiazepine pharmacotherapies).

In particular, the committee evaluated several dimensions of the effect of newly dispensed^1,2 opioid and benzodiazepine pharmacotherapies³ in the Veterans Health Administration (VHA) on all-cause mortality and suicide.

___________________

¹ The committee noted differences in types of pharmacy data measures. Pharmacy data can be categorized into three measures: prescribed (prescriber submits a prescription to a pharmacy), filled (pharmacy completed the requested prescription), and dispensed (individual picks up/is mailed the prescription from the pharmacy). The committee used dispensed pharmacy data in its analyses.

² The committee operationalized the term “initiation” in the statement of task as “newly dispensed” pharmacotherapy in analyses. In the committee’s studies, “newly dispensed” was defined as pharmacotherapy of interest not previously dispensed (or initiation of pharmacotherapy) to the individual in the last 90 days before the start date.

³ The committee used the term “pharmacotherapy” throughout the report instead of “medications,” “prescriptions,” or “treatment,” especially when referencing the committee’s analyses and results.

This report examines (1) newly dispensed opioid pharmacotherapy in veterans without and with current benzodiazepine pharmacotherapy, (2) varying levels of the initial baseline dosage and dosage escalation of dispensed opioid pharmacotherapy, and (3) newly dispensed benzodiazepine compared to alternative non-benzodiazepine pharmacotherapy in veterans with consistent opioid pharmacotherapy. The committee’s approach, results of the studies, and findings in response to the statement of task are presented next.

CONTEXT OF THE OPIOID EPIDEMIC

Multiple factors contribute to pain as a public health challenge, as outlined in the 2011 Institute of Medicine report Relieving Pain in America. These include the large number of people impacted, the disparities in pain prevalence and treatment, differences in the burden on different subgroups, the large indirect and direct economic impact on individuals and society, and the consequential impact of pain treatments. Pain has been linked to higher rates of morbidity, mortality, suicide, disability, and co-occurring psychiatric disorders. Furthermore, treatment of pain is complex, given the heterogeneity of individuals experiencing pain, the need for multimodal approaches to therapy, a lack of understanding of why individuals transition from acute to chronic pain, and the overlapping effects of pain with depression, anxiety, and other psychosocial conditions.

Prescribed Opioid Pharmacotherapy Helps Manage Pain, But Can Contribute to Adverse Outcomes, Including Mortality

In the mid-1990s, the pharmaceutical industry, professional societies, and patient advocacy groups argued that there was an epidemic of untreated pain, and wider use of opioid pharmacotherapy was encouraged. Although opioid pharmacotherapies are prescribed to control pain, they are known to produce feelings of euphoria and sedation, which may contribute to opioid misuse and increased risk of overdose and other drug-related problems. Rapid increases in opioid dosage can inhibit respiration and even lead to death. Higher amounts of opioids are associated with increased risk of OUD,⁴ overdose, mortality, and suicide. Increases in opioid prescribing coincided with increasing opioid-related mortality in the United States from the mid-1990s to around 2010. In addition to overdose deaths, studies suggest that the increasing use of prescription opioid pharmacotherapy to manage pain has also correlated with an increased risk of suicide. These trends have been compounded by the increased use of illicit opioids, which are now the main drivers of drug overdose deaths.

The co-prescribing of benzodiazepine and opioid pharmacotherapies has raised additional concerns. Whereas opioid pharmacotherapy is primarily prescribed for pain, benzodiazepine pharmacotherapy is typically prescribed for anxiety, panic disorders, insomnia, and seizures. The combination of opioid and benzodiazepine pharmacotherapies is known to have potential adverse respiratory effects, as the sedating effects of benzodiazepines can further suppress breathing in individuals experiencing opioid-induced respiratory depression. Data on causes of death in overdose fatalities published by the National Institute of Drug Abuse indicate that the percent of drug overdose deaths with concomitant benzodiazepine pharmacotherapy rose dramatically in the initial period of the opioid epidemic from 1999 to 2010 and has been falling slowly since. Although several studies examine the risk of concurrent opioid and benzodiazepine pharmacotherapy, the level of risk and the relationship of these deaths to the concomitant prescribing of opioid and benzodiazepine pharmacotherapies compared to other pharmacotherapies (active comparator) have never been studied.

Pain and Opioid Prescribing in the Veteran Population

Studies have demonstrated that a higher percentage of veterans experience pain compared to non-veterans, which is attributable to the physical demands of military service, including training and combat-related injuries.

___________________

⁴ The Diagnostic and Statistical Manual of Mental Disorders—5 defines opioid use disorder (OUD) as “a problematic pattern of opioid use leading to clinically significant impairment or distress, as manifested by at least two of the [11 diagnostic criteria], occurring within a 12-month period.”

Veterans are also more likely to experience chronic health conditions, psychological distress, and a higher incidence of suicide than non-veterans (civilians). As a result, veterans are more likely than the general population to need care for chronic pain and other conditions.

