Clinical Toxicology
ISSN: 1556-3650 (Print) 1556-9519 (Online) Journal homepage: http://www.tandfonline.com/loi/ictx20
Trends in analgesic exposures reported to Texas Poison Centers following increased regulation of hydrocodone Ashley Haynes, Kurt Kleinschmidt, Mathias B. Forrester & Amy Young To cite this article: Ashley Haynes, Kurt Kleinschmidt, Mathias B. Forrester & Amy Young (2016) Trends in analgesic exposures reported to Texas Poison Centers following increased regulation of hydrocodone, Clinical Toxicology, 54:5, 434-440, DOI: 10.3109/15563650.2016.1148720 To link to this article: http://dx.doi.org/10.3109/15563650.2016.1148720
Published online: 26 Feb 2016.
Submit your article to this journal
Article views: 298
View related articles
View Crossmark data
Citing articles: 7 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ictx20 Download by: [Imperial College London Library]
Date: 15 September 2017, At: 00:27
CLINICAL TOXICOLOGY, 2016 VOL. 54, NO. 5, 434–440 http://dx.doi.org/10.3109/15563650.2016.1148720
POISON CENTRE RESEARCH
Trends in analgesic exposures reported to Texas Poison Centers following increased regulation of hydrocodone Ashley Haynesa, Kurt Kleinschmidta, Mathias B. Forresterb and Amy Younga
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
a Emergency Medicine/Division of Toxicology, University of Texas Southwestern, Dallas, TX, USA; bTexas Department of State Health Services, Austin, TX, USA
ABSTRACT
ARTICLE HISTORY
Context: In October 2014, the Drug Enforcement Administration reclassified hydrocodone to schedule II, increasing regulations on use. The impact of rescheduling hydrocodone on opioid exposures is unclear, especially in states with special restrictions required for prescribing schedule II agents. Objective: To assess whether changes in exposures to prescription opioid analgesics and heroin as reported to poison centers occurred in the 6 months after hydrocodone rescheduling. We hypothesized that hydrocodone exposures would decrease, while less tightly regulated opioids, such as codeine and tramadol, would increase. Materials and methods: This study compares opioid analgesic exposures reported to Texas Poison Centers before and after this change in a state that requires special prescription pads for Schedule II agents. Cases included all opioid analgesic exposures reported to a statewide poison center network, comparing exposures from 6 months before to 6 months after heightened regulations. Specific opioids with large changes in reported exposures were further characterized by patient age and exposure intent. Results: Hydrocodone exposures decreased from 1567 to 1135 (28%, p ¼ 0.00017), decreasing for all ages. Codeine exposures increased significantly from 189 to 522 (176%, p ¼ 0.00014), including a 263% increase for age >20 years. Codeine misuse increased 443% and adverse drug events 327%. Oxycodone exposures increased from 134 to 189 (39%, p ¼ 0.0143), increasing only among patients age >20 years. Reported heroin exposures increased from 156 to 179 (15%, p ¼ 0.2286) and tramadol from 666 to 708 (6%, p ¼ 0.0193). Other opioid exposures changed little or had limited reports. Discussion: The increased regulation of hydrocodone was followed temporally by a decrease in reported hydrocodone exposures, but also increases in codeine, oxycodone and tramadol exposures. This may reflect a shift in prescribing practices, changes in street availability of hydrocodone or decreased drug diversion. Conclusion: The increased regulation was temporally associated with decreased hydrocodone exposures reported to Texas Poison Centers.
Received 30 October 2015 Revised 22 January 2016 Accepted 25 January 2016 Published online 25 February 2016
Introduction Opioid substance use disorder has become an increasing problem in the United States since the early 1990s.[1] A study of opioid prescribing practices in 2012 demonstrated a 35% increase in the number of opioid prescriptions during the period of 2000–2010.[2] In 2014, an epidemiologic study regarding patients in the emergency department (ED) in California and Florida reflected that nearly 10% of opioid overdose-related ED visits resulted in a near-fatal outcome, with death in 1.1%.[3] In 2013, the US Food and Drug Administration (FDA) approved a single-entity, long-acting hydrocodone product (Zohydro, Pernix Therapeutics, LLC, Morristown, NJ) for use in the treatment of pain. Concerns have arisen that this may exacerbate opioid addiction; moreover, this was approved despite the FDA’s scientific advisory panel voting against approval.[4] According to an IMS Healthcare Informatics report, hydrocodone/acetaminophen was the #2 selling prescription drug in 2014 in the US, with 119 million prescriptions.[5] A review of ED visits involving the non-medical use of prescription drugs from 2004 to 2008 in the US demonstrated that hydrocodone was one of the top three implicated CONTACT Ashley Haynes ß 2016 Taylor & Francis
[email protected]
KEYWORDS
Analgesics; codeine; drug and narcotic control; drug exposure patterns/trends; hydrocodone; opioid
agents, with a statistically significant increase over the study period.[6] Unfortunately, most available data regarding opioid overdose trends combine hydrocodone with other opioids; thus, little is known currently about trends in hydrocodone overdoses individually. Data from the National Survey on Drug Use and Health (NSDUH) from 2008 to 2010 implicated physicians as the source of 17.5–18.9% of prescription drugs used non-medically.[7] In 2014, NSDUH data from 2008 to 2011 demonstrated that while friends and family were the primary source of non-medically used opioids, persons exhibiting high-frequency, high-risk use were more likely to obtain opioid prescriptions from a physician.[8] On 24 October 2013, the FDA announced that it was considering reclassification of hydrocodone from Schedule III to Schedule II, a more tightly regulated classification.[9] The Controlled Substances Act, passed in the US in 1970, designates agents with no therapeutic use as schedule I, which are illegal. Agents with high abuse potential and potential for severe withdrawal are designated schedule II and require strict regulations compared with other classes, including limitations in the number of refills and restrictions on who may prescribe these agents and methods by which these may be transmitted to pharmacies.[10] Agents with moderate abuse
Emergency Medicine/Division of Toxicology, University of Texas Southwestern, Dallas, TX, USA
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
CLINICAL TOXICOLOGY
potential and low-to-moderate potential for withdrawal are classified as schedule III and are less tightly regulated.[10] The Drug Enforcement Administration (DEA) made this effective on 6 October 2014.[11] The change in scheduling classification means that hydrocodone combination medications will be more strictly regulated, with no refills able to be given and telephonic prescribing prohibited. This may also have additional implications in some states, such as Texas. In Texas, all Schedule II medications must be prescribed using a special form, a copy or record of which must be kept by the physician for at least 2 years. Many physicians do not carry these, as they have been previously able to prescribe hydrocodone when an opioid was indicated for acute pain. With this scheduling change, prescription opioid analgesic prescribing practices are anticipated to change in Texas, i.e. physicians might switch from prescribing hydrocodone to other prescription opioid analgesics. Moreover, the availability of hydrocodone either by prescription or on the streets may also change. As a consequence, the nature of analgesic drug exposures may be greatly affected. Heroin may also be affected; individuals unable to obtain hydrocodone might seek to obtain heroin, which may be cheaper and readily available, to prevent or treat opioid withdrawal and cravings. Several studies have demonstrated the use of heroin after prior non-medical use of prescription opioids.[12,13] This study assesses whether any changes in exposures to prescription opioid analgesics and heroin as reported to poison centers occurred in the 6 months after hydrocodone was rescheduled. We hypothesized that hydrocodone exposures would decrease, while less tightly regulated opioids, such as codeine and tramadol, would increase.
