Abstract

Objectives: to determine the effect of drugs with anti-cholinergic properties on relevant health outcomes.

Design: electronic published and unpublished literature/trial registries were systematically reviewed. Studies evaluating medications with anti-cholinergic activity on cognitive function, delirium, physical function or mortality were eligible.

Results: forty-six studies including 60,944 participants were included. Seventy-seven percent of included studies evaluating cognitive function (n = 33) reported a significant decline in cognitive ability with increasing anti-cholinergic load (P < 0.05). Four of five included studies reported no association with delirium and increasing anti-cholinergic drug load (P > 0.05). Five of the eight included studies reported a decline in physical function in users of anti-cholinergics (P < 0.05). Three of nine studies evaluating mortality reported that the use of drugs with anti-cholinergic properties was associated with a trend towards increased mortality, but this was not statistically significant. The methodological quality of the evidence-base ranged from poor to very good.

Conclusion: medicines with anti-cholinergic properties have a significant adverse effect on cognitive and physical function, but limited evidence exists for delirium or mortality outcomes.

Introduction

Drugs with anti-cholinergic properties are commonly prescribed for a variety of medical illnesses [1]. With a globally ageing population, much of this drug burden falls on the elderly. Ninety percent of older adults report taking at least one prescription medication [2]. It has been estimated that 20–50% of older people have been prescribed at least one medication with anti-cholinergic activity [3]. Younger adults may also be prescribed long-term anti-cholinergic treatment for conditions such as asthma or to manage the side-effects of medicines used to treat psychiatric disorders [3]. It has been recommended that increased care should be taken to avoid the inappropriate prescribing of anti-cholinergic drugs due to the wide spectrum of central effects such as the onset of dizziness, sedation, confusion, in addition to increasing delirium, causing a decline in cognitive and physical function [1]. Peripheral adverse effects are also commonly reported and include dry mouth, dry eyes, constipation, blurred vision and increased heart rate [1].

Much of the previous evidence has focused on a link between medications with anti-cholinergic properties and cognitive function [3, 4]. Medications with anti-cholinergic properties recognized by the anti-cholinergic cognitive burden (ACB) scale have been recently correlated with an additional 0.33 point decline in Mini-Mental State Examination (MMSE) score over 2 years [5], a 2-fold increase in cognitive impairment with as little as 60–90 days of use [6], and ∼50–80% increase in the risk of incident cognitive impairment over 6 years [7].

A decline in cognitive function and the diagnosis of mild cognitive impairment is associated with a progression to dementia within 5 years [8], making primary prevention and avoidance of anti-cholinergic medications wherever possible, of significant importance as a strategy to protect against persistent cognitive decline [9]. Similarly, it is well known that functional impairment in older adults limits independent living and impacts on their quality of life [10]. Mild cognitive impairment has also been attributed to an increased risk of falls, further increasing morbidity and reduced physical function in older people [11].

This systematic review assesses the empirical research surrounding the effect of increasing anti-cholinergic load on cognitive function, delirium, physical function and mortality. To the author's knowledge, this is the first systematic review to evaluate the association between medications with anti-cholinergic properties and delirium or physical function. This paper will also provide an important update required to review the current literature on a possible association with cognitive function and mortality.

Search strategy and selection criteria

Search methods

A PRISMA compliant systematic review was undertaken [12]. The primary search was conducted of the published literature using the electronic databases EMBASE (2002–2013) and Ovid MEDLINE (2002–2013) to Week 3 October 2013. The search terms adopted are presented in Table 1. This was adapted for the different search databases.

Table 1.

Search terms used for the electronic database searches

  1. (Anticholinergic* or Anticholinergic agent* or Cholinergic antagonist* or Anti-cholinergic* or Antimuscarinic* or Antimuscarinic agent* or Muscarinic antagonist* or Anti-muscarinic*).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui] (mortality or death or survival).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui]

  2. (Cognitive function or Cognitive disorder* or Cognitive impairment or Dementia or Delirium).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui]

  3. (Physical function or Physical activity or Function* or Activity*).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui]

  4. OR/2–4

  5. 1 and 5

  6. Limit 6 to English language

  7. Limit 6 to year = ‘2002 –Current’

 
  1. (Anticholinergic* or Anticholinergic agent* or Cholinergic antagonist* or Anti-cholinergic* or Antimuscarinic* or Antimuscarinic agent* or Muscarinic antagonist* or Anti-muscarinic*).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui] (mortality or death or survival).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui]

  2. (Cognitive function or Cognitive disorder* or Cognitive impairment or Dementia or Delirium).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui]

  3. (Physical function or Physical activity or Function* or Activity*).mp. [mp = ti, ab, sh, hw, tn, ot, dm, mf, dv, kw, nm, kf, ps, rs, ui]

  4. OR/2–4

  5. 1 and 5

  6. Limit 6 to English language

  7. Limit 6 to year = ‘2002 –Current’

 

A secondary search was conducted of the unpublished grey literature and trial registries. The following databases were accessed from January 2002 to Week 3 October 2013: open Grey, the WHO International Clinical Trials Registry Platform, Current Controlled Trials and the UK National Research Register Archive. An additional search of reference lists from all potentially eligible papers and review articles was also undertaken for completeness.

Eligibility criteria

Studies were deemed eligible if they satisfied each of the following criteria:

  • Studies investigating anti-cholinergic effects on adults. This was confirmed by cross-checking of the mentioned drugs against the 2012 updated ACB scale [13]; http://www.agingbraincare.org/tools/abc-anticholinergic-cognitive-burden-scale/). All studies included were required to indicate the dosage and duration at which medicines were used or how the anti-cholinergic load was calculated.

  • Studies investigating the effect of medicines with anti-cholinergic properties on one of the following outcomes: mortality, cognitive function, delirium or physical function.

  • Either randomised controlled trial (RCT), prospective cohort, cross-sectional or prospective case-controlled studies.

Studies were excluded if:

  • The primary exposure was not clearly stated or the drug used did not have anti-cholinergic properties.

  • The anti-cholinergic load was based on serum sample analysis alone.

  • Studies that reported the right exposure but did not report the effect of stated anti-cholinergic medicines against the selected outcomes.

  • Retrospective studies, case reports, journal editorials, literature reviews, clinical audits or studies that were not published in the English language.

  • Animal studies.

We selected to review only those studies published after 1 January 2002 to capture the results of studies evaluating systematic recognition of anti-cholinergic medications through various scales. Other systematic reviews have described the relationship between anti-cholinergics and cognitive outcomes [3, 4, 14] and included results prior to 2002. This review therefore updates these previous studies.

Identification of studies

Two reviewers (N.B., W.Y.C.) independently reviewed the study titles and/or abstracts to identify potentially eligible studies against the review eligibility criteria. Any disagreements were resolved through discussion and adjudicated by a third investigator (C.S.K.).

