Age and Ageing Advance Access published July 13, 2015 Age and Ageing 2015; 0: 1–6 doi: 10.1093/ageing/afv079
© The Author 2015. Published by Oxford University Press on behalf of the British Geriatrics Society. All rights reserved. For Permissions, please email: [email protected]
Potentially inappropriate prescribing detected by STOPP–START criteria: are they really inappropriate? ISABEL LOZANO-MONTOYA1, MANUEL VÉLEZ-DIAZ-PALLARÉS2, EVA DELGADO-SILVEIRA2, BEATRIZ MONTERO-ERRASQUIN1, ALFONSO JOSE CRUZ JENTOFT1 1
Servicio de Geriatria, Hospital Universitario Ramón y Cajal, Madrid, Spain Servicio de Farmacia Hospitalaria, Hospital Universitario Ramón y Cajal, Madrid, Spain
2
Abstract Background: the STOPP–START criteria were developed to detect potentially inappropriate prescribing (PIP) in older people. The reasons why multidisciplinary geriatric teams decide not to follow STOPP–START criteria have not been studied. Objective: to analyse compliance with the recommendations of the STOPP–START criteria in older inpatients. Design: ambispective, non-randomised study. Subjects setting: three hundred and eighty-eight consecutive patients aged 80 years or over admitted to the acute geriatric medicine unit of a University hospital. Methods: STOPP–START criteria were systematically used by a pharmacist to assess pre-admission treatments, and the multidisciplinary geriatric team decided what drugs were recommended after discharge. Two researches independently assessed how many STOPP–START recommendations were accepted by the team, and if they were not accepted, why. Results: two hundred and eighty-four PIPs were identified (0.8 per subject) according to STOPP criteria. Two hundred and forty-seven of these prescriptions (87.0%) were discontinued at discharge. STOPP recommendations were not accepted in 37 cases, mostly because the team considered other therapeutic priorities (lorazepam, n = 12; risperidone, n = 5; other, n = 18). Three hundred and ninety-seven PIPs were identified according to START criteria (1.1 per subject). START recommendations were not followed at discharge in 264 cases (66.5%). The most frequent reasons were as follows: severe disability (n = 90), the use of other effective treatments for the condition (n = 38) and high risk of severe adverse effects (n = 32). Not following START criteria was significantly associated with dependency for basic activities of daily living (ADLs) (odds ratio, OR: 0.66 for compliance with a recommendation; 0.49–0.89), dependency for instrumental ADLs (OR: 0.64; 0.48–0.85) or inability to walk (OR: 0.72; 0.54–0.98). Conclusions: potentially inappropriate drugs are usually discontinued, but many older hospitalised patients do not receive potentially recommended medications. More research on the reasons and consequences of this fact is needed. Keywords: potentially inappropriate prescribing, older people, STOPP/START
Introduction Quality and safety of medication use in older people are an increasing problem in usual medical practice. Age-related changes in pharmacokinetics and pharmacodynamics, multiple medical co-morbidities, polypharmacy and inadequate nutritional status increase the risk for older patients to have adverse drug events (ADEs) [1, 2, 3]. Potentially inappropriate prescribing (PIP) encompasses (i) the use of medications in a situation in which the risk of an ADE outweighs the clinical benefit (particularly when safer
or more effective medications are available) and (ii) the omission of clinically indicated medications in the absence of contraindication in patients with significant life expectancy [4]. Inappropriate prescribing (IP) is common in older patients [5, 6] and is associated with ADEs [7], hospitalisation and death [8, 9, 10], functional impairment [11] and wasteful utilisation of resources [12]. Explicit criteria to detect PIP have been developed to identify and improve the use of medication in older people [13, 14, 15, 16]. The concept of PIP implies that the final
1
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
Address correspondence to: I. Lozano-Montoya. Tel/Fax: (+34) 913368172. Email: [email protected]
I. Lozano-Montoya et al.
Methods Subjects and setting
This was an ambispective, non-randomised study. Data were collected from 388 consecutive patients aged 80 years or over, admitted with acute illness to the acute geriatric medicine unit of the Hospital Universitario Ramón y Cajal in Madrid, Spain. This public hospital serves an urban catchment area of around 650,000 persons. All patients were admitted via the Emergency Department, following referral by their general practitioner (GP) or self-referral. Patients were excluded if they were admitted to hospital in the previous year or were cared as outpatients in the geriatric clinic. Readmissions during the study period were also excluded. Patients who were transferred from other medical or surgical wards for comprehensive geriatric assessment were not considered. Use of explicit criteria
On admission of each patient to the acute geriatric unit, a clinical pharmacist who is an expert on the use of STOPP–START
2
criteria and in the interpretation of medical records obtains a comprehensive drug history from the patient, caregiver and other available sources (GP and both hospital and primary care medical records) and reviews pre-admission treatments to detect PIPs and POs using STOPP–START criteria. Results are discussed within the multidisciplinary geriatric team, and decisions on drug use or discontinuation are usually taken by consensus within the team. At hospital discharge, the same pharmacist informs patients and caregivers about the recommended discharge treatment and gives them computergenerated written information for all drugs prescribed [21]. For this study, and after patients had been discharged, two researchers who were not part of the attending team (a geriatrician and a clinical pharmacist trained in the use of explicit criteria) independently collected STOPP–START criteria at admission for each patient, based on all the information available in clinical records, and analysed whether the recommendations of the explicit criteria had been followed by the geriatric team. When recommendations on starting or stopping drugs had not been followed by the geriatric team, the reason of this decision was recorded. Possible reasons for non-compliance were defined from the literature, and all recommendations that were not followed were classified by consensus between three geriatricians and two clinical pharmacists. (Supplementary data, Appendix S1, available in Age and Ageing online). In the use of STOPP–START criteria, some minor issues were considered: • Acenocoumarol was included as anticoagulant treatment (warfarin is rarely used in Spain) and new oral anticoagulants with approved indications for these pathologies were considered in STOPPA9, STOPPA15, STOPPA16, STOPPA17, STOPPE5, STARTA1 and STARTA2. • Triflusal was considered an antiplatelet therapy: STOPPA17, STARTA3. • Any type of laxative prescription was considered as well as a fibre diet: START D2. • Walter and Covinsky [30] approach was used to evaluate life expectancy: STARTA5. Independent variables collected were age, gender and number of prescribed medications prior to admission (those prescribed before the hospital physician’s intervention for >3 months or with no ending date) and at hospital discharge. All prescribed drugs with pharmacologically active ingredients were included in the analysis. Oral nutritional supplements and eye drops were excluded. Multivitamins in pills were considered as a single active drug. Comprehensive geriatric assessment data were registered: Cumulative Illness Rating Scale for Geriatrics (CIRS-G) [31], Katz Index of Independence in Activities of Daily Living (Katz), Lawton Instrumental Activities of Daily Living Scale (Lawton), Mini Nutritional Assessment (MNA), Global Deterioration Scale for dementia (GDS) and Functional Ambulation Category (FAC). Number of falls in the preceding 3 months, living in a nursing home and deaths during admission were also collected.
