Abstract
Objective
To describe experiences and activities of Clinical Informatics (CI) fellows since the first fellowships were accredited in 2014.
Materials and Methods
We performed a voluntary and anonymous survey of 394 alumni and current clinical informatics fellows from the graduating classes of 2016–2024 in the summer of 2022.
Results
We received 198 responses; 2% declined to participate. Most were male (62%), White (39%), 31–40 years old (72%), from primary care specialties (54%) and nonprocedural specialties (95%), and without prior informatics experience or any careers before medicine. Most fellows (87–94%) participated significantly in operations, research, coursework, quality improvement, and clinical care activities during fellowship.
Discussion
Women, underrepresented racial and ethnic minorities, and procedural physicians were underrepresented. Many incoming CI fellows did not have an informatics background. During CI fellowship, trainees earned Master’s degrees and certificates, were exposed to many different types of CI activities, and were able to spend most of their time pursuing projects that supported their personal career goals.
Conclusion
These findings represent the most comprehensive report to date of CI fellows and alumni. Physicians without prior informatics experience who are interested in CI should be encouraged to apply because CI fellowship provides a strong foundation of informatics knowledge while supporting fellows’ personal career goals. There remains a lack of women and underrepresented minorities in CI fellowship programs; efforts to expand the pipeline are needed.
Keywords: clinical informatics, medical education, graduate medical education, academic training
BACKGROUND
In 2009 the medical subspecialty of Clinical Informatics (CI) was defined1 along with the core content2 and program requirements for fellowship training.3 The first CI fellowships were accredited by the Accreditation Council for Graduate Medical Education in 20144 and since then the number of fellowships has grown to nearly 60.5
Previous estimates demonstrate a relatively strong interest among medical students in topics related to CI, however less than a third were aware of training opportunities, including fellowship.6 There are limited data available on CI fellowship applicants,7,8 and training experience.4,9 A prior survey of CI fellows and program directors examined the process and success in searching for jobs.10 However more data are needed to help attract medical undergraduate and graduate trainees to the field. This is especially true given the ongoing shortfall and uneven distribution of physician informatics specialists. Based on the most recent published data from the American Board of Preventive Medicine there are 2313 board certified CI physicians.11 This is still well below previous workforce estimates, including the American Medical Informatics Association 10×10 goal of 1 physician and nurse informaticist for every hospital in the United States.11,12
OBJECTIVE
In this study we survey the current and past CI fellows in order to describe the population and their training experiences.
MATERIALS AND METHODS
During July through September 2022, with Mass General Brigham institutional review board (IRB) exemption, we sent an anonymous, voluntary survey to clinical informatics fellows and alumni known to the AMIA Clinical Informatics Fellows (ACIF) and Clinical Informatics Program Directors (CIPD). Study data were collected using the secure, web-based software platform Redcap (Research Electronic Data Capture)13 electronic data capture tools hosted at Mass General Brigham. Data were anonymized and exported to RStudio 2022.12.0/R v3.6.314 for analysis.
Except for the first question for electronic consent to voluntarily participate, all questions on the survey were optional. We minimized combining answers from multiple questions in order to protect respondents’ anonymity. Questions were mostly single-select multiple choice, with some free text short answer, multi-select multiple choice, and ranking, including: 10 questions on background (demographics, clinical specialty, career goals) and 14 questions on experiences during fellowship. Categorical results were reported as counts and percentages. Unanswered questions were included as “Did not respond”. Free text responses (eg, specific certificates or degrees completed during fellowship, career goals) were categorized and reviewed by the first and last authors. Multi-select multiple choice questions summarized each choice as a count and percentage. Questions asking for ranking multiple activities from most to least time, leaving unselected activities not performed, were summarized in 2 ways: any ranked activity was considered as performed and shown as a count and percentage; and the activity ranked as the most time was shown as a count and percentage.
