Study design and setting

This qualitative interview study employed semi-structured interviews to investigate the perceptions of physicians and physician assistants (PAs) regarding opioid prescribing and opioid-related harms in the ED setting. The study was conducted with a purposive sample of PAs, residents, and emergency physicians working in various (academic and non-academic) representative EDs, located in both urbanized and less urban regions in the Netherlands: Leiden University Medical Center (LUMC, an academic center with 25,000 ED visits annually), Haaglanden Medical Center Westeinde (HMC, a teaching hospital and level-I trauma center with 55,000 ED visits annually), Radboud University Medical Center (RadboudUMC, an academic level-I trauma center with 16,000 ED visits annually), and Rijnstate Hospital (a teaching hospital with 35,000 ED visits annually). Small rural hospitals were not included in our sample. However, this type of hospitals is scarce within the Dutch healthcare system. The sample thus offers a strong reflection of current practice and captures the diversity most relevant to the national context.

Participants

Participants were health care providers working in the ED, with the authority to prescribe opioids. Purposive sampling was utilized to select a diverse group of healthcare professionals, considering factors such as age, sex, profession, and years of experience, while also achieving variation across hospital types and regions. This was done to provide a comprehensive perspective on opioid prescribing, use, and misuse within the ED context. An invitation to participate in the study was emailed to all emergency physicians and PAs working in the four participating EDs (around 20 clinicians per site) several weeks prior to the interviews. In addition, the interviewer visited each ED on multiple occasions to facilitate recruitment. No financial or other incentives were offered. Nearly all clinicians who were approached agreed to participate; the few who did not participate cited scheduling conflicts or workload as the main reasons. Participants understood that their statements might be quoted in the manuscript. The participants did not have the opportunity to provide feedback on the study findings.

Interviewer role

All interviews were conducted by C.t.P., a resident physician with two years of working experience in the ED. The interviewer was known within the LUMC but had no prior relationship with participants from the other three hospitals. Participants were informed that the interviewer was a resident at the LUMC ED and that the study aimed to explore physicians’ perspectives on opioid prescribing and opioid-related issues. They had the possibility to ask additional questions before the start of the interviews, and were assured that their input would be used solely to improve understanding of the topic. The interviewer’s background in emergency care enriched the credibility and quality of the insights gathered. As participants interacted with someone who understood the complexity and nuances of their work environment, they were more inclined to engage in detailed discussions about their experiences and challenges.

Data collection

Data were collected through semi-structured interviews. The interviews were conducted using a topic guide developed for this study by all authors based upon previous literature and experience [19]. The topic guide was designed to elicit detailed responses regarding participants’ views on opioid prescribing, their experiences and attitudes regarding opioid-related harms in the ED, allowing participants to elaborate freely on their perspectives. Follow-up and probing questions were asked to clarify or deepen responses when relevant. Data saturation was achieved after 20 interviews, as no new additional significant information was captured beyond this point. To ensure saturation, five additional interviews were conducted using purposive sampling, also ensuring diversity. These interviews did not provide relevant new insights. The 25 interviews were all conducted in person by C.t.P, a resident in emergency medicine between January 2024 and May 2024.

In addition to questions regarding opioids, participant demographic information (age, gender, and work experience) was collected. These data were stored securely and retained for up to one year following data collection.

Data analysis

The interviews were audio-recorded, transcribed verbatim (using Amberscript transcription software). To develop a comprehensive understanding of each physician’s experience, the transcripts were read and reread thoroughly. Two researchers C.t.P. (female emergency medicine resident with no prior experience in qualitative research) and N.K., PhD, emergency physician, experienced in qualitative research) independently coded the transcripts with ATLAS.ti software (Version 23.2.3.27778) using the reflexive thematic analysis approach as described by Braun & Clarke [28]. This involved identifying inductive descriptive codes by highlighting recurring phrases within the physicians’ narratives. Following the initial coding, C.t.P. and N.K. compared and discussed the codes, thereby enhancing the depth of the analysis. The codes were organized into themes over the course of two thematization meetings. In these sessions, the codes were collaboratively grouped to construct themes that reflected the content and nuances of the interview data. The first meeting took place after the first 15 interviews were conducted and included researchers C.t.P., N.K., A.S. (professor in addiction psychiatry, experienced with qualitative study methods), H.E. (PhD candidate with a background in qualitative research on opioid misuse) and V.B. (medical student/ PhD candidate in research on opioid misuse). The topic guide was refined during this initial meeting, adding questions about personal experiences with opioid-related harms to deepen the discussions. The second session was held after all 25 interviews were completed and involved C.t.P., N.K., A.S. and H.E. The study results were reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ) [29]. The manuscript is reviewed and corrected using AI-based tools, specifically for grammatical accuracy and clarity.