About 9 million U.S. veterans receive health care services each year from the VHA, the largest integrated health care network in the country. VHA veterans (those who obtain care through the VHA) differ from non-VHA veterans (those who obtain health care from non-VHA health care providers), with a statistically significant higher prevalence of both physical and mental health conditions, such as posttraumatic stress disorder, anxiety, and depression even after adjusting for demographic characteristics. In addition, VHA veterans were more likely to report chronic pain than non-VHA veterans (45.2 percent versus 26.8 percent). In addition, among VHA veterans, the rate of OUD was 7 times higher than that of non-VHA veterans.

Similar to national trends, the prevalence of receiving an opioid prescription⁵ increased over time within the VHA, from about 19 percent of veterans in 2004 to 33 percent in 2012. Overall, opioid prescribing in the VHA declined thereafter. Between 2004–2009, 27 percent of VHA veterans who received opioid pharmacotherapy also received benzodiazepine pharmacotherapy. Of the drug overdose deaths observed during that period, nearly half occurred in VHA veterans concurrently prescribed benzodiazepines and opioid pharmacotherapies. Another study compared the risk of adverse outcomes (accidents and injuries, non-fatal overdose events, and death) in VHA veterans who received long-term opioid prescription⁶ (N = 924,656) to VHA veterans who received concomitant opioid and benzodiazepine pharmacotherapies (N = 110,305). VHA veterans with concomitant pharmacotherapies had a 36 percent higher risk of an adverse outcome (e.g., opioid-related overdose, self-inflicted injury, all-cause mortality).

CLINICAL GUIDELINES, POLICIES, AND LEGISLATION RELEVANT TO UNDERSTANDING THE CONTEXT OF THE OPIOID EPIDEMIC

In response to the adverse outcomes associated with the trends in prescribed opioid pharmacotherapy, several VHA clinical guidelines and policies addressing the opioid epidemic have been issued, including during the committee’s study period (2007–2019).⁷ The VHA set forth guidelines in 2003 to support clinical decision making around pain management and opioid treatment. VHA clinical guidelines continued to evolve through the years (2010, 2017, 2022) along with the agency-wide 2013 VHA Opioid Safety Initiative, which, among many efforts, increased education and monitoring and promoted effective prescribing. These efforts were complemented by other federal agency efforts, such as clinical guidelines from the Centers for Disease Control and Prevention (CDC) in 2016 and 2022 and Department of Health and Human Services in 2019. Federal legislation, such as the Veterans Access, Choice and Accountability Act (2014), the Comprehensive Addiction and Recovery Act (2016), and the VA Mission Act (2018), was passed during the study time frame, intended to help support veterans or reduce harms related to the opioid overdose crisis.

___________________

⁵ Opioid prescription in the cited study was based on the receipt of an opioid prescription, using VHA outpatient prescription fill.

⁶ Individuals receiving long-term opioid prescription were categorized as “chronic” or “high-risk.” “Chronic” was defined as receiving a 90-day or greater opioid prescription, with less than a 30-day gap in supply and within a 180-day period. “High-risk” was defined as those with ≥1 substance use disorder diagnosis (excluding tobacco) and/or an average opioid dose of ≥100 morphine milligram equivalent (MME)/day. Concomitant prescribing was defined as the receipt of both opioid and benzodiazepine prescriptions, with >7 days of overlap of both prescriptions.

⁷ In 1999, the VHA followed the American Pain Society recommendations, integrating the Pain as the 5th Vital Sign Initiative into the VHA National Pain Management Strategy, whereby routine screening, a published comprehensive pain assessment and management toolkit, and educational campaigns were launched.

CONGRESSIONAL MANDATE TO ADDRESS CONCERNS ABOUT PRESCRIBING OPIOIDS AND BENZODIAZEPINES IN THE VETERAN POPULATION

In 2019, through the Commander John Scott Hannon Veterans Mental Health Care Improvement Act of 2019,⁸ Congress mandated that the VA request an ad hoc committee of the National Academies be assembled “to evaluate the effects of opioids and benzodiazepine on all-cause mortality of veterans, including suicide, regardless of whether information relating to such deaths has been reported by the Centers for Disease Control and Prevention” (see Box S-1). The legislation stated that the request was “a revised study design to fulfill the goals of the 2019 study design of the National Academies.” The study designs developed by this current committee were based on the study designs, or target trial protocols, presented in the congressionally mandated 2019 National Academies report An Approach to Evaluate the Effects of Concomitant Prescribing Opioids on Veteran Deaths and Suicides.