RESULTS
We sent the survey to 394 people during July through September 2022, including 242 alumni and 152 current fellows, and received 198 responses (50.3%). Of the 198 responses, 4 (2%) did not consent to voluntarily participate. Table 1 shows demographics of the 194 respondents who consented. Participants mostly had male gender, white or Asian race, not Hispanic/Latinx, and 31–40 years old (at the time of answering the survey).
Table 1.
Demographics
| Category | Choices | N (%) |
|---|---|---|
| Gender | ||
| Male | 121 (62.4%) | |
| Female | 42 (21.6%) | |
| Did not respond | 29 (14.9%) | |
| Other | 2 (1.0%) | |
| Race | ||
| White | 76 (39.2%) | |
| Asian | 60 (30.9%) | |
| Did not respond | 31 (16.0%) | |
| Black | 19 (9.8%) | |
| Other | 8 (4.1%) | |
| Hispanic/Latinx | ||
| No | 159 (82.0%) | |
| Did not respond | 31 (16.0%) | |
| Yes | 4 (2.1%) | |
| Age range | ||
| 26–30 | 6 (3.1%) | |
| 31–35 | 88 (45.4%) | |
| 36–40 | 52 (26.8%) | |
| 41–45 | 13 (6.7%) | |
| 46+ | 7 (3.6%) | |
| Did not respond | 28 (14.4%) | |
| Total | 194 (100%) |
Table 2 shows the background of the 194 participants. More than half (53.6%) of participants had a clinical specialty of internal medicine, pediatrics, or family medicine; only 5.2% had procedural specialties (eg, general surgery, urology, anesthesiology, radiation oncology). About 9% were completing additional clinical training (residency or fellowship) concurrently. A minority of participants had a career prior to medicine (21.6%). Nearly half (43.8%) had some informatics experience prior to CI fellowship (43.8%). Many participants recalled starting CI fellowship with career goals of becoming a chief medical informatics officer (CMIO) or hospital administrator, hybrid clinical care and informatics, and an academic or research career (48.9%).
Table 2.
Background
| Category | Choices | N (%) |
|---|---|---|
| Clinical specialty | ||
| Internal medicine | 58 (29.9%) | |
| Pediatrics | 31 (16.0%) | |
| Did not respond | 30 (15.5%) | |
| Pathology | 16 (8.2%) | |
| Family medicine | 15 (7.7%) | |
| Emergency medicine | 12 (6.2%) | |
| Other | 10 (5.2%) | |
| Procedural specialties | 10 (5.2%) | |
| Preventive medicine | 8 (4.1%) | |
| Psychiatry | 4 (2.1%) | |
| Any careers before medicine? | ||
| No | 97 (50.0%) | |
| Did not respond | 55 (28.4%) | |
| Yes | 42 (21.6%) | |
| Any informatics experience before clinical informatics fellowship? | ||
| Yes | 85 (43.8%) | |
| No or minimal | 60 (30.9%) | |
| Did not respond | 49 (25.3%) | |
| Career goals at the beginning of clinical informatics fellowship? | ||
| Did not respond | 62 (32.0%) | |
| CMIO or hospital administration | 38 (19.6%) | |
| Hybrid clinical care and informatics | 29 (14.9%) | |
| Academic or research | 28 (14.4%) | |
| Industry, nonprofit, or nonclinical | 14 (7.2%) | |
| Clinical care | 11 (5.7%) | |
| Unsure | 10 (5.2%) | |
| Other | 2 (1.0%) | |
| Total | 194 (100%) |
Table 3 shows fellowship experiences of the 106 respondents who, at the time of the survey, reported having a job finalized for the following year; nearly all were alumni though some current fellows were also included. Most of these participants completed at least one type of structured coursework during CI fellowship (87%). About two-thirds of these participants completed a Master’s degree (most commonly a Masters in Biomedical Informatics or similar) and two-thirds completed a Certificate program; note, some of these participants completed both a degree and a certificate. The majority of participants noted exposure to the electronic health records (EHR) Epic (92.5%) and Cerner (56.6%) at some point (including careers before medicine, medical training, other clinical activities [moonlighting], or research). Nearly all fellows participated in many different types of activities during CI fellowship; 92.5% participated in 5–8 out of 8 activity categories (operations, research, coursework, quality improvement, clinical care, consulting/industry, teaching, other). Nearly all of the participants spent some time on research, clinical care, teaching, operations, coursework, and quality improvement. They spent the most time on operations (n = 38, 35.8%), research (n = 16, 15.1%), coursework (n = 15, 14.2%), and quality improvement (n = 12, 11.3%). These participants commonly led or significantly participated in the following types of projects during CI fellowship: data extraction from an EHR or building information technology (IT) systems, and EHR transition or Go Live of a digital product.
