A National Survey of Contracts, Compensation, and Workplace Culture Among Junior Surgical Oncologists

Abstract

Background:

The transition from surgical training to independent practice is challenging, and little quantitative data exist to guide junior surgeons on contracts and compensation. Qualitative evidence suggests compensation model type may influence workplace dynamics and attrition, but this has not been formally examined.

Methods:

A cross-sectional survey of Society of Surgical Oncology members in their first 8 years of practice was conducted October-November 2025 (n=221; response rate = 20.2%). Outcomes included contract and compensation characteristics, attrition risk, and workplace collegiality.

Results:

Respondents were 53% female, 68% non-Hispanic White, and 65% academic; 40% were within their first 2 years of practice. Most (69%) attempted contract negotiation and 79% succeeded in at least one domain. Nearly all (79%) had a guaranteed salary period (median 3 years), after which most transitioned to wRVU-based compensation. One in four respondents (25%) were considering leaving within the year; this rose to nearly 80% among those who strongly disagreed their compensation was fair (p<0.01). Greater productivity emphasis was inversely associated with workplace collegiality (OR 1.8 per SD increase, 95% CI 1.1–3.0; p=0.02).

Conclusions:

This national survey provides contract and compensation data for junior surgical oncologists and demonstrates associations between perceived compensation fairness and attrition risk, and between productivity-based compensation emphasis and workplace hostility. These findings suggest that how surgeons perceive their compensation — not merely its structure — may be a target for institutional efforts to improve retention and culture.

Introduction

The transition from surgical trainee to independent practice is a challenging one. New graduates face immediate demands to navigate unfamiliar territory: the shift from the structured "match" process to independently choosing a position and weighing competing interests like disease site, setting, and geography. These details likely matter considerably. Many surgeons leave their first position at great personal and institutional expense, suggesting that initial job fit is critical¹. Replacing a surgeon is estimated to cost up to a million dollars due to lost revenues and the cost of recruitment². Compensation model, in particular, may influence not only surgeon retention but also workplace culture and the ability to pursue non-clinical activities such as education, service, and research³. Simultaneously, surgeons are entering healthcare during a turbulent period—physician payments are declining while costs are rising—placing intense pressure on institutions to prioritize productivity and efficiency⁴.

Resources to guide early-career surgeons through this transition are scarce. While compensation data from sources such as the Association of American Medical Colleges (AAMC), Medical Group Management Association (MGMA), and SullivanCotter can provide benchmarks for salary expectations or work relative value unit (wRVU) targets, these resources are often expensive or difficult to access. More critically, none provide actionable guidance on contract negotiation, the construction and implications of different compensation models, or how these structural choices translate into daily clinical practice and career satisfaction. Published literature on surgeon compensation models is sparse. Even with very broad inclusion criteria, a recent systematic review identified only a few dozen studies, most of which were reviews, editorials, or analyses of benchmarking data (e.g., MGMA)⁵. The few empirical survey studies were mostly outside of general surgery (e.g., orthopedics, neurosurgery) and focused primarily on compensation and RVU targets, with little to no granularity regarding other aspects of a contract^6-8.

To fill this critical resource gap, as members of the Fellows and Young Attendings Committee (FYAC) of the Society of Surgical Oncology (SSO), we aimed to (1) characterize the employment landscape for junior surgical oncologists, including practice setting, disease sites, and practice structure (e.g., FTE, weekly schedule); (2) document what is typical in contracts and compensation models; and (3) explore how compensation models may be associated with workplace culture and attrition.

Methods

Study Design & Ethics Review

This was a cross-sectional survey study of junior members in the SSO. The study was reviewed and approved by the University of Washington Institutional Review Board.

Instrument Development

The survey instrument was developed by the SSO’s FYAC. The group created an initial version with three primary domains: (1) respondent characteristics such as demographics, geography, and years in practice; (2) characteristics of the respondent's current institution and position; and (3) detailed subjective and objective descriptions of their compensation plan. Questions were refined through iterative feedback from FYAC members and subsequently reviewed and edited by the SSO executive committee. The instrument was then uploaded into Qualtrics and piloted among members within and outside the FYAC to verify survey logic, comprehensibility, and clarity. A copy of the full instrument is included as a supplemental file.

Survey Distribution and Participant Selection

The survey was distributed via email to SSO members on October 13, 2025 and closed on November 11^th, 2025. SSO members received periodic reminder emails during this collection period, and all partial responses were recorded as complete when the survey closed. Additional information regarding distribution procedures and response rate calculations is provided in the Supplement. The target population was domestic junior surgical oncologists within their first 8 years of practice, consistent with the SSO's existing definition of junior attendings. No specific exclusion criteria were applied.

