Abstract
Context
Although off-label medications are frequently prescribed in palliative care, there are no published studies examining their use in the United States.
Objectives
We examined the frequency of off-label medication use in cancer patients admitted to an acute palliative care unit (APCU).
Methods
This prospective observational study enrolled consecutive patients with advanced cancer admitted to the APCU of a tertiary care cancer center. We collected data on all prescription events, including indications for use, from admission to discharge. Off-label use was checked against the U.S. Food and Drug Administration approved indications.
Results
Among the 201 patients, median survival was 10 days (95% confidence interval 7-13), and 85 (42%) patients died in the APCU. We documented 6276 prescription events, and 2199 (35%) were off-label. Among off-label prescriptions, central nervous system agents (n=1606, 73%), hormones and synthetic substitutes (n= 302, 14%), and autonomic drugs (n=183, 8%) were most commonly prescribed. Haloperidol (n=720, 33%), chlorpromazine (n=292, 13%), dexamethasone (n=280, 13%), glycopyrrolate (n=175, 8%), hydromorphone (n=161, 7%), and morphine (n=156, 7%) were most frequently prescribed off-label. The most common indications for off-label prescribing were delirium (n=783, 36%) and dyspnea (n=449, 20%). 70% of all off-label prescription events had strong evidence supporting use, and 19% of prescription events had moderate or weak evidence for use.
Conclusion
One-third of prescription events in the APCU were off-label, with majority of off-label use having a strong level of supporting evidence. Our findings highlight the need for more research in key areas such as delirium and dyspnea management.
Keywords: Drug Prescriptions, Neoplasms, Off-Label Use, Palliative Care, Therapeutics, Unlabeled Indication
Introduction
In the United States (U.S.), the Food and Drug Administration (FDA) determines the official product labeling for approved or “on-label” indications, while use of medications for non-FDA approved indications or in unapproved dosages or forms are considered “off-label”. While FDA-approved indications are part of the official product labeling for marketing, on-label indications do not necessarily dictate appropriate or wrongful medication use. In fact, off-label prescription use is both prevalent and legal. One U.S. study of general outpatient practices found that 21% of prescribed medications or an estimated 150 million prescriptions were off-label (1), while another study in Quebec, Canada found off-label use to be 11.8% of all prescriptions (2). Physicians cite the absence of medications intended for a particular indication as being the primary reason for off-label prescribing (3).
Despite being common, off-label use does not come without concerns. Off-label use has been studied as a potential contributing factor to preventable adverse drug events (ADEs) (4-7), with the risk for ADEs increased by 44% with off-label use compared to on-label use. This risk was even greater when off-label use was not supported by strong evidence, with off-label use lacking strong evidence found to be 80.9% out of all off-label use prescriptions (4). These findings highlight the importance of reducing indiscriminate or harmful off-label use, perhaps through research investigating safe and effective medication use or development of new drugs and formulations that lead to more available “on-label” rather than “off-label” treatments.
Off-label prescribing in palliative care signifies the lack of FDA-approved medications intended for symptom control, thus limiting optimal patient care. Palliation is not an attractive market for medication research, as seen with most clinical drug trials being focused on disease treatment rather than symptom management. The difficulty of conducting clinical trials ethically in a fragile population with a rapidly fluctuating or deteriorating clinical status creates another barrier to research, leading to fewer clinical advances (8). Furthermore, reimbursement policies based only on FDA-approved indications can hinder the delivery of quality palliative care.
Despite these challenges, palliative care physicians should strive to provide effective and evidence-based medical therapies supported by high quality research. A better understanding of patterns of off-label medication use in the palliative care setting may help to identify areas that require further research for on-label medication development. Examining off-label medication use in relation to the supporting evidence may also facilitate the expansion of FDA-approved indications of medications commonly used for symptom management.
There have been very few published studies investigating the frequency of off-label medication use in patients in the palliative care unit or in hospice. A cross sectional study with 507 hospice patients in Italy reported 4.5% of all medications used were off-label, with common indications being control of secretions, dyspnea, anorexia, and fatigue (10). A prospective study in a single palliative care center in the United Kingdom (U.K.) showed that 15% of 689 prescribing events were off-label, with pain, agitation, insomnia, and anxiety being the most frequently cited indications for off-label use (11). Currently, there are no published studies investigating the frequency of off-label use in palliative care centers or hospices in the U.S. for comparison.
We conducted this study to examine the frequency of off-label medication use in cancer patients admitted to an acute palliative care unit (APCU). We also investigated off-label use according to the strength of evidence supporting their use for particular indications. An understanding of how off-label medications are used in palliative care can highlight treatment areas that need more focused drug research or advances in medication development.
