I-821D Median Collapses 136 Days in One Week, But the Signal Is Misleading
A -9.32 sigma anomaly fired for the week of June 1, 2026, as the I-821D final-stage median fell from 170 days to 34. The drop almost certainly reflects a composition shift in which cases were approved this week, not a genuine change in USCIS throughput pace.
The I-821D weekly median dropped 136 days in a single week, from 170 days to 34 days, the week of June 1, 2026. The anomaly detector fired at z = -9.32 against the prior 4-week trailing window, the largest single-week low in the 12-week series. The 12-week trailing average of 136 days remains the more reliable baseline. A drop of this magnitude, paired with an inverted RFE (Request for Evidence) pattern and an unusually wide spread between the 25th and 75th percentiles, points to a batch of atypically fast cases clearing together this week rather than any acceleration in USCIS adjudication (USCIS's decision step) pace.
- The I-821D final-stage median fell 136 days week over week, from 170 days to 34 days, triggering an outlier flag at z = -9.32 against the prior 4-week trailing mean.
- RFE (Request for Evidence) response cases cleared in 9 days this week versus 133 days for plain cases, a 14.5x inversion of the normal pattern consistent with a batch of older RFE-response cases resolving together.
- The 25th percentile (p25) this week sat at 20 days while the 75th percentile (p75) reached 133 days, a spread of 113 days that confirms the approved pool was internally mixed rather than uniformly fast.
- The active pending backlog grew by 14 cases (+20%) over the past 31 days, with 44.9% of cases sitting at the processing stage at a median of 82 days.
- The 4-week forecast carries an R² of only 0.22, making the range-required band of ~17 to 107 days for the week of June 8 the appropriate read rather than any single projected number.
The 34-day median that needs a reality check
The I-821D final-stage median fell from 170 days (week of May 25) to 34 days (week of June 1), a 136-day single-week drop of 79.9%. The anomaly detector fired at z = -9.32 against the prior 4-week trailing mean, the most pronounced single-week departure in the 12-week series. A move of this magnitude almost always reflects which cases happened to be approved in a given week rather than a change in USCIS adjudication (the case decision step where approval, denial, or an evidence request is issued) pace. The 12-week trailing average of 136 days is a better guide for anyone estimating when their case might resolve.
When RFE cases close faster than plain cases
The most telling detail this week is the RFE inversion. Cases that received a Request for Evidence from USCIS and submitted a response cleared in a median of 9 days, while cases that went through plain processing posted a median of 133 days, a 14.5x speed ratio. Under normal conditions, RFE-stamped cases take longer because the adjudicator must review the applicant's additional documents after the response arrives. An rfe_fast anomaly flag fired alongside the outlier flag, consistent with a batch of older RFE-response cases resolving together in a single week and pulling the overall median sharply downward. The data does not tell us why these cases batched; a single week is not enough to confirm any mechanism.
14.54x RFE response cases clear faster than the plain processing cohort.
What the pipeline actually looks like right now
The underlying pending-case pipeline looks nothing like this week's headline number. According to the Stage durations (30-day raw) data, 44.9% of active I-821D cases sit at the 'Case Is Still Being Processed By USCIS' status, with a median of 82 days and a spread running from 48 days at the 25th percentile (p25) to 118 days at the 75th percentile (p75) across 37,824 cases. Another 35.7% of cases are at the biometrics (fingerprint and photo appointment) stage with a median of 20 days across 30,068 cases. The overall active pending backlog (cases not yet adjudicated) grew by 14 cases, a 20.0% increase, over the 31-day window from May 7 to June 7, 2026, driven mainly by the December 2025 filing-month group adding 5 active cases.
The filing-month cohort burndown data reinforces that recent filers are nowhere near approval. Cases filed in January 2026 (the group of cases filed in that month, now at 22 weeks of age) show a 12.5% approval rate across 32 tracked cases, while the February 2026 group (18 weeks old, 13 cases) is at 7.7%. Cohorts from March through May 2026 show 0% approval so far. The December 2025 group, now 26 weeks old across 51 tracked cases, has reached 27.5% cleared. Because the tracked cohort sizes here are small (ranging from 1 to 51 cases per filing month), individual percentages carry wide uncertainty and should not be read as a trend toward acceleration based on this week's headline reading.
Preview
Unlock the rest of this brief
Subscribers see the full analysis, including detailed forecasts and personalised case lookup.
- Full premium analysis: outlook, what it means for you, and implications.
- Personalised case lookup against this week's cohort, with percentile and remaining days.
- Estimated approval window mapped to your filing date.