Introduction
Several medications need to be prescribed cautiously in older adults, because of age-related variations in pharmacokinetics and pharmacodynamics (Anathhanam, Powis, Cracknell, & Robson, Reference Anathhanam, Powis, Cracknell and Robson2012). There is consistent evidence implicating inappropriate prescribing as a potential predictor of negative health outcomes, including adverse events, drug interactions, hospital admissions, increasing health care costs, and an increase in both morbidity and mortality in older adults (Page, Linnebur, Bryant, & Ruscin, Reference O’Mahony, O’Sullivan, Byrne, O’Connor, Ryan and Gallagher2010; Spinewine et al., Reference Shinde, Shinde, Khatri and Hande2007). The term “potentially inappropriate medications” (PIMs) refers to those medications that should not be prescribed for most older adults because the risk of adverse events outweighs the clinical benefits, particularly when there is evidence in favour of safer and more effective alternative treatments, including non-pharmacological measures (Sehgal et al., Reference Scott, Gray, Martin, Pillans and Mitchell2013) such as lifestyle modification, weight reduction, regular physical exercises, smoking cessation, and reduction in alcohol consumption (Shinde, Shinde, Khatri, & Hande, Reference Sehgal, Bajwa, Sehgal, Bajaj, Khaira and Kresse2013).
Epidemiology of PIMs
Prescription of PIMs to older adults has received significant consideration worldwide for several decades, and is a pervasive public health concern, with reported figures of 5.2 per cent to more than 85 per cent of older adults being exposed to PIMs (Bala, Narayan, & Nishtala, Reference Bala, Narayan and Nishtala2018). Studies have demonstrated significant associations between inappropriate medication use and higher health care costs (Fick, Reference Fick2001; Fick, Mion, Beers, & Waller, Reference Fick, Mion, Beers and Waller2008; Fu et al., Reference Fu, Jiang, Reeves, Fincham, Liu and Perri2007). A systematic review found that prescription of PIMs had a statistically significant effect on health care utilization, including hospitalization, inpatient and outpatient visits, and emergency department visits among older adults (Hyttinen et al., Reference Holt, Schmiedl and Thürmann2017). The prescription of PIMs in older adults has been on the rise globally, which can be attributed to the widespread increase in the prescription of medications for the management of multiple chronic medical illnesses (Ailabouni, Mangin, & Nishtala, Reference Ailabouni, Mangin and Nishtala2017). A study conducted in community dwelling older adults found that more than half of the study population were prescribed PIMs (Al Odhayani, Tourkmani, Alshehri, Alqahtani, & Mishriky, Reference Al Odhayani, Tourkmani, Alshehri, Alqahtani and Mishriky2017).
Prescribing medications to older adults is complicated in the presence of cognitive decline, multiple morbidities, and frailty (Poudel, Reference Pazan, Weiss and Wehling2015). Clinicians perceive deficits in self-efficacy (relating to knowledge, skills, and decision support), and feasibility (resource availability and work practices) as hindrances to addressing the risk–benefit ratio of prescribing medications to older adults (Anderson, Stowasser, Freeman, & Scott, Reference Anderson, Stowasser, Freeman and Scott2014). When formulating therapeutic aims for older adults, clinicians have to consider the indications for prescribing, the time-to-benefit, co-morbid conditions, concomitant medications, side effects, compliance, patient preferences, and the patent’s remaining life expectancy (Beers et al., Reference Beers, Ouslander, Rollingher, Reuben, Brooks and Beck1991; Spinewine et al., Reference Shinde, Shinde, Khatri and Hande2007). It is plausible that as more medications become available, and longevity continues to increase, there will be a further increase in the consumption of prescription medications among older adults, and the incidence of potentially inappropriate prescribing will continue to grow proportionately (Gallagher, Barry, & O’Mahony, Reference Gallagher, Barry and O’Mahony2007). In view of the high rate of prescription of PIMs in the older population, it is important to discuss the practical applications of the existing PIMs criteria, and identify methods to reduce the occurrence of PIMs. We have attempted to describe the popular explicit and implicit criteria globally for assessing appropriate prescribing in older adults, and have suggested methods to reduce inappropriate prescribing, which include meticulously reviewing the prescriptions during assessments. We have emphasized de-prescribing as an efficient way forward (Al Odhayani et al., Reference Al Odhayani, Tourkmani, Alshehri, Alqahtani and Mishriky2017).