Variables

Most variables were analyzed in their native distributions. Two exceptions were made: (1) One question asked respondents to rate the collegiality of their institutional environment on a 0 (hostile) to 100 (collegial) scale. Because the empirical distribution was heavily left-skewed, we dichotomized this variable at a cutoff of 50 for some analyses, with scores less than 50 classified as hostile. (2) Two survey items were used to assess perceived emphasis on productivity within compensation models. The first asked respondents to place their compensation model on a 0-10 continuum anchored by "salary" and "eat what you kill.” The second presented a standardized scenario involving a hypothetical 25% reduction in wRVU productivity and assessed perceived downstream effects on compensation using a six-category ordinal response scale (with additional “other” and “unsure” options). Although these items differed in format (continuous scale vs. scenario-based ordinal response), they were designed a priori to capture complementary dimensions of a shared latent construct reflecting the degree of compensation emphasis on productivity. Consistent with this conceptualization, the items were strongly correlated and demonstrated high internal consistency (Cronbach’s a = 0.88). We therefore constructed a standardized composite score by averaging z-standardized values of the two items. Combing these items improved measurement robustness by capturing both respondents explicit classification of their compensation model and their sensitivity to marginal productivity changes, thereby reducing reliance on any single framing of the construct.

Statistical Analysis

The primary analysis was descriptive, using proportions for categorical variables and medians with interquartile ranges for continuous variables. Bivariate analyses were used to compare variables using appropriate parametric and non-parametric tests. Multivariate analyses were used to examine associations between respondent characteristics and our primary outcomes of workplace collegiality and attrition likelihood. We first started with basic demographic factors (e.g., sex, years in practice) and then added compensation-related variables (productivity emphasis and perceived fairness). We used complete-case analysis throughout with no attempt to impute missing data. Statistical tests were two-sided with an alpha level of 0.05 for determining statistical significance.

Results

Sample Size, Participant, and Practice Characteristics

Between October 13th and November 11th, 2025, there were 221 partial responses to the survey for an approximate response rate of 20.2% (see Supplement for additional details regarding response rate calculation). Respondents came from 37 states, with the majority coming from the South and Northeast (59.5% of respondents). A handful of respondents (n=10, 6.1%) resided outside of the United States. The sample was slightly more female (53%) than male (47%), and predominantly non-Hispanic white (68%) (Table 1). The sample was relatively evenly distributed across graduation year, with 40% of respondents within their first 2 years of practice, another 31% 3–5 years out, and the remainder (29%) more than 5 years out. Approximately one in five (18%) respondents had already switched jobs, and this rate was 27% among those 3 or more years out of practice compared to 1% in those 2 or fewer years (p<0.001).

Table 1:

Survey Sample Characteristics

Characteristic		Sample (n)	Percent
Sex	Female	92	52.6%
	Male	82	46.9%
	Prefer not to say	1	0.6%
Race	White/Caucasian	125	71.8%
	Asian	28	16.1%
	Black/African American	8	4.6%
	Multiple/Other	9	5.2%
	Prefer not to say	4	2.3%
Hispanic		12	6.9%
Year Graduated	2023-2025	67	40.1%
	2020-2022	52	31.1%
	<2020	48	28.7%
Switched Jobs Since Starting as Faculty		31	18%
Institution Type	Academic Medical Center	143	65.0%
	Community - University Affiliated	39	17.7%
	Community - Non University Affiliated	17	7.7%
	Free Standing Cancer Center	10	4.5%
	Private Practice, Employed	1	0.5%
	Private Practice, Physician Owned	1	0.5%
	Kaiser	4	1.8%
	VA	2	0.9%
	Other	3	1.4%
Geographic Region	Northeast	41	25.2%
	Midwest	31	19.0%
	South	56	34.4%
	West	25	15.3%
	Outside United States	10	6.1%
Practice	100% Breast	62	28.1%
	Predominant (>=50%) HPB	44	19.9%
	Predominant (>=50%) Mel/Sarc	32	14.5%
	Mixed practice	56	25.3%
Schedule	Days in clinic a week	1.55	[1-2]
	Days in OR a week	2	[1.9-2.35]

HPB = Hepatopancreatobiliary; Mel/Sarc = Melanoma/Sarcoma; VA = Veterans Affairs

Sample sizes vary from n=163 to n=220 depending on question; demographic characteristics (e.g., sex, race/ethnicity, region) had lower counts due to late placement in the survey instrument.