Table 3.
Fellowship experiences
| Category | Choices | N (%) |
|---|---|---|
| Concurrent clinical training (residency or other fellowship) | ||
| No | 96 (90.6%) | |
| Yes | 9 (8.5%) | |
| Did not respond | 1 (0.9%) | |
| Structured coursework, including Master’s degree, certificate, or other | ||
| Yes | 94 (88.7%) | |
| No | 12 (11.3%) | |
| Master’s degree | ||
| No | 64 (60.4%) | |
| Yes | 42 (39.6%) | |
| Certificate | ||
| No | 73 (68.9%) | |
| Yes | 33 (31.1%) | |
| Electronic health record (EHR) exposure | ||
| Epic | 98 (92.0%) | |
| Cerner | 60 (56.6%) | |
| CPRS | 30 (28.3%) | |
| All Scripts | 18 (17.0%) | |
| Meditech | 17 (16.0%) | |
| Other | 14 (13.2%) | |
| Number of activity types (out of 8 categories: Operations, Research, Coursework, Quality improvement, Clinical care, Consulting/Industry, Teaching, Other) | ||
| 8 | 22 (20.8%) | |
| 7 | 43 (40.6%) | |
| 6 | 21 (19.8%) | |
| 5 | 12 (11.3%) | |
| 4 | 3 (2.8%) | |
| 3 | 2 (1.9%) | |
| 2 | 2 (1.9%) | |
| 1 | 0 (0.0%) | |
| 0 | 1 (0.9%) | |
| Activity | ||
| Research | 100 (94.3%) | |
| Clinical care | 97 (91.5%) | |
| Teaching | 96 (90.6%) | |
| Operations | 95 (89.6%) | |
| Coursework | 92 (86.8%) | |
| Quality improvement | 95 (89.6%) | |
| Consulting or industry | 80 (75.5%) | |
| Other | 30 (28.3%) | |
| Project types | ||
| Data extraction from an EHR or building information technology (IT) systems | 77 (72.6%) | |
| Implementing a CDS tool in an EHR | 68 (64.2%) | |
| EHR transition or Go Live of a digital product | 66 (62.3%) | |
| Resident or medical student education | 48 (45.3%) | |
| Developing ML models | 39 (36.8%) | |
| Developing an app or software | 36 (34.0%) | |
| Health informatics policy or advocacy | 29 (27.4%) | |
| Consulting or industry | 25 (23.6%) | |
| Cybersecurity or information security | 9 (8.5%) | |
| Other | 8 (7.5%) | |
| Total | 106 (100%) |
DISCUSSION
More than half of the 394 CI fellows and alumni responded to our voluntary and anonymous survey during the summer of 2022. This represents the most comprehensive study to date of CI fellows and alumni, though there is likely overlap in respondents from prior studies.7,8,13 This creates an assessment of the current state of CI fellowships and a baseline for future comparisons.
Our survey results demonstrate that although many fellows entered training with some experience, almost a third had minimal or no informatics background. This should serve as encouragement to medical students and residents who are curious about CI but do not see themselves as viable candidates for the specialty. Between the structured coursework available to nearly all fellows, and the broad set of activities and projects, the training can provide all one needs to launch an informatics career.