Sources and Selection Criteria
The current study is a narrative review of the existing criteria for appropriate prescribing in older adults. The Ovid MEDLINE®, Embase, PubMed, Scopus, and International Pharmaceutical Abstracts databases were searched using the keywords prescribing criteria, prescribing indicators, deprescribing, appropriate prescribing, and older adults (including synonyms), by the MeSH or major descriptor headings. The search was limited to studies undertaken in humans, which were published in English during the past 30 years (1987–2017), and in individuals over 65 years of age. In addition, a citation analysis with the aid of Web of Science was conducted to track prospective citing of references of the selected articles. Reference lists of retrieved articles were studied for the purpose of finding additional articles not identified in the original database searches. Studies that were pertinent to the description of appropriate prescribing in older adults were selected. The most recent studies conducted globally were prioritized. For this review, the discussion is limited to the most common criteria employed internationally to assess appropriate prescribing in clinical practice and research.
Summary of the Criteria Measuring PIMs
Several criteria have been implemented internationally to reduce the prevalence of the prescription of PIMs in older adults (Tables 1–4).
Table 1: Globally most commonly used explicit criteria
Note. PIM = potentially inappropriate medications.
Table 2: Popular country-specific criteria for assessing potentially inappropriate medications (PIMs)
Table 3: Miscellaneous country specific criteria
Table 4: Implicit criteria adopted globally for assessment of PIMs
Note. STOPP/START = Screening Tool of Older Persons’ potentially inappropriate Prescriptions/Screening Tool to Alert doctors to the Right Treatment; STRIP = Systematic Tool to Reduce Inappropriate Prescribing
Explicit Criteria
These criteria are established by expert consensus, and used to generate lists of medications to be avoided in older adults, or in the presence of specific co-morbidities. It is often easier to implement explicit criteria in routine clinical practice, because no extensive clinical judgement is required for their implementation, and the number of medications and clinical conditions specified is limited. Explicit criteria are often utilized in studies of health outcomes and prevalence associated with PIMs (Chang & Chan, Reference Chang and Chan2010).
Several of these criteria are completely explicit or have both implicit and explicit measures embedded in them, and are usually drug or disease oriented, rather than being patient oriented (Morin, Fastbom, Laroche, & Johnell, Reference Meulendijk, Spruit, Drenth-van Maanen, Numans, Brinkkemper and Jansen2015). Most explicit criteria are based on the sequential Beers criteria, which could be attributed to the fact that each updated version of the Beers criteria encompasses most of the contemporary medications and ailments (Dimitrow, Airaksinen, Kivelä, Lyles, & Leikola, Reference Dimitrow, Airaksinen, Kivelä, Lyles and Leikola2011). The explicit criteria include listings of medications to be avoided in older adults, which are perceived to have increased possibilities of negative health outcomes. Expert opinions, literature reviews, and consensus statements are typically considered in the development of the explicit criteria, because there is insufficient evidence from randomised controlled trials to guide prescribing to older adults. However, it is important to note that explicit criteria may not encompass all aspects that define the quality of prescribing for older adults (Spinewine et al., Reference Shinde, Shinde, Khatri and Hande2007), and they must be updated and validated regularly (Dimitrow et al., Reference Dimitrow, Airaksinen, Kivelä, Lyles and Leikola2011).
These criteria are re-classifications of previous international PIMs criteria, and were developed by a panel of experts through the Delphi method. Very few studies have been conducted internationally to validate the reliability of the aforementioned criteria, and the majority do not circumvent under-prescribing of medications (Chang & Chan, Reference Chang and Chan2010; Lucchetti & Lucchetti, Reference Lavan, Gallagher, Parsons and O’Mahony2017; O’Connor, Gallagher, & O’Mahony, Reference Ní Chróinín, Ní Chróinín and Beveridge2012).