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
judgment about the appropriateness of a drug in a patient is made by the attending physician, considering all the factors involved, including costs [17] and patient preferences and experiences [18]. PIPs, according to the Screening Tool of Older People’s Prescriptions (STOPP) criteria, are identified in 35–77% of older inpatients [19–22, 23]. The Screening Tool to Alert to Right Treatment (START) criteria detect potential prescribing omissions (PPOs) in 44–65% of hospitalised older people [24, 25]. Standard geriatric care uses comprehensive geriatric assessment, which includes medication review, but changes in drug prescription in older hospitalised patients are reduced by using comprehensive geriatric assessment [26]. Adding pharmaceutical care to the multidisciplinary team of an acute geriatric care unit may enable one to detect a great number of PIPs and ADEs and to optimise patients’ medical treatment, although evidence is still weak [27, 28]. However, compliance with pharmacist recommendations is low. Recently, a specially developed structured pharmacist review of medication intervention supported by computerised decision support systems was shown to improve both the appropriateness and accuracy of medication regimens of older hospitalised inpatients, but compliance with recommendations was only 54.8% [29]. To the best of our knowledge, there are no studies that analyse the rates and reasons why multidisciplinary geriatric teams with expertise in the use of drugs in older people decide not to follow the recommendations from explicit criteria. The objective of this study was to analyse the appropriateness of the STOPP–START criteria in older inpatients using a multidisciplinary comprehensive assessment as the gold standard.
PIP detected by STOPP–START criteria Descriptive statistics were used: frequencies for categorical measures and means and standard deviations for continuous measures. Contingency tables were created for discrete variables. Avalue of P < 0.05 was considered statistically significant.
Results
Table 1. Characteristics of the study population (n = 346) Characteristics
N
Mean ± SD or proportion
........................................ Age (years) Sex (women) Living in nursing home Faller (one or more falls in the 3 months prior to assessment) Died in hospital Cumulative Illness Rating Scale for Geriatrics (CIRS-G)* ADL (Katz) Severe ADL disability (Katz: F-G) IADL (Lawton) Severe IADL disability (Lawton: 0–2) Cognitive status (Global Deterioration Scale, GDS) Severe dementia (GDS: 6–7) FAC index Walking disability (FAC: 0–1) Nutritional assessment (MNA) At risk of malnutrition (MNA: 8–11) Malnourished (MNA: 0–7)
346 241 86 80
88.5 ± 6.0 69.7% 24.9% 23.1%
42 345
10.8% 2.5 ± 0.4
330 158 312 215 290 94 318 137 209 87 89
2.6 ± 2.3 47.9% 1.5 ± 2.4 68.9% 3.9 ± 2.1 32.4% 2.4 ± 2.0 43.1% 8.0 ± 3.4 41.6% 42.6%
Table 2. Compliance with the most prevalent STOPP– START criteria Compliance (%)
n
........................................ STOPP Benzodiazepines in those prone to falls (H1) Duplicate drug classes ( J) Loop diuretic as first-line monotherapy for hypertension (A3) Calcium channel blockers with chronic constipation (A8) Aspirin at dose >150 mg per day (A12) Long-term (i.e. >1 month), long-acting benzodiazepines (B7) Long-term use of NSAID (>3 months) for pain in osteoarthtitis (E4) Neuroleptic drugs (H2) Aspirin with no history of coronary, cerebral or peripheral arterial symptoms (A13) Long-term (i.e. >1 month) neuroleptics as long-term hypnotics (B8) START Calcium and vitamin D supplement in patients with known osteoporosis (E3) ACE inhibitor with chronic heart failure (A6) Statin therapy in diabetes mellitus if there are coexisting major cardiovascular risk factors present (F4) Antiplatelet therapy in diabetes mellitus with coexisting major cardiovascular risk factors (F3) Metformin with type 2 diabetes ± metabolic syndrome (F1) ACE inhibitor following acute myocardial infarction (A7) Warfarin in the presence of chronic atrial fibrillation (A1) Aspirin/clopidogrel with a history of atherosclerotic vascular disease in patients with sinus rhythm (A3) ACE inhibitor or angiotensin receptor blocker in diabetes with nephropathy (F2) Antihypertensive therapy where systolic blood pressure is consistently >160 mmHg (A4) Statin therapy with vascular disease and life expectancy >5 years (A5) Regular inhaled corticosteroid for moderate/severe asthma or COPD (B2) Regular inhaled 2-agonist or anticholinergic agent for mild-to-moderate asthma or COPD (B1) Fibre supplement for chronic, symptomatic diverticular disease with constipation (D2)
33 28 21
52 100 81
19
79
17 16
100 100
13
100
12 9
58 100
9
100
88
38
57 49
44 14
39
23
24
17
21
52
20
25
20
25
16
25
14
57
10
30
10
50
8
38
6
67
3
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
Three hundred and eighty-eight patients were included (88.8 ± 6.0 years, 68.3% female); 42 of them died in hospital (data excluded from analysis as they were not discharged with a drug treatment). Twenty-five per cent of the subjects lived in an institutional setting prior to hospital admission. Most of the subjects were severely dependent for activities of daily living (ADLs) (47.9%) and suffered from malnutrition (42.6%). The mean number of medications prescribed per patient decreased from admission (8.1 ± 3.7) to discharge (7.0 ± 3.3). The main characteristics of the study population are presented in Table 1. One hundred and seventy subjects out of 346 (49.1%) were using at least one PIP on admission according to STOPP criteria and 212 subjects (61.3%) had a PPO according to START criteria. Two hundred and eighty-four PIPs were identified (0.8 per subject) according to STOPP criteria, of which 247 (87.0%) were discontinued at discharge. The most frequently encountered STOPP criteria were the use of benzodiazepines in those prone to falls (STOPP H1) (n = 33, 11.6%), duplicate drug classes (STOPP J) (n = 28, 9.9%) and loop diuretic as first-line monotherapy for hypertension (STOPP A3) (n = 21, 7.4%) (Table 2). The suggestion to stop a drug was not accepted in 37 cases (13.0%) because of other therapeutic priorities (lorazepam, n = 12; risperidone, n = 5; other drugs, n = 18), palliative care (n = 1) and severe disability (n = 1) (Table 3). Three hundred and ninety-seven PPOs were identified according to START criteria (1.1 per subject). The most
frequent were calcium/vitamin D supplementation with osteoporosis (START E3) (n = 88, 22.2%), angiotensinconverting enzyme (ACE) inhibitor with chronic heart failure (START A6) (n = 57, 14.4%) and statin therapy with diabetes mellitus and coexisting cardiovascular risk factor (START F4) (n = 49, 12.3%). The START recommendations to introduce new drugs were not followed in 264 cases at discharge (66.5%). The most frequent reasons identified for not following recommendations to start drugs were advanced physical disability (n = 90), use of other effective treatment (n = 38) and high risk of severe adverse effects (n = 32) (Table 3). The
I. Lozano-Montoya et al. Table 3. Reasons for not following STOPP–START recommendations Criteria
Followed recommendation
Cause
Total
........................................ STOPP (317)
START (441)
Died in hospital No
Total No Total Yes Died in hospital No
Total (general)
Severe mental or physical disability Other treatment used for the disease High risk of adverse events Contraindicated drug or allergy Palliative care Conflicting drugs Unjustified cause Therapeutic prioritisation Patient refusal/no compliance
33 35 1 1 37 247 44 90 38 32 22 20 20 20 16 6 264 133 758
rates of compliance with START criteria were highly variable for different criteria (Table 2). Neither physical nor mental disabilities were associated with the degree of compliance of STOPP recommendations. Not following START criteria was significantly associated with being dependent for basic ADLs (OR: 0.66 for compliance with the recommendation; 0.49–0.89), dependent for instrumental ADLs (OR: 0.64; 0.48–0.85) or unable to walk (OR: 0.72; 0.54–0.98).
