Introduction

Some drugs are considered potentially inappropriate medications (PIMs) for older adults due to a potential inefficacy to achieve the desired therapeutic response, an increased risk of adverse events that exceeds the benefits of the drug, or the existence of a safer therapeutic option [1,2]. Concern over the harmful effects caused by some drugs used by the elderly led to the development of lists of PIMs based on explicit and implicit criteria [1]. The Beers criteria are the most widely used and known worldwide and was updated in 2015 by the American Geriatrics Society [2]. Avoiding the use of PIMs is an important, simple, and effective strategy for reducing drug therapy-related issues in the elderly [2]. However, the prevalence of the use of these drugs is approximately 40% and is commonly associated with the female gender and the practices of self-medication and polypharmacy [1-3].

The use of many different drugs by elderly patients makes them more susceptible to Potential Drug-Drug Interactions (PDDIs). These interactions might alter the pharmacological activities of the drugs involved, increase or decrease their effects or even cause undesirable clinical consequence relevant to quality of life [4,5]. These consequences can be aggravated by the involvement of one or more PIMs in the PDDI, leading to an elevated risk of complications caused by the use of these drugs [6]. In the 2015 Beers criteria was incorporated into a new area with a PDDI list that should not be overlooked and reinforced the risk of serious harm and PDDI associated with damage to the health of the elderly [2].

Based on the Beers Criteria by the American Geriatrics Society [2], we identified studies that evaluated the use of PIMs by analyzing explicit factors but did not evaluate the PDDIs involving these drugs, which is an essential intrinsic factor for therapeutic success and patient safety. In this context, this study analyzed potential interactions involving medications that are inappropriate for the elderly according to the updated Beers Criteria and the health conditions associated with these PDDIs and the use of PIMs.

Methods

Study Design, Participants, and Data Sources

This cross-sectional population-based study formed part of the in-home survey conducted by the Elderly Health Vigilance Network (EHVN). The sample consisted of 934 elderly (at least 60 years of age) individuals living in the city of Goiânia, Goiás, Brazil. It was approved by the Research Ethics Committee of the Federal University of Goiás.

The data were collected through visits to the homes of elderly individuals by duly-trained interviewers between December/2009 and April/2010. The parameter details of the sample size calculations and sampling procedures are described in previous publications [7]. Briefly, the sample calculation assumed an expected frequency of 30.0% for all outcomes, 95% confidence intervals, a significance level of 5%, an absolute precision of 5% and a Design Effect of Cluster Sampling (DEF) of 1.8. To cover possible losses, an additional 11% was included, yielding a sample size of 934 elderly individuals. Of this total, 783 responded to questions regarding the use of medication and were included in this study. The elderly was selected by cluster sampling using 912 Census Sectors (CS) from the urban area of the municipality as the primary sampling units. Polypharmacy, Self-Medication and Potentially Inappropriate Medications.

During data collection, the elderly individuals were asked about the medications they used daily, with or without prescription, and were requested to show all of the drugs they use and the respective prescriptions. Thus, it was possible to obtain the necessary data on drug use, such as the use of PIMs, polypharmacy, and self-medication, and to analyze PDDIs. To identify PIMs, the updated Beers criteria [2] were applied to the drugs mentioned by the elderly during the survey and the corresponding concentrations and doses. The practice polypharmacy was defined as the concomitant use of at least five drugs [8].

To investigate the practice of self-medication based on the drugs used by the elderly, the individuals were asked who had indicated each of the drugs and to show the prescriptions. All of the drugs taken by the individuals without a current prescription were considered to be due to the practice of self-medication. A prescription with a date older than three months was considered olden.

Potential Drug-Drug Interactions

We used the Micromedex^® tool, which is available online with restricted access, to identify PDDIs among the drugs used by the elderly as well as their severity and consequences [9] Regarding severity, the interactions were classified as contraindicated when the simultaneous administration of the drugs was not recommended, severe when the drugs were life-threatening and required immediate medical intervention, moderate when the drugs could worsen the patient’s clinical presentation , making it necessary to change the drug therapy, and mild when the patient presented changes in the clinical presentation that did not require changes in drug therapy [9].

