Potentially inappropriate medications and functional decline in elderly hospitalized patients

Potentially Inappropriate Medications and Functional Decline inElderly Hospitalized Patients Andrea Corsonello, MD,Ã Claudio Pedone, MD, Fabrizia Lattanzio, MD,Ã Maria Lucchetti, ScD,ÃSabrina Garasto, ScD,Ã Massimo Di Muzio, ScD,Ã Sergio Giunta, MD,Ã Graziano Onder, MD,zAngelo Di Iorio, MD,§ Stefano Volpato, MD,k Francesco Corica, MD,# Chiara Mussi, MD,ÃÃ and Raffaele Antonelli Incalzi, MD, ,ww on behalf of the Pharmacosur Veillance in the Elderly Care Study Group OBJECTIVES: To verify whether the use of potentially in- 95% CI 5 1.50–10.5 for the loss of !3 ADLs), but PIMs appropriate medications (PIMs) is associated with loss of independence in elderly in-patients by promoting adverse CONCLUSIONS: ADRs to any drugs more than the use of PIMs might be associated with functional decline in elderly DESIGN: Prospective observational study.
hospitalized patients, but because the power of this study PARTICIPANTS: Five hundred six patients aged 65 and was too limited to definitively exclude a direct relationship older admitted to 11 acute care medical wards.
between PIMs and functional decline, this merits further MEASUREMENTS: In-hospital loss of one or more activ- investigation. J Am Geriatr Soc 2009.
ities of daily living (ADLs) and three or more ADLs. PIMswere identified according to diagnosis-independent Beers Key words: potentially inappropriate medications; Beers criteria and ascertained by study physicians based on daily criteria; functional decline; adverse drug reactions review of medical and nurse records. The relationship be-tween risk factors and outcomes was assessed using logisticregression.
RESULTS: Overall, 104 patients (20.6%) were taking atleast one PIM at the time of admission (baseline users), and49 (9.7%) were newly prescribed at least one PIM duringtheir hospital stay. The loss of one or more ADLs occurred Functional decline is frequent in elderly hospitalized pa- tients,1 and the loss of physical function predicts wors- in 9.6% of baseline users, 16.3% of new users, and 8.5% of ening quality of life and increasing morbidity and mortality nonusers (P 5.21) and that of three or more ADLs in 7.7% and has relevant social and economic implications.2 Older of baseline users, 12.2% of new users, and 4.8% of non- patients are particularly exposed to the risk of developing users (P 5.10). The lack of association was confirmed after new functional deficits during hospitalization;3 thus, the correction for potential confounders, including ADRs. The identification of factors contributing to the development of occurrence of ADRs was strongly associated with both functional decline during a hospital stay is a relevant goal in outcomes (odds ratio (OR) 5 7.80, 95% confidence interval geriatric clinical practice. The use of potentially inappro- (CI) 5 3.53–17.3 for the loss of !1 ADLs; OR 5 3.98, priate medications (PIMs, i.e., of drugs with a particularlyhigh risk of adverse drug reactions (ADRs) in frail hospi-talized elderly people)4 might be among these factors. It has From the ÃItalian National Research Centers on Aging, Ancona and Cosenza, been demonstrated that in home-dwelling elderly subjects, Italy; Department of Geriatrics, University Campus BioMedico, Rome, Italy; chronic use of PIMs is associated with impaired physical zDepartment of Gerontology, Geriatrics and Physiatry, Catholic University of performance and greater rate of hospitalization.5,6 Accord- Sacred Heart, Rome, Italy; §Department of Medicine and Sciences of Aging,Laboratory of Clinical Epidemiology, University G. D’Annunzio, Chieti, ingly, limiting PIMs might be valuable not only from the Italy; kDepartment of Clinical and Experimental Medicine, Section of perspective of preventing ADRs. Criteria identifying PIMs Internal Medicine and Geriatrics, University of Ferrara, Ferrara, Italy; in elderly people were developed in 1991 for use in the Department of Internal Medicine, University of Messina, Messina, Italy; elderly nursing home population7 and were subsequently Unit of Geriatric Medicine, S. Agostino Estense Hospital, University of expanded and made generalizable to the whole population aged 65 and older in 19978 and revised in 2002.9 Previous Address correspondence to Andrea Corsonello, Viale della Resistenza Pal, investigations reported on the extensive use of PIMs in older Alfa Scala H, I-87036 Rende (CS), Italy. E-mail: populations in various settings, with a prevalence ranging from 12% to 40%,10–15 but the clinical relevance of PIM r 2009, Copyright the AuthorsJournal compilation r 2009, The American Geriatrics Society use in elderly in-patients has been assessed only with regard Ninth Revision, Clinical Modification.19 Information to outcomes such as in-hospital ADRs and mortality but not about destination at discharge and evolution of the main with regard to the risk of functional decline.14,15 The ev- diagnosis was also collected. For drugs taken during the idence pertaining to home-dwelling elderly people5 and the stay, to establish the duration of the therapy and which dramatic effect a hospital stay has on personal capabilities16 drugs were administered simultaneously during hospital- makes this outcome worthy of assessment in hospitalized ization, the starting and ending dates of therapy were re- corded. If the dosage was modified, the name of drug, It was hypothesized that inappropriate drug use is as- the new dosage, the initiation and discontinuation dates of sociated with greater risk of loss of personal independence the new dosage, and the main reasons that determined the in elderly hospitalized patients, and the objective of this modification of the therapy were also recorded.17 The list of study was to evaluate whether this relationship exists and, if drugs prescribed at discharge was also collected.17 so, whether a greater incidence of ADRs mediates it.
