Allowing Surgical Pathology Fellows to Release Preliminary Reports Increases Learner Independence and Satisfaction

This study was exempt from institutional review board approval.

The pathology LIS at our institution was changed from Copath (Cerner, Kansas City, Missouri) to SoftPathDx in 2017. The implementation of SoftPathDx allowed new functionality, including the ability to release preliminary pathology reports. Release of preliminary pathology reports was adopted as routine practice for cytology smear interpretation and frozen section interpretation. At the time of implementation of preliminary pathology reports in these 2 settings, broad email communication was sent to the clinical and surgical services. Once this process became routine (>1 year after implementation), the Division of Anatomic Pathology Executive Committee approved a pilot to allow surgical pathology fellows to release preliminary reports to increase their independence. Additional email communication was sent to the clinical services about this pilot if deemed necessary by the pathology subspecialty section heads.

For consecutive 1-week pilot periods, 4 board-certified (anatomic pathology, American Board of Pathology) surgical pathology fellows in the last 2 months of their ACGME-accredited 12-month fellowship were allowed to independently manage cases sent from outside institutions for confirmatory review before additional treatment was undertaken at our facility. These fellows had already completed a full 1-month rotation on this outside slide review service. In the pilot period, fellows decided whether to release a preliminary report, to share with a subspecialty pathologist in consultation and then release a preliminary report, or to not release a preliminary report and show the case directly to a general surgical pathology attending physician (Figure 1). The fellows were instructed not to release a preliminary report on cases where they had any concern about the accuracy of the outside diagnosis or the need for any additional workup with immunohistochemistry, molecular studies, etc. If the fellows decided they wanted to sign out a preliminary report on a case, they gave the case to the secretarial staff to create the preliminary report in SoftPathDx. The secretarial staff then returned the case to the fellow to finalize and sign out the report. After the preliminary reports were released in SoftPathDx, they were visible in the electronic medical record (Epic) until release of a final report. The preliminary reports were only visible to the patient care team via Epic, and the report was not sent to the patient portal until the final report was issued by a surgical pathology attending physician (which is standard for all preliminary pathology reports released at our institution).

Safety measures (Figure 2) included a comment in bold and all-capital case at the top of each report that stated, “This is a preliminary report generated by a surgical pathology fellow. This will be converted to a final report upon review by a pathology attending.” The goal was to have an attending pathologist review the case and convert the preliminary report to a final report within 2 business days. Therefore, fellows were instructed not to sign out any preliminary reports if the patient had an appointment within 2 business days of slide review. The patient appointment date is routinely added to the paperwork of all cases by secretarial staff at case check-in, so this information was available to the fellows at a glance. Additionally, fellows were given an 8-case total limit for pending preliminary reports so as to not outpace the attending pathologist’s ability to review those cases in the desired time frame. Fellows did not release any preliminary reports on the last day of their rotation so that they would be present to help finish all cases through to final report release.

The cases were assigned to a general surgical pathology attending physician, who was able to edit the report as needed before converting it to a final report. The final report replaced the previously released preliminary report in Epic, and both versions of the report (preliminary and final) were permanently stored in SoftPathDx. All final reports included a comment at the end: “This final pathology report replaces a previously released preliminary report. All clinical management decisions should be based on this final report.” Any changes deemed potentially clinically significant and/or major pathologic diagnostic changes from the preliminary to the final report (as judged by the attending pathologist) were verbally or electronically communicated to the clinical team by the fellow or attending pathologist, and the notification was documented in the LIS. If the attending pathologist deemed that more workup was required and/or disagreed with a preliminary report that could not be immediately finalized, the preliminary report could be amended to the following: “Upon review by the pathology consultant, additional ancillary testing is deemed necessary before a final report can be issued. A final pathology report will be issued upon completion of this additional workup.”

The fellows were instructed to keep a record of any cases in which the preliminary report needed to be amended in this manner, or in which the consultant disagreed with their preliminary diagnosis such that the clinical team needed to be notified, and submit this information to the rotation director at the end of the pilot study. This was not considered too onerous for the fellows because it was a very rare event, and self-reporting was double-checked by the program director after the pilot was completed. The fellows routinely keep track of diagnostic disagreements with outside diagnoses as part of the quality assurance component of this rotation, and so this process fits well with existing workflows.