Health Status Variables

For the quantification of self-referred diseases, the elderly was given a list of diseases and asked to indicate their diseases that had been diagnosed by their doctors. These diseases were categorized as diabetes, hypertension, obesity, malnutrition, high cholesterol, high triglycerides, osteoporosis, cancer, stroke, myocardial infarction, asthma/bronchitis/respiratory problems, musculoskeletal diseases, depression, memory problems, thyroid disorders, cataracts or other diseases. The number of self-referred chronic diseases was obtained by adding the number of affirmative answers to the above-mentioned question.

To investigate the need for a caregiver, the elderly was asked if they needed someone to help them with their daily activities.

To evaluate hospitalization episodes over the past year, the elderly was asked if they had been admitted to the hospital within the past 12 months. The self-rated health status of the individuals was obtained by asking the elderly to rate their health as excellent/good, regular or poor/bad. In addition, the time-comparative self-rated health status was obtained by asking the individuals the following question: “Compared to one year ago, how would you rate your health in general now?” The answers “somewhat worse” and “much worse” were considered to indicate a worse time-comparative self-rated health status.

Statistical Analysis

The bivariate analysis tested the association between the outcomes variables (use of PIM and PDDIs involving PIMs) and the other study variables by applying Pearson’s Chi-squared test. The level of significance adopted was 5%. The outcomes prevalence’s for the categories of the other variables of interest was also calculated. A multivariate Poisson regression analysis was performed according to a hierarchic analysis model. The first level included sociodemographic variables (gender, age, civil status and education level), the second level included health variables (polypharmacy, self-medication, PDDIs, number of self-referred diseases, need for caregiver and hospitalization in the past year), and the third level included the worsening time-comparative self-rated health and the self-rated health. All variables that had p<0.20 in the bivariate analyzes were included in the multivariate analyzes, and only variables with p≤0.05 remained in the final model.

Results

Of the 934 elderly individuals who participated in the survey, 783 answered all of the questions regarding drugs, showed all of the drugs they used and the respective prescriptions. As a result, these 783 individuals were included in this study, and of these, 65.1% were women, 45.7% were 60 to 69 years of aged, 49.5% were married, and 48.3% had completed primary education (Table 1).

These individuals used 2,846 drugs (mean of 3.63 drugs/elderly individual), and 12.7% of these drugs were PIMs, accounting for a group of 35 different drugs. The prevalence of PIM use among the elderly was 35.4% (277). Among all of the drugs used by the elderly, 665 PDDIs were observed. Of these, 45.1% (300) involved PIMs and 54.9% (365) other drugs, and 10.7% (32) of the interactions were between two PIMs. Moreover, 14.8% (116) of the elderly were exposed to PDDIs involving PIMs, and 4.0% (31) were exposed to at least three PDDIs involving PIMs. The investigation of the severity of these PDDIs revealed that 32.1% were severe, 64.6% were moderate, and 3.4% were mild. The PIMs most commonly involved in the PDDIs were diclofenac, amiodarone, digoxin, nifedipine and amitriptyline; in fact, these five drugs were found to be responsible for 73.7% of the PDDIs involving PIMs.

The most frequent PDDIs involving PIMs were between amiodarone and losartan (12, moderate), digoxin and spironolactone (8, severe), digoxin and furosemide (8, moderate), amiodarone and digoxin (6, severe) and atenolol and nifedipine (6, moderate). The most common effects of these PDDIs were an increase in the exposure to or toxicity of one of the drugs (33.0%), a reduction in the effect of one of the drugs involved in the interaction (29.5%), risk of bleeding (6.9%) and cardiotoxicity and/or QT interval prolongation (6.6%).

We observed associations for the use of PIMs with the practice of polypharmacy (prevalence ratio [PR]= 1.23; confidence interval [95%CI] 1.17-1.29), the presence of three or more PDDIs (PR= 1.13; 95%CI 1.07-1.18), self-referral of at least five diseases (PR= 1.13; 95%CI 1.07-1.18), episodes of hospitalization within the past year (PR= 1.07; 95%CI 1.01-1.12) and a poor/bad self-rated health status (PR= 1.19; 95%CI 1.10-1.28) (Table 1). The variables polypharmacy (PR=1.76; 95%CI 1.42-2.19), one/two PDDIs (PR=1.68; 95%CI 1.35-2.11), three or more PDDIs (PR=1.86; 95%CI 1.44-2.43), and a poor/bad self-rated health status (PR=1.44; 95%CI 1.12-1.86) remained in the final multivariate analysis model and were associated with PIM use (Table 3).

