This was a cross-sectional study of 324 people who inject drugs in Appalachian Kentucky. The study procedures included a comprehensive baseline self-report assessment. For this study, the research team recruited and interviewed participants assessing them on their demographics, substance use, health, behaviors, and environment. Participants were recruited into this study in 2 ways. The first method entailed targeted recruitment from community locations, such as local churches, stores, organizations, and housing units with higher representation of individuals who inject drugs. The researchers also identified “seed” participants for “respondent–driven sampling.” This method involved recruiting “seeds” who were given recruitment coupons and asked to use them to recruit eligible others from within their social networks to participate in the study.
The researchers used information obtained during participant interviews to identify factors associated with non-prescribed gabapentin use. For recent non-prescribed gabapentin use, participants were asked, “when was the last time that you used gabapentin to get high, for fun, to relax, or come down?” The researchers then categorized responses as having occurred within the past 90 days (~3 months) vs. all other responses/no past gabapentin use. The researchers then investigated characteristics of participants that might be associated with non-prescribed gabapentin use, which included demographics (age, gender, race), county of residence, substance use (e.g., non-prescribed buprenorphine use, heroin, methamphetamine), drug injection frequency, health problems, and healthcare service utilization (e.g., previous substance use disorder treatment, past year doctor visit).
Participants included 324 individuals from 3 separate counties in Kentucky. The majority of participants were White/Caucasian (90.7%); 50% were women, and the mean age of participants was 37.3 years (SD = 9.5). Commonly reported substance use in the past 90 days included methamphetamine (83.9% of participants), cannabis (61.9%), buprenorphine (48.9%), gabapentin (43%), alcohol (32.8%), and prescription opioids (31.3%), benzodiazepines (30.3%), and heroin (27.9%), among others.
WHAT DID THIS STUDY FIND?
Proportion of non-prescribed gabapentin users.
All study participants were people who injected drugs at the time of study enrollment, and 43% of them also reported non-prescribed use of gabapentin in the last 90 days.
Variables that predicted non-prescribed gabapentin use.
Among nearly 30 factors investigated, only 4 were found to be associated with a higher likelihood of recent non-prescribed gabapentin use once the team evaluated all of the other factors, such as demographic characteristics (e.g., gender, age), county, substance use, injection frequency, health problems, and health care utilization. The factors identified indicated that injection drug users with more severe substance use disorder, or who reported recent misuse of buprenorphine, prescription opioids, or a significant amount of physical pain were more likely to report recent non-prescribed gabapentin use.
When the researchers examined each factor one-by-one, and not altogether at once in the same statistical model, other factors were found to be associated with gabapentin use. Specifically, use of cannabis, cocaine, methamphetamine, benzodiazepines, fentanyl, and prescription stimulant medication were each associated with gabapentin use. These additional factors were no longer associated with gabapentin use when every factor was considered at the same time. This can happen when there is overlap among variables that are included in a statistical model, and when some variables represent stronger predictors of the outcome overall.
WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?
Gabapentin has been shown to have a low risk for physical dependence but high and potentially lethal misuse potential, particularly when combined with other drugs that can slow down respiration, such as opioids and alcohol. These researchers identified several factors associated and recent non-prescribed gabapentin use among injection drug users. Notably, substance use disorder severity, pain, cannabis, cocaine, methamphetamine, benzodiazepines, fentanyl, buprenorphine, and prescription stimulant and opioid medication were each associated with non-prescribed gabapentin use. However, only substance use disorder severity, pain, and non-prescribed use of buprenorphine and other prescription opioids were associated with gabapentin use when all measured factors were considered at the same time. As such, these four predictors are considered to be the most important ones to consider among these participants. The fact that these variables were most strongly related to gabapentin use might also suggest a clinical profile or type of patient that at greatest risk; that is, higher levels of pain, prescription opioid use, and more severe substance use disorder might also suggest greater psychological distress, psychosocial impairment, and unmet treatment need among individuals who are most at risk for non-prescribed gabapentin use.
Figure 2.
