You are here
Impact of an Opioid Stewardship Program In Reducing Unnecessary Opioid Use In a Community Teaching Health System
Objective: The national opioid epidemic has been declared a public health crisis. From 1999 to 2017, the Centers for Disease Control and Prevention reported about 218,000 deaths due to prescription opioid overdoses. An opioid stewardship program was implemented in two community teaching hospitals in an effort to reduce unnecessary use of intravenous (IV) opioids. The objective of this study was to assess the impact of this program on the utilization of IV hydromorphone and morphine in a community teaching health system.
Methods:This single-center observational analysis compared outcomes pre-exposure and post-exposure to the intervention. The intervention included a policy in which pharmacists implemented a next-day-at-noon stoppage of all IV hydromorphone and morphine orders upon verification. This allowed pharmacists to review the orders for appropriateness and recommend alternative nonopioid pain medications when appropriate. Hospice patients and patients receiving patient-controlled analgesia or patient-controlled epidural analgesia were excluded. The primary outcome was the average IV morphine milligram equivalents (MME) per total patient-days. Secondary outcomes included average pain score and incidence of naloxone use. For the primary and secondary outcomes, a two-tailed, unpaired t-test with a P value of < 0.05 was utilized.
Results: There were reductions of 18% and 63% in the average IV MME/total patient-days compared to baseline with IV morphine and hydromorphone utilization, respectively. The rate of naloxone use, based on number of patients treated per total patients receiving opioids, was 0.07% and 0.2% for control and intervention periods, respectively.
Conclusions: The aim of this program was to implement an effective inpatient opioid stewardship program to reduce unnecessary outpatient prescribing practices. There was a statistically significant decline in use of IV hydromorphone and morphine after the implementation of this program. The difference in naloxone use was not statistically significant. Moving forward, pharmacist documentation of interventions can be used to determine the amount of alternative nonopioid therapies used pre- and post-intervention.
Prescription opioids play a major role in the opioid epidemic, with more than 218,000 deaths in the U.S. reported by the Centers for Disease Control and Prevention (CDC) between 1999 and 2017 due to prescription opioid use.1 Moreover, overdose deaths involving prescription opioids were five times higher in 2017 than in 1999.1 The primary catalyst for the opioid crisis was an increase in prescribing rates in the late 1990s. This increase was spurred by claims from pharmaceutical companies that opioids did not carry a risk of addiction, minimizing their potential for abuse and diversion.2 Although opioid prescribing rates have been declining, the amount of opioids in morphine milligram equivalents (MME) and average number of days per prescription have continued to increase since the 1990s.1 Moreover, there continues to be a lack of adequate training in alternative therapies and pain management for health care professionals as a whole, which is also contributing to this crisis.3
The CDC reports that the most common medications involved in prescription opioid overdoses include methadone, oxycodone, and hydrocodone.4 While there is a definite link between inappropriate outpatient opioid prescribing and overdose deaths, there is also a strong relationship between inpatient opioid administration and outpatient opioid use. A retrospective study conducted by the University of Pittsburgh School of Medicine investigated the link between outpatient opioid prescribing and inpatient opioid use in opioid-naïve hospitalized patients.5 The study included opioid-naïve, nonobstetric inpatients with at least one outpatient visit in the 12 months before and after admission.5 The authors found that patients receiving hospital opioids were twice as likely to report outpatient opioid use within 90 days.5
A quality improvement project conducted by the Yale University School of Medicine aimed to assess the impact of a new standard of inpatient opioid prescribing and accompanying education of health care providers in reducing intravenous (IV) opioid exposure.6 This pilot study involved 127 patients in an adult medical unit, with a three-month intervention period compared to a six-month period of historical control. The opioid prescribing standard established the oral route as preferred and the subcutaneous route as the preferred alternative when patients were not able to tolerate oral intake. The primary outcome was the number of IV opioid doses administered per patient-day. Secondary outcomes included parenteral MME per patient-days and mean reported pain scores, among others. The study found an 84% reduction in IV opioid doses per patient-day and a 49% reduction in mean daily MME/patient-day of overall parenteral opioid exposure. Furthermore, there was no significant difference in pre-intervention and post-intervention mean reported pain score the first three days of hospitalization, with improvement seen on days 4 and 5. This study highlights the impact of effective policy changes and stewardship programs in reducing inpatient IV opioid use without compromising pain control.