Discussion Explicit instruments to detect PIPs are widely used and recommended by scientific societies [13, 32]. Physicians usually decide to stop or start drugs that are potentially inappropriate medication (PIM) or PPO considering all circumstances in individual patients, including patient preferences. However, we have found no studies that address the reasons for not following STOPP/START criteria. In this study, performed in geriatric inpatients, we have shown that explicit recommendations to discontinue PIPs are usually followed at discharge. However, two-thirds of the recommendations to start potentially appropriate treatments are not followed in this population due to different clinical reasons. Our prevalence results are consistent with previous studies of inappropriate prescribing using STOPP and START criteria that report prevalence rates of 35–77% for PIPs [19] and
4
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
Total No Total Yes
Therapeutic prioritisation Severe mental or physical disability Palliative care
58–65% for POs [24, 25]. The most frequently encountered STOPP criteria at hospital admission were the use of benzodiazepines in those at risk of recurrent falls (STOPP H1) and duplicate drug-class prescriptions (STOPP J). These results are consistent with previous studies [19]. Some contextual aspects may influence prescription. For instance, benzodiazepine use is higher in Spain than in most European countries [33] Lorazepam was the most frequent drug used in our patients that was not discontinued. However, its use was mostly limited to specific clinical conditions, mainly to treat insomnia for short periods of time at low doses (0.5 mg/day). The second drug that was usually not discontinued in those at risk of recurrent falls was risperidone. It was maintained at the lowest effective dose in patients to treat behavioural and psychological symptoms of dementia. A previous meta-analysis has shown a small but significant benefit with aripiprazole, olanzapine and risperidone in treating this syndrome [34]. Risperidone is approved for this indication in most European countries. For most other drugs, recommendations based on STOPP criteria were followed by the medical team, with a low rate of non-compliance. The most frequent PPOs at admission included calcium/ vitamin D supplementation in patients with osteoporosis (START E3), ACE inhibitors with chronic heart failure (START A6) and statin therapy with diabetes mellitus and coexisting cardiovascular risk factor (START F4). These results are similar to previous studies [19]. Two hundred and sixty-four PPOs remained at discharge. In some instances, not starting a potentially useful drug detected by the START criteria may be caused by the publication of new evidence after the explicit criteria were released. For instance, a recent meta-analysis has suggested a lack of benefit of calcium supplements in preventing fractures in community-dwelling individuals [35]. These results as well as their possible side effects such as cardiovascular events [36], kidney stones [37], constipation and gastrointestinal distress had led some authors to recommend against their use [38]. An open debate on the benefits of vitamin D exists [39]. This may lead many physicians not to start supplements of calcium/vitamin D, even if the reason is not specifically stated in the records. Compliance with recommendations based on START criteria was highly variable and overall low (Table 2, from 14 to 67% for different recommendations). Reasons for this variability are unclear. In some cases (i.e. statin therapy, F4), the reason may be low or conflicting evidence in old, complex patients [40, 41] or considering that preventive drugs do not have a role in such a population. This may also explain underuse of antiplatelet drugs or anticoagulants. In other cases, perceived risks (metformin, warfarin and ACE inhibitors) of a general thrive to avoid extreme polypharmacy may be present. However, in most instances, the reason seems to be that the population treated in an acute care geriatric unit includes many patients with severe disability, in which a more conservative end-of-life approach may be chosen by patients and caregivers. More research is needed to understand whether this reflects a problem in the concept of using explicit criteria in
PIP detected by STOPP–START criteria complex populations, or a practice that may be improved with education or other intervention. The impact of non-compliance with recommendations on clinical outcomes is unknown.
Limitations
Conclusions To the best of our knowledge, this is the first clinical study to look at the most frequent reasons for not following STOPP/ START criteria in usual clinical practice. Potentially inappropriate drugs are usually discontinued, but many patients do not receive potentially recommended medications. More research on the reasons and consequences of this fact is needed.
Key points • This is the first clinical study to look at the most frequent reasons for not following STOPP/START criteria. • PIPs are usually discontinued. • Two-thirds of the patients do not receive potentially recommended medications due to clinical reasons such as disability.
Acknowledgements All of the authors listed have had a significant role in the project, its execution and analysis, and the writing of the paper. This included the conception and design, analysis and interpretation of data.
Conflicts of interest None declared.
Supplementary data Supplementary data mentioned in the text are available to subscribers in Age and Ageing online.
