Transcript Jane Portlock and Nia Dobson

Can the evidence base shape our
solutions to polypharmacy?
Professor Jane Portlock and Mrs Nia Dobson
De-prescribing………
 A review of Reeve et al., accepted for publication in the British Journal of
Clinical Pharmacology (Aug 2015), proposes the following definition:
 ‘De-prescribing is the process of withdrawal of an inappropriate medication,
supervised by a health care professional with the goal of managing
polypharmacy and improving outcomes’.
Dispensing Health in Later Life – White Paper,
Pharmacy Voice
 GPs believe more than half their patients (50.3%) aged over 75, and taking four or more
medicines, would benefit from taking fewer medicines.
 More than nine in ten GPs (92%) would like more support from community pharmacy
teams to help their patients take their medicines correctly.
 Half (51%) of GPs are not confident their elderly patients are still taking their medicines
as prescribed just three months after their last consultation; this increases dramatically
to eight in 10 (79%) who are not confident they are being taken as prescribed six months
after their last consultation.
De-prescribing tools
 STOPP START (used by NHS Cumbria)
 Beers criteria
 Drug effectiveness summary (NHS Highland)
 NO TEARS (Welsh tool based on NHS Highland/ Cumbria)
 MAI tool (Medication Appropriateness Index)
 The Good Palliative–Geriatric Practice algorithm
 Medstopper
STOPP START
 STOPP: Screening Tool of Older People’s Potentially Inappropriate Prescriptions
 START: Screening Tool to Alert Doctors to Right (i.e. appropriate, indicated)
Treatments
 Updated in 2014 to take into account an expanding therapeutics base.
Beers Criteria
 Updated October 2015.
 Two lists of medications that may be potentially harmful for patients 65yrs+ and not
receiving palliative or hospice care
 Separate guidance on avoiding 13 combinations of medications known to cause
harmful “drug-drug interactions.”
 A list of 20 potentially problematic medications to avoid or for which doses should
be adjusted depending on an older person’s kidney function.
Does it matter which one we use?
All based on evidence of known high risk medicines (especially in elderly) and high risk
combinations.
Locality preference may guide choice e.g. NHS Highland choosing to use
STOPP/START.
As long as based on evidence, is choice of tool important?
What evidence do we have of outcomes from these
tools?
 Some very specific trials for particular populations/medications.
 Outcomes presented mainly in economic terms.
 Little evidence to prove that use of polypharmacy tools improve patient
outcomes such as quality of life, or reduction in unplanned admissions.
The war against polypharmacy – Garfinkel et al.
Study group: 119 disabled patients in six geriatric nursing departments
Control group: 71 patients of comparable age, gender and co-morbidities patients in the same
wards.
After 12 months: assessed whether any change in medications affected the death rate,
referrals to acute care facility, and costs.
 The War Against Polypharmacy: A New, Cost-Effective, Geriatric-Palliative Approach for Improving Drug
Therapy in Disabled Elderly People D. Garfinkel, S. Zur-Gil, J. Ben-Israel
Garfinkel et al. – The results
332 different drugs were discontinued in 119 patients (average of 2.8 drugs/pt)
Not associated with significant adverse effects.
Overall rate of drug discontinuation failure was 18% of all patients and 10% of all drugs.
The 1 year mortality rate was 45% in the control group but only 21% in the study group (P <
0.001, chi-square test).
The patients’ annual referral rate to acute care facilities was 30% in the control group but
only 11.8% in the study group (P < 0.002).
The intervention was associated with a substantial decrease in the cost of drugs.
Limitations of Garfinkel et al.
 Not blinded.
 Small sample size.
 Patients receiving nursing care – may not be representative of patients whose
medications are stopped by GP and who care for themselves.
 All patients were disabled – not representative of the general population.
RCT using STOPP/START criteria
 720 patients over 65 years who were admitted to hospital.
 Randomised to receive usual pharmaceutical care or screening with STOPP/START criteria
on admission followed up with written recommendations to their attending physician.
 Significantly fewer ADRs occurred in the intervention group (n=38; 21 certain/probable)
compared to the control group (n=60; 29 certain/probable).
 Number needed to screen with STOPP/START criteria to prevent 1 in hospital ADR = 16

O’Connor M, O’Sullivan D, Gallagher P et al. Prevention of adverse drug events in hospitalised older patients: A randomised controlled trial
using STOPP/START criteria. European Geriatric Medicine [Internet].2012 September[cited 2015 November 3rd];3(Supplement 1):S132[About 1
screen]. Available from: http://ac.els-cdn.com/S1878764912004664/1-s2.0-S1878764912004664-main.pdf?_tid=89b24c00-8222-11e5-958b00000aacb362&acdnat=1446552235_dc4d245ee4143fad5dec65b45dbbc674
Preventing hospital admissions by reviewing
medication (PHARM) trial.
 An open controlled study in an elderly population.
 Patients with a high risk of medication related hospitalisations based on:
 Old Age
 5+ meds
 Non adherence
 Type of medication used
 Intervention: Patient interview, review of pharmacotherapy and execution and follow up
of pharmaceutical care plan. Carried out by the patients own pharmacist and usual GP.
 Leendertse, A. J., de Koning, G. H. P., Goudswaard, A. N., Belitser, S. V., Verhoef, M., de Gier, H. J., Egberts, A. C. G.
and van den Bemt, P. M. L. A. (2013), Preventing hospital admissions by reviewing medication (PHARM) in primary
care: an open controlled study in an elderly population. Journal of Clinical Pharmacy and Therapeutics, 38: 379–387.
doi: 10.1111/jcpt.12069
PHARM trial
 Control group – received usual care – cared for by another GP other than the intervention GP.
 Primary outcome: Frequency of hospital admissions related to medication within the study period of 12 months for each
patient.
 Secondary outcomes: Survival, Quality of life, and Adverse Drug Events.
 364 intervention patients, 310 control patients.
 Less medication related hospital admissions were found in the intervention group. (n = 6; 1·6%) than in the control group
(n = 10; 3·2%) but the overall effect was not statistically significant (hazard ratio (HR) 0·50, 95% confidence interval (CI)
0·12–1·59).

