Hot topic in geriatric medicineRisk factors for adverse drug events in hospitalized elderly patients: A geriatric score
Introduction
Adverse drug events (ADEs) represent a major public health problem in the aged. The World Health Organization (WHO) defines an adverse drug event (ADE) as a detrimental response to a medication that is undesired and unintended, excluding therapeutic failure, poisoning, and intentional overdose [1]. In the United States, the number of reported ADEs increased by 160% between 1998 and 2005 [2] with an increase in fatal ADEs of 170%. Despite concerns that ADEs represent an important medical problem in the elderly, the predictive factors are still poorly understood. Risk factors reported to be independently associated with ADEs have included advancing age, sex, comorbidities, multiple drug regimens, inappropriate use of medication, poor cognitive function and anticoagulants [3]. Factors such as frailty, renal insufficiency, malnutrition or heart failure have also been mentioned. However, there is currently no consensus on which factors have the greatest impact. On the other hand, most studies have focused on the prevalence and risk analyses of inappropriate drug use in the elderly population using the Beers criteria [4], [5].
As strategies for preventing and reducing the effects of ADEs in the elderly are developed, an important component will be identifying high risk patients. Strategies to systematically detect and prevent ADEs are not always put to practice. One strategy for preventing in hospital ADEs may be to identify high risk elderly patients so that physicians can include the patient's level of risk when deciding upon the drug monitoring strategy. The aim of our study was to propose a specific score to evaluate the risk of ADEs in the elderly.
Because another score (the GerontoNet ADR risk score) [6] was recently proposed by a European group of geriatricians, it was interesting to compare the two scores.
Section snippets
Methods
To investigate this issue, a score was developed based on data from the IMEPAG study. The IMEPAG study was described in detail elsewhere [7] and its principal features are briefly described below.
Results
Sixteen geriatric centres participated in the study and 576 patients (mean age: 83.6 ± 7.9 years) were consecutively included. The mean number of drugs taken at inclusion was 9.4 ± 4.24 per patient; the median was nine (IQR: 6–12).
The most prevalent chronic diseases in the IMEPAG study were: cardiovascular (n = 412; 72%), musculoskeletal (n = 276; 48%), gastrointestinal (n = 207; 36%), genitourinary (n = 168; 29%), neurological (n = 150; 26%), ophthalmologic (n = 126; 22%), respiratory (n = 122; 21%), dementia (n
Discussion
The number of medications, antipsychotic treatment and a recent anticoagulant were identified as risk factors for ADE in our study. The most frequently involved drugs were antipsychotics, antihypertensives, anticoagulants, and analgesics. This confirms several other studies in elderly inpatients. Our score was validated by booststrapping (internal validation) and seems to be simple and easy to use in clinical practice. Discrimination of the risk score between those with low from those with
Comparison with the GerontoNet score
Recently, the GerontoNet ADR risk score was proposed by a European group of geriatricians [6]. The variables included in this score are: ≥ 4 comorbid conditions, heart failure, liver disease, number of drugs, previous adverse drug reaction (ADR) and renal failure (glomerular filtration rate less than 60 mL/min). The GerontoNet score is between 0 and 10, with an ADR risk in the validation study of 4.2–4.5% for a score of 0 to 3; 7% for a score of 4 to 5; 11.5% for a score of 6 to 7 and 28% for a
Number of drugs
The most powerful predictor of an ADE in our score is the use of seven or more drugs like in the GerontoNet ADR risk score [6]. Many studies have shown that the only significant predictor for ADE with multivariate analysis was the number of drugs. Indeed, the number of drugs taken, whatever the type of drug, could be a predictor of ADE. For Davies et al. [12], each additional drug multiplies the risk of an ADE by 1.14. It is important to note that the incidence of ADE increases with each
Comorbid conditions and age
In addition to the number of drugs, the presence of four or more comorbid conditions was a variable in the final score in the study by Onder et al. [6]. Comorbidity has been found to be the principal ADE risk factor in many other studies. In our study, the number of diseases or the number of comorbidities was not an independent risk factor because it was correlated with the number of medications. In practice, the number of prescribed drugs could be a useful measurement of comorbidity [16]. In
Anticoagulants
