Abstract
Background:
The US Department of Veterans Affairs, Department of Defense (VA/DoD) clinical guidelines recommend extended-release naltrexone (XR-NTX) as a treatment option for moderate-to-severe alcohol use disorder (AUD); however, contemporary real-world outcomes related to this guideline are lacking. This retrospective, observational, descriptive study examined treatment patterns and healthcare resource use (HCRU) among veterans with an AUD diagnosis who initiated XR-NTX.
Methods:
Veterans with incident AUD who initiated XR-NTX between 8/2014 and 11/2018 were identified. Treatment patterns and HCRU were assessed during the 1-year baseline period before and following XR-NTX initiation (the index date).
Results:
Of the 3665 VA patients (mean [SD] age: 46 [12.5] years; male: 89.7%; White: 76.9%) included in the study, time from AUD diagnosis to XR-NTX initiation was highly variable (mean [range]: 13.6 [0-50.5 months]). Patients received a mean [SD] of 6.8 [6.1] XR-NTX administrations; 44.4% received ⩾6. Mean [SD] time to XR-NTX discontinuation was 93.4 [75.7] days, and 31.3% of discontinuing patients resumed XR-NTX therapy. Of those who received other subsequent medications for AUD, 38.6% (acamprosate) to 47.8% (disulfiram) re-initiated XR-NTX. The proportion of patients with ⩾1 inpatient admissions decreased during follow-up compared with baseline (all-cause: 61.5% to 37.8%; AUD-related: 58.0%-35.4%); with a smaller decrease observed in emergency department (ED) visits. In contrast, more patients had ⩾1 outpatient visits during follow-up (all-cause: 97.5%-99.7%; AUD-related: 84.4%-92.7%). Compared with baseline, mean number of inpatient admissions and ED visits decreased during follow-up, while the number of outpatient visits increased for both all-cause and AUD-related care.
Conclusions:
Among VA patients with AUD who initiated XR-NTX, we observed reductions in all-cause and AUD-related acute care, and increases in outpatient care. This finding demonstrates a possible transition from acute, inpatient treatment to long-term, outpatient care that may reflect a reduction in disease severity. Additional research is warranted.
Introduction
Alcohol use disorder (AUD) disproportionately affects veterans relative to the general population.1,2 The US Department of Veterans Affairs/Department of Defense (VA/DoD) Clinical Practice Guideline for the Management of Substance Use Disorders recommends medications for AUD (MAUD), such as oral naltrexone, extended-release naltrexone (XR-NTX), acamprosate and disulfiram for the treatment of moderate-to-severe AUD (DSM-5), which corresponds to DSM-IV criteria for alcohol dependence.3,4 Existing studies on healthcare resource use (HCRU) among VA patients with AUD treated with XR-NTX were conducted prior to the addition of XR-NTX to VA/DoD Clinical Practice Guideline in 2015.5,6 This descriptive study explored the patient journey, from the time of AUD diagnosis to before and after XR-NTX initiation, in a contemporary US veteran population treated with XR-NTX by assessing treatment patterns and HCRU.
Methods
This was a retrospective, observational study that utilized administrative data on healthcare encounters from the VA Health Services Research & Development Corporate Data Warehouse.
Eligible veterans were required to have ⩾2 encounters during baseline, an incident AUD diagnosis between August 2014 and November 2018, no evidence of opioid use disorder, and ⩾1 XR-NTX prescription on or after their date of AUD diagnosis. The index date was the date of the first XR-NTX prescription. Treatment patterns and HCRU were assessed during baseline (1-year period prior to index date) and follow-up (until death or 1-year post-index, whichever occurred first; Figure 1). All analyses were descriptive. Means, standard deviations (SDs) and medians were reported for continuous variables and counts and percentages were summarized for categorical variables.
Study design.
Results
There were 3665 VA patients in this study (Figure 2). Most patients were male (89.7%), White (76.9%) or non-Hispanic (89.5%) with a mean age of 46 years at XR-NTX initiation (Table 1). Depression (74.1%), post-traumatic stress syndrome (52.9%) and hypertension (40.5%) were the most common comorbid conditions. Two thirds (66.3%) of patients had another non-opioid substance use disorder.
Patient disposition.
Patient demographic and clinical characteristics during the baseline period.
Abbreviations: AUD, alcohol use disorder.
