The Current Scope of Pain Medicine Fellowships: A Cross-Sectional Survey Study of Trainees, Directors, and Employers on Shaping Future Leaders

Abstract

Introduction

Since the establishment of Pain Medicine (PM) as an ACGME-recognized subspecialty in 1992, the field has undergone significant transformation. These changes brought increasing diversity in applicants` primary specialties, and the introduction of a myriad of emerging treatment paradigms. In this study, we aimed to evaluate the expectations, experiences, and perspectives of three target groups–PM trainees, program directors (PDs), and employers–to guide the evolution of PM education.

Methods

This study employed an integrated survey approach to comprehensively evaluate PM fellowship training. Surveys were distributed through professional societies, verified forums, and direct outreach over separate 3-month periods in 2023 for PD and trainee surveys and in 2024 for the employer survey. A total of 518 respondents across PDs, employers, and trainees completed surveys; overall response rate was indeterminable due to unknown denominators, while the PD survey response rate was 69.0%.

Results

The survey findings highlighted both shared and unique perspectives among the stakeholder groups. Trainees emphasized procedural volume and diversity as critical motivators for fellowship selection. Despite this focus, 70% of trainees expressed reluctance toward extending their fellowship duration, with 50% of trainees favored private practice. PDs acknowledged variability in training quality and emphasized the importance of supplemental workshops to address procedural gaps. Vast majority of PDs supported extending the fellowship duration to mitigate reliance on industry-led supplemental education. Employers underscored significant deficiencies in graduates’ preparedness for independent practice, with only 7% considering fellows adequately trained under the current model. Across all groups, there was a consensus on the need for curriculum standardization and enhanced training to align with the growing complexity of PM.

Conclusion

This study suggests that the one-year fellowship can be re-evaluated for adequacy in preparing physicians for independent practice, particularly in advanced procedures. Alternative training pathways offer additional exposure but vary in structure and oversight.

Introduction

Pain Medicine (PM), as a specialty, has experienced remarkable growth and transformation over the past four decades. Advancements in interventional procedures, minimally invasive spine techniques, and neuromodulation have expanded the therapeutic options available to pain physicians. Consequently, the scope of PM has expanded dramatically, encompassing an increasingly diverse range of conditions which includes chronic musculoskeletal disorders, neuropathic pain syndromes, Complex Regional Pain Syndrome (CRPS), oncologic pain and palliative care, and spinal disorders that were historically managed through traditional open spine surgery.^1,2 This rapid evolution has resulted in a growing disparity between the demand for comprehensive pain education and training and the current capacity of fellowship programs.

Despite the expanding complexity of PM, the duration and structure of post-residency training remain essentially unchanged. PM fellowships in the United States typically follow a one-year format, initially structured to serve the clinical needs and procedural expectations of the 1990s.^2,3 Over time, however, the breadth and depth of required knowledge and skills have outpaced the constraints of this model. Current PM training must not only address diverse clinical conditions but also prepare fellows for the technical and professional demands of modern interdisciplinary care. This has led to concerns regarding whether the current one-year fellowship framework is sufficient to train optimally competent, independent practitioners.^4–6

Discussions around increasing the duration of pain fellowship training are not novel, and the creation of a PM residency has been discussed as early as 2009.⁵ In addition, many feel that extended time in a fellowship program that offers limited advanced options would offer very little benefit to either the fellow or future employers. Simultaneously, many residents are choosing non ACGME (The Accreditation Council for Graduate Medical Education) accredited programs because of the advanced MIS (Minimally Invasive Surgery) options that are offered, and the movement to surgery centers that do not require an accredited program.⁶

In addressing this issue, we conducted a survey involving three primary stakeholder groups in PM education: program directors (PDs), trainees, and employers. These groups offer inherently complementary perspectives. Program directors provide insights into curricular structure, variability in procedural training, and the challenges associated with standardizing competency assessments. Trainees identify factors influencing career choices, perceived gaps in training, and their preparedness for independent practice. Employers contribute a post-training perspective, emphasizing clinical readiness, procedural competence, and adaptability to evolving practice demands.^7–9 Collectively, these viewpoints offer a comprehensive understanding of the current training challenges. This study systematically assesses the shortcomings of the current training model through the perspectives of these three groups.

