Key Points
Question
What are the implications of the shortage of parenteral opioids for prescription patterns and pain response among hospitalized patients with cancer?
Findings
In a cohort study of 386 patients with cancer referred to a palliative care team for pain management before and after the announcement of parenteral opioid shortages, the referring oncology and palliative care teams prescribed significantly fewer parenteral opioids after the announced shortages and more nonparenteral opioids. After parenteral opioid shortages, significantly fewer patients had achieved clinically improved pain by follow-up day 1.
Meaning
Less availability of parenteral opioids is associated with worse analgesia, which has potential implications for patient satisfaction and hospital length of stay.
This cohort study analyzes the implications of the parenteral opioid shortage for opioid prescription patterns and pain management in hospitalized patients with cancer referred to palliative care.
Abstract
Importance
The recent parenteral opioid shortage (POS) has potential implications for cancer-related pain management in hospitalized patients.
Objective
This study compared changes in opioid prescriptions and clinically improved pain (CIP) among patients treated by an inpatient palliative care (PC) team before and after our institution first reported the POS.
Design, Setting, and Participants
A cohort study of 386 eligible patients with cancer treated at a comprehensive cancer center 1 month before and after the announcement of the POS. We reviewed data from electronic health records, including patient demographics, opioid type, route of administration, and dose. Board-certified palliative care specialists assessed CIP at follow-up day 1.
Exposures
The announcement of the POS by the institution’s pharmacy and therapeutics committee on February 8, 2018.
Main Outcomes and Measures
The primary outcome was to measure the change in opioid prescription patterns of physicians, and the secondary outcome was to measure the proportion of patients who achieved CIP before and after announcement of the POS.
Results
Of 386 eligible patients, 196 were men (51%), 270 were white (70%), and the median age was 58 years (interquartile range, 46-67 years). Parenteral opioids were prescribed less frequently by the referring oncology teams after the POS (56 of 314 [18%]) vs before the POS (109 of 311 [35%]) (P < .001). The PC team also prescribed fewer parenteral opioids after the POS (96 of 336 [29%]) vs before the POS (159 of 338 [47%]) (P < .001). After the POS (vs before the POS), significantly fewer patients achieved CIP on follow-up day 1 (119 [62%] vs 144 [75%] of 193; P = .01). Multivariate analysis showed that before the POS, patients had an 89% higher chance of achieving CIP on follow-up day 1 (odds ratio, 1.89; 95% CI, 1.22-2.94; P = .005).
Conclusions and Relevance
There was a significant change in opioid prescription patterns associated with the POS. Furthermore, after the POS, fewer patients achieved CIP. These factors have potential implications for patient satisfaction and hospital length of stay.
Introduction
Pain is one of the most common symptoms among patients with cancer; it is experienced by 39% of cancer survivors, 55% of patients undergoing active cancer treatment, and 66% of patients with advanced, metastatic, or terminal-stage disease.1 The effectiveness of opioids, especially parenteral opioids, in treating cancer pain has been well recognized.2,3,4,5 Despite this evidence and consensus guidelines published by various organizations, cancer pain remains inadequately treated among hospitalized patients.6,7,8,9,10 Untreated or undertreated cancer pain not only affects patients’ physical activity, mood, and overall quality of life, but it is also associated with delayed hospital discharge and increased medical costs.11,12,13,14
Recently, there has been a national parenteral opioid shortage (POS), which includes morphine, hydromorphone, methadone, and fentanyl.14,15 There are several reasons for these shortages, including high demand for opioids,16 delays as well as reductions in the production quotas of schedule 1 and 2 drugs set by the US Drug Enforcement Agency,17 and shortages of sterile injectable products from drug manufacturers.18 As a consequence, physicians have been forced to make changes in the patterns of opioid prescription for hospitalized patients. The POS may further potentiate existing barriers to adequate cancer pain management.19 To our knowledge, no data are available regarding the association of the recent POS with cancer pain management among hospitalized patients. Drug shortages, especially of sterile injectable products, will likely persist15; therefore, knowledge obtained from this study will help health care professionals and hospital facilities to create timely, safe, and effective countermeasures.
