Prevalence and types of potentially inappropriate prescriptions among older and middle-aged community-dwelling Albanian patients

Abstract

BACKGROUND:

Prevalence of potentially inappropriate prescriptions (PIP) varies among community-dwelling patients and this difference partly depends on the methods used to evaluate improper use.

OBJECTIVE:

The aim of our study was to assess the prevalence and type of PIP among community-dwelling elderly as well as among middle-aged people by applying three different explicit tools – Ghent Older People’s Prescriptions community Pharmacy Screening (GheOP³S) tool, The European Union Potentially Inappropriate Medications EU(7)-PIM list and PRescribing Optimally in Middle-aged People’s Treatments (PROMPT).

METHODS:

Cross-sectional study among community-dwelling patients in Tirana (Albania) from 1 March to 1 June 2018. Two community pharmacists reviewed the prescriptions independently. The PIP index and the mean number of PIP / patient was calculated.

RESULTS:

In total, 241 participants were included in the study. Among 142 elderly patients, 55.6 % and 54.2 % of them were prescribed at least one PIP, respectively according to the GheOP³S tool and the EU7 PIM list. After applying PROMPT criteria, we found at least one PIP in 33.3 % of the middle-aged patients. The most commonly represented PIP drugs groups were cardiovascular drugs among older people and non-steroidal anti-inflammatory drugs among middle-aged people. Reminding and proposing the patient to undergo yearly influenza vaccination was the most common potentially prescribing omission.

CONCLUSIONS:

Prevalence of PIP is higher in elderly patients and slightly different according to the tool used for detection. However, PIP are also common in the middle-aged population.

Keywords

Potentially inappropriate prescriptions community-dwelling patients prevalence Albania

1 Introduction

Potentially inappropriate prescribing has become a global issue and is often associated with poorer health-related quality of life and adverse health outcomes [1–3]. Potentially inappropriate prescriptions (PIP) include drugs prescribed without an evidence-based indication, which present unfavourable risk-benefit or not cost-effective ratios, and are associated with adverse events or reactions [4]. Prevalence of inappropriate medication use varies among community-dwelling patients and this difference partly depends on the methods used to evaluate improper use [5]. A recent systematic review regarding potentially inappropriate prescriptions among community-dwelling older people across Europe estimated an overall PIP prevalence of 22.6 %, with a broad range (0.0–98.0 %) and 82 different screening tools [6]. The most extensively used explicit tools are STOPP/START and Beers criteria [7–9], last updated in 2014 and 2015, respectively.

A new explicit screening tool - Ghent Older People’s Prescriptions community Pharmacy Screening (GheOP³S) - was developed to be used for PIP screening in the typical community pharmacy practice, adapted to the European market. Studies suggest that GheOP³S-tool is an effective screening method for PIP and the clinical relevance of PIP-items detected is high [10, 11]. The European Union (EU)(7)-PIM (Potentially Inappropriate Medications) list is another screening tool, developed in 2015 with participation of experts from seven European countries. It includes 282 drugs from 34 therapeutic groups considered as PIM for older people and suggestions for dose adjustments and therapeutic alternatives [12].

The majority of published screening tools for PIP detection focus on older people, nonetheless suboptimal prescribing is an important issue even in other population groups. PRescribing Optimally in Middle-aged People’s Treatments (PROMPT) is the first set of prescribing criteria developed for use in middle-aged adults. This tool has been published in 2014 by researchers from United Kingdom and Ireland for its application to prescribing datasets [13]. For PIP prevention and improved patients' outcomes, current emphasis is put on continuous assessment of tools and interventions used [14].

Our objective was to assess the prevalence and type of PIP among community-dwelling elderly as well as among middle-aged people by applying three different explicit tools – GheOP³S tool, EU(7)-PIM list and PROMPT.

2 Methods

We conducted a cross-sectional study in community-dwelling patients in Tirana (Albania) from 1 March to 1 June 2018. Two groups of patients were included in our study: elderly (≥65 years) and middle-aged patients (40–64 years). The patients took their medications from eight different pharmacies located in different neighborhoods of the city.

Independently, two community pharmacists reviewed the prescriptions and collected data regarding demographic characteristics, drugs prescribed and diagnoses. The pharmacological group of each drug and its ATC code was recorded. Patients were interviewed when necessary regarding different aspects of their treatment. The GheOP³S tool and EU(7)-PIM list were both used for PIP identification among elderly patients; PROMPT criteria were applied instead for PIP detection among middle-aged patients.