Methods The data source for this epidemiologic study was the Texas Poison Center Network (TPCN), six poison centers that together service the state’s population of over 25 million. The TPCN is a telephone consultation service that assists in the management of potentially adverse exposures to many substances, including prescription opioid analgesics. The six poison centers of the TPCN use a single electronic database to collect information on all calls in a similar manner. The data variables and allowable codes in this database were standardized by the American Association of Poison Control Centers (AAPCC).[14] Cases were all exposures to prescription opioid analgesics and heroin reported to the TPCN during April 2014–March 2015. An exposure was defined by the AAPCC definition as ‘‘actual or suspected contact with any substance which has been ingested, inhaled, absorbed, applied to or injected into the body, regardless of toxicity or clinical manifestation’’.[14] Texas Health and Safety Code 161.042 requires healthcare providers who treat overdoses to Penalty Group 1 drugs to report these overdoses to the Texas Department of State Health Services (DSHS). Most prescription opioid analgesics and heroin are included in Penalty Group 1. When this law was created, the decision was made to submit these overdoses to the DSHS through the TPCN. Thus, a portion of the prescription opioid analgesic and heroin cases included in
435
this investigation may not have been reported to the TPCN for assistance in the treatment of the patient but simply as notification according to the law. Exposures involving other substances in addition to the prescription opioid analgesics or heroin and exposures not followed to a final medical outcome were included. The exposures were divided into two groups: (1) those reported during April 2014–September 2014 (the 6 months before rescheduling) and (2) those reported during October 2014–March 2015 (the 6 months after rescheduling). The number of exposures reported during these two time periods was determined for each drug and compared by calculating the ratio of the number after rescheduling to the number before rescheduling. For those drugs demonstrating the greatest change between these two time periods, a similar comparison was performed for patient age, exposure reason and medical outcome. For the analysis of patient age, the patients were grouped into those 12 years or less, 13–19 years and 20 years. Patients of unknown age were excluded from the age analysis. For exposure reason, the major categories were unintentional (i.e. accidental), intentional (i.e. suspected attempted suicide, intentional misuse or abuse), adverse reaction to the product, other, and unknown. The medical outcome or severity of an exposure is assigned by the poison center staff and is based on the observed or anticipated adverse clinical effects. Medical outcome is classified according to the following criteria: no effect (no symptoms due to exposure), minor effect (some minimally troublesome symptoms), moderate effect (more pronounced, prolonged symptoms), major effect (symptoms that are life-threatening or cause significant disability or disfigurement) and death. A portion of exposures are not followed to a final medical outcome because of resource constraints or the inability to obtain subsequent information on the patient. In these instances, the poison center staff records the expected outcome of the exposure. These expected outcomes are grouped into the following categories: not followed but judged as non-toxic exposure (symptoms not expected), not followed but minimal symptoms possible (no more than minor symptoms possible), unable to follow but judged as a potentially toxic exposure. Another medical outcome category is unrelated effect where the exposure was probably not responsible for the symptoms. For both hydrocodone and codeine, rates of exposure were determined by dividing by total exposure call volume for each month of the study. These rates of reported exposures before and after the reclassification were compared for these two drugs using a Student’s t-test. Changes in the other five most reported opioids were also assessed for statistical significance using a Student’s t-test of the monthly exposure rates. This study is exempted from ethical review by the DSHS institutional review board.
Results The number of hydrocodone exposures decreased 28%. In contrast, the number of codeine exposures increased 176%.
A. HAYNES ET AL.
The other three opioids among the five most reported exposures all had more modest increases; including a 15% increase for heroin and a 6% increase for tramadol. While reported exposures of other opioids had small fluctuations, these exposures represented very small numbers of cases and neither trends nor statistical significance were established. Table 1 compares the number of reported exposures to the various opioids before and after hydrocodone was rescheduled. The total number of opioid exposures were similar for the study periods; 2838 exposures during the first 6 months and 2816 during the second 6 months. The monthly number of reported exposures for the five most common Schedule II and III agents; hydrocodone, codeine, tramadol, oxycodone and heroin; during the 12-month period are shown in Figure 1. The monthly number of hydrocodone exposures was relatively stable during April–September 2014, declined slightly during October 2014 and declined further during Table 1. Number of prescription opioid analgesic and heroin exposures reported to the TPCN before and after hydrocodone rescheduling.
Drug Hydrocodone Tramadol Codeine Heroin Oxycodone Morphine Methadone Buprenorphine Unknown opioid Fentanyl Hydromorphone Propoxyphene Oxymorphone Pentazocine Tapentadol Butorphanol Meperidine Nalbuphine Dihydrocodeine
6 Months before rescheduling
6 Months after rescheduling
After/before ratio
1567 666 189 156 134 84 82 76 50 46 34 9 8 8 8 3 4 2 0
1135 708 522 179 186 93 77 70 64 37 40 10 10 8 8 4 1 1 1
0.72 1.06 2.76 1.15 1.39 1.11 0.94 0.92 1.28 0.80 1.18 1.11 1.25 1.00 1.00 1.33 0.25 0.50 –
A given exposure might involve more than one drug. Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015.
300
250
Reported Exposures
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
436
p Value 0.00017 0.0193 0.00014 0.2286 0.0143
November 2014 before remaining relatively stable at the lower number during December 2014–March 2015. The monthly number of codeine exposures was stable during April–September 2014 before doubling in October 2014, then remaining at this higher level through March 2015. Monthly oxycodone exposures did not undergo a change between September and October 2014; however, there was a slow upward trend over the ensuing months that was statistically significant (p ¼ 0.0143). While the increase in tramadol was statistically significant (p ¼ 0.0193), no consistent trending in the monthly data for tramadol was identified and further characterization was not done. Three of the 4 months with the most heroin use occurred within the final 4 months of the study; however, again, no clear trend was present and this was not statistically significant (p ¼ 0.2286). The decrease in hydrocodone exposures was seen in all age groups and for all exposure reasons (Table 2). Hydrocodone represented 55% of all opioid exposures in the first 6 months and 40% of all opioid exposures in the last 6 months. Exposures with major outcomes decreased by over a third and reported deaths due to hydrocodone exposures also decreased; albeit the numbers for the latter are small. When the total volume of exposures reported to Texas Poison Centers during this time period is considered, the rate of hydrocodone exposures decreased significantly from 1.9% of all exposures to 1.4% (p ¼ 0.00017) of all exposures (Table 3). To evaluate for a potential seasonal effect, the 6 months following the change were also compared with the same 6 months the preceding year. The exposures did drop significantly from 1.9% in the preceding year to 1.4% following the schedule change (p ¼ 0.000049). The increase in codeine exposures was seen across all age ranges (Table 4). The greatest escalation was in the patients aged 20 years, increasing by 262%. The increase was noted for all exposure reasons. The number of therapeutic errors more than doubled, and adverse drug reactions quadrupled. There were no deaths reported during the study period. When the total volume of exposures reported to Texas Poison Centers during this time period was examined, the rate of codeine exposures increased significantly from 0.23% of all
Hydrocodone Schedule Change
200 Hydrocodone 150
Codeine Tramadol
100
Oxycodone Heroin
50
0
Figure 1. Opioid exposures reported to Texas Poison Centers from April 2014 to March 2015.
CLINICAL TOXICOLOGY
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
exposures to 0.67% (p ¼ 0.00014) of all exposures (Table 5). Codeine was reported in 6.7% of all opioid exposures in the first 6 months, increasing to 18.5% of all opioid exposures in the last 6 months. Codeine alone represented just 15 exposures in the 6 months before and 28 exposures in the 6 months after the new regulations, while the remaining exposures were related to combination products. To evaluate for a potentially seasonal effect, the 6 months following the change were also compared with the same 6 months the preceding year. The exposures did increase significantly from 0.34% in the preceding year to 0.67% following the schedule change (p < 0.00001). Table 6 displays further details regarding oxycodone exposures. Exposures in patients aged 20 years increased by Table 2. Hydrocodone exposure characteristics reported to the TPCN before and after hydrocodone rescheduling.