The mean (median) time from AUD diagnosis to XR-NTX initiation was 13.6 (9.3) months (Table 2). During the 1-year period following XR-NTX initiation, patients received a mean of 6.8 XR-NTX injections and 44.4% received ⩾6 injections. Most patients (93.2%) discontinued XR-NTX during follow-up, with a mean time to discontinuation of 93.4 days. Among patients who discontinued XR-NTX, approximately one third (31.3%) resumed treatment with XR-NTX.
Extended-release naltrexone (XR-NTX) treatment patterns during the 1-year follow-up period.
Abbreviations: AUD, alcohol use disorder.
The follow-up period was defined as the 1-year period following the index date. In total, 8 patients who died within the year following index date are included in the table. Among these patients, the duration from index to death was 76, 107, 114, 272, 308, 346, 354 and 363 days.
Discontinuation was assessed as no subsequent record of XR-NTX in the 45-day period following the previous XR-NTX record. The date of discontinuation was assigned as the end of days of supply.
159 (4.3%) patients received ⩾20 claims for XR-NTX within the 1-year follow-up period, which might be a result due to duplicate claim records.
XR-NTX was the initial MAUD in 22% of patients (Table 3). Most patients who were initially treated with another MAUD had received oral naltrexone (75.8%), with 7.3% and 7.1% of patients having received acamprosate and disulfiram, respectively. After the index XR-NTX treatment episode, about half of the patients (46.6%) did not initiate another MAUD (Figure 3). Of those who initiated oral naltrexone, acamprosate or disulfiram after index XR-NTX treatment, 40.3%, 38.6% and 47.8% of these patients, respectively, subsequently re-initiated treatment with XR-NTX. The mean time to re-initiation ranged from 1.1 (disulfiram) to 2.9 (oral naltrexone) months.
Sequence of medications for AUD.
Abbreviations: AUD, alcohol use disorder; XR-NTX, extended-release naltrexone.
Discontinuation was assessed as no subsequent record of XR-NTX in the 45-day period following the previous XR-NTX record. The date of discontinuation was assigned as the end of days of supply.
If days of supply was missing for XR-NTX in any place of service, it was imputed as 28 days including date of initiation.
AUD treatment initiations beyond the index XR-NTX injection, for up to 2 subsequent treatment initiations.a
All-cause and AUD-related HCRU during baseline and follow-up are described in Table 4 and Figure 4. During baseline, 61.5% and 39.8% of patients had ⩾1 all-cause inpatient admission and emergency department (ED) visit(s), respectively; during follow-up, these percentages decreased to 37.8% and 35.4%. Reductions were also observed in AUD-related inpatient admissions and ED visits. After XR-NTX initiation, a similar proportion of patients had ⩾1 all-cause outpatient visits during follow-up (99.7%) vs baseline (97.5%). On average, patients had 28 all-cause and 10 AUD-related outpatient visits during the baseline period, which increased to 43 all-cause and 18 AUD-related visits during follow-up.
Healthcare resource use during the 1-year period prior to and following the index date. a
Abbreviations: AUD, alcohol use disorder; HCPCS, Healthcare Common Procedure Coding System; NDC, National Drug Code; VA, Veterans Affairs; XR-NTX, extended-release naltrexone.
The index date was defined as the date of the patient’s first prescription for XR-NTX. The baseline and follow-up periods were defined as 1 year prior to and 1 year following the index date, respectively.
In total, 8 patients who died within the year following index date were included in the follow-up period. Among these patients, the duration from index to death was 76, 107, 114, 272, 308, 346, 354 and 363 days.
Residential treatments were categorized via HCPCS (either inpatient or outpatient), stop codes (outpatient only), and bed section/location variables (inpatient only) within the VA data.
Length of stay was calculated at the admission level and patients could contribute more than 1 length of stay.
AUD-related was defined by a claim with an alcohol-specific-procedure code or a non-alcohol specific procedure code in combination with a diagnosis code for AUD on the same day.
Medication use was identified through NDC codes or HCPCS codes listed.
Mean and SD were computed among patients with at least 1 prescription for AUD medication (ie, oral naltrexone, acamprosate or disulfiram).
Healthcare resource use during the 1-year period prior to and following the index date. (A) All-cause HCRU, any admission/visit. (B) AUD-related HCRU, any admission/visit.