In this study, the primary objective was to examine the implications of the recent POS for opioid prescriptions and cancer pain improvement, which was defined as achievement of clinically improved pain (CIP), among patients referred to our inpatient palliative care (PC) team for opioid management. We hypothesized that the POS has been associated with the use of fewer parenteral and more oral and transdermal opioids and delayed achievement of CIP.
Methods
This retrospective cohort study was approved by the institutional review board at The University of Texas MD Anderson Cancer Center. The institutional review board waived the requirement for written patient informed consent. Eligible patients were 18 years old or older, had a diagnosis of either early-stage cancer or advanced-stage cancer (defined as locally advanced, recurrent, or metastatic disease), were seen by the inpatient PC team as a new consultation, and had at least 1 consecutive follow-up visit. Patients who were not taking any opioids were excluded from the study. The MD Anderson pharmacy and therapeutics committee announced the POS on February 8, 2018.
We selected 2 cohorts of patients 1 month before and after the POS was first announced so that the clinicians would be well adjusted to the countermeasures initiated by the institution at the time of POS. During these time frames, there were no changes to the structure or practices in opioid management made by the PC team. We reviewed and compared the electronic health records of 386 eligible patients at 1 month before and after the announcement of the POS. The cohort examined before the POS included 193 consecutive patients treated between December 8, 2017, and January 8, 2018. The cohort examined after the POS included 193 consecutive patients treated between March 8, 2018, and April 8, 2018.
Opioid Data
Data were collected on the opioid route of administration, type, and dose—defined as the oral morphine equivalent daily dose (MEDD) in milligrams per day—on the day of consultation and consecutive follow-up day 1. Data collected at the consultation day were considered baseline. Data on the opioid route included parenteral and nonparenteral opioids. Parenteral opioids included those administered intravenously and through patient-controlled analgesia. Nonparenteral opioids included extended release (ER) preparations of oral opioids, immediate release (IR) preparations of oral opioids, and transdermal (TD) opioids. Opioid types included codeine, morphine, hydromorphone, hydrocodone, oxycodone, methadone, tramadol, and fentanyl. Data on the MEDD were collected using standard conversion ratios.20 The MEDD was inclusive of scheduled and as-needed opioids administered to patients within the past 24-hour period.
Baseline opioid data reflected the prescription patterns of the primary oncology teams, whereas opioid data at follow-up days 1 reflected the prescription patterns of the PC team.
Clinically Improved Pain
All patients were seen at least once a day by a PC clinician—a board-certified attending physician, an advanced practice clinician, or PC fellow—at baseline and on follow-up day 1. An attending physician comanaged the patients seen by an advanced practice clinician or PC fellow. Pain is the primary goal of the first follow-up for the patients who are referred for uncontrolled pain. The diagnosis of CIP was made for a patient with (1) pain documented as “well-controlled,” “better,” or “comfortable,” (2) the absence of new/worsening pain, and (3) absence of opioid-induced neurotoxic effects on follow-up day 1. We classified pain response into 2 categories: CIP achieved or CIP not achieved.
Data were collected on patient demographics, including age, sex, and race. Clinical information was collected on patient cancer type, Eastern Cooperative Oncology Group (ECOG) performance status,21 the Cut-down, Annoyed, Guilty, Eye-opener (CAGE) questionnaire for alcoholism,22 the Edmonton Symptom Assessment Scale (ESAS),23 and current or prior history of tobacco use.
Statistical Considerations
Data were summarized by using standard descriptive statistics such as mean, SD, median, and range for continuous variables and frequency and proportion for categorical variables. Association between categorical variables was examined using a χ2 test or Fisher exact test, when appropriate. A Wilcoxon rank sum test or Kruskal-Wallis test was used to examine the difference in continuous variables between or among patients’ characteristic groups. Univariate logistic regression analysis was used to investigate 12 variables as potential predictors of CIP on follow-up day 1, which included age, sex, race, marital status, time of POS, ECOG, CAGE, ESAS all items, tobacco use, and baseline opioid route of administration, type, and MEDD. We also conducted multivariate analysis performed by using the stepwise selection method. All tests were 2-sided, and P ≤ .05 was considered to be statistically significant. All computations were carried out using SAS software version 9.4 (SAS Institute Inc).
