Prognosis: the “missing link” within the CanMEDS competency framework

  • Vincent Maida1, 2, 3Email author and

    Affiliated with

    • Paul M Cheon4

      Affiliated with

      BMC Medical Education201414:93

      DOI: 10.1186/1472-6920-14-93

      Received: 21 October 2013

      Accepted: 6 May 2014

      Published: 13 May 2014

      Abstract

      Background

      The concept of prognosis dates back to antiquity. Quantum advances in diagnostics and therapeutics have relegated this once highly valued core competency to an almost negligible role in modern medical practice. Medical curricula are devoid of teaching opportunities focused on prognosis. This void is driven by a corresponding relative dearth within physician competency frameworks. This study aims to assess the level of content related to prognosis within CanMEDS (Canadian Medical Education Directives for Specialists), a leading and prototypical physician competency framework.

      Methods

      A quantitative content analysis of CanMEDS competency framework was carried out to measure the extent of this deficiency. Foxit Reader 5.1 (Foxit Corporation), a keyword scanning software, was used to assess the CanMEDS 2005 framework documents of 29 physician specialties and 37 subspecialties across the seven physician roles (medical expert, communicator, collaborator, manager, health advocate, scholar, and professional). The keywords used in the search included prognosis, prognostic, prognosticate, and prognostication.

      Results

      Of the 29 specialties six (20.7%) contained at least one citation of the keyword “prognosis”, and one (3.4%) contained one citation of the keyword “prognostic”. Of the 37 subspecialties, sixteen (43.2%) contained at least one citation of the keyword “prognosis”, and three (8.1%) contained at least one citation of the keyword “prognostic”. The terms “prognosticate” and “prognostication” were completely absent from all CanMEDS 2005 documents. Overall, the combined citations for “prognosis” and “prognostic” were linked with the following competency roles: Medical Expert (80.3%), Scholar (11.5%), and Communicator (8.2%).

      Conclusions

      Given the fundamental and foundational importance of prognosis within medical practice, it is recommended that physicians develop appropriate attitudes, skills and knowledge related to the formulation and communication of prognosis. The deficiencies within CanMEDS, demonstrated by this study, should be addressed in advance of the launch of its updated version in 2015.

      Keywords

      Prognosis CanMEDS Competency frameworks Quoad vitam Quoad sanantionem Foreseeing Foretelling

      Background

      The birth of prognosis, using clinical signs, dates back to ancient Sumerian civilization circa 2,000 BC [1]. Hippocrates (460 BC – 370 BC) advanced the domain of prognosis by using combinations of symptoms and clinical signs to predict outcomes. Hippocratic prognostication also took into account certain environmental factors and patient characteristics but did not take into account the patient’s diagnosis. During the era of Hippocrates the core competencies of a physician consisted of diagnosis, therapeutics, and prognosis. However, the paucity of effective diagnostic and therapeutic modalities rendered the ability to prognosticate the most important role of a physician [13]. Conceptually, Hippocrates described prognosis as a two dimensional construct: quoad vitam (predictions about survival and life expectancy) and quaod sanantionem (predictions about healing and restoration of function) [2, 3].

      Over the past century monumental advances in diagnostic and therapeutic modalities have led to the marginalization of prognosis as a core competency. This trend is reflected by its withering presence within medical literature. The original comprehensive textbook of medicine, written by Sir William Osler (1849 to 1919), “The Principles and Practice of Medicine” published twenty-two editions between 1892 and 1988. Osler’s textbook described each disease under seven categories: etiology, presentation, pathology, diagnosis, therapy, prognosis, and complications. Osler’s prognostic formulations took into account symptoms, signs, and diagnostic criteria. Review of a leading contemporary forerunner, “Harrison’s Principles of Internal Medicine”, 12th edition, revealed that only 27% of the diseases had dedicated sections discussing prognosis [4].

