Potentially inappropriate medication use in institutionalized older adults according to the Beers Criteria

The need for specific care, coupled with new family arrangements, has contributed to the increasing institutionalization of elderly members. The purpose of this study was to evaluate drug use by institutionalized older adults according to Beers Criteria. This prospective, longitudinal study was conducted in the three non-profit long-stay geriatric care institutions of Campo Grande, in the Central-West region of Brazil. All subjects aged 60 years and above on November 2011 were included and followed until November 2012. Eighteen subjects were excluded and the final sample consisted of 133 individuals aged 60 to 113 years. Overall, 212 medications were used at geriatric care institution A, 532 at B, and 1329 at C. Thirty-four drugs were inappropriately prescribed 89 times at geriatric care institution A (41.98%), 49 prescribed 177 times at B (33.27%), and 91 prescribed 461 times at C (34.68%). Statistical differences in the inappropriate drug use were found between genders (p=0.007). The most commonly used potentially inappropriate medication were first-generation antihistamines (15.34%). There was a high frequency in the use of potentially inappropriate medications which can initiate marked side effects and may compromise the fragile health of institutionalized elderly. Thus, adopting the Beers Criteria in prescribing medication contributes to minimize adverse reactions and drug interactions.


INTRODUCTION
Older adults have unique medication requirements as organ functions are reduced by age-related physiological changes.These changes affect the pharmacokinetics and pharmacodynamics of drugs, making it difficult to draw a clear-cut line between risks and benefits of their use in this population (Mangoni, Jackson, 2003;Baldoni et al., 2010).
Drugs absorption may be impaired in elderly patients due to increasing gastric pH.This increase enhances the absorption of alkali drugs and reduces the absorption of acidic drugs.Aging also promotes reduced surface intestinal absorption and lower esophageal sphincter pressure and peristalsis (Baldoni et al., 2010;Ferreira, 2010).
With increasing age, the amount of water in the body decreases, adipose mass increases and lean mass decreases.Thus distribution volume is less for water soluble drugs and greater for liposoluble drugs.Consequently, liposoluble drugs tend to accumulate in adipose tissue, increasing their plasma half-life and period of action, and the risk of adverse effects.Reduced distribution volume for watersoluble may increase their initial concentration in the central compartment, resulting in higher plasma concentrations (Baldoni et al., 2010).Biotransformation can be affected by hepatic blood flow, which can be reduced by as much as half in the elderly.This results in reduced first pass metabolism and increased bioavailability of drugs (Mangoni, Jackson, 2003).There is a decrease in the activity of cytochrome P450 enzymes.Thus, some medications may continue to exert their effects for a longer than expected (Nobrega, Karnikowski, 2005).One example is diazepam, which has a half-life of 24 hours in younger patients and 90 hours in elderly patients (Mangoni, Jackson, 2003).
Renal function is an important parameter for the clearance of pharmaceutical agents, in the elderly; this function progressively declines with advancing age.Reduced renal blood flow, tubular clearance, and creatinine clearance and increase serum creatinine, result in two clinically significant effects -increased half-life and serum levels of drugs (Ferreira, 2010).
For drugs with a narrow therapeutic safety margin, such as digoxin, aminoglycosides and warfarin, serious side effects may occur in elderly patients if a dosage adjustment is not based on creatinine clearance value (Mangoni, Jackson, 2003;Baldoni et al., 2010).
In relation to pharmacodynamics, modifications in the number of drugs, receptor affinity to drugs, and signal transduction, especially for agents that depress the central nervous system such as opioid analgesics, hypnotics and sedatives cause increased sensitivity to these triggered effects (Hutchison, O'Brien, 2007;Ferreira, 2010).
The need for specific care, coupled with new family arrangements, has contributed to increasing institutionalization of elderly family members.Institutionalized older adults differ from those who live with their families as they present specific characteristics such great frailty, impaired functionality, and physical, psychological and social dependence (Gorzoni, Pires, 2006).Comorbidities and chronic degenerative diseases make the elderly prone to using concomitant medications, increasing the risk of adverse events occurrence (Rajska-Neumann et al., 2011;Liu et al., 2012).
The 2000 Population Census of Brazil produced by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatistica -IBGE) revealed less than 100,000 senior citizens living in collective households, this represents less than 1% of the elderly population.Approximately 10 years later, a survey conducted by the Institute for Applied Economic Research (Instituto de Pesquisa Econômica Aplicada -IPEA) identified 3548 long-stay geriatric care institutions in Brazil, where 83,870 seniors citizens were living, 0.5% of the elderly population (IPEA, 2011).
The need for caution in prescribing drugs to older adults has led to the development of a number of tools to assist this practice.One of these resources is the Beers Criteria of potentially inappropriate medications (PIM) use in older adults, developed by Beers et al. (1991).These criteria were established to reduce the risks of iatrogenesis and adverse reactions.In 1997, the criteria were updated by Beers to include degrees of frailty. In 2003, Fick et al. reviewed the original guidelines and published the latest update in the following decade (Fick et al., 2012).Potentially inappropriate medications (PIM) are medications or classes of medications that should be avoided in elderly patients of 60 or more years.Medications are placed in this category when they have no evidence-based indication, do not present costeffectiveness, or there are safer alternatives (Varallo, Capucho, Planeta, 2011;Beers et al., 1991).
The purpose of this study was to identify and quantify drug use in elderly institutionalized adults using the Beers Criteria (2012).

