Big Data Can Identify Risks for Near Misses Among Nurses

April 3, 2021
Laura Joszt, MA
Laura Joszt, MA

Interventions to address and prevent near miss events must be customized to a nurse's individual risk profile.

Use of big data is allowing researchers to better understand a nurse’s risks for a near miss (NM) based on workload variables. A NM is an event that could have caused harm for the patient but ultimately did not, either because it was caught or simply through chance. NMs are similar to medication errors.

The findings were published in Journal of Nursing Scholarship.

“NMs also occur as the result of the same systemic, organizational, and personal problems that cause errors, but occur more frequently, with estimates of NMs occurring 300 times more frequently than full-fledged errors,” the researchers wrote.

They conducted a literature review on past studies to understand the factors that increase medication administration errors and NMs. There were six workload factors commonly found across in-patient clinical hospital settings:

  • number of patients assigned,
  • number of medications ordered,
  • number of tasks assigned,
  • frequency of interruptions,
  • acuity of patient load, and
  • overtime hours worked.

A the medical-surgical unit of a private, medium-sized hospital participated in the study. The nursing staff was observed across shifts and times throughout the day. Data on gender, employment status, and each of the six workload factors was pulled from electronic health records, call lights, and bar-code medication administration (BCMA) systems. The data was collected over a 60-day period.

Over the study period, data was collected on the workload of 23 nurses caring for 389 patients. There were 11,595 tasks were assigned and 26,150 medications ordered. However, the study only included oral medications, which dropped the number of medications to 18,104. Tasks that could be completed by nurse assistants were also removed, dropping the number of tasks reviewed 6,605.

The data was reviewed for two-hour time periods since medications were typically not ordered more frequently than that. The majority of the nurses were female (87%) and they accounted for 88% of the two-hour blocks worked and 86% of the blocks with NMs. The three male nurses accounted for 12% of the blocks worked and 14% of the blocks with NMs.

The authors noted that while such gender disparity is common in the nursing workforce, there was no significant association between gender and NMs, nor was there a significant association between the full-time nurses (3 males and 15 females) and part-time nurses (5 females).

In total, there were 1,174 NMs associated with the oral medications flagged by the BCMA database. The majority (91%) of the NMs were of scanning the wrong dose-time-medication and the remaining NMs were flagged as the wrong patient.

Over the 60 days, the majority (78%) of two-hour time periods “were devoid of any NM,” the authors noted. Of the periods with at least 1 NM, 60% had only one NM, 27% had two NMs, 6% had three NMs, and 7% had four or more NMs.

Regarding the 6 workload factors, the researchers found considerable variation among the two-hour blocks. There was no significant correlation among the independent variables. A comparison of the workload factors among the nurses showed significant differences for each factor “indicating considerable variation in factors recorded from nurse to nurse.”

According to the researchers, the findings suggest the risk for an NM is impacted by a combination of variables, which means systemic and organizational changes that impact only a few factors would have only a moderate impact in reducing NMs. Instead, interventions to reduce NMs must be customized to each nurse’s risk profile.

They added that data on nurse workloads is already being gathered at hospitals and can be used to identify the risk for NMs.

“It is time that researchers and managers harness this vast amount of data and use it to improve the quality of care nurses are providing,” the authors concluded. “This can be done successfully if the data that are currently fragmented and locked away inside different clinical systems can be brought together and used to improve patient care.”