      The post Osler era has witnessed the relegation of prognosis to a negligible level of importance. This has occurred in tandem with decreased emphasis, if not denial, of the natural history of disease. In modern times physicians are trained and socialized with a nearly singular objective, to cure. Anything less is regarded as a failure. As a result, there exists a tendency to avoid consideration of prognosis as this would seem to promote nihilism and negativism. Thus, the need to render a prognosis has been deemed redundant, and has led to a noticeable void within medical curricula. Moreover, competency frameworks, which are the main drivers of medical curricular development scarcely touch upon topics related to prognosis. The implications of this trend for patient, family, and society are immense, not the least of which is the tendency for physicians to excessively offer late stage interventions to patients within the final phases of their lives, thus, denying them the opportunity to be managed more appropriately and effectively in a conservative palliative mode of care. In addition, when prognosis is not disclosed, treatment decisions tend to be more paternalistically driven by physicians, often associated with vested interests, rather than being truly patient-centered [4].

      At the vanguard of a fledgling renaissance for prognosis in medical practice has been Dr. Nicholas A. Christakis. His award winning book “Death Foretold-Prophecy and Prognosis in Medical Care”, published in 1999, eloquently outlines the deficiencies related to prognosis in clinical care and research together with the serious implications for patients, families and society. Dr. Christakis passionately urges that prognosis be restored as a core competency. His thesis, built upon Hippocratic principals, posits that the process of prognosis comprises two basic components, namely, foreseeing (computing and formulating the prognosis) and foretelling (disclosing and communicating the prognosis). Hence, prognosis may be viewed as both a science (foreseeing) and an art (foretelling) [4]. Thus, a complete and comprehensive approach to prognosis must involve both components.

      The impact of Christakis’s work is evidenced by the observation that over the past decade there has been a steep rise in the number of publications related to objective prognostic factors for both quoad vitam and quoad sanantionem. For the purposes of quoad vitam, the TNM (Tumor, Node, Metastasis) anatomic cancer classification system [5], and actuarial data, outlined in Surveillance, Epidemiology, and End-Results program of the National Cancer Institute (SEER) are predictive of long-term survival in patients with newly diagnosed cancer [6]. In advanced cancer Palliative Performance Scale scores (PPS) [7, 8], Palliative Prognostic Index (PPI) [9], and Palliative Prognostic Score (PaP) [10] are predictive of short-term survival. In the Intensive Care Unit setting, the Acute Physiology and Chronic Health Evaluation Score (APACHE) [11] and Simplified Acute Physiology Score (SAPS) [12] correlate with survival. Moreover, recent research is also demonstrating the potential of novel prognostic factors such as tumour-related factors (cytogenic and molecular markers) and inflammatory markers [13]. The occurrence of sentinel events in dementia cases such as dysphagia, sepsis, and pressure ulcers also portend a shortened life expectancy [14, 15]. Examples of assessment tools that facilitate quoad sanantionem include the FIM (Functional Independence Measure) [16, 17] being predictive of neurologic recovery in stroke patients, and PPS being predictive of pressure ulcer healing [18]. However, despite the existence of data on such factors, instruments, and models, the process of knowledge translation remains laggard, rendering physician utilization low.

      A paradigm shift to a competency-based approach in medical education is the result of public outcry for improved quality, comprehensiveness, and accountability [19, 20]. One of the original competency frameworks was created during the early 1990’s by the Royal College of Physicians and Surgeons of Canada (RCPSC), and was called the “Canadian Medical Education Directives for Specialists” (CanMEDS). Since 1993, CanMEDS has gone through four major development phases. The current 2005 version is based upon the seven core roles of a physician: Medical Expert, Communicator, Collaborator, Manager, Health Advocate, Scholar, and Professional. The Medical Expert role is regarded as the central role as it represents a balanced integration of the other six roles. The CanMEDS framework has been incorporated into medical undergraduate and postgraduate programs throughout Canada. Moreover, CanMEDS is foundational in defining educational objectives and outcomes of all medical specialties and subspecialties recognized by the RCPSC. CanMEDS has been recognized as a prototypical competency framework having been adopted, adapted, and modelled internationally. Currently, CanMEDS is in the midst of a three-year reform process that is scheduled to be finalized in 2015 [21].

      This study aims to quantify the level of content related to the domain of prognosis within CanMEDS 2005. It is hoped that by highlighting any deficiencies, attention may be focused on recognizing and restoring prognosis as a core physician competency, both from the perspective of a scholar (foreseeing) and a communicator (foretelling).