ACGME does not currently mandate minimum procedural volumes or post-procedural competency metrics, resulting in inconsistent fellowship experiences across institutions. Furthermore, changes in applicant backgrounds, particularly the growing representation of residents in neurology, psychiatry, and emergency medicine, necessitate adaptable curricula that account for diverse training foundations. An abundance of unregulated society and industry web-based education has further complicated training by disrupting in-person education by accelerating the shift to virtual didactics, the effectiveness of which remains unquantified.¹⁰

Together, these three complementary surveys offer an unprecedented assessment of the current state of pain fellowship training in the US. Through triangulating insights from those who direct, receive, and evaluate PM education, we aim to identify gaps in the existing model and prompt dialogue addressing potential reform. Our central question remains: Is a one-year fellowship still adequate to prepare future pain physicians, given the expanding clinical, procedural, and professional scope of the field? By bringing together perspectives from relevant stakeholders, we hope to inform future improvements in PM education that align with the realities of modern practice.

Methods

Study Design

The present investigation comprised a synthesis of three cross-sectional surveys aimed at evaluating the perspectives of key stakeholders in PM education, including PDs, trainees, and employers.^7–9 Survey instruments were developed through an iterative process by the APPD Board of Directors. Members of the APPD Research Committee, composed of academic pain physicians, identified key domains and drafted stakeholder-specific content for each of the three groups. Based on this framework, questionnaires were developed by the research committee and subsequently refined by the board through three rounds of review prior to dissemination. To facilitate accurate responses, questions were designed to be concise and clearly worded.

Surveys were administered online using a secure platform (SurveyMonkey.com). A standardized three-phase recruitment strategy was utilized across all surveys. Phase one consisted of an advanced notification Email describing the purpose, voluntary nature, and confidentiality of the survey. Phase two involved direct distribution of the survey link via Email to targeted participants or organizational representatives. Phase three consisted of follow-up reminders sent approximately one to two weeks after the initial invitation to optimize response rates and minimize nonresponse bias. All survey links were single-use, and participants could not modify their responses once submitted. Survey completion times varied but remained within a three-month window for response collection.

Participants and Recruitment

The first survey targeted all PDs of ACGME-accredited PM fellowships who were also members of the APPD. The second survey was targeted to residents applying to PM fellowship and fellows enrolled in ACGME-accredited pain programs during the 2023 cycle and the third survey focused on the pain physician employer.

In the PD and trainee survey studies, surveys were distributed through the APPD. For the PD survey, questionnaires were sent directly to pain fellowship program directors, whereas for the trainee survey, recruitment was facilitated through program directors of anesthesia, physical medicine and rehabilitation (PM&R), neurology, and emergency medicine residencies.

In employer survey, participants were recruited through multiple professional societies, including the American Academy of Pain Medicine (AAPM), the American Society of Interventional Pain Physicians (ASIPP), the North American Neuromodulation Society (NANS), and the American Society of Pain and Neuroscience (ASPN), as well as the physician-exclusive DocMatter.com online platform, a trusted network for practicing pain physicians with verified medical licenses.

Questionnaire and Data Collection

The PD, employer, and trainee surveys included 21, 13, and 12 questions, respectively, with a combination of single-choice, multiple-choice, drag-and-drop, and open-ended formats (Supplementary Document 1). Questions were aimed at eliciting stakeholder perspectives on critical aspects of PM education, including fellowship curriculum content, procedural exposure, communication training, expectations for future practice, and suggestions for improving the quality of training.

Data was collected over separate 3-month periods in 2023 for the PD and trainee surveys and in 2024 for the employer survey. A total of 283 trainees, 196 employers, and 39 PDs completed the surveys. The PD survey achieved a response rate of 69.0% (39/56), while the overall response rate could not be determined due to unknown denominators for the employer and trainee groups.

Ethics and Registration

Ethical approval was obtained from the Montefiore Medical Center Institutional Review Board (IRB protocols 2023–15590, 2023–15592, and 2024–15604), and all survey distributions were conducted in accordance with institutional and ethical standards. Participation in all three surveys was voluntary and uncompensated. We confirm that all survey participants provided written and/or electronic informed consent prior to participation in the survey.

Statistical Analysis

Survey data were exported from the survey platform and analyzed using descriptive statistics. Frequencies, percentages, and means were computed using SPSS (Version 28.02, IBM Inc) or equivalent statistical software to summarize participant demographics and response distributions. No identifying data was collected, ensuring anonymity and confidentiality throughout the study.

