European Geriatric Medicine - Supplement 1 - Volume 7 - October 2016

242 / 290

response of 4/5 (agree/strongly agree) and a 25th centile of ≥4 were
included in the final criteria.
Results: Three Delphi rounds were required. All panellists completed
all rounds. Thirty criteria were proposed for inclusion; 27 were
accepted. No new criteria were added. The first two criteria suggest
deprescribing medications with no indication or where compliance
is poor. The remaining 25 criteria include lipid-lowering therapies,
alpha-blockers for hypertension, anti-platelets, memantine, neuro-
leptics, proton pump inhibitors, H-2 receptor antagonists, anti-
spasmodics, theophylline, leukotriene antagonists, calcium supple-
ments, bone anti-resorptive therapy, selective oestrogen receptor
modulators, non-steroidal anti-inflammatories, corticosteroids, 5-
alpha reductase inhibitors, alpha-1 selective blockers, muscarinic
antagonists, oral diabetic agents, ACE-inhibitors, angiotensin receptor
blockers, systemic oestrogens, multivitamins, nutritional supplements
and prophylactic antibiotics. Anticoagulants and anti-depressants
were excluded. Despite incorporation of panellists ’ suggestions,
acetyl-cholinesterase inhibitors remained inconclusive.
Conclusion: STOPPFrail comprises 27 criteria that are potentially
inappropriate in frailer older patients with limited life expectancy.
STOPPFrail may assist in deprescribing medications in these patients.
P-784
Unavailability of drugs as a cause of inappropriately omitted
antiparkinsonian doses in Parkinson’ s disease inpatients
U. Lertxundi, A. Isla, M.A. Solinís, R. Hernández, S. Domingo-Echaburu,
J. Peral-Aguirregoitia.
Introduction: One of the most important risk factor for motor
function deterioration in Parkinson’ s Disease when patients are
admitted to hospital is wrong timing of antiparkinsonian drug
administration, which has, more recently, shown to prolong hospital
stays in about 4 days.
Method: All PD patients admitted to any of the 11 public acute care
hospitals in the Basque Country during 2011–2012 were included.
Since 2011, all hospitals share a single informatics application (e-
Osabide) for inpatient pharmacotherapeutic management including
electronic administration records. When a registry of no administra-
tion was made for a certain dose, the reason for omission was also
recorded. We chose omitted doses only due to unavailability of the
prescribed drug.
Results: We included 1.628 patients admitted 2.546 times. Of the
80.043 prescribed antiparkinsonian doses, 210 were not administered
because the prescribed drug was not available on time. In all 11 acute
health care hospitals at least one dose could not be administered
because of this reason. The most frequently implicated drug was L-
Dopa in combination (104 doses) followed by pramipexole (31 doses),
ropinirole (30 doses), rotigotine (19 doses), rasagiline (19 doses),
entacapone (3 doses), cabergoline (2 doses) and selegiline (2 doses).
Conclusions: Antiparkinsonian drug omissions occurred in all
hospitals. The most frequent omitted drug was L-Dopa in combination.
In our opinion having all possible drugs on time, as suggested by some
authors, seems unfeasible especially in small hospitals. A therapeutic
interchange protocol where each unavailable drug is converted to the
equivalent levodopa dose could be a realistic solution.
P-785
Danish register-based cohort study on the dosage association
between antidepressants and hyponatremia
K. Leth-Møller1, M. Torstensson1, S.E. Andersen2, G. Gislasson3, C. Torp-
Pedersen4, E.A. Holm1. 1Department of Geriatric Medicine, Nykøbing
Falster Hospital, Nykøbing Falster, 2Clinical Pharmacology Unit,
University Hospital Zealand, Roskilde, 3Department of Cardiology,
Gentofte Hospital, Hellerup, 4Institute of Health, Science and Technology,
Aalborg University, Aalborg, Denmark
Objective: There is a known association between some antidepres-
sants and hyponatremia. In case of drug-induced hyponatremia it is
common practice to reduce the dosage, however the effect of lowering
the dosage is not fully investigated. The aim of this study is to
investigate the association between high and low dosage of the most
commonly used antidepressants and hyponatremia.
