Objectives: To describe the continuity of opioid prescribing and prescriber characteristics among older adults with chronic noncancer pain (CNCP) who are on long-term opioid therapy (LTOT) and to evaluate the association of continuity of opioid prescribing and prescriber characteristics with the risk of opioid-related adverse events.
Study Design: Nested case-control design.
Methods: This study employed a nested case-control design using a 5% random sample of the national Medicare administrative claims data for 2012-2016. Eligible individuals experiencing a composite outcome of opioid-related adverse events were defined as cases and matched to controls using incidence density sampling. Continuity of opioid prescribing (operationalized using the Continuity of Care Index) and prescriber specialty were assessed among all eligible individuals. Conditional logistic regression was conducted to assess the relationships of interest after accounting for known confounders.
Results: Individuals with low (odds ratio [OR], 1.45; 95% CI, 1.08-1.94) and medium (OR, 1.37; 95% CI, 1.04-1.79) continuity of opioid prescribing were found to have greater odds of experiencing a composite outcome of opioid-related adverse events compared with individuals with high prescribing continuity. Fewer than 1 in 10 (9.2%) older adults starting a new LTOT episode received at least 1 prescription from a pain specialist. Receiving a prescription from a pain specialist was not significantly associated with the outcome in adjusted analyses.
Conclusions: We found that higher continuity of opioid prescribing, but not provider specialty, was significantly associated with fewer opioid-related adverse outcomes among older adults with CNCP.
Am J Manag Care. 2023;29(2):88-94. https://doi.org/10.37765/ajmc.2023.89317
- Fewer than 1 in 10 older adults starting new long-term opioid therapy received at least 1 prescription from a pain specialist.
- There was significant variability in continuity of opioid prescribing among older adults on long-term opioid therapy.
- Individuals with low and medium continuity of opioid prescribing were found to have greater odds of experiencing opioid-related adverse events compared with individuals with high prescribing continuity.
- Receiving a prescription from a pain specialist was not significantly associated with risk of opioid-related adverse outcomes.
- Improving continuity of care among older adults on long-term opioid therapy may be an effective intervention for improving patient safety.
Approximately 20% of individuals worldwide are affected by chronic noncancer pain (CNCP).1,2 CNCP commonly affects older adults and is frequently associated with musculoskeletal disorders such as degenerative spine conditions and arthritis.2 Opioid analgesics are often used to manage persistent pain; however, they may lead to adverse events such as overdoses, respiratory depression, cognitive impairment, falls, nausea, and constipation.2-4
Risk factors of opioid-related adverse outcomes for patients on long-term opioid therapy (LTOT) include treatment characteristics, such as dosage and formulation of opioids, and patient characteristics, such as disability status, comorbidities, and use of other medications.5-7 Some research has indicated that obtaining prescriptions from multiple prescribers or pharmacies may also be associated with more high-risk opioid prescriptions and lead to higher risks of an opioid overdose in the subsequent year.8,9 Prescribing continuity, or receiving prescriptions from the same provider, is fundamental to high-quality care and better patient experiences and outcomes, particularly for patients on higher-risk regimens such as LTOT. Fragmented care from multiple providers may be associated with greater risks of adverse drug events.10 Hallvik et al found that lower prescribing continuity in long-term opioid use, as measured by the Continuity of Care Index (COCI), was associated with a greater likelihood of receiving risky opioid prescriptions and having opioid-related hospitalizations.11 Lagisetty and colleagues found that those not having a usual prescriber prior to a surgery were more likely to receive high-risk opioids.12 Similarly, the involvement of a pain specialist in the management of CNCP is strongly recommended to reduce the risk of adverse outcomes associated with LTOT,3 as pain specialists often have more training, better knowledge, and fewer negative perceptions of LTOT.13,14
To evaluate the importance of appropriate patient management in safe use of LTOT, the current study aims to (1) describe the continuity of opioid prescribing and prescriber characteristics among older adults with CNCP who are on LTOT and (2) evaluate the association of continuity of opioid prescribing and prescriber characteristics with the risk of opioid-related adverse outcomes among older adults with CNCP on LTOT.