The PDDIs involving PIMs were associated with the practice of polypharmacy (PR=4.90; 95%CI 3.47- 6.91), self-referral of at least five diseases (PR= 2.83; 95%CI 1.63-4.91), episodes of hospitalization within the past year (PR= 1.46; 95%CI 1.01-2.10) and a poor/bad self-rated health status (PR= 2.91; 95%CI 1.80-4.73) (Table 2). The variables polypharmacy (PR=4.46; 95%CI 3.12-6.37) and a poor/bad self-rated health status (PR=1.93; 95%CI 1.21-3.07) remained in the final multivariate analysis model and were associated with PDDIs involving PIMs (Table 3).

Discussion

The list of PIMs according to the Beers criteria has been guiding safe drug prescriptions for the elderly population from its first version to the last update [2]. However, the findings of this and other studies reinforce the prevalent use of PIMs by this population, supporting the need to consider the risk of PDDIs involving these drugs [10-12]. The fact that the elderly constitutes the population that most frequently practices polypharmacy and is most susceptible to PDDIs, as well as the with a corresponding imminent risk, should be taken into account [13]. Approximately half of the PDDIs identified involved PIMs. The use of PIMs was strongly associated with a higher amount of PDDIs. Moreover, the most frequent consequence of these interactions was found to be the potentiation of the risks associated with PIM use. These findings confirm the warnings raised by other studies: the drugs defined by the Beers criteria might be toxic or exert adverse effects in the elderly when administered as monotherapies, and these risks might be intensified when these PIMs are involved in interactions with other drugs [2-12].

Some of the elderly individuals were exposed to an even higher risk due to the existence of interactions between the PIMs being used by the individuals. Almost all of these PDDIs potentiated the risk of use of these drugs with severe or even irreversible consequences, such as cardiac arrest [14]. This finding indicates that the health professionals prescribing the drugs and the pharmacists dispensing the drugs are unaware or do not pay attention to the safety issues involving drug use among elderly patients, suggesting a need for more assertive initiatives regarding training and continuing education for health professionals treating this patients [15].

The interactions were frequently classified as severe The constant monitoring of actions that might reduce the risk of unnecessary damage to a patient to an acceptable number must be noted. In the case of drug therapy, the risk-benefit ratio of the drugs and, in particular, the potential for interactions must be well observed and considered at the time of prescription [16], moreover, safer therapeutic alternatives must be considered [2]. Diclofenac, amiodarone, digoxin and nifedipine were found to be responsible for the largest number of moderate and severe PDDIs involving PIMs. PDDIs associated with a risk of bleeding were found for the combination of diclofenac with calcium channel blockers, Ginkgo biloba, and citalopram. According to Shin, et al. [17] the combination of diclofenac with citalopram might cause an increased risk of intracranial bleeding within 30 days. Moreover, Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) combined with diuretics and anti-hypertensive agents reduce the efficacy of these drugs, there by resulting in compromised blood pressure control [18].

PDDIs of high clinical significance occur with amiodarone primarily due to its cytochrome P450 and P-glycoprotein inhibitory activities [9-19]. When associated with beta-blockers or calcium channel blockers, as was identified in the present study, amiodarone results in a risk of hypertension and might induce severe bradycardia with a risk of cardiac arrest [20].

Digoxin was associated with several PDDIs, and most of these interactions might increase its serum concentration and toxicity (e.g., when used with diuretics (spironolactone, hydrochlorothiazide (HCTZ), or furosemide), amiodarone or fluoxetine). Effects such as bradycardia and arrhythmia with cardiovascular collapse might also occur when digoxin interacts with carvedilol and calcium [1-21].

