PDF
Citation
Print
INTRODUCTION
Global population aging is a principal driver of the increased demand for healthcare services and necessitates pharmacological interventions.1 This phenomenon is pronounced in developed nations, where the health of the geriatric population is receiving increased attention due to a rise in hospital visits and the demand for various drug therapies.23 In the United States, individuals aged ≥ 65 years constituted approximately 16.8% of the total population in 2020, and this is predicted to increase to 22% by 2050,4 leading to an increased prevalence of chronic diseases. The 2018 National Health Interview Survey reported that over half (51.8%) of US adults had at least one diagnosed chronic condition, such as arthritis, cancer, or diabetes. Furthermore, 27.2% had multiple chronic conditions, highlighting the substantial comorbidity burden in this population.5 The prevalence of complex chronic conditions among older adults, including cardiovascular disease, hypertension, diabetes, arthritis, and cancer, often requires polypharmacy for optimal management.6
Polypharmacy is defined as the use of ≥ 5 medications at discharge or ≥ 10 medications during hospital admission.7 This includes the use of appropriate medications based on current evidence-based medicine and the use of inappropriate or clinically nonbeneficial medications. Such extensive medication use significantly impacts older adults, contributing to decreased adherence to treatment plans, increased potential for drug interactions, medication errors, and preventable adverse drug events (ADEs), culminating in a range of health complications.8 A hospital-based evaluation indicated that 91% of patients experiencing ADEs were subjected to polypharmacy, in contrast to 73% in the non-ADE group.9 This was particularly evident in older patients, who presented a higher incidence of ADEs and were on an average of more medications compared to their younger counterparts.
The intersection of polypharmacy and patient health outcomes, particularly in aging populations, is currently a critical area of clinical research. The relationship between polypharmacy and its potential hazards is complex, especially among older adults.101112 Of particular concern are elevated risks of fainting and accidental falls in older patients on polypharmacy.1314 An analysis of National Health Insurance data from Korea for 2010–2019 revealed a notable prevalence of polypharmacy (41.8% for periods > 90 days and 38.1% for periods > 180 days) and hypermedication (14.4% for periods > 90 days and 9.4% for periods > 180 days) among outpatients aged ≥ 65 years.15 Similarly, an examination of the Japanese Adverse Drug Reaction Reports database highlighted an increased susceptibility to renal dysfunction among elderly individuals prescribed > 10 medications, with a reporting odds ratio of 1.82, indicating a substantially increased risk.16 Further supporting these findings, a retrospective observational study on 7.36 million elderly patients in South Korea demonstrated significant prevalence of continuous polypharmacy (47.8% for ≥ 90 days and 36.9% for ≥ 180 days) and hyper-polypharmacy (11.9% for ≥ 90 days and 7.1% for ≥ 180 days), with factors like male sex, older age, insurance status, and comorbidities including cardio-cerebrovascular disease and diabetes increasing the likelihood of polypharmacy, highlighting a critical need for policy interventions to reduce polypharmacy in this population.17
Despite previous research highlighting the perils of polypharmacy and its associated side effects in the elderly, there remains a gap in understanding whether the occurrence of polypharmacy is primarily due to chronic diseases or the aging process itself. Moreover, exploration of the differential impact of ADEs contributing to polypharmacy between older and younger individuals is inadequate. Thus, this study analyzed voluntary ADE reporting data to determine whether ADEs linked to polypharmacy were indeed more frequently reported among older adults compared with younger individuals and to discern differences in ADE patterns across these age groups. Identifying age-specific ADE patterns associated with polypharmacy may support the development of targeted strategies in clinical practice, enabling a proactive approach to managing and mitigating these risks in both older and younger adult populations.
METHODS
Data source and preprocessing
This study used South Korean spontaneous ADE reports collected from the Korea Institute of Drug Safety and Risk Management (KIDS)-Korea Adverse Event Reporting System (KAERS) Database (KIDS-KD). The reporting system was implemented in 1988 and is currently managed by KIDS. ADE reports are submitted to KIDS by medical doctors, Korean traditional medicine doctors, dentists, pharmacists, lawyers, or patients.18 Drugs mentioned in the ADE report are recorded using M codes, a drug ingredient code used by the Korean Ministry of Food and Drug Safety for drug approval and adverse event reporting. ADEs are recorded using the Medical Dictionary for Regulatory Activities (MedDRA) lowest-level terminology. KIDS-KD includes the following information: patient demographics, disease history, ADE, basic drug information, drug ingredient, drug dose, drug indication, drug and ADE causality assessment, and the presence of follow-up ADE reports.