Study Design and Data Source
This study employed a nested case-control design using a 5% random sample of the national Medicare administrative claims data for 2012-2016. The data include demographic characteristics from a 5% random sample of all Medicare beneficiaries in the United States, along with medical and pharmacy claims and other health care utilization records. The University of Mississippi Institutional Review Board (protocol #18-069) approved the study, and a data use agreement (DUA #RSCH-2018-52319) was obtained from CMS.
Study Cohort Definition
Medicare beneficiaries identified to be on a new LTOT episode between July 1, 2012, and December 31, 2016, were deemed eligible for cohort entry. Based on extant literature, a new LTOT episode was defined as presence of at least 3 prescription claims fills for opioids and a cumulative 45 days of opioid possession in any 90-day period during the study, with no history of opioid prescription fills in the 6-month period prior to start of the chronic opioid therapy episode.5,7,15 The 91st day after initiation of the LTOT episode was considered as the cohort entry date. Beneficiaries were required to be 65 years or older as of the start date of the LTOT episode and continuously enrolled in Medicare parts A, B, and D in the 12-month period prior to cohort entry. Additionally, beneficiaries were required to have no history of cancer and at least 2 claims with diagnoses for a CNCP condition within a 30-day window in the 12-month period prior to start of the chronic opioid use episode (eAppendix Table 1 [eAppendix available at ajmc.com]). Beneficiaries entering the study cohort remained in the cohort until the first occurrence of one of the following events: outcome of interest, death, cancer diagnosis, loss of Medicare eligibility, disenrollment, or end of the study period.
Multiple opioid-related adverse events (ie, opioid-induced respiratory depression [OIRD], opioid-related overdose [OD], and all-cause mortality) were evaluated in this study.6,7,16 Based on prior literature, an OIRD episode was identified based on presence of a diagnosis code for prescription opioid-related poisoning and a procedure code for life-threatening respiratory or central nervous system depression or mechanical ventilation or critical care within 1 day of the prescription opioid-related poisoning (eAppendix Table 2).6,7,14 OD was determined based on presence of a diagnosis code for opioid-related poisoning, or a diagnosis code for opioid-related adverse event with a diagnosis code for opioid overdose on the same day (eAppendix Table 3).5,7 Date of death was obtained from the Master Beneficiary Summary File. A composite outcome, the first occurrence of any of the 3 aforementioned events before the end of the study period, was used for case definition. The date of the first occurrence of an event of interest was considered as the index date.
Controls were defined as beneficiaries who entered the study cohort but did not experience the composite outcome as of the index date of their matched case. Incidence density sampling was employed to select 1 control for each case, to allow for random sampling from the pool of eligible controls, such that the time at risk of the composite outcome for each control beneficiary was equal to or more than that of their matched case. This technique allows for beneficiaries who were selected as controls to also serve as a case at a future time point.17 Moreover, it is possible for a particular beneficiary to serve as a control for more than 1 case. Cases and controls were matched on age and time of cohort entry. In addition to previously mentioned eligibility criteria, cases and controls were also required to be on LTOT (ie, at least 3 prescription fills and 45 days of supply) in the 90 days prior to index date.7
Operationalization of Continuity of Opioid Prescribing and Prescriber Specialty
Opioid prescribing continuity was determined using the COCI proposed by Bice and Boxerman, which captures “the extent to which a given individual’s total number of visits for an episode of illness or a specific time period are with a single [provider] or group of referred providers.”18 In the context of the present study, COCI was calculated based on opioid prescriptions filled; its value ranged from 0 (complete fragmentation of care) to 1 (perfect continuity of care). Beneficiaries with a COCI score of 0.467 (25th percentile) or lower were assigned to the low COCI group, those with a score of 1 (75th percentile) were assigned to the high COCI group, and the remaining were assigned to the medium COCI group. Finally, the specialty of the prescriber for each individual prescription was identified using the Medicare prescriber characteristics file. Providers who self-identified their specialty as pain medicine were classified as being a pain specialist. Each beneficiary was classified according to whether or not they had at least 1 opioid prescription from a pain specialist, with the hypothesis that prescriptions from pain specialists were indicative of a greater degree of management of pain treatment.