These results are consistent with prior studies documenting higher rates of non-prescribed gabapentin use among those seeking to become intoxicated, enhance the effects or diminish the negative effects of opioids, or manage pain. These findings also point specifically to pain, non-prescribed opioid and buprenorphine use, and substance use disorder severity as primary predictors of non-prescribed gabapentin use. This information has implications providing some avenues for follow up research, prevention, and treatment among underserved injection drug users in Kentucky and beyond. These data also inform harm reduction strategies, such as public health messaging about the added overdose risk associated with gabapentin, that could be implemented within syringe exchanges or other clinic settings frequented by injection drug users perhaps targeting those individuals especially with the clinical profile identified here.
Results of this study are based on one-time interviews with participants about past substance use behaviors. While the researchers employed common data collection techniques, there are limitations inherent with reliance on retrospective recall alone.
The researchers did not examine the reasons for non-prescribed gabapentin use. For example, gabapentin might have been used in the past 90 days as a means of managing opioid withdrawal, to enhance the effects of opioids, manage pain, some combination of these reasons, or none of these reasons. Relatedly, the researchers are unable to examine the direction of impact in their study, or cause and effect relationships.
The researchers asked about past use of gabapentin and then used this to create a variable to identify past 90-day use vs. everything else. This variable does not provide information about frequency, timing, or dosing of this use, and excludes non-prescribed gabapentin use beyond the 90-day period.
The researchers did not gather source or supply chain information regarding non-prescribed gabapentin use, so it is not clear whether individuals used their own or another’s prescription, or some other unknown source.
While it is important to study higher-risk or higher-need groups, and Kentucky is an identified high-risk region, the results of such research might not generalize broadly. For example, this was largely a rural or rurally-situated sample and the vast majority of participants regarded methamphetamine their primary substance of choice.
Results of this study differed from similar research, which might relate to sampling differences attributable to the targeted and respondent-driven sampling approached used in the current study.
The researchers did not appear to account for similarities and differences between various recruitment sites, or similarities among individuals recruited from the same “seed” within a single “network.”
For scientists:Results of this study suggest that injection drug users in rural Appalachian Kentucky report high rates of non-prescribed gabapentin use (43%). Bivariate and multivariate logistic regression methods were used to identify factors associated with past 90-day non-prescribed gabapentin use; non-prescribed of buprenorphine, prescription opioids, pain, and severe substance use disorder were identified as significant predictors in both sets of analyses. Bivariate analyses identified additional substances associated with past 90-day gabapentin use (e.g., cannabis, methamphetamine); however, these variables were not significant in the full model. Future research may benefit from more targeted and granular assessment of the 4 predictors identified in the full model, and prospective data collection to help overcome the cross-sectional limitations of the current design. Future research might build on the current variable-centered analytic approach (e.g., identification of variables associated with gabapentin use) by employing mixture modeling or cluster analytic techniques to identify subgroups of people at risk for gabapentin use (e.g., injection drug users who use opioids, buprenorphine AND have pain vs. non-prescribed opioid and buprenorphine who do NOT have pain). Replication with new samples and across larger and more geographically, racially/ethnically diverse regions could be of added value to research moving forward, as well as further analysis of the role of other drugs (e.g., methamphetamine) in predicting non-prescribed gabapentin use or complications associated with its misuse.
For policy makers: This study suggests that non-prescribed gabapentin use may be prevalent among injection drug users (43% of sample) in Appalachian Kentucky, and that risk of this is higher among individuals with more severe substance use disorders, recent non-prescribed opioid and buprenorphine use, and who experience pain. This study does not provide direct gabapentin policy implications, as there is much to learn about the reasons for non-prescribed gabapentin use, who’s most at risk and likely to benefit from policy intervention, or before knowing precisely what form policy intervention should take. For instance, if non-prescribed gabapentin use in Kentucky is largely attributable to mismanaged pain in the absence of appropriate treatment resources, then it might be beneficial to direct policy toward improving appropriate treatment access for injection drug users with unmet pain needs. This is in contrast with introducing tighter gabapentin restrictions, which could encourage pain management through other, potentially riskier, means (e.g., non-prescribed buprenorphine or prescription opioid use). In short, additional research funding and support could help ensure that policy interventions are data-driven and appropriately responsive to reducing harm and saving lives.
