Author + information
- aDivision of Interventional Cardiology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts
- bSafety, Quality, Informatics and Leadership (SQIL) Program, 2016–17, Harvard Medical School, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. Ankur Kalra, Division of Interventional Cardiology, Beth Israel Deaconess Medical Center, Harvard Medical School, 185 Pilgrim Road, Baker 4, Boston, Massachusetts 02215.
Procedure logging is an integral part of the fellowship curriculum in cardiovascular medicine and its subspecialties, particularly for the cardiovascular disease fellows who are acquiring advanced training in interventional cardiology, structural heart diseases, or clinical cardiac electrophysiology. Appropriate, comprehensive, and timely procedure logging is crucial for the program directors to assess their fellows’ development and learning as they progress in their fellowship year to become independent proceduralists. The Core Cardiology Training Statement 4 documents from the American College of Cardiology, American Heart Association, and the Accreditation Council for Graduate Medical Education (ACGME) have categorically defined the number of procedures that fellows are required to perform as primary operators for successful completion of their respective subspecialty fellowships in cardiovascular medicine (1–3). In addition, the ACGME’s Next Accreditation System and the Milestones initiative have enhanced the comprehensiveness of case logging for cardiovascular medicine fellows, thereby augmenting its significance for trainees and program directors (4). A welcome initiative from the Society of Cardiovascular Angiography and Interventions to improve the process of case/procedure logging for interventional cardiology fellows, launched through the fellows-in-training online portal under the leadership of Dr. Ehtisham Mahmud, aims to standardize the process for training programs across the United States (5).
At our interventional cardiology fellowship training program at Beth Israel Deaconess Medical Center, Harvard Medical School, we have designed an online tool that not only ensures secure, standardized, and timely case/procedure logging for the fellows, but also accounts for fellow efficiency in the cardiac catheterization laboratory and aids in the equitable distribution of percutaneous coronary interventions (PCIs) among interventional fellows. This is done using a Microsoft Excel algorithm and Microsoft Access 2013 database, both hosted on a Microsoft SharePoint Online 2013 platform (Microsoft, Redmond, Washington). Termed the “catheterization laboratory T-score,” the score is derived by a simple mathematical calculation of the number of PCIs performed by a fellow divided by the number of days spent in the cardiac catheterization laboratory during the interventional training year. The T-score takes into account days spent outside of the cardiac catheterization laboratory (for vacation, conferences, research, structural heart rotations, or any personal “off” days due to reasons not listed previously), and allows the fellows to track their efficiency in the cardiac catheterization laboratory.
Calculation of the T-Score
The denominator (catheterization laboratory days per fellow) is derived from a Microsoft Excel spreadsheet hosted in an online Microsoft SharePoint server that takes into account the number of business days the laboratory is open during the academic year during which the respective fellow is present in the catheterization laboratory. This is derived from the annual fellows’ schedule stored in a Microsoft Excel spreadsheet. This denominator gets automatically updated incrementally in a Microsoft Excel table each day at midnight from the previously mentioned spreadsheet using a “VLOOKUP” formula. The days that a respective fellow is not scheduled to be in the cardiac catheterization laboratory on a regular working day are not counted in the denominator. The numerator in the formula is calculated using data in a Microsoft Access database, also hosted in the same Microsoft SharePoint Online 2013 server. Fellows log their procedures at the end of each day in a comprehensive form in Microsoft Access. No identifiable patient data are stored in the online database. To enable auditing, each patient log entered into the database is tagged with an identifier derived from the medical record number after hashing it into a 256-bit alphanumeric string using a freely available open source SHA-256 algorithm (6). This hash is 1-way and cannot be decrypted backward. A running list of the hash strings and corresponding medical record numbers is maintained within the hospital information technology secure storage to ensure compliance with the Health Insurance Portability and Accountability Act (7) regulations. From the data that are entered by the fellows each day, the number of PCIs is obtained by a query that feeds into the Microsoft Excel T-score table using an online Office Data Connection file. The data are updated at the end of each day, after all procedures have been logged. This number is divided automatically by the aggregate number of cardiac catheterization laboratory days to generate the T-score (Figures 1A and 1B). A standard deviation from the mean is automatically calculated for each fellow each time the T-score gets updated to identify fellow(s) with an above/below average T-score and to help assign cases in the catheterization laboratory the following day.