Ethical considerations

Ethical approval was obtained from the Leiden University Medical Center (LUMC) ethics committee (protocol identification number N052). The study was conducted in accordance with the ethical standards of the institutional and national law. All participants provided written informed consent before they participated in the study. Their anonymity and confidentiality were maintained throughout the research process.

Characteristics of participants

The interviewees comprised thirteen women (52%) and twelve men (48%) and included twelve emergency physicians, ten resident physicians from various specialties (from which two emergency residents) and three physician assistants.

The average age of the participants was 35 years, ranging from 25 to 59 years. On average, the participants had ten years of experience in their current roles (median of four years of experience ranging from six months to 27 years) and on average eight years of experience in the ED (median of five years, range of several months to 27 years).

Twelve out of twenty-five (48%) had previously worked at another ED, and seven (28%) participants had prior clinical experience outside the Netherlands, mainly within Europe, and a few in the US. Additionally, participants had backgrounds in various specialties, including Surgery (20%), Internal medicine (16%), Neurology (16%) and Cardiology (16%), Intensive Care (12%), among others. Detailed participant information is provided in Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline demographics and characteristics of the participants.

https://doi.org/10.1371/journal.pone.0338421.t001

Themes

Through reflexive qualitative analysis, two main themes were developed, each encompassing three subthemes, see Table 2. These themes captured the perceptions and experiences of emergency care providers in opioid prescribing and addressing opioid-related harms at the ED.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Overview of main themes and subthemes, including illustrative quotes.

https://doi.org/10.1371/journal.pone.0338421.t002

Theme 1: Preventing opioid-related harms

This theme addresses the prevention of opioid-related harms, through thoughtful and conservative prescribing of opioids at the ED. In three subthemes, the roles of the involved healthcare providers, patients and the social context are discussed.

Subtheme 1: Role of health care providers

Opioid prescriptions at ED discharge

Shared decision-making

Communication with patients regarding their preferences and opioid-associated risks were often emphasized. Some physicians mentioned that patients often forget the provided information, and may use more opioids than prescribed, which can contribute to development of dependence. Several participants mentioned that shared-decision making can be difficult due to time constraints in the ED.

Participant #14: “You want to make a meaningful difference for the patient, and I believe this should be done through shared decision-making. This involves engaging the patient, acknowledging their knowledge and perspective, while also helping them understand the potential negative effects of opioids. The conversation is key.”

Perceived roles of other healthcare providers

Subtheme 2: Patients

Subtheme 3: Social context

Pain in society

Several participants expressed the perception that contemporary society often does not accept the experience of pain, leading to a strong desire to treat it.

Participant #1: ‘’There seems to be a trend where opioids are prescribed quite readily in the ED, as if pain is no longer considered acceptable.”

Conversely, some physicians noted a cultural tendency in the Netherlands to inadequately assess and treat pain in hospital settings, resulting in the undertreatment of pain. One of the physicians mentioned a reluctance to prescribe opioids due to the hospital’s conservative opioid prescribing policy. Participants described ambivalent concerning two contrasting cultural perspectives on pain management. Some emphasized a traditional, more stoic approach to pain, characterized by the belief that pain is something to be endured rather than eliminated. This may contribute to undertreatment. Others highlighted a more recent shift toward the idea that pain should not be accepted and must be actively treated, supported by initiatives such as including pain as a (fifth) vital sign in triage protocols. This may contribute to overtreatment. These contrasting tendencies coexist within Dutch emergency medicine, reflecting broader societal and professional debates about what constitutes appropriate pain management.

Responsible prescribing with Enhanced Electronic Patient File (EPF)

Several physicians highlighted the lack of oversight and coordination in prescribing practices between hospitals, making it challenging to track what other providers may have already prescribed. This gap increases the risk of duplicate prescriptions and facilitates “opioid shopping”. Physicians and PAs expressed a strong need for improved information sharing.

Participant #3: ‘’I think it would be beneficial to have a more unified, nationwide EPF and shared insights across EDs. Currently, there is no way to share data between facilities. For example, if a patient visits one ED and later goes to another, there’s no way of knowing.”

Theme 2: Managing opioid misuse and OUD in the ED

In this theme the focus is on addressing the challenges in the ED in patients who are already using or dependent on opioids. Again, in three subthemes, the roles of the involved healthcare providers, patients and the social context are considered, see Table 2.