The full text for all potentially eligible studies were gathered and independently re-reviewed by two reviewers (N.B., W.Y.C.) against the eligibility criteria to determine final eligibility. Any disagreements were resolved through discussion and adjudicated by three senior reviewers (C.S.K., C.F. and I.M.).

Data extraction

Data extraction were independently conducted by five reviewers (N.B., W.Y.C., M.G., I.K., C.S.K.) and verified by two senior reviewers (I.M., C.F.). Data extraction were undertaken using a pre-defined data table. Data extracted included: study design, number of participants, year of the study undertaken, selection criteria, results of each study with regards to the effect of anti-cholinergic medications on the outcomes of interest, significance of the associations were based on the statistical results reported in each study.

Risk of bias assessment

Two critical appraisal tools were used to assess methodological quality and risk of bias. The Newcastle-Ottawa scale [15] was used to assess the quality of non-randomised studies. The Cochrane Risk of Bias tool [16] was used to assess methodological quality for all RCTs.

Risk of bias assessments were conducted by two independent reviewers (N.B., W.Y.C.). In the event of disagreement on critical appraisal score, agreement was met through discussion, adjudicated by a third reviewer (C.S.K.).

Data analysis

The data extraction table were reviewed to determine the most appropriate analysis technique to answer the research question. There appeared considerable study heterogeneity in relation to population diagnosis and characteristics, medication and dose, follow-up period, outcome measurement and reporting of data. This therefore precluded the adoption of a meta-analysis to pool data. Accordingly, a qualitative narrative review of the literature, answering the research questions, was the most appropriate analysis strategy for synthesising trends in findings.

Results

Search results

The results of the search strategy are summarised in Figure 1. From a total of 7078 identified citations, 133 were deemed potentially eligible. From these 46 studies met the eligibility criteria and were included in the final review.

Figure 1.

PRISMA flow diagram of search strategy results.

Figure 1.

PRISMA flow diagram of search strategy results.

Characteristics of included studies

The characteristics of included studies are summarised in Table 2. The studies consisted of 38 cohort studies, six RCTs and two case–control study.

Table 2.