Considering the opioid epidemic and in accordance with the institution's guiding principles of people-centered care and effective stewardship, an opioid stewardship program was implemented in our institution. The goal of this program is to reduce unnecessary use of IV hydromorphone and morphine in an inpatient setting and subsequently reduce the number of prescriptions written at discharge.
The study was designed as a single-center, prospective analysis comparing outcomes pre-exposure and post-exposure to the intervention. Patients receiving patient-controlled analgesia, patient-controlled epidural analgesia, and hospice or palliative care were excluded, as were pediatric patients. Patients older than 18 years of age prescribed IV hydromorphone and morphine in the observation unit, emergency department, or inpatient unit were included. The study involved a pre-exposure or control period of two months and a post-exposure or intervention period of 14 months. The intervention encompassed a policy implemented in January 2018 in which pharmacists applied next-day-at-noon stoppage to all IV hydromorphone and morphine orders upon verification. Pharmacists daily distributed a list of patients with these orders to the medical team. This allowed pharmacists to discuss pain level with the team and make recommendations for alternative or adjunctive nonopioid pain medications when appropriate. If there was a continued need for opioid therapy, the orders were placed for another 24 hours as needed. Additionally, pharmacists utilized an opioid stewardship tool to make recommendations for IV to oral conversions and alternative nonopioid therapies when appropriate. The stewardship tool, developed in collaboration with clinical pharmacists and pharmacy residents, included quick references on safety and efficacy of nonopioid pain medications for specified indications. Throughout the intervention period, pharmacy residents led presentations to pharmacists, nurses, and physicians on alternative pain medications.
The primary outcomes of the study were IV hydromorphone and morphine utilization based on MMEs per total patient-days. The average IV MME/total patient-days for each month was calculated using total number of charges on patient charts upon dispensing of the opioid product. The conversion factor utilized to convert IV hydromorphone to IV MMEs was 6.66. Secondary outcome measures included average rate of naloxone use and pain scores based on patient satisfaction assessment upon discharge with a rating scale of never = 1, sometimes = 2, usually = 3, and always = 4. It is important to note that the assessment question on the key metrics for pain was modified during the intervention period. The rate of naloxone use was defined as the number of patients treated with naloxone out of the total number of patients receiving opioids.
A two-tailed, unpaired t-test, with P value of < 0.05, was used to determine the difference in the primary and secondary outcomes.
The two-month control period from November 2017 through December 2017 included 4,046 patients, while the 14-month intervention period from January 2018 through February 2019 included 21,945 patients. The number of patients in the control period for IV hydromorphone and morphine was 2,014 and 2,032, respectively. The number of patients in the intervention period for IV hydromorphone and morphine was 9,308 and 12,637, respectively. The mean (standard deviation [SD]) of IV MME/total patient-days for hydromorphone was reduced by 63% (9.49 [2.75] versus 3.47 [2.19], P < 0.001). The mean (SD) of IV MME/total patient-days for morphine was reduced by 18% (0.86 [0.11] versus 0.70 [0.13], P < 0.001) (Figure 1). The change in the rate (SD) of naloxone use pre- and post-intervention was not statistically significant (0.07% [0.035] versus 0.2% [0.091], P = 0.07). The pre-intervention average score on pain assessment upon discharge was 3.50 compared to 3.53 post-intervention.