Subtheme 1: Role of health care providers

Starting OUD treatment at the ED

Perceived roles of other healthcare providers

Subtheme 2: Patients

Subtheme 3: Social context

Shared Electronic Patient Files (EPF) for opioid problem detection

Several interviewed physicians advocated for a nationally shared EPF system to provide insight into prescribed medications and improve care coordination. These systems could implement alerts for increased dosages and documenting opioid misuse.

Participant #3: “If someone visits this ED and later goes to another, we have no way of knowing. I think that makes it harder to implement effective interventions.”

Participant #21: “I think we lack the tools to make a meaningful contribution to this problem on a societal level. For instance, having a nationwide EPF for the ED could help identify shoppers or patients who visit multiple EDs because of opioid intoxications. With such a system, we would have more ground to address the issue effectively and stronger grounds to refer patients or initiate conversations with them.”

Preventing opioid-related harms from an ED perspective

Our findings indicate that emergency care providers often experience a dual responsibility in preventing opioid-related harms. Cultural attitudes toward pain and addiction further complicate care delivery. On one hand, they need to manage acute pain effectively. A few physicians emphasized the importance of adequate pain treatment, stressing that this is crucial to prevent the development of chronic pain. There is ample evidence that effective early management of severe acute pain is important in preventing the development of chronic pain [31–33]. On the other hand, emergency care providers emphasize the importance of reducing the risk of opioid-related harms, such as misuse, addiction, and overdose, by prescribing opioids with caution and in minimal amounts [1,34]. We found evident variations in opioid prescribing practices across EDs between participants, particularly concerning the selection of long-acting versus short-acting formulations and prescription duration.

Cultural attitudes toward pain and opioid use differ markedly between the Netherlands and the US. In the US, aggressive pain management and pharmaceutical marketing contributed to high opioid consumption and misuse [35,36]. In contrast, European countries, including the Netherlands, have more restrictive prescribing cultures and greater emphasis on non-opioid pain management [37].

Additionally, we found that patient expectations contribute to this dilemma. Patients in acute pain often require quick and effective relief, which can pressure providers to prescribe strong painkillers. Most physicians balance this dilemma by prescribing only short courses of opioid treatment. Currently, the treatment of acute pain still appears to depend heavily on opioids, as other therapeutic options for managing this pain are perceived to be limited. Some physicians emphasize the need for well-defined, evidence-based protocols on opioid prescribing in acute care settings to support their decision-making processes. Notably, our findings on opioid prescribing at ED discharge are largely consistent with those reported in a qualitative US-based study from 2017 on the same topic, reflecting similar clinical considerations [38]. However, US physicians frequently referenced patient satisfaction scores as a factor influencing prescribing decisions, a consideration not present in our study in the Dutch healthcare context, where such metrics are not routinely used. Furthermore, opioid prescribing in US EDs is subject to stricter regulatory frameworks, including state-level legislation that imposes limits on amounts and restricts the types of opioids that may be prescribed. In this regard, Dutch emergency physicians operate with greater clinical autonomy [1,22,38].

Previous research has shown that the risk of problematic opioid use increases with the duration of opioid prescriptions. However, even short-term use of prescription opioids carries the potential for opioid-related harms like abuse and addiction [34]. Notably, the majority of patients discharged from the ED with acute pain can be effectively managed using opioid alternatives. Specifically, appropriate dosing of acetaminophen and ibuprofen often provides sufficient analgesia for most adults [39]. So, both ED physicians and patients may benefit from education on acute pain management. It is essential to manage patient expectations by focusing on pain reduction rather than complete pain elimination [31,40]. In the US, educational programs targeting ED residents have been associated with significant reductions in opioid prescribing both in the ED and at discharge [41]. Similarly, online teaching modules have been shown to improve knowledge, confidence, and adherence to guidelines among healthcare providers regarding opioid prescriptions [42].

The majority of the participants in this study emphasize that, in addition to prescribing opioids with caution, shared decision-making is crucial. They agree that it is essential to initiate discussions about opioid use, ensuring that patients are informed about the different pain management options.

Indeed, the US Centers for Disease Control and Prevention (CDC) have issued guidelines to all healthcare providers involved in opioid management and recommend shared decision-making with patients by discussing the potential side effects, risks, and benefits of opioid use [21]. This approach has shown to not only improve patient outcomes and satisfaction but also enhances the safety and effectiveness of pain treatment by fostering a collaborative relationship between patients and healthcare providers [43]. Despite these potential benefits, participants also found that, due to time constraints in a busy ED, extensive patient education and shared decision making may not always be feasible.

