黑料网

Chronic pain; Insurance claims; Buprenorphine buccal film; Buprenorphine patch; Opioids; Safety; Adverse events

Introduction

According to the World Health Organization (WHO), chronic Low Back Pain (cLBP) affects over six hundred million people worldwide, representing one of the leading causes of disability globally [1]. Most people experience chronic back pain at least once over their course of life [1, 2]. The prevalence of cLBP increases with age, affecting 20-25% of the population aged 65 or older [1, 3, 4]. In older individuals, cLBP is characterized by more frequent and longer pain episodes that require continuous monitoring [4]. Persisting pain reduces the ability to participate in family, social, and routine daily activities, negatively impacting overall well-being and mental health [1, 4]. Clinical management of cLBP requires substantial resources, particularly in the geriatric population. The estimated economic burden of cLBP in the United States is over $100 billion annually and is expected to rise with the fast-growing aging population [4].

Management of cLBP in older adults is challenging due to the high prevalence of comorbidities, multiple medication regimens, slower metabolism, and increased fall risk [5]. The current WHO guideline for treating older adults with cLBP recommends a risk-stratified approach, predominantly based on multimodal non-pharmacological strategies, such as education programs, exercises, and physical and psychological therapies [6]. Still, pharmacological treatment remains an essential part of pain management, though it is important to consider medication pharmacokinetic, safety profiles, and individual comorbidity when selecting the appropriate treatment regimen. Pharmacotherapy for cLBP in older patients employs lower therapeutic doses considering the frequency of polypharmacy, co-morbid medical disorders, and potentially decreased renal and hepatic metabolism. Although used judiciously in elderly due to the increased morbidity and mortality risk, opioids remain a mainstay in persistent refractory severe cLBP treatment [5]. The rate of opioid prescribing in the elderly population is high; with one in four Americans aged 65 and older prescribed at least one opioid in 2017, according to the Centers for Disease Control and Prevention [7]. Common side effects of opioid use in the elderly population include constipation, urinary retention, and cardiovascular and endocrine disorders [8], while central nervous system adverse events and respiratory depression represent the most serious conditions that may lead to a higher risk of death in this population [9]. Older patients with chronic pain are at an increased risk of opioid use overdose and abuse. According to the Centers for Medicare and Medicaid Services, more than 6 of every 1,000 Medicare beneficiaries were diagnosed with opioid use disorder [9], with considerable opioid-related mortality [10].

Buprenorphine represents a valuable therapeutic alternative for elderly patients with cLBP due to the potent and sustained analgesia, resulting from a very high binding affinity and partial agonism at the µ-opioid receptor, as well as the prolonged pharmacokinetic characteristics of transdermal and transmucosal delivery platforms [11, 12]. Buprenorphine is also considered to potentially have a milder safety profile than full µ-opioids, including lower rates and severity of constipation and urinary retention, lower risk of respiratory depression, and limited abuse potential [13]. In 2022, the United States Departments of Defense and Veterans Affairs added buprenorphine to the clinical practice guideline for the use of opioids as a first-line treatment for chronic pain due to its relatively lower risk for overdose and misuse [14]. Importantly, the pharmacokinetic and safety profile of buprenorphine is not altered by patient age due to the rapid conversion to non-active conjugates [13]. The bioavailability of orally administered buprenorphine is low, approximately 10%. However, the issue has been resolved by using the alternative routes. Transdermal patches provide approximately 15% and buccal film 46–65% bioavailability, while both formulations bypass first-pass metabolism [15].

Studies directly comparing buprenorphine to oral opioids for chronic pain in terms of their safety in the elderly population are lacking, although the topic is particularly important given the current opioid public health emergency [9, 11]. This study aimed to explore and compare Treatment-Emergent Adverse Event (TEAE) rates while using Schedule III (CIII) buprenorphine (Belbuca^® and buprenorphine patch) and Schedule II (CII) oral opioids among Medicare patients with cLBP using real-world claims data. Additional sub-analyses investigated safety outcomes between Belbuca^® vs. CII oral opioids and separately Belbuca^® vs. buprenorphine patches.