Discussion The intent of laws scheduling medications is to limit abuse of high-risk substances, thereby decreasing harm. This study is Table 4. Codeine exposure characteristics reported to the TPCN before and after hydrocodone rescheduling. 6 Months after rescheduling
Patient age (years) 12 13–19 20 Unknown
56 38 92 3
113 72 333 4
2.02 1.89 3.62
0.61 0.76 0.70 – 0.00 0.73 0.64 0.81 0.73 0.78 0.50 – 0.00 0.93
Exposure reason Unintentional – general Unintentional – therapeutic error Unintentional – misuse Unintentional – food poisoning Unintentional – unknown Intentional – suspected suicide Intentional – misuse Intentional – abuse Intentional – unknown Adverse reaction – drug Adverse reaction – other Other – malicious Other – withdrawal Unknown reason
22 64 1 0 0 62 8 9 7 13 0 0 0 3
54 151 2 1 1 170 43 24 19 52 0 1 0 4
2.45 2.36 2.00 – – 2.74 5.38 2.67 2.71 4.00 – – – 1.33
0.76 0.72 0.73 0.63 0.43 0.38 0.79 0.66 1.16
Medical outcome No effect Minor effect Moderate effect Major effect Death NF – judged as nontoxic NF – minimal clinical effects NF – judged potentially toxic Unrelated effect
55 30 21 6 0 7 45 22 3
105 104 84 20 0 11 135 49 14
1.91 3.47 4.00 3.33 – 1.57 3.00 2.23 4.67
6 Months after rescheduling
180 186 1,193 8
120 165 839 11
0.67 0.89 0.70
Exposure reason Unintentional – general Unintentional – therapeutic error Unintentional – misuse Unintentional – food poisoning Unintentional – unknown Intentional – suspected suicide Intentional – misuse Intentional – abuse Intentional – unknown Adverse reaction – drug Adverse reaction – other Other – malicious Other – withdrawal Unknown reason
175 247 23 0 4 709 124 98 74 64 2 0 1 46
107 188 16 0 0 517 79 79 54 50 1 1 0 43
Medical outcome No effect Minor effect Moderate effect Major effect Death NF – judged as nontoxic NF – minimal clinical effects NF – judged potentially toxic Unrelated effect
304 373 300 99 7 34 240 191 19
231 268 220 62 3 13 189 127 22
Patient age (years) 12 13–19 20 Unknown
over 50% while a decrease in reported exposures was seen in patients aged 12–19 years, although the total number of exposures in this age group was low. Four of the 5 months with the most exposures were in the last 4 months of the study. Among reasons for exposure, intentional misuse had the greatest rise, while intentional abuse saw a decline. The number of reported deaths due to oxycodone did not change.
6 Months before rescheduling
6 Months before rescheduling
Variable
437
After/before ratio
Variable
Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015. NF: Not followed/unable to follow.
Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015. NF: Not followed/unable to follow.
Table 3. Monthly hydrocodone exposures reported to the TPCN as both a raw number and a percentage of total exposures from April 2014 through March 2014. Reported hydrocodone exposures 6 Months before rescheduling
6 Months after rescheduling
Month
Number
Percent of all exposures
Month
Number
Percent of all exposures
Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14
252 275 254 265 271 250
1.88 1.93 1.83 1.91 1.90 1.79
Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15
234 184 181 186 166 184
1.63 1.40 1.43 1.41 1.37 1.37
The overall percent change in hydrocodone exposures decreased from 1.88% in the 6 months before to 1.44% after rescheduling, p ¼ 0.00017.
After/before ratio
438
A. HAYNES ET AL.
Table 5. Monthly codeine exposures reported to the TPCN as both a raw number and a percentage of total exposures from April 2014 through March 2014. Reported codeine exposures 6 Months before rescheduling
6 Months after rescheduling
Month
Number
Percent of all exposures
Month
Number
Percent of all exposures
Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14
34 36 31 29 25 34
0.25 0.25 0.22 0.21 0.18 0.24
Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15
78 87 90 97 91 79
0.57 0.66 0.71 0.74 0.75 0.59
The change in codeine exposures in the 6 months before and after hydrocodone rescheduling was statistically significant, p ¼ 0.00014.
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
Table 6. Oxycodone exposure characteristics reported to the TPCN before and after hydrocodone rescheduling. 6 Months before rescheduling
6 Months after rescheduling
13 16 102 3
15 13 155 3
1.15 0.81 1.52
Exposure reason Unintentional – general Unintentional – therapeutic error Unintentional – misuse Unintentional – food poisoning Unintentional – unknown Intentional – suspected suicide Intentional – misuse Intentional – abuse Intentional – unknown Adverse reaction – drug Adverse reaction – other Other – malicious Other – withdrawal Unknown reason
16 29 0 0 0 47 8 18 8 3 0 0 0 5
17 38 2 0 1 66 18 14 13 6 0 0 0 11
1.06 1.31 – – – 1.40 2.25 0.78 1.63 2.00 – – – 2.20
Medical outcome No effect Minor effect Moderate effect Major effect Death NF – judged as nontoxic NF – minimal clinical effects NF – judged potentially toxic Unrelated effect
21 34 24 12 1 3 19 19 1
33 38 37 12 1 2 30 33 0
1.57 1.12 1.54 1.00 1.00 0.67 1.58 1.74 0.00
Variable Patient age (years) 12 13–19 20 Unknown
After/before ratio
Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015. NF: Not followed/unable to follow.
an important first look at the events following a legislative change in hydrocodone regulation on reported opioid exposures. Our hypothesis that rescheduling of hydrocodone from Schedule III to Schedule II would decrease hydrocodone exposures while increasing those to other opioids was supported by the data. These changes occurred for all reason codes, suggesting an impact on all types of exposures and not just abuse that is the general target of these regulations. Various potential explanations exist for the decreased hydrocodone exposures after the rescheduling. One is that there was an overall decrease in opioid exposures between study periods. However, the data reflected the overall opioid exposure calls were similar between study periods. Also, the percent of exposure calls that were about opioids actually increased from the first to the second study period,
increasing from 3.4 to 3.6% of all calls. Another potential reason for the decreased hydrocodone exposures is that there was a decrease in the number of total exposure calls to the poison centers. In fact, there was a decrease in total call volume by 6.8%. However, exposures to all opioids did not decrease by a similar percentage. The more likely cause for the decrease in hydrocodone exposures is that it is being used less while other agents are being used more. Tramadol exposures were flat between study periods. However, codeine exposures increased much over the same period reflecting that codeine may have become the replacement medication for hydrocodone. The inverse relationship in exposures between these two agents was even reflected by the change in total number of these two agents over study periods. While codeine is a Schedule II medication, when in combination with other medications it is regulated at Schedule III. The increase in codeine exposures may be due to providers choosing a less tightly regulated substance (e.g. codeine with R ) to control pain. Providers acetaminophen, or Tylenol #3V may choose another agent because they do not have the required Texas Department of Public Safety prescription pads. Another factor that could decrease the use of hydrocodone is that physician assistants, nurse practitioners and residents practicing with a training license are not permitted to obtain the required prescription pads, and must seek the attending physician for assistance in prescribing Schedule II medications. Anecdotally, a preference for prescribing codeine with acetaminophen was suspected at one of the authors’ home institution, as a shortage of codeine with acetaminophen was announced just 20 days after the hydrocodone rescheduling went into effect despite no apparent changes in supply. Prior to the study period, surveys of ED admissions by the Substances and Mental Health Services Administration had indicated increasing trends in mentions of all opioid analgesics, with hydrocodone mentions more than doubling between 2004 and 2011.[15] While unlikely, the decrease in hydrocodone exposures noted in this study may be due to a downward trend that began prior to the rescheduling of hydrocodone. However, when we compared with the same months a year prior, there was a statistically significant difference between the 2 years, which would argue against a preexisting trend. Providers who were informed of the FDA’s recommendations regarding the hydrocodone schedule and the approval of sustained release hydrocodone preparations may have begun to alter their own prescribing practices due to heightened awareness. In addition, the rescheduling no
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