Independent study variables were sex, age, number of drugs prescribed, and number of diagnosed diseases. Dependent variables were number of PIP detected by each set of criteria and number of patients with at least one PIP in their prescribed treatment.

We also calculated the PIP index dividing the number of detected PIPs by the total number of prescribed drugs, [15], in order to allow comparison of prescribing appropriateness between different studies. Similarly, the mean number of PIP / patient was calculated for each screening tool, in order to compare the results.

3 Results

The study population overall included 241 participants, slightly more women (58.1 %). The mean age was 68.9 years old (SD 10.6) and nearly half of the patients were prescribed four drugs. The most common pathology diagnosed in both groups (elderly and middle-aged) was hypertension. The detailed description is found in Tables 1 and 2.

Table 1
Description of the study population

Variable Category N %

Sex Men 101 41.9 %

Women 140 58.1 %

Age (years) 40 – 54 17 7.1 %

55 – 64 82 34.0 %

65 – 74 61 25.3 %

75 – 84 60 24.9 %

≥85 21 8.7 %

Mean (SD) 68.9 (10.6)

Median (Range) 68 (45 – 92)

Number of prescribed drugs 4 121 50.2 %

5 - 6 102 42.3 %

7 – 8 16 6.6 %

9 – 11 2 0.8 %

Mean (SD) 4.81 (1.07)

Median (Range) 4 (3 – 11)

Number of diagnosed diseases 1 24 9.9 %

2 134 55.6 %

3 51 21.2 %

4 26 10.8 %

5, 6 6 2.5 %

Mean (SD) 2.41 (0.91)

Median (Range) 2 (1 – 6)

Variable	Category	N	%
Sex	Men	101	41.9 %
	Women	140	58.1 %
Age (years)	40 – 54	17	7.1 %
	55 – 64	82	34.0 %
	65 – 74	61	25.3 %
	75 – 84	60	24.9 %
	≥85	21	8.7 %
	Mean (SD)	68.9 (10.6)
	Median (Range)	68 (45 – 92)
Number of prescribed drugs	4	121	50.2 %
	5 - 6	102	42.3 %
	7 – 8	16	6.6 %
	9 – 11	2	0.8 %
	Mean (SD)	4.81 (1.07)
	Median (Range)	4 (3 – 11)
Number of diagnosed diseases	1	24	9.9 %
	2	134	55.6 %
	3	51	21.2 %
	4	26	10.8 %
	5, 6	6	2.5 %
	Mean (SD)	2.41 (0.91)
	Median (Range)	2 (1 – 6)

Table 2

Distribution of pathologies in the elderly (≥65 years) and middle-aged (40–64 years) patients

Pathology	Number of elderly patients (N = 142)	%	Number of middle-aged patients (N = 99)	%
Hypertension	129	90.8 %	77	77.8 %
Dyslipidemia	36	25.4 %	20	20.2 %
Diabetes	31	21.8 %	27	27.3 %
Atrial fibrillation	29	20.4 %	6	6.1 %
Angina pectoris	27	19.0 %	8	8.1 %
Cardiac chronic failure	24	16.9 %	8	8.1 %
Coronary arteries disease	16	11.3 %	8	8.1 %
Glaucoma	11	7.7 %	3	3.0 %
Asthma	9	6.3 %	7	7.1 %
Rheumatoid arthritis	8	5.6 %	5	5.1 %
Benign prostatic hypertrophy	8	5.6 %	2	2.0 %
Chronic obstructive pulmonary disease	5	3.5 %	4	4.0 %
Depression	4	2.8 %	0	0 %

132 different drugs were prescribed (1127 in total). Hydrochlorothiazide (ATC-code: C03AA03) was the most prescribed drug (73 patients; 30.3 % of the participants), followed by enalapril (61 patients or 25.3 % of the patients) and atenolol. Table 3 shows the top ten most prescribed drugs and their ATC code.

Table 3

Frequency of top ten drug prescriptions

ATC code	Drug	Number of patients	Frequency of drug prescription (%)
C03AA03	Hydrochlorothiazide	73	6,48
C09AA02	Enalapril	61	5,41
C07AB03	Atenolol	61	5,41
C03CA01	Furosemide	46	4,08
C09CA01	Losartan	46	4,08
N02BA01	Acetylsalicylic acid	44	3,9
A10BA02	Metformin	40	3,55
C08CA13	Lercanidipine	39	3,46
C08CA01	Amlodipine	35	3,11
C10AA05	Atorvastatin	33	2,93

After applying PROMPT criteria, we found at least one PIP in 33.3 % of the middle-aged patients. Two or more PIP were identified in 13 patients, and the highest number of PIP prescribed to the same patient was 5.