Results
Table 1 summarizes the demographic and clinical characteristics of the patients before and after the POS. The median age was 58 years (interquartile range [IQR], 46-67 years). Of 386 patients, 196 (51%) were men, and 270 (70%) were white. The most common cancers were gastrointestinal (n = 109, 28%), genitourinary (n = 49, 13%), lung (n = 46, 12%), and gynecologic (n = 44, 11%). Variables such as age, sex, race, marital status, type of cancer, positive CAGE, current or former history of smoking, and performance status did not differ significantly between the 2 cohorts. There was no difference in symptom assessment scores on ESAS except that patients reported higher median baseline scores for drowsiness before vs after the POS (3; IQR, 0-5 vs 2; IQR, 0-5; P = .04).
Table 1. Demographic and Clinical Characteristics of Patients Before and After the Parenteral Opioid Shortage .
| Characteristic | No. of Patients (%) | P Value | ||
|---|---|---|---|---|
| Before POS (n = 193) | After POS (n = 193) | Total (N = 386) | ||
| Median age (IQR), y | 58 (45-67) | 57 (48-67) | 58 (46-67) | .70 |
| Male | 92 (48) | 104 (54) | 196 (51) | .22 |
| Race | .64 | |||
| Black | 23 (12) | 27 (14) | 50 (13) | |
| Hispanic | 18 (10) | 14 (7) | 32 (8) | |
| White | 131 (68) | 139 (72) | 270 (70) | |
| Other | 21 (10) | 13 (7) | 34 (9) | |
| Cancer type | .07 | |||
| Breast | 18 (9) | 6 (3) | 24 (6) | |
| Gastrointestinal | 52 (27) | 57 (30) | 109 (28) | |
| Genitourinary | 27 (14) | 22 (11) | 49 (13) | |
| Gynecologic | 27 (14) | 17 (9) | 44 (11) | |
| Head and neck | 11 (6) | 15 (8) | 26 (7) | |
| Hematologic | 17 (9) | 21 (11) | 38 (10) | |
| Lung cancer | 17 (9) | 29 (15) | 46 (12) | |
| Other | 24 (12) | 26 (13) | 50 (13) | |
| Marital status (missing = 3) | .25 | |||
| Single | 36 (19) | 30 (16) | 66 (17) | |
| Married | 122 (64) | 136 (71) | 258 (67) | |
| Divorced/separated/widowed | 29 (14) | 26 (12) | 55 (14) | |
| Other | 6 (3) | 1 (1) | 7 (2) | |
| CAGE (missing = 41) | ||||
| Positive | 14 (9) | 17 (9) | 31 (9) | .76 |
| Smoking status (missing = 1) | ||||
| Positivea | 105 (54) | 103 (53) | 208 (54) | .17 |
| ECOG PS (missing = 82) | ||||
| ≤1 | 20 (10) | 19 (10) | 39 (10) | .27 |
| ≥2 | 122 (63) | 143 (74) | 265 (67) | |
| Baseline ESAS, median (IQR)b | ||||
| Pain | 7 (5-8) | 7 (4-8) | 7 (4-8) | .51 |
| Fatigue | 7 (5-8) | 6 (5-8) | 7 (5-8) | .90 |
| Nausea | 2 (0-5) | 1 (0-5) | 2 (0-5) | .24 |
| Depression | 0 (0-4) | 1 (0-4) | 0 (0-4) | .13 |
| Anxiety | 2 (0-5) | 2 (0-5) | 2 (0-5) | .60 |
| Drowsiness | 3 (0-5) | 2 (0-5) | 2 (0-5) | .04 |
| Appetite | 5 (3-8) | 5 (2-7) | 5 (3-8) | .30 |
| Dyspnea | 0 (0-3) | 0 (0-5) | 1 (0-4) | .06 |
| Well-being | 5 (4-7) | 5 (4-7) | 5 (4-7) | .96 |
| Sleep | 3 (1-7) | 3 (2-6) | 3 (1-7) | .89 |
| Financial distress | 0 (0-3) | 0 (0-2) | 0 (0-2) | .22 |
| Spiritual pain | 0 (0) | 0 (0) | 0 (0) | .42 |
Abbreviations: CAGE, Cut-down, Annoyed, Guilty, Eye-opener questionnaire for alcoholism; ECOG PS, Eastern Cooperative Oncology Group performance status; ESAS, Edmonton System Assessment Scale; IQR, interquartile range; POS, parenteral opioid shortage.