Study design and setting
This prospective, longitudinal study was conducted in the three non-profit long-stay geriatric care institutions (herein designated A, B, and C) of Campo Grande in the Central-West region of Brazil.Non-profit long-stay geriatric care institutions are residential nursing homes for individuals of 60 years and over who have some degree of difficulty in performing daily activities and for those whose families lack the financial, physical, or emotional resources to provide them with the necessary care (ANVISA, 2005).

Study population
Subjects aged 60 years and over (Brasil, 2003) living at one of the non-profit long-stay geriatric care institutions of Campo Grande, Brazil, in November 2011 were included in the study and followed until November 2012.
Those institutionalized after November 2011 were excluded, as were those who died or were discharged from the institutions during the study period.

Data collection
Data was collected weekly from the subjects' medical records; the variables investigated were gender, age, number of prescribed drugs and identification of their active principles, and PIM occurrence according to the Beers Criteria (Fick et al., 2012), (a) potentially inappropriate medications for use in older adults independent of diagnoses or conditions, (b) potentially inappropriate medications for use in older adults considering diagnoses and conditions, and (c) medications that, although potentially inappropriate to older adults, can be used with caution.
Quantification of the total number of used drugs and evaluation of their unsuitability took into account the number of times the same drug was prescribed, so as to identify repeated exposure to PIM.

Data treatment and interpretation
Data were expressed as absolute and relative frequencies, means and standard deviations.Statistical analyses were performed using Epi Info software, version 3.5.1, 20083.5.1, (CDC, 2009) ) and BioEstat, version 5.0 (Ayres et al., 2007).Associations between variables were compared using Pearson's Chi-squared test and prevalence ratio with 95% confidence interval (CI).

Ethical issues
The study was approved by the Federal University of Mato Grosso do Sul Research Ethics Committee (protocol number 2212/2011).
Healthcare professionals responsible for the subjects proposed their treatment and the study evaluations did not interfere with this process.

RESULTS
Table I shows the profile of the study population.During data collection, 18 subjects were excluded (16 died and two left the institutional setting).The final sample thus consisted of 133 individuals aged 60 to 113 years.Mean ages (by institution) were 77.94±7.94years (A), 76.17±11.10 (B), and 78.49±10.15 (C).
Thirty-four drugs were inappropriately prescribed 89 times at A (41.98%), 49 prescribed used 177 times at B (33.27%), and 91 prescribed 461 times at C (34.68%).In all, roughly 35% of the drugs prescribed were classified as PIM and statistical differences were found between genders for inappropriate use of drugs (Pearson's Chisquared test, P = 0.007) (Table II).
All subjects living at geriatric care institutions were treated with PIM, except one (1.33%)older adult from institution C, who did not use any medicine.
Table IV presents data on PIM use in older adults considering diagnoses or conditions.Potential interactions were also observed for PIM use on fewer than 10 occasions: cyclo-oxygenase 2 inhibitors at A (1/1.30%), B (2/1.49%) and C (6/1.80%) and diltiazem at C (1/0.30%), a calcium channel blocker, which is potentially inappropriate for older adults with heart failure and chronic constipation.