Methodology

Data source

This retrospective cohort study used US insurance claims data from the Merative MarketScan^® Medicare Supplemental and Coordination of Benefits Database. The Merative MarketScan^® databases consist of deidentified, longitudinal, patient-level closed claims and specialty data for patients in the US sourced directly from a diverse pool of payers. This study focuses on retirees with employer-sponsored Medicare Supplemental and Medicare Advantage plans and includes drug information and outcomes data for healthcare services performed in both inpatient and outpatient settings [16]. The study was performed in insurance data claimed in the period from January 1, 2018, to December 31, 2021.

Study population

The study observed Medicare beneficiaries (≥ 65 years of age) with cLBP defined as at least two diagnoses of low back pain on different dates during the six-month pre-index period. The diagnoses were identified using the International Classification of Diseases – Clinical Modification (ICD-10-CM) codes (Supplement Table A1). The population of interest considered for the study were patients prescribed oral CII opioids or CIII buprenorphine (Belbuca^® or buprenorphine patch). The respective drug codes were identified in the database based on the National Drug Codes (NDCs) (Supplement Table A2,A3). Patients were treatment-naïve, with no CIII buprenorphine and CII oral opioids in the 6-month pre-index period. Exclusion criteria were a gap in the health plan or pharmaceutical coverage during the observational period and Belbuca^® or buprenorphine patch prescriptions within the CII opioid cohort. To enable a fair comparison, CII opioid patients that received Belbuca^® or buprenorphine patch prescriptions in the post-index period were excluded from the cohort. On the other hand, CIII buprenorphine patients were allowed to have concomitant CII opioid use during the post-index period. This way, the study ensured that potential selection bias is conservative, with possibly overestimated rates of adverse events in CIII buprenorphine cohort.

Study design

The index date was defined as the first date of buprenorphine or oral CII opioid treatment. The analysis observed 6-month pre-index period and the post-index period lasted until the end of continuous healthcare and pharmaceutical coverage. Clinical characteristics were evaluated in the pre-index period, while demographic characteristics were assessed on the index date. Patients were classified into cohorts based on the index prescription. The primary analysis compared CIII buprenorphine (prescribed Belbuca^® or buprenorphine patch) and oral CII opioid patients (short-acting [SAO] and long-acting [LAO]), while sub-analyses considered the comparison of Belbuca^® vs. CII oral opioids and Belbuca® vs. buprenorphine patch patients. The study design is presented in Figure 1.

Figure 1: Study design.

Outcome measures

The main study outcome was the incidence of TEAEs. The list of relevant adverse events was comprised based on the published literature sources including the most common adverse events (≥5% rate), serious adverse events, adverse events leading to treatment discontinuation, and opioid-related adverse events [17-26]. Finally, the list comprised 44 relevant TEAEs reported in Table 1.

Table 1: List of captured adverse events of interest.

Diagnoses of TEAEs were identified in the database based on the ICD-10-CM codes (Supplement Table A4). The diagnosis was considered as TEAE if occurred during the treatment period with index medication and only among patients without a history of investigated TEAE during the pre-index period. The treatment period was calculated by summing up the drug supply periods for the index medication, excluding treatment gaps and extracting the days of prescription overlaps. In the CIII buprenorphine cohort, only buprenorphine prescription claims were considered when defining the treatment period although patients were allowed to use oral CII opioid concomitantly. The scheme of treatment period definition is presented in Figure 2. Repeated events experienced by a patient during the post-index period were counted as separate events. All-grade TEAEs were all relevant diagnoses observed within treatment periods, while serious TEAEs were defined as events claimed in inpatient or Emergency Department (ED) settings. TEAE rates were reported per 1,000 person-years. The comparison between the cohorts was performed using absolute Incidence Rate Difference (IRD) and Incidence Rate Ratio (IRR). The IRD is calculated as a crude difference of the observed incidence rates (i.e., cohort I incidence rate minus cohort II incidence rate), while IRR represents a relative difference measure calculated as a quotient of incidence rates (i.e., cohort I incidence rate divided by cohort II incidence rate).

Figure 2: Treatment period definition scheme.