Part DFComprehensive geriatric assessment: Mini Men- tal State Examination (MMSE),20 Geriatric Depression Scale(GDS),21 activities of daily living (ADLs) and instrumental ADLs,22–24 Cumulative Illness Rating Scale (CIRS),25 presence of pressure sores, urinary incontinence, fecal incontinence, use The present study used data from a collaborative observa- of urinary cathether, clinical history of falls, syncope, dizziness, tional study group, the PharmacosurVeillance in the elderly blood pressure, heart rate, body temperature, weight, height, Care, based in community and university hospitals located waist circumference, weight loss in the previous 3 months.
throughout Italy, aimed at surveying drug consumption, Part EFADRs: An ADR was defined according to the occurrence of adverse drug reactions, and quality of World Health Organization definition, which refers to any hospital care. The main objective of the study was to in- noxious, unintended, and undesired effect of a drug that vestigate the prevalence and correlates of potentially inap- occurs at doses used in humans for prophylaxis, diagnosis, propriate prescribing in elderly hospitalized patients and to or therapy. This definition excludes therapeutic failures, measure its effect on short-term and 1-year clinical out- intentional and accidental poisoning (i.e., overdose), drug abuse, errors in drug administration, and nonadherence.26 All patients consecutively admitted to the participating The study physicians evaluated the participants on each day wards (11 acute care medical wards and 3 long-term care of their hospital stay for new potential ADRs. They were and rehabilitation units) from April 1 to June 30, 2007, instructed to check the medical chart and new laboratory were asked to participate in the study. After obtaining writ- analyses and to ask the nurse, the attending physician, and ten informed consent, a study physician with specific train- the participants themselves for any new event, regardless of ing completed a questionnaire for each patient at admission its apparent relation to the drugs prescribed. The decision as to the hospital and updated it daily. A training session was to whether to consider the new event an ADR was left to conducted at the coordinating center, and an instruction their judgment. For each suspected ADR, the study physi- manual was delivered to each clinical center. In particular, the cian coded clinical description, severity, and eventual evo- training covered all of the components of the assessment (see lution. In addition, he collected detailed information about Data Recorded Included parts A–E, below), with special em- the drug(s) identified as the potential culprit. The attending phasis on how to detect ADRs (on the logical process under- physician rated the severity of each ADR as mild (the with- lying the interpretation of a new clinical event as a potential drawal of the suspected drug was not necessary), moderate ADR). Researchers judged simulated clinical cases under the (the withdrawal of the suspected drug was necessary, but guidance of two senior researchers (CA, PC). The training the ADR was not life threatening), or severe (life-threaten- lasted until the achievement of good interrater agreement in ing ADR).17 The researcher was not aware of the analysis to the identification and coding of ADRs and diagnoses.
be performed using the recorded data.
Part FFLaboratory parameters: Routine blood anal- yses performed at the time of admission and at discharge.