COCI and prescriber specialty were computed at different time points for the 2 study aims. For aim 1, COCI and the presence of at least 1 prescription from a pain specialist were identified in the 6-month period prior to the cohort entry date (ie, during the initiation of the new LTOT episode). For aim 2, COCI and the presence of at least 1 prescription from a pain specialist were assessed in the 6-month hazard period prior to the index date for each case and their matched control. A sensitivity analysis was conducted using an alternative measure of continuity of care, the Herfindahl Index (HI).19
Potential confounding variables included in the analysis were beneficiary sociodemographic characteristics, clinical characteristics, and opioid medication use characteristics. Sociodemographic characteristics included race, sex, Medicare low-income subsidy status, and region of residence. Clinical characteristics included comorbidity score, assessed using the Deyo adaptation of the Charlson Comorbidity Index,20 in addition to separate indicators for presence of multiple CNCP conditions, mental illnesses, renal insufficiency, hepatic insufficiency, Parkinson disease, chronic obstructive pulmonary disorder, sleep apnea, or other sleep disorders (eAppendix Table 4). Use of other classes of medications such as benzodiazepines and muscle relaxants during the hazard period were also included because they are likely to influence the risk of adverse reactions that were captured in the study. Additionally, history of OD and OIRD were assessed. Opioid medication use characteristics included the mean daily dose of opioids prescribed in morphine milligram equivalent (MME) units and the type of opioid prescribed. Based on CDC guidance and prior research, the mean daily dose of opioids was divided into the following categories: less than 20 MME, 20 up to 50 MME, and 50 MME or greater.21 For the type of opioid prescribed, 3 categories were created: short-acting opioids only, long-acting opioids only, and combination of short-acting and long-acting products.7,22 For aim 1, the characteristics of opioid therapy were assessed in the 6 months prior to cohort entry, whereas they were assessed in the period prior to the index date for aim 2. All other covariates were assessed during the study period for aim 1 and only prior to the index date for aim 2.
Descriptive statistics were used to report beneficiary characteristics, prescriber specialty, and continuity of opioid prescribing. Statistical comparisons between cases and controls were conducted using McNemar test, Cochran-Mantel-Haenszel test, or paired t test. Conditional logistic regression models were used to examine the relationships among continuity of opioid prescribing, presence of a prescription from a pain specialist, and the composite outcome of opioid-related adverse events accounting for the matched case-control data. All analyses were conducted using SAS version 9.4 (SAS Institute).
The characteristics of the 35,189 Medicare beneficiaries who were new LTOT users are shown in Table 1 [part A and part B]. Their mean age was 77 years, 24,342 (69%) were female, 29,321 (83%) were White, and 15,054 (43%) were enrolled in the low-income subsidy program. During the initial LTOT episode, eligible individuals filled a mean (SD) of 4.3 (1.9) opioid prescriptions for a total mean (SD) of 90.1 (35.4) days of supply; 39.0% (n = 13,721) of them had a low COCI score; and only 9.2% (n = 2884) received at least 1 prescription from a pain specialist. The departure of the distribution of raw COCI scores (Figure) from normality explains the abundance of the study population in the lowest quartile of scores.