Sub-analyses

The primary analysis compared study outcomes between CIII buprenorphine (Belbuca^® and buprenorphine patch) vs. oral CII opioid (SAO and LAO) cohorts. Sub-analyses aimed to assess the outcomes in more granularly stratified populations. Sub-analysis #1 compared TEAE rates between Belbuca^® vs. CII oral opioid cohorts, while sub-analysis #2 compared Belbuca^® vs. buprenorphine patches.

Statistical analysis

Descriptive statistics was reported summarizing continuous variables as means with standard deviations and categorical variables as numbers with proportions of the sample. The independent t-test was performed to test the difference between the compared cohorts for continuous variables, while chi-square test of independence was used for categorical variables. P-values lower than 0.05 were considered statistically significant.

All TEAE rates and IRD values were reported per 1,000 person-years, while IRR was reported as a rate ratio with 95% confidence intervals (95% CI). The incidence rate ratio test computed IRD and IRR and explored the statistical significance of TEAE rate differences between study cohorts. If TEAE occurred in only one cohort, the P-value was reported for IRD, otherwise it refers to the statistical difference between the cohorts in IRR Negative IRD values and IRR less than 1 imply that the TEAE rate was lower in the referent cohort (CIII buprenorphine in primary analysis and Belbuca^® in sub-analyses).

To control for confounders and minimize the selection bias, the Propensity-Score Matching (PSM) analysis was performed applying the nearest-neighbour matching algorithm. Demographic and clinical characteristics of patients were used as covariates in the matching process.

Statistical analyses were performed using the IBM Statistical Package for the Social Sciences (SPSS^®) and MedCalc^® statistical software.

Results

The final sample of patients before PSM consisted of 26,995 patients (1,020 in CIII buprenorphine cohort and 25,975 in CII opioid cohort). The CIII buprenorphine cohort was composed of 203 patients treated with Belbuca^® and 817 patients on buprenorphine patch. In the sample of 1,496 matched patients, 545 patients were treated with CIII buprenorphine and 951 patients with CII opioids. The subgroup analyses considered 421 patients (124 Belbuca^® matched to 297 CII opioid patients) and 124 patients (62 patients in both Belbuca^® and buprenorphine patch cohorts). The flow diagram depicting patient selection process is shown in Figure 3.

Figure 3: Patient selection flow diagram.

Primary analysis: CIII buprenorphine vs. CII opioidss

Non-matched population: Demographic characteristics of the total sample of 26,995 patients with cLBP stratified in the treatment cohorts are presented in Table 2. Medicare beneficiaries with cLBP were approximately 75 years old and predominantly females. A higher proportion of males was observed in the CII opioid cohort (34.4% vs. 43.8%, p<0.001). The majority of patients were covered by the Preferred Provider Organization (43.4% vs. 47.5%, p=0.011 in CIII buprenorphine vs. CII opioids). A higher proportion of CIII buprenorphine patients were covered by the Comprehensive plan (33.3% vs. 24.5%, p<0.001), while CII opioid patients were more commonly covered by the Health Maintenance Organization plan (17.2% vs. 20.9%, p=0.004). CII opioid-treated patients mostly resided in the North East region, while the majority of CIII buprenorphine patients were located in the North Central and South regions (p<0.001, respectively). Observing the comorbidity burden, CIII buprenorphine patients had a higher Charlson Comorbidity Index (CCI) than CII opioid patients (2.3 vs. 1.9, p<0.001), with a significantly higher proportion of patients in CCI category 4+ (24.8% vs. 18.0%, p<0.001), while more patients in CII opioid cohort were noted in CCI=0 category (26.7% vs. 35.2%, p<0.001). There were significant differences between study cohorts in almost all CCI components (except moderate or severe liver disease, hemiplegia or paraplegia, metastatic solid tumors, and AIDS/HIV), as well as all mental health disorders and other chronic pain comorbidities (p<0.05), except spine disorders. The list of clinical characteristics is presented in Table 3.

Table 2: Demographic characteristics of non-matched patients.

Table 3: Clinical characteristics of non-matched patients.

