Rescue Analgesia for Opioid-Dependent Individuals on Opioid Agonist Treatment during Hospitalization: Adherence to Guideline Treatment

Introduction: Opioid agonist treatment (OAT) is the first-line treatment for opioid use disorder (OUD). Simultaneously, opioids are essential medicines in acute pain management. The literature is scarce on acute pain management in individuals with OUD, and guidelines are controversial for patients on OAT. We aimed at analyzing rescue analgesia in opioid-dependent individuals on OAT during hospitalization in the University Hospital Basel, Switzerland. Methods: Patient hospital records were extracted from the database over 6 months (Jan-Jun) in 2015 and 2018. Of the 3,216 extracted patient records, we identified 255 cases on OAT with full datasets. Rescue analgesia was defined according to established principles of acute pain management, e.g., i) the analgesic agent is identical to the OAT medication, and ii) the opioid agent is dosed above 1/6th morphine equivalent dose of the OAT medication. Results: The patients were on average 51.3 ± 10.5 years old (range: 22–79 years), of which 64% were men. The most frequent OAT agents were methadone and morphine (34.9% and 34.5%). Rescue analgesia was not documented in 14 cases. Guideline-concordant rescue analgesia was observed in 186 cases (72.9%) and consisted mostly of NSAIDs, including paracetamol (80 cases), and identical agents such as the OAT opioid (70 cases). Guideline-divergent rescue analgesia was observed in 69 (27.1%) cases, predominantly due to an underdosed opioid agent (32 cases), another agent other than the OAT (18 cases), or contraindicated agents (10 cases). Discussion: Our analysis suggests that rescue analgesia in hospitalized OAT patients was predominantly concordant with guidelines, while divergent prescriptions seemed to follow common principles of pain medicine. Clear guidelines are needed to appropriately treat acute pain in hospitalized OAT patients.