Methods
Study setting and criteria
This was a prospective observational study involving consecutive enrollment of patients with active cancer treated in the APCU at MD Anderson Cancer Center (MDACC), with inclusion criteria being admission or transfer to the APCU between December 17, 2012 and April 22, 2013. The APCU is a closed unit, with patients receiving care from an interdisciplinary team consisting of board-certified palliative care physicians, rotating fellows and residents, and nurse practitioners, all of whom have prescribing ability. Information on APCU clinician characteristics is available from published literature (11). The unit also has a dedicated pharmacist to offer guidance on dosing and forms of medication administration.
The project protocol was submitted to the institutional review board in November 2012 and was approved. There was no direct patient interaction involved in this study, and a waiver of informed consent was obtained. The study did require written consent from treating physicians whose documentation in the medical chart was used as part of the data collection.
Data Collection and Classification
For each patient, demographic information, such as the patient's age, gender, race, and primary tumor, were collected from the medical record. Prescription medication data was also obtained from the date of admission to the date of discharge, with recording and monitoring of any newly prescribed medications during the inpatient stay. Every medication order, either initiation or modification, found in the chart was recorded as a discrete “prescription event”. Other collected data included medication dosage, frequency, route of administration, and indication for use from the prescriber. No other events, such as adverse events or drug interactions, were collected. Patients in the study had only one inpatient stay for which medication data was collected during the specified time period. For patients who were admitted multiple times, data from their first inpatient stay was collected.
All medications that were scheduled and prescribed as needed were included in the study, and a scheduled prescription was considered a separate prescribing event from an as needed prescription of the same medication. If a medication was prescribed by multiple different routes, then each route was considered a separate prescribing event, and if a medication was prescribed multiple different dosages, then each dosage was also considered a separate prescribing event. For example, 0.5 mg intravenous morphine every hour was recorded as a separate prescription from 0.2 mg intravenous morphine every hour. A medication that was prescribed for two different indications was also recorded as two different prescriptions.
To facilitate analysis, the American Hospital Formulary Service (AHFS) classification system, developed in 1959 for the organization of medications and development of formularies in both institutional and governmental settings (12), was used to group all medication prescriptions into categories that reflect similar pharmacologic and therapeutic characteristics. For example, albuterol, glycopyrrolate, and scopolamine were categorized into the autonomic drug classification. Hormones and synthetic substitutes were primarily steroid medications such as dexamethasone. Central nervous system agents included analgesics, antipyretics, antidepressants, anticonvulsants, and antipsychotic medications.
Determination of indication for use and on-label versus off-label use
The physician in charge of each patient's care was asked to prospectively document the indication(s) for which medications were prescribed within three days of the prescription event. MICROMEDEX® 2.0 (Truven Health Analytics, Inc. Ann Arbor, MI) was used as the standard reference for clinical indications and dosing approved by the FDA. We defined a medication as off-label use if the indication provided by the physician was different from the approved indication. If a medication was used for more than one indication (e.g. morphine for pain and dyspnea), we counted each indication as an individual prescription event in order to properly document the frequency of off-label use.
Determination of level of evidence for off-label use
Each prescription event was categorized into seven different groups according to licensed, or on-label, versus unlicensed, or off-label, use and the level of evidence supporting unlicensed use; a modified version of Ferner's classification for off-label medications was used (Table 1) (13). We classified the level of evidence for off-label use into high (e.g. randomized controlled trials and systematic reviews), moderate (e.g. prospective phase II trials, prospective case series, and retrospective controlled studies) and weak (e.g. retrospective case series or case reports). The MD Anderson Symptom Control and Palliative Care Handbook, intended for clinicians to be a general guide for standard palliative care practices, was also used as a resource. Developed, written, and updated every 1-5 years by expert palliative care physicians at MDACC, the handbook contains information on common palliative care issues and treatments with an aggregate of research citations.
Table 1. Categories Based on Licensed versus Off-Label Evidence for Use.
|
For example, a 5HT3 antagonist is licensed for chemotherapy-induced nausea and vomiting, not for nausea in the terminal cancer patient
Data analysis
Descriptive statistics including frequency, percentage, mean, median, minimum, maximum and interquartile ranges were used to summarize the data. Kaplan-Meier survival analysis was used to calculate the median survival time from APCU admission and its 95% confidence intervals.
Results
Study population characteristics
A total of 201 patients were enrolled in the study. The median number of days in the APCU was 6 (IQR 4-8). 42% of patients died during the APCU admission. Median survival time from APCU admission was 10 days (95% confidence interval 7-13). Median total number of new prescription events starting from admission and throughout the entire course of the APCU stay was 29 (IQR 23-37). Additional demographic information is included in Table 2.