Matched population: The sample of matched patients consisted of 1,496 patients (545 patients in CIII buprenorphine cohort and 951 patients in CII opioid cohort). Patients were well-balanced between the study cohorts and there were no differences in demographic and clinical characteristics that could impact study outcome measures (Supplement Tables A5,A6).

The rate of serious hypertension was significantly higher (IRR 0.23, p=0.036) in CII opioid cohort, while other cardiovascular TEAEs were occurred with similar incidence in both cohorts (Table 4). Out of all-grade CNS-related TEAEs, the rates of all-grade confusion, syncope, cerebrovascular accident, and sleep disturbances more commonly occurred in the CII opioid cohort (IRR 0.15, 0.35, 0.42, and 0.31, p<0.050, respectively). The rate of all-grade headache, fatigue and dehydration was significantly higher among CII opioid patients (IRR 0.18, 0.71, and 0.52, p<0.050). The rate of all-grade constipation was significantly higher in the CII opioid cohort (IRR 0.41, p=0.041), while all-grade anorexia/loss of appetite was more frequent in CIII buprenorphine cohort (IRR 2.12, p=0.032). The rates of all-grade osteoarthritis (IRR 0.34, p<0.001), pneumonia (IRR 0.26, p<0.001), cellulitis (IRR 0.52, p=0.026), and urinary discomfort (IRR 0.45, p=0.002) were significantly higher in CII opioid patients.

Table 4: All-grade TEAE rates (per 1,000 person-years) during the CIII buprenorphine and CII opioid treatment among matched sample in the primary analysis.

The reported rates of serious hypotension (IRD -30.62 per 1,000 person-years, p=0.036) and atrial fibrillation (IRR 0.19, p=0.001) were significantly lower in the CIII buprenorphine cohort (Table 5). From all investigated serious CNS-related TEAEs, the rates of serous confusion (IRR 0.07, p<0.001), syncope (IRR 0.08, p<0.001), cerebrovascular accident (IRR 0.18, p=0.003), and sleep disturbances (IRD -91.85 per 1,000 person-years, p<0.001) were significantly higher among CII opioid-treated patients. The rate of serious headache was also significantly more frequent among CII opioid patients (IRR 0.11, p=0.002). Out of GIT-related TEAEs, serious constipation and abdominal pain more commonly occurred among CII opioid patients compared to the CIII buprenorphine cohort (IRR 0.17 and 0.45, respectively, p<0.05). Serious bone fractures were more frequent among CIII buprenorphine patients (IRR 5.90, p=0.044), while serious osteoarthritis occurred more commonly in oral CII opioid-treated patients (IRR 0.28, p=0.029). The rates of serious pneumonia, cellulitis, and urinary discomfort appeared to be significantly higher among CII opioid cohort (IRR 0.34, 0.29, and 0.16, respectively, p<0.05).

Table 5: Serious TEAE rates (per 1,000 person-years) during the CIII buprenorphine and CII opioid treatment in the primary analysis.

Sub-analysis #1: belbuca^® vs. CII opioids

Non-matched population: Demographic and clinical characteristics of 26,178 patients treated with Belbuca^® (203 patients) and CII Opioids (25,976 patients) before the PSM are provided in the Supplement (Table A7,A8).

Matched population: A total number of 421 patients were identified in the final sample of matched patients (124 patients in Belbuca^® cohort and 297 patients in CII opioid cohort). As the demographic and clinical characteristics were used as a basis for PSM analysis, there were no statistical differences in these measures implying the cohorts were well-balanced. The list of demographic and clinical characteristics is shown in the Supplement (Table A9,A10).

There was no observed all-grade TEAEs with significantly higher rates in Belbuca^® compared to CII opioid-treated Medicare patients diagnosed with cLBP. Otherwise, it was demonstrated that all-grade atrial fibrillation (IRD -194.33, p=0.010), confusion (IRD -194.33, p=0.010), headache (IRR 0.12, p=0.035), constipation (IRR 0.15, p=0.001), cellulitis (IRD -242.92, p=0.004), and urinary discomfort (IRR 0.11, p<0.001) occurred more frequently in CII opioid cohort. The list of all-grade TEAE rates in sub-analysis #1 matched sample is provided in Table 6.