Methods: This is an observational Danish register-based cohort
study using nationwide registers. The study cohort consisted of all
residents in The North Denmark Region who initiated treatment with
antidepressants from January 1st 1998 to December 31th 2012. The
association between high and low dosage of different antidepressants
and hyponatremia was analysed using multivariable Poisson regres-
sion models to determine incidence rate ratios (IRR).
Results: There were 91,100 individuals included in the study, of these
13.77% (n = 12,550) had an event of hyponatremia. Using low dose as
reference IRRs and confidence intervals (CI) for the association with
hyponatremia and high doses were for citalopram (40 mg) 1.11 (CI
1.01–1.21); escitalopram (20–40 mg) 1.22 (CI 1.01–1.47); paroxetine
(40–60 mg) 1.57 (CI 1.22–2.01); sertraline (100– 200 mg) 1.96 (CI 1.65–
2.32); venlafaxine (150– 300 mg) 1.21 (CI 0.99–1.49) and mirtazapine
(22.5–45 mg) 1.14 (1.00–1.21).
Key conclusions: Our study showed an increased risk of hyponatremia
for high dosage compared to low dosage for all included antidepres-
sants, except venlafaxine.
P-786
Comparison of tools to reduce polypharmacy on hospital discharge
in a real-world elderly population
M. Debney1, R. Taylor1, K. Elliott 1, C. Lisk 1. 1Barnet Hospital, Royal Free
Nhs Foundation Trust, UK
Introduction: As the UK population ages, survival from chronic illness
increases the prescribing of medications and the burden of poly-
pharmacy. We sought to compare and determine the effectiveness of
validated tools on reducing the number of medications prescribed on
discharge from a cohort of elderly patients.
Methods: All patients over the age of 65, discharged from a UK
hospital between 16th January and 30th January 2015 were identified
from discharge records. Demographics, co-morbidities and total
number of medications on discharge were recorded. The modified
Beers Criteria and the Screening Tool of Older People’ s potentially
inappropriate Prescriptions (STOPP) criteria were retrospectively
applied.
Results: 139 patients (average age of 82 ± 10 years), (45% male) with
multiple co-morbidities (median 3, range 0–8) were identified. Each
patient was prescribed an average of eight medications. Application of
the Beers Criteria in both strict (stopping drugs listed as “inappropriate
medication use”) and broad (stopping medications listed as “inappro-
priate medication use”, “drug-disease interactions ” and “use with
caution”) and STOPP resulted in a significant reduction in average
number of medications on discharge (original 8 ± 4 medication vs.
Beers Strict Criteria 7 ± 4 (paired t-test, p < 0.05) vs. Beers Broad Criteria
7 ± 4, (paired t-test, p < 0.05) vs STOPP 7 ± 3 (paired t-test, p < 0.05)).
Conclusion: This study suggests that the implementation of poly-
pharmacy screening pre-discharge may afford a reduction in medica-
tions and should be considered as part of the discharge process for
elderly patients. Application of STOPP criteria afforded a medication
reduction much more compared to the Beers Criteria.
P-787
Our knowledge of herbal medicines as healthcare professionals
needs to be better
M. Kelshiker1, C. Lisk1. 1Barnet Hospital, Royal Free Nhs Foundation Trust,
United Kingdom
Introduction: Older adults are increasingly using herbal medicines
(HM) to boost longevity, treat pain and boost memory. It is therefore
important for health care professionals (HCPs) to have a working
knowledge about HM.
Methods: A cross sectional web-based survey of HCPs working within
the Department of Medicine at a UK district general hospital was
carried looking at their knowledge of HM.
Poster presentations / European Geriatric Medicine 7S1 (2016) S29–S259 S236