Recent anticoagulant treatment (< 3 months) was associated with the risk of ADEs in our study. In the study by Hanlon et al. [17], the main ADE risk factors in frail elderly persons after a hospital stay were the number of medications and warfarin use. The study by Gurwitz et al. [13] also found that anticoagulant use was a risk factor for all ADEs as well as preventable ADEs. Although warfarin presents safety concerns for patients, it is still often associated with severe ADEs. In a study on
In practice
Patients with a high ADE risk score and taking anticoagulants or antipsychotics should be carefully evaluated for ways of reducing their ADE risk, such as the possibility of safely reducing the number of medications to less than seven, if they are taking more than that number.The distribution of drugs associated with an ADE in our study (Table 2) is very similar to the drug groups which cause preventable admissions to hospital [25]. These treatments are frequently co-prescribed in the elderly,
Strengths and limitations
This study has the advantage of being a prospective and multicentric trial with 16 participating centres. Moreover the assessment of ADEs was standardized and reviewed weekly by investigators. Third, the results of the study were analysed by a multidisciplinary team of physicians and pharmacists.
Our study was limited by the size of the patient population and the resulting number of ADEs and preventable ADEs restricting our ability to identify risk factors. Another limitation is that our
Conclusion
Elderly patients with a high ADE geriatric score should be considered at risk and clinicians should critically review their drug regimens during hospitalization to identify unnecessary or inappropriate medications that could be discontinued. We have developed and validated a score to help clinicians identify elderly adults at a greater risk of ADE and therefore, who could be more closely monitored. Clinicians who care for elderly patients with multiple prescriptions, antipsychotics or recent
Disclosure of interest
The authors declare that they have no conflicts of interest concerning this article.
References (27)
- et al.
Identifying and preventing adverse drug events in elderly hospitalised patients: a randomised trial of a program to reduce adverse drug effects
J Nutr Health Aging
(2010) - et al.
The safety of warfarin therapy in the nursing home setting
Am J Med
(2007) - et al.
The incidence of adverse drug events in two large academic long-term care facilities
Am J Med
(2005) WHO Technical Report
(1972)- et al.
Serious adverse drug events reported to the Food and Drug Administration, 1998–2005
Arch Intern Med
(2007) - et al.
Predicting and preventing adverse drug reactions in the very old
Drugs Aging
(2005) Explicit criteria for determining potentially inappropriate medication use by the elderly: an update
Arch Intern Med
(1997)- et al.
Inappropriate medication use and health outcomes in the elderly
J Am Geriatr Soc
(2004) - et al.
Development and validation of a score to assess risk of adverse drug reactions among in-hospital patients 65 years or older. The GerontoNet ADR risk score
Arch Intern Med
(2010) - WHO Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC classification and DDD assignment, Oslo,...
Preventable and non-preventable risk factors for adverse drug events related to hospital admissions in the elderly. A prospective study
Clin Drug Invest
Multivariate prognostic models: issues in developing models, evaluating, assumptions and adequacy, and measuring and reducing errors
Stat Med
Cited by (46)
Machine Learning Methods for Predicting Adverse Drug Reactions in Hospitalized Patients
2023, Procedia Computer SciencePrognostic prediction models and clinical tools based on consensus to support patient prioritization for clinical pharmacy services in hospitals: A scoping review
2021, Research in Social and Administrative PharmacyCitation Excerpt :Internal validation was performed during the development of 14 studies.14,22,23,26–28,30,31,33,38,40–42,45 A cohort design, commonly recommended for prognostic model development, was used across 12 models,14,23,27,28,30,31,35,37,39,42,44,45 10 prospective22,24,26,29,32,34,36,38,41,43 and two retrospective studies.25,40 Only one study had a cross-sectional design.33
How hospital pharmacists prioritise patients at high-risk for medication harm
2019, Research in Social and Administrative PharmacyCitation Excerpt :The significant variables in the final BADRI model included, 1) length of stay (≥12 days), 2) use of hypoglycaemics, 3) high white cell count at admission (a proxy measure for antibiotic use), 4) hyperlipidaemia (a proxy measure for cardiovascular medications), and 5) the number of medications (>8 regular medications at admission).8 The Trivalle risk score had three variables in the final model, 1) use of anticoagulants, 2) use of antipsychotics and 3) number of medications.3 Comparing these variables with those from our study, use of anticoagulants and the number of medications prescribed, were the only common criteria.