Study design and characteristics of the included studies

Study Design; setting Country Number of participants Mean age (years) % Male Participant characteristics 
Agar [17Post hoc analysis of RCT; Hospital and community Australia 461 72 48 Patients included in Palliative Care Trial with diagnosis of cancer, known date of death, Australia—modified Karnofsky Performance Scale (AKPS) score of >60 at baseline and AKPS score falls <60 at any time during follow-up 
Ancelin [18Cohort study; Community France 372 66.2 NS Patients with age >60 years and without dementia at recruitment 
Boustani [19Cohort study; Community USA 1558 77.6 33.6 Patients with age ≥65 years and were African American 
Caeiro [20Case–control study; Hospital Portugal 74 62 55 Patients included with admission diagnosis of cerebral infarct or intracerebral haemorrhage/intraventricular haemorrhage, assessment delirium performed within 4 days after stroke onset, a Glasgow Coma Scale score ≥5 on the day of the delirium examination 
Cai [5Cohort study; Community USA 3690 72 30 Patients aged 65 and older who were living independently or with family in a community setting. Medication dispensing data defined the exposure, and a two-stage screening and diagnosis design provided the outcome assessment of cognitive impairment 
Campbell [7Cohort study; Community USA 1652 81.8 30.9 Patients with age ≥70 years who were African American, community dwelling, had normal cognitive function at baseline and enrolled in Indianapolis-Ibadan Dementia Project between 2001 and 2007 
Campbell [21Cohort study; Hospital USA 147 76.5 37 Patients with age ≥65 years who were screened to have cognitive impairment, admitted to general medical ward, English speaking and delirium-free at admission 
Cancelli [22Cohort study; Community Italy 750 75 38.7 Patients with age ≥65 years who were living independently or at an institution 
Cao [23Cross-sectional; Community USA 932 78 Patients included were aged ≥65 years, female Medicare beneficiaries from 1 September 1992, residents in Baltimore area, self-reported difficulty in two or more functional domain which included: (i) mobility and exercise tolerance, (ii) upper extremity function, (iii) complex activity heavily involving cognition and sensory input and (iv) basic self-care 
Carriere [24Cohort study; Community France 6912 73.7 40.3 Patients were recruited from electoral roll and were community dwelling, aged ≥65 years and from three French cities 
Cruce [25Cross-sectional; Hospital Canada 88 50.7 31.8 Patients with diagnosis of multiple sclerosis, aged 18–65 years, EDSS score <7.5, on stable dosage of classical anti-cholinergic drugs (oxybutynin or tolterodine) for bladder dysfunction for at least 6 months prior to assessment and had stable multiple sclerosis with no recent relapse or treatment with steroids within the past 3 months 
De Luise [26Cohort study; Hospital USA 10,603 NS 47.8 Patients admitted between January 1977 and December 2003 
Drag [27Cross-sectional study; Hospital USA 450 67.95 95.3 Patients admitted to Extended Care Centre and had completed the cognitive screen, had premorbid IQ ≥70 and did not have delirium or dementia 
Fox [5Cohort study; Community UK 12,423 75.2 40 Patients were a random sample of ≥65 years old, living at home and institutions 
Fox [28Cohort study; Community UK 244 81 28.6 Patients included standardised diagnosis of dementia, fulfilment of criteria for possible or probably Alzheimer's disease with age ≥55 years, had lived in North London or Essex (UK) and in contact with family or statutory carer for ≥4 h a week 
Gaudreau [29Cohort study; Hospital Canada 261 59.6 56 Patients included if they had histological diagnosis of cancer in consecutive admissions to the unit 
Geller [30Cohort study; Hospital USA 35 70.4 Patients were postmenopausal women, age ≥55 years, seeking treatment for overactive bladder and opting for anti-cholinergic therapy 
Gnijidic [31Cohort study; Community Australia 1705 77.2 100 Patients had to be born in English-speaking countries or learned English before the age of 12 years, community-dwelling men, age ≥70 years, living within the defined region of the New South Wales Electoral role whose cognition was intact or had mild cognitive impairment or dementia 
Gnjidic [32Cross-sectional study; Community Australia 1705 76.9 100 Patients included were male, aged ≥70 years and resident of South Wales 
Han [33Cohort study; Community USA 544 74.4 100 Patients included were male, aged ≥65 years and part of Connecticut Veterans Longitudinal Cohort with a diagnosis of hypertension. 
Harvey [34Randomised, double-blinded controlled trial; Community USA 377 39.9 73 Patients had to have a diagnosis of Schizophrenia, baseline positive and negative syndrome scale (PANSS score) of 60–120 and were aged 18–64 years and outpatients or inpatients hospitalised for <4 weeks 
Hilmer [35Cohort Study; Community USA 2172 73 47 Patients were community dwellers with age 70–79 years who had participated in the Health ABC study 
Kay [36Randomised, double-blinded controlled trial; Commercial trial centre USA 150 67.3 38 Patients were healthy subjects aged ≥60 years, English as first language and were able to follow instructions and complete computerised cognitive tests. Excluded if anti-cholinergic use was contraindicated or they suffered from dementia, depression or had MMSE ≤27 
Kersten [37RCT; Community Norway 87 85 61 Patients were long-term nursing home residents from 22 nursing homes. Have a total anti-cholinergic drug scale (ADS) of ≥3. Patients were not blind, deaf, aphasic, delirious or with severe dementia (Clinical Dementia Rating scale of 3) 
Kersten [38RCT; Community Norway 87 85 61 Patients were long-term nursing home residents from 22 nursing homes. Have a total ADS of ≥3. Patients were not blind, deaf, aphasic, delirious or with severe dementia (Clinical Dementia Rating scale of 3) 
Kolanowski [39Longitudinal study; Community USA 87 85.7 23 Patients were included if English speakers, with age ≥65 years, diagnosis of dementia using DSM-IV criteria, MMSE score ≥8 but <24, no new psychoactive drugs prescribed and presence of behavioural symptoms as reported by staff and documented in the latest minimum dataset 
Koyama [40Cohort study; Community USA 1484 87.5 Patients were community-dwelling women who had previously been enrolled on the Study of Osteoporotic Fractures from 1986 to 1988. A cohort of African-American women were later recruited from 1997 to 1998 
Kumpula [41Cohort study; Hospital Finland 1004 81.3 25 Patients included were living in 1 of 53 long-term care wards in seven hospitals in Helsinki. Exclusion due to incomplete medication data and unavailable mortality data 
Lampela [42Cohort study; Community Finland 621 81.7 29.8 Patients were randomly selected ≥75 years from previous cross-sectional data of Geriatric Multidisciplinary Strategy for the Good Care of the Elderly (GeMS) study with consent to participate 
Lipton [43Randomised, double-blinded crossover controlled trial; Community USA 129 71.2 41.8 Patients were age >65 years, and had to score 10 or less than on the short orientation memory and concentration test on enrolment 
Low [31Cohort study; Community Australia 2058 62.5 51.7 Patients were randomly selected from electoral roll 
Luukkanen [44Cohort study; Hospital and Community Finland 425 86.1 18.4 Patients were from geriatric wards, residential or nursing home residents aged over 70 years. Diagnosis of dementia using DSM-IV criteria, MMSE score 
Mangoni [45Cohort study; Hospital Netherlands 71 85 29.6 Patients were ≥65 admitted with hip fractures and scheduled for surgery 
Merchant [46Cross-sectional study; Community Singapore 2804 NS NS Patients were enrolled in the Singapore Longitudinal Aging Study, community dwellers and aged ≥55 years 
Pandharipande [47Cohort study; Hospital USA 198 55.5 52 Patients who were admitted onto medical/coronary ICU and were mechanically ventilated, without a baseline neurological disease to confound the assessment of delirium 
Pasina [48Cross-sectional study; Hospital Italy 1232 78.6 49.4 Patients were ≥65 years and admitted into internal medicine and geriatric wards participating in the Registry of Polytherapies SIMI (REPOSI study) in 2010 
Shah [49Cohort study; Community USA 896 74.8 30.7 Patients were community-dwelling older clergy without dementia who were participating in the Religious Orders Study—a longitudinal epidemiologic study of aging where participants have been assessed annually for a mean of 10 years 
Shakakibara [50Cohort study; Community Japan 62 70 40.3 Patients were consecutive subjects in neurology outpatients. All had diagnosis of overactive bladder. Exclusion criteria were anti-cholinergic agents within 2 weeks of entry into study, indwelling foley catheters, intermittent catheterisation, postvoid residual urine volume >100 ml, high prostate-specific antigen, acute urinary tract infection, closed angle glaucoma, diseases of anti-cholinergic contraindication 
Uusvaara [51Cohort study; community Finland 400 80 35 Patients were community dwelling, aged 75–90 years, had a diagnosis of cardiovascular disease and were a part of the Drugs and Evidence-Based Medicine in the Elderly (DEBATE) study cohort 
Uusvaara [52Cohort study; community Finland 400 80 35 Patients were community dwelling, aged 75–90 years, had a diagnosis of cardiovascular disease and were a part of the Drugs and Evidence-Based Medicine in the Elderly (DEBATE) study cohort 
Wagg [53Randomised, double-blinded, triple-crossover trial; Commercial trial centre UK 26 79 54 Patients with age ≥75 years with mild cognitive impairment and body mass index of 18–30 kg/m2. Excluded patients had short-form Geriatric Depression Scale score ≥5, and history of urinary retention or current medications to treat overactive bladder 
Wesnes [54Randomised, double-blinded, triple-crossover trial; Commercial trial centre UK 12 69.1 50 Patients with aged ≥65 years, willing and able to complete study test battery, had body mass index 18.0–30.0 kg/m2, 60–100 kg for males, 55–90 kg for females and a total score of ≥27 in the MMSE at first visit 
Whalley [55Cohort study; Community UK 281 77.1 57.6 Patients who took part in 1932 Scottish Mental Survey and not known to be in treatment for a major illness, had major sensory impairment, were not recently bereaved and born in 1921 
Wilson [56Cross-sectional study for RCT data; Community Australia 602 85.7 29.1 Patients were residents of residential aged care facilities with aged ≥70 years and likely to survive for the next 12 months 
Wilson [11Cross-sectional study for RCT data; Community Australia 602 85.7 29.1 Patients were residents of residential aged care facilities with aged ≥70 years and likely to survive for the next 12 months 
Yeh [57Case–control study; Community Taiwan 71 83.4 100 Patients had diagnosis of dementia as per the DSM-IV in a veteran (residential) home. Residents with primary diagnosis of major psychotic disorder, mental retardation, recent aggravation of behaviour and psychological symptoms of dementia, recent deterioration in health status or short-life expectancy were excluded 
