Drug-drug interactions and adverse drug reactions in polypharmacy among older adults: an integrative review 1

ABSTRACT Objective: to identify and summarize studies examining both drug-drug interactions (DDI) and adverse drug reactions (ADR) in older adults polymedicated. Methods: an integrative review of studies published from January 2008 to December 2013, according to inclusion and exclusion criteria, in MEDLINE and EMBASE electronic databases were performed. Results: forty-seven full-text studies including 14,624,492 older adults (≥ 60 years) were analyzed: 24 (51.1%) concerning ADR, 14 (29.8%) DDI, and 9 studies (19.1%) investigating both DDI and ADR. We found a variety of methodological designs. The reviewed studies reinforced that polypharmacy is a multifactorial process, and predictors and inappropriate prescribing are associated with negative health outcomes, as increasing the frequency and types of ADRs and DDIs involving different drug classes, moreover, some studies show the most successful interventions to optimize prescribing. Conclusions: DDI and ADR among older adults continue to be a significant issue in the worldwide. The findings from the studies included in this integrative review, added to the previous reviews, can contribute to the improvement of advanced practices in geriatric nursing, to promote the safety of older patients in polypharmacy. However, more research is needed to elucidate gaps.


Introduction
The world is on the brink of a demographic milestone. In about five years' time, the number of people aged 65 or older will outnumber children under age 5. Driven by falling fertility rates and remarkable increases in life expectancy, population ageing will continue, even accelerate. The number of people aged 65 or older is projected to grow from an estimated 524 million in 2010 to nearly 1.5 billion in 2050, with most of the increase in developing countries (1) .
Ageing, one of the most complex biological phenomena, is a multifaceted process in which several physiological changes occur at both the tissue and the whole-organism level, occurring in cascade, especially post-reproduction (2) . The changes characterized by ageing include: changes in biochemical composition of tissues; progressive decrease in physiological capacity; reduced ability to adapt to stimuli; increased susceptibility and vulnerability to disease and increased risk of death (3) .
Age related chronic diseases such as dyslipidemia, hypertension, diabetes, and depression usually require the use of multiple drugs, a state known as polypharmacy.
This refers to the use of multiple medications and/ or more medications than clinically indicated. It is estimated that more than 40% of adults aged 65 or older use 5 or more medications, and 12% use 10 or more different medications (4). However, the magnitude of the problem among older adults is still scarcely known in most countries.
It is well known in the literature that polypharmacy increases the use of inappropriate drugs, leading to the underuse of essential medicines for the appropriate control of conditions prevalent in the older adults. In addition, it sets up a barrier to treatment adherence in that it creates complex therapeutic regimens, and enables the occurrence of medication errors, drug-drug interactions, adverse reactions, and poor quality of life.
It increases morbidity, mortality, and complexity of care.
It also imposes a huge financial burden on both the older adults and health system (5) .
Furthermore, attention should be paid to the fact that the body of the older adults presents changes in their physiological functions that may lead to a differentiated pharmacokinetics and greater sensitivity to both therapeutic and adverse drug effects (5) .
Pharmacokinetics, pharmacodynamics, and clinical outcomes are affected by a number of patient-specific factors, including age, sex, ethnicity, genetics, disease processes, polypharmacy, drug dose and frequency, social factors, and many other factors (6) .
The scenario above highlights that population ageing is a global phenomenon and the practice of polypharmacy is dangerous for patients, in particular for older adults, because favors the emergence of drugdrug interactions (DDI), adverse drug reactions (ADR), side effects, longer hospital stays, iatrogenic disease and may also lead to complications that induce the patient's death. Thus, the purpose of the present study was to conduct a broader integrative review aimed at identifying and summarizing studies examining both DDIs and ADRs in older adults polymedicated.

Methods
The stages of this integrative review include: problem identification, formulating the appropriate question to be investigated; literature search with selection of articles according to predetermined criteria; data evaluation extracting data from each study summary of results; data analysis and presentation of results (7) . The following describes the steps of the integrative review for this study.
To elaborate the guiding question was applied to the PICO strategy defining population "older adults", intervention "use of multiple medications/polypharmacy" and outcome "occurrence of drug interactions and adverse drug reactions". Thus, the central question of this integrative review was: What is the scientific evidence available, demonstrating the occurrence of drug-drug interactions and adverse drug reactions in older (i.e. ≥60 years of age) polymedicated adults? seek tested interventions that address polypharmacy (9) .
Additionally, studies obtained from primary sources, represented by original scientific articles, surveys that have shown data on the occurrence of DDI and ADR in older adults (≥ 60 years of age), female and male sex, were on multiple medications (polypharmacy) were selected. The following were excluded from this review: The main descriptors adopted in the search strategy for primary studies were: older adults, polypharmacy, drug interactions, adverse drug reactions and aged, combined using the Boolean operators AND and OR.
After searched, all articles were screened by reading their title, abstract and, when necessary, the content briefly, and, thereby, identifying those papers potentially addressing the topic. The selected articles were analyzed initially and in a second stage, they were read in more detail regarding their content. Finally, the selected articles had their data synthesized. To summarize the data of the selected articles and aiming to ensure that all relevant information was extracted, we applied to each study a validated instrument by Ganong(10).
In order to determine the relevance of articles captured in the searched databases, two examiners performed the synthesis of the data of interest independently which was followed by the thematic analysis of the papers. Each item synthesized/recorded in the instrument was filed in Microsoft Word® 2007, generating a database. All disagreements were resolved by discussion.
The results and data analysis are presented in descriptive form.