Once discharged, patients were followed up every 3 Part AFDemographic and pre-admission data: Age, sex, ad- mission diagnosis, place of living information, marital status, Overall, 762 patients were screened in the survey pe- years of education, number of hospital admission in the pre- riod, but 72 (9.4%) refused to participate, leaving a final vious 12 months, family arrangement, alcohol consumption, sample of 690. Twenty-five patients who died during hos- smoking habits. For drugs taken before admission, patients pital stay were excluded from the present study, as were were asked to display the drug containers and the physician’s patients enrolled in long-term care or rehabilitation units instructions. For cognitively impaired patients, this informa- (n 5 159), leaving a final sample of 506 patients for the tion and those about adherence were checked through the analysis. The present study used only data collected from patient’s relatives or caregivers.17 Drugs were coded accord- admission to discharge from the hospital. The Ethical ing to Anatomical and Therapeutical Classification.18 Committee of the Italian National Research Center on Ag- Part BFSocioeconomic status: Occupation before re- ing (INRCA) approved the study protocol.
tiring, economic status, need of economic help for healthproblems, formal and informal assistance.
Part CFClinical problems: Diagnoses and diagnostic procedures performed during stay coded by the trained The main outcome of the present study was the loss of physician using the International Classification of Disease, physical independence in completing ADLs. Overall func- INAPPROPRIATE MEDICATIONS AND FUNCTIONAL DECLINE tional capabilities were rated using a seven-item (transfer- ADRs could mediate the effects of PIMs on functional sta- ring, ambulation, dressing, eating, toileting, bathing, and tus. All drugs causing ADRs in the database were also re- continence) ADL scale22,24 at admission and discharge.
viewed, and the relationship between increasing number or Performance on individual tasks was scored according to severity of ADRs and functional decline was investigated.
the five levels of difficulty recommended by the World All statistical analyses were performed using SPSS soft- Health Organization: without difficulty, with difficulty but ware (version 10.0; SPSS Inc., Chicago, IL).
without help, help for only part of the activity, help for totalactivity, and not able to perform.27 For each ADL item,needing help for total activity or being not able to perform it was considered to indicate loss of the explored function. To Overall, 104 of 506 patients (20.6%) were taking one or investigate the relationship between exposure to PIMs and more PIMs at the time of admission, and 49 (9.7%) were different degrees of functional impairment, two different newly prescribed one or more PIMs during their hospital outcome variables were defined: loss of one or more ADLs stay. The prevalence of baseline and newly prescribed drugs and loss of three or more ADLs from admission to dis- listed in the Beers criteria is reported in Figure 1. Of them, the most frequently used at the baseline was ticlopidine, PIMs were identified according to the diagnosis-inde- followed by doxazosin, amiodarone, and ferrous sulfate.
pendent 2003 revision of the Beers criteria.9 Patients who The most frequently newly prescribed PIMs were ticlopi- were already taking a PIM at admission (baseline users) and dine, long-acting benzodiazepines, ferrous sulphate, and those who started a PIM during their hospital stay (new Baseline users of PIMs were younger but had greater cumulative comorbidity and clinical severity scores andwere more frequently affected by hypertension, stroke, and chronic cerebrovascular disease than nonusers. New users According to the basic hypothesis, the occurrence of any had a greater prevalence of depression and a lower prev- ADR during hospital stay was considered as a potential alence of chronic obstructive pulmonary disease than non- correlate of the outcomes. Based on the available literature, users. ADRs occurred in 14.4% of baseline users and in the following potential correlates were also considered: 16.3% of new users (Table 1). The loss of one or more anthropometric and sociodemographic data (age, sex, body ADLs occurred in 9.6% of baseline users, 16.3% of new mass index, smoking habit, alcohol consumption), history users, and 8.5% of nonusers (P 5.21), and that of three or of falls during the previous year and occurrence of falls more ADLs in 7.7% of baseline users, 12.2% of new users, during the hospital stay, health status indexes (level of per- sonal independence in ADLs at admission, cognitive im- In the logistic regression analysis, the use of PIMs was pairment as ascertained based on an age- and education- not independently associated with loss of one or more or adjusted MMSE score o24, and depressed mood (GDS three or more ADLs. After also adjusting for the occurrence score 45)), selected comorbid conditions (diabetes melli- of any ADR during the hospital stay, the association be- tus, hypertension, coronary artery disease, atrial fibrilla- tween baseline or new use of PIMs and functional decline continued to be nonsignificant (Table 2).