The composite outcome was experienced by 3.2% (n = 1122) of all eligible individuals prior to the end of the follow-up period. The median (IQR) time between cohort entry and the index date for those who experienced the outcome was 254.5 (77-519) days. The mean (SD) COCI for cases was 0.70 (0.30); for controls, it was 0.75 (0.30). The presence of at least 1 prescription from a pain specialist was identified in 8.4% (n = 90) of cases and in 6.5% (n = 70) of controls. In the 90 days preceding the index date, controls had significantly fewer mean (SD) fills for opioid prescriptions compared with cases (3.7 [1.4] vs 4.5 [2.3]; P < .0001) and received prescriptions for fewer days (90.9 [27.6] vs 94.2 [34.6]; P = .0147). Cases also received higher doses of opioid medications compared with controls during this period (P < .0001).
Individuals with low COCI (odds ratio [OR], 1.61; 95% CI, 1.28-2.02) and medium COCI (OR, 1.65; 95% CI, 1.33-2.04) had greater odds of experiencing the outcome relative to those with high COCI (Table 2). Beneficiaries who received at least 1 opioid prescription from a pain specialist had 34% lower odds (OR, 0.66; 95% CI, 0.48-0.93) of the composite outcome compared with those who did not receive any opioid prescription from a pain specialist.
After adjusting for all potential confounders, low COCI (OR, 1.45; 95% CI, 1.08-1.94) and medium COCI (OR, 1.37; 95% CI, 1.04-1.79) were significantly associated with the composite outcome compared with high COCI (Table 3). Patients receiving at least 1 opioid prescription from a pain specialist had 29% (OR, 0.71; 95% CI, 0.46-1.09) lower odds of the composite outcome relative to those who did not receive any opioid prescription from a pain specialist. However, this finding was not statistically significant. Sensitivity analysis using the HI showed results similar to COCI (eAppendix Tables 5 and 6). When the primary analyses were repeated using only mortality events as the outcome, 943 cases were identified in the study cohort, and although the adjusted ORs predicting the effect of COCI (low COCI: OR, 1.29; 95% CI, 0.94-1.76; medium COCI: OR, 1.35; 95% CI, 1.00-1.82) and receiving a prescription from a pain specialist (OR, 0.91; 95% CI, 0.50-1.66) were not statistically significant, they were in the same direction as the primary findings.
This study examined the safety of LTOT in a nationally representative cohort of Medicare-eligible older adults with CNCP and evaluated the importance of prescribing characteristics on the risk of opioid-related adverse outcomes. Specifically, this study found that less than 10% of opioid prescriptions in the eligible population were prescribed by a pain specialist and that there were wide variations in COCI among older adults starting a new LTOT episode.
After accounting for confounders, individuals in the lowest quartile of continuity of opioid prescribing had nearly 50% greater odds of experiencing an adverse event than those who had high continuity of care. To our knowledge, only 1 study has previously evaluated the impact of continuity of opioid prescribing. Hallvik and colleagues used prescription drug monitoring program (PDMP) data from the state of Oregon and found that individuals with low continuity of opioid prescribing were more likely to receive riskier prescriptions and to experience any opioid-related hospitalization.11 Data from previous studies showed that patients with high continuity of care in general have lower rates of mortality23 and preventable hospitalizations,24,25 receive fewer unnecessary medical services,26 and have fewer emergency department visits,27 lower health care costs, and adverse outcomes.28 Previous research examining overdoses measures the number of unique prescribers or pharmacies used by an individual. Those using multiple providers are usually referred to as engaging in provider shopping, a behavior that may be indicative of nonmedical use or drug diversion.9,29 In fact, the number of unique prescribers of controlled substances is the fundamental basis of PDMP programs implemented in nearly all states across the United States.30,31 Provider shopping measures are also endorsed as a measure of quality of opioid prescribing by organizations such as the Pharmacy Quality Alliance.32 However, this indicator may not in fact be reflective of nonmedical use of prescription opioids, particularly among older adults, in whom prevalence of such behavior is considerably lower.33,34 The current study’s findings show that continuity of opioid prescribing may be a valid indicator of appropriateness and safety of LTOT among older adults. Our findings suggest the need for revising indicators of safety in opioid prescribing among older adults on LTOT. Measures that evaluate continuity in prescribing may perhaps be more appropriate than a simple evaluation of the number of unique prescribers. These findings are supported by existing evidence that contends that continuity of care is essential for appropriate management of CNCP,35 and they are particularly salient in light of the disruptions to the care continuum occurring as a result of the global COVID-19 pandemic.36 Future research should compare the importance of COCI with that of other commonly used indicators of high-risk opioid use, such as the number of prescribers or the NarxCare Score.37
