research

Do incident logs match the care record for the quarter-start review

A method for comparing incident logs with care records at quarter start, including how to treat missing and corrected entries.

Staffing Care Home Editorial TeamPublished 11 min read
Do incident logs match the care record for the quarter-start review
On this page8 sections

Do incident logs match the care record for the quarter-start review: research question and scope

This review asks whether an incident log entry and the care record describe the same event with the same date, time, people, and follow-up. The scope covers incidents reported during the first two weeks of the quarter, plus any older incident that was still open when the quarter began. It includes falls, medication events, missed visits, and complaints that were logged as incidents. It excludes general service notes that were never treated as incidents, unless the care record shows a safety event that should have been logged.

The purpose is reconciliation, not blame. A mismatch can mean the log is incomplete, the care record is incomplete, or the two records use different definitions. The review should identify which record is missing what, and whether the gap affects the follow-up. A mismatch that does not change the follow-up is still a documentation finding, but it is a different finding from an incident that never received a review.

Population and unit of analysis

The population is every incident entry tied to an active resident or service setting during the window. The unit is one incident, even when it appears in two systems. Link the log entry and the care record with a shared key before comparing them. If the systems do not share a key, use the resident, date, and event type to build a crosswalk and record how the link was made.

Stratify by incident type, severity, and whether the event was reported by a caregiver or discovered during a review. A fall reported at the time is different from a fall found in a weekly audit. The incident log routine describes the fields that a complete entry should carry. Use those fields as the comparison points, not as a substitute for reading the actual records.

Definitions and measures

A match means the log and the care record agree on the resident, date, event description, reporter, and follow-up status. A partial match means the core event agrees but one or more supporting fields differ. A mismatch means the records describe different events or different dates, or one record is missing entirely. Keep those categories separate, because a missing follow-up is not the same as an event that was never reported.

The primary measure is the share of incidents that fully match. Secondary measures include the share with a missing follow-up, the median time between the event and the log entry, and the number of care-record events that have no log entry. Report the denominator and the review window for each measure. A count of mismatches without the total number of incidents is not interpretable.

Evidence standard

The evidence is the incident entry, the care record, and any linked notification or follow-up note. A screenshot of one system is not enough because it cannot show whether the other system contains the same event. Use a dated export or a read-only report that shows the record as it existed during the review. If a record was corrected after the event, keep the original and the correction together.

The reviewer should record whether the evidence was available at the time of the incident or added later. A late entry can still be accurate, but it should be labeled as late. The incident completeness study explains why the timing field matters. Do not treat a corrected record as a match until the correction explains what changed and who approved it.

Sampling and collection

Sample from the incident log and from a separate list of care-record events that look like incidents. Starting only from the log will miss events that were never logged. Start with every open incident, then add a sample of closed incidents and a sample of care-record events that mention a fall, medication issue, or missed visit. That design gives the review a chance to find under-reporting, not just mismatched fields.

Collect the resident key, event date and time, reporter, description, notification, follow-up owner, and closure date. Record any field that is absent rather than filling it from the other system. If the two systems use different definitions, write the definitions down before comparing. The collection log should show the source of each field so a later reviewer can reproduce the match.

Analysis and reporting

Report the full-match rate, the partial-match rate, and the mismatch rate. Then show the most common missing field and one dated example. If the care record has events with no log entry, report that separately as an under-reporting signal. Do not average the two findings into a single score, because they point to different process changes.

Break the results down by incident type and reporting source. Medication events may match well because they trigger a pharmacy review, while missed visits may match poorly because they are handled as scheduling issues. The timesheet approval checkpoint is a useful cross-check when a missed visit also affects payroll. The report should name the process that creates the gap, not just the record that lacks a field.

Limits and sources of error

The review measures record agreement, not the quality of the response. Two records can match while the follow-up was inadequate, and a mismatch can occur even when the care was appropriate. Do not present the match rate as a safety score. Use it as a signal about whether the documentation can support a later review.

Other limits include systems that overwrite history, incidents reported through informal channels, and care records that use free text without a consistent event type. If the sample is small, avoid percentages that imply more precision than the data supports. If the review covers one service line, say so. The finding describes that setting and should not be generalized to every agency without a larger sample.

Review cadence

Run the reconciliation at each quarter start and use a smaller monthly check for open incidents. The quarter-start review should repeat the same definitions and sampling frame so the comparison is meaningful. The monthly check can focus on incidents that changed status or owner during the month. Keep a dated record of the sample and the crosswalk.

Assign the review to a named role and set a follow-up date for each recommended change. The follow-up should confirm whether the log fields, the notification step, or the closure rule actually changed. A report that is filed without a follow-up is only a description of the past. The goal is a log and a care record that tell the same story when a reviewer reads them together.

Sources

<- Back to research