      Methods

      This study employed the methodology of quantitative content analysis [22]. The entire compilation of CanMEDS 2005 text documents were accessed on the website of the Royal College of Physicians and Surgeons of Canada [21] on July 31, 2013. The online documents found under “Information by Discipline” were subjected to Foxit Reader 5.1 (Foxit Corporation), a proprietary scanning software that locates the presence of keywords within text-based documents [23]. All documents pertaining to 29 medical specialties and 37 subspecialties, across the seven physician roles (medical expert, communicator, collaborator, manager, health advocate, scholar, and professional) were subjected to the Foxit Reader 5.1 scan. The keywords used in the search included prognosis, prognostic, prognosticate, and prognostication. The presence of the particular keywords were then manually entered onto Microsoft Excel 2007 spreadsheets in order to tabulate the frequency of occurrences per specialty and subspecialty, as well as categorizing them as to which of the physician roles they were linked with. No human material or human data were used in this research.

      Results

      Among the 29 specialties there were 20 citations of the keyword “prognosis” and one for “prognostic”. (Tables  1 and 2). Neurology accounted for 13 of the 20 (65%) citations for the keyword “prognosis”. Of the 29 specialties six (20.7%) contained at least one citation of the keyword “prognosis”, and one (3.4%) contained one citation of the keyword “prognostic” (Tables  1 and 2). The combined citations for “prognosis” and “prognostic” among specialties were linked with the following competency roles: Medical Expert (81%), Scholar (14.3%), and Communicator (4.7%) (Table  1).
      Table 1

      Specialties with citations of keywords “prognosis” or “prognostic” (n = 7)

      Specialty

      Keyword

      Competency domain

      Prognosis (Number of citations)

      Prognostic (Number of citations)

      Medical expert (Number of citations)

      Communicator (Number of citations)

      Scholar (Number of citations)

      Dermatology

      1

      0

      1

      0

      0

      Medical Genetics

      2

      0

      1

      1

      0

      Neurology

      13

      0

      10

      2

      1

      Neurosurgery

      1

      0

      1

      0

      0

      Obstetrics & Gynecology

      2

      0

      2

      0

      0

      Otolaryngology-Head & Neck Surgery

      1

      0

      1

      0

      0

      Plastic Surgery

      0

      1

      1

      0

      0

      Total number of Specialties

      6

      1

      7

      2

      1

      Table 2

      Subspecialties with citations of keywords “prognosis” or “prognostic” (n = 19)

      Subspecialty

      Keyword

      Competency domain

      Prognosis (Number of citations)

      Prognostic (Number of citations)

      Medical expert (Number of citations)

      Communicator (Number of citations)

      Scholar (Number of citations)

      Cardiology (Pediatric)

      1

      0

      1

      0

      0

      Cardiology (Adult)

      1

      0

      1

      0

      0

      Critical Care Medicine (Pediatric)

      2

      0

      1

      0

      1

      Critical Care Medicine (Adult)

      1

      0

      0

      0

      1

      Developmental Pediatrics

      1

      0

      0

      0

      1

      Forensic Psychiatry

      1

      0

      0

      0

      1

      Gastroenterology

      1

      0

      1

      0

      0

      General Surgical Oncology

      14

      0

      14

      0

      0

      Geriatric Medicine

      0

      1

      1

      0

      0

      Gynecologic Oncology

      1

      0

      1

      0

      0

      Gynecologic Reproductive Endocrinology & Infertility

      1

      0

      0

      0

      1

      Hematology

      0

      2

      2

      0

      0

      Maternal Fetal Medicine

      1

      0

      0

      0

      1

      Medical Oncology

      0

      3

      2

      1

      0

      Pain Medicine

      5

      0

      5

      0

      0

      Pediatric Hematology & Oncology

      1

      0

      0

      1

      0

      Respirology (Pediatric)

      1

      0

      1

      0

      0

      Respirology (Adult)

      1

      0

      1

      0

      0

      Thoracic Surgery

      1

      0

      1

      0

      0

      Total number of Subspecialties

      16

      3

      13

      2

      6

      Table 3

      Specialties & Subspecialties with no citations of keywords “prognosis” or “prognostic”

      Specialty N=22

      Subspecialty N=18

       