According to the GheOP³S tool, 55.6 % of the elderly patients were prescribed at least one PIP. One of the patients received 6 PIP, and three other patients, 4 PIP.

When the EU7 PIM list was used for PIP identification, we found that 54.2 % of the elderly patients received at least one PIM. This figure was slightly lower than the frequency found by the GheOP³S tool. One patient had four simultaneous PIM, whereas four other patients were prescribed three PIM. Results about the number of PIP identified are provided in Table 4.

Table 4

Number of PIP / patient

	Number of PIP / patient	N	%
PIP according to PROMPT	0	66	66.7 %
	1	20	20.2 %
	2 or more	13	13.1 %
PIP according to GheOP³S tool	0	63	44.4 %
	1	48	33.8 %
	2 or more	31	21.8 %
PIP according to EU7 PIM list	0	65	45.8 %
	1	54	38.0 %
	2 or more	23	16.2 %

The mean number of PIP / patient was respectively: for PROMPT 0.64 (63 PIP / 99 patients), for GheOP³S tool 0.75, and for EU7 PIM list 0.89.

Table 5 summarizes the types of PIP identified according to the GheOP³S tool. When this tool was used to identify them, 126 PIP were found, without calculating the most common potentially prescribing omission (reminding and proposing the patient to undergo yearly influenza vaccination). Indeed, none of the older patients included in our study was advised from the pharmacist regarding influenza vaccination. PIP index was calculated 0.182 (126 / 691).

Table 5

Frequency of potentially inappropriate prescriptions according to the GheOP³S tool

PIP according to the GheOP³S tool	N	N / Total number of PIP (%)	N / Number of prescriptions with that drug (%)
Part 1a: Potentially inappropriate drugs, independent of diagnosis— drug classes
Any antidepressant≥1 year	4	3.17	50.0
Any antipsychotic≥1 month	2	1.60	25.0
Any drug for arterial vascular disorders	11	8.73
Any intermediate acting benzodiazepine at full dose or any dose≥30 subsequent days (lorazepam, alprazolam, clonazepam, temazepam)	4	3.17	28.6
Any short- or long-acting benzodiazepine	7	5.56	87.5
Any long-acting sulfonylurea derivative (glibenclamide etc.)	12	9.52	63.2
Any oral NSAID	7	5.56	10.0
Any recently marketed drug	1	0.79	100.0
Any sedating antihistaminic drug	2	1.60	66.7
Part 1b: Potentially inappropriate drugs, independent of diagnosis— specific molecules
Digoxin > 0.125 mg/day	20	15.87	80.0
Dipyridamole monotherapy (without ASA)	1	0.79	50.0
Methyldopa	7	5.56	50.0
Phenobarbital	1	0.79	100.0
Theophylline	13	10.32	68.4
Part 2a: potentially inappropriate drugs, dependent on diagnosis— drug classes
Not found
Part 2b: potentially inappropriate drugs, dependent on diagnosis— specific molecules
Not found
Part 3: PPOs
The patient is taking oral corticosteroids for 1 month and is not prescribed Ca / Vitamin D supplementation	1	0.79	7.69
The patient is not reminded and proposed to undergo yearly influenza vaccination	142	–	–
Part 4: DDIs of specific relevance
Vitamin K Antagonist + Oral NSAIDs	1	0.79	1.22
RAAS Inhibitor + Potassium sparing diuretic (spironolactone or triamterene)	6	4.76	2.9
Oral NSAID + Oral corticosteroid	1	0.79	1.2
Oral NSAID + Diuretic	4	3.17	6.9
RAAS Inhibitor + Oral NSAID	4	3.17	1.5
Oral antidiabetics / Insulin + Cardio selective beta-blocker	15	11.9	6.36
Combination of anticholinergic drugs	2	1.60	10.0

Abbreviations: NSAIDs – Non-Steroidal Anti-Inflammatory Drugs; ASA – Acetylsalicylic acid; Ca – Calcium; RAAS Inhibitor – Renin-Angiotensin-Aldosterone-System Inhibitor; PPO – Potentially Prescribing Omission; DDIs – Drug-Drug Interactions.

