Tracking the anticholinergic burden score during hospital admission in an acute hospital in England

Here we report on a retrospective study undertaken in a UK hospital that found that the overall anticholinergic drug burden did not change significantly during an inpatient stay on five wards typically caring for older people

Anticholinergic medications (ACM) are prescribed for the management of conditions such as depression, psychosis, Parkinson’s disease, overactive bladder, and chronic obstructive pulmonary disease. These are among the most prescribed medications in patients with polypharmacy. One systematic review looking at adverse effects found that certain individual ACM or increased overall exposure to ACM may increase the risks of cognitive impairment, falls and all-cause mortality in older adults,¹ though others argue that the evidence for harmful outcomes in certain groups of older patients remains uncertain or deficient.^2,3

What is the anticholinergic burden?

Anticholinergic burden (ACB) is defined as the “accumulation of higher levels of exposure due to one or more ACM and the attendant increased risk of medication-related adverse effects”.⁴ A number of anticholinergic quantification scales have been reported in the literature, providing a list of ACM and a rank of low- to high-risk based on anticholinergic activity. However, most scales are constructed using expert opinion panels, in vitro data and literature reviews. These scales each include different drugs and have variations in the rating of the included drugs, meaning variability and inconsistency among the scales exist.^5,6 As an example, it is suggested that the relationship between ACB and fracture risk might differ depending on the ACB scale used.⁷

Inappropriate prescription of ACM to older patients in primary care has been reported extensively.^8.9 In England, the NHS Business Services Authority anticholinergic burden prescribing comparator can be used to identify the number of patients at risk of anticholinergic side effects at Clinical Commissioning Group, Primary Care Network and GP Practice level, and to prioritise work in this area.¹⁰ However, ACM use in a hospital setting has been less extensively studied and with varying results. ACM prescription was found in 10% of hospitalised, older patients using a database from a French general hospital covering 14,090 hospital stays by patients aged 75 and over.¹¹ A Danish study, utilising the Anticholinergic Risk Scale, examined the association between ACM at hospital admission and mortality in older patients and found that such use is associated with short- and long-term mortality in geriatric patients, even when adjusting for other important variables such as comorbidity and activities of daily living.¹² In this Danish study, nearly two-thirds of a total of 74,589 patients received ACM. Few patients received medications with an ACB score of two or three while a score of one accounted for 88.1% of the overall anticholinergic intake. 

Previously in our Trust, a RADAR (RCHT Analysis, Data and Reporting) report had been developed that pulls daily prescribing data for ACM from the hospital’s e-prescribing system (CareFlow Medicines Management) for patients on five selected wards. These wards covered elder care, trauma (mainly older patients), stroke, and neurology. For each patient, the report generates the total ACB score, together with the names of any drugs that fall into the categories of an ACB score of 1, 2, or 3. This report was developed using the Ageing Brain Care scoring system.¹³ Patient characteristics (age and gender) are also displayed. The intention was that this report would be used by clinical staff to identify those patients with a high ACB who might be suitable for a medication review. However, a previous unpublished internal study has found that this electronically available report was not used by clinicians, mainly because of a lack of compatibility with workflow.

Our overall aim was to identify if the ACB score altered between admission and the last day on the ward for patients on those specific wards that are the subject of this RADAR report. The objectives were to report on changes to ACB score over the hospital stay and to ascertain if, in general, any particular ACM was stopped to reduce the ACB score.

Method

This was a retrospective study utilising the electronically available RADAR ACB score report that was run for November 2021 for those patients admitted to the five target wards. Data were extracted such that the report displayed any ACM that contributed to the patient’s ACB score both at admission and the final day on the ward. This extraction did not include medication listed on any discharge prescription. Patients on the same medication, for example, morphine prescribed as a standard-release oral formulation and an injectable formulation ‘as required’ accrued a score of only one. Likewise, the presence of both cyclizine lactate and cyclizine hydrochloride on the same patient’s electronic prescribing chart accrued a score of only one. Data were entered into Excel for analysis. Data for two wards that were more typical of care of the elderly patients were also analysed separately.