Table 2. Clinical characteristics of patients.
| Characteristics | Sub-categories | N (%) |
|---|---|---|
| Age, mean (range); median (IQR) | 58 (20, 86); 60 (49, 67) | |
| Sex | Female | 100 (50) |
| Race | White | 116 (58) |
| Black | 39 (19) | |
| Hispanic | 31 (15) | |
| Other | 15 (7) | |
| Cancer class | Breast | 25 (12) |
| Gastrointestinal | 46 (23) | |
| Genitourinary | 24 (12) | |
| Gynecologic | 21 (10) | |
| Head and neck | 5 (2) | |
| Hematologic | 25 (12) | |
| Respiratory | 34 (17) | |
| Other | 21 (10) | |
| Died in Hospital | 85 (42) | |
| Number of APCU days, median (IQR) | 6 (4, 8) | |
| Number of medication orders, median (IQR) | 29 (23, 37) |
N = number of patients
Off-label medication use by AHFS classification
Prescribers' indications were collected for all prescription events. Among 6276 total prescription events, 2199 (35%) were for off-label indications. There was an average of 11 off-label prescribing events per patient during one APCU admission. All medications were classified into AHFS classification groups for further analysis, as seen in Table 3. Central nervous system agents (51%), gastrointestinal drugs (18%), autonomic drugs (10%), hormones and synthetic substitutes (7%) were the most commonly prescribed. AHFS classification groups that were most commonly prescribed off-label were central nervous system agents (73%), hormones and synthetic substitutes (14%), and autonomic drugs (8%). Of these three AHFS classification groups, autonomic drugs had the highest proportion of prescription events supported by strong evidence for off-label use (95%), followed by central nervous system agents (73%) and hormones and synthetic agents (52%) (Table 3).
Table 3. Frequency of off-label use by AHFS therapeutic class.
| AHFS classification | Total prescription events N (%)1 | Off-label prescription events N (%)2 | Off-label use with strong evidence N (%)3 | Off-label use with moderate or weak evidence N (%)3 | Off-label use without evidence or contraindicated N (%)3 |
|---|---|---|---|---|---|
| Central nervous system agents | 3179 (51) | 1606 (73) | 1171 (73) | 358 (22) | 77 (5) |
| Hormones and synthetic substitutes | 467 (7) | 302 (14) | 156 (52) | 53 (18) | 93 (30) |
| Autonomic drugs | 635 (10) | 183 (8) | 174 (95) | 9 (5) | |
| Gastrointestinal drugs | 1159 (18) | 43 (2) | 3 (7) | 40 (93) | |
| Cardiovascular drugs | 192 (3) | 34 (2) | 25 (74) | 1 (3) | 8 (23) |
| Miscellaneous therapeutic agents | 33 (1) | 17 (1) | 5 (30) | 12 (70) | |
| Respiratory tract agents | 43 (1) | 5 (<1) | 1 (20) | 1 (20) | 3 (60) |
| EENT preparations | 6 (0) | 4 (<1) | 2 (50) | 2 (50) | |
| Anti-infective agents | 285 (5) | 2 (<1) | 2 (100) | ||
| Electrolytic, caloric, water balance | 111 (2) | 2 (<1) | 1 (50) | 1 (50) | |
| Local anesthetics | 5 (0) | 1 (<1) | 1 (100) | 0 | |
| Blood formation, coagulation, thrombosis | 93 (1) | 0 | 0 | ||
| Skin and mucous membrane agents | 40 (1) | 0 | 0 | ||
| Antihistamine drugs | 15 (<1) | 0 | 0 | ||
| Vitamins | 7 (<1) | 0 | 0 | ||
| Antineoplastic agents | 5 (<1) | 0 | 0 | ||
| Smooth muscle relaxants | 1 (<1) | 0 | 0 |
Abbreviation: EENT, Eye, Ear, Nose, and Throat
N = Number of prescription events
Percentage of prescription events out of all prescription events
Percentage of off-label use out of all off-label prescription events
Percentage of off-label use out of all off-label prescription events that specific AHFS class, according to associated level of evidence
Off-label medication use by frequency and level of evidence
Out of the off-label medications prescribed, 70% of prescription events had strong evidence for use, and 19% of prescription events had moderate or weak evidence for use. 4% of all prescription events or 11% of all off-label prescription events had no literature support for use or were contraindicated.