Following the journal publication rules, only the 30 references are listed here, marked in bold type. The full list of references is given in Supplementary data, Appendix S2, available in Age and Ageing online. 1. Spinewine A, Schmader KE, Barber N et al. Appropriate prescribing in elderly people: how well can it be measured and optimised? Lancet 2007; 370: 173–84. 3. Mangoni AA, Jackson SH. Age-related changes in pharmacokinetics and pharmacodynamics: basic principles and practical applications. Br J Clin Pharmacol 2004; 57: 6–14. 4. Hill-Taylor B, Sketris I, Hayden J, Byrne S, O’Sullivan D, Christie R. Application of the STOPP/START criteria: a systematic review of the prevalence of potentially inappropriate prescribing in older adults, and evidence of clinical, humanistic and economic impact. J Clin Pharm Ther 2013; 38: 360–72. 5. Conejos MM, Sánchez Cuervo M, Delgado Silveira E et al. Potentially inappropriate drug prescription in older subjects across health care settings. Eur Geriatr Med 2010; 1: 9–14. 7. Hamilton H, Gallagher P, Ryan C, Byrne S, O’Mahony D. Potentially inappropriate medications defined by STOPP criteria and the risk of adverse drug events in older hospitalized patients. Arch Intern Med 2011; 171: 1013–9. 8. Dalleur O, Spinewine A, Henrard S, Losseau C, Speybroeck N, Boland B. Inappropriate prescribing and related hospital admissions in frail older persons according to the STOPP and START criteria. Drugs Aging 2012; 29: 829–37. 13. The American Geriatrics Society 2012 Beers Criteria Update Expert Panel. American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc 2012; 60: 616–31. 14. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D. STOPP (Screening Tool of Older Person’s Prescriptions) and START (Screening Tool to Alert doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol Ther 2008; 46: 72–83. 17. Barry AR, Loewen PS, de Lemos J, Lee KG. Reasons for non-use of proven pharmacotherapeutic interventions: systematic review and framework development. J Eval Clin Pract 2012; 18: 49–55. 18. Denford S, Frost J, Dieppe P, Cooper C, Britten N. Individualisation of drug treatments for patients with longterm conditions: a review of concepts. BMJ Open 2014; 4: e004172. 19. Gallagher P, Lang PO, Cherubini A et al. Prevalence of potentially inappropriate prescribing in an acutely ill population of older patients admitted to six European hospitals. Eur J Clin Pharmacol 2011; 67: 1175–88. 20. Gallagher P, O’Mahony D. STOPP (Screening Tool of Older Persons’ potentially inappropriate Prescriptions): application to acutely ill elderly patients and comparison with Beers’ criteria. Age Ageing 2008; 37: 673–9. 21. Delgado-Silveira E, Alvarez-Diaz A, Perez-Menendez-Condea C, Muñoz-García M, Cruz-Jentoft A, Bermejo Vicedo T. Results of integrating pharmaceutical care in an Acute Geriatric Unit. Rev Esp Geriatr Gerontol 2012; 47: 49–54. 22. Onatade R, Auyeung V, Scutt G, Fernando J. Potentially inappropriate prescribing in patients on admission and discharge from an older peoples’ unit of an acute UK hospital. Drugs Aging 2013; 30: 729–37.
5
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
This study has some limitations. First, it was performed in older inpatients admitted to a specialised geriatric unit; therefore, results cannot be generalised to the entire older population in other settings. It was conducted in a single centre, and analysed decisions made by a multidisciplinary geriatric team specialised in the use of drugs in older people; so findings may be different for teams with lower expertise in this topic. Similar studies in different settings are needed to assess the differences in prescribing practices. Finally, we found no consensus lists of reasons for non-compliance of recommendations from explicit criteria. Many expected reasons (i.e. cost, therapeutic inertia and psychological reactance) were not found in our sample or may not have been properly documented in records, a limitation due to the retrospective nature of our study. Prospective studies are needed to better define and understand such reasons.
References
I. Lozano-Montoya et al.
6
33. Huerta C, Abbing-Karahagopian V, Requena G et al. Prevalence of use of benzodiazepines and related drugs in seven European databases: a cross-national descriptive study from the PROTECT-EU Project. In: 29 International Conference on Pharmacoepidemiology Therapeutic Risk Management, Montréal, Canada, 2013. https://boletinbifap. files.wordpress.com/2013/09/protect_bzd_descriptive.pdf (25 June 2014, date last accessed). 34. Maher AR, Maglione M, Bagley S et al. Efficacy and comparative effectiveness of atypical antipsychotic medications for offlabel uses in adults: a systematic review and meta-analysis. JAMA 2011; 306: 1359–69. 35. Reid IR, Bolland MJ. Calcium risk-benefit updated-New WHI analyses. Maturitas 2014; 77: 1–3. 36. Mao PJ, Zhang C, Tang L et al. Effect of calcium or vitamin D supplementation on vascular outcomes: a meta-analysis of randomized controlled trials. Int J Cardiol 2013; 169: 106–11. 38. Reid IR. Should we prescribe calcium supplements for osteoporosis prevention? J Bone Metab 2014; 21: 21–8. 39. Kmietowicz Z. Vitamin D shows no clear evidence of benefits despite hundreds of studies. BMJ 2014; 348: g2489. 40. Stone NJ, Intwala S, Katz D. Statins in very elderly adults (debate). J Am Geriatr Soc 2014; 62: 943–5. 41. Berthold HK, Gouni-Berthold I. Lipid-lowering drug therapy in elderly patients. Curr Pharm Des 2011; 17: 877–93. Received 25 June 2014; accepted in revised form 25 March 2015
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
24. O’Mahony D, Gallagher P, Ryan C et al. STOPP & START criteria: a new approach to detecting potentially prescribing in old age. Eur Geriatr Med 2010; 1: 45–51. 25. Lang PO, Hasso Y, Dramé M et al. Potentially inappropriate prescribing including under-use amongst older patients with cognitive or psychiatric co-morbidities. Age Ageing 2010; 39: 373–81. 26. Larsen MD, Rosholm JU, Hallas J. The influence of comprehensive geriatric assessment on drug therapy in elderly patients. Eur J Clin Pharmacol 2014; 70: 233–9. 27. Spinewine A, Fialová D, Byrne S. The role of the pharmacist in optimizing pharmacotherapy in older people. Drugs Aging 2012; 29: 495–510. 28. Viswanathan M, Kahwati LC, Golin CE et al. Medication therapy management interventions in outpatient settings: a systematic review and meta-analysis. JAMA Intern Med 2015; 175: 76–87. 29. ÒSullivan D, O’Mahony D, O’Connor MN et al. Impact of a structured pharmacist intervention on the appropriateness of prescribing in older hospitalized patients. Drugs Aging 2014; 31: 471–81. 31. Miller MD, Paradis CF, Houck PR et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res 1992; 41: 237–48. 32. Turner G, Clegg A. Best practice guidelines for the management of frailty: a British Geriatrics Society, Age UK and Royal College of General Practitioners report. Age Ageing 2014; 43: 744–7.