Since its inception and implementation in the interventional fellowship curriculum at our program, the T-score has significantly affected fellows’ efficiency and experience in the cardiac catheterization laboratory:
1. Distribution and prioritization of PCIs. The T-score is utilized on a daily basis to assign cases to the fellows. The interventional fellow with the lowest T-score gets priority for performing PCIs to allow “catch up” with other colleagues. The T-score model on Microsoft SharePoint platform also allows for real-time simulation to predict the T-score on the basis of the number of cases performed by each fellow, thus allowing for chronological distribution of cases among fellows throughout the day.
2. Equitable distribution. By eliminating the concept of absolute number of PCIs per fellow, which varies depending upon the actual number of days spent in the catheterization laboratory at a given time during the year, and taking into account the number of PCIs performed matched to the number of days spent in the laboratory per fellow, the T-score ensures equitable case distribution and eliminates any possible tension among fellows. For example, if fellow A has performed 32 PCIs and fellow B has performed 44 PCIs, the T-score algorithm will still prioritize fellow B over fellow A, if fellow A was in the laboratory for 30 days (T-score = 1.07) compared to fellow B who was in the laboratory for 45 days (T-score = 0.98); this result would be counterintuitive if case distribution occurred on the basis of a comparison of absolute PCI numbers (Figure 1).
3. Catheterization laboratory scheduling. The T-score aids in successful schedule implementation for the academic year, as for this model to be successful in providing equal number of cases to all fellows by the end of the training year, the number of days spent in the cardiac catheterization laboratory per fellow must be equal. Therefore, the T-score concept also inculcates discipline among fellows, as there is more sensitivity toward days missed or spent outside of the cardiac catheterization laboratory. Any day lost cannot be regained unless a “switch” is made with another fellow on research/vacation.
4. Timely case/procedure logging. As case assignments for the next day are dependent upon the dynamic T-score that gets updated each night, the fellows meticulously log their cases at the end of each day, thereby ensuring timely case/procedure logging for the program directors to review their fellows’ progress throughout the year. Timely case/procedure logging also ensures fulfillment of credentialing and graduation requirements for the fellows. These PCI numbers can be audited anytime and cross-checked with the official hospital reporting system on a monthly/quarterly basis for accuracy.
5. Didactic and fellow development (fellows’ milestones). The comprehensive logging format allows fellows to “flag” interesting cases on a daily basis and has separate fields for complex procedures like unprotected left main coronary artery interventions, thereby building a personal repository of case files for fellows that can be aggregated together for easy identification and efficient retrievability for discussion during fellows’ didactic sessions. It also allows the program directors to appraise the breadth and quality of procedural experience that fellows get exposed to during their interventional year of training, thereby facilitating a more comprehensive fellow evaluation and feedback, in sync with the Milestones initiative of the ACGME (4).
The T-score model has also evolved into a vital tool for assessing the overall efficiency of the interventional fellowship class. A “group T-score” can be calculated for sum total PCIs (updated daily) divided by the total fellow-days (also updated daily), and this metric can be utilized to derive the overall efficiency of the training program, providing an indexed comparison tool among different interventional cardiology fellowship training programs across the nation vis-à-vis absolute number of PCIs.
The implementation of the T-score model may not be generalizable to all interventional cardiology fellowship programs, as there are some that involve multiple hospital sites and deploy 1 fellow per hospital site on a rotational basis during their fellowship training period. In this case, a site-specific T-score may be applicable using the concept described previously. Also, the goal or target fellows’ T-score will depend upon the number of PCIs performed annually as well as the number of interventional cardiology fellows training in the program. However, computing the average number of working days/year for the current academic year, starting from July 1, 2015, to June 30, 2016 (252 working days, given 104 weekend days and 11 public holidays), a target annual T-score of 1.0 is the minimum each interventional cardiology fellow should strive for, to achieve a total of 250 cardiac interventions/year to meet the ACGME requirements (8).