We initially obtained 2,038,309 ADE reports submitted to KAERS from January 1, 2012 to December 31, 2021. The name of the dataset with these ADE reports is KIDS KAERS DB (2206A0046). The M codes in the KIDS-KD drug ingredient table (DRUG1.txt) were mapped to active drug ingredient (drug) names using the M code-drug ingredient linkage table (file name: OpenData_E2bIngrListC20220914.csv, downloaded on September 14, 2022 from https://nedrug.mfds.go.kr/pbp/CCBGA01/getItem?infoName=%EC%84%B1%EB%B6%84&totalPages=8&limit=10&searchYn=true&page=1&&openDataInfoSeq=42).19 This table provides drug codes as M codes and corresponding Korean and English drug names used for presenting drug ingredients in spontaneous ADE reports. The codes are updated daily by the Korean Ministry of Food and Drug Safety. The mapping output was manually examined by two experts to extract only the active drug ingredient name, resulting in 3,067 distinct drugs.
The initially collected ADE reports were filtered as follows. First, reports removed by the reporter or were not the final version were excluded Then, ADE reports with input errors for ADE start date, and drug administration start and end dates were excluded. Examples of these dates included February 30, June 31, and November 31. Afterwards, reports missing data on patient age, gender, ADE name and start date, and drug name, administration start and end date were removed. Redundant reports were ruled out based on patient age, gender, ADE information (event name, start date, and severity), and prescribed drugs (name and administration date). Reports with ADE start dates preceding the suspected causal drug administration start date were also removed. Finally, reports for patients aged < 20 years were excluded. The final number of ADE reports after filtering was 793,498 (Fig. 1).
Data analysis
The association between polypharmacy and severe ADEs in different age groups was examined by categorizing the ADE reports by patient age at the time of ADE occurrence according to 10-year age groups as well as older adults ≥ 60 years and younger adults < 60 years to examine age-related characteristics of ADEs and binarizing as older adults (aged ≥ 65 years) and younger adults (aged 20–64 years) to examine age group-related associations between polypharmacy and severe ADE risk. Polypharmacy was determined according to the maximum concurrent number of drugs. We defined polypharmacy as the concomitant use of ≥ 5 drugs. The time point of polypharmacy was when the maximum number of concomitantly used drugs by the patient was ≥ 5 according to the ADE report. Consequently, the actual date of polypharmacy differed for each report. The duration of polypharmacy was the period of time the patient used this maximum number of drugs. The threshold of 5 drugs or more for polypharmacy is a criterion used in several studies.20212223 Severe ADEs were defined as those resulting in death, life-threatening situations, hospitalization, sustained or significant disability, congenital anomalies, or other symptoms, including drug dependence and abuse.18 The rate of severe ADEs among overall ADEs reported were calculated by using the number of ADE reports as the denominator and the number of severe ADE reports as the nominator. This was calculated using ADEs reported across ages ≥ 20 years or for ADE reports of patients grouped by 10 years of age. To determine the physiological systems affected by polypharmacy, severe ADEs were classified into their respective MedDRA System Organ Classes (SOCs), and each was examined for its association with polypharmacy. For ADEs in MedDRA SOCs that were significantly associated with polypharmacy and showed a significantly different association between older and younger adults, the individual severe ADEs included in the MedDRA SOC were examined for their association with polypharmacy.
Statistical analysis
The gender distribution, number of concomitant drugs, severe ADE report frequency, and reporter type were characterized for the overall, older adult, and younger adult populations. Additionally, the reporting rate of ADEs was examined and compared between older and younger adults according to the MedDRA SOCs. The association between polypharmacy and any severe ADEs, MedDRA SOC-stratified, and individual severe ADEs were examined using disproportionality analysis. Specifically, we classified ADE reports by polypharmacy exposure and non-exposure groups. Afterwards, the proportion of severe ADE reports out of all ADE reports in each group were obtained and its ratio and 95% confidence interval (95% CI) was calculated. The calculated ratio was the proportional reporting ratio (PRR) of severe ADE reports in polypharmacy to non-polypharmacy ADE reports. The significance of the difference in severe ADE report proportions in polypharmacy and non-polypharmacy groups were evaluated using the calculated 95% CI and χ2 test. The PRR formula used is specified in equation (1).
