Pre-Operative Testing for Cataract Surgery Among Adults in LA County

Pre-operative testing for cataract surgery provides no discernible benefit to patients, increases risk of harm, and substantially raises healthcare costs-representing the quintessential example of low value care. Despite multiple randomized controlled trials demonstrating no benefit, physicians continue to routinely order pre-operative testing for most patients undergoing cataract surgery.

Supported by the American Board of Internal Medicine's Choosing Wisely™ campaign to reduce low value care, a quality improvement (QI) initiative to reduce pre-operative visits and testing for cataract surgery at LAC+USC Medical Center, a safety-net health system serving Los Angeles County's under-resourced populations and control site, Harbor-UCLA Medical Center will be evaluated.

Using Institute for Healthcare Improvement PDSA (plan, do, study, act) cycle techniques, the QI team implemented the following steps (1) review randomly sampled charts on cataract surgery patients, (2) show local data on over-testing to hospital leadership, (3) obtain buy-in from the chairs of anesthesia and ophthalmology, (4) recruit an ophthalmology resident champion, and (5) empower nurses to stop scheduling pre-operative visits for cataract surgery. On October 13th, 2015, at LAC+USC Medical Center but not Harbor-UCLA, the resident champion and department chairs emailed "new" clinical guidelines for pre-operative testing to faculty, trainees, and staff, and specifically promoted avoidance of pre-operative testing for cataract surgery, except for guideline-concordant point-of-care glucose testing for patients with diabetes and potassium testing for patients with chronic kidney disease.

Evaluation of this quality improvement initiative will consists of a quasi-experimental pre-post time series analysis. Primary outcomes include pre-operative medical visits, laboratory testing, chest X-rays and electrocardiograms within 80 days of surgery (chart review revealed the maximum duration of pre-operative testing was 80 days) for patients undergoing cataract surgery before and after the intervention.

Multivariable generalized linear models that account for clustering by clinician and adjust for patient and provider characteristics and time (in months) will be used.

.