Haloperidol (33%), chlorpromazine (13%), dexamethasone (13%), glycopyrrolate (8%), hydromorphone (7%), and morphine (7%) were the most frequently prescribed medications out of all off-label prescription events. Of these medications, glycopyrrolate, haloperidol, and chlorpromazine were most consistently prescribed for primarily off-label use. The medication most frequently used off-label, haloperidol, was prescribed 72% of the time with a strong level of evidence to treat delirium and insomnia and with moderate level evidence for anxiety and nausea and vomiting. The second most frequently prescribed off-label medication, chlorpromazine, was prescribed 78% of the time with a strong level of evidence to treat delirium; it was used to treat anxiety and insomnia with a moderate or weak level of evidence. Dexamethasone was prescribed 70% of the time with a strong level of evidence, most often to treat pain but also used to treat nausea or vomiting, appetite, and spinal cord compression. Dexamethasone was used with a moderate level of evidence to treat bowel obstruction. Glycopyrrolate was used to treat excessive secretions with what was considered a strong level of evidence, while dyspnea was treated with a strong level of evidence with morphine and hydromorphone (Table 4).
Table 4. Frequency of off-label use by drug and associated level of scientific evidence.
| Drug Name | Total prescription events N (%)1 | Off-label prescription events N (%)2 | Off-label use with strong evidence N (%)3 | Off-label use with moderate or weak evidence N (%)3 | Off-label use without evidence or contraindicated N (%)3 |
|---|---|---|---|---|---|
| Haloperidol | 723 (12) | 720 (33) | 520 (72) | 187 (26) | 13 (2) |
| Chlorpromazine | 305 (5) | 292 (13) | 229 (78) | 60 (21) | 3 (1) |
| Dexamethasone | 337 (5) | 280 (13) | 195 (70) | 13 (5) | 72 (25) |
| Glycopyrrolate | 175 (3) | 175 (8) | 170 (97) | 5 (3) | |
| Hydromorphone | 670 (11) | 161 (7) | 149 (93) | 12 (7) | |
| Morphine | 518 (8) | 156 (7) | 156 (100) | 0 | |
| Lorazepam | 350 (6) | 71 (3) | 1 (1) | 39 (55) | 31 (44) |
| Fentanyl | 182 (3) | 48 (2) | 47 (99) | 1 (1) | |
| Olanzapine | 47 (1) | 47 (2) | 46 (98) | 1 (2) | |
| Gabapentin | 33 (1) | 33 (2) | 29 (88) | 4 (12) | |
| Metoclopramide | 252 (4) | 31 (1) | 3 (10) | 28 (90) | |
| Metoprolol | 74 (1) | 22 (1) | 15 (68) | 7 (32) | |
| Ipratropium Bromide | 212 (3) | 2 (<1) | 2 (<1) | ||
| Pantoprazole | 201 (3) | 1 (<1) | 1 (<1) | ||
| Levalbuterol | 200 (3) | 1 (<1) | 1 (<1) | ||
| Bisacodyl | 164 (3) | 1 (<1) | 1 (<1) | ||
| Acetaminophen | 141 (2) | 1 (<1) | 1 (<1) | ||
| Milk and molasses enema | 84 (1) | 1 (<1) | 1 (<1) | ||
| Ondansetron | 46 (1) | 1 (<1) | 1 (<1) | ||
| Polyethylene glycol | 37 (1) | 1 (<1) | 1 (<1) | ||
| Sennosides | 184 (3) | 0 | 0 | ||
| Lactulose | 73 (1) | 0 | 0 | ||
| Enoxaparin | 55 (1) | 0 | 0 | ||
| Piperacillin/Tazobactam | 46 (1) | 0 | 0 | ||
| Aluminum/Magnesium Hydroxide/Simethicone | 37 (1) | 0 | 0 |
N=Number of prescription events
Percentage of medication prescriptions out of total 6272 prescription events. Medications (N=1126, 18%) having less than 30 total prescription events were excluded from this table.
Percentage of off-label use of specific medications out of all 2199 off-label prescription events. Medications (N=154, 7%) having less than 30 total prescription events were excluded from this table.
Percentage of off-label use out of all off-label prescription events for that specific medication, according to associated level of evidence.
Indications for off-label medication use
The greatest proportion of medications were prescribed for pain (19%), dyspnea (14%), delirium (12%), constipation (10%), nausea and vomiting (7%). Medications were most commonly prescribed for off-label use to treat delirium (36%) and dyspnea (20%). Delirium and excessive secretions were treated exclusively by medications for off-label use (Table 5).
