Our study confirmed routine use of anticholinergic drugs in more than half of older people referred over a six months period to a memory clinic. Our reported higher rate of prescribing anticholinergic medication is in sharp contrast to the observed decreasing trend during the last decade [14], the implementation of the UK national regulative, i.e. the NICE guidance on multimorbidity and polypharmacy in 2017 [15], the NICE guidance for dementia (2018) [13], as well as the 2016 Health Survey England [16], that documented widespread polypharmacy among the older UK people. Bearing in mind that our audit was conducted shortly after the above recommendations were put in place, it is not surprising that they had very limited influence upon the prescribing culture among primary and secondary/tertiary care physicians in our region, and thus have not contributed to the recommended reduction of inappropriate prescribing of anticholinergic/antimuscarinic drugs in community dwelling older people [17].  Future studies may shed further light whether the above recommendations have been both accepted by medical professionals and contribute to raising awareness of the anticholinergic side effects among service user, and determine their impact on the quality of healthcare and outcomes for older people.

Another reason for the high percentage of ACB drug use in the current study may be due to the choice of the ACB assessment tool. Thus, an earlier (paper) version of the ACB tool (2008) did not include a number of anticholinergic drugs (i.e. diazepam, and the full list of SSRIs and SNRIs antidepressants, with only paroxetine and mirtazapine included), whereas fluoxetine and diazepam were added in the 2012 ACB version. The latest web ACB tool, however, contains all 4 SSRIs and 2 SNRIs (mirtazapine and venlafaxine) antidepressants, but not duloxetine.  When assessing the ACB score on this analysed cohort, using the 2012 paper version of the tool, we detected only 55 people were prescribed ACB drugs bringing down the rate of ACB prescribing to 17.41%. Thus, only two people were identified to have been using diazepam regularly, and additional two SSRI and SNRI (paroxetine and venlafaxine) antidepressants, in contrast to the current revised ACB study in which 52 additional SSRI (n=41) and SNRI (n=11) antidepressant users were identified. When using the 2008 ACB version, the frequency of the prescribed anticholinergic drugs was, not surprisingly, significantly lower and similar to that reported to the prescribing rate of potential or definite anticholinergic properties (ACB>0) to  that described for Italian memory clinic users (15.9%; [18]), whereas the latest revision of 57.2% is similar to the ACB prescribing reported in community-dwelling Australian people who attended 9 memory clinics  (59.69%; [19]) and had a diagnosis of mild cognitive impairment or dementia [11].

Not surprisingly, the most prescribed anticholinergic drugs were those who are used for treatment of pain (i.e. amitriptyline and codein /co-codamol/morphine/tramadol), antidepressants (SSRI and SNRI), heart failure and fluid retention (digoxin/warfarin/furosemide).  Furthermore, most of these drugs have been used over a prolong period of time (i.e. amitriptyline, atenolol, diazepam), and have either not been substituted over the years, or substituted with an alternative with similar anticholinergic properties.  The higher rate of ACB prescribing among the above conditions has been further confirmed in our secondary analysis, where hypertension, diabetes, cardiovascular, renal and respiratory conditions were linked to significant increase in the ACB prescribing, i.e. up to 1.4-2.3 fold higher, whereas the group of mixed medical conditions i.e. gastrointestinal problems, skin conditions (i.e. psoriasis, pruritus, vitiligo etc.), central nervous system conditions (i.e. epilepsy, polyneuropathy, multiple sclerosis, Bell’s paralysis etc.), was the medical group were ACB drugs were not commonly prescribed (EBM-L et al, unpublished). However, in contrast to a previous study, in which most common anticholinergic classes used were tricyclic antidepressants, first-generation antihistamines, and bladder antimuscarinics [20], this was not the case in our study. Although the number of people who were on antimuscarinic medication for overactive bladder was relatively low, we noticed a differential mode of prescribing, similar to that described in a larger population-based study, with slight preference of solfenicine over oxybutynin [21].