We also redemonstrate the underrepresentation of women and minorities in the field. While not dissimilar from other medical specialties, the need is perhaps more critical in CI. As we increasingly seek to use artificial intelligence to inform clinical care, a diverse workforce will only help to avoid the propagation of biased data and ensure responsible implementation of these tools.
Informatics activities
Few fellows had careers before medicine, but nearly half reported some experience with informatics prior to CI training. This may be similar to the 44% of Bell et al’s survey respondents who reported informatics-related experience prior to CI fellowship,7 with the same limitations of comparison mentioned above. The fact that a substantial proportion of CI fellows had no prior informatics experience should encourage people without informatics experience who are interested and considering applying to CI fellowship programs.
The most common aspirations at the beginning of CI fellowship were to take on the role of CMIO or a similar administrative leader, and to take on the dual roles of clinician and clinical informatician in academic or research careers. This aligns very well with the types of activities that fellows participated in during fellowship (operations, research, coursework, and quality improvement). It is encouraging that fellows were able to pursue activities during CI fellowship that directly supported their career goals. At the same time, nearly all CI fellows participated in many different types of informatics activities and had opportunities for both structured coursework and experiential learning through projects.15,16 It is essential that CI fellowships provide fellows with a broad-based informatics foundation and well-rounded educational experiences, regardless of career goals.2,3,17 Clinical informaticians do not all need to be experts at writing code to extract data from an EHR, or at building clinical decision support (CDS) tools in an EHR; but they do need to understand the process and potential strengths, weaknesses, and limitations of different approaches, and to be able to fluently communicate with colleagues who know more or less about these processes.2 Our results provide reassurance that CI fellowships seem to be fulfilling these missions. For example, although only 7% started CI fellowship with ambitions for a career in industry, a majority of fellows (76%) participated in some consulting or industry activities during fellowship and anecdotally, fellows’ most common request to ACIF was for more exposure to industry. Fellows were not only exposed to a variety of projects, but they led or were significantly involved in many different types of projects. Less than 10% of participants led or were significantly involved in projects related to cybersecurity or information security, though it is unclear whether this was due to a lack of opportunities or fellows’ interest. This is an area of need given the growing threat of cybercrime in the US healthcare industry and in hospitals in particular.18
Diversity
We found that the majority of CI fellows were males in their 30s and either white or of Asian descent. These findings are in keeping with Van Cain’s prior survey of fellowship graduates10 as well as with publicly available data on CI applicants from the Electronic Residency Application System (ERAS)19 and prior studies by Bell et al.7,8
Our racial distribution is similar to that of United States (US) allopathic medical school applicants and matriculants.20 However, the gender distribution is different from all US medical students, with considerably fewer women in CI fellowships. One study has suggested women have more stable preferences regarding planned medical specialties,21 which may indicate this gender disparity originates prior to medical school matriculation.
The CI community must continue to try to increase the diversity in CI. The National Academy of Medicine concluded that increasing the racial and ethnic diversity of healthcare professionals is associated with more equitable access to healthcare.22 This is even more important for CI because clinical informaticians can impact healthcare at the systems and population levels, magnifying both positive and negative effects and can even cause unintended consequences.23 The current Accreditation Council for Graduate Medical Education (ACGME) Common Program Requirements note the need to recruit and retain a diverse workforce of residents/fellows, faculty, and staff.24 CI could consider trying to adapt strategies to close the gender gap in related fields, such as surgical specialties,25 science, technology, engineering, and math (STEM)26 and data science.27
Internal medicine and pediatrics were the most well-represented specialties (45.9%) and very few were proceduralists (5.2%). This may be more diverse than the previously mentioned results of Bell et al7 (54.5% internal medicine or pediatrics) and Van Cain and Hron10 (64.3%), but it is challenging to draw a meaningful conclusion with such small samples, different methods, and partially shared cohorts among these 3 studies. These specialties are dramatically overrepresented in CI when compared to the 19.0% of total active US physicians in internal medicine and pediatrics according to the American Medical Association’s 2021 Physician Masterfile. This may be related to the differential impact of the EHR on clinical work (less time spent using the EHR while in an operating room); time devoted in clinical training to gain and maintain procedural skills; and differences in physician specialty salaries. Training programs with significant time for research or quality improvement (QI) lend themselves well to incorporating informatics.
