In 2010, the American Board of Medical Specialties (ABMS) approved subspecialty certification in clinical informatics.1 The first board examination was offered jointly in 2013 by the American Board of Preventive Medicine (ABPM) and the American Board of Pathology. Physicians with an unrestricted and valid license to practice medicine in the United States or Canada, who were boarded in a primary ABMS specialty (eg, anesthesiology), and who had a specified amount of work experience in an informatics domain, could apply for approval to sit for the clinical informatics board certification examination under a practice pathway.2 Requirements of the practice pathway included at least 3 years out of the previous 5 years of at least 25% full-time effort in clinical informatics, which may include research and teaching.2 Through the year 2022, the practice pathway or successful completion of a 24-month Accreditation Council of Graduate Medical Education (ACGME) fellowship in clinical informatics are options to qualify to sit for the board examination. Beginning in 2023, the ABPM has stipulated that the practice pathway will no longer be an option; rather, ACGME-accredited fellowship training will be required for all candidates.2
There currently is a 10-year, 4-part maintenance-of-certification process cycle requiring continuation of a valid US or Canadian medical license, continuing medical education credits including patient safety modules, passing a written examination, and practice performance projects.3 The requirements for the maintenance of certification process are currently under debate among the clinical informatics diplomates.
A search of the American Board of Anesthesiology and ABPM websites in January 2016 revealed 36 individuals with active board certification in anesthesiology among the 1105 clinical informatics diplomates.4 We surveyed this group of double-boarded anesthesia informaticians about their experience and attitude regarding training in the field of clinical informatics.
A survey with 13 questions (11 structured and 2 open-ended; see Supplemental Digital Content, Survey, http://links.lww.com/AA/C291) was developed via a Delphi process between the first and second authors (K.A.P., R.H.E.).5 The survey was implemented as a Microsoft Word form with checkboxes and text fields for free text entry (Microsoft, Redmond, WA). Results were manually tabulated into an Excel workbook (Microsoft) without including subject identifiers. The study was approved by the Mayo Clinic institutional review board on February 20, 2016 (IRB #16-002766), with a waiver of written informed consent.
The Word document was circulated among 18 faculty members at the first author’s institution (Mayo Clinic, Phoenix, AZ). Feedback was obtained on issues related to the clarity and relevance of the questions, and the technical process of completing the questionnaire. No issues were identified.
The survey was sent by e-mail on February 29, 2016, to all (100%) 36 board-certified anesthesiologists who had passed their clinical informatics boards as of January 1, 2016. Reminders were sent at 1-week intervals over a period of 4 weeks to nonrespondents. There were no incentives offered for participation. Surveys were returned via e-mail, so a single response per individual could be assured. The respondent’s name was not entered on the survey itself.
The method of Blyth-Still-Casella was used to calculate exact 95% confidence intervals (CIs) for responses to the individual questions (StatXact-11; Cytel, Cambridge, MA).6 To be able to claim that an observed percentage represented the reply of “most” respondents, either the 95% lower confidence limit of that percentage had to be >50%, or the 95% upper confidence limit had to be <50%. We refer to “many” when the CI included 50%. These are independent analyses, even though the question responses represent multivariate data. Associations among potentially meaningfully related responses were therefore tested, and none were significant (see Supplemental Digital Content, Survey, http://links.lww.com/AA/C291). There were too few answers provided for the free text questions to be analyzed reliably.
Responses were received from 26 of the 36 surveyed population, each of whom answered all the structured questions (Table). All 26 had qualified for the clinical informatics board certification via the practice pathway.
Most respondents had previous clinical informatics work experience (92.3%; 95% CI, 76.6%–98.6%). Most respondents did not have formal training in informatics (15.4%; 95% CI, 5.4%–33.4%).
Most respondents had an informatics role outside of the department (76.9%; 95% CI, 57.9%–89.4%). Although many had an informatics role in the anesthesia department (69.2%; 95% CI, 49.4%–84.6%), most did not write computer code at least several days a week (80.8%; 95% CI, 62.6%–92.1%). Approximately half stated that they never wrote computer code (50.0%, 95% CI, 29.9%–70.1%). Most also worked clinically in operating rooms (92.3%; 95% CI, 76.6%–98.6%), averaging 3.0 ± 1.6 days per week.
