L E T T E R T O T H E E D I T O R
Comment on the article by Eshetie et al. entitled
“Potentially
inappropriate medication use and related hospital admissions
in aged care residents: The impact of dementia
”
Dear Editor,
We really appreciated the work by Eshetie et al1who analysed
the prevalence of potentially inappropriate medications (PIMs) according to the 2019 AGS Beers and 2015 STOPP criteria at admis-sion and at discharge in 181 patients aged≥75 years with polytherapy (mean drugs at admission 10) and discharged to residential care facilities, stratified according to history of cognitive impairment. This study showed, once more, the high burden of inappropriate prescrib-ing (IP) in daily clinical practice, recordprescrib-ing an extremely high preva-lence of STOPP PIMs at discharge irrespective of presence or absence of dementia (79.1% vs. 85.6%, respectively), without significant changes during hospital admission. Moreover, a possible or probable PIM-related hospitalization was recorded in 45 patients (24.9%) at admission, in 75.6% of cases involving STOPP PIMs. The authors report conflicting results of previous studies on the impact of hospital admission on PIM prevalence, including a Spanish cohort study on 200 patients discharged from an acute geriatric ward (GW) that did not demonstrate a change in PIM use.1
At this regard, we would like to report the results of two papers we recently published on this topic, investigating the prevalence and clinical implications of polypharmacy and IP among patients aged ≥65 years discharged from three internal medicine wards (IMWs) and two GWs of two tertiary hospitals in North-western Italy.2,3In the cross-sectional part of the study, we investigated the prevalence of, and factors associated with, PIMs and potential prescribing omissions (PPOs) identified at discharge according to 2015 STOPP/START criteria.2 Every patient underwent a multidimensional assessment, including standardized scales for frailty, cognitive impairment, func-tional status, and comorbidity. We collected a sample of 726 patients (mean age 81.5 years, 36% discharged to medium/long-term care facilities, MLTCFs) with a high burden of comorbidities and geriatric syndromes (28.5% had moderate to severe cognitive impairment).2
Table 1 compares principal characteristics of our sample2and that by Eshetie et al1; because of less stringent inclusion criteria, our sample
was generally younger and presumably had better overall health sta-tus, resulting in a lower number of prescribed medications (mean 7.5 vs.11) and a lower prevalence of STOPP PIMs (54.4% vs. 82.3%) at discharge. However, this gap significantly narrows when considering patients with hyperpolypharmacy, that presented a 74.4% STOPP PIM prevalence in our sample.2 Indeed, at multivariate analysis, a
higher number of drugs at discharge was associated with presence of
at least one STOPP PIM (OR 1.22, 95% CI 1.15–1.28), whereas GW discharge was protective (OR 0.55, 95% CI 0.40–0.75). The same associations were recorded also for PPOs.2
Even if we did not perform a formal analysis on prescribing trends during hospital stay, it is interesting to note that, while the mean number of drugs at admission was superimposable in patients from IMWs and GWs (6.8 ± 3.5 vs. 6.3 ± 3.2), IMW-discharged patients showed a significantly higher number of drugs than GW-discharged ones (8.2 ± 3.3 vs. 6.9 ± 2.9), as well as higher prevalence of PIMs (63.8% vs. 45.1%) and PPOs (53.9% vs. 35.2%).2Several factors may
contribute to these findings, including the recognized experience of geriatricians in older patients' prescribing recommendations. Another option could be the higher prevalence of patients with dementia observed in GWs in our sample2; in the work by Esethie et al., patients
with dementia had lower STOPP PIM prevalence than subjects without dementia (83.5% vs. 93.3%).1
In the prospective part of our study, we evaluated whether PIMs and PPOs were associated with 6-month mortality and unplanned hospitalization in the overall sample, and according to discharge set-ting, excluding terminally ill patients.3 Among 611 patients studied,
25.0% died and 30.9% were readmitted at least once. Not unexpect-edly, mortality was higher in MLTCF-discharged patients (32.1% vs. 21.4%), while rehospitalization rates were lower (24.4% vs. 34.3%). Neither PIMs nor PPOs were associated with mortality, while a higher number of STOPP PIMs was independently associated with readmissions in the overall sample (OR 1.23, 95% CI 1.03–1.46) and in home-discharged patients (OR 1.38, 95% CI 1.13–1.68). Among MLTCF-discharged patients, a higher drug burden (OR 1.27, 95% CI 1.13–1.42) and neurological comorbidities (excluding dementia) were associated with unplanned readmission.3The lack of an association
between STOPP PIMs and rehospitalizations in MLTCF-discharged patients might be due to local management of prescriptions and adverse events.