The 2019 report applied the TTE framework in its proposed target trial protocols to emulate, using observational data, randomized controlled trials (RCTs). The 2019 report proposed two study designs (“opioid treatment initiation” and “opioid tapering”) to address the effect of prescribed opioid pharmacotherapy on all-cause mortality and suicide mortality. The 2019 report outlined specific study components to consider: eligibility criteria, treatment strategies, treatment assignment, start and end of follow-up, outcomes, causal contrasts, and statistical approach.

STUDY CHARGE AND APPROACH

In response to the congressional mandate, the National Academies convened a 13-member committee with expertise in pain, mental health, neurology, addiction, epidemiology, pharmacoepidemiology, statistics, VHA health data systems, and VHA and non-VHA care of mental health and pain in veterans. This committee leveraged use of the TTE framework mentioned in the 2019 National Academies report, as well as refined the study designs to respond to this statement of task. The committee developed causal research questions to align with the statement of task, identified comparator groups to align with “alternative non-opioid pain treatments” and “alternative treatments for anxiety and other indications for benzodiazepines,” and refined the protocols to use available observational electronic health record (EHR) data, data from VHA administrative and clinical databases, and mortality data from the National Death Index. Considering the available data, the committee operationalized the statement of task into the following causal questions, which are then addressed in three distinct studies:

___________________

⁸ Public Law 116-171 Commander John Scott Hannon Veterans Mental Health Care Improvement Act of 2019, 116th Congress, October 17, 2020. See also appendix C for the full language of Section 204 of the Act, specific to this study.

• Among veterans receiving care in the VHA who were not consistently⁹ dispensed pain pharmacotherapy between 2007–2019, what is the effect of newly dispensed¹⁰ opioid versus non-opioid pain pharmacotherapy on all-cause mortality (primary outcome) and suicide mortality (secondary outcome) among those without and with current use of benzodiazepine pharmacotherapy within a 12-month follow-up period? (Study 1)
• Among veterans receiving care in the VHA who were newly dispensed full agonist¹¹ opioid pharmacotherapy between 2007–2019, what is the effect of different initial opioid dosage and escalation strategies on all-cause mortality (primary outcome) and suicide mortality (secondary outcome) within a 12-month follow-up period? (Study 2)
• Among veterans receiving care in the VHA who were consistently¹² dispensed opioid pharmacotherapy between 2007–2019, what is the effect of newly dispensed benzodiazepine versus alternative non-benzodiazepine pharmacotherapy for anxiety and other common indications for benzodiazepines on all-cause mortality (primary outcome) and suicide mortality (secondary outcome) within a 3-month follow-up period? (Study 3)

For all studies, the committee did not include “chronic pain” in the causal question or as a covariate for several reasons. There are multiple factors that limit the reliability of pain-relevant diagnosis codes and their utility in observational research. First and foremost, there are challenges in its measurement and capture, such as potential underreporting of the large variety of diagnoses associated with chronic pain diagnoses and the underlying painful conditions. Preliminary analysis indicated that 37.7 percent of eligible participants who were dispensed an opioid pharmacotherapy did not have a diagnosis code related to chronic pain. Accurate diagnosis can be complicated and emerge over time as a comprehensive assessment proceeds (e.g., diagnosis of low back pain). Further, providers may vary in the reporting of pain-relevant diagnostic codes, which may further contribute to unreliability. Additionally, individuals with chronic pain report multiple sites of pain and multiple pain conditions, further complicating the use of diagnosis codes to characterize the pain experience and to reliably associate diagnosis with treatment decisions. Yet, opioid pharmacotherapy is used for a wide variety of pain conditions. Given the challenges in capturing pain and chronic pain, the committee assumed that most individuals who were dispensed an opioid prescription received it to manage pain (via opioids’ known analgesic effects). The indication of pain was used to choose the non-opioid pharmacotherapies included in study 1, to control for confounding by indication. Finally, the term is not reflected in the statement of task.

The committee noted that there are several aspects of opioid prescribing that were not stated in the statement of task and not reflected in the committee’s analyses. The committee did not implement the opioid tapering target trial protocol suggested in the 2019 report, considering the following factors in its decision. First, the current statement of task did not direct the committee to examine the effect of opioid tapering on mortality. Second, the committee was aware of studies based on the 2019 report protocol that had been or were being completed and commissioned a review of these and other relevant studies conducted since the 2019 report (see Appendix E). Based on the review, the committee determined that the risk of immortal time and selection biases is high and substantially limits the extent to which retrospective observational studies can infer causality between opioid tapering and mortality. Of note, Dr. Erin Krebs and her colleagues applied the 2019 protocol in an unpublished study (see Appendix F) included in the review. The committee noted, and based on the presentation by the author, that the study’s findings and conclusions were highly sensitive to different study design elements, statistical modeling, and measurement strategies. In addition, the treatment assignment was based on pharmacy records and as such

___________________

⁹ Not dispensed any qualifying opioid or non-opioid pain pharmacotherapy in the 90-day pre-index period.