Interestingly, data from the current study demonstrated that individuals with at least 1 opioid prescription from a pain specialist were not significantly less likely to experience an opioid-related adverse outcome. Although this is a surprising finding, no previous study has directly evaluated this relationship between prescriptions from pain specialists and adverse events, as far as we know. Nevertheless, a significant amount of evidence supports the role of pain specialists in the management of CNCP.38-40 The CDC also recommends that pain specialists be consulted for the treatment of chronic pain.3
Pain specialists also have more training and knowledge around (and fewer negative perceptions of) LTOT,13,14 and they prescribe opioid medications more often than primary care providers, but their prescriptions are likely to be for lower dosages.41,42 It is possible that our findings are indicative of the fact that having at least 1 prescription from a pain specialist does not adequately capture the benefits of pain specialist involvement in CNCP management. Alternative operationalizations may be necessary to capture the benefits of pain specialist involvement in LTOT. On the other hand, the fact that the OR for the pain specialist variable changed from significant to not significant after adjustment for other covariates suggests that a more complex relationship may be at play. It is possible that after accounting for continuity of opioid prescribing, provider specialty does not significantly improve safety in LTOT. If such a hypothesis is in fact accurate, it may be a promising finding for the future of opioid prescribing in the United States. Previous studies’ findings have shown that pain specialists are not easily accessible, and those authors have called for greater numbers of allied providers such as pain-specialist pharmacists.43,44 Nonetheless, a renewed focus on continuous patient management, regardless of the provider specialty, should still hold the potential for improving the safety of patients on LTOT.
Sensitivity analyses showed that study findings were replicated when using an alternate operationalization of continuity of opioid prescribing but not when using mortality as the outcome. It is important to note that the analyses predicting mortality were conducted with fewer cases, and further research is needed to identify whether the impact of continuity of opioid prescribing is driven by certain types of outcomes over others.
These findings should be interpreted in the context of some limitations. First, this study operationalized continuity of opioid prescribing using COCI, which has limitations. The COCI assumes that repeated visits to the same provider allow for greater continuity, but it is unable to capture the actual content of the visit. It is also important to consider that perfect continuity is not always ideal. Management of chronic conditions requires involving multidisciplinary teams of providers who work together to address various aspects of care, and this cannot be captured by the continuity of care measure used in this study. Second, our study design estimated COCI during a 6-month duration prior to the index date during which individuals were required to have at least 3 prescriptions for opioid medications. However, when the number of prescriptions is low, the COCI may yield unstable estimates and unreliable results. In addition, because COCI varies over time, it may lead to heterogeneity in outcomes. Although our study did not explore such changes over time, future studies should evaluate prescribing continuity across longer durations to validate these results. Because this study used administrative claims data, it was unable to capture prescriptions paid for using cash. This study was also unable to capture individuals enrolled in Medicare Advantage plans or those younger than 65 years. Future studies should test these relationships in various populations to expand the generalizability of our findings.33 Finally, future research should also attempt to delineate the impact of plan-level variables and formulary design on continuity of opioid prescribing among older adults on LTOT.45
This study’s data showed that Medicare-enrolled older adults who initiate LTOT have moderate levels of continuity of opioid prescribing and only a small portion of them receive prescriptions from pain specialists. Additionally, higher continuity of opioid prescribing, but not provider specialty, was significantly associated with fewer opioid-related adverse outcomes among older adults with CNCP.