Validation of a score to identify inpatients at risk of a drug-related problem during a 4-year period
2018, Saudi Pharmaceutical JournalCitation Excerpt :Therefore, the aim of our previous study was to design a risk score that could be applied in the majority of inpatients, similar to other studies that have evaluated risk factors for DRP (Bedouch et al., 2015; Blix et al., 2004; Classen et al., 1997; Zopf et al., 2008). A constant finding in several studies is that a higher number of drugs is a risk factor for the development of DRP (Alhawassi et al., 2014; Angamo et al., 2017; Blix et al., 2004; Davies et al., 2009; O’Connor et al., 2012; Onder et al., 2010; Pedrós et al., 2014; Tangiisuran et al., 2014; Trivalle et al., 2011; Urbina et al., 2014; Zopf et al., 2008). The DRP rate among inpatients in the present study was lower than that reported in other studies (Bedouch et al., 2009; Blix et al., 2004).
Consumer decision making for using comprehensive medication review services
2019, Journal of the American Pharmacists AssociationCitation Excerpt :Although the medians of self-reported health status were the same (median good), our respondents used fewer prescriptions than U.S. older adults in general (median 4 vs. 6).46,47 Literature suggested that the number of prescriptions was a risk factor for having MRPs.4,48-51 Because U.S. older adults had a higher prescription use compared with our sample, they may perceive themselves to be more susceptible to MRPs and may have higher benefit expectancy from using CMRs.
Diabetes mellitus increases risk of adverse drug reactions and death in hospitalised older people: the SENATOR trial
2024, European Geriatric Medicine
- 1
Composition of the Iatrogénie Médicamenteuse Évitable chez les Personnes Âgées en soins de suite et réadaptation Gériatriques (IMEPAG) Group: Hélène Agostini (URC Bicêtre), Joël Ankri (hôpital Sainte-Périne), Béatrice Barbier (URC Bicêtre), Alain Baulon (hôpital Sainte-Périne), Laurent Becquemont (URC Bicêtre), Patrick Bocquet (hôpital Corentin-Celton), Séverine Brémont (URC Bicêtre), Hervé Cazorla (hôpital Corentin-Celton), Nathalie Charasz (hôpital Broca), Jean-Philippe David (hôpital Émile-Roux), Philippe Davrinche (hôpital Corentin-Celton), Pierre Démolis (Afssaps), Claude Gallinari (hôpital Charles-Foix), Didier Guillemot (Institut Pasteur), Brigitte Hamon (hôpital Paul-Brousse), Olivier Henry (hôpital Émile-Roux), Marie-Pierre Hervy (hôpital Bicêtre), Muriel Kunstler (URC Bicêtre), Alexia Latierce (URC Bicêtre), Fabien Lesourd (Siège AP–HP), Isabelle Marie (URC Bicêtre), Anne-Marie Mathieu (hôpital Corentin-Celton), Sylvie Meaume (hôpital Charles-Foix), Muriel Palisson (hôpital Joffre), Jean-Guy Périlliat (hôpital Dupuytren), Marie-Laure Pibarot (Siège AP–HP), François Piette (hôpital Charles-Foix), Maité Rabus (hôpital Dupuytren), Anne-Sophie Rigaud (hôpital Broca), Michel Roger (hôpital Sainte-Périne), Georges Sebbane (hôpital Sevran), Christiane Verny (hôpital Bicêtre).