Most respondents stated that they would take the examination if they had to do it over again (96.2%; 95% CI, 82.0%–99.8%). Although many indicated that they had received value from having passed the boards (69.2%; 95% CI, 49.4%–84.6%), most did not experience a change in the role because of their achievement (26.9%; 95% CI, 11.6%–46.5%).
There was a uniform opinion that the specialty should encourage anesthesiologists to be interested in or participate in informatics (100.0%; 95% CI, 88.4%–100.0%). However, most had skepticism (“unlikely” or “highly unlikely”) that anesthesia residents would elect to complete a full 2-year ACGME-accredited fellowship to become eligible to sit for the clinical informatics board after the practice pathway expires (92.3%; 95% CI, 76.6%–98.6%).
Our survey found that all respondents thought that the specialty should encourage interest in anesthesia informatics. Nearly all the respondents expressed satisfaction with having gone through the process of becoming board-certified in clinical informatics. Most had previous experience in informatics, with knowledge gained from practical experience at work, not via formal training. However, there was skepticism that a substantive number of anesthesia residents will choose to enroll in a full-time, 24-month ACGME-accredited fellowship in the future to allow sitting for the examination.
Because the window to qualify for board certification via the practice pathway was extended (in November 2016) from 2017 to 2022, anesthesia residents and faculty currently have additional opportunities to qualify. For example, departments interested in cultivating informatics expertise could establish a hybrid part-time faculty position alongside a 2 days a week nonaccredited informatics fellowship, similar to what is currently done for some postresidency regional anesthesia training programs. Such training can be included in the practice time requirement.2 Such informatics training might also be included as part of existing ACGME anesthesia fellowship programs leading to ABMS subspecialty certification (eg, critical care medicine or pediatric anesthesiology). Alternatively, a new attending physician with some previous informatics experience could be hired as part-time clinical faculty with an informatics role within the department, with time assigned for completion of a Master’s degree in clinical informatics. Such a program could be completed locally, if available, or via distance learning. If the participant already had significant experience, gaps in knowledge could be filled in through the American Medical Informatics Association 10 × 10 virtual course series.7 For motivated individuals, most of the didactic information can be learned from Shortliffe and Camino’s textbook on biomedical informatics.8 Because the breadth of the examination is wide and covers material outside anesthesia informatics,9 studying will be required for most individuals to pass the examination.
In conclusion, this survey of recent clinical informatics diplomates indicates that anesthesia informatics training is valuable and should be encouraged, but that alternatives to the ACGME informatics fellowship pathway will need to be pursued.
Name: Karl A. Poterack, MD.
Contribution: This author helped obtain the data and write the article.
Name: Richard H. Epstein, MD.
Contribution: This author helped design the study, perform the analysis, and write the article.
Name: Franklin Dexter, MD, PhD.
Contribution: This author helped perform the analysis and write the article.
This manuscript was handled by: Maxime Cannesson, MD, PhD.
1. Detmer DE, Shortliffe EHClinical informatics: prospects for a new medical subspecialty. JAMA. 2014;311:2067–2068.
2. American Board of Preventative Medicine. Subspecialty Requirements: Clinical Informatics. Available at: http://www.theabpm.org/wp-content/uploads/2017/05/ClinicalInformatics.pdf
. Accessed September 17, 2017.
3. American Board of Preventative Medicine. Maintenance of Certification. Available at: https://www.theabpm.org/maintain-certification/
. Accessed November 21, 2017.
4. American Medical Informatics Association. 2015 Clinical Informatics Diplomates. Available at: https://www.amia.org/clinical-informatics-board-review-course/2015-diplomates
Accessed September 22, 2017
5. Wikipedia. Delphi method. https://en.wikipedia.org/wiki/Delphi_method
. Accessed November 21, 2017.
6. Casella GRefining binomial confidence intervals. Can J Stat. 1986;14:113–129.
7. American Medical Informatics Association. AMIA 10x10 Courses. Available at: https://www.amia.org/education/10x10-courses
. Accessed September 22, 2017.
8. Shortliffe E, Cimino JBiomedical Informatics: Computer Applications in Health Care and Biomedicine. 2014.4th ed. New York, NY: Springer;
9. The American Board of Preventative Medicine. Clinical Informatics 201 Examination Blueprint. Available at: https://www.theabpm.org/wp-content/uploads/2017/09/CI-Content-Outline.pdf
. Accessed September 22, 2017.