Nifedipine, was also associated with several PDDIs, including some serious interactions observed in the study, and with carbamazepine and phenytoin, which are drugs with low therapeutic index inducers of CYP3A4, that reduce nifedipine exposure [9-16]. A poor/bad self-rated health status was associated with PIM use and with the presence of PDDIs involving PIMs. These drugs might have a negative impact on the health of the elderly, as reflected by a worsening of the general health status and a negative assessment. The negative contribution related to the use of PIMs and the PDDIs involving these drugs must be investigated [22,23]. However, in clinical practice, the undesired effects of PIMs and the consequences of PDDIs are rarely investigated; in contrast, the practice of adding one more drug to treat a complaint is more frequent, which further complicates the health status of an elderly patient, particularly when PDDIs are present.

The practice of polypharmacy by the elderly was strongly associated with PIM use. This result is similar to the results reported in the literature [24,25], most likely because a PIM is more likely to be included in a prescription containing many drugs. Bao, et al. [25] also highlighted that this association might lead to disease worsening, hospitalizations, and a higher risk of PDDIs. Moreover, these results draw attention to the importance of promoting treatments not involving drugs, such as a balanced diet and physical exercise, whenever possible; these alternatives should be exhausted before prescribing drugs. Finally, the results underscore the promotion of rational drug use by the elderly.

The results of this study have very clear implications and demonstrate the need to reduce PIMs use. These observations may help health managers and health vigilance managers evaluate the quality and safety of the health assistance provided to the elderly population. Some studies have warned about the importance of reducing the use of these drugs and recommended the implementation of safer therapies or the use of lower doses of these drugs whenever possible.

If a prescription containing PIM with potential for interactions is maintained after a rigorous and individualized analysis, the elderly patient must be monitored for dose adjustment, changes in the dosing schedule, control of serum levels of the drugs, and monitoring of clinical conditions, to enable early detection of signs and symptoms of PDDIs [8-24].

In conclusion, based on the findings of this study, more than one-third of the elderly individuals used PIMs, which exposes them to a risk that might exceed the benefits of the drug therapy. Approximately half of the identified PDDIs involved PIMs, and half of those PDDIs had the same effects that deemed the drug potentially inappropriate for use by the elderly as a consequence. PIM use was associated with the practice of polypharmacy, the number of PDDIs, and a poor health self-assessment. The risk of use of these drugs and the severity of the PDDIs involving them are well described but, nonetheless, seem to be neglected in clinical practice. These drugs continue to be frequently prescribed for the elderly.

Acknowledgement

To the coordinator of the city’s Elderly Health Vigilance Network for planning the study’s activities. To all the investigators in the Elderly Health Vigilance Network of Goiânia for participating in the planning and development of the study.

Conflict of Interests

The authors declare no conflicts of interest.

Author Contributions

All authors participated effectively in the development of the study, such as the study design, data collection, and preparation of the manuscript. All approved the final version of the manuscript, and assume responsibility for the integrity of the results.

Sponsor’s Role

The Research Support Foundation of the State of Goiás did not influence the design and performance of the study or the data collection, project management, or data analysis and interpretation. Likewise, it did not influence the study’s conclusions, drafting of the manuscript, or choice of journal for publication.

Funding sources:

State of Goias Research Foundation (FAPEG), Research Program for SUS: Shared Health Management - PPSUS.