Table 5. Off-label drugs used for each treatment indication.
| Treatment indications | Total prescriptions N (%)1 | Off-label prescription events N (%)2 | Most common off-label drug (N) | 2nd most common off- label drug (N) | 3rd most common off-label drug (N) | 4th most common off-label drug (N) |
|---|---|---|---|---|---|---|
| Delirium | 783 (12) | 783 (36) | Haloperidol (463) | Chlorpromazine (229) | Lorazepam (33) | Midazolam (19) |
| Dyspnea | 888 (14) | 449 (20) | Morphine (155) | Hydromorphone (149) | Dexamethasone (66) | Fentanyl (47) |
| Excessive Secretions | 177 (3) | 177 (8) | Glycopyrrolate (168) | Acetylcysteine (2) Guaifenesin (2) Haloperidol (2) | Octreotide (1) Polyethylene glycol (1) Scopolamine (1) | |
| Anxiety | 309 (5) | 168 (8) | Haloperidol (89) | Chlorpromazine (44) | Olanzapine (28) | Hydromorphone (2) Metoclopramide (2) |
| Pain | 1216 (19) | 165 (8) | Dexamethasone (154) | Gabapentin (4) | Glycopyrrolate (2) Haloperidol (2) | Chlorpromazine (1) Methylprednisolone (1) Pregabalin (1) |
| Nausea and Vomiting | 415 (7) | 137 (6) | Haloperidol (97) | Dexamethasone (22) | Octreotide (10) | Hydromorphone (2) Polyethylene glycol (2) |
| Insomnia | 98 (2) | 89 (4) | Haloperidol (57) | Chlorpromazine (6) | Mirtazapine (5) | Olanzapine (4) |
| Neuropathic pain | 43 (1) | 43 (2) | Gabapentin (29) | Pregabalin (11) | Amitriptyline (3) | |
| Bleeding | 35 (1) | 29 (1) | Lorazepam (26) | Glycopyrrolate (1) Haloperidol (1) Moxifloxacin (1) | ||
| Fever | 153 (2) | 16 (1) | Hydrocortisone (12) | Dexamethasone (2) | Hydrocodone/ Acetaminophen (1) Methylprednisolone (1) | |
| Cough | 41 (1) | 7 (<1) | Prednisone (3) | Hydromorphone (2) | Lidocaine (1) Morphine (1) | |
| Constipation | 630 (10) | 6 (<1) | Metoclopramide (3) | Haloperidol (1) Octreotide (1) Levalbuterol (1) | ||
| Hypertension | 143 (2) | 3 (<1) | Isosorbide mononitrate (2) | Furosemide (1) | ||
| Infection | 301 (5) | 2 (<1) | Hydrocortisone (1) Octreotide (1) | |||
| Seizures | 158 (3) | 2 (<1) | Metoclopramide (1) Olanzapine (1) | |||
| Dyspepsia | 86 (1) | 1 (<1) | Ipratropium Bromide (1) | |||
| Inflammation | 60 (1) | 1 (<1) | Hydromorphone (1) | |||
| GI prophylaxis | 188 (3) | 0 | ||||
| DVT prophylaxis | 45 (1) | 0 | ||||
| Anticoagulation | 43 (1) | 0 |
Abbreviation: DVT, deep vein thrombosis; GI, gastrointestinal
N=Number of prescription events
Percentage of medication prescriptions out of total 6272 prescription events. 464 (7%) medications had less than 40 total prescription events were excluded from this table.
Percentage of off-label use of specific medications out of all 2199 off-label prescription events. 123 (6%) medications had less than 40 total prescription events were excluded from this table.
Discussion
This study shows that one-third of the prescription events in our APCU were off-label, with the majority of off-label use supported by a strong level of evidence per study definition. Central nervous system agents (e.g. haloperidol, chlorpromazine), hormones and synthetic substitutes (e.g. dexamethasone), and autonomic agents (e.g. glycopyrrolate) were the most commonly prescribed off-label medications. Delirium and dyspnea were the most common indications for off-label use and represent key areas for further research at the end-of-life.
Despite our study's focus on off-label medication use, it is important to note that a majority of prescriptions were on-label. Specifically, opioids, benzodiazepines, bronchodilators, metoclopramide and laxatives, and H2 receptor blockers were used for pain, seizures, dyspnea, nausea and vomiting, constipation, and gastrointestinal prophylaxis according to their FDA-approved indications, respectively. Our findings are consistent with other studies demonstrating that medication use for constipation, nausea and vomiting, and dyspnea are often on-label (10).