Introduction
The WHO lists opioids as essential medicines for the treatment of acute pain, pain caused by cancer, and opioid use disorder (OUD), although access is often inadequate [1]. OUD is defined as a "problematic pattern karger@karger.com www.karger.com/ear of opioid use leading to clinically significant impairment or distress" [2] and involves misuse of heroin and prescribed opioid medications. In Switzerland, which counts 8.5 million inhabitants, as well as in Europe, the prevalence of OUD is stable at 0.1-0.8% [3,4]. Opioid agonist treatment (OAT) is the first-line and the most commonly used treatment for OUD [4]. OAT effectively minimizes harms related to illicit opioid use, especially reducing illicit substance use, overdoses, the transmission of bloodborne viruses, hospital-treated incidents of somatic diseases, and mortality [5]. In Switzerland, the prescription of opioids for OUD has been regulated by federal and state laws since 1975. All physicians may prescribe methadone, buprenorphine, or long-acting morphine for oral OAT, but each patient must be registered by health authorities [6]. In duly justified cases, a formulation of codeine and dihydrocodeine can be prescribed for OAT. Moreover, heroin-assisted treatment involving supervised injectable diacetylmorphine or oral diacetylmorphine is also available as a regular OAT option for those who could not benefit from or be reached by conventional OAT or other treatments. Fentanyl, hydromorphone, oxycodone, and pethidine are not approved for OAT. OAT is covered by health insurance and is provided by office-based practitioners and specialized treatment centers. It is available for opioid-dependent individuals aged 18 years or older who show recent opioid use on toxicology screening. Currently in Switzerland, almost 80% of people with OUD are enrolled in OAT, which represents one of the highest levels of coverage worldwide [7]. No restrictions exist regarding duration of treatment and dosing. However, the dosage should be tailored to the individual needs of the patient, including withdrawal symptoms, subjective comfort, and concomitant heroin or other opioid use. In 2018, an estimated 18,000 patients received oral OAT throughout Switzerland, while approximately 1,600 obtained diacetylmorphine (pharmaceutical heroin) in specialized centers [6].
Due to the aging of the OAT population, health conditions and treatment needs have changed. Thus, next to diseases linked to the use of illicit drugs, such as infectious diseases, a large range of somatic comorbidities has emerged, including cardiovascular diseases, cancer, kidney, and lung disease [8]. Pain can result from problems on a non-acute basis (e.g., somatic diseases) or acute basis (e.g., injuries). In a Norwegian cohort of 156 long-term OAT patients, more than 50% of the participants reported joint or chest pain [9]. Even if the role of opioids in the treatment of pain has been challenged due to the opioid crisis, they still represent a cornerstone in pain management [10]. However, pain management remains challenging in OAT patients due to high opioid tolerance, cross-tolerance between opioids, and individual differences in the required dose [11,12].
Among other standards of care, effective attenuation of pain is crucial in patient care [13], independently of the presence or absence of pain in the medical history. However, in hospital settings, individuals with OUD are less likely than other patients to receive effective analgesics and to obtain adequate pain relief [14,15], probably due to knowledge deficiencies, misinformation, and misconceptions or preconceptions of hospital health professionals about accepted principles of pain management practice and pain issues [16]. Thus, common misconceptions among health providers are that OAT already provides sufficient analgesia, that relapse to illicit opioid use may result from prescribing opioid analgesics, that the additional use of opioid analgesics in OAT patients may lead to respiratory and central nervous system depression, and that complaints of pain may represent a manipulative way to obtain opioids [11]. Independent of the reasons for undertreated pain in individuals with OUD, treatment of pain in this population remains challenging because of the concomitant use of opioids for OAT, the presence of comorbidities, and the lack of evidence-based guidelines for pain management among these patients [17][18][19].
The literature is scarce on acute pain management in individuals with OUD, and recommendations concern mostly patients receiving methadone-or buprenorphinebased OAT [20,21]. Similarly, doses of medication asneeded for episodic or breakthrough pain (so-called rescue medication) in individuals on OAT are rarely mentioned [22]. Thus, most Swiss clinicians follow the general recommendations on pain management that are available, such as that individuals on OAT should obtain proper pain treatment as any other patient [14] and that hospitalized patients receiving OAT should continue OAT at the usual dosage during inpatient treatment [23] as abstinence should not be enforced [19,23]. Further but without evidence in opioid-dependent individuals, pain should be aggressively treated with conventional analgesics (i.e., paracetamol, NSAIDs, steroids, COX-2 inhibitors, coanalgesics such as gabapentin/pregabalin) in a multimodal way [11,19,21,24]. Combination products containing fixed doses of an opioid and paracetamol should not be used [11]. Although no literature exists on acute pain treatment when the OAT agent is diacetylmorphine, pharmacological reasoning suggests to adopt the same principles as with long-acting morphine [19].
The present study aimed at investigating whether acute pain treatment of opioid-dependent individuals on OAT 254 during a hospital stay is concordant with existing guidelines. We used rescue medication as proxy.

Materials and Methods
We performed a secondary analysis of existing hospital data from the University Hospital Basel (UHBS), a 600-bed hospital serving about 1 million residents in the Northwest of Switzerland. Data were extracted from paper and electronic patient records.

Patient Selection
The electronic patients' data management system of the UHBS was searched for eligible records with the terms "mental and behavioral disorder" and "dependence syndrome" in the field "diagnose," independently of the substances used (i.e., cannabinoids, cocaine, hypnotics, opioids, sedatives, other stimulant and psychotropic agents), and the mention of OAT. A selection was extracted with the qualifier "opioid use disorder." We chose two 6-month periods (01/Jan/2015 to 30/Jun/2015 and 01/Jan/2018 to 30/Jun/2018) covering the same seasons to avoid selection bias due to influenza-related admissions. We excluded the years 2016/2017 to avoid missing information because a comprehensive electronic information system including patient records had been progressively rolled out since 2016. Full electronic patient records are available since 2018 [25].

