Limited National Potential for Reducing Anesthesia Clinician Staffing for Cataract Surgery: A Historical Cohort Study Using a Large United States Anesthesia Registry

Figures

• 

  Figure 1. Flow diagram to identify elective cataract cases Cases with implausible durations for primary elective cataract surgery were excluded. The five-minute threshold corresponds to the 0.01 percentile, and the 90-minute threshold to the 0.9725 percentile. Cases starting during late evening and early morning hours were also excluded, as they likely represent either portions of urgent or emergent surgery or miscoding, given that elective cataract surgeries would be expected to start rarely after 8 PM or before 5 AM. Of the 754,718 cataracts identified in the dataset, 131,765 (17.5%) were excluded.

• 

  Figure 2. Determination of the number of simultaneously running ORs (i.e., number of anesthesia clinicians providing care) The number of simultaneously running cases (horizontal bars), equivalent to the number of anesthesia clinicians providing care, was inferred from the number of overlapping cases in each 1-minute bin of the day (red numbers at the bottom of the graph). The effective start of the workday was the first time the maximum number of simultaneous running ORs occurred, 6:30 AM in the example. That value corresponds to the number of anesthesia clinicians needed to staff the ORs performing cataract surgery. The process continued throughout the workday until 3:00 PM. The time until the OR with the shortest list ended corresponds to the first time when the number of ORs simultaneously running dropped by one and did not increase for the remainder of the workday.

• 

  Figure 3. Distribution of the median proportions among facilities The median proportions for the 66.7th percentile exceeding six hours were examined and confirmed to follow a normal distribution by the Shapiro-Wilk test (W=0.99, P=0.24). In the upper panel (A), the logits of the observed proportions closely follow the expected Z scores for a normal distribution. In the lower panel (B), the histogram of the logits closely follows the normal distribution (red line).