Abstract
Background/Aim: In a super-aging society, understanding the prescription status of drug burden index (DBI) drugs that have anticholinergic and sedative effects in patients with frailty to consider proper medical intervention and promote appropriate drug use for older adults is important. This study evaluated the association between frailty and the use of DBI drugs in older hospitalized patients using hospital electronic medical records.
Patients and Methods: This cross-sectional study included patients admitted to the Kameda Medical Center between October 1, 2016 and October 31, 2017. Patients with a Barthel Index of <90 or Mini-Mental State Examination score of <18 or otherwise were classified into the frailty and non-frailty groups, respectively. DBI drugs fall into nine categories based on previous studies, and 162 drugs marketed in Japan were included. Patients using DBI drugs were considered DBI drug users; otherwise, patients were considered DBI drug non-users. Comparisons of the DBI drug proportions in both groups were performed using logistic regression analysis while adjusting for patient background factors and calculating the adjusted odds ratios and 95% confidence intervals.
Results: The proportion of DBI drug users was significantly lower in the frailty group compared to the non-frailty group (adjusted odds ratio=0.32, 95% confidence interval=0.24-0.42, p<0.001).
Conclusion: Hospitalized older patients with frailty in Japan may be associated with a lower risk of DBI drug use and may use drugs with caution. In clinical practice, drug treatment for older patients may be implemented in consideration of various patient backgrounds, including frailty.
Introduction
In 2022, Japan had the highest aging population rate worldwide, with 29.0% of its population aged ≥65 years (1, 2). The increasing number of older patients requiring nursing care will exacerbate the pressure on social security, particularly long-term care services. As older adult individuals age, they transition from a state of health to one requiring care through a frailty clinical phase characterized by decreased physiological reserves and increased vulnerability (3-5). Frailty is defined as a condition where older adults may experience a decline in multiple functions, resulting in increased adverse clinical outcomes (3-7). One characteristic of frailty is its reversibility through proper medical intervention (3, 5).
In a super-aging society, one of the issues that has been emphasized along with frailty is that older people are prone to adverse drug events due to multiple medications or changes in pharmacokinetics caused by decreasing physiological function (8). The criteria for Potentially Inappropriate Medications have been proposed in Japan, the United States, and Europe and are being used to promote the appropriate use of medicines for older adults (9-12). Among the drugs on these lists, those with anticholinergic and sedative effects are known to have effects on the central nervous system and body functions. A measure that considers dosage is the drug burden index (DBI) developed by Himiler et al. (13), which is a measure of drug load based on the total exposure of an individual to anticholinergic and sedative drugs. A higher DBI score is associated with an increased risk of falls and decreased function and cognition (14).
It is important to understand the prescription status of DBI drugs in patients with frailty to consider proper medical intervention and promote appropriate drug use in older adults. Studies on frailty and DBI drugs have been inconsistent, with some reports from other countries finding a statistically significant association between high DBI scores and frailty (15-17) and others finding none (18). There are reports of an association between DBI drug administration and frailty in community-dwelling older adults in Japan (19). However, as some overseas reports indicate that the proportion and types of inappropriate drugs prescribed differ between outpatients and inpatients (20), the association between frailty and DBI drugs in inpatients in Japan needs to be clarified. The distribution of DBI scores among patients with frailty is also unknown.
This cross-sectional study aimed to evaluate the association between frailty and the use of DBI drugs in older hospitalized patients using hospital electronic medical records.
Patients and Methods
Study design. This was a cross-sectional study using electronic medical records.
Patients. Patients admitted to the Kameda Medical Center between October 1, 2016, and October 31, 2017, were included. Only the first hospitalization was included for patients who were hospitalized multiple times during the study period. The exclusion criteria were as follows: <65 years; missing data on the date of discharge or Barthel Index; hospitalized for <2 days; and did not use drugs during hospitalization.
Ethics approval and consent to participate. This study was conducted in compliance with the Ethical Guidelines for Medical Research Involving Human Subjects and was approved by the Ethical Review Committee of Tokyo University of Science and the Clinical Research Review Committee of Kameda General Hospital (Approval No. of Tokyo University of Science: 17025, Approval No. of Kameda Medical Center: 18-035).
