Original article
Concordance between patient-reported and physician-reported sexual function after radical prostatectomy

https://doi.org/10.1016/j.urolonc.2017.09.017Get rights and content

Highlights

  • Evaluate how surgeons self-report their outcomes after radical prostatectomy.

  • Real-time feedback was provided to surgeons for quality improvement purposes.

  • Agreement between patients and physicians reporting outcomes was high.

  • Implementation of a real-team feedback for reporting patient outcomes is feasible.

Abstract

Purpose

Accurately tracking health-related quality-of-life after radical prostatectomy is critical to counseling patients and improving technique. Physicians consistently overestimate functional recovery. We measured concordance between surgeon-assessed and patient-reported outcomes and evaluated a novel method to provide feedback to surgeons.

Materials and methods

Men treated with radical prostatectomy self-completed the International Index of Erectile Function-6 questionnaire at each postoperative visit. Separately, physicians graded sexual function on a 5-point scale. International Index of Erectile Function -6 score<22 and grade ≥3 defined patient-reported and physician-assessed erectile dysfunction (ED), respectively. Feedback on concordance was given to physicians starting in May 2013 with the implementation of the Amplio feedback system. Chi-square tests were used to assess agreement proportions and linear regression to evaluate changes in agreement after implementation.

Results

From 2009 to 2015, 3,053 men completed at least 1 postprostatectomy questionnaire and had a concurrent independent physician-reported outcome. Prior to implementation of feedback in 2013, patients and physicians were consistent as to ED 83% of the time; in 10% of cases, physicians overestimated function; in 7% of cases, physicians, but not patients reported ED. Agreement increased after implementation of feedback but this was not statistically significant, likely owing to a ceiling effect. Supporting this hypothesis, increase in agreement postfeedback was greater during late follow-up (≥12 mo), where baseline agreement was lower compared to earlier follow-up.

Conclusions

Agreement was higher than expected at baseline; implementation of feedback regarding discrepancies between patient-reported and physician-assessed outcomes did not further improve agreement significantly. Our observed high rate of agreement may be partly attributed to our institutional practice of systematically capturing patient-reported outcomes as part of normal clinical care.

Introduction

Treatment of localized prostate cancer with radical prostatectomy (RP) carries risks of functional impairment in urinary and sexual health-related quality-of-life (HRQOL) [1]. Patient-reported outcomes (PRO) are now critical components of clinical trials used to direct patient-centered care [2]. Providing PRO to individual surgeons helps them to counsel patients, improve surgical technique, and guide follow-up care decisions.

Systematic assessment of PRO is still lacking, however, though the American Urological Association guidelines for localized prostate cancer recommend an assessment of overall health and functional status to guide treatment and follow-up care [3], assessment of patient functional status has historically been sparse. In the Cancer of the Prostate Strategic Urological Research Endeavor (CaPSURE) and American College of Surgeons National Cancer Data Base cohorts, 22% to 64% of men had no documented assessment of urinary or sexual functional status [4], [5].

Even when physicians do assess and document patient functional status, research has demonstrated that physicians’ reports are often discordant with patient experience. For instance, in 1 study, surgeon and patient sexual function assessments were concordant for only 55% of patients [5]. These discordant physician assessments consistently underestimate the functional limitations that men experience across multiple domains before and after prostate cancer treatment [6].

Increasing awareness of the importance of PRO assessment in routine clinical care has failed to increase agreement between physician- and patient-assessed functional status. In separate reports from the CaPSURE cohort that were over a decade apart, there was no temporal convergence in the agreement of patient- and physician-reported outcomes [7]. We hypothesized that actively providing surgeons with systematic feedback as to their patients’ self-reported sexual function outcomes would improve the concordance between patient- and physician-reported outcomes.

Section snippets

Amplio feedback system

Starting January 2009, PRO were systematically collected as part of routine clinical care at Memorial Sloan Kettering Cancer Center (MSKCC). Patients with prostate cancer prospectively completed a validated HRQOL questionnaire assessing erectile, urinary, and bowel function as well as global quality-of-life before RP and at regular follow-up intervals [8]. The survey was administered using an interactive secure online form completed before clinical appointments via e-mail or immediately before

Results

In the period from January 2009 to April 2015, a total of 3,053 men completed at least 1 post-RP HRQOL self-assessment and had a concurrent, independently reported sexual function outcome from the physician. Patient characteristics are described in Table 1. Patients who received surgery after implementation of Amplio concordance feedback more often received robot-assisted laparoscopic surgery (71% vs. 45%; P<0.0001) and fewer patients harbored low-risk Gleason grade 3 + 3 prostate cancer (21%

Discussion

Despite the importance of accurately assessing functional status in prostate cancer treatment decision-making and survivorship, there have been few studies examining the issue of patient-physician agreement on outcomes and possible overestimation of functional recovery by surgeons. Prior work from Litwin et al. and Sonn et al. using the CaPSURE cohort demonstrated significant and persistent discordance between physician- and patient-reported functional assessments using the University of

Conclusions

In a large institutional cohort of men undergoing RP, standardized and systematic collection of patient- and physician-reported outcomes demonstrated a high degree of interrater agreement. Actively providing feedback to surgeons on their degree of agreement with patient self-assessments did not further improve the degree of agreement. There is ample reason to believe that the systematic collection of PRO as part of routine clinical care contributes to improved delivery of patient-centered

References (22)

  • B.A. Spencer et al.

    Variations in quality of care for men with early-stage prostate cancer

    J Clin Oncol

    (2008)
  • Cited by (2)

    • A 25-year perspective on advances in the study of the epidemiology, disparities, and outcomes of urologic cancers

      2021, Urologic Oncology: Seminars and Original Investigations
      Citation Excerpt :

      Within data science, development of methods harnessing data mining, machine learning, and natural language progressing represent important technological advances to facilitate the analysis of the complex and extensive data captured in ADs and linked genetic and imaging databases. Multiple studies have identified that provider-reported and patient-reported outcomes (PRO) can differ in important ways [30-33]. The addition of PRO would provide critical perspective on the benefits and burdens of treatment beyond Common Terminology Criteria for Adverse Events (CTCAE) grading and survival outcomes.

    Funding: Supported by funds from David H. Koch provided through the Prostate Cancer Foundation, Sidney Kimmel Center for Prostate and Urologic Cancers, P50-CA92629 National Cancer Institute SPORE Grant, P30-CA008748 National Institutes of Health cancer center support Grant, and T32 CA82088-15 National Cancer Institute training Grant.

    View full text