According to the GheOP³S tool, digoxin with a dose higher than 0.125 mg/day was inappropriate in twenty patients, followed by the combination of oral antidiabetics or insulin with cardio selective beta-blockers in 15 patients.

According to the EU7 PIM list, 106 out of 691 drugs were identified as potentially inappropriate (PIP index 0.153), provided in Table 6. After applying EU7 PIM list, the drug most commonly responsible for PIP was digoxin, which was potentially inappropriately prescribed in all patients (20), making up 18.9 % of all PIP. In 16 patients, nifedipine was considered potentially inappropriate for the increased risk of hypotension and myocardial infarction. Other drugs involved in PIP were theophylline (13 patients) and glibenclamide (12 patients).

Table 6

Frequency of potentially inappropriate prescriptions according to the EU7 PIM list

PIM according to EU 7 PIM list	Main reason	N	N / Total number of PIP (%)	N / Number of prescriptions with that drug (%)
Glibenclamide	Risk of protracted hypoglycemia	12	11.3	63.2
Acenocoumarol	Risk of bleeding, especially in people with difficult control of INR value	7	6.6	70.0
Dipyridamole	Risk of vasodilatation and orthostatic hypotension	2	1.9	100.0
Digoxin	Elevated glycoside sensitivity in older people (women > men); risk of intoxication	20	18.9	80.0
Amiodarone	Associated with QT interval problems and risk of provoking torsade de pointes	10	9.4	76.9
Trimetazidine	Can cause or worsen parkinsonian symptoms (tremor, akinesia, hypertonia)	1	0.9	100.0
Nifedipine	Increased risk of hypotension; myocardial infarction	16	15.1	84.2
Verapamil	May worsen constipation; risk of bradycardia	6	5.7	0.5
Diltiazem	May worsen constipation; risk of bradycardia	6	5.7	0.5
Chlorpromazine	Risk of orthostatic hypotension and falls; may lower seizure thresholds in patients with seizures or epilepsy	1	0.9	100.0
Risperidone > 6 weeks	Problematic risk-benefit profile for the treatment of behavioral symptoms of dementia; increased mortality, with higher dose, in patients with dementia	1	0.9	33.3
Diazepam	Risk of falling with hip fracture; prolonged reaction times; psychiatric reactions (can also be paradoxical, e.g. agitation, irritability, hallucinations, psychosis); cognitive impairment; depression	4	3.8	100.0
Lorazepam > 1 mg/day	Risk of falling with hip fracture; prolonged reaction times; psychiatric reactions (can also be paradoxical, e.g. agitation, irritability, hallucinations, psychosis); cognitive impairment; depression	3	2.8	23.1
Amitriptyline	Peripheral anticholinergic side effects (e.g. constipation, dry mouth, orthostatic hypotension, cardiac arrhythmia); central anticholinergic side effects (drowsiness, inner unrest, confusion, other types of delirium); cognitive deficit; increased risk of falling	2	1.9	33.3
Fluoxetine	CNS side effects (nausea, insomnia, dizziness, confusion); hyponatremia	1	0.9	100.0
Piracetam	No efficacy proven; unfavorable risk/benefit profile	1	0.9	100.0
Theophylline	Higher risk of CNS stimulant effects	13	12.3	68.4

Drugs that contributed more often to potentially inappropriate prescribing among middle-aged according to PROMPT criteria were NSAIDs (naproxen, ibuprofen) for more than 3 months, present in 22.2% of prescriptions in 14 patients suffering from rheumatoid arthritis. Other drugs were oral corticosteroids for more than 3 months without concomitant bisphosphonates (11 patients) and benzodiazepines for more than 4 weeks (9 patients diagnosed with cancer). The PIP index for the PROMPT criteria was calculated 0.134 (63 PIP / 469 prescribed drugs).

Table 7 provides an overview of PIP identified by using PROMPT criteria.