© The Author 2015. Published by Oxford University Press on behalf of the British Geriatrics Society. All rights reserved. For Permissions, please email: [email protected]
Potentially inappropriate prescribing detected by STOPP–START criteria: are they really inappropriate? ISABEL LOZANO-MONTOYA1, MANUEL VÉLEZ-DIAZ-PALLARÉS2, EVA DELGADO-SILVEIRA2, BEATRIZ MONTERO-ERRASQUIN1, ALFONSO JOSE CRUZ JENTOFT1 1
Servicio de Geriatria, Hospital Universitario Ramón y Cajal, Madrid, Spain Servicio de Farmacia Hospitalaria, Hospital Universitario Ramón y Cajal, Madrid, Spain
2
Abstract Background: the STOPP–START criteria were developed to detect potentially inappropriate prescribing (PIP) in older people. The reasons why multidisciplinary geriatric teams decide not to follow STOPP–START criteria have not been studied. Objective: to analyse compliance with the recommendations of the STOPP–START criteria in older inpatients. Design: ambispective, non-randomised study. Subjects setting: three hundred and eighty-eight consecutive patients aged 80 years or over admitted to the acute geriatric medicine unit of a University hospital. Methods: STOPP–START criteria were systematically used by a pharmacist to assess pre-admission treatments, and the multidisciplinary geriatric team decided what drugs were recommended after discharge. Two researches independently assessed how many STOPP–START recommendations were accepted by the team, and if they were not accepted, why. Results: two hundred and eighty-four PIPs were identified (0.8 per subject) according to STOPP criteria. Two hundred and forty-seven of these prescriptions (87.0%) were discontinued at discharge. STOPP recommendations were not accepted in 37 cases, mostly because the team considered other therapeutic priorities (lorazepam, n = 12; risperidone, n = 5; other, n = 18). Three hundred and ninety-seven PIPs were identified according to START criteria (1.1 per subject). START recommendations were not followed at discharge in 264 cases (66.5%). The most frequent reasons were as follows: severe disability (n = 90), the use of other effective treatments for the condition (n = 38) and high risk of severe adverse effects (n = 32). Not following START criteria was significantly associated with dependency for basic activities of daily living (ADLs) (odds ratio, OR: 0.66 for compliance with a recommendation; 0.49–0.89), dependency for instrumental ADLs (OR: 0.64; 0.48–0.85) or inability to walk (OR: 0.72; 0.54–0.98). Conclusions: potentially inappropriate drugs are usually discontinued, but many older hospitalised patients do not receive potentially recommended medications. More research on the reasons and consequences of this fact is needed. Keywords: potentially inappropriate prescribing, older people, STOPP/START
Introduction Quality and safety of medication use in older people are an increasing problem in usual medical practice. Age-related changes in pharmacokinetics and pharmacodynamics, multiple medical co-morbidities, polypharmacy and inadequate nutritional status increase the risk for older patients to have adverse drug events (ADEs) [1, 2, 3]. Potentially inappropriate prescribing (PIP) encompasses (i) the use of medications in a situation in which the risk of an ADE outweighs the clinical benefit (particularly when safer
or more effective medications are available) and (ii) the omission of clinically indicated medications in the absence of contraindication in patients with significant life expectancy [4]. Inappropriate prescribing (IP) is common in older patients [5, 6] and is associated with ADEs [7], hospitalisation and death [8, 9, 10], functional impairment [11] and wasteful utilisation of resources [12]. Explicit criteria to detect PIP have been developed to identify and improve the use of medication in older people [13, 14, 15, 16]. The concept of PIP implies that the final
1
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
Address correspondence to: I. Lozano-Montoya. Tel/Fax: (+34) 913368172. Email: [email protected]
I. Lozano-Montoya et al.
Methods Subjects and setting
This was an ambispective, non-randomised study. Data were collected from 388 consecutive patients aged 80 years or over, admitted with acute illness to the acute geriatric medicine unit of the Hospital Universitario Ramón y Cajal in Madrid, Spain. This public hospital serves an urban catchment area of around 650,000 persons. All patients were admitted via the Emergency Department, following referral by their general practitioner (GP) or self-referral. Patients were excluded if they were admitted to hospital in the previous year or were cared as outpatients in the geriatric clinic. Readmissions during the study period were also excluded. Patients who were transferred from other medical or surgical wards for comprehensive geriatric assessment were not considered. Use of explicit criteria
On admission of each patient to the acute geriatric unit, a clinical pharmacist who is an expert on the use of STOPP–START
2
criteria and in the interpretation of medical records obtains a comprehensive drug history from the patient, caregiver and other available sources (GP and both hospital and primary care medical records) and reviews pre-admission treatments to detect PIPs and POs using STOPP–START criteria. Results are discussed within the multidisciplinary geriatric team, and decisions on drug use or discontinuation are usually taken by consensus within the team. At hospital discharge, the same pharmacist informs patients and caregivers about the recommended discharge treatment and gives them computergenerated written information for all drugs prescribed [21]. For this study, and after patients had been discharged, two researchers who were not part of the attending team (a geriatrician and a clinical pharmacist trained in the use of explicit criteria) independently collected STOPP–START criteria at admission for each patient, based on all the information available in clinical records, and analysed whether the recommendations of the explicit criteria had been followed by the geriatric team. When recommendations on starting or stopping drugs had not been followed by the geriatric team, the reason of this decision was recorded. Possible reasons for non-compliance were defined from the literature, and all recommendations that were not followed were classified by consensus between three geriatricians and two clinical pharmacists. (Supplementary data, Appendix S1, available in Age and Ageing online). In the use of STOPP–START criteria, some minor issues were considered: • Acenocoumarol was included as anticoagulant treatment (warfarin is rarely used in Spain) and new oral anticoagulants with approved indications for these pathologies were considered in STOPPA9, STOPPA15, STOPPA16, STOPPA17, STOPPE5, STARTA1 and STARTA2. • Triflusal was considered an antiplatelet therapy: STOPPA17, STARTA3. • Any type of laxative prescription was considered as well as a fibre diet: START D2. • Walter and Covinsky [30] approach was used to evaluate life expectancy: STARTA5. Independent variables collected were age, gender and number of prescribed medications prior to admission (those prescribed before the hospital physician’s intervention for >3 months or with no ending date) and at hospital discharge. All prescribed drugs with pharmacologically active ingredients were included in the analysis. Oral nutritional supplements and eye drops were excluded. Multivitamins in pills were considered as a single active drug. Comprehensive geriatric assessment data were registered: Cumulative Illness Rating Scale for Geriatrics (CIRS-G) [31], Katz Index of Independence in Activities of Daily Living (Katz), Lawton Instrumental Activities of Daily Living Scale (Lawton), Mini Nutritional Assessment (MNA), Global Deterioration Scale for dementia (GDS) and Functional Ambulation Category (FAC). Number of falls in the preceding 3 months, living in a nursing home and deaths during admission were also collected.