Patient characteristics. The patient background was determined using age at admission, sex, Barthel Index, Mini-Mental State Examination (MMSE) (21) 90 days before the date of admission, and comorbidities in the year before the date of admission. The Barthel Index is an objective tool used to evaluate the activities of daily living on a 100-point scale based on 10 items, including transfers, bathing, and eating (22). The MMSE is the most commonly used cognitive assessment scale in clinical and research settings, with a maximum score of 30 points (23). Comorbidities included cardiac disease (I20-25, I50, J81), cerebrovascular disease (I60-69, G45), dementia (F00-F03 and G30), liver disease (B18, K70.0-K70.3, K70.9, K71, K73-74, K76.0,B15.0, B16.0 B16.2, B19.0, K70.4, K72, K76.6, I85), diabetes (E10-14), renal disease (I12, I13, N00-N05, N07, N11, N14, N17-N19,Q61), depression (F32-34), hypertension (I10-11), cancer (C00-C97), and Parkinson’s disease (G20) (24-26).
Frailty definition. Frailty identification was based on the report of Rozzini et al. (27). Patients with a Barthel Index of <90 or MMSE score of <18 were classified into the frailty group; otherwise, patients were classified into the non-frailty group (27).
Study outcomes. The medications used by eligible patients during their hospital stay were also investigated. DBI drugs, which have anticholinergic and sedative effects, fall into nine drug categories based on previous studies, and 162 drugs marketed in Japan were included (28, 29). Dosage forms included oral and injectable drugs (13). To exclude “as-needed” medications (29), DBI drug use was defined as the use of DBI drugs for ≥2 days during hospitalization (13) and was considered an outcome. If DBI drugs were used in each patient’s hospitalization period, the patient was considered a DBI drug user; otherwise, the patient was considered a DBI drug non-user.
Statistical analysis. For the comparison of patient backgrounds between the frailty and non-frailty groups, the Wilcoxon rank-sum test was used for age, and the χ2 test was used for the others. DBI drug use during hospitalization in the frailty and non-frailty groups was calculated, including the number of patients using DBI drugs, DBI drugs used (classified into nine drug categories), and DBI drugs used most frequently. The χ2 test was used to compare the number of patients using DBI drugs in the frailty and non-frailty groups. The DBI score was calculated for each patient based on the medications used throughout the entire hospitalization period, and the Wilcoxon rank-sum test was used to compare the DBI scores between the frailty and non-frailty groups. The DBI score was calculated using the following formula based on previous studies (13, 28):
D represents the daily dose of the drug obtained from electronic medical records. δ represents the lowest daily dose. In accordance with previous studies, the lowest daily dose approved in Japan was obtained from the package insert (28, 29). A higher DBI score indicated a higher drug burden. A DBI score of 0 indicated no DBI drug use. A DBI score of 0.5 indicated that the patient was using one type of DBI drug at the lowest daily dose listed on the label. The DBI score did not exceed 1 for the use of one DBI type of drug. A DBI score of ≥1 indicates that two or more types of DBI drugs are used together. Finally, comparisons of DBI drug user proportions in both groups were evaluated using logistic regression analysis while adjusting for patient background factors and calculating the adjusted odds ratios and 95% confidence intervals. As a sensitivity analysis, the frailty group was further divided into two subgroups, and comparisons were made between mild frailty, severe frailty, and non-frailty of DBI user’s proportions. Mild frailty was defined as having either a Barthel Index of <90 or an MMSE score of <18, while severe frailty was defined as having both a Barthel Index of <90 and an MMSE score of <18 (27). Homogeneity was tested using the Mantel-Haenszel test. Additionally, covariates were entered into the multivariate logistic regression analysis to identify any independent related factors associated with frailty and DBI use.
The statistical software R version 4.0.0 (The R Foundation for Statistical Computing, Vienna, Austria) was used. Statistical significance was set at p<0.05.
Results
Patients characteristics. Overall, 10,661 patients were admitted to the Kameda Medical Center during the study (Figure 1). After excluding patients <65 years, patients with missing discharge data or Barthel Index, patients with a hospital stay <2 days, and patients who did not receive drugs during their hospital stay, a total of 3,454 eligible patients were included in the study. Of the eligible patients, 765 (22.1%) were in the frailty group, and 2,689 (77.9%) were in the non-frailty group. Regarding patient background, the mean age was 80.4±8.4 years in the frailty group and 74.3±6.6 years in the non-frailty group (Table I). Significant differences were found in age, sex, and comorbidities other than depression upon admission (all p<0.01).
Patient exclusion flowchart.
Patient characteristics.