After the pilot study, all preliminary reports were reviewed by the program director and compared with the corresponding final report to manually assess for any diagnostic disagreements. SoftPathDx was used to generate data regarding amendments, turnaround time, and case volume. Participating fellows, attending pathologists, and secretarial staff were surveyed via email for feedback regarding their experience with the process. The fellow survey included the following questions:

1. What benefits did you perceive to your learning?

2. What did you learn about yourself as a pathologist by doing this exercise?

3. Did you identify any gaps in your procedural or pathology knowledge, and did this exercise help you fill those gaps?

4. Did you identify any benefits to workflow and/or patient care using the new procedure as opposed to the “standard” workflow?

5. What areas did you identify that we could improve to make the experience better?

6. Do you have any additional comments?

Attending pathologists and secretarial staff were asked via email to provide thoughts and suggestions about their experience with the pilot in an open-ended fashion.

After the data from the pilot period were analyzed and deemed favorable, a permanent process was devised and implemented. Surgical pathology fellows board certified in anatomic pathology were eligible for preliminary report sign-out after their first month on general surgical pathology (frozen section rotation) was completed for performance evaluation. Preliminary reports could be generated during the fellow’s rotation on the outside slide confirmatory review service, as described above (4-week rotation). All rotation performance evaluations to date were reviewed by the clinical competency committee for approval of preliminary report sign-out privileges before the fellow started the outside slide review rotation. During the first week on service, an evaluation form was included with all outside cases reviewed by the fellow to assess the accuracy of the fellow’s diagnosis, the fellow’s report, and the fellow’s overall case management skills. The evaluation form included the following questions:

1. Was the fellow’s diagnosis fundamentally correct (ie, no major diagnostic errors)?

2. Was the fellow’s report accurate, complete, and without typographical errors?

3. Do you have concerns about the fellow being allowed to release preliminary reports?

4. Would you like to discuss any concerns with the rotation director?

After the first week on the rotation, the evaluation forms were reviewed by the rotation director, who decided to approve or deny preliminary report privileges for week 2, with input from the program directors if needed. If denied for week 2, the evaluation process continued and forms were reevaluated weekly by the rotation director until the end of the 4-week rotation. Once approved, the fellow could independently manage cases as described in the pilot study above. The evaluation forms were included in all cases with a preliminary report released to monitor for any concerns or issues, and the evaluations were reviewed by the rotation director weekly.

For the first 12 months, all preliminary reports were reviewed by the program director and compared with the corresponding final report to manually assess for any disagreements. SoftPathDx was used to generate data regarding turnaround time and case volume.

Participating fellows (n = 4) released 59 preliminary reports out of 101 total cases reviewed (58%), and elected to show the remaining cases to an attending pathologist without releasing a preliminary report. The number of preliminary reports by each fellow was variable: 14 to 26 cases, which represented 46% to 79% of cases they reviewed during their pilot week. The released preliminary reports included 19 gastrointestinal, 9 breast, 7 genitourinary, 5 gynecologic, 5 pulmonary, 5 head and neck, 4 cytopathology, and 3 bone/soft tissue cases, and 1 case each for endocrine and dermatopathology (Figure 3). The fellows shared cases with a subspecialty pathologist before releasing a preliminary report in 20 of 59 preliminary cases (34%). This compares with a share rate of approximately 20% for general surgical pathology attending physicians on this service (based on practice data generated from the LIS). The distribution of cases shown to a subspecialist was roughly similar to overall subspecialty case distribution (based on practice data generated from the LIS), and included 6 gastrointestinal, 4 breast, 3 gynecologic, 3 genitourinary, and 2 pulmonary cases, and 1 case each for endocrine and cytopathology. In 5 of 59 cases (8%), the attending pathologist chose to show a subspecialty pathologist after the preliminary was released by the fellow. Fifty-five of 59 preliminary reports (93%) were finalized within the goal of less than 2 business days. The median time from accessioning to preliminary report was 0.2 day (range, 0.1–4.0 days). For cases with a preliminary report, the median time from accessioning to final report was 1.0 day (range, 0.1–10.1 days), compared with a median of 1.1 days (range, 0.1–25.9 days) for cases where no preliminary report was released.

There was 1 case with a difference in diagnosis between the preliminary and final report that was deemed potentially significant, but this discrepancy did not adversely impact patient care. This was a parotid core biopsy with a preliminary diagnosis of oncocytoma, which was shown to a subspecialist head and neck pathologist after preliminary report release. The diagnosis was changed to low-grade oncocytic neoplasm with a broader differential diagnosis that included benign and malignant entities. The change in diagnosis occurred before the patient’s appointment, and thus had no impact on patient care.

