Use of potentially inappropriate medications in hospitalized elderly at a teaching hospital: a comparison between Beers 2003 and 2012 criteria.

OBJECTIVES: To detect the prevalence and pattern of use of Potentially inappropriate medications (PIMs) in hospitalized elderly patients of a tertiary care teaching hospital using Beers 2012 criteria and to compare the same with Beers 2003 criteria.
MATERIALS AND METHODS: Prescriptions of the elderly patients aged 65 years and above were collected from the medicine ward and analyzed. PIMs were identified with help of Beers 2003 and Beers 2012 criteria and comparison was made between the two criteria. Predictors associated with use of PIM were identified using bivariate and multivariate logistic regression analysis.
RESULTS: A total of 210 patients received 2,267 drugs. According to Beers 2003 criteria, 60 (28.57%) elderly patients received at least one PIM and 2.9% drugs were prescribed inappropriately. According to Beers 2012 criteria, 84 (40%) elderly received at least one PIM while 22 (10.47%) received multiple PIMs and about 5% drugs were prescribed inappropriately. The most commonly prescribed PIM was mineral oil-liquid paraffin (30, 14.3%) followed by spironolactone (25, 11.9%), digoxin (19, 9%), and benzodiazepines (14, 6.7%). There was a significant association between the number of patients receiving more than six drugs and the use of PIMs (P < 0.01). Use of more than 10 drugs was a significant predictor for use of PIMs in the elderly.
CONCLUSION: The study shows high prevalence of prescribing PIMs in hospitalized elderly patients. Beers 2012 criteria are more effective in identifying PIMs than Beers 2003 criteria.

Introduction

The world population is growing; what is new is the rapid pace of aging. The world's elderly population which was 550 million in 1996 is expected to approach 1.2 billion by the year 2025. As a result of decline in birth rate and increased life expectancy, the older population in most countries is growing faster than the population as a whole. Lesser developed nations are aging faster than their developed counterparts.[1] India's population aged 60 years and above, is projected to increase over the next decade from 8.3 to 10.7% in 2021 according to the United Nations Population Division. This is due to increased life expectancy at birth which is 67.3 years for males and 69.6 years for females. This number may also rise by almost 3 years during the next decade.[2]

The elderly population is the highest consumer of drugs. High degree of caution is required while prescribing in this vulnerable group. Due to age related changes in the physiology of elderly, these people also experience altered pharmacokinetics and dynamics of drugs. They have multiple comorbidities and are hospitalized more often which increases the chances of polypharmacy and inappropriate prescribing. Potentially inappropriate medication (PIM) is defined as “a drug in which the risk of an adverse event outweighs its clinical benefit, particularly when there is a safer or more effective alternate therapy for the same condition”.[3] Several studies have reported increased incidence of adverse drug reactions with PIMs.[45] Identifying potentially inappropriate medications is, thus, important to minimize the pharmacotherapy related hazards in this vulnerable population.

Several criteria have been designed for identification of PIMs. These criteria can be drug based explicit criteria like Beers[6] and Screening Tool of Older Persons’ potentially inappropriate Prescriptions (STOPP)[7] and patient based implicit criteria like Medication Appropriateness Index.[8] Beers criteria was the first set of explicit criteria for inappropriate drug use in elderly developed in 1991[9] by Mark H. Beers at the University of California. This criteria was subsequently updated in 1997[10] and 2003.[11]

Beers 2003 criteria is the most widely used explicit tool for identifying PIMs in elderly patients.[1213] Recently, Beers 2003 criteria has been used to identify PIM use in India both in outpatient setting and hospitalized elderly.[141516] With new medications being continually added to the market, a need for frequent updates of such tools is warranted.[17] Therefore, in 2011 the American Geriatrics Society (AGS) updated the Beers criteria by using an evidence-based approach after Dr. Beers passed away in 2009. The 2003 Beers criteria classified PIMs in three categories which included drugs to be avoided in elderly in general, drugs exceeding recommended dose and drugs to be avoided with specific comorbidity.[11] The latest version of the Beers criteria has undergone some major changes like addition of new drugs and deleting drugs which are no longer used. A new category of PIMs is also added for the drugs which are to be used with caution in the elderly.[18]

This study was undertaken to identify the prevalence and pattern of use of PIMs in elderly patients hospitalized in medicine department using the Beers 2003 criteria. During the course of the study, the revised Beers 2012 criteria were released, which was expected to be more relevant in identifying PIMs. Hence, a comparison between the Beers 2003 and Beers 2012 criteria was also made.