CLINICAL TOXICOLOGY
longer permits phone-in hydrocodone prescriptions without a physical prescription, which should reduce calls placed by patients impersonating physicians for the purpose of obtaining hydrocodone. Securement and monitoring of hydrocodone in pharmacies and healthcare facilities is also increased, which likely impacted rates of diversion. All of these may have affected the price and availability of hydrocodone on the streets. In addition, if fewer hydrocodone tablets are present in homes, accidental ingestions by children or persons for whom the medication is prescribed may also decrease. Codeine has a complicated metabolism and thus its use in pediatric patients has been discouraged by the American Academy of Pediatrics (AAP).[16,17] Codeine metabolism occurs primarily via cytochrome p450 2D6 and is highly variable in the population.[18] Those who rapidly metabolize codeine to the active metabolite, morphine, may have profound symptoms of intoxication at therapeutic doses. Those who are unable to metabolize codeine may have very minimal pain relief even at higher therapeutic doses. Despite concerns over this complexity, codeine exposures in our study doubled after the hydrocodone rescheduling. The reasons for this are not clear but may include the simplicity in using standard prescriptions versus the special prescription pads, a change in drug acquisition costs or insurance coverage, street availability, or other reasons. Most codeine exposures were due to the less-tightly regulated combination medications. This leads to the consideration that providers may have been choosing less tightly-regulated agents once hydrocodone regulations changed. The increase in codeine-related adverse drug events and therapeutic errors may also support this. The number of oxycodone exposures increased after the hydrocodone rescheduling. In 2012, a report by Kenan et al. [2] demonstrated a 35% increase in the number of opioid prescriptions and a 69% increase in prescription sizes for both oxycodone and hydrocodone between 2000 and 2012. Thus, the increase in oxycodone exposures we are reporting may simply support Kenan et al.’s observation of a steady increase. Figure 1 demonstrates an upward trend in monthly oxycodone exposures preceding the regulatory change, though the trend does seem to be more exaggerated after the change. Another explanation may be that if providers are required to use a special prescription pad, they may prescribe a medication they believe is stronger with no difference in required effort. This study has various public health implications. This study reflects that regulatory changes may impact opioid exposures, including limiting availability of these substances for recreational use and accidental exposure to children in the home. Another possible impact is the potential for undertreated pain if providers intentionally avoid more tightlyregulated substances. While use of the more regulated agent may decrease, concern can be raised regarding increases in exposures to other opioids. This includes increases in heroin use in the population of patients who abuse opioids. Heroin use is associated with risks above the non-medical use of prescription opioids, such as risks for blood-borne diseases, endocarditis and risks associated with contaminants. Certainly this is an area for future investigation. This study has limitations. Refills on existing hydrocodone prescriptions written prior to 6 October 2014 were permitted
439
for refilling until 8 April 2015; hence, the effects of rescheduling on hydrocodone exposures may be underestimated.[11] In spite of Texas Health and Safety Code 161.042, all potentially adverse exposures to prescription opioid analgesics and heroin might not be reported to the TPCN. Thus, the cases included in this investigation might not be representative of all such exposures that occur in the state. Moreover, Texas Health and Safety Code 161.042 might influence the reporting of prescription opioid analgesics and heroin to the TPCN in some manner that might affect the current investigation. In addition, codeine is also contained in non-analgesic preparations, such as cold medications; thus, a rise in codeine exposures reported may be due to other medications as well. Reported exposures may involve multiple agents, which is a confounder. Accuracy of poison center data is reliant on the coding by the call agents (called Specialists in Poison Information) documenting the exposure. This study covered only 6 months before and 6 months after hydrocodone was rescheduled and included cases from only a single state, albeit a state with a population of over 25 million. Thus, a relatively small number of cases were included in this investigation, particularly for many of the specific prescription opioid analgesics. The impact of rescheduling hydrocodone might have changed after March 2015 and the decrease in exposures may not be sustained. Lastly, the data represent only temporal relationships between the changes in hydrocodone regulations and exposures to these agents, but cannot confirm a causal relationship given other nationwide interventions underway to curb prescription drug abuse.
Conclusions The increased regulation of hydrocodone in the US was followed by a large decrease in hydrocodone exposures reported to the TPCN. Benefits of these decreased exposures may be limited by increases in exposures to other opioid agents. Further evaluation after more time has elapsed will be necessary to demonstrate whether any lasting benefit endures.
Disclosure statement The authors report no declarations of interest.
References [1]
[2]
[3]
[4]
Ling W, Mooney L, Hillhouse M. Prescription opioid abuse, pain and addiction: clinical issues and implications. Drug Alcohol Rev. 2011;30:300–305. Kenan K, Mack K, Paulozzi L. Trends in prescriptions for oxycodone and other commonly used opioids in the United States, 2000–2010. Open Med. 2012;6:e41–e47. Hasegawa K, Brown DF, Tsugawa Y, et al. Epidemiology of emergency department visits for opioid overdose: a population-based study. Mayo Clin Proc. 2014;89:462–471. Manchikanti L, Atluri S, Candido KD, et al. Zohydro approval by food and drug administration: controversial or frightening? Pain Phys. 2014;17:E437–E450.
440
[5]
[6]
[7]
[8]
[9]
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
[10]
[11]
A. HAYNES ET AL.
Aitken M. Medicines use and spending shifts: a review of the use of medicines in the US in 2014. IMS Institute for Healthcare Informatics. 2015. Control CD. Prevention. Emergency department visits involving nonmedical use of selected prescription drugs - United States, 2004–2008. MMWR Morb Mortal Wkly Rep. 2010;59:705–709. Wang KH, Fiellin DA, Becker WC. Source of prescription drugs used nonmedically in rural and urban populations. Am J Drug Alcohol Abuse. 2014;40:292–303. Jones CM, Paulozzi LJ, Mack KA. Sources of prescription opioid pain relievers by frequency of past-year nonmedical use United States, 2008–2011. JAMA Intern Med. 2014;174:802–803. FDA. Statement on proposed hydrocodone relcassification from Janet Woodcock, MD, Director, Center for Drug Evaluation and Research. 2013. Available from: http://wwwfdagov/Drugs/ DrugSafety/ucm372089htm. US DOJ D. Title 21 Code of Federal Regulations 1308.11 through 1308.15. 2015. Available from: http://wwwdeadiversionusdojgov/21cfr/ cfr/2108cfrthtm. Drug Enforcement Administration DoJ. Schedules of controlled substances: rescheduling of hydrocodone combination products from schedule III to schedule II. Final rule. Fed Regist. 2014;79:49661–49682.
[12]
[13]
[14]
[15]
[16]
[17] [18]
Jones CM. Heroin use and heroin use risk behaviors among nonmedical users of prescription opioid pain relievers – United States, 2002–2004 and 2008–2010. Drug Alcohol Depend. 2013;132: 95–100. Kolodny A, Courtwright DT, Hwang CS, et al. The prescription opioid and heroin crisis: a public health approach to an epidemic of addiction. Annu Rev Public Health. 2015;36:559–574. AAPCC. American Association of Poison Control Centers National Poison Data System (NPDS) Reference Manual Part 2 - System Information Manual. 2007. US Department of Health and Human Serrvices, Substance Abuse and Mental Health Services Administration, Office of Applied Studies, Drug Abuse Warning Network (DAWN) National and Metro Tables, 2004–2011 (‘‘All misuse and abuse’’). 2013. Available from: http://wwwsamhsagov/data/DAWNaspx#DAWN. Use of codeine- and dextromethorphan-containing cough remedies in children. American Academy of Pediatrics. Committee on Drugs. Pediatrics. 1997;99:918–920. AAP Publications Retired or Reaffirmed October 2006. Pediatrics. 2007;119:450. Zhou SF. Polymorphism of human cytochrome P450 2D6 and its clinical significance: part II. Clin Pharmacokinet. 2009;48: 761–804.