pulmonary disease, peripheral vascular disease, stroke or According to the fully adjusted model, the following transient ischemic attack, chronic cerebrovascular disease, variables qualified as independent correlates of the out- renal failure, and malignancies), and CIRS comorbidity and comes or were close to the threshold of significance: the severity scores. The overall number of drugs, excluding occurrence of ADRs (odds ratio (OR) 5 7.80, 95% CI 5 PIMs, and the length of hospital stay were calculated and 3.53–17.3 for the loss of !1 ADLs; OR 5 3.98, 95% CI 5 1.50–10.5 for the loss of !3 ADLs), age (for each 1- First, the prevalence of the use of PIMs in the popu- year increase: OR 5 1.11, 95% CI 5 1.05–1.18 and lation studied was calculated, and baseline users of one or OR 5 1.11, 95% CI 5 1.03–1.18, respectively), CIRS more PIMs and new users were compared with nonusers.
comorbidity score (OR 5 1.21, 95% CI 5 0.98–1.50 and Contingency tables with chi-square were used for categor- ical variables and one-way analysis of variance for contin- number of drugs excluding PIMs (OR 5 1.05, 95% uous variables. Age, sex, and variables significantly CI 5 0.99–1.11 and OR 5 1.07, 95% CI 5 0.99–1.15, re- distinguishing users from nonusers at Po.1 were entered spectively). Chronic cerebrovascular disease (OR 5 2.82, into the multivariable logistic regression model to identify 95% CI 5 1.18–6.77) was significantly associated with the independent correlates of the outcomes. Functional status loss of one or more ADLs, whereas the number of lost ADLs at admission was also considered as a potential confounder.
at admission was negatively associated with the loss of three The number of drugs used during the hospital stay exclud- or more ADLs (OR 5 0.76, 95% CI 5 0.60–0.97).
ing PIMs was also entered into the final multivariable mod- Patients experiencing an ADR during their hospital els, given that polypharmacy is known to be associated with stay were more likely to be female (72.4% vs 52.0%, ADRs.28 The presence of collinearity between variables en- P 5.003) and cognitively impaired (62.1% vs 49.1%, tered in the final regression model was excluded by assess- P 5.06), used a greater number of drugs (15.2 Æ 7.9 vs ing the effect of individual variables on the standard error of 9.9 Æ 4.8, P 5.001), had greater overall comorbidity (CIRS regression coefficients, as reported by Hosmer and Leme- comorbidity score 4.5 Æ 1.8 vs 3.6 Æ 1.9, P 5.001), and show.29 This analysis was repeated including the occurrence more frequently had congestive heart failure (37.9% vs of any ADR during the hospital stay to verify whether 20.3, P 5.002) and renal failure (27.5% vs 14.5%, Baseline use
Figure 1. Prevalence of baseline (upper panel) and new (lower panel) use of potentially inappropriate medications. ÃExcept when usedto control atrial arrhythmias. ÃÃExcept phenobarbital when used to control seizures.
P 5.011). Of the 58 patients experiencing one or more with the loss of three or more ADLs. Although, as expected, ADRs during their hospital stay, 29 had a single ADR, 17 ADRs were strongly correlated with the outcome, drugs had two ADRs, and 12 had three or more ADRs. The di- considered inappropriate caused few of them (6/106), in- agnosed ADR was judged to be moderate to severe in 22 of dicating that the ADRs, more than PIMs, could be associ- 58 patients (37.9%). The greater the severity of ADRs, the greater likelihood of experiencing a functional loss during These data are in contrast with others showing an as- the hospital stay, whereas having two or more ADRs was sociation between use of PIMs and disability.5,30 The most associated only with loss of three or more ADLs (Table 3).
likely explanation for this discrepancy lies in the fact that When reviewing drugs causing ADRs in the database, it those studies were cross-sectional, and some of the Beers was found that only a minority of ADRs could be ascribed drugs might simply be more frequently used in frail patients.