      Adolescent Medicine

      Anatomic Pathology

      Child & Adolescent Psychiatry

      Anesthesiology

      Clinical Immunology & Allergy

      Cardiac Surgery

      Clinical Pharmacology & Toxicology

      Diagnostic Radiology

      Colorectal Surgery

      Emergency Medicine

      Endocrinology & Metabolism

      General Pathology

      Forensic Pathology

      General Surgery

      General Internal Medicine

      Hematologic Pathology

      Geriatric Psychiatry

      Internal Medicine

      Infectious Diseases

      Medical Biochemistry

      Neonatal Perinatal Medicine

      Medical Microbiology

      Nephrology

      Neuropathology

      Neuroradiology

      Nuclear Medicine

      Occupational Medicine

      Ophthalmology

      Pediatric Emergency Medicine

      Orthopedic Surgery

      Pediatric Radiology

      Pediatrics

      Pediatric Surgery

      Physical Medicine & Rehabilitation

      Rheumatology

      Psychiatry

       

      Public Health and Preventive Medicine

       

      Radiation Oncology

       

      Urology

       

      Vascular Surgery

       

      Among the 37 subspecialties there were 34 citations of the keyword “prognosis” and six for “prognostic” (Table  2). General Surgical Oncology accounted for 14 of the 34 (41.2%) citations for the keyword “prognosis”. Of the 37 subspecialties, sixteen (43.2%) contained at least one citation of the keyword “prognosis”, and three (8.1%) contained at least one citation of the keyword “prognostic” (Table  2). The combined citations for “prognosis” and “prognostic” among subspecialties were linked with the following competency roles: Medical Expert (80%), Scholar (15%), and Communicator (5%) (Table  2).

      Table 3 summarizes the 22 specialties and the 18 subspecialties that were found to have no citations of the keywords “prognosis” or “prognostic”. The following keywords had zero citations: prognosticate, prognostication, foresee, foretell, and survival estimate. Vascular Surgery was found to have one citation with the keyword life expectancy.

      Discussion

      CanMEDS 2005 is deficient in its content pertaining to the domain of prognosis. Only 26 of the 66 (39.4%) combined medical specialties and subspecialties were found to have at least one citation related to the keywords prognosis and prognostic. Overall, there were only 61 citations for the keywords prognosis and prognostic. The distribution of the total citations was skewed as 13 (21.3%) were associated with the specialty of Neurology, and 14 (22.9%) were associated with the subspecialty of General Surgical Oncology. The deficiencies within CanMEDS are further emphasized by the observation that the following specialties and subspecialties had zero citations associated with prognosis despite being heavily involved in the management of terminally ill patients: Radiation Oncology, Internal Medicine, General Internal Medicine, General Surgery, Colorectal Surgery, and Nephrology. Moreover, the following specialties and subspecialties had zero citations associated with prognosis despite being heavily involved in the management of patients who are receiving rehabilitative services: Orthopedic Surgery, Physical Medicine and Rehabilitation, and Rheumatology. These findings are consistent with published data that reports more than 90% of physicians are “reluctant to make predictions” about a patient’s illness [4], and even when provided with objective prognostic estimates, discuss it with patients and/or substitute decision makers 15% of the time [24]. Failure to disclose prognoses to terminally ill patients is emerging as a medical-legal issue [25]. In a number of cases where prognostic information was not disclosed, the rulings were in favour of the plaintiff [26]. In the United States of America, the states of California and New York have “Palliative Care Information Acts” that mandate the disclosure of prognostic information by healthcare professionals in the setting of terminal illness [27]. Moreover, failure to prognosticate is associated with numerous significant concerns that apply to the complete spectrum of bioethics [4, 28].

      Most of the keyword citations in this study were linked to the competency role of “Medical Expert” (80.3%), while 11.6% were linked with the “Scholar” role, and 8.1% were linked with the “Communicator” role. Although the CanMEDS “Medical Expert” role is an integration of the six other roles, the discussion of a complex and multifaceted issue such as prognosis requires detailed discussion under more than one competency role. Thus, competency frameworks such as CanMEDS must acknowledge that a comprehensive and effective approach to prognosis may only be achieved by articulating the need for objective computation of prognostic estimates through the “Scholar” role, together with skillful disclosure to the patient and/or substitute decision maker through the “Communicator” role. In other words, the “Scholar” role promotes “Foreseeing”, while the “Communicator” role promotes “Foretelling”. Foreseeing may be regarded as a series of scholarly activities that begin with the knowledge translation of available data on prognostic factors, instruments, and tools, followed by the computation of a prognostic estimate. Successful Foretelling is dependent on the physician being effective at “breaking bad news” [29]. Thus, physicians must develop skills to deliver prognostic estimates in a gentle and sensitive, yet, confident manner. In addition, physicians must be able to detect emotions and respond empathetically, while being able to gauge the patient’s understanding of the delivered prognostic estimate [30]. Given the inherent uncertainty and probabilistic nature, prognostic quotations should never be stated in exact terms, but rather as ranges (days to weeks, weeks to months, months to years), or as median survival [1, 4].