Study Design; setting Country Number of participants Mean age (years) % Male Participant characteristics 
Agar [17Post hoc analysis of RCT; Hospital and community Australia 461 72 48 Patients included in Palliative Care Trial with diagnosis of cancer, known date of death, Australia—modified Karnofsky Performance Scale (AKPS) score of >60 at baseline and AKPS score falls <60 at any time during follow-up 
Ancelin [18Cohort study; Community France 372 66.2 NS Patients with age >60 years and without dementia at recruitment 
Boustani [19Cohort study; Community USA 1558 77.6 33.6 Patients with age ≥65 years and were African American 
Caeiro [20Case–control study; Hospital Portugal 74 62 55 Patients included with admission diagnosis of cerebral infarct or intracerebral haemorrhage/intraventricular haemorrhage, assessment delirium performed within 4 days after stroke onset, a Glasgow Coma Scale score ≥5 on the day of the delirium examination 
Cai [5Cohort study; Community USA 3690 72 30 Patients aged 65 and older who were living independently or with family in a community setting. Medication dispensing data defined the exposure, and a two-stage screening and diagnosis design provided the outcome assessment of cognitive impairment 
Campbell [7Cohort study; Community USA 1652 81.8 30.9 Patients with age ≥70 years who were African American, community dwelling, had normal cognitive function at baseline and enrolled in Indianapolis-Ibadan Dementia Project between 2001 and 2007 
Campbell [21Cohort study; Hospital USA 147 76.5 37 Patients with age ≥65 years who were screened to have cognitive impairment, admitted to general medical ward, English speaking and delirium-free at admission 
Cancelli [22Cohort study; Community Italy 750 75 38.7 Patients with age ≥65 years who were living independently or at an institution 
Cao [23Cross-sectional; Community USA 932 78 Patients included were aged ≥65 years, female Medicare beneficiaries from 1 September 1992, residents in Baltimore area, self-reported difficulty in two or more functional domain which included: (i) mobility and exercise tolerance, (ii) upper extremity function, (iii) complex activity heavily involving cognition and sensory input and (iv) basic self-care 
Carriere [24Cohort study; Community France 6912 73.7 40.3 Patients were recruited from electoral roll and were community dwelling, aged ≥65 years and from three French cities 
Cruce [25Cross-sectional; Hospital Canada 88 50.7 31.8 Patients with diagnosis of multiple sclerosis, aged 18–65 years, EDSS score <7.5, on stable dosage of classical anti-cholinergic drugs (oxybutynin or tolterodine) for bladder dysfunction for at least 6 months prior to assessment and had stable multiple sclerosis with no recent relapse or treatment with steroids within the past 3 months 
De Luise [26Cohort study; Hospital USA 10,603 NS 47.8 Patients admitted between January 1977 and December 2003 
Drag [27Cross-sectional study; Hospital USA 450 67.95 95.3 Patients admitted to Extended Care Centre and had completed the cognitive screen, had premorbid IQ ≥70 and did not have delirium or dementia 
Fox [5Cohort study; Community UK 12,423 75.2 40 Patients were a random sample of ≥65 years old, living at home and institutions 
Fox [28Cohort study; Community UK 244 81 28.6 Patients included standardised diagnosis of dementia, fulfilment of criteria for possible or probably Alzheimer's disease with age ≥55 years, had lived in North London or Essex (UK) and in contact with family or statutory carer for ≥4 h a week 
Gaudreau [29Cohort study; Hospital Canada 261 59.6 56 Patients included if they had histological diagnosis of cancer in consecutive admissions to the unit 
Geller [30Cohort study; Hospital USA 35 70.4 Patients were postmenopausal women, age ≥55 years, seeking treatment for overactive bladder and opting for anti-cholinergic therapy 
Gnijidic [31Cohort study; Community Australia 1705 77.2 100 Patients had to be born in English-speaking countries or learned English before the age of 12 years, community-dwelling men, age ≥70 years, living within the defined region of the New South Wales Electoral role whose cognition was intact or had mild cognitive impairment or dementia 
Gnjidic [32Cross-sectional study; Community Australia 1705 76.9 100 Patients included were male, aged ≥70 years and resident of South Wales 
Han [33Cohort study; Community USA 544 74.4 100 Patients included were male, aged ≥65 years and part of Connecticut Veterans Longitudinal Cohort with a diagnosis of hypertension. 
Harvey [34Randomised, double-blinded controlled trial; Community USA 377 39.9 73 Patients had to have a diagnosis of Schizophrenia, baseline positive and negative syndrome scale (PANSS score) of 60–120 and were aged 18–64 years and outpatients or inpatients hospitalised for <4 weeks 
Hilmer [35Cohort Study; Community USA 2172 73 47 Patients were community dwellers with age 70–79 years who had participated in the Health ABC study 
Kay [36Randomised, double-blinded controlled trial; Commercial trial centre USA 150 67.3 38 Patients were healthy subjects aged ≥60 years, English as first language and were able to follow instructions and complete computerised cognitive tests. Excluded if anti-cholinergic use was contraindicated or they suffered from dementia, depression or had MMSE ≤27 
Kersten [37RCT; Community Norway 87 85 61 Patients were long-term nursing home residents from 22 nursing homes. Have a total anti-cholinergic drug scale (ADS) of ≥3. Patients were not blind, deaf, aphasic, delirious or with severe dementia (Clinical Dementia Rating scale of 3) 
Kersten [38RCT; Community Norway 87 85 61 Patients were long-term nursing home residents from 22 nursing homes. Have a total ADS of ≥3. Patients were not blind, deaf, aphasic, delirious or with severe dementia (Clinical Dementia Rating scale of 3) 
Kolanowski [39Longitudinal study; Community USA 87 85.7 23 Patients were included if English speakers, with age ≥65 years, diagnosis of dementia using DSM-IV criteria, MMSE score ≥8 but <24, no new psychoactive drugs prescribed and presence of behavioural symptoms as reported by staff and documented in the latest minimum dataset 
Koyama [40Cohort study; Community USA 1484 87.5 Patients were community-dwelling women who had previously been enrolled on the Study of Osteoporotic Fractures from 1986 to 1988. A cohort of African-American women were later recruited from 1997 to 1998 
Kumpula [41Cohort study; Hospital Finland 1004 81.3 25 Patients included were living in 1 of 53 long-term care wards in seven hospitals in Helsinki. Exclusion due to incomplete medication data and unavailable mortality data 
Lampela [42Cohort study; Community Finland 621 81.7 29.8 Patients were randomly selected ≥75 years from previous cross-sectional data of Geriatric Multidisciplinary Strategy for the Good Care of the Elderly (GeMS) study with consent to participate 
Lipton [43Randomised, double-blinded crossover controlled trial; Community USA 129 71.2 41.8 Patients were age >65 years, and had to score 10 or less than on the short orientation memory and concentration test on enrolment 
Low [31Cohort study; Community Australia 2058 62.5 51.7 Patients were randomly selected from electoral roll 
Luukkanen [44Cohort study; Hospital and Community Finland 425 86.1 18.4 Patients were from geriatric wards, residential or nursing home residents aged over 70 years. Diagnosis of dementia using DSM-IV criteria, MMSE score 
Mangoni [45Cohort study; Hospital Netherlands 71 85 29.6 Patients were ≥65 admitted with hip fractures and scheduled for surgery 
Merchant [46Cross-sectional study; Community Singapore 2804 NS NS Patients were enrolled in the Singapore Longitudinal Aging Study, community dwellers and aged ≥55 years 
Pandharipande [47Cohort study; Hospital USA 198 55.5 52 Patients who were admitted onto medical/coronary ICU and were mechanically ventilated, without a baseline neurological disease to confound the assessment of delirium 
Pasina [48Cross-sectional study; Hospital Italy 1232 78.6 49.4 Patients were ≥65 years and admitted into internal medicine and geriatric wards participating in the Registry of Polytherapies SIMI (REPOSI study) in 2010 
Shah [49Cohort study; Community USA 896 74.8 30.7 Patients were community-dwelling older clergy without dementia who were participating in the Religious Orders Study—a longitudinal epidemiologic study of aging where participants have been assessed annually for a mean of 10 years 
Shakakibara [50Cohort study; Community Japan 62 70 40.3 Patients were consecutive subjects in neurology outpatients. All had diagnosis of overactive bladder. Exclusion criteria were anti-cholinergic agents within 2 weeks of entry into study, indwelling foley catheters, intermittent catheterisation, postvoid residual urine volume >100 ml, high prostate-specific antigen, acute urinary tract infection, closed angle glaucoma, diseases of anti-cholinergic contraindication 
Uusvaara [51Cohort study; community Finland 400 80 35 Patients were community dwelling, aged 75–90 years, had a diagnosis of cardiovascular disease and were a part of the Drugs and Evidence-Based Medicine in the Elderly (DEBATE) study cohort 
Uusvaara [52Cohort study; community Finland 400 80 35 Patients were community dwelling, aged 75–90 years, had a diagnosis of cardiovascular disease and were a part of the Drugs and Evidence-Based Medicine in the Elderly (DEBATE) study cohort 
Wagg [53Randomised, double-blinded, triple-crossover trial; Commercial trial centre UK 26 79 54 Patients with age ≥75 years with mild cognitive impairment and body mass index of 18–30 kg/m2. Excluded patients had short-form Geriatric Depression Scale score ≥5, and history of urinary retention or current medications to treat overactive bladder 
Wesnes [54Randomised, double-blinded, triple-crossover trial; Commercial trial centre UK 12 69.1 50 Patients with aged ≥65 years, willing and able to complete study test battery, had body mass index 18.0–30.0 kg/m2, 60–100 kg for males, 55–90 kg for females and a total score of ≥27 in the MMSE at first visit 
Whalley [55Cohort study; Community UK 281 77.1 57.6 Patients who took part in 1932 Scottish Mental Survey and not known to be in treatment for a major illness, had major sensory impairment, were not recently bereaved and born in 1921 
Wilson [56Cross-sectional study for RCT data; Community Australia 602 85.7 29.1 Patients were residents of residential aged care facilities with aged ≥70 years and likely to survive for the next 12 months 
Wilson [11Cross-sectional study for RCT data; Community Australia 602 85.7 29.1 Patients were residents of residential aged care facilities with aged ≥70 years and likely to survive for the next 12 months 
Yeh [57Case–control study; Community Taiwan 71 83.4 100 Patients had diagnosis of dementia as per the DSM-IV in a veteran (residential) home. Residents with primary diagnosis of major psychotic disorder, mental retardation, recent aggravation of behaviour and psychological symptoms of dementia, recent deterioration in health status or short-life expectancy were excluded 