Extracted Data
One researcher (M.K.) extracted patient data (sex, age, year of hospitalization), medical data (reason for hospitalization, main diagnosis, length of hospital stay, leaving hospital against medical advice), and pharmacotherapy data (current OAT treatment, rescue analgesia including agent[s], strength, route and frequency of administration, administration of other medicines) from the patient records. In case of split doses, the amount for 24 h was calculated. Extracted items and sentences or fragments of text from medical and nursing records were copied and pasted manually into an Excel spreadsheet on a personal laptop that was password-protected. All entered data were anonymized. All files were stored on a passwordprotected server with access to study investigators only.

Outcome Measures and Conversion to Morphine Milligram Equivalents
The primary outcome is the concordance of rescue analgesia with guidelines, which comes true when at least one of the following occurs: 1) the rescue opioid analgesic agent is identical to the OAT medication [19,20,26,27]; 2) the rescue opioid analgesic agent is not contraindicated ( Table 1); 3) the daily dose of the rescue opioid analgesic agent corresponds to at least 1/6th of the OAT medication in morphine equivalent dose (MED) [28]. MED was calculated according to the recommendations of the UHBS [29] (Table 2); less than 1/6th of the OAT MED was defined as underdosed rescue opioid analgesic; 4) the rescue analgesic agent is a nonsteroidal anti-inflammatory analgesic (NSAIDs such as paracetamol or metamizole); 5) there is no rescue analgesia, and the nurse calls the physician on duty to prescribe pain medication; 6) when the OAT medication is not approved in this indication (i.e., codeine, fentanyl, oxycodone, pethidine), the rescue analgesic agent is morphine in any dose.

Statistical Analysis
Sentences and fragments of patient medical and nursing records were categorized inductively into categories. Entries of a randomly selected sample of 5% of the patients were double-checked for plausibility and correctness. Data were stored and analyzed using Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA, USA). Descriptive statistics were used and presented as percentages, means, and medians where appropriate. Hospital stay outlier was defined as values that fall outside the distribution. Comparison of the demographic variables between the two cohorts (2015 and 2018) was done with χ 2 and t tests as appropriate. A p value below 0.05 was considered significant. Because patients may have more than one treatment case in one or both observation periods, we analyzed patient data or cases where appropriate.

Results
Of the 3,216 eligible patient records, 413 were identified with the qualifier "opioid use disorder." After eliminating 45 duplicates and excluding 81 erroneously coded cases (emergency department visit without Table 1. Analgesic opioids that may be prescribed in patients on opioid agonist treatment (OAT) with Yes: combination is recommended; (Yes): combination is possible but alternative exists; No: combination is not recommended (reason in parentheses), adapted from [19] and [28] OAT hospitalization, incomplete datasets) and 27 patients not engaged in OAT at hospital admission, we extracted 260 cases. Data analysis was performed with 255 cases with full datasets after exclusion of further three cases with a missing OAT agent in the hospital records and two cases with confusing rescue analgesics due to constant changes in agents and dosage. Cases were equally distributed over the 6-month periods in 2015 (98 cases) and in 2018 (157 cases). There were no major differences in demographic variables, such as sex, age, and diagnosis, or in the OAT medication or rescue analgesics so that pooled data are presented. Mean duration of stay was significantly shorter in 2018 (9.7 ± 9.7 days) than in 2015 (13.5 days ±13.4; p = 0.008).
The mean age of the patients was 51.3 ± 10.5 years (range: 22-79 years); 163 (64%) were men. Most admissions to the hospital occurred via the emergency department in 207 (81.2%) cases or electively in 47 (18.4%) cases. The most frequent reasons for hospitalization were infections, such as abscess, pneumonia, and sepsis (39.2%); exacerbation of chronic obstructive pulmonary disease (8.2%); liver or gallbladder disorder (7.1%); cancer (5.9%); or a fracture due to a fall (3.1%). Pain accompanied the reason for hospitalization in 65 (25.5%) cases. The mean length of hospital stay was 10.8 ± 9.9 days (range: 1-77 days; median: 8 days) after eliminating an outlier with a stay of 100 days. This patient had been transferred from the University Psychiatric Clinics Basel with multiorgan failure.
Autonomous discharge from hospital against medical advice occurred in 14 (5.5%) patients.