      A useful paradigm for incorporating prognosis into patient-centered decision making is shown in Figure  1. The process begins with a discussion about diagnosis. This is followed by a discussion pertaining to the natural history of the particular disease or affliction. The physician then computes a prognostic estimate and then communicates it to the patient. This is followed by a discussion of the available options along with success rates, benefits, risks, and burdens. Finally, the patient arrives at a decision that is consistent with his/her preferences, wishes, and values.
      http://static-content.springer.com/image/art%3A10.1186%2F1472-6920-14-93/MediaObjects/12909_2013_925_Fig1_HTML.jpg
      Figure 1

      A paradigm for incorporating prognosis into patient-centered decision making.

      Conclusions

      The provision of a truly patient-centered and ethically sound approach to healthcare is dependent upon prognosis being both formulated and communicated to the patient and/or substitute decision maker. This study has demonstrated deficient levels of content related to the domain of prognosis within the CanMEDS 2005 competency framework. It is recommended that in advance of the launching of CanMEDS 2015, reforms occur that enhance content pertaining to prognosis within the outcome documents of all medical specialties and subspecialties. In order to hold true to the Hippocratic definition of prognosis as a two dimensional construct, both quoad vitam and quoad sanantionem should be discussed. It is also recommended that foreseeing should be discussed within the context of the Scholar role while foretelling should be discussed within the context of the Communicator role. Ultimately, increased attention to prognosis within CanMEDS and other competency frameworks carries the potential to promote medical curricular reform, thus moving towards restoring prognosis as a core physician competency.

      Declarations

      Acknowledgements

      There was no funding for this project.

      Authors’ Affiliations

      (1)
      University of Toronto
      (2)
      McMaster University
      (3)
      Division of Palliative Medicine, William Osler Health System
      (4)
      University of Toronto

