Surgical Team Composition and Outcomes After Radical Prostatectomy for Prostate Cancer

Abstract

Introduction:

Physician assistants (PAs) and nurse practitioners (NPs) are increasingly integrated into surgical teams. Although their training varies, existing literature often considers PAs and NPs interchangeable when assessing their impact. The aim of this study was to measure the effects of PA vs NP team composition on surgical outcomes in men with prostate cancer.

Methods:

We performed a retrospective cohort study using a 20% national sample of traditional Medicare beneficiaries undergoing radical prostatectomy between January 2010 and December 2020. The cohort included patients treated by independent urology groups with at least 1 PA or NP. Groups were stratified into 3 mutually exclusive categories: PA only, NP only, and both PA and NP. Generalized estimating equations measured the association between provider composition and surgical outcomes (ie, postoperative complications, prolonged length of hospitalization, and readmission within 90 days), adjusting for patient-level and group-level variables.

Results:

Over the study period, 424 independent urology groups with at least 1 PA or NP managed 8975 men undergoing radical prostatectomy. Patient age, race, and comorbidity were similar across team compositions. After adjustment, there was no association between team composition and complications, prolonged length of hospitalization, or readmission (P > .05 for all outcomes). For example, complications occurred in 9.8%, 11.4% and 11.5% of men managed by the PA-only, NP-only, and both PA and NP groups, respectively.

Conclusion:

Among independent urology practices performing radical prostatectomy, outcomes did not differ between practices employing only PAs, only NPs, or both PAs and NPs. These findings suggest that employing PAs, NPs, or a combination is not associated with differences in postoperative outcomes.

Introduction

Rising demand for urologic care by an aging population is projected to outpace the supply of urologists in the United States.^1-3 As such, by 2035, the demand for urologists will fall short of the supply by 46%.⁴ In response, the American Urological Association has identified the urologic workforce shortage as a federal advocacy priority.⁵ Their 2024 census data demonstrate that more than 60% of US counties lack a practicing urologist and that the workforce is aging, with nearly 30% of urologists aged 65 years. One proposed solution has been the strategic integration of physician's assistants (PAs) and nurse practitioners (NPs) into urologic care models.

Prior work demonstrates that PAs and NPs can safely deliver high-quality urologic care across diverse care settings.^6-12 Despite increasing integration of these clinicians into care teams across many specialties, distinct training paradigms suggest the possibility that one may be better suited depending on the role of the position. PAs are trained in a disease-centered model across a broad range of populations, while NPs stem from a nursing background, with its holistic focus and emphasis on specific populations. Existing literature generally has treated PAs and NPs interchangeably, ignoring their training differences and its potential effects on outcomes.^8,9,12-17 Given the looming workforce shortage, urology groups will invariably rely on advance practice clinicians to meet growing clinical demands. Whether training differences between PAs and NPs are material to hiring decisions, however, is unknown.

To better understand this issue, we examined the association between care team composition (urology groups with PAs only vs NPs only vs both PAs and NPs) and surgical outcomes after radical prostatectomy for prostate cancer. We hypothesized that outcomes would be agnostic to care team composition, supporting the idea of hiring the most outstanding candidate for an open position, independent of training background. Addressing this knowledge gap is essential to inform evidence-based workforce expansion.¹⁸

Methods

Data and Study Population

We performed a retrospective cohort study of traditional Medicare beneficiaries undergoing radical prostatectomy between January 1, 2010, and December 31, 2020, using a 20% national sample of Medicare claims. Follow-up was available through December 31, 2021. Patients were required to be 66 years of age or older and continuously enrolled in Medicare Parts A and B for the 12-month period before surgery to allow for comorbidity measurement.¹⁹ Furthermore, patients were required to be entitled to Parts A and B for 90 days after surgery to assess surgical outcomes.

Independent urology groups were identified using the Medicare Data on Provider Practice and Specialty file, a calendar year‑level file that assigns each national provider identifier—encompassing physicians, PAs, and NPs—a specialty code and the Tax Identification Number (TIN), under which the provider bills Medicare. For each calendar year, national provider identifiers were aggregated to the TIN level, capturing all providers billing under a given TIN as either their primary or secondary billing affiliation. A TIN was designated an independent urology group if more than 50% of its billing providers in that year were urologists. PAs and NPs were considered part of a given urology group if they billed under that same TIN in that calendar year.

The rationale for including only men managed by independent urology groups is the lack of specific specialty codes designating focused PA and NP clinical care roles. In addition, PAs and NPs provide services that are either billed by a supervising physician or are considered nonbillable and therefore not identified directly by claims data. As such, it can be assumed that PAs and NPs that are solely employed by independent urology groups encompass roles specific to the care of patients of those urology groups, which contrasts with PAs and NPs employed in large multispecialty groups where the deployment of these clinicians is variable. Independent urology groups were classified by annual team composition—urologists with PAs only, urologists with NPs only, urologists with PAs and NPs. Team composition, measured at the level of the independent urology group-year, served as the main exposure of interest.

