Figure 2: Box and whisker plot comparison of median FICO scores between trainees and attendings across geographical regions. Regions are defined by the United States Census Bureau (Appendix Table 1). No adjustments were made for multiplicity. Three applicants from each category were excluded due to each applicant being outside of the above regions. Data are analyzed by Mann-Whitney tests. (n for each group can be found in Appendix Table 2, **p < 0·01, ***p< 0·001, ****p < 0·0001).

This FICO score analysis of 3,843 trainees and attendings who applied for a loan is the first of its kind and underscores the statistically significant difference between the credit profiles of both groups. With the average debt of medical students being $200,000 not including the additional ~$50,000 from debt acquired from medical school, it is understandable that trainees rely on a more easily accessible form of revolving credit card debt [1,2]. Obtaining a credit card versus personal debt is easier especially with the understanding that salaries will rise five to ten-fold upon completion of residency. In line with the data, trainees in the cohort tended to get loans to consolidate credit card debt more frequently than attendings. Credit card debt has a more harmful impact on the credit score and borrowing power as it is revolving debt, tends to have higher interest rates than personal fixed debt, and the interest compounds unlike fixed debt. Though both impact the calculation of the credit score, credit reporting agencies view revolving debt as a barometer for borrower risk and utilize credit utilization as a determinant for the FICO score with a high utilization weighing negatively. As trainees graduate, less funding is needed for relocation as some contracts may include relocation stipends and sign-on bonuses. Often, funds are needed later in a career for family planning, buying into a practice, or advancing education. Thus, as physicians attempt to expand their careers and skills, they will often encounter paying off student debt and potentially add practice debt [7].

In our dataset, surgical attendings possessed a lower FICO score than medical attendings, although the difference was not statistically significant and based on a small group of surgical attending applicants. Surprisingly, studies demonstrated that 32% of academic surgeons would not recommend the same choice to their children stating the educational debt affected their productivity, quality of life, and career choices [6,19]. A recent survey of 2,212 students from 91 countries found that 69% of medical students were concerned about their income as an attending [20,21]. Alarmingly, 25% of US medical students were considering quitting training [20,21].

One can only assume that debt plays a significant factor as can be seen one year after New York University (NYU) made tuition free for medical school [22]. Applications to NYU medical school rose 47% with nearly a 50% jump in underrepresented minorities who often come from financially disadvantaged backgrounds [22]. The trend of medical schools providing more financial assistance may alter the debt acquired by trainees. With more financial education, trainees may opt for fixed versus revolving debt, thus improving their overall FICO scores and access to lower interest rates, decreasing the long-term total debt.

The significant increase in FICO scores after one year in practice can be theorized to be either due to paying down revolving credit card debt, obtaining higher credit limits, or both thus lowering the credit utilization ratio which factors into the FICO score. Trainees often have more payment delinquencies that also create drag on their FICO scores. We hypothesize that the increased income of attendings allows for more diligent on-time debt repayment habits that may also contribute to the FICO increase for attendings after one year in practice. Though a cross-sectional snapshot in time and not prospective, the FICO scores out to 30 years in practice appeared to remain rather stable in our dataset. Furthermore, with this FICO analysis up to 30+ years, the overall median score difference of 53 points for physicians did cross FICO score categories after graduation (from Fair to Good). Interestingly, the median FICO score of attendings is six points lower than that of the average American (718) despite the attending salary being nearly five-fold greater than the average American citizen suggesting that a closer look into this phenomenon may be warranted [23]. Furthermore, the average age of USA trainees is roughly 30 yet their credit score is 59 points lower than the average American [23]. Though some medical residents may delay major life decisions such as marriage, family, and buying a house until completion of training, many do not. Therefore, they may have similar financial obligations to their non-physician peers all while having a much lower FICO score and decreased access.

Family medicine physicians typically receive lower salaries due to billing based on clinic visits which often carry a lower reimbursement. This potentially translates into a decreased ability to pay down debt, revolving and fixed, affecting their credit profile as can be seen by having a lower median FICO score (702) than peers in other specialties [21,24,25].