The major role that prescription opioids play in the opioid epidemic highlights the need for comprehensive solutions to reduce exposure in the inpatient setting. Reducing unnecessary opioid use can help mitigate long-term harm caused by persistent opioid use beyond hospitalization. There was a statistically significant decline in use of IV hydromorphone and morphine after the implementation of the opioid stewardship program. Since the amount of opioid exposure was calculated using number of charges compared to amount administered, it may be an overestimation of use. While the question asked upon discharge based on the key metrics for pain was different pre- and post-intervention, the lack of statistically significant change is suggestive of the continued focus on adequate pain management.
A large cohort study by researchers at the University of Michigan Medical School showed the similar prevalence of long-term opioid use following minor and major surgical procedures among opioid-naïve patients.7 The authors also concluded that new and persistent opioid use after surgery is associated with behavioral and pain disorders rather than pain level.7 This shows the importance of identifying risk factors for persistent opioid use such as tobacco, alcohol use, substance-abuse disorders, and conditions such as anxiety, depression, and preoperative pain disorders. These behavioral and pain disorders can be important confounders in this study and need to be better assessed in the future as part of baseline patient characteristics. Additionally, it is possible there were other confounders or concomitant interventions, such as the general cultural shift to reduce opioid use in response to the epidemic. This can also contribute to the phenomenon of regression toward the mean. On the contrary, initiatives to improve pain management and increase patient satisfaction may contribute to excessive opioid prescribing.
Adoption of an opioid stewardship program centered on daily re-evaluation of IV opioid therapy, coupled with education of prescribers and nursing staff, was associated with a reduction of inpatient exposure to IV opioids. Future directions include measuring other secondary outcomes such as use of nonopioid alternative therapies through pharmacists' documentation and continuation of data collection and analysis accounting for important confounders. Continued engagement with physicians through educational sessions including grand round discussions is critical. Moreover, the multidisciplinary opioid stewardship committee is working to incorporate more nonpharmacological modalities in the inpatient setting.
The authors of this study would like to thank the Holy Cross Health staff for allowing the coordination of this study. The authors also would like to acknowledge the preceptors and previous investigators who offered guidance and completed previous phases of the study and ensured its success.
Disclosure: The authors report no financial or commercial interest in regard to this article.
- Centers for Disease Control and Prevention. Opioid overdose: overview. August 1, 2017. Available at: www.cdc.gov/drugoverdose/data/prescribing/overview.html. Accessed August 3, 2018.
- National Institutes of Health. Opioid overdose crisis. January 2019. Available at: www.drugabuse.gov/drugs-abuse/opioids/opioid-overdose-crisis. Accessed October 18, 2019.
- Bonnie RJ, Ford MA, Phillips JK. Evidence on strategies for addressing the opioid epidemic. In: Phillips JK, Ford MA, Bonnie RJ, eds. Pain Management and the Opioid Epidemic: Balancing Societal and Individual Benefits and Risks of Prescription Opioid Use: Committee on Pain Management and Regulatory Strategies to Address Prescription Opioid Abuse. Washington, D.C.: The National Academies Press; 2017. doi: https://doi.org/10.17226/24781.
- Centers for Disease Control and Prevention. Opioid overdose: overdose deaths involving prescription opioids. August 13, 2019. Available at: www.cdc.gov/drugoverdose/data/prescribing/overdose-death-maps.html. Accessed November 3, 2019.
- Donohue JM, Kennedy JN, Seymour CW, et al. Patterns of opioid administration among opioid-naive inpatients and associations with post discharge opioid use: a cohort study. Ann Intern Med 2019;171(2):81–90.
- Ackerman AL, O’Connor PG, Doyle DL, et al. Association of an opioid standard of practice intervention with intravenous opioid exposure in hospitalized patients. JAMA Intern Med 2018;178(6):759–763.
- Brummett CM, Waljee JF, Goesling J, et al. New persistent opioid use after minor and major surgical procedures in US adults. JAMA Surg 2017;152(6):e170504.