where a = number of severe ADE reports where patients used ≥ 5 concomitant drugs, b = number of all ADE reports where patients used ≥ 5 concomitant drugs, c = number of severe ADE reports where patients used < 5 concomitant drugs, and d = number of all ADE reports where patients used < 5 concomitant drugs
The number of ADE reports in each age group and polypharmacy presence combination had to be ≥ 5 to use the χ2 test. The interaction between age and polypharmacy was examined using Poisson regression to compare the strengths of association between these variables in older and younger adults. Furthermore, the number of ADE reports, as well as frequently reported drugs, ADE symptoms, and drug-ADE symptom pairs overall and in severe ADE reports were examined using 10-year age groups to determine ADE report characteristics across the various age groups. P values < 0.05 were considered significant. All statistical analyses were performed using R Statistical Software (version 4.0.2).24
RESULTS
ADE report characteristics
During the study period, 2,038,309 ADE reports were submitted to KAERS. After applying the exclusion criteria, the final number of study ADE reports was 793,498 (Fig. 1). In detail, 104,810 reports were removed due to deletion or replacement with updated ADE reports by the reporter, 1,031,761 due to missing data, and 10,533 due to duplication. Report deduplication was performed by removing all reports except one with identical patient age, gender, ADE name, start date, and severity status, and drug name, administration start, and end dates. The mean age of subjects with ADE reports was 56.30 years (standard deviation [SD] = 16.07), and there was a higher number of females (n = 469,739, 59.20%) than males (n = 323,759, 40.80%). The characteristics of reports on older adults (n = 264,763 (100%), number of females = 144,454 (54.56%), number of polypharmacy ADE reports = 6,819 (2.58%), number of severe ADE reports = 16,376 (6.19%) and younger adults (n = 528,735 (100%), number of females = 325,285 (61.52%), number of polypharmacy ADE reports = 14,132 (2.67%), number of severe ADE reports = 24,230 (4.58%) were examined (Table 1). The demographics of patients having ADE reports with polypharmacy (reports, n = 20,951) showed that 13,048 (62.27%) reports were for females, 6,819 (32.55%) were for older adults of ≥ 65 years, and 1,965 (9.38%) for severe ADEs (Supplementary Table 1).
Table 1
Demographics of subjects with reported ADEs
Overall, ADEs of the gastrointestinal disorders MedDRA SOC were reported most often (40.37% of all study ADE reports), followed by skin and subcutaneous tissue disorders (27.81%) and nervous system disorders (17.27%). This trend persisted in the older adult population (38.45%, 21.21%, and 15.40%, respectively) and the younger adult population (41.33%, 31.11%, and 18.21%, respectively) (Table 1).
Tables 2-4 and Supplementary Tables 2, 3, 4, 5, 6 show the prevalence of suspected drug, ADE, suspected drug-ADE pairs, overall drug-ADE pairs for all and severe study ADE reports, and Supplementary Table 7 shows the prevalence of ADEs for polypharmacy ADE reports by 10 year age groups. In addition, the prevalence of suspected drug, ADE, and suspected drug-ADE pairs with ≥ 5 ADE reports for adults 20–59 years of age and ≥ 60 years of age were examined through Supplementary Tables 8, 9, 10 and 11, 12, 13 using overall and severe ADE reports. The prevalence of ADEs with ≥ 5 ADE reports were also examined using polypharmacy ADE reports in adults 20–59 years of age and ≥ 60 years of age in Supplementary Table 14. The most frequently reported drugs for overall ADE reports in adults aged ≥ 20 years were the analgesics tramadol (n = 81,362, 10.25%), fentanyl (n = 67,318, 8.48%), and acetaminophen (n = 32,897, 4.15%) (Table 2). Patients aged ≥ 60 years more frequently reported ADEs associated with irinotecan (n = 5,327, 1.49% in patients ≥ 60 years of age versus n = 3,363, 0.77% in patients 