Week Ahead
Five questions to ask on Sunday about any data milestone week
A ten-minute Sunday routine that tells you whether this week's lock, dry run or transfer is on track, and what to do first on Monday.
3 min readRead the brief →
Blog · New every day at 10:00 ET
The Readiness Brief
One practical brief for clinical data, programming and submission teams, every morning at 10:00 Eastern.
1 brief published · Next brief: · Standards Decoded ·
Latest
Today's brief
A few minutes to read. Three things to do today at the end, one question to check yourself, and links to the Navigator tool that applies it.
Weekly rhythm
A different lens every day
Each weekday has its own format, so you know what is coming tomorrow. Today's is highlighted.
- SundayWeek AheadA short checklist to set up the week before Monday arrives.
- MondayStandards DecodedOne element of SDTM, ADaM, Define-XML or terminology, explained for the people who have to deliver it.
- TuesdayLock RoomDatabase lock, dry runs and the decisions that have to be closed before them.
- WednesdayReviewer's EyeWhat an agency reviewer opens first, and how to make it easy for them.
- ThursdayCapacity MathEstimates, staffing and QC effort, with the arithmetic shown.
- FridayFailure FridayOne common failure, the early signal and the fix.
- SaturdayWeekend Deep ReadA longer look at one guideline or standard and what it changes in practice.
Archive
All briefs
Newest first. Filter by format.
Week Ahead
The clocks changed overnight. Do your dates know?
Daylight saving time ended in the US this morning, so one local hour happened twice. A week-ahead check of how your study stores dates, times and partial dates.
3 min readRead the brief →
Weekend Deep Read
Clean validation, wrong data, what conformance rules cannot check
A validation report with no errors says the data follow the standard's rules. It does not say the data are right. A longer look at FDA's validator rules, CDISC CORE and the gap between them and correctness.
3 min readRead the brief →
Failure Friday
The late SAP amendment that arrives after the dry run
A change to the statistical analysis plan after the dry run is common and usually legitimate. It fails when it is treated as a small edit instead of a change with an impact analysis.
3 min readRead the brief →
Capacity Math
Independent double programming, what it costs and where to spend it
Double programming every output roughly doubles the effort and still misses some errors. A risk-based QC plan spends the independent effort where an error would change a conclusion.
3 min readRead the brief →
Reviewer's Eye
Can a reviewer get from your table back to the CRF?
Traceability is the reviewer's ability to follow a number from a results table to the analysis data, the tabulation data and the collected record. A five-step test you can run on your own outputs.
3 min readRead the brief →
Lock Room
The real cost of unlocking a database
Unlocking after database lock is possible and sometimes right. It is never cheap. What it costs, who decides, and how to make the decision quickly when it comes.
3 min readRead the brief →
Standards Decoded
What define.xml is actually for
Define-XML is not a file to generate at the end. It is the reviewer's dictionary for every dataset, variable and value, and its quality shows how well the data were specified.
3 min readRead the brief →
Week Ahead
Interim analysis week, a checklist for the people who are not on the DMC
An interim analysis puts unblinded results in front of a few people and keeps them away from everyone else. The week runs on access, data cut rules and rehearsal.
3 min readRead the brief →
Weekend Deep Read
Estimands for programmers, what ICH E9(R1) changes in ADaM and outputs
The estimand framework is often treated as a statistician's topic. In practice it decides which dates, flags and derivations the analysis datasets must carry. A practical reading.
3 min readRead the brief →
Failure Friday
Controlled terminology drift, the quiet failure in pooled data
CDISC Controlled Terminology changes every quarter. Studies that never pin a version end up with values that disagree across datasets, studies and the define.xml.
3 min readRead the brief →
Capacity Math
Why programming estimates come in low, and the range that fixes it
Single-number estimates for SDTM, ADaM and TLF work are reliably optimistic. A three-point range and one rule about adding them up give a date you can defend.
3 min readRead the brief →
Reviewer's Eye
The reviewer's guide is read first, so write it first
The data reviewer's guides are the reviewer's map to your datasets. What they are for, what FDA expects in them, and the faults that make them useless.
3 min readRead the brief →
Lock Room
The dry run that tests the wrong thing
A dry run that only proves the programs run is a rehearsal of the easy part. Here is what it should prove instead, and how to triage what it finds.
3 min readRead the brief →
Standards Decoded
TS, the smallest SDTM dataset with the biggest consequence
Why a missing or wrong Trial Summary dataset can stop an electronic submission at the door, and how to check yours in fifteen minutes.
3 min readRead the brief →
Week Ahead
Five questions to ask on Sunday about any data milestone week
A ten-minute Sunday routine that tells you whether this week's lock, dry run or transfer is on track, and what to do first on Monday.
3 min readRead the brief →
No briefs in this format yet. The next one follows the weekly rhythm.
Follow daily
Get the next brief at 10:00 ET
No account and no email list. Pick whichever suits you; nothing about you is sent anywhere.
- Add a daily 10:00 ET reminderA calendar file with one repeating event. Your calendar shows it in your own time zone.
- RSS feedFor a feed reader. A brief appears in the feed at its publish time.
Open a brief and your reading streak is counted here, on this device only.