Results

This study included a total of 518 respondents across three stakeholder groups (program directors, employers, and trainees) who completed separate survey instruments. Thirty-nine pain fellowship PDs out of 56 ACGME-accredited PM fellowship programs completed the survey, with a response rate of approximately 69.0%. It also gathered data from 196 employers of pain physicians, although the total number of recipients was unknown and therefore a precise response rate could not be calculated. Among employer respondents, 49.5% practiced in private settings, 32.7% in academic settings, and 17.9% in private practices with academic affiliations. Most employers were located in suburban areas (39.5%), followed by large cities (29.2%), urban centers (25.6%), and rural regions (5.6%), with the highest regional representation from the Northeast (29.9%), Midwest (24.2%), and Southeast (20.6%).

In addition, 283 PM trainees participated in the survey, of whom 80.0% were residents applying to PM fellowship during the 2023 application cycle, and 20.0% were current fellows in training during 2023; because the survey was distributed through PDs to both fellows and residents, the total eligible pool was unknown, and a response rate could not be determined. Findings provide a comprehensive understanding of the perceived strengths, weaknesses, and unmet needs in PM education, highlighting areas of alignment and divergence among key stakeholders.

Preparedness for Independent Practice After Fellowship

A consistent concern across all three groups was the inadequacy of the current 12-month fellowship to fully prepare fellows for independent clinical practice. Only 7% of employers expressed belief that graduates were ready for unsupervised practice immediately after fellowship, indicating that 93% of expected new hires required additional on-the-job training. Employers estimated that this period should have lasted between 6 and 12 months or more, with 65% explicitly stating that new physicians require at least six months of supervision post-fellowship. Similarly, 82% of PDs believed that fellows are only partially prepared upon graduation, with 59% of those reporting that a minimum of six additional months of training is needed for independent practice (Figure 1).

Figure 1.

Alignment across stakeholder groups regarding unpreparedness for independent practice, advanced procedural training, soft skill importance, and curriculum gaps in pain medicine fellowship education. Soft skills: integrity, professionalism, empathy, and communication as top attributes of soft skills. Curriculum gaps: All groups identified behavioral health, billing and coding, practice management, and radiology as areas that are insufficiently covered.

In addition to perceptions regarding readiness, trainees’ future practice plans provide further insight into the implications of fellowship training. A notable portion of trainees reported plans to practice in private settings, with 67.4% indicating private practice intentions, including 49.5% in independent private practice and 17.9% in private practices with academic partnerships, with the remaining 32.7% anticipating careers in academic institutions.

Advanced Procedures as a Core Fellowship Component

All stakeholder groups emphasized the importance of advanced interventional procedures in fellowship training. Over 90% of employers ranked proficiency in procedures such as spinal cord stimulation (SCS) trials, permanent implantations, and radiofrequency ablation (RFA) as essential for clinical readiness. PDs echoed this sentiment, with 87% agreeing that mastery of these advanced interventions should be a hallmark of fellowship success. Basic procedural competency was assumed and, less frequently, a point of contention (Figure 1).

Trainees demonstrated some flexibility regarding procedural training timelines, with 31% agreeing that some advanced skills could be refined after fellowship through continued education or workshops. In comparison, 67% of respondents felt that these competencies should be mastered during formal training (Table 1).

Table 1.

Key Differences in Perspectives

	Employers	Program Directors	Trainees
Preparedness for Independent Practice	Only 7% believe grads are ready; most need months to a year more	Believe current fellows are partially ready, with 6+ months more ideal	Feel somewhat confident, but recognize advanced skill gaps
Learning Advanced Procedures Post-Fellowship	Prefer fellows enter practice already trained in key procedures	Optimistic that certain skills can be learned later	Share PDs’ view, see fellowship as foundation, but open to postgrad learning
Two-Year Fellowship Model	Not directly addressed	Many support it to reduce industry dependence	Majority (~70%) oppose a 2-year model, though 30% are open
Top Qualities in a Fellow	Emphasize soft skills like empathy, communication, professionalism	Value integrity, professionalism, diligence, and team collaboration	See soft skills as important, but put more weight on procedural exposure
Institutional Reputation	Low priority	Moderate importance	Highly valued when selecting a program
Preferred Learning Format	Accept external education as a practical necessity	Actively promote use of online platforms and external workshops	Strong preference for fellowship-based learning, with supplemental use