20–59 years of age), gemcitabine (n = 5,175, 1.45% versus n = 2,135, 0.49% in 20–59 years), and furosemide (n = 6,559, 1.83% versus n = 2,478, 0.57% in 20–59 years) compared to younger patients 20–59 years of age, while patients 20–59 years reported a relatively higher frequency for metoclopramide (n = 5,517, 1.27% versus n = 1,998, 0.56% in ≥ 60 years), ketorolac (n = 5,936, 1.36% versus n = 1,932, 0.54 in ≥ 60 years), and cefazedone (n = 5,478, 1.26% versus n = 2,133, 0.60% in ≥ 60 years) compared to patients ≥ 60 years of age (Supplementary Table 8). The most frequently reported ADEs across all age groups were nausea (n = 200,596, 25.28%), vomiting (n = 86,042, 10.84%), dizziness (n = 73,689, 9.29%), itching (n = 73,453, 9.26%), and urticaria (n = 65,940, 8.31%) (Table 3). Adults aged ≥ 60 years more frequently reported anorexia (n = 4,007, 1.12% in 60 years versus n = 2,424, 0.56% in 20–59 years), hypokalemia (n = 5,958, 1.66% versus n = 2,069, 0.48% in 20–59 years), and thrombocytopenia (n = 3,933, 1.10% versus n = 1,999, 0.46% in 20–59 years) than 20–59 year adults, whereas those aged 20–59 years more frequently reported angioedema (n = 5,868, 1.35% in 20–59 years versus n = 1,666, 0.47% in ≥ 60 years), skin eruption (n = 4,159, 0.95% versus n = 3,171, 0.89% in ≥ 60 years), and palpitation (n = 3,376, 0.78% versus n = 1,621, 0.45% in ≥ 60 years) than adults ≥ 60 years (Supplementary Table 9). The most frequent suspected drug–ADE pairs reported across age groups were tramadol–nausea (n = 48,304, 6.09%) or –vomiting (n = 24,331, 3.07%), fentanyl–nausea (n = 43,845, 5.53%) or –dizziness (n = 15,481, 1.95%), and acetaminophen–nausea (n = 13,368, 1.68%) (Table 4). Pairs more often reported in older adults aged ≥ 60 years compared to younger adults 20–59 of years age included furosemide–hypokalemia (n = 5,652, 1.58% in ≥ 60 years versus n = 1,869, 0.43% in 20–59 years), propacetamol–hypotension (n = 2,062, 0.58% versus n = 808, 0.19% in 20–59 years), and tramadol-constipation (n = 1,256, 0.35% versus n = 707, 0.16% in 20–59 years), whereas morphine-nausea (n = 3,955, 0.91% in 20–59 years versus n = 1,692, 0.47% in ≥ 60 years), iopromide-itching (n = 2,836, 0.65% versus n = 1,721, 0.48% in ≥ 60 years), and oxycodone-dizziness (n = 2,456, 0.56% versus n = 1,285, 0.36% in ≥ 60 years) were more frequently reported in younger adults than older adults (Supplementary Table 10).
Table 2
Most frequently reported 20 drugs suspected to be associated with ADEs by 10-year age groups (number of drugs = 1,740)
Table 3
Most frequent 20 ADEs with a suspected drug reported by 10-year age groups (number of ADEs = 3,910)
Table 4
Most frequently reported 20 suspected drug–ADE pairs by 10-year age groups (number of pairs = 108,390)
Overall, the most frequently reported drugs for severe ADEs were cisplatin (n = 1,573, 0.20%), paclitaxel (n = 1,282, 0.16%), and oxaliplatin (n = 1,181, 0.15%) (Supplementary Table 3). Severe ADE reports associated with apixaban (n = 353, 0.10% in ≥ 60 years versus n = 25, 0.01% in 20–59 years), piperacillin (n = 348, 0.10% versus n = 184, 0.04% in 20–59 years), tazobactam (n = 344, 0.10% versus n = 181, 0.04% in 20–59 years), and decitabine (n = 406, 0.11% versus n = 102, 0.02% in 20–59 years) were reported at a higher frequency for adults aged ≥ 60 years compared to younger adults 20–59 years of age, while severe ADE reports for younger adults were more frequently associated with amoxicillin (n = 349, 0.08% in 20–59 years versus n = 141, 0.04% in ≥ 60 years), ketorolac (n = 324, 0.07% versus n = 164, 0.05% in ≥ 60 years), and ibuprofen (n = 282, 0.06% versus n = 96, 0.03% in ≥ 60 years) in comparison to adults ≥60 years (Supplementary Table 11). The most frequently reported severe ADEs were hypotension (n = 5,843, 0.74%), dyspnea (n = 4,981, 0.63%), and anaphylaxis (n = 3,061, 0.39%) for the total population (Supplementary Table 4). Increased blood creatinine (n = 573, 0.16% in ≥ 60 years versus n = 