Interestingly, a considerably higher percentage of off-label use was seen in our study compared to the prevalence of 5-15% reported in other European studies (9, 10). This discrepancy may be explained by differences in patient populations, treatment preferences, and local regulatory approval processes among the studies. In a U.K. study, some medications used for agitation, such as diazepam and haloperidol, were considered to be licensed for use (10), while an Italian hospice study did not mention off-label use for agitation or delirium at all (9). In addition, higher numbers could have been driven by the fact that our study was conducted in an APCU having high incidence of delirium and dyspnea, both of which are indications with few on-label medications for treatment in the U.S. Reassuringly, we found that a vast majority of off-label medications use was based on available evidence, although 4% of all prescription events were off-label medications used without evidence or with contraindication. We suspect that the majority of these incidents were due to imprecise wording by prescribers when describing indications for use rather than imprudent prescribing. However, uninformed off-label prescribing would be a concern that warrants further study, especially due to risk for adverse effects.
Notably, nearly 75% of off-label prescription events were central nervous system agents, with over one-third being treatment for delirium. Two neuroleptics, haloperidol and chlorpromazine, were most commonly used for delirium in our study population. Haloperidol has often been considered the medication of choice for delirium, with its use being the most studied (14, 15), although chlorpromazine and the newer neuroleptics are also useful for agitated delirium (16-18). There is still insufficient evidence to support neuroleptics and benzodiazepines over placebo in treating delirium and agitation in the palliative care setting (17, 18), with one recent study suggesting that haloperidol and risperidone were associated with an unfavorable risk-benefit ratio compared to placebo (19). Adequately powered double-blind placebo-controlled trials are clearly needed to support pharmacologic interventions for this devastating syndrome.
Dyspnea was the second most common indication for off-label use, with opioids (morphine, hydromorphone, fentanyl) and corticosteroids being the main medications prescribed. While the benefits of opioids in treating dyspnea have been noted in a number of studies, a Cochrane review of 26 randomized controlled trials concluded that there was only low-quality evidence supporting the use of opioids for dyspnea; increased risk for adverse events such as constipation, nausea and vomiting, and drowsiness were also found (20-22). This observation is concerning given the high frequency of off-label use of opioids for dyspnea in our palliative care unit. Corticosteroids represent another therapeutic option that is even less studied than opioids but shows some promise in relieving dyspnea based on a recent double-blind, randomized controlled trial (23). Taken together, our study highlights the prevalence of off-label use of these two major classes of medications and the opportunities for further research in dyspnea management.
Excessive secretions (or death rattle) was another common indication for off-label use, with glycopyrrolate being the primary agent used in our study. Past studies differ in their support of glycopyrrolate over other anticholinergic medications such as scopolamine or atropine in reducing secretions (24-27). The largest multicenter study on death rattle treatment is a randomized prospective trial that found no difference in effectiveness among three anticholinergics – scopolamine, atropine, and hyoscine butylbromide (28). While the usual adverse effects related to anticholinergics have not been reported in death rattle treatment due to the patients' unconscious state when used, there have been insufficient evidence to conclude a benefit of either pharmacological or non-pharmacological interventions over placebo according to a Cochrane review of four randomized controlled studies (29). Until properly designed placebo-controlled trials are available to confirm the efficacy of these agents, palliative care physicians have no definitive interventions to address these excessive secretions.
To our knowledge, this is the first study to examine off-label use in a palliative care unit in the United States. We enrolled a large cohort of consecutive patients and documented all medications systematically and all indications prospectively. However, a limitation of this study is that it was conducted at a single APCU at a tertiary care cancer center, which may limit its generalizability. Imprecise prescriber wording for indications of medication use, such as benzodiazepines for “bleeding” when it was likely for distress related to bleeding, and some overestimation of off-label use due to repeated prescription events for the same indication, especially opioids for dyspnea, may have affected the study results to some degree. Also, the number of prescribers, prescriber characteristics and knowledge of off-label drug use, and institutional formularies, all of which could affect off-label prescribing, were not examined in this study but would be relevant topics for further investigation. Future studies should also examine off-label use in other palliative care settings, such as outpatient clinics and home hospice.
Compared to other fields of medicine, palliative care is relatively unique because [1] patients are often in acute distress necessitating a trial of medications even if the evidence is limited; [2] with symptom management being the primary goal, many agents such as opioids and corticosteroids are easily available and considered to be effective; and [3] it is difficult to conduct randomized trials in a frail population. These may explain the high prevalence of off-label use. In reality, even if high quality trials could confirm the benefits of existing off-label medications, there may be a lack of financial incentive to undergo the arduous approval process for FDA label expansion for many of these older medications that are already generic. As a result, off-label medication use in palliative care setting is likely to remain high.
Monitoring medication appropriateness can be challenging in the setting of off-label prescribing (30). Limiting off-label use with weak supporting evidence can be one potential safeguard against inappropriate prescribing at the end of life. Another strategy would be to fill the gap of evidence for off-label use with more research. Meanwhile, physicians must still strive for safe and effective prescribing by being conscientious and well-informed, using literature evidence if possible, to support off-label prescribing.