Materials and Methods

A prospective study was carried out in a tertiary care teaching hospital after obtaining approval of the Institutional Review Board. All patients aged 65 years and above were included in the study after obtaining their written consent. Patients hospitalized in the medicine wards from July 2011 to June 2012 were included. Data was collected in a case record form which included the patients’ demographic details, complete history, clinical data, and the drugs prescribed during period of hospitalization. The PIMs were identified with help of the Beers 2003 and 2012 criteria.

Statistical analysis

Data was analyzed using Statistical Package for Social Sciences (SPSS) software version 20. Association between polypharmacy and PIM use was determined using Fisher's exact test. Predictors (comorbid conditions, no. of drugs prescribed, and duration of hospital stay) associated with PIM use were identified using bivariate analysis (independent variables). Multivariate logistic regression was then applied to study the influence of these variables on PIM use. Value of P < 0.05 was considered statistically significant.

Results

A total of 210 patients were enrolled, among which 91 (43.3%) were males and 119 (56.7%) were females. The mean age was 69.34 ± 5.26 years (range 65-85 years). The total number of drugs prescribed in 210 patients was 2267 with 9.52 ± 2.75 drugs prescribed per patient. The average number of days of hospitalization in the medicine ward for elderly patient was 8.18 ± 2.89 days. The number of drugs prescribed increased significantly with the increase in hospital stay (P < 0.001). Most common disease conditions observed were (based on ICD-10 version 2010[19]) circulatory system disorders (67%) followed by respiratory system disorders (22%), infectious diseases (18%), and genitourinary system disorders (14%).

Hypertension (51.4%) was the most common comorbid condition affecting elderly followed by diabetes mellitus (24.3%), ischemic heart disease (22.9%), and chronic obstructive pulmonary disorder (16.7%). Table 1 shows pattern of PIM use compared to patient characteristics. Aspirin (58.1%) was the most frequently prescribed drug followed by ondansetron (53.8%), pantoprazole (50.95%), atorvastatin (49.05%), and clopidogrel (44.76%).

Table 1

Pattern of potentially inappropriate medication (PIM) use and patient characteristics (n = 210)

PIMs

Based on Beers 2012 criteria, 109 out of the 2,267 (4.8%) drugs prescribed were inappropriate and 84 out of 210 (40%) patients received at least one PIM. Based on Beers 2003 criteria, 66 (2.9%) drugs were prescribed inappropriately and 60 (28.57%) patients received at least one PIM. The most commonly prescribed PIM was mineral oil/liquid paraffin (14.3%), followed by spironolactone (11.9%), digoxin (9%), and benzodiazepines (6.7%).

Table 2 depicts frequency of PIMs based on Beers 2003 and 2012 criteria.

Table 2

Pattern of PIM use in the hospitalized elderly using Beers 2003 and 2012 criteria (n = 210)

Multiple PIMs

Out of the 84 patients who were prescribed PIMs considering Beers 2012 criteria, 19 (22.61%) patients received two PIMs and three (3.57%) received three PIMs. Of the 60 patients receiving PIMs as per Beers 2003 criteria, six (10%) were prescribed two PIMs. There was a significant association between the number of patients receiving more than six drugs (polypharmacy) and use of PIMs (P < 0.01). Bivariate analysis identified age 70-79 years (odds ratio (OR): 2.351; confidence interval (CI): 1.26-4.40; P = 0.009) and increased number of medications use (7-10 and ³11) during hospital stay (OR: 0.279; CI: 0.03-0.35; P < 0.05 and OR: 10.26; CI: 0.03-0.35; P < 0.001 respectively) as positive predictors of PIM use for Beers 2012 criteria. Gender, age, duration of hospital stay, and comorbid conditions did not predict PIM use. Multivariate logistic regression analysis suggested number of drugs (7-10 and ³11) as positive predictor for PIM use (OR: 4.68; P < 0.01 and OR: 10.26; P < 0.001, respectively) for Beers criteria 2012 [Table 3].