Includes both former and current smokers.
The severity at the time of assessment of each symptom is rated from 0 to 10 on a numerical scale, with 0 indicating that a symptom is absent and 10 indicating that it is of the worst possible severity.
Changes in opioid route, type, and MEDD before and after the POS are summarized in Table 2. Parenteral opioids were prescribed less frequently at baseline by the referring oncology teams after the POS (56 of 314 [18%]) vs before the POS (109 of 311 [35%]) (P < .001). The PC team also prescribed fewer parenteral opioids after the POS (96 of 336 [29%]) vs before the POS (159 of 338 [47%]) (P < .001). The total frequency exceeds the sample size because some patients were prescribed more than 1 route or type of opioid.
Table 2. Opioid Route, Type, and Dose Before and After Parenteral Opioids Shortage.
| Factor | Opioid Prescription Frequency, No. (%) | |||||
|---|---|---|---|---|---|---|
| Baselinea | PC Follow-up Day 1 | |||||
| Before POS | After POS | P Value | Before POS | After POS | P Value | |
| Route of opioidsb | n = 311 | n = 314 | n = 338 | n = 336 | ||
| Parenteral | 109 (35) | 56 (18) | <.001 | 159 (47) | 96 (29) | <.001 |
| Nonparenteral | 202 (65) | 258 (82) | 179 (53) | 240 (71) | ||
| Type of opioidsb | .02 | .01 | ||||
| Morphine | 100 (32) | 91 (29) | 131 (39) | 139 (41) | ||
| Fentanyl | 28 (9) | 34 (10) | 46 (14) | 45 (13) | ||
| Oxycodone | 34 (11) | 54 (17) | 19 (6) | 39 (12) | ||
| Hydromorphone | 89 (29) | 67 (21) | 108 (32) | 69 (20) | ||
| Hydrocodone | 22 (7) | 29 (9) | 9 (2) | 12 (4) | ||
| Tramadol | 26 (8) | 17 (5) | 11 (3) | 12 (4) | ||
| Methadone | 12 (4) | 20 (6) | 14 (4) | 20 (6) | ||
| Codeine | 0 (0) | 2 (1) | 0 (0) | 0 (0) | ||
| MEDD, median (IQR) | 45 (20-105) | 60 (30-120) | .04 | 60 (25-105) | 60 (30-128) | .08 |
Abbreviations: IQR, interquartile range; MEDD, morphine equivalent daily dose (mg/day); PC, palliative care; POS, parenteral opioid shortage.
Baseline result reflects the prescription patterns of the referring oncology teams.
The total frequency exceeds the sample size since some patients were prescribed more than 1 route or type of opioid.
After the POS, the type of opioids prescribed by the referring oncology teams changed significantly, with less use of morphine and hydromorphone and more use of oxycodone and hydrocodone (Table 2). Opioids prescribed by the PC team before and after the POS also changed significantly on follow-up day 1 (Table 2).
After the POS began, the median MEDD prescribed by the referring oncology team was significantly higher after the POS (60 mg/d; IQR, 30-120 mg/d) vs before the POS (45 mg/d; IQR, 20-105 mg/d) (P = .04). The MEDD prescribed by the PC team before and after the POS did not change significantly (Table 2).
After the POS began, significantly fewer patients achieved CIP on follow-up day 1 (119 of 193 [62%] after the POS vs 144 of 193 [75%] before the POS; P = .01). The results of univariate and multivariate analyses of risk factors of CIP on follow-up day 1 are reported in Table 3. After adjusting for age (odds ratio [OR], 1.02; 95% CI, 1.01-1.04; P = .01), patients treated before the POS had an 89% higher chance of achieving CIP than those treated after the POS (OR, 1.89; 95% CI, 1.22-2.94; P = .005).