ISSN: 1556-3650 (Print) 1556-9519 (Online) Journal homepage: http://www.tandfonline.com/loi/ictx20
Trends in analgesic exposures reported to Texas Poison Centers following increased regulation of hydrocodone Ashley Haynes, Kurt Kleinschmidt, Mathias B. Forrester & Amy Young To cite this article: Ashley Haynes, Kurt Kleinschmidt, Mathias B. Forrester & Amy Young (2016) Trends in analgesic exposures reported to Texas Poison Centers following increased regulation of hydrocodone, Clinical Toxicology, 54:5, 434-440, DOI: 10.3109/15563650.2016.1148720 To link to this article: http://dx.doi.org/10.3109/15563650.2016.1148720
Published online: 26 Feb 2016.
Submit your article to this journal
Article views: 298
View related articles
View Crossmark data
Citing articles: 7 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ictx20 Download by: [Imperial College London Library]
Date: 15 September 2017, At: 00:27
CLINICAL TOXICOLOGY, 2016 VOL. 54, NO. 5, 434–440 http://dx.doi.org/10.3109/15563650.2016.1148720
POISON CENTRE RESEARCH
Trends in analgesic exposures reported to Texas Poison Centers following increased regulation of hydrocodone Ashley Haynesa, Kurt Kleinschmidta, Mathias B. Forresterb and Amy Younga
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
a Emergency Medicine/Division of Toxicology, University of Texas Southwestern, Dallas, TX, USA; bTexas Department of State Health Services, Austin, TX, USA
ABSTRACT
ARTICLE HISTORY
Context: In October 2014, the Drug Enforcement Administration reclassified hydrocodone to schedule II, increasing regulations on use. The impact of rescheduling hydrocodone on opioid exposures is unclear, especially in states with special restrictions required for prescribing schedule II agents. Objective: To assess whether changes in exposures to prescription opioid analgesics and heroin as reported to poison centers occurred in the 6 months after hydrocodone rescheduling. We hypothesized that hydrocodone exposures would decrease, while less tightly regulated opioids, such as codeine and tramadol, would increase. Materials and methods: This study compares opioid analgesic exposures reported to Texas Poison Centers before and after this change in a state that requires special prescription pads for Schedule II agents. Cases included all opioid analgesic exposures reported to a statewide poison center network, comparing exposures from 6 months before to 6 months after heightened regulations. Specific opioids with large changes in reported exposures were further characterized by patient age and exposure intent. Results: Hydrocodone exposures decreased from 1567 to 1135 (28%, p ¼ 0.00017), decreasing for all ages. Codeine exposures increased significantly from 189 to 522 (176%, p ¼ 0.00014), including a 263% increase for age >20 years. Codeine misuse increased 443% and adverse drug events 327%. Oxycodone exposures increased from 134 to 189 (39%, p ¼ 0.0143), increasing only among patients age >20 years. Reported heroin exposures increased from 156 to 179 (15%, p ¼ 0.2286) and tramadol from 666 to 708 (6%, p ¼ 0.0193). Other opioid exposures changed little or had limited reports. Discussion: The increased regulation of hydrocodone was followed temporally by a decrease in reported hydrocodone exposures, but also increases in codeine, oxycodone and tramadol exposures. This may reflect a shift in prescribing practices, changes in street availability of hydrocodone or decreased drug diversion. Conclusion: The increased regulation was temporally associated with decreased hydrocodone exposures reported to Texas Poison Centers.
Received 30 October 2015 Revised 22 January 2016 Accepted 25 January 2016 Published online 25 February 2016
Introduction Opioid substance use disorder has become an increasing problem in the United States since the early 1990s.[1] A study of opioid prescribing practices in 2012 demonstrated a 35% increase in the number of opioid prescriptions during the period of 2000–2010.[2] In 2014, an epidemiologic study regarding patients in the emergency department (ED) in California and Florida reflected that nearly 10% of opioid overdose-related ED visits resulted in a near-fatal outcome, with death in 1.1%.[3] In 2013, the US Food and Drug Administration (FDA) approved a single-entity, long-acting hydrocodone product (Zohydro, Pernix Therapeutics, LLC, Morristown, NJ) for use in the treatment of pain. Concerns have arisen that this may exacerbate opioid addiction; moreover, this was approved despite the FDA’s scientific advisory panel voting against approval.[4] According to an IMS Healthcare Informatics report, hydrocodone/acetaminophen was the #2 selling prescription drug in 2014 in the US, with 119 million prescriptions.[5] A review of ED visits involving the non-medical use of prescription drugs from 2004 to 2008 in the US demonstrated that hydrocodone was one of the top three implicated CONTACT Ashley Haynes ß 2016 Taylor & Francis
[email protected]
KEYWORDS
Analgesics; codeine; drug and narcotic control; drug exposure patterns/trends; hydrocodone; opioid
agents, with a statistically significant increase over the study period.[6] Unfortunately, most available data regarding opioid overdose trends combine hydrocodone with other opioids; thus, little is known currently about trends in hydrocodone overdoses individually. Data from the National Survey on Drug Use and Health (NSDUH) from 2008 to 2010 implicated physicians as the source of 17.5–18.9% of prescription drugs used non-medically.[7] In 2014, NSDUH data from 2008 to 2011 demonstrated that while friends and family were the primary source of non-medically used opioids, persons exhibiting high-frequency, high-risk use were more likely to obtain opioid prescriptions from a physician.[8] On 24 October 2013, the FDA announced that it was considering reclassification of hydrocodone from Schedule III to Schedule II, a more tightly regulated classification.[9] The Controlled Substances Act, passed in the US in 1970, designates agents with no therapeutic use as schedule I, which are illegal. Agents with high abuse potential and potential for severe withdrawal are designated schedule II and require strict regulations compared with other classes, including limitations in the number of refills and restrictions on who may prescribe these agents and methods by which these may be transmitted to pharmacies.[10] Agents with moderate abuse
Emergency Medicine/Division of Toxicology, University of Texas Southwestern, Dallas, TX, USA
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
CLINICAL TOXICOLOGY
potential and low-to-moderate potential for withdrawal are classified as schedule III and are less tightly regulated.[10] The Drug Enforcement Administration (DEA) made this effective on 6 October 2014.[11] The change in scheduling classification means that hydrocodone combination medications will be more strictly regulated, with no refills able to be given and telephonic prescribing prohibited. This may also have additional implications in some states, such as Texas. In Texas, all Schedule II medications must be prescribed using a special form, a copy or record of which must be kept by the physician for at least 2 years. Many physicians do not carry these, as they have been previously able to prescribe hydrocodone when an opioid was indicated for acute pain. With this scheduling change, prescription opioid analgesic prescribing practices are anticipated to change in Texas, i.e. physicians might switch from prescribing hydrocodone to other prescription opioid analgesics. Moreover, the availability of hydrocodone either by prescription or on the streets may also change. As a consequence, the nature of analgesic drug exposures may be greatly affected. Heroin may also be affected; individuals unable to obtain hydrocodone might seek to obtain heroin, which may be cheaper and readily available, to prevent or treat opioid withdrawal and cravings. Several studies have demonstrated the use of heroin after prior non-medical use of prescription opioids.[12,13] This study assesses whether any changes in exposures to prescription opioid analgesics and heroin as reported to poison centers occurred in the 6 months after hydrocodone was rescheduled. We hypothesized that hydrocodone exposures would decrease, while less tightly regulated opioids, such as codeine and tramadol, would increase.