This study highlights several areas in end-of-life care that would benefit from further research of off-label use of medications and their effectiveness, such as larger randomized controlled trials. Notably, the most common indications for off-label use in our study – delirium, dyspnea, and excessive secretions – are unfortunately also the most distressing symptoms at end-of-life. Additional studies in these areas may not only help to clarify if existing off-label medications are truly superior to placebo but also pave for the way for development of novel interventions to provide evidence-based, patient-centered care.
Acknowledgments
Disclosures: Eduardo Bruera is supported in part by National Institutes of Health grants RO1NRO10162-01A1, RO1CA122292-01, RO1CA124481-01. David Hui is supported in part by an American Cancer Society Mentored Research Scholar Grant in Applied and Clinical Research (MRSG-14-1418-01-CCE), a National Institutes of Health grant (R21CA186000-01A1), and the Andrew Sabin Family Fellowship Award. This study is also supported by the MD Anderson Cancer Center Support Grant (CA016672). The funding sources were not involved in the conduct of the study or development of the submission.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Radley DC, Finkelstein SN, Stafford RS. Off-label prescribing among office-based physicians. Arch Intern Med. 2006;166(9):1021–1026. doi: 10.1001/archinte.166.9.1021. [DOI] [PubMed] [Google Scholar]
- 2.Eguale T, Buckeridge DL, Winslade NE, et al. Drug, patient, and physician characteristics associated with off-label prescribing in primary care. Arch Intern Med. 2012;172(10):781–788. doi: 10.1001/archinternmed.2012.340. [DOI] [PubMed] [Google Scholar]
- 3.Golocorbin Kon S, Ilikovic I, Mikov M. Reasons for and frequency of off-label drug use. Med Pregl. 2015;68(1-2):35–40. doi: 10.2298/mpns1502035g. [DOI] [PubMed] [Google Scholar]
- 4.Diesel G. Review of adverse events following off-label use of medicines. Vet Rec. 2011;168(8):205–207. doi: 10.1136/vr.d1128. [DOI] [PubMed] [Google Scholar]
- 5.Eguale T, Buckeridge DL, Verma A, et al. Association of off-label drug use and adverse drug events in an adult population. JAMA Intern Med. 2016;176(1):55–63. doi: 10.1001/jamainternmed.2015.6058. [DOI] [PubMed] [Google Scholar]
- 6.Good CB, Gellad WF. Off-label drug use and adverse drug events: turning up the heat on off-label prescribing. JAMA Intern Med. 2016;176(1):63–64. doi: 10.1001/jamainternmed.2015.6068. [DOI] [PubMed] [Google Scholar]
- 7.Smithburger PL, Buckley MS, Culver MA, et al. A multicenter evaluation of off-label medication use and associated adverse drug reactions in adult medical ICUs. Crit Care Med. 2015;43(8):1612–1621. doi: 10.1097/CCM.0000000000001022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Toscani F. Prescribing in palliative care: a quest for appropriateness. Palliat Med. 2013;27(4):293–294. doi: 10.1177/0269216313479843. [DOI] [PubMed] [Google Scholar]
- 9.Toscani F, Di Giulio P, Campi R, et al. Off-label prescriptions in Italian hospices: a national survey. J Pain Symptom Manage. 2009;38(3):365–371. doi: 10.1016/j.jpainsymman.2008.11.014. [DOI] [PubMed] [Google Scholar]
- 10.Atkinson CV, Kirkham SR. Unlicensed uses for medication in a palliative care unit. Palliat Med. 1999;13(2):145–152. doi: 10.1191/026921699676057177. [DOI] [PubMed] [Google Scholar]
- 11.Hui D, Kilgore K, Nguyen L, et al. The accuracy of probabilistic versus temporal clinician prediction of survival for patietns with advanced cancer: a preliminary report. Oncologist. 2011;16(11):1642–1648. doi: 10.1634/theoncologist.2011-0173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chase K, Meyer JD, Kelly WN. Total formulary review--the easy way. Hosp Pharm. 1984;19(3):159–162. [PubMed] [Google Scholar]
- 13.Ferner R. Prescribing licensed medicines for unlicensed indications. Prescribers' Journal. 1996;36(2):73–78. [Google Scholar]