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
judgment about the appropriateness of a drug in a patient is made by the attending physician, considering all the factors involved, including costs [17] and patient preferences and experiences [18]. PIPs, according to the Screening Tool of Older People’s Prescriptions (STOPP) criteria, are identified in 35–77% of older inpatients [19–22, 23]. The Screening Tool to Alert to Right Treatment (START) criteria detect potential prescribing omissions (PPOs) in 44–65% of hospitalised older people [24, 25]. Standard geriatric care uses comprehensive geriatric assessment, which includes medication review, but changes in drug prescription in older hospitalised patients are reduced by using comprehensive geriatric assessment [26]. Adding pharmaceutical care to the multidisciplinary team of an acute geriatric care unit may enable one to detect a great number of PIPs and ADEs and to optimise patients’ medical treatment, although evidence is still weak [27, 28]. However, compliance with pharmacist recommendations is low. Recently, a specially developed structured pharmacist review of medication intervention supported by computerised decision support systems was shown to improve both the appropriateness and accuracy of medication regimens of older hospitalised inpatients, but compliance with recommendations was only 54.8% [29]. To the best of our knowledge, there are no studies that analyse the rates and reasons why multidisciplinary geriatric teams with expertise in the use of drugs in older people decide not to follow the recommendations from explicit criteria. The objective of this study was to analyse the appropriateness of the STOPP–START criteria in older inpatients using a multidisciplinary comprehensive assessment as the gold standard.
PIP detected by STOPP–START criteria Descriptive statistics were used: frequencies for categorical measures and means and standard deviations for continuous measures. Contingency tables were created for discrete variables. Avalue of P < 0.05 was considered statistically significant.
Results
Table 1. Characteristics of the study population (n = 346) Characteristics
N
Mean ± SD or proportion
........................................ Age (years) Sex (women) Living in nursing home Faller (one or more falls in the 3 months prior to assessment) Died in hospital Cumulative Illness Rating Scale for Geriatrics (CIRS-G)* ADL (Katz) Severe ADL disability (Katz: F-G) IADL (Lawton) Severe IADL disability (Lawton: 0–2) Cognitive status (Global Deterioration Scale, GDS) Severe dementia (GDS: 6–7) FAC index Walking disability (FAC: 0–1) Nutritional assessment (MNA) At risk of malnutrition (MNA: 8–11) Malnourished (MNA: 0–7)
346 241 86 80
88.5 ± 6.0 69.7% 24.9% 23.1%
42 345
10.8% 2.5 ± 0.4
330 158 312 215 290 94 318 137 209 87 89
2.6 ± 2.3 47.9% 1.5 ± 2.4 68.9% 3.9 ± 2.1 32.4% 2.4 ± 2.0 43.1% 8.0 ± 3.4 41.6% 42.6%
Table 2. Compliance with the most prevalent STOPP– START criteria Compliance (%)
n
........................................ STOPP Benzodiazepines in those prone to falls (H1) Duplicate drug classes ( J) Loop diuretic as first-line monotherapy for hypertension (A3) Calcium channel blockers with chronic constipation (A8) Aspirin at dose >150 mg per day (A12) Long-term (i.e. >1 month), long-acting benzodiazepines (B7) Long-term use of NSAID (>3 months) for pain in osteoarthtitis (E4) Neuroleptic drugs (H2) Aspirin with no history of coronary, cerebral or peripheral arterial symptoms (A13) Long-term (i.e. >1 month) neuroleptics as long-term hypnotics (B8) START Calcium and vitamin D supplement in patients with known osteoporosis (E3) ACE inhibitor with chronic heart failure (A6) Statin therapy in diabetes mellitus if there are coexisting major cardiovascular risk factors present (F4) Antiplatelet therapy in diabetes mellitus with coexisting major cardiovascular risk factors (F3) Metformin with type 2 diabetes ± metabolic syndrome (F1) ACE inhibitor following acute myocardial infarction (A7) Warfarin in the presence of chronic atrial fibrillation (A1) Aspirin/clopidogrel with a history of atherosclerotic vascular disease in patients with sinus rhythm (A3) ACE inhibitor or angiotensin receptor blocker in diabetes with nephropathy (F2) Antihypertensive therapy where systolic blood pressure is consistently >160 mmHg (A4) Statin therapy with vascular disease and life expectancy >5 years (A5) Regular inhaled corticosteroid for moderate/severe asthma or COPD (B2) Regular inhaled 2-agonist or anticholinergic agent for mild-to-moderate asthma or COPD (B1) Fibre supplement for chronic, symptomatic diverticular disease with constipation (D2)
33 28 21
52 100 81
19
79
17 16
100 100
13
100
12 9
58 100
9
100
88
38
57 49
44 14
39
23
24
17
21
52
20
25
20
25
16
25
14
57
10
30
10
50
8
38
6
67
3
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
Three hundred and eighty-eight patients were included (88.8 ± 6.0 years, 68.3% female); 42 of them died in hospital (data excluded from analysis as they were not discharged with a drug treatment). Twenty-five per cent of the subjects lived in an institutional setting prior to hospital admission. Most of the subjects were severely dependent for activities of daily living (ADLs) (47.9%) and suffered from malnutrition (42.6%). The mean number of medications prescribed per patient decreased from admission (8.1 ± 3.7) to discharge (7.0 ± 3.3). The main characteristics of the study population are presented in Table 1. One hundred and seventy subjects out of 346 (49.1%) were using at least one PIP on admission according to STOPP criteria and 212 subjects (61.3%) had a PPO according to START criteria. Two hundred and eighty-four PIPs were identified (0.8 per subject) according to STOPP criteria, of which 247 (87.0%) were discontinued at discharge. The most frequently encountered STOPP criteria were the use of benzodiazepines in those prone to falls (STOPP H1) (n = 33, 11.6%), duplicate drug classes (STOPP J) (n = 28, 9.9%) and loop diuretic as first-line monotherapy for hypertension (STOPP A3) (n = 21, 7.4%) (Table 2). The suggestion to stop a drug was not accepted in 37 cases (13.0%) because of other therapeutic priorities (lorazepam, n = 12; risperidone, n = 5; other drugs, n = 18), palliative care (n = 1) and severe disability (n = 1) (Table 3). Three hundred and ninety-seven PPOs were identified according to START criteria (1.1 per subject). The most
frequent were calcium/vitamin D supplementation with osteoporosis (START E3) (n = 88, 22.2%), angiotensinconverting enzyme (ACE) inhibitor with chronic heart failure (START A6) (n = 57, 14.4%) and statin therapy with diabetes mellitus and coexisting cardiovascular risk factor (START F4) (n = 49, 12.3%). The START recommendations to introduce new drugs were not followed in 264 cases at discharge (66.5%). The most frequent reasons identified for not following recommendations to start drugs were advanced physical disability (n = 90), use of other effective treatment (n = 38) and high risk of severe adverse effects (n = 32) (Table 3). The
I. Lozano-Montoya et al. Table 3. Reasons for not following STOPP–START recommendations Criteria
Followed recommendation
Cause
Total
........................................ STOPP (317)
START (441)
Died in hospital No
Total No Total Yes Died in hospital No
Total (general)
Severe mental or physical disability Other treatment used for the disease High risk of adverse events Contraindicated drug or allergy Palliative care Conflicting drugs Unjustified cause Therapeutic prioritisation Patient refusal/no compliance
33 35 1 1 37 247 44 90 38 32 22 20 20 20 16 6 264 133 758
rates of compliance with START criteria were highly variable for different criteria (Table 2). Neither physical nor mental disabilities were associated with the degree of compliance of STOPP recommendations. Not following START criteria was significantly associated with being dependent for basic ADLs (OR: 0.66 for compliance with the recommendation; 0.49–0.89), dependent for instrumental ADLs (OR: 0.64; 0.48–0.85) or unable to walk (OR: 0.72; 0.54–0.98).