Table 3

Bivariate analysis and multivariate logistic regression: Predictors of PIM use (Beers 2012 criteria, n = 210)

Discussion

Worldwide the number of people above 65 years is on the rise due to increased longevity. Beers criteria, one of the widely used explicit criteria for identifying PIMs, have undergone a major reformation in 2012[6] as compared to the previous version of 2003 and is expected to be more competent in identifying PIMs. Some drugs like ferrous sulfate, guanadrel, guanathedine, mesoridazine, pseudoephedrine (for hypertension), and others which are no longer used have been omitted in the Beers 2012 criteria. Others like aspirin (for primary prevention of cardiac events), some anti-arrhythmic drugs, belladonna derivatives, some benzodiazepines, antipsychotics (for behavioral problems of dementia), metoclopramide, spironolactone, and many more which are being prescribed have been added in the recent edition. Few drugs to be prescribed with caution in the elderly have also been added as a new category in the revised Beers criteria [Additional File 1].

Out of four categories of PIMs in Beers 2012 criteria, the first forms a major group where the PIMs are divided into therapeutic categories. Seventy (64.22%) PIMs in our study fell under this category. Use of mineral oil/liquid paraffin is considered as PIM according to both the criteria (2003 and 2012) due to possibility of aspiration pneumonia in elderly.[610] As it is still widely used in our setup, it was the most common PIM in the first category. Second category includes drugs given at inappropriate doses; from which, spironolactone and digoxin contributed around 30% of PIMs. Under the drugs that may exacerbate a preexisting condition which comprised the third category, antipsychotics (in patients with behavioral problems of dementia) and calcium channel blockers (in patients with constipation) were considered inappropriate contributing to around 5% of total PIMs. Of the drugs to be used with caution in the elderly, selective serotonin reuptake inhibitor; escitalopram were prescribed in two patients.

Beers 2003 criteria have been instrumental in revealing high prevalence of PIM use between 25 and 45% in the developed part of the world.[182021] Use of PIMs has been recently reported from South India using Beers 2003 criteria where 4.1% of total drugs could be considered as potentially inappropriate as compared to 2.9% in our study. This difference could be due to difference in study population.[22] Our findings are however consistent with an earlier study which included patients from different specialties and both outdoor as well as hospitalized patients, reporting PIM use in 27.25% patients as compared to 28.57% in our study using Beers 2003 criteria;[16] but somewhat higher compared to a recent study reporting 24.6% prevalence in hospitalized elderly.[23] We have previously reported inappropriate prescribing in 23.58% of elderly patients in the outpatient setting using Beers 2003 criteria.[14]

Polypharmacy was the most significant predictor of PIM use; higher incidence of PIM use with higher number of drugs prescribed in hospitalized elderly patients. There was no statistically significant association when prevalence of PIM was correlated with variables like age, gender, duration of hospital stay, and associated comorbidities. There was a significant association between PIM prescribing and the diagnosis of acute decompensated heart failure and cerebrovascular accident according to Beers 2003 and 2012 criteria. In addition, Beers 2012 criteria also showed a significant association with age group less than 80 years and PIM prescribing.

As this study was restricted to medicine department, several drugs enlisted in Beers 2012 criteria were not prescribed like tricyclic antidepressants, antipsychotics (conventional and atypical), ergot derivatives, uterine relaxants (isoxsuprine), estrogens (with or without progestins), skeletal muscle relaxants (carisoprodol, chlorzoxazone, cyclobenzaprine, metaxalone, methocarbamol, orphenadrine), and anticancer agents (carboplatin, cisplatin, vincristine). Hence, the findings of this study are not applicable to other specialties.

To our knowledge this is the first study of PIM use with Beers 2012 criteria. As expected, we found Beers 2012 more competent than Beers 2003 criteria in identifying the PIMs-2.9% viz-a-viz 4.8% of total drugs as PIM and 28.57% viz-a-viz 40% of hospitalized elderly receiving PIM. This may be attributed to the significant changes made in the updated Beers criteria. The revised Beers criteria are definitely a superior tool compared to earlier version of 2003.

Regardless of different editions of Beers criteria used for evaluating the use of PIMs in the elderly, there is a high prevalence of PIM use in hospitalized elderly which must be brought to the attention of the clinicians.