Discussion
Among VA patients receiving XR-NTX, 1 in 5 patients received XR-NTX as their initial MAUD. Most patients in this study had received oral naltrexone prior to XR-NTX. On average, over a year elapsed between AUD diagnosis and XR-NTX initiation, during which time VA patients had high levels of acute care (e.g. inpatient, ED visits). A 3-month period of XR-NTX persistence was observed, followed by a break from pharmacotherapy or a switch to an alternative MAUD treatment, followed by a resumption of XR-NTX. These findings are consistent with Harris et al, 5 who found that the proportion of individuals who reached 90 days without a gap in medication possession was highest among VA patients receiving XR-NTX (20.2%) relative to other MAUDs (range, 9.4%-19.1%). A similar study among VA patients with AUD reported that patients treated with XR-NTX had better adherence rates than those treated with other MAUD. 6
Following initiation of XR-NTX, the present study found that inpatient admissions and ED visits were reduced, while use of outpatient services increased, particularly for AUD-related encounters. This may indicate a transition to less resource-intensive care. While this present study is focused on HCRU only among veterans with AUD treated with XR-NTX, a prior study comparing HCRU across MAUD among veterans over the same timeframe found that XR-NTX (vs oral MAUD) was associated with longer time on treatment and fewer hospitalizations among veterans with AUD.6,7 Additionally, a real-world, retrospective study of US veterans by Kauf et al 8 examined HCRU for patients receiving XR-NTX versus other MAUDs and found that MAUD treatment appeared to reduce inpatient admissions and increase outpatient care, regardless of agent.
The present study has limitations. This VA population was majority male and White, and older on average than the US population; thus, the generalizability of these findings to the general US population with AUD is limited. Patients with co-occurring opioid use disorder were also excluded from the analysis. It is possible that HCRU was underestimated if patients in this study received care outside the VA healthcare system. Incident AUD was approximated and was not necessarily a patient’s first-ever AUD diagnosis. Adherence to XR-NTX may also have been underestimated because XR-NTX administrations provided under the manufacturer’s hospital-based inpatient free trial program were not captured in the data. Additionally, residential treatments were categorized via Healthcare Common Procedure Coding System (either inpatient or outpatient), stop codes (outpatient only) and bed section/location variables (inpatient only), which may differ from other claims databases. It is not possible to determine if the increase in outpatient visits was due to visits related to XR-NTX treatment, as no specific outpatient code exists for XR-NTX treatment in the dataset and different clinics may use different codes. As with all retrospective studies utilizing claims data, there is potential for coding errors or omissions.
Conclusions
In this real-world study of veterans with AUD, reductions in all-cause and AUD-related inpatient admissions and ED visits, and increases in outpatient visits, were observed following XR-NTX initiation. This transition from acute to long-term outpatient care utilization may reflect a reduction in severity of illness. Additional research regarding the impact of XR-NTX on health and functional outcomes in patients with AUD is warranted.
Footnotes
Acknowledgements
Medical writing support was provided by Rebecca Jarvis, PhD of Envision Pharma Group and funded by Alkermes, Inc. This manuscript was developed in accordance with Good Publication Practice 2022 guidelines.
Funding:
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was sponsored by Alkermes, Inc. Authors had full control of the content and made the final decision on all aspects of this publication.
Declaration of conflicting interests:
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: RG is an employee of Alkermes, Inc. JL, AKO and MAS and were employees of Alkermes, Inc., during the conduct of the study and may own stock in the company. SS, ES and AL are employees of Analysis Group, Inc., a consulting firm that received funding from the study’s sponsor to conduct this study. KW serves on the Scientific Advisory Board for Pear Therapeutics and is a member of the Alcohol Clinical Trials Initiative (ACTIVE) Workgroup, which has been supported previously, but not in the past 36 months, by Abbott/AbbVie, Amygdala Neurosciences, Arbor Pharmaceuticals, GSK, Indivior, Janssen, Lilly, Pfizer, and Schering Plough; but in the past 36 months, its activities were supported by Alkermes, Inc., Dicerna, Ethypharm, Lundbeck, Mitsubishi, and Otsuka. SL, LS and KD have no financial conflicts of interest to disclose.
Author Contributions
RG, JL, AKOS, ES, MAS, KW, KD: Developed conceptualization of the study. JL, AKOS, SS, ES, ALSL, LS: Provided resources for the study. SS, AL, SL, LS: Performed study investigation and formal analysis of the data. RG, AKOS, JL, ES: Provided supervision and project administration. ES: performed investigation and data validation. SS, AL: Performed data validation and data curation. RG: Contributed towards writing of the original draft. All authors contributed to the study methodology and manuscript writing and editing. All authors reviewed and approved the final draft.
Ethics Approval and Consent to Participate
Institutional Review Board approval for this study was received by the Research & Development Committee (R&DC) of the Southeast Louisiana Veterans Health Care System (Study ID: 664). As these data are deidentified and there was no contact with patients, informed consent was not needed for this study.