DISCUSSION
Our results reveal the existence of risks related to medication use by institutionalized older adults in the three non-profit long-stay geriatric care institutions investigated.By providing a broad overview of PIM use by institutionalized older adults, this investigation encourages the development of mechanisms to evaluate riskminimizing processes, so as to increase the likelihood of positive therapeutic outcomes for the geriatric population.
All used drugs were considered independent of formulation type or administration route, since the Beers Criteria (Fick et al., 2012) does not discriminate between these parameters.The male predominance in the surveyed institutions contrasts with the greater life expectancy pattern seen in women in Brazil (IBGE, 2011) and in other studies on institutionalized older adults (Aguiar et al., 2008;Fochat et al., 2012).However Correr et al. (2007) found elderly males in the majority thus corroborating our finding.Our study also revealed greater use of PIM by males, but interpretation of this finding should take into account that drugs such as doxazosin (an alpha-blockers) PIM use in our study was higher than seen by other investigators in Brazil.This may be because the study was longitudinal and used the updated Beers Criteria, which includes a larger number of PIM. Correr et al. (2007) and Aguiar et al. (2008) found 13.5% and 28.7% rates of PIM use, respectively. In Ireland, O'Mahony et al. (2010) described PIM prescription at 60% in long-stay geriatric care institutions.
Only one subject did not use PIM.This is a worrying finding, given the often compromised pharmacokinetic and pharmacodynamic profile of elderly individuals, which can compound the adverse effects of PIM.On the other hand, potentially inappropriate medication can be used in situations where the benefits outweigh the risks.Lin et al. (2011), applying the Beers Criteria during an investigation of a rural community in Taiwan found that one third of elderly subjects had been prescribed at least one PIM (Fick et al., 2003).On the same island, Liu  (Gallagher et al., 2008).
The high number of prescriptions for firstgeneration anti-histamines, tricyclic antidepressants, and antipsychotics found at the geriatric care institutions investigated raises concerns.Caution is required in prescribing these agents, irrespective of patient clinical condition, as these drugs have a pronounced anticholinergic effect, progressively reduced clearance with advancing age, and increased tolerance when used as hypnotics.They can also increase the risks of confusion, dry mouth, constipation, blurred vision, urinary retention, and tachycardia (DiPiro et al., 2011).
Drugs with pronounced anticholinergic effects are also categorized as PIM for older adults due to drugdisease or drug-syndrome interactions stemming from increased tissue responsiveness and central cholinergic hypofunction and dysfunction in old age and dementia (Bartus, 2000).Kim, Heo, Lee (2010) confirmed these effects while compiling a list of potentially inappropriate medications for elderly Koreans, with the aid of the Delphi method (Ablah et al., 2013).
First-and second-generation antipsychotics increase the risks of stroke and cardiac QT-interval prolongation by altering the electrical properties of cardiac cells and causing hypotension (Risch, Groom, Janowsky, 1982).They can also trigger symptoms of Parkinson's by antagonizing dopamine receptors (Holt, Schmiedl, Petra, 2010).In this category, olanzapine and thioridazine should not be prescribed to patients with seizures, as these drugs reduce the neuronal excitability threshold (Muench, Hamer, 2010), they are PIM that can be prescribed with caution, owing to their potential to trigger or exacerbate the syndrome of inappropriate antidiuretic hormone secretion and hyponatremia, requiring that sodium levels be monitored when initiating the drug regime or changing doses.Monitoring is also necessary with the use of carbamazepine, mirtazapine, serotonin-noradrenaline reuptake-inhibiting antidepressants, and selective serotonin reuptake inhibitors (Fick et al., 2012).
Metoclopramide, an antiemetic and prokinetic drug with antidopaminergic action, can cause extrapyramidal side effects, including tardive dyskinesia and Parkinsonian symptoms, which contraindicate its use unless indicated for gastroparesis (Kim, Heo, Lee, 2010).
Non-steroidal anti-inflammatory drugs (NSAID) were also prescribed, despite their enhanced side effects in elderly patients, which include gastrointestinal bleeding, ulcer induction, kidney failure, high blood pressure, and cardiovascular changes.Because NSAID promote fluid retention and exacerbate heart failure by decreasing prostaglandin production, they must not be prescribed to patients with heart failure (Sostres, Gargallo, Lanas, 2009).