¹⁰ The committee operationalized the term “initiation” in the statement of task as “newly dispensed” pharmacotherapy in analyses. In the committee’s studies, “newly dispensed” was defined as pharmacotherapy of interest not previously dispensed (or initiation of pharmacotherapy) to the individual in the last 90 days before the start date.

¹¹ There are two types of opioids: full agonist and partial opioid agonist. Full opioid agonists fully activate the mu receptors in the brain, enabling the opioid to have “full” effect, such as codeine, oxycodone, and fentanyl. Partial opioid agonists also activate mu receptors in the brain but to a lesser degree, such as butorphanol or tapentadol.

¹² The committee defined as three or more dispensed opioids at least 21 days apart within a 180-day period for at least 84-day supply or two dispensed opioids of 90-day supply within 180 days.

may not accurately reflect whether a tapering was intended or actually took place. Based on these observations, not available at the time of the 2019 report, the committee concluded that additional efforts to implement such a study is unlikely to yield more definitive results and, for these reasons, the committee chose not to implement the tapering trial suggested in the 2019 report. The committee focused the main report to reflect findings from the analyses addressing the statement of task.

The statement of task also did not direct the committee to compare mortality of veterans prescribed opioids and benzodiazepines in the VHA to veterans prescribed opioids and benzodiazepines in other health care systems. Thus, comparing VHA and non-VHA providers was not part of the committee’s work. In addition, the committee noted that the focus of the statement of task is to examine the risk, not the benefits, of prescribing opioid and benzodiazepine pharmacotherapies. Beyond non-opioid pain pharmacotherapy, the committee did not investigate the impact of other alternative treatments for pain, such as non-pharmacologic approaches as this was outside the scope of the statement of task.

Furthermore, the committee noted that the statement of task specifically mandated the study period (2007–2019) to understand the implications of opioid treatment on mortality in veterans. Therefore, the findings of this report are not intended to be applicable to current practices or behaviors, including those of the VHA. In addition, these retrospective studies are inherently forensic, reflecting the effect of care that occurred in the past rather than being indicative of the mortality risks of current practices, particularly given the ongoing changes in opioid prescribing practices, the policy landscape, mental health care, and the nature of the opioid overdose epidemic in the intervening years. As reflected in the statement of task and expert perspectives, many unanswered questions remain about the causal effects of past opioid prescribing practices on mortality, especially in light of ecologic data indicating that U.S. opioid- and benzodiazepine-pharmacotherapy–related mortality has continued to climb in recent years.

In sum, the committee sought to address the statement of task in the least biased and most internally valid way, which calls on the committee to utilize best practices and recent advances in the modern causal inference literature, including TTE techniques. The committee applied this strategy to estimate causal effects of several key opioid and benzodiazepine pharmacotherapy prescribing decision points—to initiate opioid pharmacotherapy, increase opioid dosages over time, and co-prescribe opioid and benzodiazepine pharmacotherapies. The committee was explicit in the causal objective, which can reduce ambiguity in the scientific question and errors in data analysis and clarify assumptions required for causal inference. The committee adhered to best practices from the modern causal inference literature, including (1) designing the observational analysis to emulate a target trial and (2) selecting confounding adjustment variables using expert knowledge and clear articulation of the causal structure of the study question.

ANALYTIC APPROACH

In line with the 2019 report, the current committee employed the TTE framework. The committee considered two designs: experimental or RCTs and non-experimental, or observational, studies. RCTs are experimental studies in which participants are randomly assigned, or randomized, to treatment strategies. Randomization minimizes confounding and assures balance across treatment groups. While RCTs¹³ are the gold standard for assessing the causal effects of medical treatments, barriers to implementation include prohibitively high costs and time demands, ethical dilemmas around the random allocation to medications with potentially negative side effects, potential for mistreatment of historically vulnerable populations, and potentially small sample sizes in the case of rare medical disorders or outcomes. In the last 3 decades, notable progress has been made in estimating the causal effects of medical treatment beyond the confines of RCTs, such as through observational study designs. In observational studies, the investigator does not randomize individuals into different treatment strategies, but rather observes to examine outcomes in each treatment strategy. With no randomization, there is a risk of confounding. The committee recognized other approaches to adjust for confounding (e.g., instrumental variables, regression discontinuity)

___________________

¹³ A key advantage of RCTs is protocol-based outcomes ascertainment, which is a limitation with observational studies. This ascertainment can help reduce bias due to misclassification of outcomes, exposure, and confounders.

in observational data, but the committee leveraged the TTE framework. The TTE framework is an advance in the modern causal inference literature to approach designing an observational study to mimic a hypothetical randomized experiment, or target trial. The TTE framework helps to ensure the observational study compares realistic clinical treatment strategies, avoids biases related to mishandling time zero, and improves upon conventional methods to address baseline and time-varying confounding. However, the impact of unmeasured confounding and immortal time bias still remains a risk in observational studies, unlike RCTs; sensitivity analyses and alternative assumptions about confounding should be taken into consideration.