Author Affiliations: Department of Pharmacy Administration (SR, MS, KB, JPB, SM, YY) and Center for Pharmaceutical Marketing and Management (SR, KB, JPB, YY), University of Mississippi School of Pharmacy, University, MS; Department of Anesthesiology, School of Medicine, University of Mississippi Medical Center (IE), Jackson, MS; Department of Epidemiology, School of Public Health, University of Alabama at Birmingham (GM), Birmingham, AL; Sally McDonnell Barksdale Honors College, University of Mississippi (MJM), University, MS.
Source of Funding: This work was supported by the National Institute on Drug Abuse (R15DA046036; PI: Yi Yang).
Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (SR, KB, JPB, GM, YY); acquisition of data (SR, JPB, YY); analysis and interpretation of data (SR, MS, KB, JPB, SM, GM, MM, YY); drafting of the manuscript (SR, MS, KB, SM, GM, MM); critical revision of the manuscript for important intellectual content (SR, MS, KB, JPB, SM, IE, GM, MM, YY); statistical analysis (SR, MS, KB, JPB, MM); obtaining funding (SR, YY); administrative, technical, or logistic support (JPB, IE, YY); and supervision (IE, YY).
Address Correspondence to: Yi Yang, MD, PhD, University of Mississippi School of Pharmacy, Faser 225, University, MS 38655. Email: [email protected].
1. Els C, Jackson TD, Kunyk D, et al. Adverse events associated with medium- and long-term use of opioids for chronic non-cancer pain: an overview of Cochrane reviews. Cochrane Database Syst Rev. 2017;10:CD012509. doi:10.1002/14651858.CD012509.pub2
2. American Geriatrics Society Panel on Pharmacological Management of Persistent Pain in Older Persons. Pharmacological management of persistent pain in older persons. J Am Geriatr Soc. 2009;57(8):1331-1346. doi:10.1111/j.1532-5415.2009.02376.x
3. Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain—United States, 2016. JAMA. 2016;315(15):1624-1645. doi:10.1001/jama.2016.1464
4. Baldini A, Von Korff M, Lin EHB. A review of potential adverse effects of long-term opioid therapy: a practitioner’s guide. Prim Care Companion CNS Disord. 2012;14(3):PCC.11m01326. doi:10.4088/PCC.11m01326
5. Dunn KM, Saunders KW, Rutter CM, et al. Opioid prescriptions for chronic pain and overdose: a cohort study. Ann Intern Med. 2010;152(2):85-92. doi:10.7326/0003-4819-152-2-201001190-00006
6. Zedler B, Xie L, Wang L, et al. Risk factors for serious prescription opioid-related toxicity or overdose among Veterans Health Administration patients. Pain Med. 2014;15(11):1911-1929. doi:10.1111/pme.12480
7. Salkar M, Ramachandran S, Bentley JP, et al. Do formulation and dose of long-term opioid therapy contribute to the risk of adverse events among older adults? J Gen Intern Med. 2022;37(2):367-374. doi:10.1007/s11606-021-06792-8
8. Cepeda MS, Fife D, Chow W, Mastrogiovanni G, Henderson SC. Opioid shopping behavior: how often, how soon, which drugs, and what payment method. J Clin Pharmacol. 2013;53(1):112-117. doi:10.1177/0091270012436561
9. Carey CM, Jena AB, Barnett ML. Patterns of potential opioid misuse and subsequent adverse outcomes in Medicare, 2008 to 2012. Ann Intern Med. 2018;168(12):837-845. doi:10.7326/M17-3065
10. Beadles CA, Voils CI, Crowley MJ, Farley JF, Maciejewski ML. Continuity of medication management and continuity of care: conceptual and operational considerations. SAGE Open Med. 2014;2:2050312114559261. doi:10.1177/2050312114559261
11. Hallvik SE, Geissert P, Wakeland W, et al. Opioid-prescribing continuity and risky opioid prescriptions. Ann Fam Med. 2018;16(5):440-442. doi:10.1370/afm.2285
12. Lagisetty P, Bohnert A, Goesling J, et al. Care coordination for patients on chronic opioid therapy following surgery: a cohort study. Ann Surg. 2020;272(2):304-310. doi:10.1097/SLA.0000000000003235
13. Varrassi G, Müller-Schwefe GHH. The international CHANGE PAIN physician survey: does specialism influence the perception of pain and its treatment? Curr Med Res Opin. 2012;28(5):823-831. doi:10.1185/03007995.2012.674499
14. McCarberg BH, Patel AA, Benson CJ, et al. Physician management of moderate-to-severe acute pain: results from the Physicians Partnering Against Pain (P3) study. J Opioid Manag. 2013;9(6):401-406. doi:10.5055/jom.2013.0182
15. Ramachandran S, Salkar M, Bentley JP, Eriator I, Yang Y. Patterns of long-term prescription opioid use among older adults in the United States: a study of Medicare administrative claims data. Pain Physician. 2021;24(1):31-40.