DSM-IV, Diagnostic and Statistical Manual of Mental Disorders-IV; EDSS, Extended Disability Status Scale; ICU, intensity care unit; IQ, intelligence quotients; kg, kilograms; kg/m2, kilograms per square meter; MMSE, Mini-Mental State Examination; NS, not stated; PANSS, Positive and Negative Syndrome Scale; RCT, Randomised Controlled Trial; UK, United Kingdom; USA, United States of America; yrs, years.

In total, 60,944 participants, with mean age range of 39.9–87.5 years were included. This consisted of 25,225 males and 32,543 females; cohort gender proportions were not stated in two studies [18, 46]. Thirty-three studies were conducted in community dwellings and 14 studies in hospital settings, one study was conducted across both settings [44]. Participants in the hospital settings were admitted for a variety of medical reasons including cancer [29], general frailty/long-term care [41] and acute bladder symptoms [25, 30]. Supplementary data available in Age and Ageing online, Table S1 illustrate the estimates of anti-cholinergic load or burden to estimate the ACB in each included study.

Results of risk of bias assessment

The results of the critical appraisal and risk of bias assessments are presented in Supplementary data available in Age and Ageing online, Tables S2–S4. The findings indicate that the evidence-base was largely moderate in methodological quality.

Supplementary data available in Age and Ageing online, Table S3 present the appraisal results of the single case–control study. The results indicated that whilst demonstrating a number of key strengths, the evidence-base was unclear on the validity of the comparison between cases and control participants, with unacceptable non-response rates demonstrated for the study cohort.

The Cochrane Risk of Bias tool for RCTs demonstrated the evidence presented with a moderate risk of bias. Several studies poorly demonstrated the randomisation procedures, and were limited by incomplete analysis of the dataset through intention-to-treat principles, and rarely adjusted analyses for missing data, thereby reducing the strength of these statistical analyses.

Data synthesis

A summary of the results of each included study is presented in Table 3 with a more comprehensive summary as Supplementary data available in Age and Ageing online, Table S5.

Table 3.

