Lee JS, Verbeek PR, Schull MJ, et al. Paramedics assessing Elders at Risk for Independence Loss (PERIL): Derivation, reliability and comparative effectiveness of a clinical prediction rule. CJEM. 2016;18(2):121-132.
We’ve all responded to the dreaded “lift assist” call where the patient needs more than just help getting back in bed. EMS providers are among the few healthcare practitioners who actually enter people’s homes and make first-hand assessments on site, particularly for elderly patients.
Researchers at the Sunnybrook Research Institute and three ambulance services from Ontario, Canada, devised and tested the validity and reliability of a rule for determining elderly patients who have a high risk for independence loss: the Paramedics assessing Elders at Risk for Independence Loss (PERIL) checklist.
After determining the reliability and clinical prediction usefulness of the PERIL checklist, they compared it to the Identify Seniors at Risk (ISAR) checklist, which has been previously used and validated in the ED but not validated for paramedic use. They also compared PERIL with the paramedic global assessment of risk (i.e., a medic’s “gut feeling”).
Method: The PERIL checklist consists of 43 seemingly “clinically sensible” variables. It was developed through a strong consensus-building process involving various experts: a clinical epidemiologist, a decision rule expert, two geriatricians, a paramedic with research experience and paramedic focus groups.
The checklist’s prediction rule was studied with multivariate logistic regression. Thirty-eight dichotomous environmental variables were reviewed along with five questions from the ISAR. Descriptive statistics were used after collecting baseline demographic data and basic health data.
Once PERIL was created, existing one-hour continuing education sessions were used to train 1,185 paramedics in the three Ontario ambulance services. These sessions included covering the study rationale, objectives and use of the PERIL data form.
A simulated patient encounter was used to determine a prediction value and to improve interrater reliability.
Outcome measures studied included patient return to the ED, hospitalization and/or death within 30 days of the paramedic encounter.
Results: In total, 1,065 patients were enrolled in the study and 71% (756) of these patients had a complete data set. Inter-observer reliability was determined to be good or excellent and had advantages compared to clinical judgement. The PERIL checklist performed better than the proxy measure of the tried and true paramedic “gut feeling.” ISAR was an acceptable alternative, but adherence was lower than the PERIL checklist.
Using multivariate logistic regression, researchers identified four key elements as significant markers for patients at high risk for independence loss:
- Are there potential problems in the home?
- Was there a repeat call within 30 days?
- Is the patient male?
- Does the patient lack social support?
Each “yes” answer to these questions increases the patient’s risk of losing independence.
The researchers concluded that the PERIL checklist could be used by paramedics and emergency physicians to target preventive services or interventions for high-risk populations.
Discussion and limitations: The authors report a 28.3% loss of data due to information systems (which is actually fairly low). Implementation of electronic charts will greatly assist research like this in the future.
The PERIL checklist, as devised and validated in this study, is a promising new tool for EMS providers. This study’s conclusions suggest that quick and targeted training can improve overall care for elderly patients.
Visit www.pcrfpodcast.org for audio commentary.
What we already know: Patients over the age of 65 often use EMS for non-urgent events. They are often isolated, vulnerable, refuse transportation and have a high mortality. Lift assist calls are extremely dangerous and high-risk non-transports should be carefully screened.
What this study adds: New evidence for a possible decision-making checklist could assist field personnel in identifying high-risk elderly patients.
Multivariate regression: Research method by which researchers systematically compare multiple variables to that which they are studying in an attempt to find a relationship.
Interrater reliability: The extent to which to independent evaluators agree on their observations or ratings.