Nearly two-thirds (65%) reported working at an academic medical center, with community practice (university- or non-university-affiliated, 25.5%) representing the other dominant setting. The survey did not ask about complex general surgical oncology versus breast surgical oncology fellowship, although 28% of respondents reported a 100% breast practice. The remainder had eclectic clinical practices. One in five respondents (20%) reported a practice that was >50% hepatopancreatobiliary (HPB); 15% reported a practice that was >50% melanoma or sarcoma, but more commonly, 25% of respondents reported a mixed practice with no dominant disease site. Median operating room days per week were 2 and median clinic days were 1.6.

In general, respondents had positions that were well matched to what they had desired (Figure 1). The majority reported that their current position was "somewhat" or "exactly" what they wanted with respect to disease site (85%), setting (82%), practice mix (81%), and location (69%), Location had slightly higher rates of dissatisfaction compared to the other domains, with 24% of respondents reporting that their position was "somewhat" or "very" different from what they had desired.

Figure 1: Alignment Between Current and Desired Career Characteristics Among Junior Surgical Oncologists.

Sample size = 214 (academic setting, practice mix, location) and 215 for disease site

Contract Negotiation, Details, and Compensation Model

Most (69%) respondents attempted to negotiate part of their contract, with no statistical difference between males and females (71% vs. 66%, p=0.28). In descending order, the most common aspects of a contract that respondents attempted to negotiate were salary, signing bonus, and startup funds (Table 2). Among those who attempted to negotiate, 79% were successful in at least one domain. There were differential rates of success across domains; for example, less than half (43%, 42/98) of respondents who attempted to negotiate their salary were successful, whereas nearly three-quarters (72%, 51/71) of those who attempted to negotiate their signing bonus were successful.

Table 2:

Negotiation & Contract Basics

Negotiation
Attempted to negotiate any portion of contract (n, %)	144	(69%)
Component (n)	Attempted to Negotiate	Successfully negotiated
Salary	98	42
Signing Bonus	71	51
Start Up Funds	42	27
FTE	37	21
Block	24	7
Clinical Support	22	7
Other	21	13
Call	14	9
Contract Basics
Clinical FTE	Median [IQR]	0.8	[0.6-0.9]
	1.0 FTE (n, %)	30	16%
	<=0.5 FTE (n, %)	45	24%
Had guarantee period (n, %)	144	79%
Median [IQR] period, years	3 [ 2-3]
After guarantee, had RVU target (n, %)	119	66%
CME	Median [IQR]	$5,000	[$3,500-6000]
	Range	$0-$18K+1
Call2	Sample (n)	< 1x/month	1x/month	2-3x/month	4+x/month
Disease site only	103	30	13	17	41
Service line	84	25	19	14	24
General surgery	58	9	8	28	11
Trauma	6		1	3	2
Critical Care	1				1
Other	12	2	2	3	5
Inpatient Staff & Trainees	Sample (n)	Percent3
Students	131	62%
Residents	181	85%
Fellows	78	37%
APP	129	61%
Outpatient Staff & Trainees
Medical assistant	191	90%
OR Scheduler	170	80%
Clinic Scheduler	167	78%
Nurse Navigation	151	71%
Students	92	43%
Residents	129	61%
Fellows	65	31%
APP	149	70%
Familiarity with compensation model (n, %)	Not familiar at all	11	6%
	Slightly familiar	26	13%
	Moderate familiar	61	31%
	Very familiar	63	32%
	Extremely familiar	38	19%

¹A handful of participants reported unlimited CME funds

²Proportions are not provided as we do not know whether or not a lack of response indicated the respondent skipped the question or that they took no call.

³Proportion = sample (n) / sample that reported any inpatient or outpatient support (n=213); we assume that respondents who did not check any boxes did not answer the question; as a result this proportion may be overestimated if respondents truly had no support although this anticipated to be a small sample

Most (79%) respondents reported having a guaranteed period, with a median (interquartile range [IQR]) of 3 (2–3) years. The median (IQR) clinical full-time equivalent (cFTE) across the entire sample was 0.8 (0.6–0.9). Less than one in four (24%) had a clinical FTE of 0.5 or less. Median cFTE was 0.7 for those at an academic medical center, 0.8 at a community university-affiliated hospital, and 0.9 at a community non-university-affiliated hospital. Slightly over half (54%) of respondents considered research a significant part of their current practice. Median clinical and research FTE was 0.7 and 0.2, respectively, among those who considered research a significant part of their practice, compared to 0.8 and 0, respectively, for those who did not. Call responsibilities were heterogeneous. Most commonly, respondents took call for their disease site (e.g., HPB) and/or their service line (e.g., surgical oncology). Among those taking call, 29% took general surgery call, most commonly 2 to 3 times per month.