245, 0.06% in 20–59 years), pneumonia (n = 474, 0.13% versus n = 136, 0.03% in 20–59 years), and anemia (n = 369, 0.10% versus n = 281, 0.06% in 20–59 years) were reported in adults aged ≥ 60 years more frequently than in adults of 20–59 years; and reports for abdominal pain (n = 510, 0.12% in 20–59 years versus n = 245, 0.07% in ≥ 60 years), angioedema (n = 1,142, 0.26% versus n = 308, 0.09% in ≥ 60 years), and chest discomfort (n = 464, 0.11% versus n = 258, 0.07% in ≥ 60 years) were more frequent in adults of 20–59 years than in adults ≥ 60 years of age (Supplementary Table 12). The suspected drug–severe ADE pairs reported most frequently in the total population were propacetamol–hypotension (n = 955, 0.12%), fentanyl-hypotension (n = 692, 0.09%), and cefaclor–anaphylaxis (n = 530, 0.07%) (Supplementary Table 5), and those more frequently reported in populations aged ≥ 60 years than in 20–59 years of age were acetaminophen-hypotension (n = 180, 0.05% in ≥ 60 years versus n = 99, 0.02% in 20–59 years), midazolam-hypotension (n = 125, 0.03% n = 39 versus 0.01% in 20–59 years), and glimepiride–hypoglycemia (n = 56, 0.02% versus n = 6, 0.00% in 20–59 years) (Supplementary Table 13). Ibuprofen-dyspnea (n = 139, 0.03% in 20–59 years versus n = 24, 0.01% in ≥ 60 years) or –angioedema (n = 81, 0.02% versus n = 6, 0.00% in ≥ 60 years) and acetaminophen-dyspnea (n = 122, 0.03% versus n = 40, 0.01% in ≥ 60 years) were reported more frequently in adults aged 20–59 years than in adults aged ≥ 60 years (Supplementary Table 13).
The most frequent ADE symptoms reported with polypharmacy, where ≥ 5 drugs were concomitantly used by the patient, were nausea (n = 5,652, 0.71%), vomiting n = 2,035, 0.26%), and dizziness (n = 1,871, 0.24%) (Supplementary Table 7). ADEs of higher reporting frequency in older adults ≥ 60 years than younger adults of 20–59 years of age included hypotension (n = 243, 0.07% in older, n = 135, 0.03% in younger adults), neutropenia (n = 230, 0.06% in older, n = 191, 0.04% in younger adults), and fever (n = 199, 0.06% in older, n = 194, 0.04% in younger adults), while ADEs of higher frequency in younger adults than older adults presented as itching (n = 675, 0.15% in younger, n = 459, 0.13% in older adults), rash (n = 556, 0.13% in younger, n = 370, 0.10% in older adults), and urticaria (n = 575, 0.13% in younger versus n = 264, 0.07% in older adults) (Supplementary Table 14).
The number of ADE reports in age groups per 10 years of age was 53,984, 81,527, 122,007, 177,985 174,961, 134,512, and 48,522 for those aged 20–29, 30–39, 40–49, 50–59, 60–69, 70–79, and ≥ 80 years, respectively (Fig. 2). The proportion of severe ADEs among all ADEs reported for each age group consistently increased with older age starting from the 30s age group (Fig. 2). The mean number of drugs used by the patient in the ADE report was 1.46 (SD = 1.46). There were 40,606 severe ADE reports, comprising 5.12% of all study reports. The overall number of ADE reports involving polypharmacy (≥ 5 drugs used by the patient) was 20,951 (2.64% of study reports) of which 1,965 (9.38%) were severe ADEs (Table 1, Fig. 2).
Association between polypharmacy and severe ADEs
Polypharmacy was significantly associated with severe ADE reporting (PRR, 1.88; 95% confidence interval [CI], 1.80–1.96; χ2
P < 0.001), and this was observed across all age groups. In older adults, this association had a PRR of 2.07 (95% CI, 1.94–2.21) and a PRR of 1.75 (95% CI, 1.66–1.86) in younger adults. The strength of this association was significantly higher in the older than in the younger population (Poisson regression P < 0.001; Fig. 3). The association between polypharmacy and severe ADE reporting was analyzed by each SOC, revealing that older adults had a stronger association with ADEs of cardiac, renal and urinary MedDRA SOCs than younger adults, whereas younger adults had a stronger association with ADEs of skin and subcutaneous disorder MedDRA SOCs (Fig. 4).