Surgical Outcomes

All outcomes were measured at the patient level and included complications during the index admission (ie, postoperative complications), prolonged hospitalization during the index admission (ie, prolonged length of stay), and readmission within 90 days of surgery.²⁰ Postoperative complications were characterized as a composite of the occurrence of 1 or more of the Agency for Healthcare Research and Quality's Patient Safety Indicators during the index hospitalization. Prolonged length of stay was defined as those men requiring hospitalization duration in the top 25th percentile in a given calendar year, allowing for secular changes in practice patterns.

Statistical Analysis

Patient and urology group characteristics were compared across team composition (PA only, NP only, both PAs and NPs) using chi-square tests for categorical variables and analysis of variance for continuous variables. Generalized estimating equations were used to measure the association between team composition and surgical outcomes. This approach accounts for clustering of patient outcomes within urology groups (ie, some groups may have better outcomes than others). Models were adjusted for patient (age, race [White, Black, Hispanic, and other], rural residence, comorbidity, and socioeconomic status) and urology group (number of urologists, average annual prostatectomy volume among urologists performing the procedure, and density of advanced practice clinicians) variables. Comorbidity was characterized using the Charlson Comorbidity Index, derived using claims from the 12-month period before surgery. Socioeconomic status was measured at the zip code level and categorized into tertiles. The density of advanced practice clinicians was estimated by ratio of PAs and NPs to urologists within a group in a calendar year. Finally, a fixed effect for treatment year was included to account for secular changes in outcomes. Adjusted percentages for each outcome were derived by urology group team composition.

All analyses were performed using SAS, and statistical significance was assessed at α = 0.05. This study was deemed not regulated by the institutional review board because it used deidentified claims. The requirement for informed consent was waived.

Results

Over the study period, we identified 8975 Medicare beneficiaries undergoing radical prostatectomy by 424 unique independent urology groups deploying at least 1 advanced practice clinician. Patient and group characteristics by team composition are summarized in Table. Patient age, race, and comorbidity were well balanced across groups by care team composition. However, NP-only groups were more likely to manage men in the lowest tertile for socioeconomic status (38.4% vs 34.6% and 32.5% in PA-only and both PA and NP groups, respectively, P < .01). In addition, men managed by urology groups with both PAs and NPs were larger in size and had a higher density of advance practice clinicians compared with PA-only and NP-only groups.

Table.

Characteristics of Patients Undergoing Prostatectomy by Independent Surgical Groups by Practice Composition and Surgical Specialty

Characteristics	Only PA	Only NP	Both NP and PA	P-value
Group years, N	707	563	759
Procedures, N	2496	1842	4637
Age, mean (SD)	70.1 (3.4)	69.9 (3.3)	70.0 (3.4)	.27
Non-White, N (%)	274 (11.0)	219 (11.9)	577 (12.4)	.19
Nonmetropolitan area, N (%)	566 (22.7)	448 (24.3)	868 (18.7)	< .01
Socioeconomic status, N lowest tertile (%)	864 (34.6)	708 (38.4)	1505 (32.5)	< .01
Charlson Comorbidity Score, N 3+ (%)	883 (35.4)	673 (36.5)	1664 (35.9)	.73
Group size, mean (SD)	14.6 (14.7)	16.5 (16.8)	29.7 (28.4)	< .01
Annual procedure volume per surgeon, mean (SD)	2.6 (2.0)	2.4 (1.3)	2.6 (1.6)	< .01
PA and NP density, mean (SD)	0.3 (0.3)	0.3 (0.3)	0.4 (0.3)	< .01

NP, nurse practitioner; PA, physician assistant.

As illustrated in Figure, team composition was not associated with postoperative complications, prolonged length of stay, and readmission after radical prostatectomy after adjusting for differences in patient and urology group characteristics. For example, postoperative complications occurred in 9.8% (95% CI 8.3, 11.4), 11.5% (95% CI 10.4, 12.8), and 11.4% (95% CI 9.9, 13.1) of men managed by the PA-only, NP-only, and both PA and NP groups, respectively.

Figure.

Outcomes by care team composition for prostatectomy, adjusted for age, sex, race (White, Black, Hispanic, and other), place of residence (metropolitan areas >1 million, metropolitan areas <1 million, nonmetropolitan areas >5000, and nonmetropolitan areas <5000), Charlson Comorbidity Score, socioeconomic status tertiles (low, medium, and high), year of surgery, surgical group size, surgical group prostatectomy volume, and surgical group PA and NP density. Adjusted outcomes were similar across care team composition for patients undergoing prostatectomy. NP, nurse practitioner; PA, physician assistant.

Discussion

In this national study of Medicare beneficiaries, urology care team composition, defined by group employment of PAs, NPs, or both, was not associated with postoperative complications, prolonged length of stay, or readmission after radical prostatectomy for prostate cancer. In this clinical context, training distinctions between advanced practice clinicians did not affect patient outcomes, lending support to hiring the best candidate for an open position as opposed to gravitating toward a specific discipline.