- John A. Ambrose, MD ( and )
- Sundararajan Srikanth, MD
RESPONSE: What Should Be the Appropriate Procedure for Filling Out a Procedure Log?
For a fellow to satisfactorily complete interventional cardiology training, the Accreditation Council for Graduate Medical Education mandates that he or she must perform ≥250 therapeutic and interventional cases documented in a case list that is attested to by the program director (PD). The fellow must participate in procedural planning, critical technical manipulation during the case, and post-procedure management.
Dr. Kalra and colleagues addressed the question of accounting for the fellow’s laboratory time and how to ensure that there is an equitable case distribution. They developed an online procedure log tool called the T-score, which in programs with multiple fellows should ensure that 1 fellow is not performing more cases or more challenging cases than another. Although this approach may be reasonable in large fellowships, published data indicate that the vast majority of interventional fellowships in the United States have only 1 to 2 fellows/year in 1 or multiple hospital settings. Thus, the approach by Dr. Kalra and colleagues—although laudable—may not be necessary or broadly applicable. Furthermore, in programs with multiple interventional faculty members, how would the PD know that the procedure log accurately reflects the level of participation of a fellow in each case?
Although the emphasis of the interventional T-score as explained by Dr. Kalra and colleagues is on improving efficiency and experience of the fellow, there is more to be desired from a log. Adding qualitative to quantitative aspects would ensure that the PD has an adequate assessment of each fellow’s strengths and weaknesses, which can then be addressed accordingly. Although 1 type of procedure log may not fit all sizes and types of programs, it must include the necessary information to help assess and train the fellow. To fulfill the Accreditation Council for Graduate Medical Education’s mandate, we believe logs of the future should include more than the usual data, including information on procedural judgment and the fellow’s technical skills. This might require a form filled out by the supervising faculty member after each or most cases that includes aspects of procedural planning and judgment, guide catheter and wire manipulation, or stent deployment. This information could then be simply added to the log with a check mark (satisfactory, not satisfactory, and not applicable) for each aspect.
The remarkable advances in technology including data storage and communications have made it easier to achieve the aims of any educational endeavor. We might also take a page from our surgical colleagues to develop more efficient and comprehensive methods of capturing a trainee’s experience. This may include a cloud storage–based accumulation of case logs along with comments from the supervising faculty. One could also develop a smartphone-based application that could quickly populate a log just by reading a QR code from the patient’s identification tag. There is also the concept of the electronic logbook, which has been developed in the United Kingdom and Ireland. The individual cases of each surgeon are stored in a portfolio with the goal of providing seamless integration to any vehicle adopted by any training body/organization that requires data for assessing surgical training. Needless to say, the privacy of these data would have to be maintained independent of the format utilized. These enhanced storage modalities promise expansion of the range and applicability of the traditional procedure log book beyond that suggested by the prosaic dictionary definition of a “log” as an official record of the event.
The authors are grateful for the enthusiastic participation of Drs. Ali Andalib, Marie-France Poulin, and Ronnie Ramadan, interventional cardiology fellows in the 2015–16 class, in the implementation of the T-score model at the Beth Israel Deaconess Medical Center interventional cardiology fellowship program. The authors are also grateful for mentoring and support from Drs. Jeffrey J. Popma and Donald E. Cutlip, Program Director and Division Chief of Interventional Cardiology at Beth Israel Deaconess Medical Center, respectively.
- 2016 American College of Cardiology Foundation
- Halperin J.L.,
- Williams E.S.,
- Fuster V.
- King S.B. 3rd.,
- Babb J.D.,
- Bates E.R.,
- et al.
- Sinha S.S.,
- Julien H.M.,
- Krim S.R.,
- et al.,
- for the American College of Cardiology Fellow-In-Training Section Leadership Council
- Palaniswamy C.
- ↵The Society for Cardiovascular Angiography and Interventions. Fellows in training online portal. Available at: http://www.scai.org/FIT/about-fit-portal#.Vugmthj-hG9. Accessed March 15, 2016.
- ↵XORBIN. SHA-256 hash calculator. Available at: http://www.xorbin.com/tools/sha256-hash-calculator. Accessed March 15, 2016.
- Harold J.G.,
- Bass T.A.,
- Bashore T.M.,
- et al.