Managing problematic opioid use at the ED

EDs serve as a potential point for intervention in problematic opioid use, offering opportunities for identification, initiating treatment, and linking patients to follow-up care. Stigma surrounding substance use was recognized as a possible influential factor shaping clinicians’ attitudes and prescribing decisions. Several participants noted that a history of substance use was often not explicitly explored during ED encounters, potentially reflecting discomfort around the topic. This stigma may contribute to under recognition of opioid misuse risk and missed opportunities for early intervention. Addressing this requires targeted education and communication training to normalize discussions about substance use.

Participants reported more challenges in identifying problematic opioid use, attributing these difficulties to the time constraints inherent in the ED environment and limited access to comprehensive information regarding patients’ current and past medication histories.

Insight into patients’ current medication use was frequently emphasized as essential for preventing “opioid shopping” and minimizing the risk of patients obtaining prescriptions from multiple sources. It was also noted that when a patient’s record indicated possible opioid misuse or “shopping,” clinicians sometimes preferred to consider non-opioid alternatives first, suspecting drug-seeking behavior. While not entirely unjustified, this represents a difficult clinical and ethical balance in acute pain management.

The integration of PDMPs in the US has led to a significant decrease in opioid prescriptions with a cumulative effect of interventions like education on alternatives to opioids and modifications to electronic patient file processes [44]. In several US states, prescribers are required by law to consult the PDMP before issuing an opioid prescription. PDMPs offer a model to help healthcare providers and pharmacists identify potential drug misuse through shared data [45]. They are designed to promote safe prescribing practices and to reduce high-risk drug harms like prescription shopping, overdoses and mortality. The impact of PDMPs on reducing overdose mortality and misuse is often inconsistent and the effectiveness depends on their implementation and integration with wider public health strategies [46,47].

Adopting similar systems in the Netherlands, such as expanding access to nationwide electronic patient files, may provide a practical solution to recognize problematic prescription opioid use.

Our study participants commonly acknowledged the EDs’ potential role in initiating care for patients with problematic opioid use. However, many ED physicians indicated that organizing follow-up care for patients with opioid addiction presents a significant challenge and is often perceived as impractical within the ED setting. Emergency care providers commonly reported a lack of resources and expertise to effectively manage and refer these patients. Limited collaboration with GPs and addiction specialists was identified as an issue. Additionally, as emergency care providers are accustomed to managing patients around the clock, the unavailability of GPs during after-hours was frequently highlighted as a gap in care. It is expected to worsen over time as the national shortage of GPs increases, driven by population aging, higher care demands, and limited workforce capacity. Addiction care services are generally unavailable outside regular working hours, leaving the ED as the only accessible point of contact for many patients, restricting timely access to specialized support for patients with opioid-related problems.

In the US several models have been developed to ensure continuity of care after ED discharge of patients with opioid addiction, including direct referrals to local addiction treatment services, or providing a short-term prescription for buprenorphine and referring patients to an outpatient clinic within the hospital [23,48,49]. The second option allows emergency physicians to initiate treatment and referring patients to a dedicated addiction clinic, ensuring a transition from ED care to ongoing treatment [18].

US-based studies have shown promising results from the initiation of buprenorphine and methadone in the ED for managing opioid dependency, with studies demonstrating its effectiveness in reducing overdoses and linking patients to long-term care [23,50]. Buprenorphine can be effectively used to manage opioid withdrawal and serve as a starting point for medication-assisted therapy [49]. However, participants in this study noted that this approach might be less relevant to the Netherlands due to the overall lower incidence of opioid overdoses and withdrawal cases in Dutch EDs [3].

Tailoring opioid stewardship interventions to the organization of care, such as ensuring GPs have the time, skills, and resources for patient-centered deprescribing, can help address rising trends in opioid prescriptions, hospital admissions, and mortality.

Strengths and limitations of the study

This study offers valuable insights into the experiences and perceptions of ED physicians and PAs regarding opioid prescribing and management in the Netherlands. To our knowledge, this is the first study of its kind conducted in Europe. By including perspectives from physicians across various specialties, all working in the ED, the study captures a broad view of the challenges and opportunities in this setting. However, certain limitations must be acknowledged. Social desirability bias may have affected participants’ responses, particularly on sensitive issues like opioid prescribing and addiction care. Physicians may have provided answers they perceived as more socially or professionally acceptable rather than fully disclosing their dilemmas or uncertainties. This could have led to an underestimation of the challenges they experience in practice.

Further, the workload of some participants impacted the interviews; busier physicians often had less time, resulting in shorter and less detailed responses compared to those who had more time to participate. This discrepancy could have influenced the findings by underrepresenting the perspectives of physicians working in high-pressure settings, where decision-making around opioid prescribing may be even more complex.