This was a cross-sectional study of 324 people who inject drugs in Appalachian Kentucky. The study procedures included a comprehensive baseline self-report assessment. For this study, the research team recruited and interviewed participants assessing them on their demographics, substance use, health, behaviors, and environment. Participants were recruited into this study in 2 ways. The first method entailed targeted recruitment from community locations, such as local churches, stores, organizations, and housing units with higher representation of individuals who inject drugs. The researchers also identified “seed” participants for “respondent–driven sampling.” This method involved recruiting “seeds” who were given recruitment coupons and asked to use them to recruit eligible others from within their social networks to participate in the study.
The researchers used information obtained during participant interviews to identify factors associated with non-prescribed gabapentin use. For recent non-prescribed gabapentin use, participants were asked, “when was the last time that you used gabapentin to get high, for fun, to relax, or come down?” The researchers then categorized responses as having occurred within the past 90 days (~3 months) vs. all other responses/no past gabapentin use. The researchers then investigated characteristics of participants that might be associated with non-prescribed gabapentin use, which included demographics (age, gender, race), county of residence, substance use (e.g., non-prescribed buprenorphine use, heroin, methamphetamine), drug injection frequency, health problems, and healthcare service utilization (e.g., previous substance use disorder treatment, past year doctor visit).
Participants included 324 individuals from 3 separate counties in Kentucky. The majority of participants were White/Caucasian (90.7%); 50% were women, and the mean age of participants was 37.3 years (SD = 9.5). Commonly reported substance use in the past 90 days included methamphetamine (83.9% of participants), cannabis (61.9%), buprenorphine (48.9%), gabapentin (43%), alcohol (32.8%), and prescription opioids (31.3%), benzodiazepines (30.3%), and heroin (27.9%), among others.
WHAT DID THIS STUDY FIND?
Proportion of non-prescribed gabapentin users.
All study participants were people who injected drugs at the time of study enrollment, and 43% of them also reported non-prescribed use of gabapentin in the last 90 days.
Variables that predicted non-prescribed gabapentin use.
Among nearly 30 factors investigated, only 4 were found to be associated with a higher likelihood of recent non-prescribed gabapentin use once the team evaluated all of the other factors, such as demographic characteristics (e.g., gender, age), county, substance use, injection frequency, health problems, and health care utilization. The factors identified indicated that injection drug users with more severe substance use disorder, or who reported recent misuse of buprenorphine, prescription opioids, or a significant amount of physical pain were more likely to report recent non-prescribed gabapentin use.
When the researchers examined each factor one-by-one, and not altogether at once in the same statistical model, other factors were found to be associated with gabapentin use. Specifically, use of cannabis, cocaine, methamphetamine, benzodiazepines, fentanyl, and prescription stimulant medication were each associated with gabapentin use. These additional factors were no longer associated with gabapentin use when every factor was considered at the same time. This can happen when there is overlap among variables that are included in a statistical model, and when some variables represent stronger predictors of the outcome overall.
WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?
Gabapentin has been shown to have a low risk for physical dependence but high and potentially lethal misuse potential, particularly when combined with other drugs that can slow down respiration, such as opioids and alcohol. These researchers identified several factors associated and recent non-prescribed gabapentin use among injection drug users. Notably, substance use disorder severity, pain, cannabis, cocaine, methamphetamine, benzodiazepines, fentanyl, buprenorphine, and prescription stimulant and opioid medication were each associated with non-prescribed gabapentin use. However, only substance use disorder severity, pain, and non-prescribed use of buprenorphine and other prescription opioids were associated with gabapentin use when all measured factors were considered at the same time. As such, these four predictors are considered to be the most important ones to consider among these participants. The fact that these variables were most strongly related to gabapentin use might also suggest a clinical profile or type of patient that at greatest risk; that is, higher levels of pain, prescription opioid use, and more severe substance use disorder might also suggest greater psychological distress, psychosocial impairment, and unmet treatment need among individuals who are most at risk for non-prescribed gabapentin use.
Figure 2.