The main strength of our study is its prospective design and the systematic use of standardized scales, which reduces the bias inherent to the retrospective collection of information derived from real-world clinical practice and allows to correct for the potential confounding effect of geriatric syndromes. Yet, some potential limitations should be discussed. First, the prevalence of STOPP PIMs could be over-estimated both in our study2,3and in that by Eshetie et al1because
the use of STOPP criteria in very old people with geriatric syndromes, Received: 9 December 2020 Revised: 28 January 2021 Accepted: 29 January 2021
DOI: 10.1111/bcp.14761
TAB L E 1 Main chara cteristics of the pro tocols and subject s stud ied by Esh etie et al 1 and by Bo et a l 2 Eshetie et al 1 Bo et al 2 Setting and population Patients aged ≥ 75 years admitted to general medical wards, taking ≥ 5 medications and discharged to long-term care facilities, not in palliative care. Recruitment site: 5 major hospitals of South Australia Recruitment period: June – July 2017 Patients aged ≥ 65 years admitted to internal medicine or geriatric wards, irrespective of number of medications, discharge setting and palliative care status. Recruitment site: 2 major hospitals of North-western Italy Recruitment period: March – June 2017 PIM criteria applied 2015 STOPP criteria (62 out of 80 criteria evaluated) 2019 AGS Beers criteria (excluding “Drugs to be used with caution in older adults ”) 2015 STOPP criteria (80 out of 80 criteria evaluated) No evaluation of Beers criteria Sample Overall With dementia Without dementia Overall Number of patients 181 91 90 726 Age, years N.A. 87 (81.7 – 90.9) a 88.4 (82.9 – 92.1) a 81.5 (7.0) b Female gender, n (%) 99 (54.7%) 47 (51.7%) 52 (57.8%) 347 (47.8%) Dementia diagnosis, n (%) 91 (50.3%) -207 (28.5%) Admitted from (medium-or) long-term care facility, n (%) 135 (74.6%) 67 (73.6%) 68 (75.6%) 70 (9.7%) Discharge to (medium-or) long-term care facility, n (%) 181 (100%) c 91 (100%) c 90 (100%) c 261 (36.0%) Number of drugs at admission 10 (N.A.) b 9.5 (3.5) b 10.9 (3.4) b 6.5 (3.4) b Number of drugs at discharge 11 (N.A.) b 10.2 (3.4) b 11.4 (3.5) b 7.5 (3.2) b Polypharmacy prevalence at admission, n (%) 181 (100%) c 91 (100%) c 90 (100%) c 598 (82.4%) Hyperpolypharmacy prevalence at admission, n (%) 83 (45.9%) 36 (39.6%) 47 (52.2%) 133 (18.3%) Prevalence of at least one Beers or STOPP PIM at discharge, n (%) N.A. N.A. (86%) N.A. (92.2%) N.A. c Prevalence of at least one Beers PIM at discharge, n (%) 154 (85.1%) 77 (84.6%) 77 (85.6%) N.A. c Prevalence of at least one STOPP PIM at discharge, n (%) 149 (82.3%) 72 (79.1%) 77 (85.6%) 395 (54.4%) Abbreviations: AGS, American Geriatrics Society; N.A., not available; STOPP, Screening Tool of Older Persons' Prescriptions. aMedian (25th – 75th percentile). bMean (standard deviation). cAccording to study protocol.
multimorbidity and polypharmacy may induce a high number of PIMs that the geriatrician in charge may consider irrelevant for the single patient. In recent years, several criteria were suggested for use in vul-nerable patients, including the STOPPFrail, recently updated.4Despite
their name, these criteria were developed for use in older patients approaching the end of life, with the main goal to reduce drug-related morbidity and burden without compromising health or quality of life.4 Their application requires careful evaluation of end of life status (target patients typically meet all the following: functional dependency and/or severe chronic disease and/or terminal illness; severe irrevers-ible frailty; high risk of acute medical complications and clinical deterioration; death in the following 12 months not surprising)4; thus,
STOPPFrail were not evaluated in either study.1–3
Our study might have been underpowered to identify a direct relationship between PIMs and cause of admission, also considering that STOPP PIMs were responsible of just 18.8% of total hospitaliza-tions in the sample of institutionalised older patients studied by Eshetie et al.1
Even if interventions on PIM recognition and reconciliation are meant to reduce drug-related hospital admissions, we could demon-strate only an association with reduced all-cause hospitalization in the overall sample and in home-discharged patients. In real-world older patients with multiple chronic conditions, it may be extremely challenging to identify a direct causal association between PIMs and unplanned readmissions, and this rigorous methodological approach would probably underestimate the potential clinical benefit of PIM reduction. Indeed, multiple concurring factors intervening in a short period of time may make it difficult to trace back the first pharmaco-logical harm that triggered the sequence of events leading to hospitali-zation. Therefore, even if the lower hospitalization rate in patients without STOPP PIMs we observed in our study cannot be fully ascribed to reduced drug-related hospital admissions, these findings maintain some clinical relevance, demonstrating once more that adher-ence to well-acknowledged criteria of appropriate prescribing in older subjects is associated with overall clinical benefit. In accordance with previous studies,5–9and combining the strengths of our work (i.e., a large sample size and correction for important prognostic variables)2,3 with the proof of a direct implication of PIMs in hospital admission by Eshetie et al1there is a strong rationale that reducing polypharmacy and IP may reduce hospitalizations in vulnerable older patients. The results of recently concluded randomized clinical trials10,11will help clinicians to disentangle the impact of STOPP PIM reduction on hard clinical outcomes, including drug-related admissions.
