eduKateSG · Why Science?
Read adverse-event reports as safety signals with explicit limits
Use FDA MAUDE through MDR identifiers, reporter types, device fields, event narratives, supplemental reports, update dates, data files and rate limitations.
Postmarket surveillance helps regulators and health professionals notice possible medical-device problems after products are in use. The FDA MAUDE database contains medical device reports submitted by manufacturers, importers, device user facilities and voluntary reporters.
A report can describe a death, injury, malfunction or other event, but FDA cautions that the data have major limits. Reports can be incomplete, inaccurate, unverified, duplicated or influenced by reporting practices, and they do not supply a dependable denominator for device use.
Did you know? FDA explicitly says MAUDE data should not be used to calculate adverse-event rates, compare occurrence rates across devices or evaluate rate changes over time. A larger count can reflect use volume or reporting behaviour rather than greater risk.
This guide teaches safety-data literacy. A suspected device problem may require clinical care and formal reporting; a database search is not medical diagnosis or advice.
Chapter index
Chapters 1–9: report identity, device fields and events
- 1. MAUDE contains medical device reports
- 2. A report does not prove causation
- 3. Mandatory and voluntary reporting differ
- 4. Event types are categories
- 5. Device identity can be messy
- 6. Report dates answer different questions
- 7. Initial and supplemental reports connect
- 8. Narratives are source statements
- 9. Counts are not incidence rates
Chapters 10–17: duplicates, denominators, coverage and surveillance
- 10. Trends are difficult to infer
- 11. Comparing devices by raw counts is unsafe
- 12. The searchable window has boundaries
- 13. Monthly updates create a retrieval date
- 14. Exemptions and alternative reporting affect coverage
- 15. MAUDE is one postmarket signal source
- 16. Reporting suspected problems supports surveillance
- 17. Safety-data literacy supports public health careers
An adverse-event report is a signal record, not a proven causal conclusion. The database houses MDRs submitted by mandatory reporters and voluntary reporters concerning suspected device-related events and product problems. This supports safety awareness without turning raw counts into incidence claims.
Try this
Evidence question: Which report type and source are shown? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Record the MDR report key and reporter category. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for maude contains medical device reports, its identifier or version context, and one limit that must remain visible.
MAUDE collects mandatory and voluntary medical-device reports from several reporter types. FDA warns that an MDR alone generally cannot establish that a device caused or contributed to an event. Narratives may be incomplete or unverified. It also keeps reporter knowledge and FDA verification limits visible.
Try this
Evidence question: What causal statement would exceed the record? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Use “reported event” language. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for a report does not prove causation, its identifier or version context, and one limit that must remain visible.
Reporting rates cannot be calculated when device-use denominators and complete reporting are unavailable. Manufacturers, importers and device user facilities have defined mandatory duties, while clinicians, patients and consumers can submit voluntary reports. Search dates and report identifiers make observations auditable.
Try this
Evidence question: Who submitted this report? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Preserve reporter type and occupation fields. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for mandatory and voluntary reporting differ, its identifier or version context, and one limit that must remain visible.
Narratives, corrections and duplicate reports require careful record-level reading. Reports may be coded as death, injury, malfunction or other categories under reporting rules. A category is not a clinical adjudication. A suspected problem should be handled through appropriate medical and reporting channels.
Try this
Evidence question: What event type is coded? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Read the narrative and follow-up status. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for event types are categories, its identifier or version context, and one limit that must remain visible.
An adverse-event report is a signal record, not a proven causal conclusion. Brand names, model numbers, product codes and manufacturer names can be incomplete, misspelled or changed. This supports safety awareness without turning raw counts into incidence claims.
Try this
Evidence question: Which identifiers are independently verifiable? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Cross-check AccessGUDID and clearance records. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for device identity can be messy, its identifier or version context, and one limit that must remain visible.
CHAPTER 6 OF 17 · Read narratives carefully
6. Report dates answer different questions
MAUDE collects mandatory and voluntary medical-device reports from several reporter types. Date received, event date, report date and manufacturer awareness date describe different moments and may be missing. It also keeps reporter knowledge and FDA verification limits visible.
Try this
Evidence question: Which date is being plotted? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Use the exact field label. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for report dates answer different questions, its identifier or version context, and one limit that must remain visible.