The pilot study was endorsed by all 4 fellows (100%) as a positive experience, greatly increasing their independence and responsibility. Specific comments regarding the experience by fellows and attending pathologists are summarized in Figure 4. One fellow stated, “This was the truest form of autonomy I have experienced in all of my residency/fellowship [within the anatomic pathology realm] thus far!” Faculty and allied health staff feedback was also very positive. Based on these pilot data, the process was endorsed by practice leadership and approval was given to make this a permanent process.

During the first year of adoption of fellow preliminary report release as a permanent process, eligible fellows (n = 8) released 123 preliminary reports out of 1260 total cases reviewed (9.8%). The preliminary reports included 63 gastrointestinal, 25 breast, 14 genitourinary, 9 gynecologic, 3 bone/soft tissue, 2 endocrine, 2 pulmonary, 2 dermatopathology, 2 head and neck, and 1 cytology (Figure 3). Upon detailed review of preliminary reports by the program director, there were no major discrepancies between the preliminary reports and the final reports released by attending pathologists. There were no adverse effects on overall turnaround time (ie, time of accessioning to time of final report release): the mean and median number of days from accessioning to final report release for cases with a preliminary report were 2.7 and 2.0 days (range, 0.2–11.9 days), compared with a mean of 4.0 and median of 3.0 days for cases without a preliminary report (range, 0.1–51.3 days). Seven of 123 cases (5.7%) were shared with a subspecialty pathologist before preliminary report release, and 44 of 123 cases (35.8%) were shared after preliminary report release and before the final report was completed. The number of reports released by each fellow was quite variable (2–48 per rotation), with a mean of 15.4 and a median of 8 per rotation.

Processes that lead to a meaningful increase in independence for anatomic pathology trainees can be difficult to implement. Although sign-out of final reports by trainees would greatly facilitate transition to independent practice, federal Medicare/Medicaid billing requirements do not allow for trainees in ACGME accredited pathology programs to bill for final pathology reports.^8,11  Training programs can choose to bring their fellows on as junior attending staff and sign out final pathology reports, but this requires the programs to forego ACGME accreditation.⁸  Hiring and credentialing fellows as junior faculty for a 1-year fellowship may not be attractive to many medical centers, as it requires significant time and effort for a temporary position, especially if any visa issues come into play for foreign medical graduates. The overall environment of medical education has also been moving swiftly toward one of increased supervision, leaving many attending pathologists uncomfortable giving their trainees the amount of autonomy that would have been routine 20 years ago.⁷  Perception of autonomy by trainees and the amount of decision-making ability given to trainees may vary based on the experience level of the medical educator, with more experienced educators generally being willing to give trainees more autonomy.⁶  With time, this may create an epidemic of “helicopter attendings” with a cohort of trainees who consequently suffer from lack of autonomy and independence. This could lead to many undesirable effects when trainees enter independent pathology practice, including increased anxiety, prolonged turnaround time, low case volume, ordering of unnecessary ancillary testing, and an overall feeling of being overwhelmed. Therefore, meaningful increased independence must be implemented in pathology training programs, and creative solutions are necessary.

A recent change in the LIS system at our institution allowed for the release of preliminary pathology reports into the electronic medical record. This was quickly implemented in several parts of our practice, including preliminary frozen section diagnosis and preliminary cytology smear interpretation. As this has become a routine part of our practice, we hypothesized that preliminary report release by trainees might be a meaningful way to provide progressive responsibility and graded supervision. The final report could then be released and subsequently billed by a staff attending pathologist, thus falling in line with current Medicare/Medicaid billing and ACGME requirements.^8,11  In our experience detailed above, preliminary report release is an effective way to increase surgical pathology fellow independence in a safe learning environment, and we have adopted this as our standard practice for a subset of routine pathology cases. As we gain more experience with this process, we hope to expand it to other training programs and other types of specimens in the near future. It is important for any program desiring to implement a similar process to carefully consider the types of cases in which fellows are allowed to release preliminary reports, and what implication that might have for the clinical teams and patients. We felt cases sent for confirmatory review were an appropriate case type for preliminary report release, because any discrepancies could be resolved before the patient’s appointment. Cytology smear interpretation and surgical pathology specimens are also good candidates for these types of preliminary reports, because they provide potentially valuable interim information, but clinical services usually do not have follow-up appointments with the patients until the final pathology report is released. Biopsies for primary diagnosis of malignancy may be the most high-stakes type of specimens routinely encountered in pathology. Any discrepancies or errors in biopsy interpretation released in a preliminary report could have significant ramifications for the clinical team and the patient. Thus, implementation of preliminary reports on biopsy services should be undertaken only after careful consideration, and optimally should be supported by clear understanding of the process by the clinical team with their endorsement of the process.