Implicit Criteria
When utilising implicit criteria for reducing inappropriate prescribing, the focus is generally on the patient, rather than on medications or diseases per se. Implicit criteria may be the most sensitive approach, as they account for the patients’ preferences; nonetheless, they are time consuming, and the outcomes are dependent on the prescriber’s knowledge and attitudes and can have low reliability (Spinewine et al., Reference Shinde, Shinde, Khatri and Hande2007). Implicit criteria may be employed as a supplement, but not as a substitute for clinical judgment, when optimising medication use in older adults (Pattanaworasate, Emmerton, Pulver, & Winckel, Reference Pardo-Cabello, Manzano-Gamero, Zamora-Pasadas, Gutiérrez-Cabello, Esteva-Fernández and Luna-Del Castillo2010).
The ideal criteria should consider the management of co-morbid disorders, under-prescribing of guideline-recommended medications, drug interactions, polypharmacy, patient preferences, life expectancy, and clinical information about the older adults (Basger, Chen, & Moles, Reference Basger, Chen and Moles2008).
The Way Forward
The health and functional status of the geriatric population vary widely; so a “one size fits all” approach to prescribing is sub-optimal for meeting individual patient needs (Bpacnz , 2010; Hanlon & Schmader, Reference Hanlon and Schmader2013). Individual assessments that review the need for continuing each medication help in simplifying treatment regimens, and may decrease the prescribing of PIMs. A Cochrane review in 2013 demonstrated that medication reviews of inpatients, led by physicians, pharmacists, and other health care professionals, resulted in a 36 per cent reduction in emergency department visits (Christensen & Lundh, Reference Christensen and Lundh2016). As the illness progresses, and if it is evident that the therapy is not appropriate, a tailored approach for discontinuing medications may be favoured (Holmes, Hayley, Alexander, & Sachs, Reference Hill-Taylor, Walsh, Stewart, Hayden, Byrne and Sketris2006). The physician has a limited role in effective prescribing in clinical practice as the prevailing professional and organisational culture towards quality influences the outcome to a larger extent. Interactive and continuous education, which includes discussion of evidence, local consensus, feedback on performance (by peers), and personal and group learning techniques facilitate appropriate prescribing (Grol & Grimshaw, Reference Grol and Grimshaw2003).
De-prescribing
De-prescribing is an initiative to decrease the use of redundant medications, especially PIMs, and it encourages the use of non-pharmacological alternatives, supervised by a health care professional, with the objective of managing polypharmacy and improving health outcomes (Reeve, Gnjidic, Long, & Hilmer, Reference Pugh, Hanlon, Zeber, Bierman, Cornell and Berlowitz2015). Rational withdrawal of medications in older adults may be one of the best clinical decisions for significant clinical benefits, including improved adherence. It can also reduce the inevitable negative consequences of polypharmacy, including medication burden and costs of complex medication regimens (Bpacnz , 2010; Ní Chróinín, Ní Chróinín, & Beveridge, Reference Nauta, Groenhof, Schuling, Hugtenburg, van Hout and Haaijer-Ruskamp2015). Appropriate cessation of medications in older adults encompasses factors such as the patient’s residual life expectancy, avoiding preventive treatments for those with a reduced survival prognosis, excluding medications with questionable evidence of effectiveness, and promoting the prescription of medications with favourable risk–benefit ratios. (O’Mahony & Gallagher, Reference O’Connor, Gallagher and O’Mahony2008; Scott, Gray, Martin, Pillans, & Mitchell, Reference Schuling, Gebben, Veehof and Haaijer-Ruskamp2013). In 2003, Woodward proposed the following five principles of de-prescribing: review all current medications, identify medications to be targeted for cessation, prepare a de-prescribing regimen, discuss with patients and carers, and frequent review and support (Woodward, Reference Winit-Watjana, Sakulrat and Kespichayawattana2003)
Scientific Evidence of Benefits of De-prescribing