Qualities Defining a Successful Fellow

Beyond procedural skills, all groups agreed on the vital importance of “soft” skills in clinical practice. Employers (78%) and PDs (85%) ranked integrity, professionalism, empathy, and communication as top attributes in evaluating fellows. Trainees similarly valued these qualities but placed comparatively more emphasis on procedural and academic competencies, with 60% ranking procedural training as their highest priority. The alignment between employers and PDs underscores the multifaceted demands on pain fellows, who must balance technical expertise with strong interpersonal and ethical conduct to manage complex patients and multidisciplinary teams effectively (Box 1 and Table 1).

Box 1.

Highlights

Highlights
Fellowship duration for Preparedness of Independent Practice All three groups express concern that the current 12-month fellowship format is insufficient to cover all necessary training, especially advanced procedures. Employers estimate new hires often need 6–12+ months of additional on-the-job training. PDs agree, and many suggest fellowship expansion. Trainees want more exposure but are hesitant about longer formal training.
Advanced Procedures Are Core to Fellowship Training Basic and advanced interventional procedures (SCS trials, implants, and RFA) are consistently ranked as the most important part of fellowship training by all groups. Basic procedural proficiency is assumed, while advanced interventions are seen as differentiators in clinical readiness.
Qualities of a Good Fellow Soft skills such as integrity, professionalism, communication, and empathy are top-ranked traits by employers and PDs. Trainees also acknowledge their importance, though they tend to emphasize procedural and academic training more heavily.
Curriculum Gaps Identified All groups agree that behavioral health, billing/coding, practice management, and radiology are under-taught. These gaps often push fellows to seek industry or society workshops to complete their training.
External Educational Resources Are Valued PDs overwhelmingly support and encourage fellows to participate in professional society workshops and industry-sponsored procedural events. Employers see such external opportunities as necessary, and trainees prefer in-program learning but are open to supplemental workshops when needed.
Openness to Fellowship Reform A significant number of PDs believe that extending the fellowship beyond 12 months could reduce dependence on industry for procedural education. A meaningful minority of trainees support a 2-year model, particularly if it guarantees exposure to advanced procedures. However, 70% are not in favor of a formal extension.

Curriculum Gaps and External Learning

Significant consensus exists regarding notable gaps in fellowship curricula. All groups identified behavioral health, billing and coding, practice management, and radiology as areas that are insufficiently covered. For example, 88% of PDs and 81% of employers noted that behavioral health integration is lacking, while 75% of trainees reported minimal exposure to practice management. PDs overwhelmingly support participation in professional society workshops and industry-sponsored procedural courses, with 92% encouraging such involvement to supplement fellowship education. Employers echoed this view, and 70% of trainees indicated a willingness to engage in external workshops, when necessary, despite preferring in-program learning whenever possible (Box 1 and Figure 1).

Openness to Fellowship Reform

The prospect of extending fellowship duration to address educational gaps generated mixed opinions. Among PDs, nearly half of them support a two-year fellowship model to enhance clinical preparation and reduce reliance on industry-sponsored procedural training. In contrast, the majority of trainees (69.6%) expressed opposition to extending fellowship training to two years (No 40.6%, Unsure 28.9%), citing concerns such as delayed career progression and financial burden. However, 47.6% indicated that they would nonetheless apply for a two-year fellowship if advanced procedures were included (Box 1 and Table 1).

Discussion

Findings from this study highlight ongoing concerns regarding the adequacy of the current one-year fellowship model in preparing pain physicians for independent practice. Employers and PDs consistently reported that graduates often need additional months of supervised training, especially to gain proficiency in advanced procedures. Trainees acknowledged similar gaps but expressed reluctance to extend fellowship duration, citing concerns regarding delayed career progression and financial burdens. This mismatch underscores the need for more flexible approaches to competency development and results in a transitional post-fellowship period during which many graduates require structured mentorship before achieving full independence.