Methods The data source for this epidemiologic study was the Texas Poison Center Network (TPCN), six poison centers that together service the state’s population of over 25 million. The TPCN is a telephone consultation service that assists in the management of potentially adverse exposures to many substances, including prescription opioid analgesics. The six poison centers of the TPCN use a single electronic database to collect information on all calls in a similar manner. The data variables and allowable codes in this database were standardized by the American Association of Poison Control Centers (AAPCC).[14] Cases were all exposures to prescription opioid analgesics and heroin reported to the TPCN during April 2014–March 2015. An exposure was defined by the AAPCC definition as ‘‘actual or suspected contact with any substance which has been ingested, inhaled, absorbed, applied to or injected into the body, regardless of toxicity or clinical manifestation’’.[14] Texas Health and Safety Code 161.042 requires healthcare providers who treat overdoses to Penalty Group 1 drugs to report these overdoses to the Texas Department of State Health Services (DSHS). Most prescription opioid analgesics and heroin are included in Penalty Group 1. When this law was created, the decision was made to submit these overdoses to the DSHS through the TPCN. Thus, a portion of the prescription opioid analgesic and heroin cases included in
435
this investigation may not have been reported to the TPCN for assistance in the treatment of the patient but simply as notification according to the law. Exposures involving other substances in addition to the prescription opioid analgesics or heroin and exposures not followed to a final medical outcome were included. The exposures were divided into two groups: (1) those reported during April 2014–September 2014 (the 6 months before rescheduling) and (2) those reported during October 2014–March 2015 (the 6 months after rescheduling). The number of exposures reported during these two time periods was determined for each drug and compared by calculating the ratio of the number after rescheduling to the number before rescheduling. For those drugs demonstrating the greatest change between these two time periods, a similar comparison was performed for patient age, exposure reason and medical outcome. For the analysis of patient age, the patients were grouped into those 12 years or less, 13–19 years and 20 years. Patients of unknown age were excluded from the age analysis. For exposure reason, the major categories were unintentional (i.e. accidental), intentional (i.e. suspected attempted suicide, intentional misuse or abuse), adverse reaction to the product, other, and unknown. The medical outcome or severity of an exposure is assigned by the poison center staff and is based on the observed or anticipated adverse clinical effects. Medical outcome is classified according to the following criteria: no effect (no symptoms due to exposure), minor effect (some minimally troublesome symptoms), moderate effect (more pronounced, prolonged symptoms), major effect (symptoms that are life-threatening or cause significant disability or disfigurement) and death. A portion of exposures are not followed to a final medical outcome because of resource constraints or the inability to obtain subsequent information on the patient. In these instances, the poison center staff records the expected outcome of the exposure. These expected outcomes are grouped into the following categories: not followed but judged as non-toxic exposure (symptoms not expected), not followed but minimal symptoms possible (no more than minor symptoms possible), unable to follow but judged as a potentially toxic exposure. Another medical outcome category is unrelated effect where the exposure was probably not responsible for the symptoms. For both hydrocodone and codeine, rates of exposure were determined by dividing by total exposure call volume for each month of the study. These rates of reported exposures before and after the reclassification were compared for these two drugs using a Student’s t-test. Changes in the other five most reported opioids were also assessed for statistical significance using a Student’s t-test of the monthly exposure rates. This study is exempted from ethical review by the DSHS institutional review board.
Results The number of hydrocodone exposures decreased 28%. In contrast, the number of codeine exposures increased 176%.
A. HAYNES ET AL.
The other three opioids among the five most reported exposures all had more modest increases; including a 15% increase for heroin and a 6% increase for tramadol. While reported exposures of other opioids had small fluctuations, these exposures represented very small numbers of cases and neither trends nor statistical significance were established. Table 1 compares the number of reported exposures to the various opioids before and after hydrocodone was rescheduled. The total number of opioid exposures were similar for the study periods; 2838 exposures during the first 6 months and 2816 during the second 6 months. The monthly number of reported exposures for the five most common Schedule II and III agents; hydrocodone, codeine, tramadol, oxycodone and heroin; during the 12-month period are shown in Figure 1. The monthly number of hydrocodone exposures was relatively stable during April–September 2014, declined slightly during October 2014 and declined further during Table 1. Number of prescription opioid analgesic and heroin exposures reported to the TPCN before and after hydrocodone rescheduling.
Drug Hydrocodone Tramadol Codeine Heroin Oxycodone Morphine Methadone Buprenorphine Unknown opioid Fentanyl Hydromorphone Propoxyphene Oxymorphone Pentazocine Tapentadol Butorphanol Meperidine Nalbuphine Dihydrocodeine
6 Months before rescheduling
6 Months after rescheduling
After/before ratio
1567 666 189 156 134 84 82 76 50 46 34 9 8 8 8 3 4 2 0
1135 708 522 179 186 93 77 70 64 37 40 10 10 8 8 4 1 1 1
0.72 1.06 2.76 1.15 1.39 1.11 0.94 0.92 1.28 0.80 1.18 1.11 1.25 1.00 1.00 1.33 0.25 0.50 –
A given exposure might involve more than one drug. Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015.
300
250
Reported Exposures
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
436
p Value 0.00017 0.0193 0.00014 0.2286 0.0143
November 2014 before remaining relatively stable at the lower number during December 2014–March 2015. The monthly number of codeine exposures was stable during April–September 2014 before doubling in October 2014, then remaining at this higher level through March 2015. Monthly oxycodone exposures did not undergo a change between September and October 2014; however, there was a slow upward trend over the ensuing months that was statistically significant (p ¼ 0.0143). While the increase in tramadol was statistically significant (p ¼ 0.0193), no consistent trending in the monthly data for tramadol was identified and further characterization was not done. Three of the 4 months with the most heroin use occurred within the final 4 months of the study; however, again, no clear trend was present and this was not statistically significant (p ¼ 0.2286). The decrease in hydrocodone exposures was seen in all age groups and for all exposure reasons (Table 2). Hydrocodone represented 55% of all opioid exposures in the first 6 months and 40% of all opioid exposures in the last 6 months. Exposures with major outcomes decreased by over a third and reported deaths due to hydrocodone exposures also decreased; albeit the numbers for the latter are small. When the total volume of exposures reported to Texas Poison Centers during this time period is considered, the rate of hydrocodone exposures decreased significantly from 1.9% of all exposures to 1.4% (p ¼ 0.00017) of all exposures (Table 3). To evaluate for a potential seasonal effect, the 6 months following the change were also compared with the same 6 months the preceding year. The exposures did drop significantly from 1.9% in the preceding year to 1.4% following the schedule change (p ¼ 0.000049). The increase in codeine exposures was seen across all age ranges (Table 4). The greatest escalation was in the patients aged 20 years, increasing by 262%. The increase was noted for all exposure reasons. The number of therapeutic errors more than doubled, and adverse drug reactions quadrupled. There were no deaths reported during the study period. When the total volume of exposures reported to Texas Poison Centers during this time period was examined, the rate of codeine exposures increased significantly from 0.23% of all
Hydrocodone Schedule Change
200 Hydrocodone 150
Codeine Tramadol
100
Oxycodone Heroin
50
0
Figure 1. Opioid exposures reported to Texas Poison Centers from April 2014 to March 2015.
CLINICAL TOXICOLOGY
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
exposures to 0.67% (p ¼ 0.00014) of all exposures (Table 5). Codeine was reported in 6.7% of all opioid exposures in the first 6 months, increasing to 18.5% of all opioid exposures in the last 6 months. Codeine alone represented just 15 exposures in the 6 months before and 28 exposures in the 6 months after the new regulations, while the remaining exposures were related to combination products. To evaluate for a potentially seasonal effect, the 6 months following the change were also compared with the same 6 months the preceding year. The exposures did increase significantly from 0.34% in the preceding year to 0.67% following the schedule change (p < 0.00001). Table 6 displays further details regarding oxycodone exposures. Exposures in patients aged 20 years increased by Table 2. Hydrocodone exposure characteristics reported to the TPCN before and after hydrocodone rescheduling.