In a trial composed of 119 older adults, 332 medications (2.8 medications per patient on an average) were discontinued utilising an algorithm, leading to a decline in mortality by 24 per cent, a significant reduction in the referral rates to acute care facilities, and reduction in health care costs (Garfinkel, Zur-Gil, & Ben-Israel, Reference Garfinkel, Zur-Gil and Ben-Israel2007; Scott et al., Reference Schuling, Gebben, Veehof and Haaijer-Ruskamp2013). In a similar study conducted in Israel, 58 per cent of medications were withdrawn with an 81 per cent success rate, without major untoward effects, and with almost 90 per cent of the patients reporting a holistic improvement in health (Garfinkel & Mangin, Reference Garfinkel and Mangin2010; Scott et al., Reference Schuling, Gebben, Veehof and Haaijer-Ruskamp2013). A systematic review conducted in Australia in 2008 observed that withdrawal of benzodiazepines and psychotropics diminished the number of falls, and improved cognition and psychomotor functioning in older adults (Iyer, Naganathan, McLachlan, & Le Couteur, Reference Hyttinen, Taipale, Tanskanen, Tiihonen, Tolppanen and Hartikainen2008). Likewise, a randomised controlled study in the United Kingdom in 2009 demonstrated a decline in mortality when antipsychotics were withdrawn in nursing home patients presenting with dementia (Ballard et al., Reference Ballard, Hanney, Theodoulou, Douglas, McShane and Kossakowski2009). In a recent clinical trial, it was observed that patients with a lower remaining life expectancy could safely discontinue statins, and the discontinuation was associated with a better quality of life and a decrease in medication costs (Kutner et al., Reference Koria, Zaidi, Peterson, Nishtala, Hannah and Castelino2015). The List of Evidence-baSed depreScribing for CHRONic patients (LESS-CHRON) criteria constitute the first explicit criteria to assist clinicians in deprescribing PIMs. Each of the 27 criteria consists of indications for which the medications are prescribed, clinical situations that offer an opportunity to de-prescribe, clinical variables to be monitored, and the minimum time to follow the patient after de-prescribing (Rodríguez-Pérez et al., Reference Renom-Guiteras, Meyer and Thürmann2017). The Current medication, Elevated risk, Assess, Sort, Eliminate (CEASE) de-prescribing framework, a de-prescribing five step protocol has been developed by the Australian Deprescribing Network, which includes taking a comprehensive medication history, identifying PIMs, determining whether the PIMs can be terminated, planning the withdrawal regimen (tapering where necessary), and the provision of monitoring, support, and documentation. It focuses on engaging patients throughout the sequence, with the aim of improving long-term health outcomes (Reeve, Shakib, Hendrix, Roberts, & Wiese, 2014a)
Barriers to De-prescribing
For most prescribers, prescribing medications is much easier than de-prescribing, possibly because of insufficient awareness of de-prescribing (Lai & Fok, Reference Kutner, Blatchford, Taylor, Ritchie, Bull and Fairclough2017). A study surveying family physicians in Vancouver observed that they were reluctant to de-prescribe the medications prescribed by another practitioner or specialist, and many physicians felt that they lacked the knowledge and skills to de-prescribe in a safe and effective manner out of fear of initiating an adverse effect (Harriman, Howard, & McCracken, Reference Harriman, Howard and McCracken2014). In another study, it was noticed that the physicians were not in favour of discontinuing medications because they usually followed the prescribing guidelines, and de-prescribing often requires discussing the patient’s limited life expectancy, which is challenging (Schuling, Gebben, Veehof, & Haaijer-Ruskamp, Reference San-José, Agustí, Vidal, Barbé, Torres and Ramírez-Duque2012). A systematic review in 2013 explored the views of patients and observed that the fear of non-specific consequences makes patients reluctant to agree to cessation (Reeve et al., Reference Reeve, Shakib, Hendrix, Roberts and Wiese2013). Sudden withdrawal of a medication could result in a physiological response, termed as “withdrawal reaction”, which could be prevented (or minimised) by tapering the dose before withdrawing a medication. Ceasing a particular medication may result in alteration of the pharmacokinetics and pharmacodynamics of other medications. The potential for negatively and irreversibly affecting the medical condition is of a greater concern (Reeve et al, 2014b). Other barriers to de-prescribing include lack of time and support, the anxiety of withdrawal reactions, and unfortunate experiences with cessation of medications in the past. To overcome these barriers to de-prescribing, it is imperative to educate the prescribers and patients about the problems of inappropriate prescribing and to develop guidelines for de-prescribing (Lai & Fok, Reference Kutner, Blatchford, Taylor, Ritchie, Bull and Fairclough2017).