Discussion Explicit instruments to detect PIPs are widely used and recommended by scientific societies [13, 32]. Physicians usually decide to stop or start drugs that are potentially inappropriate medication (PIM) or PPO considering all circumstances in individual patients, including patient preferences. However, we have found no studies that address the reasons for not following STOPP/START criteria. In this study, performed in geriatric inpatients, we have shown that explicit recommendations to discontinue PIPs are usually followed at discharge. However, two-thirds of the recommendations to start potentially appropriate treatments are not followed in this population due to different clinical reasons. Our prevalence results are consistent with previous studies of inappropriate prescribing using STOPP and START criteria that report prevalence rates of 35–77% for PIPs [19] and
4
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
Total No Total Yes
Therapeutic prioritisation Severe mental or physical disability Palliative care
58–65% for POs [24, 25]. The most frequently encountered STOPP criteria at hospital admission were the use of benzodiazepines in those at risk of recurrent falls (STOPP H1) and duplicate drug-class prescriptions (STOPP J). These results are consistent with previous studies [19]. Some contextual aspects may influence prescription. For instance, benzodiazepine use is higher in Spain than in most European countries [33] Lorazepam was the most frequent drug used in our patients that was not discontinued. However, its use was mostly limited to specific clinical conditions, mainly to treat insomnia for short periods of time at low doses (0.5 mg/day). The second drug that was usually not discontinued in those at risk of recurrent falls was risperidone. It was maintained at the lowest effective dose in patients to treat behavioural and psychological symptoms of dementia. A previous meta-analysis has shown a small but significant benefit with aripiprazole, olanzapine and risperidone in treating this syndrome [34]. Risperidone is approved for this indication in most European countries. For most other drugs, recommendations based on STOPP criteria were followed by the medical team, with a low rate of non-compliance. The most frequent PPOs at admission included calcium/ vitamin D supplementation in patients with osteoporosis (START E3), ACE inhibitors with chronic heart failure (START A6) and statin therapy with diabetes mellitus and coexisting cardiovascular risk factor (START F4). These results are similar to previous studies [19]. Two hundred and sixty-four PPOs remained at discharge. In some instances, not starting a potentially useful drug detected by the START criteria may be caused by the publication of new evidence after the explicit criteria were released. For instance, a recent meta-analysis has suggested a lack of benefit of calcium supplements in preventing fractures in community-dwelling individuals [35]. These results as well as their possible side effects such as cardiovascular events [36], kidney stones [37], constipation and gastrointestinal distress had led some authors to recommend against their use [38]. An open debate on the benefits of vitamin D exists [39]. This may lead many physicians not to start supplements of calcium/vitamin D, even if the reason is not specifically stated in the records. Compliance with recommendations based on START criteria was highly variable and overall low (Table 2, from 14 to 67% for different recommendations). Reasons for this variability are unclear. In some cases (i.e. statin therapy, F4), the reason may be low or conflicting evidence in old, complex patients [40, 41] or considering that preventive drugs do not have a role in such a population. This may also explain underuse of antiplatelet drugs or anticoagulants. In other cases, perceived risks (metformin, warfarin and ACE inhibitors) of a general thrive to avoid extreme polypharmacy may be present. However, in most instances, the reason seems to be that the population treated in an acute care geriatric unit includes many patients with severe disability, in which a more conservative end-of-life approach may be chosen by patients and caregivers. More research is needed to understand whether this reflects a problem in the concept of using explicit criteria in
PIP detected by STOPP–START criteria complex populations, or a practice that may be improved with education or other intervention. The impact of non-compliance with recommendations on clinical outcomes is unknown.
Limitations
Conclusions To the best of our knowledge, this is the first clinical study to look at the most frequent reasons for not following STOPP/ START criteria in usual clinical practice. Potentially inappropriate drugs are usually discontinued, but many patients do not receive potentially recommended medications. More research on the reasons and consequences of this fact is needed.
Key points • This is the first clinical study to look at the most frequent reasons for not following STOPP/START criteria. • PIPs are usually discontinued. • Two-thirds of the patients do not receive potentially recommended medications due to clinical reasons such as disability.
Acknowledgements All of the authors listed have had a significant role in the project, its execution and analysis, and the writing of the paper. This included the conception and design, analysis and interpretation of data.
Conflicts of interest None declared.
Supplementary data Supplementary data mentioned in the text are available to subscribers in Age and Ageing online.
