Potentially Inappropriate Medications
Conditions
Keywords
Electronic Health Records, Hospital Medication Systems, Geriatrics, Pharmacy, Electronic Prescribing, High-risk drugs
Brief summary
This study will assess whether a modification in the default dose and frequency (the first option a provider sees) during electronic prescribing of a high-risk drug can impact prescribing behavior and subsequent changes in average dose for the targeted high-risk drug, when prescribed to a hospitalized patient aged ≥65 years. In this cluster randomized crossover (CRXO) trial we will randomize a non-intrusive nudge intervention, which involves modifying the default dose for high-risk drugs when prescribed electronically to hospitalized patients aged ≥65 years. The CRXO trial involves 10 sites in an urban health system: five sites will start the trial under the intervention/control during a first time period (T1) after which they switch intervention/control status (T2). The primary outcome is prescription rate of a lower default dose (i.e. the geriatric standard) for 8 high-risk drugs. This study will inform the effectiveness of EHR-based nudge interventions to reduce inappropriate prescribing of high-risk drugs for elderly patients. Analyses ongoing, expected to finalize spring 2023
Interventions
BEHAVIORALEHR-based nudge interventions
Change/introduction of default drug and frequency during electronic prescribing of high-risk drugs for a patient aged ≥65 years
Sponsors
New York City Health and Hospitals Corporation
Icahn School of Medicine at Mount Sinai
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
HEALTH_SERVICES_RESEARCH
Masking
DOUBLE (Subject, Caregiver)
Masking description
Participants and care providers will be unaware of the intervention. However, prescribers will be able to prescribe all available doses, without any restrictions
Intervention model description
This cluster randomized crossover (CRXO) trial involves 10 sites in an urban health system: five clinics will start the trial under the intervention/control during a first time period (T1) then will switch to intervention/control status (T2). The primary outcome is prescription rate of a lower default dose for 8 high-risk drugs. Other inputs used for trial design were: baseline rate of prescriptions meeting the geriatric standard (sources: local data, Geriatrics At Your Fingertips), within-period intraclass correlation coefficient (ICC, correlation between two randomly chosen observations in the same cluster), cluster autocorrelation (ratio of between-period ICC to within-period ICC), and coefficient of variation of cluster sizes (required when unequal sample size per cluster is expected). The aim is for ≥80% power to detect a ≥50% increase in prescriptions meeting the geriatric standard; this is based on previous studies and on the researchers' preliminary data on Trazodone.
Eligibility
Sex/Gender
ALL
Age
65 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* a patient must be hospitalized * 65-years old or older * receive one of the targeted medications
Exclusion criteria
\- none
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Prescription rate of modified default dose for eight high-risk drugs | 24 weeks | Rate at which the modified default dose (geriatric standard) and frequency for eight high-risk drugs are prescribed. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Mean per-patient dose | 24 weeks | Mean per-patient dose of the relevant individual high-risk drugs. |
| Rate of inpatient falls | 24 weeks | Inpatient falls (with and without injury) |
| Rate of 30-day readmissions | 30 days post-discharge regarding hospitalization during which a high-risk drug was prescribed | 30-day readmission rates for patients aged ≥65 years who received any of the eight high-risk drugs during their hospitalization. |
| Drug-specific prescription rate | 24 weeks | Drug-specific prescription rate of the modified default dose for each of the eight high-risk medications and type of order (frequency/type of order). |
| Cost of hospital stay | 24 weeks | Cost of hospitalization in USD for patients aged ≥65 years who received any of the eight high-risk drugs during their hospitalization. |
| Spillover effects to patients aged <65 | 24 weeks | The rate of prescriptions of the newly defaulted (geriatric) dose for any of the eight high-risk drugs among patients aged \<65 years. |
| Length of hospital stay | 24 weeks | Duration of hospitalization in days for patients aged ≥65 years who received any of the eight high-risk drugs during their hospitalization. |
Countries
United States
Outcome results
None listed