In the current sample, for example, patients prescribedBeers drugs were more frequently affected by stroke andother cerebrovascular diseases. Furthermore, ticlopidine, the most used of the Beers drugs in this study, was usually This study shows that the use of PIMs, as ascertained based prescribed to patients with a greater atherosclerotic burden; on Beers criteria, is not associated with decline in personal two or more cardio- or cerebrovascular conditions were capabilities in elderly patients in acute care medical wards, found in 86.8% of ticlopidine users, versus 60.3% in non- even if the data suggest that some correlation might exist users, (Po.001). This further supports the possibility that, INAPPROPRIATE MEDICATIONS AND FUNCTIONAL DECLINE Table 1. Sociodemographic and Clinical Characteristics of Patients According to the Use of One or More PotentiallyInappropriate Medications (PIMs) Alcohol consumption 40.5 L wine equivalent, % Number of lost activities of daily living at admission, mean Æ SDGeriatric Depression Scale score 45, % Mini-Mental State Examination score o24, % Cumulative Illness Rating Scale score, mean Æ SD History of falls during the previous year, % Number of drugs during stay (excluding PIMs), mean Æ SDAny adverse drug reaction during stay, % in previous studies, Beers drugs might be associated with Table 2. Relationship Between Use of Potentially Inap- disability as a marker of an at-risk condition rather than propriate Medications (PIMs) and Functional Decline The strong association between ADRs and functional decline adds to the knowledge of the spectrum of ADR- related negative outcomes. Although ADRs might also be a marker of frailty, as suggested by the clinical profile of patients experiencing ADR in the current study, the analysisof individual ADRs suggests that more than one-third were judged moderate to severe (were severe enough to affect the level of personal independence). ADRs were the main correlate of severe functional decline in a logistic model Additionally adjusted for occurrence of any adverse drug reaction during stay corrected for the global number of drugs used and the use of Beers drugs. Furthermore, ADRs were frequently multiple ( !2 in 50% of the patients experiencing them) and impor-tant, such as gastrointestinal bleeding, respiratory failure, Adjusted for age, sex, number of lost activities of daily living at admission, and delirium, thus, seeming severe enough to affect Geriatric Depression Scale score 45, Cumulative Illness Rating Scale co- personal capabilities. Further supporting this view is the morbidity score, hypertension, chronic obstructive pulmonary disease, stroke fact that disability and mortality share most of their or transient ischemic attack, cerebrovascular disease, and number of drugsexcluding PIMs.
predictors in different epidemiological models and clinical contexts.31–33 Accordingly, the repeatedly reported link Table 3. Relationship Between Number or Severity of Table 4. Summary of Beers Listed (Upper Panel) and All Adverse Drug Reactions (ADRs) and Functional Decline Other Drugs (Lower Panel) Causing Adverse Drug Reac- Adjusted for age, sex, number of lost activities of daily living at admission, Geriatric Depression Scale score 45, Cumulative Illness Rating Scale comorbidity score, hypertension, chronic obstructive pulmonary disease, stroke or transient ischemic attack, cerebrovascular disease, number of drugs excluding potentially inappropriate medications (PIMs), and use of PIMs.
between ADRs and mortality makes it reasonable that ADRs also mark frail patients at risk of disability.34 This has been proved in selected populations; ADRs such as hy- poglycemia in people with diabetes mellitus or hypotension in populations with and without hypertension (e.g., patients with heart failure) are associated with accelerated func- tional decline.35,36 Moreover, ADRs are associated with longer stay, and longer stay, in turn, might per se carry a greater risk of functional decline,28 given that the acute care hospital frequently poses several barriers to mobility.37 Fi- nally, ADRs are more common in patients with renal or tachycardia (1), bradycardia (2),atrioventricular block (2) liver impairment, which are important indicators of im- paired homeostatic mechanisms.38 Renal dysfunction fre- quently remains unrecognized in elderly people and predicts the risk of developing ADRs to hydrosoluble drugs.28 Thus, several reasons support the hypothesis that ADRs might contribute to functional decline. Accordingly, prevention of in-hospital ADRs is expected to preserve personal capabil- phlebitis (3), renal failure (1), tremor(1), vomiting (1) ities. These findings seem robust, because a wide set of potential confounders was taken into account, and the study was a prospective one, although definitive proof of the relationship between ADRs and functional decline would require that interventions preventing ADRs also The fact that 42 of 506 patients lost more than three ADL deserves consideration. Only a detailed analysis of individual case histories would allow disease-related func- tional decline to be distinguished from procedural-related functional decline. Among procedural-related disabilities should be considered those attributable to ADRs, inade- quate or improper therapy, lack of nutritional support, and other factors suggesting a poor-quality process of care, al- though only one disease with a recognized disabling po- tential, chronic cerebrovascular disease, was among the independent correlates of the outcome, whereas ADRs (a procedural-related factor) emerged as the main correlate.
