A vaping study links records: matched files can still contain matching errors
Separate correctly linked records, false links and missed links before interpreting a study built from several data sources.
No Vape · Updated

Find what linked means in this paper
The phrase matched records can describe different operations. Linking one person’s records across sources asks whether they concern the same person. Matching different people on measured characteristics builds a comparison group. Several papers describing one study create another relationship again. Read the methods before borrowing a familiar meaning from the headline.
The UK Office for National Statistics distinguishes false-positive links from false-negative missed links and warns that a match rate alone is not a linkage-quality measure. We use that methodological distinction, not a claim about any vaping dataset or a rule governing US services. The source provides no quitting outcome or assurance about a named linkage project.
UK Office for National Statistics: Data linkage and matching policy — quality standards
Draw a source-to-link map
Create a small reading card with one row for each original source, the entity being linked and the stated matching process. Add a separate row for any quality assessment the authors report. Leave identifying variables at their public descriptive level; there is no need to obtain names, dates of birth or private records to understand the method.
In an invented study, one file contains enrollment records and another contains follow-up contacts. Two different adults are accidentally joined, while a changed contact detail leaves another adult unlinked. These are opposite linkage errors. The example contains no actual participants, error rate or result, and does not show which error is more important in a real analysis.
Keep linkage coverage and correctness separate
An unlinked record is not automatically a person who did not quit. A linked record is not automatically a verified outcome. Check how the report handles those stages before repeating a final percentage. If the paper reports only the number linked, describe that number as linkage coverage rather than supplying an unreported accuracy estimate.
Ask whether the authors examined who was more difficult to link and how that limits the reported comparison. Do not decide the direction of a resulting bias without evidence. Even a plausible explanation, such as changed contact information, remains only a possibility unless the study establishes it. Your role is to locate the stated method and uncertainty.
Write the narrower claim you can support
A useful summary identifies the linked sources, the intended entity relationship, the linked subset and the quality checks actually reported. If correctness is unresolved, retain that uncertainty beside the result. Avoid redesigning the linkage, setting a universal threshold or recalculating outcomes from a screenshot. Those tasks require the actual authorized data and specialist methods.
Personal tracking offers a different kind of information. Your notes can describe what you observed without being combined with another person’s records or used as a substitute for research follow-up. For a care decision, seek qualified interpretation of the full evidence. This reading card prevents the convenient phrase linked data from implying complete, error-free observation.
Sources and scope
Sources inform the guidance. The worksheets and examples are No Vape editorial suggestions, not tested treatments.
Written with AI assistance and checked against the cited sources. No clinical review is claimed. This guide covers practical planning, not individual medical treatment. How we research and write