Adjustment for Patient-related Factors Needed When Evaluating Spine Surgeon Performance

Subscribe to the latest updates from Orthopaedics Advances in Motion

Thank you for subscribing!

Error: Please enter a valid email address.

Email Address Submit

Medicine continues to shift toward alternative payment models, which give clinicians incentives to provide high-quality, cost-efficient care. Patient-reported outcome performance measures (PRO-PMs) are attractive to payers as one way to evaluate clinicians, including surgeons who perform costly, high-volume orthopedic procedures.

The National Institutes of Health developed the Patient-Reported Outcome Measurement Information System (PROMIS^®) to address limitations of older PRO-PMss such as Short Form 36. Government and commercial payers are optimistic about using PROMIS, but there is little understanding about the influence of individual surgeons on PROMIS scores.

Aditya V. Karhade, MD, MBA, orthopaedic surgery resident, and Daniel G. Tobert, MD, spine surgeon in the of the Orthopaedic Spine Center at Massachusetts General Hospital, and colleagues have conducted the first study of surgeon variance in PROMIS scores after primary lumbar decompression, the most common spine procedure. In The Spine Journal, they show that variance in scores was influenced by patient-related factors and not by the individual surgeon.

Study Methods

This prospective study, conducted at two academic medical centers and two community medical centers, included 636 patients treated by nine surgeons. All surgeons were fellowship-trained, six in orthopedic surgery and three in neurosurgery. The median time in practice was 12 years (range, 4–31).

The surgeons each operated on at least 30 patients who satisfied the following criteria:

• Index surgery between January 1, 2016, and April 30, 2019
• Age ≥18 years
• Primary one-level (75% of patients) or two-level (25%) posterior lumbar decompression for lumbar disc herniation (LDH, 57%) or lumbar spinal stenosis (LSS, 43%)

Primary Outcome

The primary outcome was the change from the preoperative PROMIS physical function score equal to or greater than the minimal clinically important difference (MCID). There is no standard MCID; in this study, it was defined using a distribution-based method as a four point improvement. 76% of patients had a change in score that reached that threshold.

A multivariable analysis showed no significant differences between surgeons in patients' rates of reaching the MCID.

Risk Adjustment

Patients were significantly more likely to reach the MCID if they had LDH compared with LSS (OR, 2.58). Risk factors that made reaching the MCID significantly less likely were:

• Higher body mass index (OR, 0.93)
• Two or more medical comorbidities (OR, 0.38)
• Higher (better) preoperative PROMIS physical function scores were less likely to reach MCID (patients who were already doing well before surgery were less likely to see the significant benefits of surgery)
• Higher rate of unemployment in zip code of residence, used as an indirect measure of sociodemographics (OR, 0.82)

Measuring Performance Accurately

Surgeon-level performance measurement would give patients objective evidence to select their surgeon, provide important feedback to surgeons and potentially improve mortality rates, as demonstrated in a study published in Heart. The enthusiasm for this concept is tempered by the long-standing concern that, in response, surgeons would be less willing to operate on high-risk patients.

This study provides evidence that patient-related risk factors should be incorporated into any alternative payment model that aims to distinguish surgeon-level performance using PROMIS. The ability to measure surgeon-level performance adequately is also dependent on sample size and data quality.

view original journal article Subscription may be required

Learn more about the Orthopaedic Spine Center

Refer a patient to the Department of Orthopaedics