Results
A total of 409 references were identified and fortyseven were included in the final analysis. For details, see the flow diagram ( Figure 1).   (37) Medication Safety Review Clinic patients Intervention study Report baseline drug-related problem found in outpatient clinic participants, as well as the correlates of having at least one drug-related problem in this population.    (21) Were identified 215 drug combinations with the potential to cause clinically significant interaction, out of which 83.3% had clinical significance C (specified agents may interact in a clinically significant manner; monitoring therapy is suggested), 16.3% clinical significance D (the two medications may interact in a clinically significant manner), and 0.4% clinical significance X (contraindicated combination).

Rahmawati
et al. Indonesia, 2010 (22) Of the 100 cases, 65 % cases had experienced potential DDIs range from 1 to 17. Of total 204 DDIs incidences, 25% were of significance level 1 and 39% of significance level 2. The study showed that the number of potential DDIs increased as the number of medications used per day increased. Geriatric patients taking nine or more medication tended to have more DDIs (6.8±5.5) in comparison to those with one to two medications. Wright et al. Canada, 2011 (23) Of the 7,100 patients admitted to hospital because of hypotension while receiving a calcium-channel blocker, 176 had been prescribed a macrolide antibiotic during either the risk or control intervals. Erythromycin was most strongly associated with hypotension, followed by clarithromycin.

Steinman
et al. U.S, 2011 (24) Over the one-year study period, 126 patients (33%) developed 167 adverse drug events. The risk of adverse drug events was not associated with any of the geriatric conditions, and the strongest factor associated with adverse drug events was the number of drugs added to a patient's medication regimen during the 1-year's study period.

Steinman
et al. US, 2011 (25) Over the 12-month follow-up period, 497 ADRs occurred in 269 participants, including 187 ADRs considered preventable and 127 considered severe. Many geriatric conditions were not associated with risk of ADRs. Santos et al. Brazil, 2011 (26) The 36 possible ADRs found were related to gastrointestinal tract, skin and the nervous system. A total of 63 DDIs were found on patient prescriptions.
(the Figure 3 continue in the next page...) Rev. Latino-Am. Enfermagem 2016;24:e2789  (28) During the 18 year study period, 6,903 admissions for hyperkalemia were identified, 306 of which occurred within 14 days of antibiotic use. Of these, 248 (81%) cases were matched to 783 controls. 10.8% of spironolactone users received at least one prescription for trimethoprim/sulfamethoxazole. Compared with amoxicillin, prescription of trimethoprim/ sulfamethoxazole was associated with a marked increase in the risk of admission to hospital for hyperkalemia. Varallo et al. Brazil, 2011 (29) Were reported 167 different drugs, of which 58 were responsible for 99 ADRs. Of these ADRs, 4 were related to the use of potentially inappropriate medicines, 82 to the use of drugs other than potentially inappropriate medicines and 13 of both categories of drugs (potentially inappropriate medicines and non potentially inappropriate medicines). The majority (57.6%) of ADRs identified were classified as "possible".
Gallagher et al. Ireland, 2011 (30) Prescription of medications with potential for clinically important DDIs was significantly reduced in the intervention group at discharge and during follow-up.
Hanlon et al. U.S, 2011 (31) Among depressed patients who received antidepressants, 43.1% had potential inappropriate use due primarily to problems seen with drug-drug and drug-disease interactions.

Skaar and O'Connor
U.S, 2011 (32) Were identified 25 DDIs of clinical importance in ambulatory settings. The potential serious interactions included aged ≥85 years, annual income >$50,000 and higher numbers of prescriptions: 8-10 and ≥11 drugs.  (34) Among 58,429 older people receiving phenytoin, were identified 796 case patients hospitalized for phenytoin toxicity and 3,148 matched controls. Following multivariable adjustment for potential confounders, were observed a more than doubling of the risk of phenytoin toxicity following the receipt of trimethoprim/sulfamethoxazole. In contrast, were observed no such risk with amoxicillin.
Budnitz et al. U.S, 2011 (35) On the basis of 5,077 cases identified in the sample, there were an estimated 99,628 emergency hospitalizations for adverse drug events in United States. Emergency department visits that resulted in hospitalization, as compared with visits that did not result in hospitalization, were more likely to involve unintentional overdoses (65.7% vs. 45.7%) and five or more concomitant medications (54.8% vs. 39.9%). Four medications or medication classes were implicated alone or in combination in 67.0% of hospitalizations: warfarin (33.3%), insulins (13.9%), oral antiplatelet agents (13.3%), and oral hypoglycemic agents (10.7%). High-risk medications were implicated in only 1.2% of hospitalizations.
Coupland et al. U.K, 2011 (36) 54,038 (89.0%) patients received at least one prescription for an antidepressant during follow-up. Selective serotonin reuptake inhibitors were associated with the highest adjusted hazard ratios for falls and hyponatraemia compared with when antidepressants were not being used. The group of other antidepressants was associated with the highest adjusted hazard ratios for all cause mortality, attempted suicide/self harm, stroke/transient ischemic attack, fracture, and epilepsy/seizures, compared with when antidepressants were not being used. Chan et al. Taiwan, 2012 (37) Problem domain: interactions (2 categories) (12%). Category: potential interaction (12%). Problem domain: adverse reactions (9%).Category: side effect suffered (non allergic) (9%).