Fig. 3
Association between polypharmacy and severe ADEs. Effect of age and polypharmacy interaction on severe ADE risk was statistically significant at P < 0.001. A proportional reporting ratio > 1 indicates that polypharmacy is associated with an increased risk of severe ADEs.
ADE = adverse drug event, PRR = proportional reporting ratio, CI = confidence interval.
Fig. 4
MedDRA SOC-grouped severe ADEs significantly associated with polypharmacy. Shown are those where the strength of the association differed significantly between subjects aged ≥ 65 and < 65 years. The interaction P between age and polypharmacy obtained with Poisson regression are shown in parenthesis. A proportional reporting ratio > 1 indicates the association of polypharmacy with an increased risk of severe ADEs in the MedDRA SOC.
MedDRA SOC = Medical Dictionary Regulatory Activity System Organ Class, ADE = adverse drug event, PRR = proportional reporting ratio, CI = confidence interval.
Finally, analysis of associations between polypharmacy and severe ADEs included in MedDRA SOCs significantly associated with polypharmacy showed significantly distinct patterns in older and younger adults (Fig. 5). There was a significant association between polypharmacy and urticaria, an ADE of the skin and subcutaneous disorder MedDRA SOC in younger adults (PRR, 2.03; 95% CI, 1.62–2.54). In contrast, older adults on polypharmacy did not have a significantly higher risk of this ADE (PRR, 0.56; 95% CI, 0.23–1.34). The risk of urticaria was significantly higher with polypharmacy in younger adults (Poisson regression P = 0.005). Similarly, younger adults showed an increased likelihood of itching (PRR, 2.06; 95% CI, 1.61–2.63), whereas older adults exhibited an PRR of 0.87 (95% CI, 0.45–1.69). This different trend of association between the different age groups was significant (Poisson regression P = 0.017). Regarding individual ADEs in cardiac disorder MedDRA SOCs, younger adults reported a non-significant association with cardiac arrest (PRR, 2.14; 95% CI, 0.78–5.87), and older adults showed a significantly stronger association (PRR, 12.13; 95% CI, 7.03–20.94). Acute renal failure, a renal and urinary disorder MedDRA SOC, showed a low risk with polypharmacy in younger adults although not significant (PRR, 0.51; 95% CI, 0.07–3.69), whereas the risk was significantly increased in older adults (PRR, 6.37; 95% CI, 3.81–10.64).
Fig. 5
Severe ADEs in the MedDRA SOCs where the degree of association with polypharmacy was significant. Shown are those where the strength of association differed significantly between subjects aged ≥ 65 and < 65 years. The interaction P between age and polypharmacy obtained with Poisson regression are shown in parenthesis. A proportional reporting ratio > 1 indicates the association of polypharmacy with an increased risk of severe ADEs.
ADE = adverse drug event, PRR = proportional reporting ratio, CI = confidence interval, MedDRA SOC = Medical Dictionary Regulatory Activity System Organ Class.
DISCUSSION
Our investigation of 793,498 spontaneous ADE reports from 2012 to 2021 yielded significant insights into the interplay between polypharmacy and ADEs across different age groups. We identified complex patterns of medication management, particularly in older adults. Our findings revealed a trend where the overall frequency of reported ADEs diminished with advancing age, although the incidence of serious adverse reactions escalated, implying that while older adults may experience fewer ADEs in general, they are more likely to be severe. Specifically, among the 20,951 ADE reports involving polypharmacy, 9.38% were classified as severe, which was significantly higher than the proportion of severe ADEs in reports not involving polypharmacy at 5.00%, highlighting the grave implications of simultaneous prescription of multiple medications. Our statistical analysis further strengthens the argument against polypharmacy, particularly in older adults. The association between polypharmacy and severe ADE reporting is undeniable, with the χ2 test yielding a P < 0.001 and a PRR that underscores a higher risk in older compared with younger adults (Supplementary Data 1). This differential risk profile (PRR 2.07 vs. 1.75 for older and younger adults, respectively) indicates an age-related vulnerability necessitating a careful reevaluation of prescribing practices for older adults (Supplementary Data 1). However, it is important to note that our dataset does not distinguish general population sizes across age groups, limiting our ability to draw definitive conclusions about the frequency of ADEs by age. While the trends we observe are suggestive, they should be interpreted with caution given these limitations.
