Delivery rate is the metric most health plans track first for the H.R.1 notice window — and it is the wrong one to stop at. A notice that reaches a mailbox but is not understood produces the same outcome as a notice that never arrives: a member who does not know what to do, and who is at risk of procedural disenrollment when enforcement begins January 1, 2027. With roughly fifty days left before the August 31 window closes, comprehension measurement is the piece most plans have not built yet, and it is still possible to build it.
Why delivery data alone is misleading
A high delivery rate can coexist with a comprehension crisis. In Arkansas's 2018–19 rollout, notices were sent, but the majority of coverage losses were procedural — driven by members who did not understand what was required of them or how to act, not by members who were actually ineligible. Delivery answers the question 'did it arrive.' It does not answer 'did the member know what to do next.' Those are different questions, and only the second one predicts your disenrollment risk.
Signal one: call-center and member-services themes
Your call center is already collecting comprehension data — it just may not be tagged as such. Pull call themes and inbound reasons related to the notice: how many calls are basic clarification questions ('what does this letter mean,' 'am I required to do this,' 'what happens if I do nothing') versus calls where the member already understood the notice and is taking a specific next step. A high ratio of clarification calls to action-oriented calls is a direct signal that the written notice is not doing its job, and it is a signal you can get within days of mailing, not months.
Signal two: readability testing before you mail, not after
Standard readability scoring — checking sentence length, jargon density, and grade level against plain-language benchmarks — should happen before content goes out, but it is worth re-running against your actual mailed notice, not just the draft, since edits during legal or compliance review often reintroduce complexity. If your English notice tests above a sixth- to eighth-grade reading level, or your Spanish version reads as a literal translation rather than natural language, comprehension will suffer regardless of delivery rate.
Signal three: small-sample member testing
Where feasible, a small round of direct testing — showing the notice to a handful of members or community members from your target population and asking them to explain in their own words what it says and what they would do next — surfaces comprehension gaps that readability formulas miss. This does not need to be a large formal study; even five to ten conversations, run through a community partner, FQHC, or navigator relationship, can reveal whether a notice's core message is landing.
Signal four: exemption-claim and response-rate patterns by language and channel
Break your response and exemption-claim rates out by language and by channel. If your Spanish-language notices are producing a meaningfully lower response rate than your English notices, or your mail-only segment is responding at a much lower rate than your SMS or portal segment, that gap is comprehension and access data, not noise. It tells you exactly where to concentrate remediation in the weeks remaining before August 31.
Building comprehension tracking into your existing dashboard
None of these four signals require a new measurement system. Most plans already have a business-intelligence or reporting dashboard tracking delivery and response rates for the notice window; the work is adding comprehension-specific fields to it — call-theme tags, readability scores by version, response rate broken out by language and channel — rather than building a parallel process. Framing this as an extension of existing reporting, rather than a new project competing for the same fifty days of runway, is usually the difference between comprehension measurement actually happening and it staying on a to-do list until after the window closes.
What to do with fifty days left
Comprehension measurement only matters if it changes something before the window closes. Pull call-center themes and response-rate-by-language data now, this week, not in September. Where a gap shows up, revise the notice language or add a supplementary, simplified follow-up message targeted at the segment showing weak comprehension — a second channel with clearer language often closes a gap that a repeat of the original notice will not. The plans that measure comprehension now, while there is still runway to act on it, are the ones whose January enforcement numbers will reflect actual eligibility rather than actual confusion.