This finding seems consistent with an important role of INAPPROPRIATE MEDICATIONS AND FUNCTIONAL DECLINE procedural factors as potential determinants of functional needed to assess the respective role of individual ADRs and to decline of hospitalized elderly people. Previous reports verify to what extent ADRs mediate the reported disabling about highly prevalent inadequate nutrition and hydration potential of selected diseases24,48 in the acute care setting.
of elderly hospitalized patients are in keeping with this hy- Furthermore, the present results suggest that disability should pothesis.39,40 Furthermore, the positive effect of in-hospital be included with mortality and length of stay among the po- comprehensive geriatric assessment and intervention pro- tential consequences of ADRs for clinical and epidemiolog- grams41 and of strategies designed to prevent functional ical purposes, as well as among the modifiable outcomes in decline42 are consistent with a concurrent disabling role of programs designed to prevent ADRs in elderly in-patients.
practice-related modifiable factors. These findings and con-siderations might improperly depict the acute care hospital as a disabling setting. It is worthy to observe that 9.7% ofpatients showed improvement in one or more ADLs, and The PharmacosurVeillance in the elderly Care study was 77.5% had ADLs unchanged with regard to preadmission partially supported by a grant from the Italian Ministry of levels. Nevertheless, the risk of in-hospital functional de- cline increases markedly with age,3 and the results of the current study reinforce the message that clinicians should closely monitor the functional status of elderly hospitalized A. Corsonello, M. Lucchetti, M. Di Muzio, G. Gioia Lobbia, S. Giunta, S. Garasto, C. Carbone, C. Greco, P.
This study has several limitations. First, some propor- Fabbietti, R. Firmani, M. Nacciariti, F. Lattanzio (Italian tion of patients who died might have experienced functional National Research Center on Aging (INRCA), Ancona and decline before dying. Thus it is likely that the analytical procedure underestimates the incidence of the outcome and the strength of its relationship with ADRs and Beers drugs, INRCA AnconaFG. Cadeddu, R. Ganzetti.
although it was judged that, an intermediate functional INRCA AppignanoFE. Giannandrea, A. Marrocchi, assessment lacking, using the composite outcome ‘‘death or disability’’ might have provided an unreliable estimate University of ChietiFA. Di Iorio, M. Abate, A. Leon- of true correlates of incident disability. Second, the Beers list of PIMs is questionable for several aspects.43,44 For in- INRCA CasatenovoFE. Guffanti, E. Marchi, P. Castel- stance, although amiodarone has limited effectiveness in maintaining sinus rhythm in older adults, together with INRCA CosenzaFB. Mazzei, L. Pranno, C. Zottola, greater risk of QT interval prolongation and torsades de E. Feraco, A. Madeo, R. Gallo, G. Nicotera, P. Scarpelli, A.
pointes,9 it is among recommended drugs for the treatment of atrial fibrillation in patients with heart failure.45 Addi- University of FerraraFS. Volpato, C. Maraldi, M. Cav- tionally, the Beers list does not address several other issues, alieri, G. Guerra, F. Sioulis, L. Testoni, S. Postorivo, R. Fellin.
such as duplicate drug class prescriptions, drug–drug inter- INRCA FermoFD. Postacchini, G. Cruciani, R. Bru- actions, inappropriate duration of therapy, and underuse of nelli, F. Sorvillo, M. Francavilla, R.L. Marzullo, C. Giuli.
lifesaving drugs, such as warfarin to prevent cardioembolic University of MessinaFF. Corica, C. Nicita Mauro, G.
events in patients with atrial fibrillation or beta-blockers in patients with coronary artery disease.46 Accordingly, Beers- University of ModenaFC. Mussi, R. Scotto, G. Salvioli.
listed PIMs refer to truly inappropriate medications only to some extent. Third, changes in functional status from ad- University Campus Biomedico, RomaFC. Pedone, mission to discharge were recorded; thus, some patients might have experienced functional decline before the onsetof ADRs. Thus, it cannot be definitively concluded that the Conflict of Interest: The editor in chief has reviewed the observed relationship between ADRs and functional decline conflict of interest checklist provided by the authors and has was a causal one. Furthermore, the use of physician-re- determined that the authors have no financial or any other ported ADRs probably underestimated the extent of ADRs.
kind of personal conflicts with this paper.
In addition, only one assessor assessed ADRs for the center, Author Contributions: Corsonello, Pedone, and Anto- although based on the good interrater agreement achieved nelli Incalzi: study concept and design, acquisition of sub- in the training session, it is likely that this is a minor lim- jects and data, analysis and interpretation of data, and itation. Given the low incidence of ADRs to PIMs, the preparation of manuscript. Lattanzio, Lucchetti, Di Muzio, effect, if any, of ADRs to PIMs rather than that of ADRs as a and Giunta: study concept and design, analysis and inter- whole on functional decline could not be assessed, although pretation of data, and preparation of manuscript. Garasto, it should also be considered that PIMs may increase the Onder, Di Iorio, Volpato, Corica, and Mussi: data collec- generic risk of ADRs by interacting with other drugs not tion and analysis and interpretation of data.
considered inappropriate.47 Finally, because of its limited power, this study may lack precision in estimates of theassociations observed.