Discussion The intent of laws scheduling medications is to limit abuse of high-risk substances, thereby decreasing harm. This study is Table 4. Codeine exposure characteristics reported to the TPCN before and after hydrocodone rescheduling. 6 Months after rescheduling
Patient age (years) 12 13–19 20 Unknown
56 38 92 3
113 72 333 4
2.02 1.89 3.62
0.61 0.76 0.70 – 0.00 0.73 0.64 0.81 0.73 0.78 0.50 – 0.00 0.93
Exposure reason Unintentional – general Unintentional – therapeutic error Unintentional – misuse Unintentional – food poisoning Unintentional – unknown Intentional – suspected suicide Intentional – misuse Intentional – abuse Intentional – unknown Adverse reaction – drug Adverse reaction – other Other – malicious Other – withdrawal Unknown reason
22 64 1 0 0 62 8 9 7 13 0 0 0 3
54 151 2 1 1 170 43 24 19 52 0 1 0 4
2.45 2.36 2.00 – – 2.74 5.38 2.67 2.71 4.00 – – – 1.33
0.76 0.72 0.73 0.63 0.43 0.38 0.79 0.66 1.16
Medical outcome No effect Minor effect Moderate effect Major effect Death NF – judged as nontoxic NF – minimal clinical effects NF – judged potentially toxic Unrelated effect
55 30 21 6 0 7 45 22 3
105 104 84 20 0 11 135 49 14
1.91 3.47 4.00 3.33 – 1.57 3.00 2.23 4.67
6 Months after rescheduling
180 186 1,193 8
120 165 839 11
0.67 0.89 0.70
Exposure reason Unintentional – general Unintentional – therapeutic error Unintentional – misuse Unintentional – food poisoning Unintentional – unknown Intentional – suspected suicide Intentional – misuse Intentional – abuse Intentional – unknown Adverse reaction – drug Adverse reaction – other Other – malicious Other – withdrawal Unknown reason
175 247 23 0 4 709 124 98 74 64 2 0 1 46
107 188 16 0 0 517 79 79 54 50 1 1 0 43
Medical outcome No effect Minor effect Moderate effect Major effect Death NF – judged as nontoxic NF – minimal clinical effects NF – judged potentially toxic Unrelated effect
304 373 300 99 7 34 240 191 19
231 268 220 62 3 13 189 127 22
Patient age (years) 12 13–19 20 Unknown
over 50% while a decrease in reported exposures was seen in patients aged 12–19 years, although the total number of exposures in this age group was low. Four of the 5 months with the most exposures were in the last 4 months of the study. Among reasons for exposure, intentional misuse had the greatest rise, while intentional abuse saw a decline. The number of reported deaths due to oxycodone did not change.
6 Months before rescheduling
6 Months before rescheduling
Variable
437
After/before ratio
Variable
Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015. NF: Not followed/unable to follow.
Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015. NF: Not followed/unable to follow.
Table 3. Monthly hydrocodone exposures reported to the TPCN as both a raw number and a percentage of total exposures from April 2014 through March 2014. Reported hydrocodone exposures 6 Months before rescheduling
6 Months after rescheduling
Month
Number
Percent of all exposures
Month
Number
Percent of all exposures
Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14
252 275 254 265 271 250
1.88 1.93 1.83 1.91 1.90 1.79
Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15
234 184 181 186 166 184
1.63 1.40 1.43 1.41 1.37 1.37
The overall percent change in hydrocodone exposures decreased from 1.88% in the 6 months before to 1.44% after rescheduling, p ¼ 0.00017.
After/before ratio
438
A. HAYNES ET AL.
Table 5. Monthly codeine exposures reported to the TPCN as both a raw number and a percentage of total exposures from April 2014 through March 2014. Reported codeine exposures 6 Months before rescheduling
6 Months after rescheduling
Month
Number
Percent of all exposures
Month
Number
Percent of all exposures
Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14
34 36 31 29 25 34
0.25 0.25 0.22 0.21 0.18 0.24
Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15
78 87 90 97 91 79
0.57 0.66 0.71 0.74 0.75 0.59
The change in codeine exposures in the 6 months before and after hydrocodone rescheduling was statistically significant, p ¼ 0.00014.
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
Table 6. Oxycodone exposure characteristics reported to the TPCN before and after hydrocodone rescheduling. 6 Months before rescheduling
6 Months after rescheduling
13 16 102 3
15 13 155 3
1.15 0.81 1.52
Exposure reason Unintentional – general Unintentional – therapeutic error Unintentional – misuse Unintentional – food poisoning Unintentional – unknown Intentional – suspected suicide Intentional – misuse Intentional – abuse Intentional – unknown Adverse reaction – drug Adverse reaction – other Other – malicious Other – withdrawal Unknown reason
16 29 0 0 0 47 8 18 8 3 0 0 0 5
17 38 2 0 1 66 18 14 13 6 0 0 0 11
1.06 1.31 – – – 1.40 2.25 0.78 1.63 2.00 – – – 2.20
Medical outcome No effect Minor effect Moderate effect Major effect Death NF – judged as nontoxic NF – minimal clinical effects NF – judged potentially toxic Unrelated effect
21 34 24 12 1 3 19 19 1
33 38 37 12 1 2 30 33 0
1.57 1.12 1.54 1.00 1.00 0.67 1.58 1.74 0.00
Variable Patient age (years) 12 13–19 20 Unknown
After/before ratio
Before rescheduling: April 2014–September 2014. After rescheduling: October 2014–March 2015. NF: Not followed/unable to follow.
an important first look at the events following a legislative change in hydrocodone regulation on reported opioid exposures. Our hypothesis that rescheduling of hydrocodone from Schedule III to Schedule II would decrease hydrocodone exposures while increasing those to other opioids was supported by the data. These changes occurred for all reason codes, suggesting an impact on all types of exposures and not just abuse that is the general target of these regulations. Various potential explanations exist for the decreased hydrocodone exposures after the rescheduling. One is that there was an overall decrease in opioid exposures between study periods. However, the data reflected the overall opioid exposure calls were similar between study periods. Also, the percent of exposure calls that were about opioids actually increased from the first to the second study period,
increasing from 3.4 to 3.6% of all calls. Another potential reason for the decreased hydrocodone exposures is that there was a decrease in the number of total exposure calls to the poison centers. In fact, there was a decrease in total call volume by 6.8%. However, exposures to all opioids did not decrease by a similar percentage. The more likely cause for the decrease in hydrocodone exposures is that it is being used less while other agents are being used more. Tramadol exposures were flat between study periods. However, codeine exposures increased much over the same period reflecting that codeine may have become the replacement medication for hydrocodone. The inverse relationship in exposures between these two agents was even reflected by the change in total number of these two agents over study periods. While codeine is a Schedule II medication, when in combination with other medications it is regulated at Schedule III. The increase in codeine exposures may be due to providers choosing a less tightly regulated substance (e.g. codeine with R ) to control pain. Providers acetaminophen, or Tylenol #3V may choose another agent because they do not have the required Texas Department of Public Safety prescription pads. Another factor that could decrease the use of hydrocodone is that physician assistants, nurse practitioners and residents practicing with a training license are not permitted to obtain the required prescription pads, and must seek the attending physician for assistance in prescribing Schedule II medications. Anecdotally, a preference for prescribing codeine with acetaminophen was suspected at one of the authors’ home institution, as a shortage of codeine with acetaminophen was announced just 20 days after the hydrocodone rescheduling went into effect despite no apparent changes in supply. Prior to the study period, surveys of ED admissions by the Substances and Mental Health Services Administration had indicated increasing trends in mentions of all opioid analgesics, with hydrocodone mentions more than doubling between 2004 and 2011.[15] While unlikely, the decrease in hydrocodone exposures noted in this study may be due to a downward trend that began prior to the rescheduling of hydrocodone. However, when we compared with the same months a year prior, there was a statistically significant difference between the 2 years, which would argue against a preexisting trend. Providers who were informed of the FDA’s recommendations regarding the hydrocodone schedule and the approval of sustained release hydrocodone preparations may have begun to alter their own prescribing practices due to heightened awareness. In addition, the rescheduling no
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