Executive summary of the results of the included studies by outcome of interest

Study Result interpretation 
Outcome 1: cognitive function 
 Ancelin [18Use of anti-cholinergic drugs is associated with mild cognitive impairment but not increased risk of dementia 
 Boustani [19Use of anti-cholinergic drugs was associated with a higher risk of incident cognitive impairment 
 Cai [6Use of at least three medications with ACB score of 1 for 90 days, or use of at least one medication with ACB score of 3 for 60 days, increases the risk of mild cognitive impairment 
 Campbell [21Use of anti-cholinergic drugs is associated with a significant increase in incident cognitive impairment 
 Cancelli [22Use of anti-cholinergic drugs is associated with a significant increase in cognitive impairment 
 Cao [23Use of anti-cholinergic drugs is associated with a significant risk of cognitive impairment 
 Carriere [24Use of anti-cholinergic drugs increases risk for cognitive impairment 
 Cruce [25Use of anti-cholinergic drugs for bladder symptoms in patients with MS has a negative impact on cognitive function 
 Drag [27Use of anti-cholinergic drugs is not associated with lower performance on cognitive measures 
 Fox [5Use of anti-cholinergic drugs is associated with increased risk of cognitive impairment 
 Fox [28Use of anti-cholinergic drugs in patients with Alzheimer's disease is not associated with deterioration in cognition 
 Geller [30Trospium chloride use is associated with significant difference in cognition 
 Gnijidic [31Use of anti-cholinergic drugs is not associated with increased risk of limitations in cognitive performance, mild cognitive impairment or dementia 
 Han [33Use of anti-cholinergic medications is associated with reduction in cognitive function 
 Harvey [34Use of atypical antipsychotics is not associated with significant risk of cognitive impairment 
 Kay [36Use of darifenacin is not associated with cognitive impairment but oxybutynin leads to cognitive impairment 
 Kersten [37Reduction of anti-cholinergic medications has no significant effects on cognitive function improvement 
 Kersten [38Increasing ADS scores is not associated with decrease in cognitive function 
 Kolanowski [39Use of anti-cholinergic medication is not associated with cognitive impairment 
 Koyama [40Higher anti-cholinergic load was significantly associated with poorer cognitive function at 10-year follow-up 
 Lampela [42Use of anti-cholinergic medications is associated with cognitive impairment 
 Lipton [43Use of darifenacin is not associated with significant difference in cognitive function 
 Low [58Use of anti-cholinergic medication is associated with lower level of complex attention in the young-old but not with greater cognitive decline 
 Merchant [46Use of anti-cholinergic drugs is associated with increased risk of cognitive impairment 
 Pasina [48Cumulative effects of anti­cholinergic drugs as assessed by ACB scale and ARS is associated with cognitive impairment 
 Shah [49There is a gradation in annual rate of cognitive function decline amongst incident users compared with never users. However, there was no significant difference between prevalent users and never users 
 Shakakibara [50Imidafenacin has no effect on cognitive function 
 Uusvaara [52DAPs may be associated with specific impairments in cognitive functioning. 
 Wagg [53Use of solifenacin is not associated with increased risk of cognitive impairment but significant differences are observed for oxybutynin 
 Wesnes [54Use of solifenacin is not associated with increased risk of cognitive impairment but significant differences are observed for oxybutynin 
 Whalley [55Use of anti-cholinergic drugs is associated with increased risk of cognitive impairment but not dementia 
 Wilson [56Use of anti-cholinergic drugs is not associated with increased risk of cognitive impairment 
 Yeh [57Reduction in anti-cholinergic drugs did not show in cognitive function improvement 
Outcome 2: delirium 
 Caeiro [20Use of anti-cholinergic drugs is associated with increased risk of delirium 
 Campbell [7Use of anti-cholinergic drugs is not associated with a significant difference in delirium 
 Gaudreau [29Use of anti-cholinergic drugs is not associated with significant difference in delirium 
 Luukkanen [44Use of anti-cholinergic drugs is not associated with a risk of development of delirium 
 Pandharipande [47Use of anti-cholinergic drugs is not associated with the development of delirium 
Outcome 3: physical function 
 Gnjidic [32No significant difference in chair stands, walking speed, narrow walk, balance and instrument activities of daily living 
 Han [33Use of anti-cholinergic drugs is associated with poorer performance on the instrument activities of daily living 
 Hilmer [35Use of anti-cholinergic drugs is associated with poorer performance on the instrument activities of daily living 
 Kersten [38Higher ADS scores are associated with higher ADL scores with no significant differences 
 Lampela [42Higher anti-cholinergic scores are associated with reduced ADL and IADL scores 
 Pasina [40Cumulative effects of anti-cholinergic drugs assessed by ACB and ARS scale is associated with functional impairment 
 Wilson [56Use of anti-cholinergic drugs is associated with greater use of mobility aids 
 Yeh [57Reduction in anti-cholinergic burden did not show benefits in functional outcome improvements 
Outcome 4: mortality 
 Agar [17Use of drugs with anti-cholinergic properties is not associated with any difference in mortality 
 De Luise [26Tiotropium use is associated with lower mortality 
 Fox [5There was a dose–response effect of ACB score associated with mortality at 2 years 
 Kumpula [41Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
 Luukkanen [44Use of anti-cholinergic drugs is not associated with an increased risk of mortality 
 Mangoni [45Use of anti-cholinergic drug is associated with increased mortality 
 Uusvaara [51Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
 Wilson [11Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
 Wilson [56Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
Study Result interpretation 
Outcome 1: cognitive function 
 Ancelin [18Use of anti-cholinergic drugs is associated with mild cognitive impairment but not increased risk of dementia 
 Boustani [19Use of anti-cholinergic drugs was associated with a higher risk of incident cognitive impairment 
 Cai [6Use of at least three medications with ACB score of 1 for 90 days, or use of at least one medication with ACB score of 3 for 60 days, increases the risk of mild cognitive impairment 
 Campbell [21Use of anti-cholinergic drugs is associated with a significant increase in incident cognitive impairment 
 Cancelli [22Use of anti-cholinergic drugs is associated with a significant increase in cognitive impairment 
 Cao [23Use of anti-cholinergic drugs is associated with a significant risk of cognitive impairment 
 Carriere [24Use of anti-cholinergic drugs increases risk for cognitive impairment 
 Cruce [25Use of anti-cholinergic drugs for bladder symptoms in patients with MS has a negative impact on cognitive function 
 Drag [27Use of anti-cholinergic drugs is not associated with lower performance on cognitive measures 
 Fox [5Use of anti-cholinergic drugs is associated with increased risk of cognitive impairment 
 Fox [28Use of anti-cholinergic drugs in patients with Alzheimer's disease is not associated with deterioration in cognition 
 Geller [30Trospium chloride use is associated with significant difference in cognition 
 Gnijidic [31Use of anti-cholinergic drugs is not associated with increased risk of limitations in cognitive performance, mild cognitive impairment or dementia 
 Han [33Use of anti-cholinergic medications is associated with reduction in cognitive function 
 Harvey [34Use of atypical antipsychotics is not associated with significant risk of cognitive impairment 
 Kay [36Use of darifenacin is not associated with cognitive impairment but oxybutynin leads to cognitive impairment 
 Kersten [37Reduction of anti-cholinergic medications has no significant effects on cognitive function improvement 
 Kersten [38Increasing ADS scores is not associated with decrease in cognitive function 
 Kolanowski [39Use of anti-cholinergic medication is not associated with cognitive impairment 
 Koyama [40Higher anti-cholinergic load was significantly associated with poorer cognitive function at 10-year follow-up 
 Lampela [42Use of anti-cholinergic medications is associated with cognitive impairment 
 Lipton [43Use of darifenacin is not associated with significant difference in cognitive function 
 Low [58Use of anti-cholinergic medication is associated with lower level of complex attention in the young-old but not with greater cognitive decline 
 Merchant [46Use of anti-cholinergic drugs is associated with increased risk of cognitive impairment 
 Pasina [48Cumulative effects of anti­cholinergic drugs as assessed by ACB scale and ARS is associated with cognitive impairment 
 Shah [49There is a gradation in annual rate of cognitive function decline amongst incident users compared with never users. However, there was no significant difference between prevalent users and never users 
 Shakakibara [50Imidafenacin has no effect on cognitive function 
 Uusvaara [52DAPs may be associated with specific impairments in cognitive functioning. 
 Wagg [53Use of solifenacin is not associated with increased risk of cognitive impairment but significant differences are observed for oxybutynin 
 Wesnes [54Use of solifenacin is not associated with increased risk of cognitive impairment but significant differences are observed for oxybutynin 
 Whalley [55Use of anti-cholinergic drugs is associated with increased risk of cognitive impairment but not dementia 
 Wilson [56Use of anti-cholinergic drugs is not associated with increased risk of cognitive impairment 
 Yeh [57Reduction in anti-cholinergic drugs did not show in cognitive function improvement 
Outcome 2: delirium 
 Caeiro [20Use of anti-cholinergic drugs is associated with increased risk of delirium 
 Campbell [7Use of anti-cholinergic drugs is not associated with a significant difference in delirium 
 Gaudreau [29Use of anti-cholinergic drugs is not associated with significant difference in delirium 
 Luukkanen [44Use of anti-cholinergic drugs is not associated with a risk of development of delirium 
 Pandharipande [47Use of anti-cholinergic drugs is not associated with the development of delirium 
Outcome 3: physical function 
 Gnjidic [32No significant difference in chair stands, walking speed, narrow walk, balance and instrument activities of daily living 
 Han [33Use of anti-cholinergic drugs is associated with poorer performance on the instrument activities of daily living 
 Hilmer [35Use of anti-cholinergic drugs is associated with poorer performance on the instrument activities of daily living 
 Kersten [38Higher ADS scores are associated with higher ADL scores with no significant differences 
 Lampela [42Higher anti-cholinergic scores are associated with reduced ADL and IADL scores 
 Pasina [40Cumulative effects of anti-cholinergic drugs assessed by ACB and ARS scale is associated with functional impairment 
 Wilson [56Use of anti-cholinergic drugs is associated with greater use of mobility aids 
 Yeh [57Reduction in anti-cholinergic burden did not show benefits in functional outcome improvements 
Outcome 4: mortality 
 Agar [17Use of drugs with anti-cholinergic properties is not associated with any difference in mortality 
 De Luise [26Tiotropium use is associated with lower mortality 
 Fox [5There was a dose–response effect of ACB score associated with mortality at 2 years 
 Kumpula [41Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
 Luukkanen [44Use of anti-cholinergic drugs is not associated with an increased risk of mortality 
 Mangoni [45Use of anti-cholinergic drug is associated with increased mortality 
 Uusvaara [51Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
 Wilson [11Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 
 Wilson [56Use of drugs with anti-cholinergic properties is associated with a non-significant trend towards increased mortality 