Table 3. Univariate and Multivariate Analysis of Risk Factors of Clinically Improved Pain at Follow-up Day 1.
| Variable | Univariatea | Multivariateb | ||
|---|---|---|---|---|
| OR (95% CI) | P Value | OR (95% CI) | P Value | |
| Time (before POS vs after POS) | 1.83 (1.18-2.82) | .01 | 1.89 (1.22-2.94) | .005 |
| Agec | 1.02 (1.01-1.04) | .01 | 1.02 (1.01-1.04) | .01 |
| IV routed | 1.78 (1.09-2.92) | .02 | NA | NA |
| MEDDe | 0.997 (0.995-0.999) | .01 | NA | NA |
Abbreviations: IV, intravenous; MEDD, morphine equivalent daily dose (mg/d); NA, not assessed; OR, odds ratio; POS, parenteral opioid shortage.
Univariate logistic regression analysis was used to investigate 12 variables as potential risk factors of pain improvement on follow-up day 1. Variables included age; sex; race; marital status; time of POS; ECOG (Eastern Cooperative Oncology Group performance status); CAGE (Cut-down, Annoyed, Guilty, Eye-opener questionnaire for alcoholism); ESAS (Edmonton System Assessment Scale) all items; tobacco use; and route, type, and dose of opioids at baseline.
Multivariate analysis was performed by using the stepwise selection method.
Compared with 1 unit increase.
Compared with oral vs transdermal route at follow-up day 1.
Compared with 1 unit mg/d increase.
Discussion
Our findings show that the POS was associated with fewer parenteral opioids prescribed by both primary oncology and PC teams. After the POS, a lower proportion of patients achieved CIP at PC follow-up day 1. To our knowledge, this is the first study to show the implications of the recent POS for the frequency of parenteral opioid prescriptions and CIP achievement among hospitalized patients.
After the POS, primary oncology and PC teams relied more on the use of nonparenteral (ER, TD, and IR) opioids (Table 2), which was probably an attempt to compensate for the shortage of parenteral opioids. These findings are concerning because patients with cancer who experience severe pain often require aggressive opioid titration, which is not feasible with ER or TD preparations.5 The ER or TD preparations take several days to achieve an effective analgesic dose, which can result in delayed CIP.5 For those patients with severe pain, the parenteral route has been proven to be a fast, reliable, and safe mode for opioid dose titration.24,25,26
After the POS, the primary oncology teams prescribed significantly higher MEDDs, which suggests that even higher doses of nonparenteral opioids were not sufficient to treat severe pain. Oral opioid formulations may have highly variable bioavailability in patients with cancer who have persistent nausea and vomiting or those with certain gastrointestinal issues such as bowel obstruction, malabsorption syndrome, or cachexia. In such patients, even the TD route is not preferred to treat acute cancer pain owing to its slow absorption and risk for potential toxic effects if the dose is titrated before the peak levels are achieved. More research is needed to fully understand this issue.
Our findings suggest that the approach of both primary oncology and PC teams was not sufficiently aggressive in terms of opioid management on the first day of their involvement. At baseline, the median ESAS pain score, which reflects the primary oncology team interventions, was 7 in both cohorts. These findings suggest that parenteral opioids were not aggressively titrated, which was associated with less CIP achievement.
Univariate analysis of risk factors of CIP on follow-up day 1 indicated that CIP achievement was significantly associated with the time of the POS, MEDD, opioid route, and older age. Multivariate analysis confirmed that the POS was independently associated with CIP on follow-up day 1. These findings strongly suggest that the POS is posing challenges for both the primary oncology and PC teams. Unnecessary patient suffering and delayed hospital discharge may both be unintended consequences associated with the POS.
Limitations
Despite a robust data set, there were several limitations to this study. First, the data were from a single institution where dedicated PC services are available. Therefore, these findings cannot be generalized to community-based programs with no access to PC. Second, owing to the retrospective nature of the study, we had to rely on the clinician’s documentation to define CIP. However, the PC team members who assessed CIP were not biased at the moment of their assessment because they were not aware of the study design. Therefore, in our opinion, the study has a qualitative value. Finally, in this retrospective study, we were unable to assess patients’ response to cancer treatment, which could have resulted in different symptom burden and the need for opioid therapy. It is reassuring that the demographic and clinical variables, including ESAS pain and other symptoms, CAGE positivity, and history of tobacco use, were not different between the 2 cohorts, which suggests that both cohorts were quite similar. Also, the fact that our PC team structure and the procedures for drug administration did not change and that we followed a consistent clinical care pathway suggest that the POS was a significant reason for the observed differences.