Most respondents had significant support (staff and trainees) on the outpatient and inpatient side. On the outpatient side, most respondents had access to medical assistants (90%), schedulers (OR – 80%, clinic – 78%), nurse navigation (71%), and/or a nurse practitioner (NP) or physician assistant (PA) (70%). On the inpatient side, over half had an NP or PA (61%) and 85% had residents. Median (IQR) annual continuing medical education (CME) funds were $5,000 ($3,500–$6,000), although the range extended from $0 to $18,000, with a handful of participants reporting unlimited CME funds.

Respondents reported relatively high familiarity with their compensation model, with 81% reporting at least moderate familiarity and over half (51%) reporting they were very or extremely familiar. Approximately two-thirds (66%) of respondents reported an RVU target after their guarantee period. A narrative question asking respondents to describe their RVU target received heterogeneous responses. A minority (35%) could be interpreted as indexed to a 1.0 FTE, but among those, median wRVU targets were 6,770 (IQR 6,000–7,667). Another 37% provided a target for which it was unclear whether it was adjusted to a 1.0 FTE; among these, the median wRVU target was 5,916 (IQR 4,925–6,470). Approximately 20% did not answer the narrative question or were unsure, and the remainder (7%) provided a description (e.g., 65% of MGMA) but no numerical value. Perceived compensation fairness was mixed: just over half (52%) somewhat or strongly agreed their compensation model was fair, 20% were neutral, and 28% disagreed.

Perceived emphasis on productivity within compensation models varied across measurement approaches. When respondents were asked to explicitly locate their compensation model on a continuum from salary to “eat what you kill,” responses spanned the full range of the scale, with the modal response favoring a salary-based model (n=38, 22%) (Figure 3a). In contrast, responses to a standardized vignette assessing the impact of a hypothetical 25% reduction in wRVUs on compensation were more concentrated, with the most common response indicating no change in compensation (n=59, 34%), while 1–25% reductions were also commonly reported; 22% of respondents selected “unsure” (Figure 3b). Despite differences in distributional shape across items, both measures were strongly correlated and demonstrated high internal consistency, supporting their aggregation into a standardized composite score (Figure 3c).

Figure 3: Survey Items Assessing Compensation Models Financial Emphasis on Productivity.

3a – Scale of productivity emphasis ranging from 0 (pure salary) to 10 (collections based / eat what you kill)

3b – Perceived impact of reducing wRVUs by 25% on compensation

3c – Standardized scale combing both measures

Attrition, Hostility, and Compensation Models

One in four (25%) respondents were considering switching jobs in the next year. We found no statistical association between respondents' sex, race, or institution type (e.g., academic medical center) and consideration of switching jobs. There was an association with years in practice, with those 3 or more years into practice more likely to be considering a switch compared to those 2 or fewer years (33% vs. 10%, p<0.01). Perceived fairness of compensation was associated with consideration of switching jobs (Figure 2), with the association persisting after controlling for years in practice, workplace collegiality, and compensation emphasis on productivity (p<0.01).

Figure 2: Association Between Perception of Compensation Model and Consideration of Switching Jobs in the Next 12 Months.

In general, respondents placed their workplace close to the collegial end of the 0 (hostile) to 100 (collegial) scale, with a mean and median of 81 and 90, respectively. Only 10% of respondents placed their institution closer to the hostile end (<50). There was a negative correlation (p<0.01) between productivity emphasis of the compensation model and perceived collegiality (Figure 4). As a dichotomous variable, a one standard deviation increase in productivity emphasis was associated with a 1.8-fold increase in the odds of a hostile work environment (95% CI 1.1–3.0, p=0.02). There was no association between respondents' sex, race, graduation year, institution type, or productivity emphasis and report of a hostile work environment.

Figure 4: Association Between Emphasis on Productivity and Workplace Collegiality.

Discussion

This cross-sectional study of junior surgical oncologist in the SSO provides important insights into the characteristics of initial employment. The sample included respondents from around the country across diverse demographics, disease sites, and practice settings. Despite a notoriously difficult job market, this sample had success obtaining surgical oncology positions with desired characteristics. A typical position is largely clinical with limited protected time for research; compensation models were usually guaranteed for 3 years, after which most individuals transition to a wRVU target. However, dichotomizing practices into having or not having a target is a blunt measure, and respondents reported a wide range of clinical productivity emphasis. This emphasis may in turn impact workplace culture, with higher productivity emphasis associated with worsening collegiality and higher odds of a hostile workplace. Compensation models may also be associated with surgeon attrition, with this study showing an inverse relationship between perceived fairness of the compensation model and interest in changing jobs.