Opioid Agonist Treatment
Of the 255 patients with the mention that an OAT had been established before hospital admission, the documented OAT agent was not approved in 14 cases (pethidine in 8 cases; oxycodone in 4 cases; fentanyl and codeine in 1 case each). Information on the dosage of the OAT agent was missing for 10 cases (5 morphine; 4 methadone; 1 oxycodone). A single opioid for OAT was mentioned in 202 (79.2%) cases, of which 12 (4.7%) of them were diacetylmorphine. The administration of two different OAT agents to prevent nighttime withdrawal symptoms was observed in 53 (20.8%) cases, of which 46 (86.8%) had diacetylmorphine as the OAT agent, together with morphine (31/46; 67.4%) or methadone (15/46; 32.6%). The remaining pairs were methadone with morphine (4/53) or oxycodone (1/53) and morphine with pethidine (2/53).
The most frequently prescribed opioids for OAT were methadone (89; 34.9%) and morphine (88; 34.5%), mostly in liquid and extended-release form, respectively (Table 3). Mean MED (excluding diacetylmorphine) was 702.1 ± 607 mg daily (range: 12-3,120 mg). The mean dose of diacetylmorphine was 430 ± 238 mg (range: 50-1,000 mg) fractioned into two daily administrations. Diacetylmorphine was mostly administered orally (26 cases; Table 3). The factors are to be used in the formula: strength per unit X number of units per day X MED conversion factor = MED per day. Example: Fentanyl TTS 100 μg/h corresponds to morphine 240 mg/day (factor 2.4). For diacetylmorphine (all routes of administration), the conversion factor was extrapolated from an equivalence factor between methadone and heroin of 4.85 published by Seidenberg [32]. sc, subcutaneous; im, intramuscular; iv, intravenous.

Rescue Analgesia
Of the 255 cases, 14 had explicitly no documented rescue analgesics, of which 6 were prescribed the OAT agent diacetylmorphine; 5, methadone; 2, morphine; and 1, the non-approved OAT agent pethidine. Four records mentioned that the nurse had to contact the ward physician on duty when the patient was complaining about pain. Opioid rescue analgesia was prescribed in 161 (63.1%) cases. Morphine was the most often prescribed opioid rescue analgesic agent (114 cases; in 70 cases as a single agent; Table 3) at a mean dose of 284.8 mg daily (range: 10-2,400 mg/day). Methadone was the opioid rescue analgesic agent in 21 cases (as a single agent in 11 cases; Table 3) at a mean dose of 59 mg daily (range: 4-200 mg; data not shown). NSAIDs, including paracetamol, were prescribed as single therapeutic class in 80 (31.3%) cases in the form of paracetamol (51 cases; 45 as a single agent and 6 with ibuprofen); in the form of metamizole (22 cases; 13 as a single agent and 9 either with paracetamol [7 cases] or ibuprofen [2 cases]); or in the form of ibuprofen alone (7 cases; data not shown). NSAIDs, including paracetamol, were prescribed in addition to any opioid in 62 cases (Table 3). No co-analgesics were prescribed.