      References

      1. Mackillop WJ: Chapter 2. In Prognosis in Advanced Cancer. Edited by: Glare P, Christakis NA. New York: Oxford University Press; 2008.
      2. Hippocrates A: Book of Prognostics. 2004. Digireads.com
      3. Bloomfield AL: Prognosis. Arch Intern Med 1956, 97:267–268.View Article
      4. Christakis NA: Death foretold-Prophecy and prognosis in medical care. Chicago: The University of Chicago Press; 1999.
      5. Sobin LH, Wittekind C: TNM classification of malignant tumors. 6th edition. New York: Wiley-Liss; 2002.
      6. Surveillance Epidemiology and End Results: SEER cancer statistics review, 1975–2001 http://​seer.​cancer.​gov/​csr/​1975_​2001/​
      7. Downing M, Lau F, Lesperance M, Karlson N, Shaw J, Kuziemsky C, Bernard S, Hanson L, Olajide L, Head B, Ritchie C, Harrold J, Casarett D: Meta-analysis of survival prediction with Palliative Performance Scale. J Palliat Care 2007,23(4):245–52.
      8. Chan EY, Wu HY, Chan YH: Revisting the palliative performance scale: change in scores during disease trajectory predicts survival. Palliat Med 2012. Epub ahead of print
      9. Subramaniam S, Thorns A, Ridout M, Thirukkumaran T, Osborne TR: Accuracy of prognosis by PPI in hospice inpatients with cancer: a multicentre prospective study. BMJ Supportive & Palliative Care 2013. doi:10.1136/bmjspcare-2012–000239 [published online 18 April 2013]
      10. Maltoni M, Caraceni A, Brunelli C, Broeckaert B, Christakis N, Eychmueller S, Glare P, Nabal M, Vigano A, Larkin P, De Conno F, Hanks G, Kaasa S: Prognostic factors in advanced cancer patients: evidence-based clinical recommendations–a study by the Steering Committee of the European Association for Palliative Care. J Clin Oncol 2005,23(25):6240–8.View Article
      11. Knaus WA, Draper EA, Wagner DP, Zimmerman JE: APACHE II: a severity of disease classification system. Crit Care Med 1985,13(10):818–29.View Article
      12. Le Gall JR, Lemeshow S, Saulnier F: A new simplified acute physiology score (SAPS II) based on a European/North American Multicentre Study. JAMA 1993,270(24):2957–63.View Article
      13. Gospodarowicz MK, O’Sullivan B, Sobin LH: Prognostic factors in cancer. 3rd edition. New York: Wiley-Liss; 2006.
      14. Mitchell SL, Teno JM, Kiely DK, Shaffer ML, Jones RN, Prigerson HG, Volicer L, Givens JL, Hamel MB: The clinical course of advanced dementia. N Engl J Med 2009,361(16):1529–38.View Article
      15. Maida V, Ennis M, Kuziemsky C, Corban J: Wounds and survival in noncancer patients. J Palliat Med 2010,13(4):453–9.View Article
      16. Maeshima S, Osawa A, Hayashi T, Tanahashi N: Factors associated with prognosis of eating and swallowing diability after stroke: a study from a community-based stroke care system. J Stroke Cerebrovasc Dis 2013. Epub ahead of print
      17. Lo A, Tahair N, Sharp S, Bayley MT: Clinical utility of the AlphaFIM® instrument in stroke rehabilitation. Int J Stroke 2012,7(2):118–24.
      18. Maida V, Ennis M, Kesthely C: Clinical parameters associated with pressure ulcer healing in patients with advanced illness. J Pain Symptom Manage 2013. in print
      19. Englander R, Cameron T, Ballard AJ, Dodge J, Bull J, Aschenbrener CA: Toward a common taxonomy of competency domains for the health professions and competencies for physicians. Academic Medicine 2013,88(8):1088–94.View Article
      20. Frank JR, Danoff D: The CanMEDS initiative: Implementing and outcomes-based framework of physician competencies. Medical Teacher 2007, 29:642–647.View Article
      21. Royal College of Physicians and Surgeons of Canada - Information by Discipline. http://​www.​royalcollege.​ca/​rc/​faces/​oracle/​webcenter/​portalapp/​pages/​ibd.​jspx?​lang=​en&​_​afrLoop=​1319250417890967​4&​_​afrWindowMode=​0&​_​afrWindowId=​null
      22. Krippendorff K: Content Analysis: An Introduction to Its Methodology. 2nd edition. Thousand Oaks, CA: Sage; 2004.
      23. Foxit Reader 5.1 (Foxit Corporation). http://​www.​foxitsoftware.​com/​company/​press.​php?​action=​view&​page=​201110265722.​html
      24. The SUPPORT, Investigators P: A controlled trial to improve care for seriously ill hospitalized patients: the study to understand prognoses and preferences for outcomes and risks of treatment (SUPPORT). JAMA 1995,274(20):1591–8.View Article
      25. Rich BA: Prognostication in clinical medicine: prophecy or professional responsibility? J Leg Med 2002,23(3):297–358.View Article
      26. Annas GJ: Informed consent, consent, cancer, and truth in prognosis. N Engl J Med 1994,330(3):223–225.View Article
      27. Astrow AB, Popp B: The palliative care information Act in real life. N Engl J Med 2011,364(20):1885–1887.View Article
      28. Maida V, Chidwick P: Prognosis: Important information when obtaining informed consent. J Palliative Care Med 2011, 1:1.
      29. Buckman R: Breaking bad news: Why is it so difficult? BMJ 1984, 288:1597–8.View Article
      30. Back AL, Arnold RM: Discussing prognosis: “How much do you want to know?” talking to patients who are prepared for explicit information. J Clin Onc 2006,24(25):4209–12.View Article
      31. Pre-publication history

        1. The pre-publication history for this paper can be accessed here: http://​www.​biomedcentral.​com/​1472-6920/​14/​93/​prepub

      Copyright

      © Maida and Cheon; licensee BioMed Central Ltd. 2014

      This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://​creativecommons.​org/​licenses/​by/​2.​0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://​creativecommons.​org/​publicdomain/​zero/​1.​0/​) applies to the data made available in this article, unless otherwise stated.