CHAPTER 7 OF 17 · Read narratives carefully
7. Initial and supplemental reports connect
Reporting rates cannot be calculated when device-use denominators and complete reporting are unavailable. A reporter may submit follow-up information that updates an initial MDR. Counting each row without relationship checks can duplicate an event. Search dates and report identifiers make observations auditable.
Try this
Evidence question: Is this a supplement to an earlier report? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Link report numbers before counting. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for initial and supplemental reports connect, its identifier or version context, and one limit that must remain visible.
Narratives, corrections and duplicate reports require careful record-level reading. Event and manufacturer narratives describe reported circumstances and investigations. They can contain redactions, uncertainty and differing perspectives. A suspected problem should be handled through appropriate medical and reporting channels.
Try this
Evidence question: Whose narrative is this? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Separate reported facts, investigation and conclusion. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for narratives are source statements, its identifier or version context, and one limit that must remain visible.
An adverse-event report is a signal record, not a proven causal conclusion. FDA states that MDR data lack reliable denominators for device use and are affected by under-reporting, duplicate reporting and reporting bias. This supports safety awareness without turning raw counts into incidence claims.
Try this
Evidence question: What denominator would be needed? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Do not calculate rates from report counts alone. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for counts are not incidence rates, its identifier or version context, and one limit that must remain visible.
MAUDE collects mandatory and voluntary medical-device reports from several reporter types. Changes in reporting requirements, publicity, market share, device age and data processing can change counts independently of underlying risk. It also keeps reporter knowledge and FDA verification limits visible.
Try this
Evidence question: Which reporting change could affect the series? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Annotate the timeline with system changes. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for trends are difficult to infer, its identifier or version context, and one limit that must remain visible.
CHAPTER 11 OF 17 · Understand counting limits
11. Comparing devices by raw counts is unsafe
Reporting rates cannot be calculated when device-use denominators and complete reporting are unavailable. Different numbers of devices in use and different reporting behaviour make report totals unsuitable for comparing adverse-event occurrence rates. Search dates and report identifiers make observations auditable.
Try this
Evidence question: What exposure data are missing? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Avoid ranking products from MAUDE counts. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for comparing devices by raw counts is unsafe, its identifier or version context, and one limit that must remain visible.
CHAPTER 12 OF 17 · Understand counting limits
12. The searchable window has boundaries
Narratives, corrections and duplicate reports require careful record-level reading. The MAUDE interface focuses on reports within a recent ten-year window, while older and special program data are available through MDR data files. A suspected problem should be handled through appropriate medical and reporting channels.
Try this
Evidence question: Which years and files were searched? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: State the coverage explicitly. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for the searchable window has boundaries, its identifier or version context, and one limit that must remain visible.
CHAPTER 13 OF 17 · Connect wider safety evidence
13. Monthly updates create a retrieval date
An adverse-event report is a signal record, not a proven causal conclusion. The MAUDE search is updated monthly with reports received through the prior month. Results can change with new and supplemental reports. This supports safety awareness without turning raw counts into incidence claims.
Try this
Evidence question: What was the data cutoff? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Save access date and update note. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for monthly updates create a retrieval date, its identifier or version context, and one limit that must remain visible.
CHAPTER 14 OF 17 · Connect wider safety evidence
14. Exemptions and alternative reporting affect coverage
MAUDE collects mandatory and voluntary medical-device reports from several reporter types. FDA has permitted exemptions, variances and alternative reporting arrangements in defined circumstances; some data are found in separate files. It also keeps reporter knowledge and FDA verification limits visible.
Try this
Evidence question: Which reporting route applies? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Check MDR data files and program notes. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. If a field is missing, write “not shown” and follow the official documentation rather than inventing a value.
Evidence checkpoint
Record the evidence object for exemptions and alternative reporting affect coverage, its identifier or version context, and one limit that must remain visible.
CHAPTER 15 OF 17 · Connect wider safety evidence
15. MAUDE is one postmarket signal source
Reporting rates cannot be calculated when device-use denominators and complete reporting are unavailable. FDA evaluates MDRs with recalls, inspections, studies, complaints and other information. A signal may prompt investigation without proving a rate or cause. Search dates and report identifiers make observations auditable.