References

1. Fick DM, Mion LC, Beers MH (2008) Health Outcomes associated with potentially inappropriate medication use in older adults. Res Nurs Health 31: 42-51.
2. The American Geriatrics Society 2015 Beers Criteria Update Expert Panel. (2015) American Geriatrics Society 2015 Updated Beers Criteria for Potentially Inappropriate Medication Use in Older Adults. J Am Geriatr Soc 63: 2227-2246.
3. Gallagher P, Lang PO, Cherubini A, Topinková E, Cruz-Jentoft A, et al. (2011) Prevalence of potentially inappropriate prescribing in an acutely ill population of older patients admitted to six European hospitals. Eur J Clin Pharmacol 67: 1175-1188.
4. Goodman & Gilman's the pharmacological basis of therapeutics, 12th (ed.) New York: McGraw-Hill, 2011.
5. Hines LE, Murphy JE (2011) Potentially harmful drug-drug interactions in the elderly: a review. Am J Geriatr Pharmacother 9: 364-377.
6. Delafuente JC (2003) Understanding and preventing drug interactions in elderly patients. Critical reviews in oncology/hematology 48: 133-143.
7. Santos TRA, Lima DM, Nakatani AY, Pereira LV, Leal GS, et al. (2013) Medicine use by the elderly in Goiania, Midwestern Brazil. Rev Saude Pub 47: 94-103.
8. Linjakumpu T, Hartikainen S, Klaukka T, Veijola J, Kivelä SL, et al. (2002) Use of medications and polypharmacy are increasing among the elderly. J Clin Epidemiol 55: 809-817.
9. MicromedexR Healthcare Series [database online]. Version 5.1. Greenwood Village, Colo: Thomson Micromedex; 2007. Available at < Accessed January 10, 2016.
10. Gnjidic D, Le Couteur DG, Pearson SA, McLachlan AJ, Viney Re, t al. (2013) High risk prescribing in older adults: Prevalence, clinical and economic implications and potential for intervention at the population level. BMC Public Health 13: 115-124.
11. Kanaan AO, Donovan JL, Duchin NP, Field TS, Tjia J, et al. (2013) Adverse drug events after hospital discharge in older adults: Types, severity, and involvement of Beers criteria medications. J Am Geriatr Soc 61: 1894-1899.
12. Davidoff AJ, Miller GE, Sarpong EM, Yang E, Brandt N, et al. (2015) Prevalence of potentially inappropriate medication use in older adults using the 2012 Beers criteria. J Am Geriatr Soc 63: 486-500.
13. Shi S, Morike K, Klotz U (2008) The clinical implications of ageing for rational drug therapy. Eur J Clin Pharmacol 64: 183-199.
14. Cuentro VD, Andrade MA, Gerlack LF (2014) Prescrições medicamentosas de pacientes atendidos no ambulatório de geriatria de um hospital universitário: estudo transversal descritivo. Ciência & Saúde Coletiva 19: 3355-3364.
15. Hanlon JT, Sloane RJ, Pieper CF (2011) Association of adverse drug reactions with drug-drug and drug-disease interactions in frail older outpatients. Age Ageing 40: 274-277.
16. Correr CJ, Pontarolo R, Ferreira LC (2007) Riscos de problemas relacionados com medicamentos em pacientes de uma instituição geriátrica. Rev Bras Cienc Farm 43: 55-62.
17. Shin JY, Park MJ, Lee SH (2015) Risk of intracranial haemorrhage in antidepressant users with concurrent use of non-steroidal anti-inflammatory drugs: nationwide propensity score matched study. BMJ 351: 3517-3524.
18. Curtis LH, Ostbye T, Sendersky V, Hutchison S, Dans PE, et al. (2004) Inappropriate prescribing for elderly Americans in a large outpatient population. Arch Intern Med 164: 1621-1625.
19. Spriet I, Meersseman W, De Hoon J, von Winckelmann S, Wilmer A, et al. (2009) Mini-series: II. Clinical aspects. Clinically relevant CYP450-mediated drug interactions in the ICU. Intens care medic 35: 603-612.
20. Frishman WH, Sonnenblick EH (1994) Beta-Adrenergic Blocking Drugs, em: Schlant RC, Alexander RW-Hurst’ s The Heart, 8nd (ed.) New York: McGraw-Hill.
21. Boustani M, Campbell N, Munger S (2008) Impact of anticholinergics on the aging brain: A review and practical appllication. Aging Health 4: 311-320.
22. Onder G, Landi F, Cesari M (2003) Inappropriate medication use among hospitalized older adults in Italy: results from the Italian Group of Pharmaco epidemiology in the Elderly. Eur J Clin Pharmacol 59: 157-162.
23. Passarelli MC, Jacob-Filho W, Figueras A (2005) Adverse drug reactions in an elderly hospitalised population: inappropriate prescription is a leading cause. Drugs Aging 4: 767-777.
24. Chen LL, Tangiisuran B, Shafie AA (2012) Evaluation of potentially inappropriate medications among older residents of Malaysian nursing homes. Int J Clin Pharm 34: 596-603.
25. Bao Y, Shao H, Bishop T (2012) Inappropriate medication in a national sample of US elderly patients, receiving home health care. J Gen Intern Med 27: 304-310.