These results are consistent with prior studies documenting higher rates of non-prescribed gabapentin use among those seeking to become intoxicated, enhance the effects or diminish the negative effects of opioids, or manage pain. These findings also point specifically to pain, non-prescribed opioid and buprenorphine use, and substance use disorder severity as primary predictors of non-prescribed gabapentin use. This information has implications providing some avenues for follow up research, prevention, and treatment among underserved injection drug users in Kentucky and beyond. These data also inform harm reduction strategies, such as public health messaging about the added overdose risk associated with gabapentin, that could be implemented within syringe exchanges or other clinic settings frequented by injection drug users perhaps targeting those individuals especially with the clinical profile identified here.
Results of this study are based on one-time interviews with participants about past substance use behaviors. While the researchers employed common data collection techniques, there are limitations inherent with reliance on retrospective recall alone.
The researchers did not examine the reasons for non-prescribed gabapentin use. For example, gabapentin might have been used in the past 90 days as a means of managing opioid withdrawal, to enhance the effects of opioids, manage pain, some combination of these reasons, or none of these reasons. Relatedly, the researchers are unable to examine the direction of impact in their study, or cause and effect relationships.
The researchers asked about past use of gabapentin and then used this to create a variable to identify past 90-day use vs. everything else. This variable does not provide information about frequency, timing, or dosing of this use, and excludes non-prescribed gabapentin use beyond the 90-day period.
The researchers did not gather source or supply chain information regarding non-prescribed gabapentin use, so it is not clear whether individuals used their own or another’s prescription, or some other unknown source.
While it is important to study higher-risk or higher-need groups, and Kentucky is an identified high-risk region, the results of such research might not generalize broadly. For example, this was largely a rural or rurally-situated sample and the vast majority of participants regarded methamphetamine their primary substance of choice.
Results of this study differed from similar research, which might relate to sampling differences attributable to the targeted and respondent-driven sampling approached used in the current study.
The researchers did not appear to account for similarities and differences between various recruitment sites, or similarities among individuals recruited from the same “seed” within a single “network.”
For scientists:Results of this study suggest that injection drug users in rural Appalachian Kentucky report high rates of non-prescribed gabapentin use (43%). Bivariate and multivariate logistic regression methods were used to identify factors associated with past 90-day non-prescribed gabapentin use; non-prescribed of buprenorphine, prescription opioids, pain, and severe substance use disorder were identified as significant predictors in both sets of analyses. Bivariate analyses identified additional substances associated with past 90-day gabapentin use (e.g., cannabis, methamphetamine); however, these variables were not significant in the full model. Future research may benefit from more targeted and granular assessment of the 4 predictors identified in the full model, and prospective data collection to help overcome the cross-sectional limitations of the current design. Future research might build on the current variable-centered analytic approach (e.g., identification of variables associated with gabapentin use) by employing mixture modeling or cluster analytic techniques to identify subgroups of people at risk for gabapentin use (e.g., injection drug users who use opioids, buprenorphine AND have pain vs. non-prescribed opioid and buprenorphine who do NOT have pain). Replication with new samples and across larger and more geographically, racially/ethnically diverse regions could be of added value to research moving forward, as well as further analysis of the role of other drugs (e.g., methamphetamine) in predicting non-prescribed gabapentin use or complications associated with its misuse.
For policy makers: This study suggests that non-prescribed gabapentin use may be prevalent among injection drug users (43% of sample) in Appalachian Kentucky, and that risk of this is higher among individuals with more severe substance use disorders, recent non-prescribed opioid and buprenorphine use, and who experience pain. This study does not provide direct gabapentin policy implications, as there is much to learn about the reasons for non-prescribed gabapentin use, who’s most at risk and likely to benefit from policy intervention, or before knowing precisely what form policy intervention should take. For instance, if non-prescribed gabapentin use in Kentucky is largely attributable to mismanaged pain in the absence of appropriate treatment resources, then it might be beneficial to direct policy toward improving appropriate treatment access for injection drug users with unmet pain needs. This is in contrast with introducing tighter gabapentin restrictions, which could encourage pain management through other, potentially riskier, means (e.g., non-prescribed buprenorphine or prescription opioid use). In short, additional research funding and support could help ensure that policy interventions are data-driven and appropriately responsive to reducing harm and saving lives.