Our study delineates specific ADEs that exhibit age-dependent discrepancies. Older patients of 60 years of age or more tend to show a higher frequency of ADEs associated with chemotherapy or cardiovascular drugs compared to younger patients of 20–59 years of age. This was the case for overall as well as severe ADE reports. Younger patients showed a trend of reporting ADEs from anti-inflammatory and non-opioid analgesics more frequently in comparison to older patients. The ADE reporting trend across ages in polypharmacy ADE reports was similar with that in overall study ADE reports where nausea, vomiting, and dizziness were most frequently reported for both types of reports. Comparing the most frequently reported 20 ADEs between older and younger age groups, a difference in trend was observed. Older patients showed a higher reporting tendency for symptoms related to the immune system (neutropenia, fever), and younger patients a tendency to report skin reactions (itching, rash, urticaria).
Furthermore, younger adults showed a pronounced association with skin-related ADEs, such as urticaria and itching, which were notably diminished or inverted in the older cohort. Conversely, severe outcomes, like cardiac arrest and acute renal failure, were markedly more common in older adults, particularly those on polypharmacy regimens. The stark contrast in PRRs between age groups for these severe conditions emphasizes the criticality of age-specific prescribing guidelines to mitigate the heightened risks associated with polypharmacy in older adults.
Frequently used drug classes, including nonsteroidal antiinflammatories and proton pump inhibitors, increase the risk of cardiovascular adverse events.2526 Ozenberger et al.27 reported that opioid analgesics and corticosteroids increased the risk of major adverse cardiovascular events (MACE), including myocardial infarction, coronary heart disease, cardiac arrest, or death, by approximately two-fold in older adults. Thus, patients on multiple drugs associated with MACE, including opioid analgesics, are also at increased risk of MACE. Our study supported the association of opioid and corticosteroid drugs, including tramadol and prednisolone, respectively, with cardiac arrest in older adults on polypharmacy.
Formica et al.28 reported a higher number of drug prescriptions in older adults with acute kidney injury, and Ernst et al.29 found that a higher number of drugs and use of nonsteroidal antiinflammatories increased the risk of impaired kidney function. Chao et al.30 revealed an association between polypharmacy with cardiovascular drugs (antihypertensive, lipid-lowering, and antiplatelet agents) and renal injury, consistent with our finding of a significant association between polypharmacy including cardiovascular drugs (amlodipine, atorvastatin, clopidogrel) and acute renal failure in older adults.
Our study significantly contributes to understanding the complex dynamics between polypharmacy and ADEs and specifically focuses on the differential effects across various age groups. It underscores the increased susceptibility of older adults to severe ADEs, calling for a critical reevaluation of medication prescription and management in older adults. Our findings highlight the need for a more tailored and cautious approach to pharmacotherapy for older populations, suggesting that personalized medication management greatly reduces the incidence of serious adverse reactions and enhances patient care. However, our research has certain limitations that may influence the interpretation of our results. Firstly, the use of data from spontaneous ADE reports may have introduced bias and underreporting, potentially skewing the perceived prevalence and severity of ADEs. Secondly, the retrospective study design limits the inference of causality between polypharmacy and the identified ADEs. Moreover, the absence of detailed data on specific drug interactions, and dosages, may affect the applicability of our findings across different populations. Thirdly, the high missing rate of clinical patient information including age, gender, prescribed or administered drugs, and ADE symptoms in the spontaneous ADE report data may decrease the generalizability and quality of the study results. This issue is well known in previous studies, however, using ADE reports with the most information was considered necessary for this study to obtain ADE cases where polypharmacy most likely contributed to severe ADEs in the elderly population.3132 Thus, further research is necessary to address these gaps, emphasizing the importance of incorporating comprehensive data on medication regimens and patient characteristics in future analyses. This is crucial for obtaining a deeper and more accurate understanding of the risk factors associated with polypharmacy, ultimately guiding the development of safer and more effective medication management strategies for patients of all ages.
Our study underscores the critical need for improved medication management in older adults due to their heightened risk of severe ADEs and highlights the criticality of ongoing research to refine our understanding of the impact of polypharmacy. This dual focus not only emphasizes the immediate need for better clinical practices but also sets the stage for future investigations to enhance the safety and efficacy of pharmacotherapy in the aging population.
XML Download