Discussion
This secondary analysis of hospital records from OAT patients in an acute care setting showed that prescription of rescue analgesia was concordant with guidelines in 72.9% of cases. Mainly, the analgesic agent was an NSAID, including paracetamol, or identical to the OAT agent. The principles to treat pain in patients receiving OAT with another opioid other than the OAT agent [19,26,30], that is, with a short-acting opioid [31,32], have been advocated to clearly separate OUD treatment from the treatment of pain [19]. In our study, continuing preexisting OAT at unchanged doses and administering supplemental analgesia with a different opioid were strictly observed in 18 (7.1%) cases. This reflects probably the controversy of published recommendations on the management of acute pain in patients receiving OAT, e.g., in Australia where a non-opioid approach (aspirin, paracetamol, NSAIDs) should be considered first, but adding an opioid analgesic or increasing the dose of the OAT medication up to 30% might also be appropriate [33]. In Switzerland, dose escalation of the OAT medication is recommended before an additional opioid is given or the OAT opioid is rotated [34].
The high rate of guideline-concordant rescue analgesia observed in this study should be interpreted in light of further clinical principles. NSAIDs, including paracetamol, were prescribed as a single therapeutic class in 80 cases (as a single or multiple agent) and in addition to any opioid in 62 cases, which corresponds to 55.7% of all cases. Acute pain treatment with paracetamol, ibuprofen, or metamizole has some legitimacy in individuals on OAT; however, opioids remain as the agent of choice to effectively relieve moderate-to-severe pain [11]. Thus, when focusing on rescue opioids, the rate in the presented study dropped considerably to 92 (36.1%) guideline-concordant treated cases, with morphine taking the lion's share. Nevertheless, morphine and methadone were correctly added as opioid rescue analgesics in individuals engaged in OAT with these agents, that is, in 78.4% of individuals with the OAT morphine and in 58.4% of individuals with the OAT methadone, which corresponds to the current recommendation of maintaining the OAT agent and increasing the dose in case of pain [34].
Although pain severity data could not be considered in the analysis, the present data suggest that OAT individuals might be at increased risk for acute pain undertreatment during hospitalization. Addressing this risk of undertreatment is crucial, considering that undertreating acute pain is likely to cause anxiety, increase tension between patients and hospital staff, impede subsequent pain control [11], or encourage hospital discharge against medical advice [35]. In a qualitative study with 30 individuals on methadone maintenance treatment in Canada and hospitalized in the past 12 months [36], participants reported inadequate pain management, especially at dosages that did not account for their needs. Finally, inadequate pain management is a general issue in OAT patients across various care settings, including outpatient care [37].
In this study, rescue analgesia diverged from guidelines in 27 (14) po (8) po (6) tab ( In case of multiple opioids, the agent with the highest dose is considered; in case of multiple opioids including diacetylmorphine, the latter is considered; in case of dose searching, the last dose is considered. ER, extended release; IR, immediate release; liq, liquid; tab, tablet; po, oral; par, parenteral. from underdosing of the analgesic opioid to less than 1/6th of the OAT medication in MED or adding 1 or even 2 different opioids in sufficient doses. Because opioid rotation was not explicitly mentioned in the patient records, we suppose that prescribing a rescue opioid agent that is different from the OAT medication is more likely to represent a misconception in the treatment of pain in individuals with OAT. In the 10 contraindicated cases, buprenorphine was the most often involved agent (6 cases) and was coprescribed to morphine or methadone. The prescription of rescue morphine or methadone in sufficient doses with buprenorphine as the OAT agent might indicate a strategy to rotating the substitute. However, the strong affinity of buprenorphine for mu receptors causes tight binding and therefore competition at the mu receptor, displacing other opioids, such as methadone and morphine. Thus, switching buprenorphine to another opioid would implicate a "washout" phase with a high probability of withdrawal symptoms. In this sense, some authors recommend cessation of buprenorphine when more than 2 days of acute pain management are needed [38]. Thus, selecting morphine as rescue opioid analgesic in case of buprenorphine as the OAT agent is likely to represent a misconception in the treatment of pain in individuals with OAT. However, controversy exists about standards, and some contraindicated combinations may not be universally accepted in practice. As an example, some authors demonstrated that patients on both buprenorphine and methadone can effectively and safely be managed with morphine or fentanyl in the first 24 h after surgery [39]. Nevertheless, the results of this study suggest that rescue analgesia was primarily based on common pain guidelines with treatment forms of low or no addiction potential [34]. Underdosing and inadequate prescription of a contraindicated analgesic agent or of 2 different opioids are alarming and denote a certain misconception of acute pain treatment in general. However, this knowledge could be easily acquired by continuous training. Hospital discharge against medical advice was observed in 5.2% of all cases, which is higher than the prevalence of 1-2% of all hospital discharges [40] but far lower than a reported prevalence of 25-30% for people who use illicit substances [41]. In that literature review, in-hospital methadone maintenance treatment was associated with reduced hospital discharge against medical advice. However, given the high rate of OAT in Switzerland, these findings are hardly comparable. Nevertheless, poor retention in care might indicate difficulties in the management of pain in this population.
From the 255 patient records mentioning OAT before hospitalization, methadone and morphine were used in 69.4% of the cases, with liquid methadone and slow release morphine being the most often prescribed. This corresponds to the most commonly prescribed OAT agents in Switzerland [42], with a growing share of morphine for individuals who respond poorly to methadone [43]. Diacetylmorphine was the OAT agent in 58 (22.7%) cases. It was administered at doses between 0.05 and 1 g daily, mostly by mouth alone (26 cases) and predominantly in combination with other opioids (46 cases). This frequency is not surprising as preparations of diacetylmorphine in the tablet form are current practice in Switzerland [43]. Nevertheless, hospital records were mostly unclear in terms of whether the preexisting OAT was continued or not at unchanged doses during hospitalization, as recommended by authorities [23]. The information was not standardized in the records and had to be deduced from manually written medical and nursing records, and several interpretations or assumptions had to be made. There was also no explicit mention in the medical records of a person or an institution responsible for the OAT to contact in case of questions. However, obtaining this information might In case of multiple rescue analgesics (e.g., morphine + metamizole + paracetamol), only one agent was considered in the order of importance, with opioid > NSAID, including paracetamol. No case could end up in both "concordant" and "divergent" categories.
Rescue Analgesia for Opioid-Dependent Individuals on Opioid Agonist Treatment be challenging because currently in Switzerland, people enrolled in OAT are not registered in a nationwide database but in cantonal health authorities' records. Thus, the general lack of unequivocal information in patient records may aggravate the difficulty for the hospital staff to adequately treat acute pain in patients on OAT and may explain why agents that are not approved for OAT, such as fentanyl, hydromorphone, oxycodone, and pethidine, were prescribed for OAT. However, it is unclear whether or not the patients who received non-approved OAT agents were truly on OAT. Thus, we claim that guideline-divergent acute pain treatment in hospitalized individuals on OAT is multifactorial; it includes incomplete or imprecise anamnesis; little experience of the hospital staff with patients on OAT and the pharmacological needs of these individuals, especially in case of acute pain; and the absence of evidence-based guidelines. In a survey with Swiss physicians actively involved in OAT, lack of training and/or information was the main barrier to win the younger generation of physicians for OAT treatment [42]. In an observational study from Canada, hospitalized individuals on methadone maintenance treatment reported that physicians and nurses lacked experience in OAT and put low priority on OAT or methadone withdrawal [36]. One way to address these issues is the introduction of clinical pharmacists or advanced practice nurses on the wards who specialize in the care of patients with OUD.
Our patients were on average 51.3 ± 10.5 years old (range: 22-79 years). This corresponds to the globally observed aging of opioid-maintained populations on OAT, with an increase of patients aged 50 or older [44]. Because a globally growing number of OAT individuals can be expected in the future, it is likely that inadequate acute pain treatment will be even more conflictual, even more with older OAT patients who are more prone to developing age-related and degenerative diseases, and often used to receive higher opioid doses for OAT [45]. Thus, targeted interventions are urgently needed to guarantee adequate treatment for Rescue analgesia according to the corresponding OAT agents morphine (n = 86), methadone (n = 84), diacetylmorphine (n = 52), and buprenorphine (n = 6) as the guideline-concordant (light grey) or guidelinedivergent (dark grey) agent for 228 cases after exclusion of non-approved OAT and 14 cases with no rescue analgesia. In case of multiple rescue analgesics (e.g., morphine + paracetamol), only one agent was considered in the order of importance, with opioid > NSAID, including paracetamol. hospitalized individuals on OAT, such as the information and education of hospital staff, the dissemination of clear guidelines, and the implementation of structured medical and nursing records. The introduction of fully electronic patient records may be the foundation stone.