- 14.Bruera E, Neumann CM. The uses of psychotropics in symptom management in advanced cancer. Psychooncology. 1998;7(4):346–358. doi: 10.1002/(SICI)1099-1611(199807/08)7:4<346::AID-PON364>3.0.CO;2-8. [DOI] [PubMed] [Google Scholar]
- 15.Lacasse H, Perreault MM, Williamson DR. Systematic review of antipsychotics for the treatment of hospital-associated delirium in medically or surgically ill patients. Ann Pharmacother. 2006;40(11):1966–1973. doi: 10.1345/aph.1H241. [DOI] [PubMed] [Google Scholar]
- 16.Breitbart W, Strout D. Delirium in the terminally ill. Clin Geriatr Med. 2000;16(2):357–372. doi: 10.1016/s0749-0690(05)70061-6. [DOI] [PubMed] [Google Scholar]
- 17.Candy B, Jackson KC, Jones L, et al. Drug therapy for delirium in terminally ill adult patients. Cochrane Database Syst Rev. 2012;11:Cd004770. doi: 10.1002/14651858.CD004770.pub2. [DOI] [PubMed] [Google Scholar]
- 18.Kishi T, Hirota T, Matsunaga S, Iwata N. Antipsychotic medications for the treatment of delirium: a systematic review and meta-analysis of randomised controlled trials. J Neurol Neurosurg Psychiatry. 2016;87(7):767–774. doi: 10.1136/jnnp-2015-311049. [DOI] [PubMed] [Google Scholar]
- 19.Agar MR, Lawlor PG, Quinn S, et al. Efficacy of oral risperidone, haloperidol, or placebo for symptoms of delirium among patients in palliative care: a randomized clinical trial. JAMA Int Med. 2017;177(1):34–42. doi: 10.1001/jamainternmed.2016.7491. [DOI] [PubMed] [Google Scholar]
- 20.Barnes H, McDonald J, Smallwood N, Manser R. Opioids for the palliation of refractory breathlessness in adults with advanced disease and terminal illness. Cochrane Database Syst Rev. 2016;3:Cd011008. doi: 10.1002/14651858.CD011008.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bruera E, MacEachem T, Ripamonti C, Hanson J. Subcutaneous morphine for dyspnea in cancer patients. Ann Intern Med. 1993;119(9):906–907. doi: 10.7326/0003-4819-119-9-199311010-00007. [DOI] [PubMed] [Google Scholar]
- 22.Bruera E, Sala R, Spruyt O, et al. Nebulized versus subcutaneous morphine for patients with cancer dyspnea: a preliminary study. J Pain Symptom Manage. 2005;29(6):613–618. doi: 10.1016/j.jpainsymman.2004.08.016. [DOI] [PubMed] [Google Scholar]
- 23.Hui D, Kilgore K, Frisbee-Hume S, et al. Dexamethasone for Dyspnea in Cancer Patients: A Pilot Double-Blind, Randomized, Controlled Trial. J Pain Symptom Manage. 2016;52(1):8–16.e1. doi: 10.1016/j.jpainsymman.2015.10.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Likar R, Rupacher E, Kager H, et al. Efficacy of glycopyrronium bromide and scopolamine hydrobromide in patients with death rattle: a randomized controlled study. Wien Klin Wochenschr. 2008;120(21-22):679–683. doi: 10.1007/s00508-008-1094-2. [DOI] [PubMed] [Google Scholar]
- 25.Lawrey H. Hyoscine vs glycopyrronium for drying respiratory secretions in dying patients. Br J Community Nurs. 2005;10(9):421–424. 426. doi: 10.12968/bjcn.2005.10.9.19688. [DOI] [PubMed] [Google Scholar]
- 26.Back IN, Jenkins K, Blower A, Beckhelling J. A study comparing hyoscine hydrobromide and glycopyrrolate in the treatment of death rattle. Palliat Med. 2001;15(4):329–336. doi: 10.1191/026921601678320313. [DOI] [PubMed] [Google Scholar]
- 27.Bennett M, Lucas V, Brennan M, et al. Using anti-muscarinic drugs in the management of death rattle: evidence-based guidelines for palliative care. Palliat Med. 2002;16(5):369–374. doi: 10.1191/0269216302pm584oa. [DOI] [PubMed] [Google Scholar]
- 28.Wildiers H, Dhaenekint C, Demeulenaere P, et al. Atropine, hyoscine butylbromide, or scopolamine are equally effective for the treatment of death rattle in terminal care. J Pain Symptom Manage. 2009;38(1):124–33. doi: 10.1016/j.jpainsymman.2008.07.007. [DOI] [PubMed] [Google Scholar]
- 29.Wee B, Hillier R. Interventions for noisy breathing in patients near to death. Cochrane Database Syst Rev. 2008;1:Cd005177. doi: 10.1002/14651858.CD005177.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.O'Malley PG. What does off-label prescribing really mean? Arch Intern Med. 2012;172(10):759–760. doi: 10.1001/archinternmed.2012.789. [DOI] [PubMed] [Google Scholar]