Similarly, to a previous study [14] where the prescription of antidepressants was double the other ACB drugs, we found high rate of prescription of antidepressants, with amitriptyline relatively more frequently prescribed and used largely for neuropathic pain control and insomnia (as based on the prescribed dosage of 10-20mg daily) rather as an antidepressant. However, the use of newer antidepressants (e.g. SSRIs and SNRIs) was more common (nearly in 17% of our sample), with no clinical evidence for treatment of enduring mental health problems, i.e. depression and/or anxiety. The relatively high rate of antidepressant prescribing is in accord with the latest reports of estimated 5-16% of adults receiving antidepressants in Europe and the USA annually [22].   The reason for the increased prescribing of antidepressants in community dwelling older people may be attributable to primary care physicians seeing people in distress and crises, and also being confident in using the antidepressants beneficial properties (i.e. sedative effects helping insomnia and anxiety symptoms and gain weight) [23]. In addition, the sedative properties of the low dose antidepressants (i.e mirtazapine) seems to have been adopted as a safer non-addictive alternative to avoid benzodiazepines [23], as documented via the sparse use of benzodiazepines in our older people referred to out memory clinic. Namely only 6 people were prescribed diazepam (1.88%).

Shortly after completing the audit, in November 2019 the Medicines & Healthcare products Regulatory Agency (MHRA, responsible for regulating all medicines and medical devices in the UK by ensuring they work and are acceptably safe)  issued an alert to healthcare professionals, recalling all unexpired stock of certain batches of prescription-only Ranitidine medicines used to treat conditions such as heartburn and stomach ulcers. This was a precautionary measure due to possible contamination with NDMA (N-nitrosodimethylamine), which has been identified as a risk factor in the development of certain cancers. Nevertheless, the ranitidine discontinuation should not affect overtly the extent of the ACB prescribing in older people, as based on our study, since only 12 of our patients had been issued with this medication, representing 3.75% of all the ACB prescription.

Chronic pain is rather common among older people, with the opioid-based pain management being much more prevalent in this age group (reviewed in [24]). It is, thus, not surprising that opioids featured among the most frequently prescribed anticholinergic drugs (18.44%). However, our study was limited to patients’ repeated medication charts from primary care and did not include opioids available over the counter, and, thus, may be an underestimate of the use of these drugs.  Additional 11.25% patients used low doses amitriptyline for musculoskeletal and chronic non-cancer pain, despite the recommendations to avoid tertiary tricyclics such as amitriptyline due to concerns over adverse side effects. Although the recommendations are to consider nortriptyline instead of amitriptyline [25], in our study, only one patient was prescribed nortriptyline.

Polypharmacy is the major risk factor for adverse side effects in older people (i.e. nausea light-headedness, muscle weakness and pain, bleeds, confusion, falls, sedation, altered clothing capacity etc.), which can lead to serious, but preventable injuries and illnesses [6]. Furthermore, the cumulative exposure (combined score) to anticholinergic and sedative drugs is significantly associated with an increased risk for frailty [95%CI OR 3.54 (1.47-8.57)] [10]. One of the interventions aimed at reducing this risk is the detection and reduction of inappropriate and harmful prescriptions. The responsibility of the drug reviews, thus, lies upon all clinicians involved in the medical care of older people. In memory clinical setting, we propose regular addressing of risks/benefits of the anticholinergic medication by both independent clinician reviews, as well as at multidisciplinary team meetings and actioned accordingly. Risks of anticholinergic cognitive burden should be highlighted in the letter to GP. If deemed appropriate, GP should be asked to review the medication in the care plan, actions that have been significantly missing in the referrals to secondary/tertiary care.

Though comorbidities proportionally increase with ageing, regular review of repeat drug prescriptions, especially those with known anticholinergic burden needs to be incorporated regularly in routine clinical setting. In addition, we propose immediate action on long standing use of ACB drugs especially opioids and tricyclic antidepressant and when only one of these ACB drugs is used. Regular pharmacological reviews will decrease the anticholinergic burden, improve the lives of older people, and also decrease their potential impact on their cognitive functioning. However, although opportunities for quality improvement of anticholinergic/muscarinic prescribing in older adults remain and they need to be explored and addressed individually for each patient, these may well be hampered by cost effectiveness and social value judgements.