Limitations
Our study is limited by our overall response rate as well as the number of respondents who chose not to answer specific questions, especially with respect to gender, race and ethnicity. Still, our findings are comparable to previously published data on CI applicants and fellows. Unfortunately, because publicly available ERAS data are reported based on the residency review committee under which CI programs are accredited (eg, internal medicine-based programs separate from pediatrics-based programs), it is impossible to discern the demographics of the full applicant pool. Moreover, ERAS data represent applicants to CI programs and not those that matriculated successfully, further limiting our ability to compare the groups.
CONCLUSION
Our findings help establish a baseline of CI fellowship experiences and fellow diversity that can be used to measure change in the future. Physicians without prior informatics experience who are interested in CI should not feel discouraged from applying. CI fellowships provide a strong foundation of informatics knowledge while supporting each fellow’s personal career goals. The CI community needs innovative strategies to recruit diverse applicants to the field in order to expand representation of women and underrepresented minorities in the workforce.
ACKNOWLEDGMENTS
The authors would like to thank the clinical informatics alumni and fellows who responded to this survey. This work would not have been possible without their participation.
Contributor Information
Ellen Kim, Department of Radiation Oncology, Brigham & Women’s Hospital, Boston, Massachusetts, USA.
Melissa Van Cain, Department of Pediatrics, The University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, USA; Department of Medical Informatics, The University of Oklahoma-Tulsa, Tulsa, Oklahoma, USA.
Jonathan Hron, Department of Pediatrics, Division of General Pediatrics, Boston Children’s Hospital, Boston, Massachusetts, USA; Department of Pediatrics, Harvard Medical School, Boston, Massachusetts, USA.
FUNDING
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
AUTHOR CONTRIBUTIONS
All of the authors have had substantial contributions to the conception and design of the work; acquisition, analysis, or interpretation of data for the work; drafting and revising the manuscript for important intellectual content; final approval of the version to be published; and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
CONFLICT OF INTEREST STATEMENT
The authors do not have any competing interests to declare.
DATA AVAILABILITY
In order to preserve respondents’ privacy, these data are only available to the study authors.
REFERENCES
- 1. Detmer DE, Lumpkin JR, Williamson JJ.. Defining the medical subspecialty of clinical informatics. J Am Med Inform Assoc 2009; 16 (2): 167–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Gardner RM, Overhage JM, Steen EB, et al. ; AMIA Board of Directors. Core content for the subspecialty of clinical informatics. J Am Med Inform Assoc 2009; 16 (2): 153–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Safran C, Shabot MM, Munger BS, et al. ; AMIA Board of Directors. Program requirements for fellowship education in the subspecialty of clinical informatics. J Am Med Inform Assoc 2009; 16 (2): 158–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Longhurst CA, Pageler NM, Palma JP, et al. Early experiences of accredited clinical informatics fellowships. J Am Med Inform Assoc 2016; 23 (4): 829–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. AMIA.org. Informatics Academic Programs Search. AMIA – Am Med Inform Assoc2023. https://amia.org/careers-certifications/informatics-academic-programs. Accessed February 5, 2023.