Study Population, Data Sources, and Statistical Approach

The study population was defined as veterans receiving care in the VHA from January 1, 2007, to December 31, 2019.¹⁴ Data were drawn from national VHA data files and linked to the VA Corporate Data Warehouse, U.S. Veterans Eligibility Trends and Statistics, and Centers for Medicare & Medicaid Services (CMS) Medicare Parts A, B, C, and D data. The CMS Medicare data were used to supplement VHA data. The National Academies Institutional Review Board approved the study.

Each of the three studies was designed as a comparison between the exposure of interest (individuals receiving opioid pharmacotherapy, for example) to another exposure (individuals receiving non-opioid pain pharmacotherapy, for example). In identifying pharmacotherapy for either the treatment or comparator groups in each study, the committee relied on clinical expertise, Food and Drug Administration-approved indications, VHA drug classifications, and a review of existing literature. Pharmacotherapy data were drawn from either VHA EHR or Medicare Part D pharmacy claims data; medications selected for analyses were not limited to those on the CMS or VHA formularies. Use of an active comparator reduces the risk of confounding by the indication for therapy. However, that risk still exists, as different drugs are used differently and have different safety concerns. A perfect active comparator to opioid pharmacotherapy or benzodiazepine pharmacotherapy does not exist. However, the comparator groups selected are less biased than including individuals who were not an active comparator. Per the statement of task, all-cause mortality (primary outcome) and suicide mortality (secondary outcome) were the study outcomes.

The causal effects on the study outcomes were estimated using both per protocol effect and intent-to-treat effect (ITT) methodologies. The per protocol approach estimated the effect of being newly dispensed and continuously adhering¹⁵ to the treatment of interest during follow-up. The ITT approach characterized the effect of newly dispensed treatments from initiation to the end of the study, irrespective of whether an individual changed their treatment during follow-up (e.g., cross-over to another exposure group or treatment discontinuation). Unadjusted and adjusted pooled logistic regressions (studies 1 and 3) and the parametric G-formula (study 2) were the statistical methods, hazard ratios (study 1 and 3) and risk ratio (study 2) were estimated, and statistical inference is represented by the calculated confidence intervals. All analyses were conducted using SAS 9.4 and SAS Enterprise Guide. To strengthen causal inferences, the committee adjusted for many potential covariates, which included demographic characteristics (e.g., age, sex, race/ethnicity, marital status, disability status), supplemental insurance to VHA coverage (i.e., Medicare, TRICARE, or private insurance), housing security (e.g., history of homelessness status), physical and mental health conditions/disorders (e.g., acute painful injury, anxiety, chronic obstructive pulmonary disease, cardiovascular disease, substance use disorders, diabetes), military status (e.g., service era, military branch, combat service), body mass index, health care utilization in prior year (e.g., VHA outpatient visits, inpatient hospital admissions, nursing home stay), VHA facility-level characteristic (e.g., urbanicity), and specific pharmacotherapies that the committee considered as potential confounders. The committee considered all potential confounders, the validity of measures of covariates in the available datasets, and then sought parsimony in the final models to minimize imprecision introduced by including variables that were not confounders.

___________________

¹⁴ Latest possible index date was December 31, 2018, in studies 1 and 2 and October 1, 2019, in study 3.

¹⁵ Adherence to treatment of interest in the context of this report indicates that individuals continuously followed the treatment protocol of the specific study.

FINDINGS

The results and findings of the three studies are presented next. Each of the studies examined all-cause mortality and suicide mortality and focused on different aspects of opioid prescribing. Study 1 evaluates veterans who are newly dispensed opioid pharmacotherapy to those veterans newly dispensed non-opioid pain pharmacotherapy, without and with currently dispensed benzodiazepine pharmacotherapy. Study 2 evaluates two opioid prescribing strategies: varying initial dosages of opioid pharmacotherapy and varying opioid dosage escalation. Study 3 compares veterans who are newly dispensed benzodiazepine pharmacotherapy to those newly dispensed alternative non-benzodiazepine pharmacotherapy, among veterans who are consistently dispensed opioid pharmacotherapy.

Study 1: The Effect of Newly Dispensed Opioid Pharmacotherapy on All-Cause Mortality and Suicide Mortality, Among Veterans Without and With Currently Dispensed Benzodiazepine Pharmacotherapy

The first study addresses parts (a) and (d) of the statement of task. Specifically, the study focused on (a) the effect of opioid prescribing for pain compared to alternative non-opioid pain treatments, and (d) the effect of initiating opioids in individuals who are already taking benzodiazepines. The available data were stratified into two populations, based on current pharmacotherapy status: without¹⁶ (study 1a) and with¹⁷ (study 1b) current benzodiazepine pharmacotherapy.