16. Zedler BK, Saunders WB, Joyce AR, Vick CC, Murrelle EL. Validation of a screening risk index for serious prescription opioid-induced respiratory depression or overdose in a US commercial health plan claims database. Pain Med. 2018;19(1):68-78. doi:10.1093/pm/pnx009
17. Lubin JH, Gail MH. Biased selection of controls for case-control analyses of cohort studies. Biometrics. 1984;40(1):63-75.
18. Bice TW, Boxerman SB. A quantitative measure of continuity of care. Med Care. 1977;15(4):347-349. doi:10.1097/00005650-197704000-00010
19. Pollack CE, Hussey PS, Rudin RS, Fox DS, Lai J, Schneider EC. Measuring care continuity: a comparison of claims-based methods. Med Care. 2016;54(5):e30-e34. doi:10.1097/MLR.0000000000000018
20. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-619. doi:10.1016/0895-4356(92)90133-8
21. Calculating total daily dose of opioids for safer dosage. CDC. Accessed May 26, 2021. https://www.cdc.gov/drugoverdose/pdf/calculating_total_daily_dose-a.pdf
22. Shah A, Hayes CJ, Martin BC. Characteristics of initial prescription episodes and likelihood of long-term opioid use — United States, 2006-2015. MMWR Morb Mortal Wkly Rep. 2017;66(10):265-269. doi:10.15585/mmwr.mm6610a1
23. Pereira Gray DJ, Sidaway-Lee K, White E, Thorne A, Evans PH. Continuity of care with doctors—a matter of life and death? a systematic review of continuity of care and mortality. BMJ Open. 2018;8(6):e021161. doi:10.1136/bmjopen-2017-021161
24. Cheng SH, Chen CC, Hou YF. A longitudinal examination of continuity of care and avoidable hospitalization: evidence from a universal coverage health care system. Arch Intern Med. 2010;170(18):1671-1677. doi:10.1001/archinternmed.2010.340
25. Nyweide DJ, Anthony DL, Bynum JPW, et al. Continuity of care and the risk of preventable hospitalization in older adults. JAMA Intern Med. 2013;173(20):1879-1885. doi:10.1001/jamainternmed.2013.10059
26. Romano MJ, Segal JB, Pollack CE. The association between continuity of care and the overuse of medical procedures. JAMA Intern Med. 2015;175(7):1148-1154. doi:10.1001/jamainternmed.2015.1340
27. Kern LM, Seirup JK, Rajan M, Jawahar R, Stuard SS. Fragmented ambulatory care and subsequent emergency department visits and hospital admissions among Medicaid beneficiaries. Am J Manag Care. 2019;25(3):107-112.
