Your panel and productivity report, line by line

The report is a set of fractions

Once a month someone hands you a page of numbers about your patients: panel size, continuity, visits per session, no-shows, referrals, imaging orders, wRVUs. Most residents find the one number that looks bad and move on. Read it differently. Every line is a fraction. Somebody chose what counts in the numerator, what counts in the denominator, and how far back to look. Change any of those and the number moves, even if your care did not. This module walks the report line by line with a fictional sample and the arithmetic behind each figure. It does not tell you what your numbers should be. It shows you how to read them well enough to ask good questions at your practice's business meeting.

Panel and empanelment

Your panel is the set of unique patients assigned to you as their primary care physician. Empanelment is the process of making that assignment, usually by applying a lookback window and an attribution rule to the visit history.

Two choices decide your panel size. The first is the lookback window. A widely used convention, described by Murray, Davies, and Boushon in Family Practice Management, counts patients seen by anyone in the practice in the last 18 months; they note a 12-month window tends to undercount. Eighteen months is a convention, not a rule, and your practice may use something else. The second is attribution: whose patient someone is after seeing several physicians. The same article assigns patients who saw only one physician to that physician, then uses the physician seen most often, then the one who did the last physical, then the one seen last. Many EHRs simply use the PCP field entered at registration. The article also gives a supply-demand identity: panel size × visits per patient per year = visits per day × clinic days per year. Illustratively, 20 visits a day × 200 days ÷ 3 visits per patient per year is about 1,333 patients.

Why it matters: If your panel looks small, check the window and the attribution rule first. A patient who saw you twice and your co-resident three times may sit on their panel, and your continuity is computed against that list.

Two continuity formulas, one patient, two answers

A fictional patient on a PGY-2's panel had 10 visits in the lookback window: 6 with the resident, 3 with a co-resident on the same team, and 1 with a faculty physician on a same-day slot.

Bottom line: One patient, 0.60 on one index and 0.40 on the other, and two patients with identical UPC can differ on COC. Before comparing continuity across residents or quarters, confirm the same index, lookback, and minimum visits per patient. Patients with one visit score 1 by definition, so many methods exclude them.

A sample monthly report (fictional)

Every figure below is invented for teaching. None is a benchmark. Visits per session = completed visits ÷ half-day sessions. No-show rate = no-shows ÷ scheduled appointments; some practices also count same-day cancellations, so ask which version you are reading. Referral rate = referrals ÷ completed visits. Diagnostic utilization is usually per 100 visits, for example advanced imaging orders ÷ visits × 100. wRVUs per visit = total work RVUs ÷ completed visits. The attending column shows the format, not a target.

Where 1.44 wRVUs per visit comes from

The fictional PGY-2 above completed 98 established-patient visits: 8 coded 99212, 60 coded 99213, and 30 coded 99214. Work RVUs: 99212 = 0.70, 99213 = 1.30, 99214 = 1.92.

Bottom line: wRVUs per visit is mostly a coding-mix number, and coding mix follows documentation of medical decision making or total time. It is a fair question whether your notes support the level you chose. It is not a target to code toward: the code has to match the note.

Putting your numbers next to an attending's

A resident's report and an attending's use the same headings and measure different things. The schedule differs. Residents see fewer patients per session by design, in program-set templates with precepting time built in. PGY-1 templates are lighter still. The billing differs. Under Medicare's teaching physician rules, a resident's visit is billed by the teaching physician. The primary care exception (42 CFR 415.174) lets a resident with more than six months of training furnish lower- and mid-level E/M visits without the teaching physician in the room, while the teaching physician directs no more than four residents at a time and reviews each visit. The wRVUs on your report are usually attributed to you for tracking, not billed under your name. The panel differs. It is newer, turns over at graduation, and is shared with a team, which pulls continuity down however hard you work.

How to avoid it: Compare your report to your own prior months, and to residents at the same PGY level in the same practice using the same report definitions. When a number moves, ask whether the template, the attribution rule, or the lookback window changed before assuming your care changed.

Read the denominator

Your no-show rate drops from 14% to 9% in one month. Nothing about your patients or your scheduling changed that you know of.

Reading your report

• Panel size depends on the lookback window and the attribution rule; 18 months is a common convention, not a rule.
• UPC and COC can give the same patient different continuity scores (0.60 and 0.40 for a 6-3-1 visit split), so confirm which index your report uses.
• wRVUs per visit is driven by coding mix: a 99214 carries 1.92 work RVUs and a 99213 carries 1.30, and the code must match the documentation.
• A resident's report reflects a training template and teaching physician billing; compare it to your own trend and same-PGY peers, not to an attending.

Do this next: Find your most recent practice report and write down, for each line, the lookback window and the denominator. Any you cannot fill in are your questions for the next business meeting.