Strengths and Limitations
This study acknowledges some strengths. First, the years 2015 and 2018 were covered, that is, before and after the transition to fully electronic records in 2017 at the University Hospital Basel. By doing so, entry errors in the medical records due to the adaption to a new system were avoided. Second, two waves of hospital staff members following half-yearly personal fluctuations were covered. Thus, the observations made result from average practice from hospital staff members and are generalizable to Switzerland. Third, even if the progressive policy model of Switzerland is unique, the presented results can be extrapolated to all countries that have legitimized OAT and have comparable hospital staff structure at disposal.
This study has also some limitations. First, it could not be confirmed whether patients were truly enrolled in OAT. However, the indication of opioids' names and doses, coupled with the notes taken in the patient records, was generally sufficient to classify the individuals on OAT or not. Thus, the population presented is representative. Second, the MED was calculated according to locally recommended conversion factors. However, as the same evidence is used worldwide in clinical practice, the differences to other conversion factors may be marginal. Only the equivalence to heroin was not available from conversion tables and was therefore extracted from an elderly publication [46]. This might mirror the current situation in most countries where diacetylmorphine is not available for OAT. Third, and most importantly, the present study did not include any data from pain assessments. Although the assessment of pain is performed in the clinic, the timing is not systematically assessed, and thus, the relationship with the administration of rescue pain medication is unclear. Consequently, it is not clear to what extent and severity the patients suffered from pain and to which degree the pain medication was effective. Nevertheless, the results presented depict the prescription practice of rescue analgesia, without further exploration about its efficacy. Fourth, only pain medication available in Switzerland was recorded, which might differ from other countries. As an example, nefopam is not available in Switzerland or Germany. Fifth, the reason for hospitalization (such as a fracture) was available in the hospital records but with no explicit details if it influenced the prescribed dose of rescue medication. Consequently, we cannot exclude that doses of rescue medication were prescribed differently according to the reason for hospitalization. Nevertheless, in individuals engaged in OAT, we can assume that the dose of rescue medication should be calculated according to guidelines, that is, in case of opioids, be at least 1/6th of the MED. Thus, we claim that our estimation of 72.9% guideline-concordant rescue analgesia including 36.1% guideline-concordant opioids rescue in OAT individuals represents the minimum pain coverage according to general principles of pain management. Finally, prevalence of other pain management strategies was not investigated in this study. Nevertheless, the results presented denote guideline-divergent rescue analgesia for 19.9% of patients on OAT, which is alarming.

Statement of Ethics
The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. The study protocol was reviewed and approved by the Ethics Committee of Northwestern Switzerland, approval number: EKNZ 2018-0205. Written informed consent from participants was not required in accordance with national guidelines.

Conflict of Interest Statement
The authors have no conflicts of interest to declare.

Funding Sources
Funding for this study was provided by the authors and their institutions.

Author Contributions
I.A. and K.D. designed the study. C.J. and F.G. provided technical assistance. K.H. provided pharmaceutical assistance. I.A. analyzed the data and prepared the manuscript. All the co-authors reviewed the manuscript. All authors approved the final manuscript.

Data Availability Statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.