Table 7

Frequency of potentially inappropriate prescriptions according to the PROMPT criteria

PIP according to PROMPT	N	N / Total number of PIP (%)	N / Number of prescriptions with that drug
Gastrointestinal system
Esomeprazole or omeprazole in combination with clopidogrel	1	1.59	7.69
Cardiovascular system
Cardio-selective calcium channel blockers (verapamil, diltiazem) in combination with β-blockers	1	1.59	0.7
Respiratory system
Theophylline as monotherapy in asthma or COPD	3	4.76	15.8
Oral corticosteroids for more than 3 months without concomitant bisphosphonates	11	17.5	84.6
Central nervous system
Tricyclic antidepressants TCA (amitriptyline, nortriptyline) as first-line treatment in depression	4	6.35	57.1
Benzodiazepines for a long period (more than 4 weeks)	9	14.3	40.9
Strong opioids (buprenorphine, diamorphine, fentanyl, morphine, oxycodone) without the concomitant prescription of laxatives	5	7.9	71.4
Endocrine system
Oral sulphonyl urea with long duration (glibenclamide) in the management of diabetes	7	11.1	36.8
Musculoskeletal system
Non-Steroidal Anti-Inflammatory (NSAIDs) (celecoxib, diclofenac, naproxen) for more than 3 months	14	22.2	20.0
The use of two drugs in the same pharmacological group
Metformin and Vildagliptin + Metformin	3	4.76	6.8
Lorazepam and Clonazepam	2	3.17	34.4

4 Discussion

Our study has found high prevalence of PIP among the elderly population, similar according to both sets of criteria used to identify them (respectively, 55.6 % according to GheOP³S and 54.2 % according to EU7 PIM list). There are variations among studies regarding the prevalence of PIP in community patients; however, figures are similar to our findings. A study conducted in Brazil [16] showed that the EU7 PIM list had higher sensitivity compared to other tools and PIP prevalence was 59.5 %. Using the same set of criteria, Grina et al. detected a prevalence of 57.2 % PIP among Lithuanian elderly [17].

Using PROMPT criteria, we found a PIP prevalence of 33.3 % among middle-aged patients. Detecting PIP with the same set of criteria, The Irish Longitudinal Study on Ageing (TILDA) found PIP in 42.9% of community-dwelling middle-aged people. Possible explanations for this variation include differences in socio-demographic and health conditions between the study populations, different drug markets and supply among different countries [18].

According to the systematic review by Tommelein et al. [6], drug groups most often involved in PIP among community-dwelling older people across Europe were anxiolytics (ATC-code: N05B), antidepressants (N06A), and nonsteroidal anti-inflammatory and anti-rheumatic products (M01A). In our study, the profile of drugs commonly involved in PIP among older people was different, involving mainly drugs used in cardiovascular diseases such as digoxin, beta-blockers and nifedipine. This result may be partly explained by the different tools used for PIP detection, as well as epidemiological factors of the population and local medicines policies. Surprisingly, among middle-aged participants, identified PIP involved almost the same groups of drugs as in the previous study, precisely NSAIDs (M01A) and benzodiazepines (N05B).

As various studies show, pharmacist-led interventions may improve prescribing appropriateness in community-dwelling older adults [19–21] and using PIP detection tools might help in this activity. Our study suggests that all three screening tools used for PIP identification may be useful instruments for pharmacists in the process of treatment review and optimization, for elderly as well as middle-aged patients.

5 Conclusion

Prevalence of PIP is higher in elderly patients and slightly different according to the tool used for detection. More than half of elderly patients‘ prescriptions are potentially inappropriate (55.6 % GheOP³S vs. 54.2 % EU7 PIM list). However, PIPs are common even in the middle-aged population, approximately 1 in 3 patients when PROMPT criteria are applied. The most commonly represented potentially inappropriate prescriptions drugs groups were cardiovascular drugs among older people and NSAIDs among middle-aged. Reminding and proposing the patient to undergo yearly influenza vaccination was the most common potentially prescribing omission.

Conflict of interest

The authors declare no conflict of interest.

Footnotes

Acknowledgments

We thank all pharmacies that agreed to participate in this study.

References

Wallace

McDowell

Bennett

Fahey

Smith

. 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):271–7.

Brown

Hutchison

Painter

Martin

. Predictive Validity of the Beers and Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions (STOPP) Criteria to Detect Adverse Drug Events, Hospitalizations, and Emergency Department Visits in the United States. J Am Geriatr Soc. 2016;64(1):22–30.

Pérez

Moriarty

Wallace

McDowell

Redmond

Fahey

. Prevalence of potentially inappropriate prescribing in older people in primary care and its association with hospital admission: Longitudinal study. BMJ. 2018;363:k4524.