Table 6: All-grade TEAE rates (per 1,000 person-years) during the Belbuca® and CII opioid treatment among matched sample in sub-analysis #1.

There were no serious TEAEs that occurred significantly more in Belbuca^®- vs. CII opioid-treated Medicare patients diagnosed with cLBP. Significantly lower rates of fatigue (IRR 0.20, p=0.43), constipation (IRR 0.12, p=0.035), osteoarthritis (IRD -340.08, p<0.001), and urinary discomfort (IRD -194.33, p=0.010). All serious TEAE rates among sub-analysis #1 matched sample are listed in Table 7.

Table 7: Serious TEAE rates (per 1,000 person-years) during the Belbuca® and CII opioid treatment among matched sample in sub-analysis #1.

Sub-analysis #2: Belbuca^® vs. Buprenorphine patches

Non-matched population: Demographic and clinical characteristics of 1,020 patients before the PSM, including 203 patients initially prescribed Belbuca® and 817 patients initially prescribed buprenorphine patches, are provided in the Supplement (Table A11,A12).

Matched population: The sample of matched patients included 62 patients in each of the compared cohorts. Patient demographic and clinical characteristics were well-balanced with no remaining statistical differences between the cohorts (Supplement Table A13,A14).

There were differences observed between study cohorts in 7/44 all-grade TEAEs. Belbuca^® treatment was associated with lower rates of all-grade cerebrovascular accident (IRD -248.26 per 1,000 person-years, p=0.038), and opioid abuse and dependence (IRR 0.07, p<0.001), dehydration (IRR 0.08, p=0.002), and nausea and vomiting (IRR 0.25, p=0.025). Contrary, all-grade confusion (IRD 520.38, p=0.004), fatigue (IRR 2.68, p=0.012), and osteoarthritis (IRR 10.71, p<0.001) more commonly occurred in Belbuca^® cohort. The list of all-grade TEAE rates in sub-analysis #2 matched sample is reported in Table 8.

Table 8: All-grade TEAE rates (per 1,000 person-years) during the Belbuca® and buprenorphine patch treatment among matched sample in the sub-analysis #2.

Serious confusion and osteoarthritis occurred only in the Belbuca^® cohort (IRDs 462.56 [p=0.006] and 867.30 [p<0.001] per 1,000 person-years, respectively). However, the rates of serious opioid abuse and dependence (IRD -806.84 per 1,000 person-years, p<0.001), dehydration (IRR 0.08, p=0.002), abdominal pain (IRD -496.52, p=0.003), anorexia/loss of appetite (IRD -372.39, p=0.011), and pneumonia (IRR 0.112, p=0.016) were significantly higher in the buprenorphine patch cohort. All serious TEAE rates in sub-analysis #2 matched samples are reported in Table 9.

Table 9: Serious TEAE rates (per 1,000 person-years) during the Belbuca® and buprenorphine patch treatment among matched sample in the sub-analysis #2.

Discussion

This retrospective study indicates to a more favourable safety and tolerability profile of CIII buprenorphine compared to oral CII opioids in the population of Medicare patients with cLBP based on reporting TEAEs. Out of more than 40 serious TEAEs that were tracked during the treatment, CIII buprenorphine had significantly lower rates of 13 events. Among those, serious hypotension and sleep disturbances occurred only in oral CII opioid patients. On the contrary, only a higher rate of serious bone fractures was reported in the CIII buprenorphine cohort. Results of sub-analysis #1 that compared TEAE rates between Belbuca^® and oral CII opioid cohorts were in line with the findings reported in the primary analysis. Belbuca^® showed significantly lower rates of 4 out of 44 serious TEAEs, and none of the 44 serious TEAEs occurred more frequently in the Belbuca^® cohort. In sub-analysis #2, safety outcomes were compared between different buprenorphine formulations – buccal film (Belbuca^®) and transdermal patches. Results were mostly similar between the cohorts, but Belbuca^®-treated patients yielded a slightly better safety profile with a significantly lower rate of 5 out of 44 serious TEAEs. Serious opioid abuse and dependence, abdominal pain, and anorexia/loss of appetite occurred only in the buprenorphine patch cohort. On the other hand, confusion and osteoarthritis were serious TEAEs with higher rates in the Belbuca^® cohort. This analysis also explored all-grade TEAE rates in the primary analysis and sub-analyses. Still, all-grade TEAE outcomes were considered less reliable than those occurring in the ED or requiring hospitalization, as they may indicate comorbid conditions or coinciding events. Also, all-grade TEAEs were considered less significant for reimbursement purposes from the payer’s perspective.

