Polypharmacy and the prescription of potentially inappropriate medicines in patients with COPD and co-morbidities

O. J. Mercer, E. H. Baker, S. McIvor, S. L. Kennie, H. K. Lamplough. St George’s, University of London, London, United Kingdom.

Introduction: Chronic obstructive pulmonary disease (COPD) is a highly prevalent disease, with patients often having multiple co-morbidities.(1) This predisposes patients to the prescription of multiple medicines, thus increasing the risk of adverse reactions and other consequences. (2) Potentially inappropriate prescribing is the use of a drug where risks outweigh benefits incorporating failure to prescribe a safer alternative or potentially beneficial drug, incorrect dosing or duration of the prescription and increased risk of drug-drug and drug-disease interactions.(3) This study aims to determine the prevalence of prescriptions for potentially inappropriate medicines in patients admitted for an exacerbation of COPD.

Method: Patients admitted with a physician diagnosis of acute exacerbation of COPD between 1/9/2014 and 31/12/2014, who survived for ≥1 year after discharge were included. Long-term medicines prescribed at discharge and co-morbid diagnoses were identified from electronic patient records. Potentially inappropriate medicines (PIMs) were identified using Beers criteria for potentially inappropriate medication use in older adults (4) and the screening tool of older persons’ prescriptions (STOPP). (5)

Results: 120 COPD patients (62 men, 58 women; 73±10 years) were included. 36 (30%) were taking ≥5 medicines and 61 (51%) were taking ≥10 medicines. At least one PIM was identified in 68 (55%) patients using Beers criteria and 39 (33%) patients using STOPP criteria. In combination, Beers and STOPP identified at least one PIM in 75 (63%) patients. The total number of PIMS (Beers+STOPP minus duplicates) determined length of index admission (0 PIMs, 2.8±3.2 days; 1 PIM, 4.3±4.6 days; >1 PIM, 6.3±7.0 days, p=0.017) and had a borderline effect on readmissions in the subsequent year (0 PIMs, 2.3±1.9; 1 PIM, 3.1±2.7; >1 PIM, 3.6±3.1, p=0.086). On univariate analysis of variance, number of PIMs (p=0.011), but not total number of drugs (p=0.322) predicted readmission rate, with an interaction effect between number of PIMs and drugs prescribed (p=0.013).

Conclusion: Prescription of PIMs is common in people with COPD and is associated with length of stay and readmission rate. Admission provides an opportunity for medication review and cessation of PIMs. Further studies are required to determine whether this could reduce duration of hospitalisation or probability of readmission.

References:

1. Almagro P et al. (2010). Respir Med 104: 253-259

2. Maher RL et al. (2014). Expert Opin Drug Saf 13: 57-65

3. Spinewine A et al. (2007). Lancet 370: 173-184

4. American Geriatrics Society (2015). J Am Geriatr Soc 63: 2227-2243

5. O’Mahony D et al. (2015). Age Ageing 44: 213-218