A C K N O W L E D G E M E N T
No funding was received for the present work.
C O M P E T I N G I N T E R E S T S
There are no competing interests to declare.
C O N T R I B U T O R S
All authors contributed to the writing and revision of the manuscript and approved the final manuscript.
Gianluca Isaia Enrico Brunetti Mario Bo
Section of Geriatrics, Department of Medical Sciences, University of Turin, A.O.U. Città della Salute e della Scienza, Molinette, Turin, Italy
Correspondence Enrico Brunetti, Section of Geriatrics, Department of Medical Sciences, University of Turin, A.O.U. Città della Salute e della Scienza, Molinette, corso Bramante 88, 10126, Turin, Italy. Email: enrico.brunetti@unito.it
O R C I D
Gianluca Isaia https://orcid.org/0000-0002-0249-9170
Enrico Brunetti https://orcid.org/0000-0003-2028-2319
Mario Bo https://orcid.org/0000-0003-2945-0243
R E F E R E N C E S
1. Eshetie TC, Roberts G, Nguyen TA, Gillam MH, Maher D, Kalisch Ellett LM. Potentially inappropriate medication use and related hospital admissions in aged care residents: the impact of dementia. Br J Clin Pharmacol. 2020;86(12):2414-2423. https://doi.org/10. 1111/bcp.14345
2. Bo M, Gibello M, Brunetti E, et al. Prevalence and predictors of inap-propriate prescribing according to STOPP/START criteria version 2 in older patients discharged from geriatric and internal medicine wards: a prospective observational multicenter study. Geriatr Gerontol Int. 2019;19(1):5-11. https://doi.org/10.1111/ggi.13542
3. Brunetti E, Aurucci ML, Boietti E, et al. Clinical implications of poten-tially inappropriate prescribing according to STOPP/START version 2 criteria in older polymorbid patients discharged from Geriatric and Internal Medicine Wards: a prospective observational multicenter study. J Am Med Dir Assoc. 2019;20(11):1476.e1-1476.e10. https:// doi.org/10.1016/j.jamda.2019.03.023
4. Curtin D, Gallagher P, O'Mahony D. Deprescribing in older people approaching end-of-life: development and validation of STOPPFrail version 2. Age Ageing. 2020;afaa159. Epub ahead of print. https://doi. org/10.1093/ageing/afaa159
5. Moriarty F, Bennett K, Cahir C, Kenny RA, Fahey T. Potentially inap-propriate prescribing according to STOPP and START and adverse outcomes in community-dwelling older people: a prospective cohort study. Br J Clin Pharmacol. 2016;82(3):849-857. https://doi.org/10. 1111/bcp.12995
6. Wallace E, McDowell R, Bennett K, Fahey T, Smith SM. Impact of potentially inappropriate prescribing on adverse drug events, health related quality of life and emergency hospital attendance in older people attending general practice: a prospective cohort study. J Gerontol A Biol Sci Med Sci. 2017;72(2):271-277. https://doi.org/10. 1093/gerona/glw140
7. Fabbietti P, Di Stefano G, Moresi R, et al. Impact of potentially inap-propriate medications and polypharmacy on 3-month readmission among older patients discharged from acute care hospital: a prospec-tive study. Aging Clin Exp Res. 2018;30(8):977-984. https://doi.org/ 10.1007/s40520-017-0856-y
8. van Der Stelt CA, Windsant-van den Tweel AV, Egberts AC, et al. The association between potentially inappropriate prescribing and medication-related hospital admissions in older patients: a nested case control study. Drug Saf. 2016;39(1):79-87. https://doi.org/10. 1007/s40264-015-0361-1
9. Counter D, Millar JWT, McLay JS. Hospital readmissions, mortality and potentially inappropriate prescribing: a retrospective study of older adults discharged from hospital. Br J Clin Pharmacol. 2018;84(8): 1757-1763. https://doi.org/10.1111/bcp.13607
10. Lavan AH, O'Mahony D, Gallagher P, et al. The effect of SENATOR (Software ENgine for the Assessment and optimisation of drug and non-drug Therapy in Older peRsons) on incident adverse drug reac-tions (ADRs) in an older hospital cohort - Trial Protocol. BMC Geriatr. 2019;19(1):1-12. https://doi.org/10.1186/s12877-019-1047-9
11. Crowley EK, Sallevelt BTGM, Huibers CJA, et al. Intervention protocol: OPtimising thERapy to prevent avoidable hospital Admis-sion in the Multi-morbid elderly (OPERAM): a structured medication review with support of a computerised decision support system. BMC Health Serv Res. 2020;20(1):220. https://doi.org/10.1186/ s12913-020-5056-3