This study has several limitations. First, in compliance with the ECOA, post-hoc data analysis was limited and excluded race, color, religion, national origin, sex, marital status, and age [26]. Comparisons were only made between trainee and attending status, geographical status, specialties, and years in training. Potential further studies may consider gaps during training as a surrogate for age after application approval and acceptance to maintain compliance with ECOA. Only voluntary, non-obligatory reporting of the aforementioned categories could be surveyed. Second, given the retrospective nature of the data, limited conclusions and associations can be made. Third, the number of surgical applicants was limited in both the trainee and attending groups.  In addition, the number of physicians in various subspecialties was limited, decreasing the applicability to the entire population of physicians. Furthermore, the data consist of approved physician-borrowers applying for a personal loan potentially skewing the median FICO score. Finally, no data were available on the socioeconomic support system of the applicants and may represent a biased analysis such that those without alternative financial reserves were the ones applying. Future, prospective, longitudinal studies, while remaining compliant with ECOA, may help shed light on the appropriate representation of trainees and attendings in terms of their FICO scores and comparisons between and within groups.

Among trainees and attendings, there was a statistically significant difference between the median FICO scores. The difference became statistically significant after one year as an attending. Between both groups, credit card consolidation was the number one reason for a borrowing application. The overall median FICO score of attendings is six points lower than the average American despite the significant difference in annual salary between attendings and the American citizen. The debt accrued may be one factor contributing to the trend of U.S. physicians leaving medicine though physicians’ credit profiles remain rather stable after graduation suggesting that physicians are reliable borrowers [27].

Key Summary Points:

• S. physicians incur a substantial and growing educational debt burden that may affect financial wellness, access to credit, and long‑term workforce retention, yet objective data describing physician credit profiles are lacking. 
• Credit scores are a widely used proxy for financial health and borrowing access but had not previously been systematically examined among U.S. physician trainees and attendings.
• This study asked whether physician trainees and attendings differ in credit profile, as measured by FICO scores, across training status, specialty, geography, and years in practice.
• Among 3,843 physician borrowers, attendings had a significantly higher median FICO score than trainees (712 vs 659), with scores rising notably within the first year of independent practice and remaining relatively stable over subsequent decades.
• Despite high educational debt and delayed earning potential, physicians demonstrated overall good credit profiles and reliability as borrowers, suggesting opportunities for earlier financial interventions during training to improve long‑term financial well‑being.

Declaration of Interests
Funding: Funding was provided by Doc2Doc Lending.

Conflicts of Interest: CR, PP, WS, JY are consultants for Doc2Doc Lending. CLR receives consulting fees from GLG, Augmend Health, Celeri Health, Shiratronics, and is a limited partner in OneSixEight Ventures. CW reports receiving (in the last 36 months) consulting fees unrelated to this work from Alosa Health, Analysis Group, Atheneum, Berkshire Hathaway Home Companies, Chronius, FVC Health, GLG, Guidepoint, NuvoAir, Ogilvy, Philips, Simbo, Substack, Tell Health, The New York Times, and The Wall Street Journal, income unrelated to this book from book rights to Doubleday Books, and speaking fees from the Harry Walker Agency. CW is supported by a career development grant from the Agency for Healthcare Research and Quality (grant # K08HS029197). ABJ reported receiving personal fees from Analysis Group, Freakonomics MD, The Wall Street Journal, The New York Times, and The Los Angeles Times; speaking fees from the AAE Speakers Bureau and Harry Walker Agency; and royalties from Doubleday Books outside the submitted work.

Author Contributions: PP, RJY, WS, and CLR devised, analyzed, and wrote the paper.  JJH, TF, CW, ABJ assisted with writing, revising, and providing expertise.

Ethics/Ethical Approval: Ethics committee approval was not required for this study as it did not involve patients and all data was de-identified.

Data Sharing: All de-identified data will be available upon publication and request from Doc2Doc Lending with data not to be reproduced or published without the authorized permission of Doc2Doc Lending and to be used for only academic purposes. A signed data access agreement must be signed prior to sharing of data. Requests should be sent to crobi151@jh.edu and will be considered on a case-by-case basis.