Among drugs for blood pressure control, alphablockers can increase the risks of syncope related to bradycardia and orthostatic hypotension.Methyldopa, a central-acting alpha agonist antihypertensive that decreases sympathetic discharge, is inappropriate given its potential to exacerbate depression (Kim, Heo, Lee, 2010).Immediate-release nifedipine, a calcium-channel blocker, induces hypotension and vasodilatation, lowers systemic vascular resistance, and compromises O 2 demand by myocardial cells (Kowey et al., 2000;Kim, Heo, Lee, 2010).Direct vasodilators can exacerbate syncope episodes in patients with a history of this condition, and should therefore be used with caution (Kim, Heo, Lee, 2010;Fick et al., 2012).
Among antiarrhythmics categorized as PIM, amiodarone is associated with toxicity, thyroid disorders, and cardiac QT prolongation and should therefore be avoided as first-line treatment for atrial fibrillation.Diltiazem can exacerbate heart failure by promoting urinary retention (Kowey et al., 2000;Gallagher et al., 2008).
The use of digoxin at doses higher than 0.125 mg/day for heart failure is not associated with additional benefits and may increase the risk of toxicity, as renal clearance is slower in the elderly and this drug has a low therapeutic index (Winit-Watjana, Sakulrat, Kespichayawattana, 2008;Mangoni, Jackson, 2003;Baldoni et al., 2010;Pinto, Ferre, Pinheiro, 2012).
Restrictions are also placed on hypnotics and sedatives, particularly barbiturates, given the high rates of physical dependence, tolerance to sleep benefits, and risk of toxicity even at low doses (Holt, Schmiedl, Petra, 2010;Fick et al., 2012).
In Brazil, flunitrazepam, nitrazepam, and bromazepam are prescribed and were included in this investigation.These drugs were not considered by the Beers Criteria because there is no record of them at the Food and Drug Administration (FDA, 2013).
The non-benzodiazepine hypnotic eszopiclone, indicated for treatment of insomnia, is categorized as a PIM by the Beers Criteria, regardless of patient clinical condition.Zopiclone, a chiral drug used in racemic form and exhibiting pharmacological activity related to the eszopiclone enantiomer (Zuo et al., 2013), is also classified as a PIM and should be avoided for chronic use because of its agonistic properties toward benzodiazepine receptors and side effects similar to these (Holt, Schmiedl, Petra, 2010).
No evidence of effectiveness has been found for Ergot mesylates, prescribed for headaches, memory impairment, cerebrovascular disease, and peripheral vascular disorders (Holt, Schmiedl, Petra, 2010).
Mineral oil by oral route should be avoided, since it reduces the cough reflex and increases the risk of aspiration and lipid pneumonia (Albuquerque Filho, 2006).
Pioglitazone (an antidiabetic agent) and cilostazol (a vasodilator) are considered potentially inappropriate medication for the elderly by reason of drug-disease or drug-syndrome interactions capable of worsening the disease or syndrome in patients with heart failure by exacerbating this condition while promoting urinary retention (Fick et al., 2012).
PIM to be prescribed with caution include aspirin, used for primary prevention of cardiac events-despite a lack of evidence of benefits over risks in individuals aged 80 years and older, requires monitoring to prevent hemorrhagic events (Gallagher et al., 2008).
We concluded that there was a high frequency of potentially inappropriate medication use which can initiate marked side effects such as hypotension, constipation, extrapyramidal effects, sedation, weakness, renal failure, sleep disorders and can compromise the fragile health of institutionalized elderly individuals.Therefore adopting the Beers Criteria in prescribing could contribute to minimize adverse reactions and drug interactions.
Awareness is required by all healthcare professionals with regard to changes experienced in old age.Drug prescription should address changes in pharmacokinetics and pharmacodynamics taking place during the aging process, so as not to compromise the health status of elderly individuals by inappropriate prescription.To improve care therapy in the elderly, safer alternatives should be sought within the same therapeutic class, there should be greater disclosure of lists like this between prescribers; and PIM lists or evaluation tools should be developed, which are appropriate to Brazil.

FIGURE 1 -
FIGURE 1 -Classification of potentially inappropriate medication to be used with caution in older adults.

TABLE II -
Potentially inappropriate medication for the elderly

TABLE I -
Characteristics of the study population M: Male; F: Female