In conclusion, this study shows that ADRs more than PIMs, as codified by the Beers criteria, might be associated 1. Sager MA, Franke T, Inouye SK et al. Functional outcomes of acute medical with functional decline in elderly patients admitted to acute illness and hospitalization in older persons. Arch Intern Med 1996;156: care medical wards. Research on larger populations is 2. Fried TR, Bradley EH, Williams CS et al. Functional disability and health care 26. World Health Organization. International Drug Monitoring: The Role of the expenditures for older persons. Arch Intern Med 2001;161:2602–2607.
Hospital. Geneva, Switzerland: World Health Organization, 1969. WHO 3. Covinsky KE, Palmer RM, Fortinsky RH et al. Loss of independence in ac- tivities of daily living in older adults hospitalized with medical illnesses: In- 27. Heikkinen E, Waters WE, Brzezinski ZJ. The Elderly in Eleven Countries: A creased vulnerability with age. J Am Geriatr Soc 2003;51:451–458.
Sociodemographic Survey. Copenhagen, Denmark: World Health Organiza- 4. Lindley CM, Tully MP, Paramsothy V et al. Inappropriate medication is a major cause of adverse drug reactions in elderly patients. Age Ageing 1992;21:294–300.
28. Corsonello A, Pedone C, Corica F et al. Concealed renal insufficiency and 5. Landi F, Russo A, Liperoti R et al. Impact of inappropriate drug use on physical adverse drug reactions in elderly hospitalized patients. Arch Intern Med performance among a frail elderly population living in the community. Eur J 29. Hosmer D, Lemeshow S. Applied Logistic Regression. New York, NY: John 6. Lin HY, Liao CC, Cheng SH et al. Association of potentially inappropriate medication use with adverse outcomes in ambulatory elderly patients with 30. Hanlon JT, Fillenbaum GG, Kuchibhatla M et al. Impact of inappropriate drug chronic diseases: Experience in a Taiwanese medical setting. Drugs Aging use on mortality and functional status in representative community dwelling 7. Beers MH, Ouslander JG, Rollingher I et al. Explicit criteria for determining 31. Ravaglia G, Forti P, Lucicesare A et al. Development of an easy inappropriate medication use in nursing home residents. Arch Intern Med prognostic score for frailty outcomes in the aged. Age Ageing 2008;37: 8. Beers MH. Explicit criteria for determining potentially inappropriate medi- 32. Hoogerduijn JG, Schuurmans MJ, Duijnstee MS et al. A systematic review of cation use by the elderly. An update. Arch Intern Med 1997;157:1531–1536.
predictors and screening instruments to identify older hospitalized patients at 9. Fick DM, Cooper JW, Wade WE et al. Updating the Beers criteria for poten- risk for functional decline. J Clin Nurs 2007;16:46–57.
tially inappropriate medication use in older adults: Results of a US consensus 33. Cacciatore F, Abete P, Maggi S et al. Disability and 6-year mortality in panel of experts. Arch Intern Med 2003;163:2716–2724.
elderly population. Role of visual impairment. Aging Clin Exp Res 2004;16: 10. Aparasu RR, Mort JR. Inappropriate prescribing for the elderly: Beers criteria- based review. Ann Pharmacother 2000;34:338–346.
34. Classen DC, Pestotnik SL, Evans RS et al. Adverse drug events in hospitalized 11. Willcox S, Himmelstein D, Woolhandler S. Inappropriate drug prescribing for patients: Excess length of stay, extra costs, and attributable mortality. JAMA the community-dwelling elderly. JAMA 1994;272:292–296.
12. Spore D, Mor V, Larrat P et al. Inappropriate drug prescriptions for elderly 35. Lassmann-Vague V. Hypoglycaemia in elderly diabetic patients. Diabetes residents of board and care facilities. Am J Public Health 1997;87:404–409.
Metab 2005;31(Spec No 2):5S53–5S57.