Hudhra

Beçi

Petrela

Xhafaj

García-Caballos

Bueno-Cavanillas

. Prevalence and factors associated with potentially inappropriate prescriptions among older patients at hospital discharge. J Eval Clin Pract. 2016;22(5):707–13.

Guaraldo

Cano

Damasceno

Rozenfeld

. Inappropriate medication use among the elderly: A systematic review of administrative databases. BMC Geriatr. 2011;11:79.

Tommelein

Mehuys

Petrovic

Somers

Colin

Boussery

. Potentially inappropriate prescribing in community-dwelling older people across Europe: A systematic literature review. Eur J Clin Pharmacol. 2015;71(12):1415–27.

O’Mahony

O’Sullivan

Byrne

O’Connor

Ryan

Gallagher

. STOPP/START criteria for potentially inappropriate prescribing in older people: Version 2. Age and Ageing. 2015;44(2):213–8.

American Geriatrics Society 2015 Beers Criteria Update Expert Panel (2015) American Geriatrics Society 2015 Updated Beers Criteria for Potentially Inappropriate Medication Use in Older Adults. Journal of the American Geriatrics Society. 63(11):2227–46.

Curtin

Gallagher

O’Mahony

. Explicit criteria as clinical tools to minimize inappropriate medication use and its consequences. Ther Adv Drug Saf. 2019;10:2042098619829431.

10.

Tommelein

Petrovic

Somers

Mehuys

van der Cammen

Boussery

. Older patients’ prescriptions screening in the community pharmacy: Development of the Ghent Older People’s Prescriptions community Pharmacy Screening (GheOP³S) tool. J Public Health (Oxf). 2016;38(2):e158–70.

11.

Kympers

Tommelein

Van Leeuwen

Boussery

Petrovic

Somers

. Detection of potentially inappropriate prescribing in older patients with the GheOP³S-tool: Completeness and clinical relevance. Acta Clin Belg. 2019;74(2):126–36.

12.

Renom-Guiteras

Meyer

Thürmann

. The EU(7)-PIM list: A list of potentially inappropriate medications for older people consented by experts from seven European countries. Eur J Clin Pharmacol. 2015;71(7):861–75.

13.

Cooper

Ryan

Smith

, et al. The development of the PROMPT (PRescribing optimally in middle-aged people’s treatments) criteria. BMC Health Serv Res. 2014;14:484.

14.

Clyne

Fitzgerald

Quinlan

Hardy

Galvin

Fahey

Smith

. Interventions to Address Potentially Inappropriate Prescribing in Community-Dwelling Older Adults: A Systematic Review of Randomized Controlled Trials. J Am Geriatr Soc. 2016;64(6):1210–22.

15.

Onatade

Auyeung

Scutt

Fernando

. Potentially inappropriate prescribing in patients on admission and discharge from an older peoples’ unit of an acute UK hospital. Drugs Aging. 2013;30(9):729–37.

16.

Novaes

da Cruz

Lucchetti

ALG

Leite

ICG

Lucchetti

. Comparison of four criteria for potentially inappropriate medications in Brazilian community-dwelling older adults. Geriatr Gerontol Int. 2017;17(10):1628–35.

17.

Grina

Briedis

. The use of potentially inappropriate medications among the Lithuanian elderly according to Beers and EU(7)-PIM list - a nationwide cross-sectional study on reimbursement claims data. J Clin Pharm Ther. 2017;42(2):195–200.

18.

Moriarty

Cahir

Bennett

Hughes

Kenny

Fahey

. Potentially inappropriate prescribing and its association with health outcomes in middle-aged people: A prospective cohort study in Ireland. BMJ Open. 2017;7(10):e016562.

19.

Riordan

Walsh

Galvin

Sinnott

Kearney

Byrne

. The effect of pharmacist-led interventions in optimising prescribing in older adults in primary care: A systematic review. SAGE Open Med. 2016;4:2050312116652568.

20.

Whitman

DeGregory

Morris

Mohile

Ramsdale

. Pharmacist-led medication assessment and deprescribing intervention for older adults with cancer and polypharmacy: A pilot study. Support Care Cancer. 2018;26(12):4105–13.

21.

McNicholl

Gandhi

Hare

Greene

Pierluissi

. A Pharmacist-Led Program to Evaluate and Reduce Polypharmacy and Potentially Inappropriate Prescribing in Older HIV-Positive Patients. Pharmacotherapy. 2017;37(12):1498–506.