A recently published study evaluated the safety of CIII buprenorphine compared to CII oral opioids in a younger population of commercially insured adults with cLBP [27]. The study identified more common occurrences of opioid poisoning, constipation, nausea/vomiting, seizures, and coronary artery disease in patients treated with CII oral opioids. Still, significantly greater rates of dizziness, cholecystitis, and, unexpectedly, opioid abuse/dependence were seen among CIII buprenorphine patients. The authors noted that all opioid abuse/dependence events occurred among patients treated with the buprenorphine patch rather than Belbuca^®. In this retrospective study, Belbuca^® patients experienced a milder safety profile than CII opioids, with lower rates of respiratory depression, seizures, osteoarthritis, and atrial fibrillation in patients using the buccal film [27]. A Systematic Literature Review (SLR) evaluated the safety and efficacy of the CIII buprenorphine for chronic pain treatment based on randomized controlled trials [15]. Based on over 30 studies, including four on buprenorphine buccal film and two assessing specifically an elderly population, buprenorphine demonstrated efficacy in pain relief and was generally well tolerated. Based on SLR study results, when juxtaposed with safety data for full µ-opioid receptor agonists, buprenorphine exhibited favourable safety. As the study concluded, the buccal film seemed to confer additional safety advantages compared with the transdermal patch, with lower reported rates of nausea, vomiting, constipation, headache, dizziness, and somnolence [15]. A case series report retrospectively observed cases of clinically significant buprenorphine toxicity in the intensive care, high-dependency unit, and acute pain service databases of an academic hospital in Australia [28]. The study described six cases of buprenorphine-related respiratory and neurological depression among opioid-naïve elderly hospitalized patients with additional risk factors such as concurrent comorbidities and ingestion of other central nervous system depressants. In all six patients, buprenorphine was used for the treatment of acute pain and administered sublingually, with the addition of a transdermal patch for one patient. The study highlights the need for caution when buprenorphine is used in patients with reduced respiratory or neurological reserve and risk factors. While these findings are concerning, it should also be considered that the hospital units manage about 5 thousand patients each year [28], so the fact that six patients with serious buprenorphine-related toxicity were seen over two years suggests the relative rare incidence and the safety profile of buprenorphine. It should also be noted that sublingual forms of buprenorphine are not approved for pain management by the US Food and Drug Administration and the doses in the products are much higher than buprenorphine products approved for use in pain management [29]. Several well-sourced narrative reviews emphasized the suitability of buprenorphine as the preferred treatment for chronic pain in older patients requiring around-the-clock analgesia [11, 12, 29-31]. The reports based their recommendation on favourable pharmacokinetic characteristics and fast clearance that is not affected by patient age, renal failure, or liver impairment. An open-label single-arm study assessed the long-term safety and efficacy of buprenorphine buccal film in patients with moderate to severe chronic pain [32]. Treatment with the buccal film occurred for 48 weeks after the after rollover subjects completed a 12-week clinical trial. Buprenorphine buccal film was well tolerated during the study, with treatment-related adverse events seen in 23% of patients during the titration phase and 14% of patients during the long-term treatment phase. Adverse events led to treatment discontinuation in 3% of patients in both phases. The principal adverse events were nausea, constipation, and headache. No adverse events related to respiratory depression were observed during the study [32]. Finally, the consensus statement of an International Expert Panel on opioids in the management of chronic severe pain identifies buprenorphine as “the top-line choice for opioid treatment in the elderly” based on the distinct benefits in the treatment of cancer pain, non-cancer-related pain, neuropathic pain, its use in patients with renal and hepatic impairment and immunocompromised patient, but also its cost and availability [33].