CLINICAL TOXICOLOGY
longer permits phone-in hydrocodone prescriptions without a physical prescription, which should reduce calls placed by patients impersonating physicians for the purpose of obtaining hydrocodone. Securement and monitoring of hydrocodone in pharmacies and healthcare facilities is also increased, which likely impacted rates of diversion. All of these may have affected the price and availability of hydrocodone on the streets. In addition, if fewer hydrocodone tablets are present in homes, accidental ingestions by children or persons for whom the medication is prescribed may also decrease. Codeine has a complicated metabolism and thus its use in pediatric patients has been discouraged by the American Academy of Pediatrics (AAP).[16,17] Codeine metabolism occurs primarily via cytochrome p450 2D6 and is highly variable in the population.[18] Those who rapidly metabolize codeine to the active metabolite, morphine, may have profound symptoms of intoxication at therapeutic doses. Those who are unable to metabolize codeine may have very minimal pain relief even at higher therapeutic doses. Despite concerns over this complexity, codeine exposures in our study doubled after the hydrocodone rescheduling. The reasons for this are not clear but may include the simplicity in using standard prescriptions versus the special prescription pads, a change in drug acquisition costs or insurance coverage, street availability, or other reasons. Most codeine exposures were due to the less-tightly regulated combination medications. This leads to the consideration that providers may have been choosing less tightly-regulated agents once hydrocodone regulations changed. The increase in codeine-related adverse drug events and therapeutic errors may also support this. The number of oxycodone exposures increased after the hydrocodone rescheduling. In 2012, a report by Kenan et al. [2] demonstrated a 35% increase in the number of opioid prescriptions and a 69% increase in prescription sizes for both oxycodone and hydrocodone between 2000 and 2012. Thus, the increase in oxycodone exposures we are reporting may simply support Kenan et al.’s observation of a steady increase. Figure 1 demonstrates an upward trend in monthly oxycodone exposures preceding the regulatory change, though the trend does seem to be more exaggerated after the change. Another explanation may be that if providers are required to use a special prescription pad, they may prescribe a medication they believe is stronger with no difference in required effort. This study has various public health implications. This study reflects that regulatory changes may impact opioid exposures, including limiting availability of these substances for recreational use and accidental exposure to children in the home. Another possible impact is the potential for undertreated pain if providers intentionally avoid more tightlyregulated substances. While use of the more regulated agent may decrease, concern can be raised regarding increases in exposures to other opioids. This includes increases in heroin use in the population of patients who abuse opioids. Heroin use is associated with risks above the non-medical use of prescription opioids, such as risks for blood-borne diseases, endocarditis and risks associated with contaminants. Certainly this is an area for future investigation. This study has limitations. Refills on existing hydrocodone prescriptions written prior to 6 October 2014 were permitted
439
for refilling until 8 April 2015; hence, the effects of rescheduling on hydrocodone exposures may be underestimated.[11] In spite of Texas Health and Safety Code 161.042, all potentially adverse exposures to prescription opioid analgesics and heroin might not be reported to the TPCN. Thus, the cases included in this investigation might not be representative of all such exposures that occur in the state. Moreover, Texas Health and Safety Code 161.042 might influence the reporting of prescription opioid analgesics and heroin to the TPCN in some manner that might affect the current investigation. In addition, codeine is also contained in non-analgesic preparations, such as cold medications; thus, a rise in codeine exposures reported may be due to other medications as well. Reported exposures may involve multiple agents, which is a confounder. Accuracy of poison center data is reliant on the coding by the call agents (called Specialists in Poison Information) documenting the exposure. This study covered only 6 months before and 6 months after hydrocodone was rescheduled and included cases from only a single state, albeit a state with a population of over 25 million. Thus, a relatively small number of cases were included in this investigation, particularly for many of the specific prescription opioid analgesics. The impact of rescheduling hydrocodone might have changed after March 2015 and the decrease in exposures may not be sustained. Lastly, the data represent only temporal relationships between the changes in hydrocodone regulations and exposures to these agents, but cannot confirm a causal relationship given other nationwide interventions underway to curb prescription drug abuse.
Conclusions The increased regulation of hydrocodone in the US was followed by a large decrease in hydrocodone exposures reported to the TPCN. Benefits of these decreased exposures may be limited by increases in exposures to other opioid agents. Further evaluation after more time has elapsed will be necessary to demonstrate whether any lasting benefit endures.
Disclosure statement The authors report no declarations of interest.
References [1]
[2]
[3]
[4]
Ling W, Mooney L, Hillhouse M. Prescription opioid abuse, pain and addiction: clinical issues and implications. Drug Alcohol Rev. 2011;30:300–305. Kenan K, Mack K, Paulozzi L. Trends in prescriptions for oxycodone and other commonly used opioids in the United States, 2000–2010. Open Med. 2012;6:e41–e47. Hasegawa K, Brown DF, Tsugawa Y, et al. Epidemiology of emergency department visits for opioid overdose: a population-based study. Mayo Clin Proc. 2014;89:462–471. Manchikanti L, Atluri S, Candido KD, et al. Zohydro approval by food and drug administration: controversial or frightening? Pain Phys. 2014;17:E437–E450.
440
[5]
[6]
[7]
[8]
[9]
Downloaded by [Imperial College London Library] at 00:27 15 September 2017
[10]
[11]
A. HAYNES ET AL.
Aitken M. Medicines use and spending shifts: a review of the use of medicines in the US in 2014. IMS Institute for Healthcare Informatics. 2015. Control CD. Prevention. Emergency department visits involving nonmedical use of selected prescription drugs - United States, 2004–2008. MMWR Morb Mortal Wkly Rep. 2010;59:705–709. Wang KH, Fiellin DA, Becker WC. Source of prescription drugs used nonmedically in rural and urban populations. Am J Drug Alcohol Abuse. 2014;40:292–303. Jones CM, Paulozzi LJ, Mack KA. Sources of prescription opioid pain relievers by frequency of past-year nonmedical use United States, 2008–2011. JAMA Intern Med. 2014;174:802–803. FDA. Statement on proposed hydrocodone relcassification from Janet Woodcock, MD, Director, Center for Drug Evaluation and Research. 2013. Available from: http://wwwfdagov/Drugs/ DrugSafety/ucm372089htm. US DOJ D. Title 21 Code of Federal Regulations 1308.11 through 1308.15. 2015. Available from: http://wwwdeadiversionusdojgov/21cfr/ cfr/2108cfrthtm. Drug Enforcement Administration DoJ. Schedules of controlled substances: rescheduling of hydrocodone combination products from schedule III to schedule II. Final rule. Fed Regist. 2014;79:49661–49682.
[12]
[13]
[14]
[15]
[16]
[17] [18]
Jones CM. Heroin use and heroin use risk behaviors among nonmedical users of prescription opioid pain relievers – United States, 2002–2004 and 2008–2010. Drug Alcohol Depend. 2013;132: 95–100. Kolodny A, Courtwright DT, Hwang CS, et al. The prescription opioid and heroin crisis: a public health approach to an epidemic of addiction. Annu Rev Public Health. 2015;36:559–574. AAPCC. American Association of Poison Control Centers National Poison Data System (NPDS) Reference Manual Part 2 - System Information Manual. 2007. US Department of Health and Human Serrvices, Substance Abuse and Mental Health Services Administration, Office of Applied Studies, Drug Abuse Warning Network (DAWN) National and Metro Tables, 2004–2011 (‘‘All misuse and abuse’’). 2013. Available from: http://wwwsamhsagov/data/DAWNaspx#DAWN. Use of codeine- and dextromethorphan-containing cough remedies in children. American Academy of Pediatrics. Committee on Drugs. Pediatrics. 1997;99:918–920. AAP Publications Retired or Reaffirmed October 2006. Pediatrics. 2007;119:450. Zhou SF. Polymorphism of human cytochrome P450 2D6 and its clinical significance: part II. Clin Pharmacokinet. 2009;48: 761–804.