13. Hanlon JT, Schmader KE, Boult C et al. Use of inappropriate prescription 36. Mukai S, Lipsitz LA. Orthostatic hypotension. Clin Geriatr Med 2002;18: drugs by older people. J Am Geriatr Soc 2002;50:26–34.
14. Onder G, Landi F, Liperoti R et al. Impact of inappropriate drug use among 37. Lazarus BA, Murphy JB, Coletta EM et al. The provision of physical hospitalized older adults. Eur J Clin Pharmacol 2005;61:453–459.
activity to hospitalized elderly patients. Arch Intern Med 1991;151:2452– 15. Onder G, Landi F, Cesari M et al. Investigators of the GIFA Study. Inappro- priate medication use among hospitalized older adults in Italy: Results from 38. McLean AJ, Le Couteur DG. Aging biology and geriatric clinical pharmacol- the Italian group of pharmacoepidemiology in the elderly. Eur J Clin Pharma- ogy. Pharmacol Rev 2004;56:163–184.
39. Singh H, Watt K, Veitch R et al. Malnutrition is prevalent in hospitalized 16. Wong RY, Miller WC. Adverse outcomes following hospitalization in acutely medical patients: Are housestaff identifying the malnourished patient? Nutri- ill older patients. BMC Geriatr 2008;8:10.
17. Carosella L, Pahor M, Pedone C et al. Pharmacosurveillance in hospitalized 40. Lancaster KJ, Smiciklas-Wright H, Heller DA et al. Dehydration in patients in Italy: Study design of the Gruppo Italiano di Farmacovigilanza black and white older adults using diuretics. Ann Epidemiol 2003;13: nell’Anziano (GIFA). Pharmacol Res 1999;40:287–295.
18. Pahor M, Chrischilles EA, Guralnik JM et al. Drug data coding and analysis in 41. Wieland D, Ferrucci L. Multidimensional geriatric assessment: Back to the epidemiologic studies. Eur J Epidemiol 1994;10:405–411.
future. J Gerontol A Biol Sci Med Sci 2008;63A:272–274.
19. PHS-HCF. International Classification of Diseases, Ninth Revision, Clinical 42. Siebens H, Aronow H, Edwards D et al. A randomized controlled trial of Modifications. Washington, DC: Public Health Service, Health Care Financing exercise to improve outcomes of acute hospitalization in older adults. J Am 20. Folstein MF, Folstein SE, McHigh PR. ‘‘Mini mental state’’: A practical method 43. Gallagher PF, Barry PJ, Ryan C et al. Inappropriate prescribing in an acutely ill for grading the cognitive state of patients for the clinician. J Psychiatr Res population of elderly patients as determined by Beers’ criteria. Age Ageing 21. Lesher EL, Berryhill JS. Validation of the Geriatric Depression Scale-Short 44. O’Mahony D, Gallagher PF. Inappropriate prescribing in the older population: Form among inpatients. J Clin Psychol 1994;50:256–260.
Need for new criteria. Age Ageing 2008;37:138–141.
22. Katz S, Ford AB, Moskowitz RW et al. Studies of illness in aged: The index of 45. ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial ADL, a standardized measure of biological and psychosocial function. JAMA fibrillation. J Am Coll Cardiol 2006;48:e149–e246.
46. Gallagher P, Barry P, O’Mahony D. Inappropriate prescribing in the elderly.
23. Lawton MP, Brody EM. Assessment of older people: Self-maintaining and J Clin Pharm Ther 2007;32:113–121.
instrumental activities of daily living. Gerontologist 1969;9:179–186.
47. Hosia-Randell HM, Muurinen SM, Pitka¨la¨ KH. Exposure to potentially 24. Antonelli Incalzi R, Corsonello A, Pedone C et al. Construct validity of ac- inappropriate drugs and drug-drug interactions in elderly nursing home tivities of daily living scale: A clue to distinguish the disabling effects of COPD residents in helsinki, Finland: A cross-sectional study. Drugs Aging 2008;25: and congestive heart failure. Chest 2005;127:830–838.
25. Conwell Y, Forbes NT, Cox C et al. Validation of a measure of physical illness 48. Di Fazio I, Franzoni S, Frisoni GB et al. Predictive role of single diseases and burden at autopsy: The Cumulative Illness Rating Scale. J Am Geriatr Soc their combination on recovery of balance and gait in disabled elderly patients.
J Am Med Dir Assoc 2006;7:208–211.

Source: http://www.geriatria.unimo.it/PDF/Mussi%20paper.pdf