Strengths and limitations

To our knowledge, this is the first real-world evidence analysis in the US healthcare setting that explored a wide spectrum of TEAEs and compared their rates in the elderly Medicare beneficiaries with cLBP between CIII buprenorphine and oral CII opioids. From a standpoint of the current opioid crisis and the lack of effective approaches to combat ever-rising OUD rates, this study attempts to fill some of the existing knowledge gaps, provides important insights into real-world clinical practice and evidence on the safety of commonly used treatment options, and may serve as a basis for further research and prospective clinical studies. The comprehensive analysis explored safety outcomes in patients treated with particular buprenorphine formulations, Belbuca^® vs. oral CII opioids and Belbuca^® vs. buprenorphine transdermal patches. The criteria applied during the patient selection process and statistical analyses aimed to minimize the research bias and provide reliable comparison between the study cohorts. Patients were required to have continuous insurance coverage in order to minimize the bias due to missing data in retrospective claims. The PSM was used to adjust for the covariates between studied cohorts and minimize their impact on the safety outcomes. Moreover, for patients with a positive history of explored events in the pre-index period, the events occurring during the treatment period were not classified as TEAEs and were not taken into account when reporting the safety outcomes. In this way, more reliable association between the index treatment and TEAEs was established.

However, several study limitations should be considered when interpreting the results. The main limitation is related the nature and characteristics of real-world insurance claims and related restrictions of coding systems. These data are primarily collected for administrative purposes, with possible data entry errors such as miscoding or duplicate claims. The impact of this limitation has been reduced by performing thorough data cleaning, a careful patient selection process, and bias-controlling methods such as PSM. Additionally, the chronicity of low back pain could not be determined based on the diagnostic claims and was assumed in patients who had low back pain diagnosis at two separate claims during the pre-index period. The study findings’ generalizability should be also considered. This retrospective analysis was performed on the Medicare population and results may not be applicable to patients with other types of insurance. Furthermore, the sub-analyses that compared the occurrence of TEAEs within Belbuca^®- vs. CII opioids-, and vs. buprenorphine patch-treated patients yielded statistical significance between cohorts, but they were conducted on small samples of Medicare patients with cLBP (sub-analysis #1: 124 vs. 297 patients, and sub-analysis #2: 62 vs. 62 patients) and the findings should be carefully interpreted. The authors were unable to require Medicare patients to be opioid-naïve in the pre-index period, as almost all cLBP patients in our analysis had a positive history of CII opioids utilization. Finally, although we tried to enhance the association between the treatment and TEAEs by capturing only events that occurred while supplied with the medicine and in the patients with no history of investigated TEAE during the pre-index period, we were unable to confirm that the event was caused by the treatment in the insurance claim database. Additionally, as buprenorphine patients were allowed to have concomitant CII opioid treatment, there is a possibility that some of the serious TEAEs were caused by CII opioids or their concomitant use in the buprenorphine cohort. In addition, the impact of average daily doses between CII opioids and CIII buprenorphine cohorts could not be considered since buprenorphine has no MME conversion factor value. No formal conclusions comparing safety between CII opioids and CIII buprenorphine may be drawn given no head-to-head clinical trials have been conducted.

Conclusion

This study demonstrated a higher occurrence of serious TEAEs in the CII opioid cohort vs. the buprenorphine cohort including confusion, syncope, headache, urinary discomfort, constipation, cerebrovascular accident, atrial fibrillation, osteoarthritis, cellulitis, pneumonia, abdominal pain, sleep disturbances, and hypotension. In general, CIII buprenorphine also seems better tolerated among the Medicare population than CII opioids, with the exception of the higher rate of serious bone fractures. Based on the real-world retrospective claims, Belbuca^® may be indicative of better safety and tolerability than CII opioids, with lower rates of serious TEAEs including fatigue, constipation, osteoarthritis, and urinary discomfort. No serious TEAEs were significantly more common in Belbuca^® than CII opioids. Finally, Belbuca^® may have significantly higher rates of confusion and osteoarthritis, while significantly lower rates of serious TEAEs of opioid abuse and dependence, abdominal pain, loss of appetite, dehydration, and pneumonia compared to the buprenorphine patch.