Conclusions
Our findings suggest that more aggressive titration with nonparenteral opioids and more frequent assessment of pain intensity, especially during the first few days of hospitalization might result in better pain control. More studies should be conducted on the techniques to achieve this in a way that is logistically feasible. Certain alternative drugs, such as nonopioid analgesics (cannabinoids, gabapentin, and ketamine) and less common parenteral opioids (nalbuphine and buprenorphine), might be used to treat cancer pain during the POS. However, their use is limited owing to lack of evidence in cancer pain, complex opioid rotations, and risk of toxic effects.14,15,27 The main priority is to ensure the availability of parenteral opioids for hospitalized patients with cancer.
There has been a significant change in opioid routes of administration associated with the POS. Furthermore, after the POS, fewer patients in this study achieved CIP. These factors have potential implications for patient satisfaction and hospital length of stay.
References
- 1.van den Beuken-van Everdingen MH, Hochstenbach LM, Joosten EA, Tjan-Heijnen VC, Janssen DJ. Update on prevalence of pain in patients with cancer: systematic review and meta-analysis. J Pain Symptom Manage. 2016;51(6):1070-1090.e9. doi: 10.1016/j.jpainsymman.2015.12.340 [DOI] [PubMed] [Google Scholar]
- 2.Koyyalagunta D, Bruera E, Solanki DR, et al. . A systematic review of randomized trials on the effectiveness of opioids for cancer pain. Pain Physician. 2012;15(3)(suppl):ES39-ES58. [PubMed] [Google Scholar]
- 3.Jennings AL, Davies AN, Higgins JP, Gibbs JS, Broadley KE. A systematic review of the use of opioids in the management of dyspnoea. Thorax. 2002;57(11):939-944. doi: 10.1136/thorax.57.11.939 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Bruera E, Macmillan K, Pither J, MacDonald RN. Effects of morphine on the dyspnea of terminal cancer patients. J Pain Symptom Manage. 1990;5(6):341-344. doi: 10.1016/0885-3924(90)90027-H [DOI] [PubMed] [Google Scholar]
- 5.Mercadante S. Parenteral routes for opioid administration in cancer pain management. Eur J Intern Med. 2016;34:e39-e40. doi: 10.1016/j.ejim.2016.06.030 [DOI] [PubMed] [Google Scholar]
- 6.Bandieri E, Romero M, Ripamonti CI, et al. . Randomized trial of low-dose morphine versus weak opioids in moderate cancer pain. J Clin Oncol. 2016;34(5):436-442. doi: 10.1200/JCO.2015.61.0733 [DOI] [PubMed] [Google Scholar]
- 7.Stjernswärd J. WHO cancer pain relief programme. Cancer Surv. 1988;7(1):195-208. [PubMed] [Google Scholar]
- 8.National Comprehensive Cancer Network NCCN clinical practice guidelines in oncology—adult cancer pain 2018; https://www.nccn.org/professionals/physician_gls/pdf/pain.pdf. Accessed May 4, 2018.