Ethics

Health Research Authority criteria for research and service evaluation were considered. This was a retrospective assessment involving no changes to the service delivered to patients, and we used the NHS Health research authority tool (www.hra-decisiontools.org.uk/research/index.html) which helped confirm that no ethical approval was required for this project.

Results

Over an approximate three-week period in November 2021, there were 262 episodes of patients admitted to the five wards. Sixty-two of these episodes were on the two wards grouped together for further analysis as they were considered to be more representative of care of the elderly patients. Excluding those whose admission was apparently less than one day, this left a total of 212 patient episodes (mean age 70 years, range 21–99, 100 male), of which 59 (mean age 74 years, range 23–97 years, 27 male) were on the two subset wards. Overall, the duration of stay ranged from one day (29%) to 5 days and longer (27%). 

Overall, there was an increase in the total ACB score between admission and final day on the ward for the 212 patient spells from 322 (mean 1.52 per spell) to 456 (mean 2.15 per spell), and also across the subgroup of 59 spells from 105 (mean 1.78 per spell) to 120 (mean 2.03 per spell). This overall increase in ACB score was also seen when considering only those patient spells that were for 5 days or longer. The number of patients with an ACB score of 2 or ≥3 was greater on the final day on the ward than at admission when considering all 212 spells (Table 1). For the subgroup of 59 patient spells, it was only the number of patients with an ACB score of ≥3 that increased during the acute stay.

However, 9% (20/212) and 15% (9/59) of patient spells did show a decrease in their ACB score (Table 2). Across all patient spells, the most frequent medicines that were ceased were fentanyl, morphine, furosemide and co-dydramol.

Discussion

We examined the ACB of medication for patients admitted to a select group of wards caring for, in the main, older people. This analysis has shown an overall increase in ACB score during the acute inpatient stay for 212 patient spells from a mean of 1.52 per spell to 2.15 per spell. There were some instances where a reduction did occur – 9% of 212 patient spells and 15% of the subgroup of 59 spells. 

Several studies with varying results have tracked changes to ACB during the hospital stay of older people. A study, utilising the Anticholinergic Risk Scale (ARS),¹⁴ described the burden of prescribed ACM in all older adults admitted as an emergency to any specialty in a large hospital in the UK. These authors looked at how ARS scores changed from admission to discharge and evaluated associations between both admission ARS and change in ARS score and hospital outcomes, primarily inpatient and post-discharge mortality.¹⁵ They found that from 33,360 patients included, just under one-third were prescribed an anticholinergic on admission, with 3266 (9.8%), 2479 (7.4%) and 4438 (13.3%) patients scoring 1, 2 or >3 respectively on the ARS. In our much smaller study, we found 46/212 (22%), 34/212 (16%) and 51/212 (24%) patients scoring 1, 2 or >3 respectively on the scale we used. These UK authors¹⁵ did find a statistically significant reduction in mean ARS from admission to discharge in all specialties. Interestingly, the largest absolute and relative reductions in mean ARS scores were seen in patients discharged by Geriatric Medicine and Trauma and Orthopaedics, although they report that patients experiencing either an increase or a decrease in ARS score from admission to discharge were more likely to have a prolonged (>10 days) hospital stay. Our five wards would be similarly classified as Geriatrics and Trauma and Orthopaedics, although we found an overall increase in mean ACB score. However, we only had ten patients with a prolonged stay and two of these had a decrease in ACB score

A similar study in New Zealand measured the ACB using the total Anticholinergic Drug Scale (ADS) score for 224 patients on presentation to and at discharge from a geriatric unit.¹⁷ Despite medication changes occurring during the hospital stay, there was no significant change in ADS score between admission and discharge. Compared with admission, 35% patients had a reduced ACB; 28% patients had an increased ACB, whereas 37% had no change on discharge. 