Following the journal publication rules, only the 30 references are listed here, marked in bold type. The full list of references is given in Supplementary data, Appendix S2, available in Age and Ageing online. 1. Spinewine A, Schmader KE, Barber N et al. Appropriate prescribing in elderly people: how well can it be measured and optimised? Lancet 2007; 370: 173–84. 3. Mangoni AA, Jackson SH. Age-related changes in pharmacokinetics and pharmacodynamics: basic principles and practical applications. Br J Clin Pharmacol 2004; 57: 6–14. 4. Hill-Taylor B, Sketris I, Hayden J, Byrne S, O’Sullivan D, Christie R. Application of the STOPP/START criteria: a systematic review of the prevalence of potentially inappropriate prescribing in older adults, and evidence of clinical, humanistic and economic impact. J Clin Pharm Ther 2013; 38: 360–72. 5. Conejos MM, Sánchez Cuervo M, Delgado Silveira E et al. Potentially inappropriate drug prescription in older subjects across health care settings. Eur Geriatr Med 2010; 1: 9–14. 7. Hamilton H, Gallagher P, Ryan C, Byrne S, O’Mahony D. Potentially inappropriate medications defined by STOPP criteria and the risk of adverse drug events in older hospitalized patients. Arch Intern Med 2011; 171: 1013–9. 8. Dalleur O, Spinewine A, Henrard S, Losseau C, Speybroeck N, Boland B. Inappropriate prescribing and related hospital admissions in frail older persons according to the STOPP and START criteria. Drugs Aging 2012; 29: 829–37. 13. The American Geriatrics Society 2012 Beers Criteria Update Expert Panel. American Geriatrics Society updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc 2012; 60: 616–31. 14. Gallagher P, Ryan C, Byrne S, Kennedy J, O’Mahony D. STOPP (Screening Tool of Older Person’s Prescriptions) and START (Screening Tool to Alert doctors to Right Treatment). Consensus validation. Int J Clin Pharmacol Ther 2008; 46: 72–83. 17. Barry AR, Loewen PS, de Lemos J, Lee KG. Reasons for non-use of proven pharmacotherapeutic interventions: systematic review and framework development. J Eval Clin Pract 2012; 18: 49–55. 18. Denford S, Frost J, Dieppe P, Cooper C, Britten N. Individualisation of drug treatments for patients with longterm conditions: a review of concepts. BMJ Open 2014; 4: e004172. 19. Gallagher P, Lang PO, Cherubini A et al. Prevalence of potentially inappropriate prescribing in an acutely ill population of older patients admitted to six European hospitals. Eur J Clin Pharmacol 2011; 67: 1175–88. 20. Gallagher P, O’Mahony D. STOPP (Screening Tool of Older Persons’ potentially inappropriate Prescriptions): application to acutely ill elderly patients and comparison with Beers’ criteria. Age Ageing 2008; 37: 673–9. 21. Delgado-Silveira E, Alvarez-Diaz A, Perez-Menendez-Condea C, Muñoz-García M, Cruz-Jentoft A, Bermejo Vicedo T. Results of integrating pharmaceutical care in an Acute Geriatric Unit. Rev Esp Geriatr Gerontol 2012; 47: 49–54. 22. Onatade R, Auyeung V, Scutt G, Fernando J. Potentially inappropriate prescribing in patients on admission and discharge from an older peoples’ unit of an acute UK hospital. Drugs Aging 2013; 30: 729–37.
5
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
This study has some limitations. First, it was performed in older inpatients admitted to a specialised geriatric unit; therefore, results cannot be generalised to the entire older population in other settings. It was conducted in a single centre, and analysed decisions made by a multidisciplinary geriatric team specialised in the use of drugs in older people; so findings may be different for teams with lower expertise in this topic. Similar studies in different settings are needed to assess the differences in prescribing practices. Finally, we found no consensus lists of reasons for non-compliance of recommendations from explicit criteria. Many expected reasons (i.e. cost, therapeutic inertia and psychological reactance) were not found in our sample or may not have been properly documented in records, a limitation due to the retrospective nature of our study. Prospective studies are needed to better define and understand such reasons.
References
I. Lozano-Montoya et al.
6
33. Huerta C, Abbing-Karahagopian V, Requena G et al. Prevalence of use of benzodiazepines and related drugs in seven European databases: a cross-national descriptive study from the PROTECT-EU Project. In: 29 International Conference on Pharmacoepidemiology Therapeutic Risk Management, Montréal, Canada, 2013. https://boletinbifap. files.wordpress.com/2013/09/protect_bzd_descriptive.pdf (25 June 2014, date last accessed). 34. Maher AR, Maglione M, Bagley S et al. Efficacy and comparative effectiveness of atypical antipsychotic medications for offlabel uses in adults: a systematic review and meta-analysis. JAMA 2011; 306: 1359–69. 35. Reid IR, Bolland MJ. Calcium risk-benefit updated-New WHI analyses. Maturitas 2014; 77: 1–3. 36. Mao PJ, Zhang C, Tang L et al. Effect of calcium or vitamin D supplementation on vascular outcomes: a meta-analysis of randomized controlled trials. Int J Cardiol 2013; 169: 106–11. 38. Reid IR. Should we prescribe calcium supplements for osteoporosis prevention? J Bone Metab 2014; 21: 21–8. 39. Kmietowicz Z. Vitamin D shows no clear evidence of benefits despite hundreds of studies. BMJ 2014; 348: g2489. 40. Stone NJ, Intwala S, Katz D. Statins in very elderly adults (debate). J Am Geriatr Soc 2014; 62: 943–5. 41. Berthold HK, Gouni-Berthold I. Lipid-lowering drug therapy in elderly patients. Curr Pharm Des 2011; 17: 877–93. Received 25 June 2014; accepted in revised form 25 March 2015
Downloaded from http://ageing.oxfordjournals.org/ at University of Cambridge on August 13, 2015
24. O’Mahony D, Gallagher P, Ryan C et al. STOPP & START criteria: a new approach to detecting potentially prescribing in old age. Eur Geriatr Med 2010; 1: 45–51. 25. Lang PO, Hasso Y, Dramé M et al. Potentially inappropriate prescribing including under-use amongst older patients with cognitive or psychiatric co-morbidities. Age Ageing 2010; 39: 373–81. 26. Larsen MD, Rosholm JU, Hallas J. The influence of comprehensive geriatric assessment on drug therapy in elderly patients. Eur J Clin Pharmacol 2014; 70: 233–9. 27. Spinewine A, Fialová D, Byrne S. The role of the pharmacist in optimizing pharmacotherapy in older people. Drugs Aging 2012; 29: 495–510. 28. Viswanathan M, Kahwati LC, Golin CE et al. Medication therapy management interventions in outpatient settings: a systematic review and meta-analysis. JAMA Intern Med 2015; 175: 76–87. 29. ÒSullivan D, O’Mahony D, O’Connor MN et al. Impact of a structured pharmacist intervention on the appropriateness of prescribing in older hospitalized patients. Drugs Aging 2014; 31: 471–81. 31. Miller MD, Paradis CF, Houck PR et al. Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale. Psychiatry Res 1992; 41: 237–48. 32. Turner G, Clegg A. Best practice guidelines for the management of frailty: a British Geriatrics Society, Age UK and Royal College of General Practitioners report. Age Ageing 2014; 43: 744–7.