These findings add nuance to a growing body of literature surrounding the integration of PAs and NPs in surgical care. PAs and NPs differ in training pathways, licensing requirements, and clinical deployment,¹⁶ such that PAs have greater involvement in operative care and NPs are more frequently employed in ambulatory settings.^15-17 Widely cited commentary has examined the theoretical interchangeability of PAs and NPs based solely on training, legal frameworks, and functional roles, but these analyses have been largely conceptual.¹⁴ Our data lend empirical support to this supposition in the context of prostate cancer. Furthermore, prior work from the primary care literature has also demonstrated no meaningful differences in scope of practice or impact on outcomes between PAs or NPs.²¹ In that study, PAs and NPs were deployed similarly across inpatient clinical environments, with no systematic differences observed in patient case load, core clinical tasks, or functional autonomy. Likewise, there were no differences detected in overall care quality between PAs and NPs as reflected by patient experience metrics.

Findings from this study should be interpreted in the context of its limitations. First, PA and NP involvement was assessed at the urology group level rather than at the patient level. Specifically, we are not able to examine the specific roles played by each provider type in patient care. Unmeasured differences in supervision, professional experience, or institutional workflow are common in contemporary surgical practice and may influence care delivery, thus attenuating detectable differences in outcomes by team composition. More granular assessment of individual PA and NP roles and behaviors within the care team may reveal differences not captured in this analysis, highlighting an opportunity for future study. Nonetheless, in the context of this study, urology groups are deploying advanced practice clinicians in a manner that is nondifferential to patient outcomes. Second, claims-based ascertainment of complications does not permit severity grading for closer examination of differences by provider type. While we found no difference in complication rates by provider type, future work using clinical registry data with Clavien-Dindo grading would be a valuable extension of this analysis. Finally, this analysis was limited to Medicare beneficiaries undergoing radical prostatectomy and may not be generalizable to other procedures or non-Medicare populations. Outcomes and team care dynamics may differ for procedures with less standardized perioperative pathways. However, it appears urology groups deploy these clinicians in a thoughtful manner. Indeed, prior work demonstrates that integrating advanced practice clinicians into surgical groups enhances ambulatory access.⁹

These findings provide context to prior studies demonstrating improved perioperative outcomes with higher levels of PA and NP integration.^8,9 A single-center study of patients undergoing radical prostatectomy demonstrated that combined PA and NP involvement improves perioperative efficiency, including a shorter length of stay and earlier discharge, without increasing costs.²² In parallel, demand for their involvement across all phases of radical prostatectomy care has grown, with Medicare claims data demonstrating a 7% increase in intraoperative PA and NP assistance between 2004 and 2014, accompanied by an 11% decline in second-surgeon assistance during the same period.^11,12 Nursing-focused literature has similarly highlighted the expanding role of advanced practice nurses in the management of common prostatectomy-related complications.^15,23 Contrary to prior work that compares urology groups with and without PAs and NPs or studies that use staffing ratios to capture their involvement, our present analysis focuses on groups that already employ PAs and NPs and isolates variations in team composition. This distinction is crucial for workforce planning. As the field of urology faces persistent workforce shortages, the ability to flexibly integrate PAs, NPs, or both, particularly without compromising care quality, is paramount.

Conclusion

In this study of Medicare beneficiaries undergoing radical prostatectomy, postoperative outcomes did not differ across independent groups employing only PAs, only NPs, or both PAs and NPs. These findings suggest that employing PAs, NPs, or a combination of both is not associated with differences in postoperative outcomes, supporting flexibility in workforce and hiring strategies within surgical urology practices. As urology groups increasingly employ PAs and NPs to address rising patient demand for urologic services in an aging population, future research should assess how specific roles shape outcomes to better define optimal models for PA and NP integration in urological care.

FUNDING

Funding was received from AHRQ APP grant and NCI consolidation grant.

CONFLICT OF INTEREST DISCLOSURES

The authors have not declared any conflicts of interest.

ETHICS STATEMENT

This study was deemed not regulated by the institutional review board because it used deidentified claims. The requirement for informed consent was waived.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the Centers for Medicare & Medicaid Services (CMS), but restrictions apply to the availability of these data, which were used under a CMS data use agreement for the current study and therefore are not publicly available. Data may be obtained from CMS/ResDAC by qualified researchers following CMS approval and execution of an appropriate data use agreement. Analytic code and/or aggregate results may be available from the corresponding author upon reasonable request, to the extent permitted by the CMS data use agreement.

AUTHOR CONTRIBUTIONS

Conceptualization: Modi, Ellimoottil, Hollenbeck, Shahinian. Data curation: Sun, Hollenbeck, Shahinian. Formal analysis: Sun, Shay, Kaufman, Shahinian. Funding acquisition: Hollenbeck, Shahinian. Investigation: Modi, Shahinian. Methodology: Sun, Shay, Hollenbeck, Shahinian. Supervision: Kaufman, Hollenbeck, Shahinian. Validation: Kaufman, Hollenbeck. Writing—original draft: Montgomery. Writing—review & editing: Montgomery, Sun, Modi, Ellimoottil, Hollenbeck, Shahinian.