A study based in the UK and Europe,¹⁸ described changes in the ACB in 549 patients admitted to hospital with a diagnosis of delirium, chronic cognitive impairment, or falls. They utilised an adapted 2012 revision of the original ACB scale.¹³ Key findings were that 21.1% of patients had their ACB score reduced, 19.7% had their ACB increased, 22.8% of ACM-naïve patients were discharged on ACM, and there was no change in the ACB scores in 59.2% of patients. The European study also observed that the same medications, while stopped in some patients, were started in others, and that more than one in five patients who were not taking anticholinergics when admitted were prescribed them by discharge.¹⁸ Compared with this European-based study¹⁸ and the New Zealand study,¹⁷ we found 9% had a reduced ACB, 37% had an increased ACB, and 53% had no change by the final day on the ward. 

A specialist multidisciplinary team based in a UK Emergency Department was able to perform targeted medication reviews and significantly reduced anticholinergic drug exposure in frail older patients as measured by the ACB scale.¹⁹ Interestingly, only 2.3% (n=3/129) of ACB-naive patients were started on an anticholinergic drug (that is, ACB score 0 pre- to 1 post-review) and there were no other examples of patients experiencing an increase in ACB score during admission.

The importance of this topic of ACB is highlighted in the national Getting It Right First Time report, which recommends that older patients should have an initial review of medicines management when they are admitted to hospital. This report notes that the admission might be triggered by adverse drug reactions and the risks and benefits of drugs need to be reviewed. This can be done using a structured approach such as the STOPP-START tool, or the anticholinergic burden score to assess the risk of drugs that contribute to falls and delirium.²⁰

It is recognised that the provision of guidelines and education alone do not seem to be sufficient to ensure the best medicines review and optimisation in older people. Whereas evidence shows an improvement in the quality of prescribing and deprescribing via the use of multidisciplinary teams, geriatric case conferences, medication review by pharmacists and the use of information technology to support medication decisions.²¹ In the context of reviewing and possibly reducing ACB score, we have in place in our Trust an electronic tool that identifies possible opportunities for review. However, we know this RADAR report is not utilised.

A strength of our study was the use of an e-prescribing system, which facilitated the accurate extraction of prescribed medication. We recognise the limitations of this retrospective study of patients admitted to a single acute trust during a relatively short follow-up period. During the pandemic, these five wards may have held outlier patients not under the care of the elderly team and so any review of ACM may not have been a priority. It is important to note that what was prescribed on the e-prescribing system at admission may be different to medicines identified at the reconciliation (clerking in) process, that is, some ACM might have been ceased/withheld at admission to the ward and we did not record this. Also, we looked only at prescribed medication, and we recognise that, especially for ‘as required’ medication, these might not have actually been administered to patients. In particular, those patients on Trauma would have had analgesic requirements (weak or strong opioids) accruing an ACB score typically of one per different opioid prescribed and this continued throughout the hospital stay with little opportunity to reduce the score, although these opioids might not then have continued into discharge medication. Finally, we did not record patient comorbidities.

Conclusion

In this study, the overall ACB did not change significantly during an inpatient stay on five wards typically caring for older people. It might be appropriate to raise prescribers’ and pharmacy team awareness of these practices such that there is more of a focus on ACB and the potential for corresponding iatrogenic effects.

Key points

• Anticholinergic medication is associated with adverse clinical outcomes, including delirium and cognitive decline. 
• Various anticholinergic burden (ACB) or risk scales have been devised to aid medication reviews. 
• Anticholinergic medication use in a hospital setting has been less extensively reported than in primary care, although some studies have tracked changes in ACB during hospital admission.
• In this retrospective study, there was an increase in the total ACB score between admission and the final day on the ward for the 212 patient spells from 322 (mean 1.52 per spell) to 456 (mean 2.15 per spell).
• It might be appropriate to raise prescribers’ awareness of these practices such that there is more of a focus on ACB, and the potential for corresponding iatrogenic effects. 