A substantial proportion of trainees reported plans to work in private practice (49.6%), compared to 37.4% in academic institutions. This finding is particularly noteworthy given that many trainees, according to the trainee survey results, may not possess sufficient clinical expertise to independently manage a practice immediately upon completion of a fellowship. These results underscore the importance of structured mentorship and continued skill development, particularly for those pursuing careers in less supervised private practice environments. Graduates who start a solo practice would need formal mentorship arrangements but may be unaware of personal skills limitations in this arena.

Across all surveyed groups, interventional procedures were consistently identified as a core component of fellowship training, with employers and PDs emphasizing the importance of proficiency in basic procedures such as RFA, as well as SCS and other advanced interventions as essential for clinical readiness.^7–9, Trainees agreed with this focus, although many thought it was acceptable to enhance certain skills following fellowship through workshops or continuing medical education. Beyond technical abilities, qualities such as professionalism, communication, and integrity were universally regarded as essential for successful practice. The results indicated widespread concern regarding readiness for independent practice, with the majority of employers and PDs identifying the need for additional training and/or supervised mentorship post-fellowship, particularly in the first six months of practice. Employers did not directly address the issue of fellowship length in the surveys, but their low confidence in graduate readiness implicitly supports calls for expanded training, but not necessarily by extending time, while many often discuss the need for an interventional track. This would require PDs to increase and diversify the procedures for which they are credentialed. This may require significant political discussions within academic facilities where the surgical “turf” is often complex.

Alternative training pathways, such as unaccredited fellowships, have gained increasing attention.¹¹ These are often affiliated with private pain or interventional spine practices, operate in an “apprenticeship-style” structure, with a strong emphasis on procedural skills and have no oversight from any sort of credentialing or educational entity to ensure didactics and basic therapy are being covered. Still, oversight is eventually determined by credentialing bodies in private hospitals or surgery centers, and in these settings, the presence of an ACGME fellowship often has no significance, and the number of supervised procedures performed is often the significant factor in approval to offer a therapy. While these pathways may accelerate procedural competency, they also introduce significant variability in training quality and limit standardization across graduates, posing challenges for establishing uniform competency benchmarks in pain medicine.Meanwhile, the North American Spine Society (NASS) established the Interventional Spine and Musculoskeletal Medicine (ISMM) fellowship pathway, which launched its national match in 2020 and mainly attracts applicants from PM&R backgrounds.¹² NASS describes the ISMM fellowships as “unique from ACGME PM fellowships, which require training in inpatient acute pain management, palliative care, psychiatry, pediatric pain management, and intubation”, which could be concerning as they intentionally omit vital training for a comprehensive pain program.¹³ Additionally, ISMM fellowships are focused only on the treatment of spinal and musculoskeletal pain conditions. Industry-sponsored and society-based “innovator” or certification programs further contribute to this evolving landscape, providing concentrated procedural training outside the traditional ACGME framework.¹⁴ While these initiatives offer valuable technical exposure, their role in ensuring standardized, comprehensive preparation remains uncertain.^15,16 These offerings may also entice pain fellowship candidates to choose non-accredited programs which focus on interventional techniques only as there may be no reimbursement or employment differences in practice. A lack of ACGME oversight may further increase training variability and complicate the assessment of competency outcomes. In addition, the non-accredited fellowships are detracting trainees from ACGME programs that, though imperfect, have more predictable and regulated education. Furthermore, the availability of such pathways may influence trainee decision-making, as comparable employment opportunities may be perceived without completing ACGME-accredited programs, potentially contributing to increased variability across training pathways.

The results demonstrated significant gaps in fellowship training, particularly in behavioral health, psychiatry, non-interventional treatments, practice management, billing and coding, and radiology, with strong support from all groups for additional training opportunities. This finding highlights the challenge of delivering a fully comprehensive curriculum within the constraints of a one-year program and the critical role of lifelong learning and continuing medical education in PM.¹⁷

Overall, these data underscore the need for thoughtful reforms that strike a balance between comprehensive clinical training and the preferences of trainees, as well as workforce demands. The discussion of changing or altering an educational training program or fellowship is not unique to PM and is consistent with ongoing discussions across multiple medical specialties. Similar discussions are observed across multiple medical specialties. The authors submit that we should study the evolution of training programs that have converted fellowship residencies in hopes to address their educational gap, as this may be a pathway for the field of PM to improve training quality and our clinical footprint.^18,19

Importantly, the mismatch between trainee preferences and employer expectations suggests that extending fellowship duration alone may not be sufficient. Alternative strategies such as competency-based progression, structured mentorship following training, and flexible or track-based educational models may better address both educational deficiencies and workforce realities.