Try this
Evidence question: Which corroborating source is needed? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Follow recalls and safety communications. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Before comparing two records, check that their object types, methods, versions and units really align.
Evidence checkpoint
Record the evidence object for maude is one postmarket signal source, its identifier or version context, and one limit that must remain visible.
CHAPTER 16 OF 17 · Connect wider safety evidence
16. Reporting suspected problems supports surveillance
Narratives, corrections and duplicate reports require careful record-level reading. Patients, clinicians and facilities can report device problems through FDA channels. Urgent medical needs should be handled clinically first. A suspected problem should be handled through appropriate medical and reporting channels.
Try this
Evidence question: What action is appropriate now? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Separate care, reporting and research. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Finish by naming one supported statement and one stronger statement that still needs another evidence layer.
Evidence checkpoint
Record the evidence object for reporting suspected problems supports surveillance, its identifier or version context, and one limit that must remain visible.
CHAPTER 17 OF 17 · Connect wider safety evidence
17. Safety-data literacy supports public health careers
An adverse-event report is a signal record, not a proven causal conclusion. Epidemiologists, engineers, clinicians and regulators interpret reports with denominators, verification and system context. This supports safety awareness without turning raw counts into incidence claims.
Try this
Evidence question: Which limitation changes the conclusion most? Answer in three parts: identify the visible field or file, explain how it bears on the question, and state the boundary of that evidence. Preserve the database’s own object names and accession rather than flattening everything into the word ‘data’.
Try this: Write a report-level evidence receipt. Turn the result into a compact claim-evidence-reasoning note, then ask a partner to retrace it. Keep the identifier, retrieval date and source page in the note so the route can be repeated.
Evidence checkpoint
Record the evidence object for safety-data literacy supports public health careers, its identifier or version context, and one limit that must remain visible.
Contents · Previous chapter · Continue at the Science Learning Hub
A safety signal stays useful when the report and its limitations remain inseparable
MAUDE provides valuable public reports for signal detection and investigation. Its strength comes from surfacing possible problems, while its limitations prevent raw counts from becoming unsupported risk rankings.
Save the MDR report key, reporter type, event type, device identifiers, dates, initial or supplemental status, narrative source, product code, search terms, data window, update date and limitations statement.
| Reading object | What it can support | Question to keep visible |
|---|---|---|
| MDR report | Submitted account of a suspected device event | Who reported and was it updated? |
| Event type | Regulatory report category | Does the narrative support the interpretation? |
| Device identity | Brand, model, product code or UDI link | Which fields were verified? |
| Report count | Number of matching records | What denominator and reporting biases are missing? |
Sources and next reading
Substantive resource claims in this article were checked against the primary pages below. A structured record is a route into evidence; it does not by itself prove a scientific conclusion. Recheck the live record, evidence labels and release context before using it in assessed work or research.
For a classroom enquiry, save a short research receipt containing the question, search terms, selected object, exact identifier, version, access date and the evidence boundary you observed. Add the file or page that supports each statement. This small routine turns browsing into accountable science learning and gives a teacher, classmate or future analyst a route they can independently follow.
Before sharing the conclusion, ask another reader to retrace the route without verbal hints. If they reach a different version, file or interpretation, improve the receipt and narrow the claim. Reproducibility begins with this practical check. It teaches careful reading, supports fair collaboration and keeps uncertainty visible while evidence moves from an online service into notes, analysis and a final explanation.
A useful final review separates five questions. What object was found? Which organisation or repository maintains it? What exact evidence was inspected? Which statement does that evidence support? What remains uncertain? Answer each question in one sentence, then connect every answer to a visible field, file or source page. This review catches common mistakes: citing a collection when a file supplied the value, treating the newest release as the analysed release, confusing permission with scientific suitability, or turning a discovery count into a conclusion. It also gives students a calm way to improve work. A missing detail becomes a next research step instead of an invitation to guess. When the route is complete, the final explanation can stay simple because the identifiers, dates, versions, limitations and supporting evidence are already organised underneath it.
Primary sources
- FDA MAUDE search — accessed 25 October 2026.
- About MAUDE — accessed 25 October 2026.
- FDA medical device reporting — accessed 25 October 2026.
- FDA MDR data files — accessed 25 October 2026.
- openFDA device event API — accessed 25 October 2026.