References

1. Ruxton K, Woodman RJ, Mangoni AA. Drugs with anticholinergic effects and cognitive impairment, falls and all-cause mortality in older adults: A systematic review and meta-analysis. Br J Clin Pharmacol 2015;80:209–20. 
2. Wang K et al. Anticholinergics and clinical outcomes amongst people with pre-existing dementia: A systematic review. Maturitas 2021;151:1–14. 
3. Mehdizadeh D et al. Associations between anticholinergic medication exposure and adverse health outcomes in older people with frailty: A systematic review and meta‑analysis. Drugs Real World Outcomes 2021;8:431–58. 
4. Boustani M et al. Impact of anticholinergics on the aging brain: a review and practical application. Aging Health 2008;4(3):311–20. 
5. Naples JG et al. Concordance between anticholinergic burden scales. J Am Geriatr Soc 2015;63:2120–4. 
6. Al Rihani SB et al Quantifying anticholinergic burden and sedative load in older adults with polypharmacy: A systematic review of risk scales and models. Drugs Aging 2021;38:977–94. 
7. Ogawa Y, Hirai T, Mihara K. A meta-analysis of observational studies on anticholinergic burden and fracture risk: evaluation of conventional burden scales. J Pharm Health Care Sci 2021;7:30. 
8. Myint PK et al. Total anticholinergic burden and risk of mortality and cardiovascular disease over 10 years in 21,636 middle-aged and older men and women of EPIC-Norfolk prospective population Study. Age Ageing 2015;44:219–25. 
9. Hanlon P et al. Assessing risks of polypharmacy involving medications with anticholinergic properties. Ann Fam Med 2020;18:148–55. 
10. NHS Business Services Authority. The AHSN Network. July 2017. Medicines Optimisation Polypharmacy Prescribing Comparators. www.nhsbsa.nhs.uk/sites/default/files/2018-02/PolyPharmacy%20Specification%20v1%200%20July%202017_0.pdf (accessed June 2022)
11. Ferret L et al. Inappropriate anticholinergic drugs prescriptions in older patients: analysing a hospital database. Int J Clin Pharm 2018;40:94–100. 
12. Sørensen SR et al. Use of drugs with anticholinergic properties at hospital admission associated with mortality in older patients: A Danish nationwide register‑based cohort study. Drugs Real World Outcomes 2022;9:129–140. 
13. Aging Brain Care. Anticholinergic Cognitive Burden Scale. www.idhca.org/wp-content/uploads/2018/02/DESAI_ACB_scale_-_Legal_size_paper.pdf (accessed June 2022). 
14. Rudolph JL et al. The anticholinergic risk scale and anticholinergic adverse effects in older persons. Arch Intern Med 2008;168:508–13. 
15. Herrero-Zazo M et al. Anticholinergic burden in older adult inpatients: patterns from admission to discharge and associations with hospital outcomes. Ther Adv Drug Saf 2021;12:20420986211012592. 
16. Carnahan RM et al. The Anticholinergic Drug Scale as a measure of drug-related anticholinergic burden: associations with serum anticholinergic activity. J Clin Pharmacol 2006;46:
    1481–6. 
17. Lee MSS, Hanger HC. Audit of anticholinergic medication changes in older hospitalised patients using the Anticholinergic Drug Scale. Intern Med J 2017;47:689–94. 
18. Weichert I et al. Anticholinergic medications in patients admitted with cognitive impairment or falls (AMiCI). The impact of hospital admission on anticholinergic cognitive medication burden. Results of a multicentre observational study. J Clin Pharm Ther 2018;43:682–4. 
19. Neilson V, Palmer S. The effectiveness of a multidisciplinary frailty team in reducing anticholinergic burden in frail older patients: A quantitative service evaluation. Geriatr Nurs 2021;42(4):
    943–7. 
20. Hopper A. Geriatric Medicine. GIRFT Programme National Specialty Report. Getting It Right First Time. 2021. 
21. Topinková E et al. Evidence-based strategies for the optimization of pharmacotherapy in older people. Drugs Aging. 2012;29:477–94.