Study 1a: The Effect of Newly Dispensed Opioid Pharmacotherapy on All-Cause Mortality and Suicide Mortality Among Veterans Without Current Benzodiazepine Pharmacotherapy

This study had 5,636,207 unique veterans, of whom 71 percent were newly dispensed non-opioid pain pharmacotherapy and 29 percent newly dispensed opioid pharmacotherapy. Study results are included in Table S-1.

Finding 1a-1. Among veterans without current benzodiazepine pharmacotherapy, the 12-month risk of all-cause mortality was 10,273 deaths per 100,000 person-years for those newly dispensed opioid pharmacotherapy and 6,473 deaths per 100,000 person-years for those newly dispensed non-opioid pain pharmacotherapy (aHR:¹⁸ 1.61; 95% CI: 1.59–1.63).

TABLE S-1 Study 1a Results. The Effect of Newly Dispensed Opioid Pharmacotherapy^† on All-Cause Mortality, Including Suicide Mortality, Among VHA Veterans Without Current Benzodiazepine Pharmacotherapy^†† (2007–2019; N = 5,636,207)

¹ Confidence Intervals

^† Defined as first occurrence of ≥7-day supply of a newly dispensed opioid or non-opioid pain pharmacotherapy. “Newly dispensed” is defined as having no exposure of opioid or non-opioid pain pharmacotherapy 90 days prior to the first occurrence.

^†† Without current benzodiazepine pharmacotherapy is defined as receiving <5 tablets of benzodiazepines within 90 days prior to the first occurence.

NOTE: No cells are reported representing 10 or fewer veterans; results are per protocol effect.

___________________

¹⁶ Defined as <5 tablets of benzodiazepine pharmacotherapy within 90 days pre-index period.

¹⁷ Defined as ≥ 5 tablets of benzodiazepine pharmacotherapy within 90 days pre-index period.

¹⁸ aHR: adjusted hazard ratio.

Finding 1a-2. Among veterans without current benzodiazepine pharmacotherapy, the 12-month risk for suicide mortality was 112 deaths per 100,000 person-years for those newly dispensed opioid pharmacotherapy and 95 deaths per 100,000 person-years for those newly dispensed non-opioid pain pharmacotherapy (aHR: 1.19; 95% CI 1.07–1.32).

Study 1b: The Effect of Newly Dispensed Opioid Pharmacotherapy on All-Cause Mortality and Suicide Mortality Among Veterans With Current Benzodiazepine Pharmacotherapy

This study had 319,990 unique veterans with current benzodiazepine pharmacotherapy, of whom 65 percent were newly dispensed non-opioid pain pharmacotherapy and 35 percent newly dispensed opioid pharmacotherapy. Study 1b results are included in Table S-2.

Finding 1b-1. Among veterans with current benzodiazepine pharmacotherapy, the 12-month risk for all-cause mortality was 11,877 deaths per 100,000 person-years for those newly dispensed opioid pharmacotherapy and 7,441 deaths per 100,000 person-years for those newly dispensed non-opioid pain pharmacotherapy (aHR: 1.59; 95% CI: 1.53–1.65).

Finding 1b-2. Among veterans with current benzodiazepine pharmacotherapy, the 12-month risk for suicide mortality was 246 deaths per 100,000 person-years for those newly dispensed opioid pharmacotherapy and 238 deaths per 100,000 person-years for those newly dispensed non-opioid pain pharmacotherapy (aHR: 1.03; 95% CI: 0.81–1.31).

Study 2: The Effect of Initial Opioid Dosage and Opioid Dosage Escalation Strategies on All-Cause Mortality and Suicide Mortality

The second study addresses part (b) of the statement of task, to examine the effects of higher doses relative to lower doses of opioids. The study emulated two trials: one examined the effect of different initial opioid dosage and the other, the effect of different opioid dosage escalation strategies.

Study 2a: The Effect of Initial Opioid Dosage Strategies on All-Cause Mortality and Suicide Mortality

Study 2a results are included in Table S-3.

TABLE S-2 Study 1b Results. The Effect of Newly Dispensed Opioid Pharmacotherapy^† on All-Cause Mortality and Suicide Mortality Among VHA Veterans with Current Benzodiazepine Pharmacotherapy^†† (2007–2019; N = 319,990)

¹ Confidence Intervals

^† Defined as first occurrence of ≥7-day supply of a newly dispensed opioid or non-opioid pain pharmacotherapy. “Newly” is defined as having no exposure of opioid or non-opioid pain pharmacotherapy 90 days prior to the first occurrence.

^†† With current benzodiazepine pharmacotherapy is defined as receiving ≥5 tablets of benzodiazepine pharmacotherapy within 90 day prior to the first occurrence.

NOTE: No cells are reported representing 10 or fewer veterans; results are per protocol effect.