Interestingly, the number of reports released by each fellow per rotation was quite variable (ranging from 2 to 48). We believe this is likely a reflection of several external factors, including the number of trainees on service and case volume fluctuations. However, it is likely that trainee-specific factors may also come into play, including confidence, efficiency, anxiety, risk aversion, and prior experience (ie, prior pathology subspecialty fellowship training, case volume of prior training programs). The aforementioned environment of increasing supervision and decreased autonomy may already be having effects on some trainees’ willingness to embrace independence when it is offered to them. Therefore, encouragement by program leadership may be necessary, especially for fellows who are tentative about taking on the responsibility of releasing preliminary reports. We have recently implemented a specific and clear expectation regarding the minimum number of preliminary reports during the rotation (15-case minimum) to help encourage tentative trainees to take the leap.

Our data indicate that fellows show good judgment when independently managing cases for preliminary report sign-out, and that this can be done in a safe fashion when appropriate guardrails are put in place. We observed just 1 potentially significant discrepancy between preliminary and final report out of 182 cases (<1%), and this did not lead to any adverse consequences on patient care. This has allowed our fellows to practice the process of final report editing and sign-out, and has also allowed them to use critical thinking skills when deciding how to manage a case. The choice to release a report, order more testing, or show the case to a colleague is a decision made by practicing pathologists dozens of times per day; however, many pathology trainees do not get enough practice with independently making this fundamental decision before the completion of training. We believe any process that successfully increases the trainee’s ability to make this decision will ease transition to independent practice.

We were able to implement this practice without any detrimental effects on turnaround time. In fact, cases without preliminary reports had a longer turnaround time, likely in part because of the more complex and difficult nature of those cases, and the need to call for additional material (tissue blocks, etc) in a subset. Our fellows were directed not to release preliminary reports on cases where they anticipated the need for additional material, or where they had any disagreement with the outside diagnosis, so the cases with preliminary reports were biased toward simpler cases. Fellows are also likely to naturally tend to release preliminary reports on easy cases they feel comfortable with, especially when first starting out with the process. The subspecialties represented by preliminary reports released by the fellows were very similar to the overall case mix in our outside review area, with a possible slight overrepresentation of gastrointestinal pathology in the preliminary reports released: 45% of preliminary reports were gastrointestinal pathology cases, which represent 33% of all cases in the area (based on practice data generated from the LIS). This could indicate an increased comfort level of the fellows with gastrointestinal pathology cases.

There were some notable differences between our pilot study data and the data from the first 12 months of implementation. The longer average turnaround time during the first 12 months likely reflects higher case volume during this time, as the pilot study took place during May and June 2020 while COVID-19 pandemic–related volume decreases were still widely apparent. This may also reflect the fact that the fellows in the pilot were more experienced: they were in the last 2 months of their surgical pathology fellowship and had already completed one full block on the outside slide review rotation. The differences in case-sharing rates are more surprising, and seem to be more dependent on individual fellows. The fellows who participated in the pilot seemed more likely to show a case to a subspecialist before sign-out, whereas the fellows during the first year seemed more comfortable releasing the report themselves. This could reflect the more routine nature of the process after the pilot period. It is interesting that the rate of attending pathologists asking a fellow to show a case to a subspecialist before finalizing the report is more than one-third, but does not routinely lead to significant change in the diagnosis. This is likely because general surgical pathology attending physicians at our institution all have a specific subspecialty expertise, and they may not be as comfortable with all case types compared with general surgical pathology fellows still in training.

This study has several limitations. The ability of the fellows to self-select cases likely led to selection bias, as mentioned above, and therefore the difficulty and complexity of cases selected for preliminary report sign-out may not be entirely reflective of cases encountered in daily practice. Although the LIS is a useful means to track data about preliminary report use, including amended reports, this is practical only after the rotation is complete and is not helpful to monitor fellow performance in real time. Relying on fellows to self-report amendments is not ideal; the case evaluations completed by the signing pathologist on all cases with a preliminary report are reviewed every week by the rotation director to monitor for issues, but an automated system would be more desirable.

In conclusion, we have devised and implemented a system of preliminary report sign-out for pathology trainees that we believe is safe and effective at providing a meaningful increase in autonomy. We hope to expand this process at our institution to aid in the training of the next generation of pathologists.