Baillargeon
et al. U.S, 2012 (38) Exposure to any antibiotic agent within the 15 days of the event/index date was associated with an increased risk of bleeding. All six specific antibiotic drug classes examined [azole antifungals, macrolides, quinolones, cotrimoxazole, penicillins and cephalosporins] were associated with an increased risk of bleeding.

Somers et al.
Belgium, 2012 (40) The number of drugs was significantly correlated with higher Medication Appropriateness Index scores. Good agreement between the scores of the geriatrician and the clinical pharmacist was found: the κ values of the 8 questions ranged between 0.54 and 0.77 and the overall mean κ value was 0.71. Were found the highest κ values for drug-disease interactions (0.77), dosage (0.74), and ADR (0.74), and the lowest values for directions (0.54) and indication (0.65). Bakken et al. Norway, 2012 (41) The prevalence of potentially inappropriate medicines increased from 24% to 35%; concomitant use of ≥ 3 psychotropic/ opioid drugs and drug combinations including non-steroid anti-inflammatory drugs increased significantly. Serious DDI were scarce both on admission and discharge (0.7%).

Rodrigues MCS, Oliveira C.
A total of 690 drug-related problem per patient were identified (mean=9.9), 393 (57%) were classified as actual drugrelated problem per patient and 297 (43%) were classified as potential drug-related problem per patient. The three most common types identified were: wrong drug (n=151), unnecessary drug therapy (n=136) and adverse drug reaction (n=118).
Therefore, there is a need for a clear cut-off point that defines polypharmacy worldwide. A definition focusing on whether the medication is clinically indicated may be more appropriate (9) than the number of ingested medicines.
Non-adherence to treatment is a common problem in older adults. DDIs and ADRs during hospitalization have been reported to be associated with nonadherence, which are also common among older adults who are discharged from hospital and are using several drugs for their chronic diseases. Studies examining readmissions due to DDIs and consequent ADRs were also performed (44,51,55) . Therefore, early detection and recognition of clinically important interactions by healthcare professionals are vital for monitoring the occurrence of DDIs and ADRs in the continuum of health care.
Older adults usually do better use of medicines when their care is managed by a multidisciplinary team, consisting of a physician (geriatrician), clinical pharmacist and nurse. The involvement of a dentist in this team seems to be relevant, as demonstrated in a study (32) . criteria (11,16,30) in the Ireland and United Kingdom, the Norwegian General Practice criteria (NORGEP) (41,43) , and the instrument Medication Appropriateness Index (MAI) (40,57) have been explored in studies. The studies Thus, the reviewed studies reinforced the notion that polypharmacy is multifactorial and is associated with negative health outcomes, as reported previously in two studies reviewed (8)(9) , and in the article of experts opinion that present information specifically of 12 studies about DDI and ADR (58) .
One aspect noted and that needs to be investigated further, relates to examine how self-medication with over-the-counter drugs and complementary medications contribute to increases the risk of DDIs and ADRs, hospitalization and death of older adults. Other gap noted in literature relates to the methods utilized by primary care providers when assessing polypharmacy.
Additionally, little information is available about the incidence/prevalence of DDIs and ADRs among older adults in developing countries.
So, this review points out the relevance of conducting more studies to explore different aspects, considering the need to develop preventive practices to guarantee the safety of older adults with regard to DDIs and ADRs.

Conclusion and implications for advanced practice in geriatric nursing
This study has identified that early detection and recognition of clinically important DDI and ADR by healthcare professionals are vital to identify patients who are at higher risk for such events and require more cautious pharmacotherapy management to avoid negative outcomes. Thus, the potential risk to DDI and ADR can be managed by professionals with appropriate prescriptions, monitoring, and patient education in the continuum of care of older adults, i.e. through best practices.
In this sense, the professionalization on advanced nursing practice is essential, as a requirement for acquisition of knowledge, skills training and skills for making safe and effective care decisions, for example, aimed at health care of older adults commonly exposed to polypharmacy. Thus, this integrative review can help to increase awareness and discussion, to implement universal health coverage and universal access to health care of the older adults in order to guarantee the quality of care by geriatric nurses.