TABLE S-3 Study 2a Results. Among VHA Veterans Newly Dispensed Full Agonist Opioid Pharmacotherapy, the Estimated Risk of All-Cause Mortality and Risk Ratio by Initial Opioid Dosage (2007–2019; N = 5,639,108)

¹ SE: standard error.

^a Initial dosage strategy is calculated as the average morphine milligram equivalent (MME)/day during the index month. The MME is a measure of the intensity of opioid exposure; it standardizes opioid prescriptions considering differences in the potency, quantity, and strength of different types of dispensed opioids. The committee calculated it using CDC conversions.

^b Confidence intervals (CI) were calculated using the percentile method from 400 bootstrap samples.

NOTE: In conducting the analysis, no cells are reported representing 10 or fewer veterans.

Finding 2a-1. Among veterans with newly dispensed opioid pharmacotherapy, the 12-month risk of all-cause mortality in each of the initial opioid dosage strategy was 3.95 percent for low, 4.89 percent for medium, and 5.80 percent for high. (RR¹⁹_{medium vs low}: :1.24; 95% CI: 1.20–1.28] and RR_{high vs low:} 1.47: 95% CI: 1.44–1.50]).

Finding 2a-2. Among veterans with newly dispensed opioid pharmacotherapy, the committee is unable to draw conclusions about the effect of those who received a high or medium initial daily opioid dosage level, compared to veterans with a low initial daily opioid dosage level on suicide mortality within 12-months of follow-up, due to the lower number of suicide deaths in the eligible population.

Study 2b: The Effect of Escalating Opioid Dosage Strategies on All-Cause Mortality and Suicide Mortality

Study 2b results are included in Table S-4.

Finding 2b-1. Among veterans with newly dispensed opioid pharmacotherapy, the 12-month risk of all-cause mortality in each of the dosage escalation strategies was 2.70 percent for stable, 4.56 percent for slow, and 4.69 percent for fast over the first six months. (RR_{slow vs stable}: 1.69; 95% CI: 1.53–1.80] and RR_{fast vs stable}: 1.74; 95% CI: 1.57–1.95]).

Finding 2b-2. The committee is unable to draw conclusions about the effect of escalating dosages of opioid pharmacotherapy on suicide mortality within 12 months of follow-up, due to the lower number of suicide deaths in the eligible population.

___________________

¹⁹ RR: risk ratio.

TABLE S-4 Study 2b Results. Among VHA Veterans Newly Dispensed Full Agonist Opioid Pharmacotherapy, the Estimated Risk Of All-Cause Mortality and Risk Ratio by Dosage Escalation Strategies (2007–2019; N = 5,639,108)

¹ SE: standard error.

^a Dosage escalation strategy is calculated as the average morphine milligram equivalent (MME)/day change in 6 months. The MME is ideal to measure the intensity of opioid exposure, as it standardizes opioid prescriptions considering differences in the potency, quantity, and strength of different types of dispensed opioids. The committee calculated it using CDC conversions.

^b Confidence intervals were calculated using the percentile method from 400 bootstrap samples.

NOTE: In conducting the analysis, no cells are reported representing 10 or fewer veterans.

Study 3: The Effect of Benzodiazepine Pharmacotherapy Co-Prescribing on All-Cause Mortality and Suicide Mortality Among VHA Veterans with Consistently Dispensed Opioid Pharmacotherapy²⁰

The third study addresses the objective outlined in part (c) of the statement of task. Specifically, the study focused on the effect of newly dispensed benzodiazepine pharmacotherapy in VHA veterans consistently dispensed opioid pharmacotherapy compared to VHA veterans who were consistently dispensed alternative non-benzodiazepine pharmacotherapy.

There were 637,793²¹ unique veterans who were consistently dispensed opioid pharmacotherapy and eligible for inclusion. About 19.4 percent of the sample were newly dispensed benzodiazepine pharmacotherapy, and about 88.6 percent were newly dispensed alternative non-benzodiazepine pharmacotherapy. Study 3 results are included in Table S-5.

Finding 3-1. Among veterans consistently dispensed opioid pharmacotherapy, the 3-month risk for all-cause mortality was 7,493 deaths per 100,000 person-years for those newly dispensed benzodiazepine pharmacotherapy and 4,392 per 100,000 person-years for those newly dispensed alternative non-benzodiazepine pharmacotherapy (aHR: 1.71; 95% CI: 1.60–1.82).²²

Finding 3-2. Among veterans consistently dispensed opioid pharmacotherapy, the 3-month risk for suicide mortality was 117 deaths per 100,000 person-years for those newly dispensed benzodiazepine pharmacotherapy and 83 deaths per 100,000 person-years for those newly dispensed alternative non-benzodiazepine pharmacotherapy (aHR: 1.42; 95% CI: 0.87–2.31).²³

CONCLUSIONS

The committee, through three studies, evaluated the effect of dispensed opioid and benzodiazepine pharmacotherapies on all-cause mortality and suicide mortality. The studies focused on effects of newly dispensed opioid or benzodiazepine pharmacotherapy, the co-prescribing of opioids and benzodiazepine pharmacotherapy, and different

___________________

²⁰ Consistently dispensed opioid pharmacotherapy is also known as “long-term opioid treatment.”