The negotiation and contract information provided in this article will be informative to trainees, training programs, and prospective employers. We are aware of no other national data source for surgeons that provides information about junior surgeon expectations as they relate to FTE, OR and clinic schedule, guarantee period, call, inpatient and outpatient support, and CME. These descriptive data suggest that respondents appear to be finding positions aligned with their interests, with adequate access to OR time, clinic, and support staff. These findings are consistent with recently published American College of Surgeons workplace standards, which recommend a minimum of two full OR days and one full clinic day per week for full-time clinical surgeons, with additional access as dictated by specialty or productivity requirements².

One area worth highlighting, however, is the clinical emphasis of these positions. Our sample was likely biased towards academia, yet very few respondents have more than 20% effort dedicated to research. Surgical oncology training, in particular, emphasizes research. Most fellowship applicants have completed 2 or more dedicated research years during their general surgery training, and many have completed additional gap years at various stages⁹. The fellowship itself also requires 4 months of research. The fact that most will then go on to largely clinical jobs calls into question this one-size-fits-all approach, which has significant financial implications for the trainee⁹. Recent work has shown that the overwhelming majority of surgical oncology fellows have limited academic productivity after fellowship with less than 10% obtaining federal funding and many having zero academic footprint after leaving training¹⁰. Future research should expand the population of this survey to include a broader range of experience levels and disciplines—not just within general surgery but across the house of surgery.

Perhaps the most intriguing results of this study are the quantitative associations we found between compensation model, surgeon risk of attrition, and workplace collegiality. A qualitative study published in 2024 interviewed surgeons at various academic institutions and showed that compensation model may influence patient decision-making and workplace culture³. Quantitative studies in orthopedics and otolaryngology have shown that compensation model may influence surgical clinical decision-making^11,12. To our knowledge, this is the first quantitative study to examine compensation model, surgeon attrition, and workplace collegiality, and the results are quite stark. Perceived fairness of compensation had a strong association with interest in changing jobs: nearly 80% of those who strongly disagreed that they had a fair compensation model were considering switching jobs in the next 12 months, even after controlling for the compensation model's clinical productivity emphasis. If true, this means that the compensation model itself is not necessarily the problem—rather, it is perceptions of the compensation model that matters. Division and department leaders may find this informative as they counsel and mentor their junior surgeons. Finally, the relationship we found between productivity emphasis and workplace hostility quantitatively validates qualitative findings previously identified in this space. Future work should validate our two-item scale for productivity emphasis, verify the association between compensation model and attrition and workplace hostility, and quantitatively assess the relationship between clinical productivity emphasis and surgical threshold for intervention.

This study has several limitations. First, it was a voluntary cross-sectional survey of junior members of a single professional society. As such, the findings are subject to nonresponse and selection bias. Physicians with particularly strong positive or negative experiences related to compensation or workplace environment may have been more likely to respond, and the modest response rate may limit generalizability. Second, all measures are self-reported and reflect perceptions rather than objective institutional data, which may be influenced by recall bias and individual interpretation. Third, while the survey instrument was developed through iterative expert review and pilot testing, most measures—including the composite measure of productivity emphasis—have not been externally validated and should be interpreted as exploratory. Fourth, the cross-sectional design precludes causal inference, and reverse causality may explain several of the observed associations. For example, individuals more likely to leave their practice for a variety of reasons may perceive their compensation model as less fair, rather than compensation structure itself driving attrition risk. Finally, despite a relatively short survey (~10 minutes), there was some late attrition, and the placement of demographic questions at the end of the instrument limited the ability to examine certain outcomes across respondent subgroups.

In conclusion, this national survey study of junior SSO members provides data on negotiation and contract details. It also demonstrates associations between perceived compensation fairness and surgeon interest in changing jobs, as well as between compensation model emphasis on productivity and workplace hostility. These data will inform trainees, junior surgeons, and their employers about national benchmarks and can guide future research on the impact of compensation model on surgical decision-making and workplace dynamics.

Synopsis:

In this national survey of 221 junior surgical oncologists, data on contracts and compensation are provided, and associations are reported between perceived compensation fairness and risk of attrition as well as compensation model and workplace collegiality.