28. Amjad H, Carmichael D, Austin AM, Chang CH, Bynum JPW. Continuity of care and health care utilization in older adults with dementia in fee-for-service Medicare. JAMA Intern Med. 2016;176(9):1371-1378. doi:10.1001/jamainternmed.2016.3553
29. Cepeda MS, Fife D, Chow W, Mastrogiovanni G, Henderson SC. Assessing opioid shopping behaviour: a large cohort study from a medication dispensing database in the US. Drug Saf. 2012;35(4):325-334. doi:10.2165/11596600-000000000-00000
30. Prescription drug monitoring programs: a guide for healthcare providers. Substance Abuse and Mental Health Services Administration. Winter 2017. Accessed May 3, 2021. https://store.samhsa.gov/sites/default/files/d7/priv/sma16-4997.pdf
31. Prescription drug monitoring programs (PDMPs). CDC. Accessed May 3, 2021. https://www.cdc.gov/drugoverdose/pdf/pdmp_factsheet-a.pdf
32. PQA opioid measure set. Pharmacy Quality Alliance. January 2021. Accessed May 3, 2021. https://www.pqaalliance.org/assets/Measures/PQA_Opioid_Core_Measure_Set_Description.pdf
33. Adewumi AD, Maravilla JC, Alati R, et al. Multiple opioid prescribers: a genuine quest for treatment rather than aberrant behaviour. a two-decade population-based study. Addict Behav. 2020;108:106458. doi:10.1016/j.addbeh.2020.106458
34. Neutel CI, Skurtveit S, Berg C, Sakshaug S. Multiple prescribers in older frequent opioid users—does it mean abuse? J Popul Ther Clin Pharmacol. 2013;20(3):e397-e405.
35. Satterwhite S, Knight KR, Miaskowski C, et al. Sources and impact of time pressure on opioid management in the safety-net. J Am Board Fam Med. 2019;32(3):375-382. doi:10.3122/jabfm.2019.03.180306
36. Hadeed N, Fendrick AM. Enhance care continuity post COVID-19. Am J Manag Care. 2021;27(4):135-136. doi:10.37765/ajmc.2021.88508
37. Emara AK, Grits D, Klika AK, et al. NarxCare scores greater than 300 are associated with adverse outcomes after primary THA. Clin Orthop Relat Res. 2021;479(9):1957-1967. doi:10.1097/CORR.0000000000001745
38. Hanna MN, Speed TJ, Shechter R, et al. An innovative perioperative pain program for chronic opioid users: an academic medical center’s response to the opioid crisis. Am J Med Qual. 2019;34(1):5-13. doi:10.1177/1062860618777298
39. Vadivelu N, Kai AM, Kodumudi V, Berger JM. Challenges of pain control and the role of the ambulatory pain specialist in the outpatient surgery setting. J Pain Res. 2016;9:425-435. doi:10.2147/JPR.S86579
40. Patwardhan A, Matika R, Gordon J, Singer B, Salloum M, Ibrahim M. Exploring the role of chronic pain clinics: potential for opioid reduction. Pain Physician. 2018;21(6):E603-E610.
41. Pan K, Blankley AI, Hughes PJ. An examination of opioid prescription for Medicare Part D patients among family practice prescribers. Fam Pract. 2019;36(4):467-472. doi:10.1093/fampra/cmy090
42. Alamanda VK, Wally MK, Seymour RB, Springer BD, Hsu JR; Prescription Reporting With Immediate Medication Utilization Mapping Group. Prevalence of opioid and benzodiazepine prescriptions for osteoarthritis. Arthritis Care Res (Hoboken). 2020;72(8):1081-1086. doi:10.1002/acr.23933
43. Wiznia DH, Zaki T, Maisano J, Kim CY, Halaszynski TM, Leslie MP. Influence of medical insurance under the Affordable Care Act on access to pain management of the trauma patient. Reg Anesth Pain Med. 2017;42(1):39-44. doi:10.1097/AAP.0000000000000502
44. Atkinson TJ, Gulum AH, Forkum WG. The future of pain pharmacy: driven by need. Integr Pharm Res Pract. 2016;5:33-42. doi:10.2147/IPRP.S63824
45. Andersen M, Lorenz V, Pant A, Bray J, Alexander G. Association of opioid utilization management with prescribing and overdose. Am J Manag Care. 2022;28(2):e63-e68. doi:10.37765/ajmc.2022.88829