²¹ During the study period (2007–2019), a veteran could enter the study more than one time and contribute more than one episode. Thus, the number of veterans in the benzodiazepine group (N = 123,684) and alternative non-benzodiazepine comparator (N = 565,481) will not sum to 637,793.

²² Alternative non-benzodiazepine pharmacotherapy can be used to treat conditions treated by benzodiazepines, such as anxiety.

²³ Alternative non-benzodiazepine pharmacotherapies can be used to treat conditions treated by benzodiazepines, such as anxiety.

TABLE S-5 Study 3. The Effect of Benzodiazepine Pharmacotherapy Co-Prescribing^† on All-Cause Mortality, Including Suicide Mortality, Among VHA Veterans with Consistently Dispensed Opioid Pharmacotherapy^†† (2007–2019; N = 637,793)

^†≥1 tablet of benzodiazepine pharmacotherapy.

^†† Consistently dispensed opioid pharmacotherapy: Dispensed ≥3 opioid prescriptions within a 180-day window with (1) ≥two 21-day gaps between prescriptions and (2) an 84+ day supply within the 180-day window OR two 90-day opioid prescriptions within 180-day window.

^††† Alternative non-benzodiazepine pharmacotherapy includes pharmacotherapies used to manage anxiety or other common indications for benzodiazepine pharmacotherapy.

NOTE: In conducting the analysis, no cells are reported representing 10 or fewer veterans.

opioid dosage treatment strategies. The committee employed the TTE framework to reduce biases, such as due to immortal time, and confounding. In addition, the committee leveraged the large observational datasets (e.g., health care, clinical, and administrative data) available within the VA and VHA, supplemented by CMS data, and the National Death Index for mortality data. Based on the study findings, the committee concludes the following in veterans who received care from the VHA between 2007–2019:

Conclusion 1. The committee concludes that there is an increased risk of all-cause mortality among veterans newly dispensed opioid pharmacotherapy compared to those newly dispensed non-opioid pain pharmacotherapy.

Conclusion 2. The committee concludes that there is an increased risk of all-cause mortality among veterans co-prescribed opioid and benzodiazepine pharmacotherapies compared to those co-prescribed non-opioid pain and benzodiazepine pharmacotherapies or to those co-prescribed opioid and alternative non-benzodiazepine pharmacotherapies.

Conclusion 3. The committee concludes that there is an increased risk of all-cause mortality among veterans who initiated opioid treatment at higher dosage levels compared to lower levels.

Conclusion 4. The committee concludes that there is an increased risk of all-cause mortality among veterans whose treatment strategy was either slow or fast opioid dosage escalation compared to stable dosage.

Regarding suicide mortality, the results were less conclusive. Based on the study findings, the committee concludes the following in veterans who received care from the VHA between 2007–2019:

Conclusion 5. The committee concludes that there is an increased risk of suicide mortality among veterans newly dispensed opioid pharmacotherapy compared to those newly dispensed non-opioid pain pharmacotherapy.

Conclusion 6. The committee concludes there is no evidence of a difference in the risk of suicide mortality among veterans newly dispensed opioid pharmacotherapy and currently dispensed benzodiazepine pharmacotherapy compared to those newly dispensed non-opioid pain pharmacotherapy and currently dispensed benzodiazepine pharmacotherapy.

Conclusion 7. The committee concludes that there is some evidence of a higher risk of suicide mortality among veterans co-prescribed opioid and benzodiazepine pharmacotherapy, although the estimate is imprecise at 3 months.

Conclusion 8. Given the lower number of suicide deaths in the eligible population the committee is unable to conduct analyses to address the question of the effect of different opioid dosage treatment strategies on suicide mortality.

Overall, the committee’s findings are consistent with concerns regarding the initiation of opioid pharmacotherapy and concurrent prescribing of benzodiazepine pharmacotherapy. Specifically, concerns about both the prescribing of opioid pharmacotherapy and the interaction between the two pharmacotherapies first appeared in the VA/Department of Defense clinical practice guidelines in 2010 and continued to be raised in the 2017 and 2022 guidelines. In addition, the committee’s findings are based on data on veterans treated by the VHA, as specified in the statement of task; the analysis does not compare the experiences of veterans and prescribing practices of non-VHA providers in other health care systems (which is not reflected in the statement of task). It is important to note that the committee’s interpretation of the results, which occurred in the past, is not intended to influence restriction for any medication nor is it to be viewed as an absolute contraindication for the clinical use of benzodiazepines with opioids.

This page intentionally left blank.