- 9.Paice JA, Lacchetti C, Bruera E. Management of chronic pain in survivors of adult cancers: ASCO clinical practice guideline summary. J Oncol Pract. 2016;12(8):757-762. doi: 10.1200/JOP.2016.014837 [DOI] [Google Scholar]
- 10.Bennett M, Paice JA, Wallace M. Pain and opioids in cancer care: benefits, risks, and alternatives. Am Soc Clin Oncol Educ Book. 2017;37:705-713. doi: 10.14694/EDBK_180469 [DOI] [PubMed] [Google Scholar]
- 11.Te Boveldt N, Vernooij-Dassen M, Burger N, Ijsseldijk M, Vissers K, Engels Y. Pain and its interference with daily activities in medical oncology outpatients. Pain Physician. 2013;16(4):379-389. [PubMed] [Google Scholar]
- 12.Kroenke K, Theobald D, Wu J, Loza JK, Carpenter JS, Tu W. The association of depression and pain with health-related quality of life, disability, and health care use in cancer patients. J Pain Symptom Manage. 2010;40(3):327-341. doi: 10.1016/j.jpainsymman.2009.12.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Goldberg GR, Morrison RS. Pain management in hospitalized cancer patients: a systematic review. J Clin Oncol. 2007;25(13):1792-1801. doi: 10.1200/JCO.2006.07.9038 [DOI] [PubMed] [Google Scholar]
- 14.Bruera E. Parenteral opioid shortage—treating pain during the opioid-overdose epidemic. N Engl J Med. 2018;379(7):601-603. doi: 10.1056/NEJMp1807117 [DOI] [PubMed] [Google Scholar]
- 15.Hollingsworth H, Herndon C. The parenteral opioid shortage: causes and solutions. J Opioid Manag. 2018;14(2):81-82. doi: 10.5055/jom.2018.0434 [DOI] [PubMed] [Google Scholar]
- 16.US Food & Drug Administration Current and resolved drug shortages and discontinuations reported to FDA. https://www.accessdata.fda.gov/scripts/drugshortages/default.cfm#M. Accessed May 5, 2018.
- 17.US Government Accountability Office GAO-15-202, Drug shortages, better management of the quota process for controlled substances needed; coordination between DEA and FDA should be improved. February 2015; https://www.gao.gov/assets/670/668252.pdf. Accessed May 15, 2018.
- 18.Woodcock J, Wosinska M. Economic and technological drivers of generic sterile injectable drug shortages. Clin Pharmacol Ther. 2013;93(2):170-176. [DOI] [PubMed] [Google Scholar]
- 19.Kwon JH. Overcoming barriers in cancer pain management. J Clin Oncol. 2014;32(16):1727-1733. doi: 10.1200/JCO.2013.52.4827 [DOI] [PubMed] [Google Scholar]
- 20.Bruera EE. A. The MD Anderson Symptom Control and Palliative Care Handbook. Houston, TX: University of Health Science Center at Houston; 2015. [Google Scholar]
- 21.Oken MM, Creech RH, Tormey DC, et al. . Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5(6):649-655. doi: 10.1097/00000421-198212000-00014 [DOI] [PubMed] [Google Scholar]
- 22.Mayfield D, McLeod G, Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry. 1974;131(10):1121-1123. [DOI] [PubMed] [Google Scholar]
- 23.Bruera E, Kuehn N, Miller MJ, Selmser P, Macmillan K. The Edmonton Symptom Assessment System (ESAS): a simple method for the assessment of palliative care patients. J Palliat Care. 1991;7(2):6-9. [PubMed] [Google Scholar]
- 24.Caraceni A, Hanks G, Kaasa S, et al. ; European Palliative Care Research Collaborative (EPCRC); European Association for Palliative Care (EAPC) . Use of opioid analgesics in the treatment of cancer pain: evidence-based recommendations from the EAPC. Lancet Oncol. 2012;13(2):e58-e68. doi: 10.1016/S1470-2045(12)70040-2 [DOI] [PubMed] [Google Scholar]
- 25.Mercadante S. The use of rapid onset opioids for breakthrough cancer pain: the challenge of its dosing. Crit Rev Oncol Hematol. 2011;80(3):460-465. doi: 10.1016/j.critrevonc.2010.12.002 [DOI] [PubMed] [Google Scholar]
- 26.Mercadante S. Intravenous morphine for management of cancer pain. Lancet Oncol. 2010;11(5):484-489. doi: 10.1016/S1470-2045(09)70350-X [DOI] [PubMed] [Google Scholar]
- 27.Davis MP, McPherson ML, Mehta Z, Behm B, Fernandez C. What parenteral opioids to use in face of shortages of morphine, hydromorphone, and fentanyl. Am J Hosp Palliat Care. 2018;35(8):1118-1122. doi: 10.1177/1049909118771374 [DOI] [PubMed] [Google Scholar]