The secondary outcomes were not statistically significantly different.
 The study was underpowered, which may explain the negative results.
 Post hoc analysis : intervention was statistically significant for patients with five diseases or more:
 Five diseases, HR 0·28 (95% bootstrap CI: 0·056–0·73)
 Eight diseases, HR 0·11 (95% CI: 0·013–0·34).
Medication reviews by Clinical Pharmacists at Hospitals
lead to improved patient outcomes – a systematic review.
 Mini Review – to identify, assess and summarise the literature investigating the effect of pharmacist
led medication review for hospitalised patients.
 31 studies met inclusion criteria, of which 21 were descriptive and 10 were controlled studies. 6 of the
controlled studies (n=10) were RCTs.
 The 10 controlled studies generally showed positive effects on medicine use and cost and satisfaction
with service. There were positive, but insignificant effects on health service use.
 Several outcomes were statistically insignificant – predominantly associated with a low sample size or
low acceptance rates.
 The study concluded that there was a need to ensure rigorous design, large sample sizes and
comparable outcome measures for patient health outcomes.
 Graabæk T, Kjeldsen LJ. Medication Reviews by Clinical Pharmacists at Hospitals Lead to Improved Patient Outcomes: A
Systematic ReviewEur J Hosp Pharm 2013;20:A195 doi:10.1136/ejhpharm-2013-000276.541
Discontinuing statins in terminally ill patients
 Study looked at 381 patients who had been taking statins for 3 months or more for primary or
secondary prevention and were not expected to live longer than 1 year.
 189 were randomised to discontinue statin therapy, 192 to continue.
 Results:
 Proportion of patients who died within 60 days were not statistically significant between groups.
 Survival time was similar between groups.(229 days for discontinuation vs 190 days for continuation).
 No significant difference in time to first CV event – 24 patients experienced a CV event (discontinuation
13, continuation, 11)
 Total QOL score was higher in those who had discontinued therapy.

Kutner JS, Blatchford PJ, Taylor DH, et al. Safety and Benefit of discontinuing Statin Therapy in the Setting of Advanced, Life-Limiting
illness: A randomised clinical trial. JAMA Internal Medicine.2015; 175(5):691-700.
NHS Highland – Six high risk medicines/combinations
 Study identified six high risk medicines/combinations of high risk medicines that should be
targeted to reduce iatrogenic disease in older people.
 n= 3643
 440 patients (94% > 60yrs) pharmacists recommended medicines stopped/amended.
 Of these, GP’s acted on the advice for 214 patients.
 226 patients – no action by GP. Of these, 22 adverse events occurred within a year of the
advice being given, 3 resulting in hospital admission and 19 needing a further GP
consultation.
 Researchers believe 21 of these adverse events could have been avoided, if GP had
followed advice.
NHS Highland – Six high risk medicines/combinations
 When the pharmacists recommendation was accepted and the medicine was stopped or
amended – the original medicine had to be restarted in 16% of cases.
 Reason given in 30 of 35 cases. In all 30, medicines were restarted by patient request due to
loss of symptom control without the medicine.
 When the pharmacists recommendation was acted upon – no adverse events occurred.

Morrison C & MacRae Y. Promoting safer use of high-risk pharmacotherapy: impact of pharmacist-led targeted medication reviews. Drugs —
Real World Outcomes 2015. doi:10.1007/s40801-015-0031-8.
Is polypharmacy always hazardous?
 Retrospective cohort analysis found that admissions were more common for patients with
multiple medications, but admission risk varied with the number of conditions.
 Patients with 1 condition were more likely to have an unplanned admission if they took 4-6
medicines compared with 1-3 medicines. (Odds ratio 1.25, 95%CI 1.11-1.42)
 When the number of medicines increased to 10 or more the odds ratio increased to 3.42(95%CI
2.72 -4.28) when compared to 1-3 medicines.
 In patients with multiple conditions, only the most extreme levels of polypharmacy are associated
with increased risk of admission.
 Authors concluded that the perception that polypharmacy is always hazardous should be
tempered with clinical judgement and assessment of the conditions for which the
medications are prescribed.

Payne RA et al. Is polypharmacy always hazardous? A retrospective cohort analysis using linked electronic health records from primary and secondary
care. Br J Clin Pharmacol 2014; 77:1073–1082.
What next?
 Studies of de-prescribing are in progress internationally.
 In Europe a multinational randomised controlled trial (RCT) that uses electronic decision
support to guide de-prescribing recently started in older people who are taking multiple
drugs for chronic diseases (www.prima-eds.eu/).
 Further trials testing the clinical effects of interventions to reduce polypharmacy are under
way in:
 Australia: (the Opti-Med study, a blinded RCT based on the Good Palliative-Geriatric Practice
algorithm),
 Canada: (Effect of Medication Minimization on Mortality and Hospitalization in Long
Term Care Residents (WiseMed)—an open RCT),
 The Netherlands: (Discontinuing Inappropriate Medication in Nursing Home Residents
(DIM-NHR)—a cluster RCT).