ABS, anti-cholinergic burden score; ACB, anti-cholinergic burden; ADL, activities of daily living; ADS, anti-cholinergic drug scale; ARS, anti-cholinergic risk scale; IADL, instrumental activities of daily living; MS, multiple sclerosis.

Anti-cholinergic effect on cognitive function

Thirty-three studies reported the impact of medications with anti-cholinergic properties on cognitive function [5–7, 18,19, 22–25, 27, 28, 30, 31, 33, 34, 36–40, 42, 43, 46, 48–50, 52–58]. This was evaluated with a number of tools, most commonly the MMSE.

There was a repeated finding of an association between anti-cholinergic medications and a significant decline in cognitive ability, as demonstrated by 23 studies [5–7, 18, 19, 22–25, 28, 30, 31, 33, 40, 42, 43, 46, 48, 52–55, 58].

Ten studies reported no significant association between medicines with anti-cholinergic properties and cognitive function [27, 28, 34, 36–39, 49, 50, 57]. A number of differences in study design and variable definitions may have accounted for the study results, such as differences in medications evaluated (single versus scale-based identification of anti-cholinergics), characteristics of the control group, duration of study and measurement of dose–effect.

Anti-cholinergic effect on delirium

Five studies assessed the impact of anti-cholinergic burden on delirium [20, 21, 29, 44, 47]. Only one study demonstrated a significant association between drugs with anti-cholinergic properties and delirium [20]. Delirium, as assessed by the Delirium Rating Scale, was more likely in people prescribed medicines with anti-cholinergic properties prior to a stroke (OR: 11.3; 95% CI: 1.19–108.2) or during hospitalisation (OR: 5.82; 95% CI: 1.96–17.2), compared with those not prescribed anti-cholinergics [20].

Luukkanen et al. [44], Pandharipande et al. [47], Campbell et al. [21] and Gaudeau et al. [29] reported contrary findings, reporting no association between the use of medicines with anti-cholinergic properties and delirium.

Anti-cholinergic effect on physical function

Eight studies assessed the impact of medications with anti-cholinergic properties on physical function [32, 33, 35, 38, 40, 42, 56, 57]. Of five studies reported that anti-cholinergic drugs were associated with reduced physical function [32, 33, 35, 40, 42].

Three studies reported no association between ACB and physical function [38, 56, 57]. Wilson et al. [56] assessed the level of mobility. They reported that participants could walk without the use of a walking aid in 46% of the Drug Burden Index (DBI) category 0, 37.7% of those in the DBI category <1 and 34% of those in the DBI category > 1. Whilst this is only one aspect of function, this does provide some conflicting evidence against the evidence-base.

Anti-cholinergic effect on mortality

Nine studies investigated the effect of anti-cholinergic medications on mortality [5, 11, 17, 26, 41, 44, 45, 51, 56]. Six studies reported that the use of drugs with anti-cholinergic properties was not statistically associated with increased mortality [11, 17, 41, 44, 51, 56].

Three studies reported contrary findings [5, 26, 48]. Fox et al. [5] reported that after adjusting for key variables including baseline MMSE score and number of non-anti-cholinergic medications, every additional point on the ACB scale increased the odds of death by 26% (OR: 1.26, 95% CI: 1.20–1.32) [5]. De Luise et al. [26] reported a risk ratio of tiotropium use and total mortality of 0.77 (95% CI: 0.56–0.78). However, the population considered in this study is notably different from other assessments of anti-cholinergic use with a focus on respiratory disease (Table 2).

Discussion

This is the first systematic review to assess the effects of medications with anti-cholinergic properties on delirium and physical function, and an important update on cognitive function and mortality. The findings indicate that medicines with anti-cholinergic properties have a negative effect on cognitive function. The results also indicated no significant association between anti-cholinergic load and either mortality or delirium. Single- or limited-drug studies in the past have supported the relationship between these medicines and delirium. Using anti-cholinergic drug scales to identify all medications with anti-cholinergic properties did not appear to confirm such an association. Finally, this review identified that the use of medications with anti-cholinergic properties may be associated with a deterioration in physical function.