This was a cross-sectional study of 324 people who inject drugs in Appalachian Kentucky. The study procedures included a comprehensive baseline self-report assessment. For this study, the research team recruited and interviewed participants assessing them on their demographics, substance use, health, behaviors, and environment. Participants were recruited into this study in 2 ways. The first method entailed targeted recruitment from community locations, such as local churches, stores, organizations, and housing units with higher representation of individuals who inject drugs. The researchers also identified “seed” participants for “respondent–driven sampling.” This method involved recruiting “seeds” who were given recruitment coupons and asked to use them to recruit eligible others from within their social networks to participate in the study.
The researchers used information obtained during participant interviews to identify factors associated with non-prescribed gabapentin use. For recent non-prescribed gabapentin use, participants were asked, “when was the last time that you used gabapentin to get high, for fun, to relax, or come down?” The researchers then categorized responses as having occurred within the past 90 days (~3 months) vs. all other responses/no past gabapentin use. The researchers then investigated characteristics of participants that might be associated with non-prescribed gabapentin use, which included demographics (age, gender, race), county of residence, substance use (e.g., non-prescribed buprenorphine use, heroin, methamphetamine), drug injection frequency, health problems, and healthcare service utilization (e.g., previous substance use disorder treatment, past year doctor visit).
Participants included 324 individuals from 3 separate counties in Kentucky. The majority of participants were White/Caucasian (90.7%); 50% were women, and the mean age of participants was 37.3 years (SD = 9.5). Commonly reported substance use in the past 90 days included methamphetamine (83.9% of participants), cannabis (61.9%), buprenorphine (48.9%), gabapentin (43%), alcohol (32.8%), and prescription opioids (31.3%), benzodiazepines (30.3%), and heroin (27.9%), among others.
WHAT DID THIS STUDY FIND?
Proportion of non-prescribed gabapentin users.
All study participants were people who injected drugs at the time of study enrollment, and 43% of them also reported non-prescribed use of gabapentin in the last 90 days.
Variables that predicted non-prescribed gabapentin use.
Among nearly 30 factors investigated, only 4 were found to be associated with a higher likelihood of recent non-prescribed gabapentin use once the team evaluated all of the other factors, such as demographic characteristics (e.g., gender, age), county, substance use, injection frequency, health problems, and health care utilization. The factors identified indicated that injection drug users with more severe substance use disorder, or who reported recent misuse of buprenorphine, prescription opioids, or a significant amount of physical pain were more likely to report recent non-prescribed gabapentin use.
When the researchers examined each factor one-by-one, and not altogether at once in the same statistical model, other factors were found to be associated with gabapentin use. Specifically, use of cannabis, cocaine, methamphetamine, benzodiazepines, fentanyl, and prescription stimulant medication were each associated with gabapentin use. These additional factors were no longer associated with gabapentin use when every factor was considered at the same time. This can happen when there is overlap among variables that are included in a statistical model, and when some variables represent stronger predictors of the outcome overall.
WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?
Gabapentin has been shown to have a low risk for physical dependence but high and potentially lethal misuse potential, particularly when combined with other drugs that can slow down respiration, such as opioids and alcohol. These researchers identified several factors associated and recent non-prescribed gabapentin use among injection drug users. Notably, substance use disorder severity, pain, cannabis, cocaine, methamphetamine, benzodiazepines, fentanyl, buprenorphine, and prescription stimulant and opioid medication were each associated with non-prescribed gabapentin use. However, only substance use disorder severity, pain, and non-prescribed use of buprenorphine and other prescription opioids were associated with gabapentin use when all measured factors were considered at the same time. As such, these four predictors are considered to be the most important ones to consider among these participants. The fact that these variables were most strongly related to gabapentin use might also suggest a clinical profile or type of patient that at greatest risk; that is, higher levels of pain, prescription opioid use, and more severe substance use disorder might also suggest greater psychological distress, psychosocial impairment, and unmet treatment need among individuals who are most at risk for non-prescribed gabapentin use.
Figure 2.