To effectively address the educational deficiencies within the domain of pain medicine, we propose the execution of comprehensive research initiatives alongside the periodic administration of surveys by the APPD or other research entities. Furthermore, due to the rapid emergence of interventional procedures in pain medicine, most pain fellowships cannot keep pace with this innovation. Therefore, the authors suggest that national pain societies should be the arena for the introduction of novel interventional education, as they can support unbiased education in large group settings in real time. These efforts are crucial for capturing contemporary advancements and insights from all associated groups, thereby facilitating the enhancement of pain medicine education across the United States.^18,19

As calls for reform to intensify, efforts will need to strike a balance between the need for comprehensive procedural and non-procedural-based education within the constraints of limited time and resources. Approaches such as competency-based assessments, structured partnerships with external organizations, improving residency baseline skills and competency, extending fellowship training, or creating a dedicated PM residency are still being considered to address these challenges.^20–22

Several limitations of this study should be acknowledged. First, the potential for response bias must be considered, particularly among trainees, whose responses may have been influenced by financial concerns, career timing, or perceptions regarding the implications of training reform. Differences in experience levels among trainees, including both current fellows and residents without formal pain training, may have further influenced responses to clinically oriented questions. Second, the representativeness of the sample is limited. While response rates could be estimated for program directors due to a defined sampling frame, they could not be determined for employer and trainee groups. Additionally, the trainee cohort was relatively small and heterogeneous. Third, the use of group-specific survey instruments may limit direct comparability across stakeholders and introduce variability in interpretation. Finally, the absence of inferential statistical analysis further constrains the strength and generalizability of the findings.

The authors submit that pain education requires transformation across the globe. Although pain medicine education remains a topic of international debate, our authors and survey respondents are based in the United States and are therefore not positioned to draw conclusions regarding solutions in other countries. However, we believe that our manuscript can serve as a model for how other countries can conduct their own national investigation. We hope that our document serves as a model for future research, and ultimately, for collaborative reformation.

Conclusion

The findings from this multi-stakeholder survey underscore critical gaps in current PM fellowship training. While trainees prioritize procedural exposure and career readiness, PDs and employers highlight the insufficiency of existing models in equipping fellows for independent practice. Collaborative efforts among educational institutions, professional societies, and industry partners are crucial to standardizing and enhancing fellowship curricula, ensuring alignment with the field’s evolving landscape and practitioner expectations.

Funding Statement

The authors have no sources of funding to declare for this manuscript.

Disclosure

Dr Scott Pritzlaff reports personal fees from SPR, Medtronic, Bioness; royalties from Wolters Kluwer and Oxford University Press; grants from Medtronic, Biotronik, Boston Scientific, Nevro, Abbott; is the co-founder of FlywheelRx, outside the submitted work. Dr Micheal E Schatman reports is a senior medical advisor for Apurano Pharma, outside the submitted work. Dr Trent Emerick reports stock option/equity from Vanish Therapeutics, Inc, Nocix Technologies, Glimmer Health, and royalties from Springer, outside the submitted work. Dr Daniel Pak is a paid consultant for Mainstay Medical, Nevro Corp, and Averitas Pharma. Dr Timothy R Deer reports personal fees for consultant, advisory board and/or research from Abbott, SpineThera, Saluda, Cornerloc, Boston Scientific (Reliviant), PainTeq, Spinal Simplicity, SPR Therapeutics, Biotronik, Aurora, and Nervonik, outside the submitted work. Dr Karina Gritsenko is a consultant for and advisory board of Pacira and Vertex. Dr Peter Staats reports personal fees as a part-time CMO from electroCore, grants for research and/or consulting from Biotronik, Saluda; a share holder of SPR, outside the submitted work. Dr Maged Guirguis is a consultant for Boston Scientific, Saluda Medical, Averitas Pharma, Avanos Medical, and Vivex biologics, outside the submitted work. Dr William Caldwell reports personal fees as a consultant from Boston Scientific, outside the submitted work. Dr Sayed E Wahezi receives research funding from Boston Scientific, Abbott, and Vertos. He is also a consultant for Boston Scientific and a key opinion leader for TENEX Health. The authors declare that they have no other known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.