Number of DBI drugs used by frail and non-frail patients. The number of DBI drug users was 78 (10.2%), 670 (24.9%), and 748 (21.7%) in the frailty group, non-frailty group, and total number of patients, respectively (Table I). Among the DBI drug users, the largest number of patients in both groups used one type of DBI drug (frail patients: 52, 6.8%; non-frail patients: 485, 18.0%), followed by two types (frail patients: 17, 2.2%; non-frail patients, 126, 4.7%). Patients who used one DBI drug and those who used two DBI drugs together accounted for approximately 90% of the DBI drug users.
The top drugs in the nine DBI drug categories used in both groups were benzodiazepines, muscarinic receptor antagonists, and antiepileptic drugs, which accounted for approximately 70% of the total DBI drug use (Table II). The top five most frequently used DBI drugs were zolpidem, pregabalin, etizolam, tamsulosin, and brotizolam in the frailty group, and pregabalin, zolpidem, tamsulosin, brotizolam, and etizolam in the non-frailty group (Table III).
Data of each therapeutic category for drug burden index drugs in the frailty and non-frailty groups.
The top five drug burden index drugs in the frailty and non-frailty groups.
DBI scores of frail and non-frail patients. In both groups, the largest proportion of patients had a DBI score of 0 (frailty group: 89.8%; non-frailty group: 75.1%), followed by patients with a DBI score of 0-0.5 (frailty group: 5.5%; non-frailty group: 13.8%; Figure 2). Thus, 95.3% of the frailty group and 88.9% of the non-frailty group had a DBI score of ≤0.5. The Wilcoxon’s rank-sum test for DBI scores showed that the non-frailty group had a significantly higher DBI score than the frailty group (p<0.001; Table I).
Drug burden index (DBI) scores of frail and non-frail patients.
Comparison of the proportion of DBI drug users in frailty and non-frailty patients. A comparison of the proportion of DBI drug users between the frailty and non-frailty groups is shown in Table IV. The multivariate logistic regression analysis adjusted for age, sex, and comorbidities (heart disease, cerebrovascular disease, dementia, liver disease, diabetes, renal disease, hypertension, cancer, and Parkinson’s disease) showed that the proportion of DBI drug users was significantly lower in the frailty group than in the non-frailty group (adjusted odds ratio=0.32, 95% confidence interval=0.24-0.42, p<0.001).
Multivariate logistic regression analysis comparing the proportion of drug burden index (DBI) users in the frailty and non-frailty groups, a sensitivity analysis divided into mild frailty and severe frailty according to severity.
As a result of sensitivity analysis, the multivariate logistic regression analysis adjusted for age, sex, and comorbidities showed that the proportion of DBI drug users was significantly lower in the mild frailty group than in the non-frailty group (adjusted odds ratio=0.35, 95% confidence interval=0.27-0.47, p<0.001). However, this was not applicable for the severe frailty group (adjusted odds ratio=0.26, 95% confidence interval=0.05-1.27, p for homogeneity=0.668). Table V shows the results of the multivariate logistic regression analysis, which revealed that higher age, cerebrovascular disease, dementia, diabetes mellitus, cancer, and Parkinson’s disease were independent related factors for frailty and DBI use.
Multivariate logistic regression analysis for frailty associated drug burden index use related factors.
Discussion
This cross-sectional study used electronic medical records to evaluate the association between frailty and DBI drug use in older Japanese inpatients. The results showed that in the assessment of DBI drug use, the frailty group was significantly less likely to use DBI drugs during hospitalization than the non-frailty group. These findings suggest that hospitalized older patients with frailty may be at a lower risk of DBI drug use and may use drugs with more caution.
Approximately 20% of the patients in both groups used DBI drugs. In studies on community-dwelling older adults in Japan, the proportion of DBI drug users ranged from 24% to 50% (19, 28, 29). One possible reason for the lower percentage of DBI drug users compared to previous studies may be that prescription medications were scrutinized upon admission.
Approximately 90% of patients in both the frailty and non-frailty groups had a DBI score of ≤0.5. In a study of community older adults in Japan, approximately 90% of patients had a DBI score of ≤0.5, which supports the findings of our study (28, 29). Our study suggests that DBI drug usage in hospitalized patients and in communities of older adults is absent or at a minimum dose; hence, drug treatment is given with consideration.
