Mobility Atlas research: power-wheelchair-batteries-guide
Data and method for the article at mobilitatlas.com/blog/power-wheelchair-batteries-guide/
License: CC BY 4.0 (credit "Mobility Atlas")
Prepared: 10 October 2026

WHAT WAS READ
FDA MAUDE medical device reports that mention a battery, a charger or charging, fire, smoke, melting or sparks, from three sets: (a) product code ITI (powered wheelchairs) received 1 January 2019 to 31 December 2023, found by openFDA text search for battery, batteries, charger, charging or fire (88 reports); (b) product code ITI received from 1 January 2024 to the latest openFDA update read on 10 October 2026 whose text matches BATTER, CHARG, FIRE, SMOKE, MELT or SPARK (48 reports); (c) product code INI (mobility scooters) received from 1 January 2016 to the same date, same text match (87 reports). Pool: 223; 3 not relevant (N), 44 mention a battery but describe another problem (O), 176 battery-related.
All FDA recall records for product codes ITI (71) and INI (22), downloaded through openFDA on 10 October 2026.

HOW IT WAS CODED
1. A codebook (codebook.txt) was written before any report was coded. Its SHA-256 fingerprint, recorded at 2026-10-10T14:55:00Z, is d284301026d747c362c9389ea0b54724fb2bcf868c7c10da4a05c357cd48b1cc. Any later change to the codebook would change this fingerprint.
2. The reports were split into chunks. Two coders (AI language models working separately, each in a private scratch folder, each without access to the other's answers, told to read every report and not to use keyword scripts) coded every report in every chunk against the codebook.
3. Agreement between the two coders: device 85.2% (kappa 0.786); category 95.5% (kappa 0.925); injury 99.6% (kappa 0.988); charging 100.0% (kappa 1.0); lithium 100.0% (kappa 1.0).
4. Every report where the coders differed was decided by the lead analyst under the written rules below (42 reports marked adjudicated). A consistency pass then re-applied the rules across the whole table and changed 16 more (marked consistency).
5. Counts in the article were produced by script from the final table, not by hand. Some sub-counts (parts, places or words named in reports) are keyword counts over the report text; the exact word patterns are listed under KEYWORD COUNTS. Different wording would give somewhat different counts.

ADJUDICATION RULES
D1: DEVICE: S when the report text says scooter; otherwise C when it says chair, wheelchair or power chair, or names a power chair model; otherwise U. Product code alone decides nothing.
B1: Sparks at motor connectors after wrong cabling, or any problem outside the battery and charging circuit, is O.
B2: A fire reported where the device was present is F even when an investigator later says the device did not cause it; the note records that.
B3: Bloated or leaking batteries with only an odour reported by the user are L; heat named only as the maker's likely cause does not make it H.
B4: Overvoltage error codes from highly charged batteries or regenerative braking, with outdated firmware, are O: the report blames the control system, not a battery fault.
B5: Sparks or a burning smell with no flames or smoke reported are H.
B6: A hospital bed or other non-mobility device is N.
B7: INJURY is n when the report states only an impact (hit the street).
B8: CHARGING is y when the text says the device was plugged in to charge when the problem began, including when the user unplugged it on noticing the problem (consistency pass).
B9: LITHIUM is y only when the text names a lithium chemistry; a model name alone (for example Carbon) is not enough (consistency pass).

KEYWORD COUNTS (Python regular expressions, matched against the upper-case report text without its header line)
KW = {
 'charger': r'CHARGER',
 'battery': r'BATTER',
 'wiring': r'\bWIR(E|ES|ING)\b|HARNESS|CABLE|CONNECTOR|\bPLUG\b|TERMINAL',
 'charger_or_wiring': r'CHARGER|\bWIR(E|ES|ING)\b|HARNESS|CABLE|CONNECTOR|\bPLUG\b|TERMINAL',
 'smoking_oxygen': r'CIGAR|PIPE SMOKING|CARELESS SMOKING|WAS SMOKING|OXYGEN|\bO2\b|LIGHTER',
 'aftermarket_modified': r'AFTERMARKET|MODIFIED|MODIFICATION|OTHER SUPPLIER|THEIR BATTERIES',
}

QUALITY NOTES (disclosed in full)
- The article went through three fact-check passes by checkers who had not written it (a full pass with a reader review, then a control pass over the changes), each recomputing numbers from the files here and reading quotes against their sources; every error found was corrected before publication.

AMOUNTS
Every report text was searched for money figures; none gives one. Prices in the article are prices shown on makers', dealers' and retailers' pages and in the Mobility Atlas catalog, listed in power-wheelchair-batteries-guide-documents.csv and the article sources.

LIMITS
- Reports are not rates. Nobody publishes how many devices of each model are in use.
- A report is an allegation or a description; the FDA says the submission of a report "itself does not necessarily demonstrate that the device caused or contributed to the adverse outcome or event."
- The pool was found by searching for fire words among others, so fire reports are over-represented compared with, for example, batteries that slowly lose range.
- Few reports name the battery chemistry, so the data say almost nothing about lithium batteries.

FILES
codebook.txt        the codebook as fingerprinted
coded_reports.csv   one row per coded report: both coders' answers, the final code, and how a disagreement was resolved (resolution, rule_applied)
report_pool.csv     the reports in the pool with product code, brand, date received and event type
The full report texts are public: search MAUDE or openFDA by report key.
power-wheelchair-batteries-guide-documents.csv  the documents and pages read, with links