These results are consistent with prior studies documenting higher rates of non-prescribed gabapentin use among those seeking to become intoxicated, enhance the effects or diminish the negative effects of opioids, or manage pain. These findings also point specifically to pain, non-prescribed opioid and buprenorphine use, and substance use disorder severity as primary predictors of non-prescribed gabapentin use. This information has implications providing some avenues for follow up research, prevention, and treatment among underserved injection drug users in Kentucky and beyond. These data also inform harm reduction strategies, such as public health messaging about the added overdose risk associated with gabapentin, that could be implemented within syringe exchanges or other clinic settings frequented by injection drug users perhaps targeting those individuals especially with the clinical profile identified here.
Results of this study are based on one-time interviews with participants about past substance use behaviors. While the researchers employed common data collection techniques, there are limitations inherent with reliance on retrospective recall alone.
The researchers did not examine the reasons for non-prescribed gabapentin use. For example, gabapentin might have been used in the past 90 days as a means of managing opioid withdrawal, to enhance the effects of opioids, manage pain, some combination of these reasons, or none of these reasons. Relatedly, the researchers are unable to examine the direction of impact in their study, or cause and effect relationships.
The researchers asked about past use of gabapentin and then used this to create a variable to identify past 90-day use vs. everything else. This variable does not provide information about frequency, timing, or dosing of this use, and excludes non-prescribed gabapentin use beyond the 90-day period.
The researchers did not gather source or supply chain information regarding non-prescribed gabapentin use, so it is not clear whether individuals used their own or another’s prescription, or some other unknown source.
While it is important to study higher-risk or higher-need groups, and Kentucky is an identified high-risk region, the results of such research might not generalize broadly. For example, this was largely a rural or rurally-situated sample and the vast majority of participants regarded methamphetamine their primary substance of choice.
Results of this study differed from similar research, which might relate to sampling differences attributable to the targeted and respondent-driven sampling approached used in the current study.
The researchers did not appear to account for similarities and differences between various recruitment sites, or similarities among individuals recruited from the same “seed” within a single “network.”
For scientists:Results of this study suggest that injection drug users in rural Appalachian Kentucky report high rates of non-prescribed gabapentin use (43%). Bivariate and multivariate logistic regression methods were used to identify factors associated with past 90-day non-prescribed gabapentin use; non-prescribed of buprenorphine, prescription opioids, pain, and severe substance use disorder were identified as significant predictors in both sets of analyses. Bivariate analyses identified additional substances associated with past 90-day gabapentin use (e.g., cannabis, methamphetamine); however, these variables were not significant in the full model. Future research may benefit from more targeted and granular assessment of the 4 predictors identified in the full model, and prospective data collection to help overcome the cross-sectional limitations of the current design. Future research might build on the current variable-centered analytic approach (e.g., identification of variables associated with gabapentin use) by employing mixture modeling or cluster analytic techniques to identify subgroups of people at risk for gabapentin use (e.g., injection drug users who use opioids, buprenorphine AND have pain vs. non-prescribed opioid and buprenorphine who do NOT have pain). Replication with new samples and across larger and more geographically, racially/ethnically diverse regions could be of added value to research moving forward, as well as further analysis of the role of other drugs (e.g., methamphetamine) in predicting non-prescribed gabapentin use or complications associated with its misuse.
For policy makers: This study suggests that non-prescribed gabapentin use may be prevalent among injection drug users (43% of sample) in Appalachian Kentucky, and that risk of this is higher among individuals with more severe substance use disorders, recent non-prescribed opioid and buprenorphine use, and who experience pain. This study does not provide direct gabapentin policy implications, as there is much to learn about the reasons for non-prescribed gabapentin use, who’s most at risk and likely to benefit from policy intervention, or before knowing precisely what form policy intervention should take. For instance, if non-prescribed gabapentin use in Kentucky is largely attributable to mismanaged pain in the absence of appropriate treatment resources, then it might be beneficial to direct policy toward improving appropriate treatment access for injection drug users with unmet pain needs. This is in contrast with introducing tighter gabapentin restrictions, which could encourage pain management through other, potentially riskier, means (e.g., non-prescribed buprenorphine or prescription opioid use). In short, additional research funding and support could help ensure that policy interventions are data-driven and appropriately responsive to reducing harm and saving lives.