Disclosure

This study was funded by Collegium Pharmaceutical, Inc. 

1. World Health Organization (2023) .
2. DelSole E, Warnick E, Galetta MS, Divi SN, Goyal DKC, et al. (2021) . Topics in Pain Management 36.

, Crossref

3. Vadalà G, Russo F, De Salvatore S, Cortina G, Albo E, et al. (2020) . J Clin Med: 9.

, ,

4. de Souza IMB, Sakaguchi TF, Yuan SLK, Matsutani LA, do Espírito-Santo AS, et al. (2019) . Clinics (Sao Paulo) 74: e789.

, ,

5. Fu JL, Perloff MD (2022) . Drugs Aging 39: 523-550.

, ,

6. World Health Organization (2023) .
7. Centers for Disease Control and Prevention (2019) .
8. Chau DL, Walker V, Pai L, Cho LM (2008) . Clin Interv Aging 3: 273-278.

, ,

9. Tilly J, Skowronski S, Ruiz S (2017) .
10. Larney S, Bohnert AS, Ganoczy D, Ilgen MA, Hickman M, et al. (2015) . Drug Alcohol Depend 147: 32-37.

, ,

11. Adler J, Mallick-Searle T, Garofoli M, Zimmerman A (2024) . J Multidiscip Healthc 17: 1375-1383.

, ,

12. Vadivelu N, Hines RL (2008) . Clin Interv Aging 3: 421-430.

, ,

13. Khanna IK, Pillarisetti S (2015) . J Pain Res 8: 859-870.

, ,

14. Sandbrink F, Murphy JL, Johansson M, Olson JL, Edens E, et al. (2023) . Ann Intern Med 176: 388-397.

, ,

15. Pergolizzi JV, Jr., Raffa RB (2019) . J Pain Res 12: 3299-3317.

, ,

16. Merative (2023) .
17. Papaleontiou M, Henderson CR, Jr., Turner BJ, Moore AA, Olkhovskaya Y, et al. (2010) . J Am Geriatr Soc 58: 1353-1369.

, ,

18. Oosten AW, Oldenmenger WH, Mathijssen RH, van der Rijt CC (2015) . J Pain 16: 935-946.

, ,

19. Moore RA, McQuay HJ (2005) . Arthritis Res Ther 7: R1046.

, ,

20. Furlan AD, Sandoval JA, Mailis-Gagnon A, Tunks E (2006) . Cmaj 174: 1589-1594.

, ,

21. Food and Drug Administration (2019) .
22. Food and Drug Administration (2019) .
23. Food and Drug Administration (2021) .
24. Els C, Jackson TD, Kunyk D, Lappi VG, Sonnenberg B, et al. (2017) . Cochrane Database Syst Rev 10: Cd012509.

, ,

25. Eisenberg E, McNicol ED, Carr DB (2005) Efficacy and safety of opioid agonists in the treatment of neuropathic pain of nonmalignant origin: systematic review and meta-analysis of randomized controlled trials. Jama 293: 3043-3052.

, ,

26. Busse JW, Wang L, Kamaleldin M, Craigie S, Riva JJ, et al. (2018) . Jama 320: 2448-2460.

, ,

27. Stanicic F, Grbic D, Vukicevic D, Zah V (2024) . J Comp Eff Res: e230183.

, ,

28. Richards S, Torre L, Lawther B (2017) . Anaesth Intensive Care 45: 256-261.

, ,

29. Hale M, Garofoli M, Raffa RB (2021) . J Pain Res 14: 1359-1369.

, ,

30. Dalal S, Chitneni A, Berger AA, Orhurhu V, Dar B, et al. (2021) . Health Psychol Res 9: 27241.

, ,

31. Gudin J, Fudin J (2020) . Pain Ther 9: 41-54.

, ,

32. Hale M, Urdaneta V, Kirby MT, Xiang Q, Rauck R (2017) . J Pain Res 10: 233-240.

, ,

33. Pergolizzi J, Böger RH, Budd K, Dahan A, Erdine S, et al. (2008) . Pain Pract 8: 287-313.

, ,