The use of benzodiazepines was found to be the highest among all DBI drug categories. There were almost no differences between the two groups in terms of DBI drug categories or the top five drugs. This finding is consistent with the results of a previous study (19). Therefore, benzodiazepines are the most commonly used DBI drugs in Japan, and it is unlikely that the trend of DBI drugs used by the frailty and non-frailty groups will differ.
The frailty group had a significantly smaller proportion of DBI users than the non-frailty group. A study of community-dwelling older adults in Japan compared the total number of patients with frailty and patients requiring nursing care with that of non-frailty patients and reported that patients with frailty and patients requiring nursing care had a higher rate of DBI drug use than non-frailty patients (19). In Japan, no comparison has been made between patients with frailty and non-frailty patients and the results of a previous study (19) may have been drawn from the results of patients requiring nursing care. Furthermore, a previous overseas study reported that 20-50% of inappropriate medications were discontinued before admission owing to hospitalization (20). One possible reason for the lower percentage of DBI drug users in the frailty group might be influenced by the unique healthcare context in Japan. In Japan, all citizens are covered under a universal health insurance system. In 2016, one year before this study period, the Japanese government introduced a “drug comprehensive evaluation adjustment surcharge” as a countermeasure against polypharmacy in hospitals (30). This system provides financial incentives to hospitals that reduce the number of drugs prescribed to patients who were taking six or more types of oral medications before hospitalization (30). Previous studies in Japan have reported that the number of medications taken by patients decreased significantly following hospitalization (31). The implementation of this policy might contribute to a reduction in DBI drug use, particularly among patients with frailty, who are more likely to undergo prescription adjustments upon hospitalization. As a result, this may have led to the observed lower use of DBI drugs among patients with frailty compared to non-frailty patients.
Study limitations. First, the number of patients with frailty, especially patients with severe frailty, was small compared to the number of non-frailty patients. As the records were from only one hospital, the sample size was limited. Due to the small number of patients with severe frailty, the point estimates in the logistic regression analysis of DBI drug users were lower for patients with severe frailty compared to patients with mild frailty. However, no significant differences were observed between non-frailty and patients with severe frailty. Future studies with larger population sizes incorporating frailty and drug data are warranted. Second, this study clarified the relationship between frailty and DBI drugs in patients admitted to Japanese hospitals. However, we focused on acute care hospitals. Therefore, a survey of chronic care hospitals where the situation of inpatients is different from our hospital is required in future studies. Furthermore, this study focused exclusively on patients in acute care hospitals; it was not possible to confirm the use of DBI drugs before hospitalization. Future studies should aim to collect and analyze data from both before and after hospitalization to confirm whether the use of DBI drugs changes throughout hospitalization. Third, the definition of frailty should also be considered. In this study, the definition was based on the report by Rozzini et al. to identify flaws in hospital electronic medical records (27). However, further research using different definitions of frailty is required (4, 32-34).
Conclusion
A study on the proportion of DBI users among older hospitalized Japanese patients with frailty and non-frail older hospitalized Japanese patients showed that the frailty group had a significantly lower proportion of DBI drug use than the non-frailty group. These findings suggest that hospitalized older patients with frailty may be at a lower risk for DBI drug use and may be using drugs with caution. In clinical practice, drug treatment for older patients may be implemented in consideration of various patient backgrounds, including frailty.
Acknowledgements
We would like to thank the staff of the Kameda Medical Center for data collection.
Footnotes
Authors’ Contributions
Conceptualization: Yukako Morisaki, Misuzu Takashima, Ayako Maeda-Minami, Sayaka Izumi, Masanori Suzuki, Ryohkan Funakoshi, and Yasunari Mano; acquisition of data: Sayaka Izumi, Masanori Suzuki, and Ryohkan Funakoshi; analysis: Yukako Morisaki, Misuzu Takashima, and Ayako Maeda-Minami; interpretation: Yukako Morisaki, Misuzu Takashima, Ayako Maeda-Minami, Ryohkan Funakoshi, and Yasunari Mano; drafting of the manuscript: Misuzu Takashima and Ayako Maeda-Minami. All the Authors have read and approved the final version of this manuscript.
Funding
This work was supported by Statistics Professors Training Programs (The Ministry of Education, Culture, Sports, Science and Technology (MEXT) Grant).
Conflicts of Interest
The Authors declare that they have no competing interests in relation to this study.
- Received February 3, 2025.
- Revision received February 25, 2025.
- Accepted February 26, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.