During medication reviews, it is essential to minimise or discontinue the utilization of inappropriate medications, commence or optimise the utilization of appropriate medications, account for a cautious dosage of medications, consider the impact of renal function on drug clearance, and review any drug interactions (Masnoon, Shakib, Kalisch-Ellett, & Caughey, Reference Mann, Böhmdorfer, Frühwald, Roller-Wirnsberger, Dovjak and Dückelmann-Hofer2018).
Ongoing Research
The Canadian Frailty Network, through the Networks of Centres of Excellence Program, is dedicated to improving the health care of older Canadians living with frailty, and as a part of its mandate, convened a stakeholders’ meeting to seek their perspectives on appropriate medication prescription. The priorities identified were: (1) augmented efforts towards developing innovations focused on facilitating prescribing of appropriate medications, and/or de-prescribing in older adults living with frailty; (2) facilitating research for developing or improving models that facilitate pharmacists to be actively involved in the process of monitoring and assessing use of PIMs; and (3) encouraging further research into the values and preferences held by older adults living with frailty with respect to medication use (Muscedere et al., Reference Morin, Fastbom, Laroche and Johnell2017).
De-prescribing is an area of continuing research, as clinicians recognise the significance of a parallel strategy to re-evaluate the prescription of medications. There is ongoing research to supplement the beneficial evidence for deprescribing by focusing on relevant patient outcomes such as a reduction in falls, hospital admissions, and mortality; and improvement in sleep quality, cognitive function, independence in activities of daily living, and quality of life (Lai & Fok, Reference Kutner, Blatchford, Taylor, Ritchie, Bull and Fairclough2017). De-prescribing may be more beneficial than continuing intensive treatment regimens in older adults presenting with severe co-morbidities (e.g., patients presenting with end-stages of dementia or with a poor functional status); this has motivated researchers to address de-intensification of medical therapy, which implies discontinuation of medications in situations in which the potential problems outweigh the benefits (Green & Leff, Reference Green and Leff2016). The Canadian Deprescribing Network continues to develop and advance de-prescribing across Canada, in a collaboration with a wide range of stakeholders to bring about real transformation in Canadian health care (Tannenbaum et al., Reference Steinman, Beizer, DuBeau, Laird, Lundebjerg and Mulhausen2017). De-prescribing has the potential to improve health outcomes; however, the clinical benefits and associated risks can be determined only after the development and validation of a systematic de-prescribing process. An account of the reduction in mortality and morbidity will necessitate large randomized controlled trials, requiring hundreds or even thousands of participants in each arm, so the conduct of these trials may, unfortunately, not be feasible (Reeve et al., 2014b). To achieve appropriate polypharmacy, de-prescribing cannot be considered in isolation for optimising the medications of older adults, as potentially inappropriate omissions have also been found to be prevalent in this vulnerable population (Cadogan, Ryan, & Hughes, Reference Cadogan, Ryan and Hughes2016). Consideration should be given to integrating the de-prescribing process with other interventions to reduce PIM prevalence.
Strength of the Review
This review provides a glimpse of the benefits of and gaps in the existing criteria for inappropriate prescribing, with respect to the latest scientific evidence. The review also describes the potential methods useful in prescribing appropriate medications, and emphasizes de-prescribing.
Limitation of the Review
The literature search was restricted to manuscripts published in English. In addition, the search terms may not be adequate, although the most-relevant criteria were included, and a manual search of the reference lists from the articles searched was also performed. Most of the explicit criteria are based on the Beers criteria, which may have produced a bias and possibly false conclusions.
Conclusion
Inappropriate prescribing of medications in older adults remains a major international health concern. Excluding the Beers criteria and the Screening Tool of Older Persons’ potentially inappropriate Prescriptions/Screening Tool to Alert doctors to the Right Treatment (STOPP/START) criteria, most of the existing criteria for measuring PIMs are not comprehensive, and are generally not being used globally as a criterion for prescribing in all older adults. A way forward to reduce PIMs is to encourage de-prescribing, which is a positive, patient-centred intervention, and requires shared decision making, informed patient consent, close monitoring of effects, and consideration of the cumulative risk from multiple medications caused by pharmacokinetic and pharmacodynamic interactions—the same prescribing principles that apply when the therapy is initiated. The development of evidence-based de-prescribing guidelines and the inclusion of de-prescribing modules in all chronic disease guidelines are a priority for the adequate care of older adults.
