Inappropriate Medication Prescriptions in Elderly Adults Surviving an Intensive Care Unit Hospitalization

Alessandro Morandi, MD, MPH; Eduard Vasilevskis, MD; Pratik P. Pandharipande, MD, MSCI; Timothy D. Girard, MD, MSCI; Laurence M. Solberg, MD; Erin B. Neal, PharmD; Tyler Koestner, MS; Renee E. Torres, MS; Jennifer L. Thompson, MPH; Ayumi K. Shintani, PhD, MPH; Jin H. Han, MD, MSc; John F. Schnelle, PhD; Donna M. Fick, PhD; E. Wesley Ely, MD, MPH; Sunil Kripalani, MD, MSc

Disclosures

J Am Geriatr Soc. 2013;61(7):1128–1134 

In This Article

Methods

Study Design and Population

This prospective cohort study was nested in a larger long-term cohort study (NCT00392795) that enrolled critically ill individuals admitted with respiratory failure or shock to the medical, surgical, or cardiovascular ICU at Vanderbilt University Hospital. Individuals were excluded from the parent study if they were moribund, had respiratory failure or shock for longer than 72 hours before enrollment, were unable or unlikely to participate in cognitive testing during follow-up (because of blindness, deafness, inability to speak English, active substance abuse, or psychotic disorder), or were at high risk for severe cognitive impairment before the time of screening (individuals admitted after cardiopulmonary arrest or with documented acute neurological injury, those with chronic neurological disease that prevented independent living, and those who had undergone cardiac surgery in the 3 months before screening). Only individuals enrolled in the parent study who were aged 60 and older and were discharged alive from the hospital were included in the current study. The age cutoff of 60 was chosen, consistent with previous research,[15] to include individuals at high risk of polypharmacy and inappropriate medication prescribing. Individuals discharged to hospice were excluded because of common use of PIMs for symptom control (i.e., these PIMs are rarely AIMs in the hospice population). Informed consent was obtained from an available surrogate at enrollment in the parent study; individuals provided consent before hospital discharge, after their critical illness had improved and they were deemed competent to consent. The institutional review board at Vanderbilt University approved the study protocol.

Demographics and Clinical Characteristics

Demographic characteristics, Acute Physiology And Chronic Health Evaluation (APACHE) II severity-of-illness score,[16] ICU admission diagnoses, type of ICU, and comorbidities according to the Charlson Comorbidity Index[17] were recorded at study enrollment. Trained research personnel used the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU)[18] to assess individuals for delirium daily until hospital discharge or study Day 30. Information on length of stay in the ICU and hospital, discharge location (home vs other), and discharging hospital service (medical vs surgical) was also recorded from the medical record.

Medication Abstraction and Classification

Medical charts (including physician notes and medication administration records) were reviewed to identify PIMs using the 2003 Beers criteria,[19] which were supplemented with additional medications identified by reviewing the medication safety literature published since 2003,[6,20–22] considering articles reporting the association between medication prescription, adverse events, and medication safety in elderly adults. Although a formal review with the Delphi approach was not completed, an evidence-based approach was applied to the selection of these medications, as suggested in the Institute of Medicine standards for practice guidelines (http://www.iom.edu/Reports/2011/Clinical-Practice-Guidelines-We-Can-Trust.aspx) and as used in the recent Beers update.[23] Most of the medications included in the list have been added to the updated Beers Criteria.[23] A clinical panel comprising a hospitalist (EEV), a geriatrician (LS), and a clinical pharmacist (EN) reviewed all PIMs at hospital discharge to identify AIMs. Similar to an approach used previously,[15] the panel reviewed hospital discharge medications, participant medical history, and laboratory data to determine whether each discharge PIM was actually inappropriate (an AIM) based on the clinical circumstances of the individual. A PIM was considered an AIM when at least two of the three panel members considered its risk–benefit profile to be unfavorable based on the individual's specific circumstances and criteria specified in the Medication Appropriateness Index,[4, 24]including indication, dosage, and likely effectiveness, as well as drug–drug interactions, drug–disease interactions, unnecessary duplication, and duration of treatment. A medication did not need to have caused harm to be considered an AIM. This approach was designed to mirror multidisciplinary clinical decision-making on rounds as opposed to independent assessments by individual clinicians, so agreement between individual clinicians was not calculated.

Each PIM and AIM was classified into one of the following 12 mutually exclusive categories based on medication class and side effects: benzodiazepines, non-benzodiazepine sedatives, typical antipsychotics, atypical antipsychotics, opioids, anticholinergics, antidepressants, drugs causing orthostasis, nonsteroidal antiinflammatory drugs, antiarrhythmics, muscle relaxants, and others. A complete list of medications reviewed, according to their classification, is available in Appendix S1.

To determine where specific types of AIMs were initiated, medications were abstracted from the medical record at five distinct time points—before admission (outpatient medications recorded at the time of admission), ward admission (outpatient medications continued at admission plus newly prescribed inpatient medications), ICU admission, ICU discharge, and hospital discharge.

Statistical Analysis

Participant demographic and clinical variables were summarized using medias and interquartile ranges for continuous variables and proportions for categorical variables. PIMs and AIMs were described as the total number prescribed in all participants at different time points. For each discharge PIM category, the percentage of PIMs that were determined to be AIMs were calculated, and this was considered to be the positive predictive value (PPV) for that PIM category; PIMs with higher PPVs could be useful when screening for possible AIMs (yielding more true positives), whereas PIMs with lower PPVs would yield more false positives (PIMs that were appropriately prescribed).

Multivariable Poisson regression models with generalized estimating equations were used to analyze risk factors for the number of PIMs and AIMs per participant at discharge. PIMs and AIMs were analyzed as the number prescribed per participant (continuous variables) rather than as present or absent (dichotomous variables) to preserve statistical power. Age, number of preadmission PIMs, Charlson comorbidity score, total days of delirium, hospital length of stay, discharge disposition (home or not home), and discharge service (medical or surgical), determined a priori according to prior publications[9,25] and clinical relevance, were included in both models. All covariates were included in both models, regardless of statistical significance.

R (version 2.11.1, www.r-project.org) was used for all statistical analyses. Two-sided P < .05 was considered statistically significant.

processing....