- 6. Banerjee R, George P, Priebe C, Alper E.. Medical student awareness of and interest in clinical informatics. J Am Med Inform Assoc 2015; 22 (e1): e42–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Bell DS, Baldwin K, Bell EJ, et al. Characteristics of the national applicant pool for clinical informatics fellowships (2016–2017). AMIA Annu Symp Proc AMIA Symp 2018; 2018: 225–31. [PMC free article] [PubMed] [Google Scholar]
- 8. Bell DS, Baldwin K, Bell Iii EJ.. Characteristics of the national applicant pool for clinical informatics fellowships (2018-2020). AMIA Annu Symp Proc2022; 2022: 241–8. [PMC free article] [PubMed] [Google Scholar]
- 9. Palma JP, Hron JD, Luberti AA.. Early experiences with combined fellowship training in clinical informatics. J Am Med Inform Assoc 2020; 27 (5): 788–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Van Cain M, Hron J. Clinical informatics fellowship graduates jobs survey. 2021. https://amia.org/education-events/amia-2021-virtual-clinical-informatics-conference/. Accessed March 13, 2013.
- 11. Safran C, Detmer DE.. Computerized physician order entry systems and medication errors. JAMA 2005; 294 (2): 179; author reply 180–81. [DOI] [PubMed] [Google Scholar]
- 12. Hersh W, Williamson J.. Educating 10,000 informaticians by 2010: the AMIA 10x10 program. Int J Med Inform 2007; 76 (5–6): 377–82. [DOI] [PubMed] [Google Scholar]
- 13. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG.. Research electronic data capture (REDCap) – a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009; 42 (2): 377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. R Core Team. R: The R Project for Statistical Computing. 2021. https://www.r-project.org/. Accessed February 2, 2023.
- 15. Subash M, Sakumoto M, Bass J, et al. The emerging role of clinical informatics fellows in service learning during the COVID-19 pandemic. J Am Med Inform Assoc 2021; 28 (3): 487–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Jin DP, Samuel S, Bowden K, Mohan V, Gold JA.. Just-in-time electronic health record retraining to support clinician redeployment during the COVID-19 surge. Appl Clin Inform 2022; 13 (5): 949–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Silverman HD, Steen EB, Carpenito JN, Ondrula CJ, Williamson JJ, Fridsma DB.. Domains, tasks, and knowledge for clinical informatics subspecialty practice: results of a practice analysis. J Am Med Inform Assoc 2019; 26 (7): 586–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Jalali MS, Kaiser JP.. Cybersecurity in hospitals: a systematic, organizational perspective. J Med Internet Res 2018; 20 (5): e10059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. ERAS Statistics. AAMC. https://www.aamc.org/data-reports/data/eras-statistics-data. Accessed May 14, 2023.
- 20. Smith MM, Rose SH, Schroeder DR, Long TR.. Diversity of United States medical students by region compared to US census data. Adv Med Educ Pract 2015; 6: 367–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Burkhardt J, DesJardins S, Gruppen L.. Diversity of the physician workforce: specialty choice decisions during medical school. PLoS One 2021; 16 (11): e0259434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Morris DB, Gruppuso PA, McGee HA, Murillo AL, Grover A, Adashi EY.. Diversity of the national medical student body – four decades of inequities. N Engl J Med 2021; 384 (17): 1661–8. [DOI] [PubMed] [Google Scholar]
- 23. Takshi S. Unexpected inequality: disparate-impact from artificial intelligence in healthcare decisions. J Law Health2021; 34 (2): 215–51. [PubMed] [Google Scholar]
- 24. Accreditation Council for Graduate Medical Education (ACGME). Common Program Requirements (Fellowship). 2023. https://www.acgme.org/globalassets/pfassets/programrequirements/cprfellowship_2023v2.pdf.Accessed March 2, 2023.
- 25. Bennett CL, Baker O, Rangel EL, Marsh RH.. The gender gap in surgical residencies. JAMA Surg 2020; 155 (9): 893–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Wang M-T, Degol JL.. Gender gap in science, technology, engineering, and mathematics (STEM): current knowledge, implications for practice, policy, and future directions. Educ Psychol Rev 2017; 29 (1): 119–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Berman FD, Bourne PE.. Let’s make gender diversity in data science a priority right from the start. PLoS Biol 2015; 13 (7): e1002206. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
In order to preserve respondents’ privacy, these data are only available to the study authors.