The negative effect of increased anti-cholinergic load on adverse cognitive outcomes revealed the strongest association throughout the evidence-base. This is in keeping with previous literature reviews which have examined the effect of anti-cholinergic burden on cognition [3, 4]. A previous review [3] noted that the effect of anti-cholinergic drugs on cognition is not always due to one anti-cholinergic drug alone, but instead an accumulation of a number of drugs with anti-cholinergic properties is an important consideration. Therefore, in this systematic review, it was decided to only include studies which quantified this load; either through dosage of the drug used in an RCT, or through using scales to quantify ACB.

Of those studies which did not report a negative effect of increased anti-cholinergic load on cognitive function, three of the five studies had relatively shorter follow-up periods (<2 years) or were cross-sectional studies. As studies included in the review had a highly variable follow-up period, ranging from a few weeks to 12 years, it was difficult to interpret whether the studies with short follow-ups would have progressed to report a significant association with cognitive decline should they have included a longer follow-up period.

There is little support that anti-cholinergic medications increase the risk of mild cognitive impairment, which then presents a risk of developing dementia. In addition, there is little, if any, evidence for a non-reversible impact of anti-cholinergics on cognition. Consequently, the findings of this review on cognition should be interpreted with caution until the evidence-base develops in this area.

In contrast to the continued evidence supporting anti-cholinergic burden and its effect on cognitive function, its effect on developing the more acute form of cognitive impairment, delirium, was less coherent. Once again these studies were heterogeneous in their quantification of anti-cholinergic load and reporting the outcome. Not all studies used the Confusion Assessment Method which is thought to be the most accurate and currently recommended diagnostic tool for delirium by the National Institute of Clinical Excellence guidance in the UK. This review did not find an association between anti-cholinergic load with delirium as reported in previous reviews [14], although this review was based on a larger, more contemporary literature dataset. One reason for this may be the improvement in characterising delirium not used in more historic studies. In addition what may have been seen in previous studies is the miscoding of delirium for cognitive impairment, therefore the diagnosis of delirium may not have been made, rather the term ‘dementia’ used to categorise all participants with cognitive impairment.

The majority of studies which reported the effect of anti-cholinergic medications on physical function reported a significant inverse relationship; the higher the anti-cholinergic burden the lower the physical functioning [33, 35, 56]. Yet again there were different methods of quantifying anti-cholinergic load used and different methods of measuring physical function. Avoiding anti-cholinergic medications may therefore preserve and maximise function and prevent acute adverse events such as falls.

The effect of anti-cholinergic burden on mortality present inconclusive findings. The majority of studies reported no statistical association between these variables. However, in a large prospective cohort study undertaken by Fox et al. [5], an adjusted statistically significant negative effect of increased anti-cholinergic load with increased mortality was reported [59]. Whilst this difference in results may be due to the heterogenic nature of studies available, further research should be conducted in this area. Furthermore, these mortality figures should be viewed with caution given that the follow-up periods for these studies were insufficient, ranging from 8.9 weeks [17] to 3.3 years [51].

This paper is the first systematic review to examine the effect of anti-cholinergic medication load (excluding those measured by serum anti-cholinergic alone) on cognitive function, physical function, delirium and mortality over a large time frame spanning decades of research from published and unpublished sources. However, the included articles contained a number of limitations. First, due to the data available, the analysis focused on estimating the presence or absence of significant associations in drug response rather than estimating effect size which was not possible in this instance. Secondly, the MMSE was the major measure of cognitive change. This could be argued as an inappropriate tool for this means in its sensitivity and scope, whilst mortality, given the multi-factorial cause of this end-point, may have been insensitive to evaluation specifically against anti-cholinergeric agents. The approach in accounting for covariates, both confounders and effect modifiers, varied considerably across the included studies. There was variability in age and seriousness of medical morbidity for which the various drugs were prescribed; these factors are critical. Studies did not consider sub-clinical disease, which may have confounded any associations, but medication use was associated with health status, so despite adjusting for many health-related factors, the possibility of residual confounders between health status and outcome could not be excluded. Indication bias may have limited this review in those participants with, for example, cognitive impairment may have been more likely to be exposed to anti-cholinergic medications because of the presence of cognitive impairment. The ratings of anti-cholinergic exposure varied considerably across the included studies. In some studies, this was presented on a standard scale, in others as a dichotomous (yes/no) scale. Additionally, the reliability of these ratings across studies was difficult to establish. Managing these confounders should be considered when designing future studies in this area. Finally, the principle limitation to this review is the variability in study designs. This may be a contributing factor to the different effects being reported on physical function, delirium and mortality. However, by managing these studies separately during the analysis, and considering the appraisal of study quality, the distinction between higher and lower quality evidence was made during the interpretation of findings.

Conclusions

This systematic review provides strong evidence for the adverse effect of increased anti-cholinergic load on cognition. The results also show consistent evidence that medicines with anti-cholinergic properties may be associated with reduced physical function. The effect on delirium and mortality appear less well-defined across the literature. Further evidence is required to truly establish their association with increasing anti-cholinergic burden.

Key points

• Medicines with anti-cholinergic properties have a significant adverse effect on cognitive and physical function.

• Medicines with anti-cholinergic properties appear to have limited effect on delirium or mortality outcomes.

• The assessment of medicines with anti-cholinergic properties should be further evaluated with people at risk of poor outcome.

Supplementary data

Supplementary data mentioned in the text are available to subscribers in Age and Ageing online.

Ethical considerations

No ethical approvals were required for this study design.

Conflicts of interest

None declared.

References

1
Ness
J
Hoth
A
Barnett
M
Shorr
R
Kaboli
P
Anti-cholinergic medications in community-dwelling older veterans: prevalence of anti-cholinergic symptoms, symptom burden and adverse drug events
Am J Geriatr Pharamacother
 , 
2006
, vol. 
4
 (pg. 
42
-
51
)
2
Gurwitz
J
Field
T
Harrold
L
Incidence and preventability of adverse drug events amongst older persons in the ambulatory setting
JAMA
 , 
2003
, vol. 
289
 (pg. 
1107
-
16
)
3
Campbell
N
Boustani
M
Limbil
T
, et al.  . 
The cognitive impact of anti-cholinergics: a clinical review
Clin Interv Aging
 , 
2009
, vol. 
4
 (pg. 
225
-
33
)
4
Tannenbaum
C
Paquette
A
Hilmer
S
Holroyd-Leduc
J
Camahan
R
A systematic review of amnestic and non-amnestic mild cognitive impairment induced by anticholinergic, antihistamine, GABAergic and opioid drugs
Drugs Aging
 , 
2012
, vol. 
29
 (pg. 
639
-
58
)
5
Fox
C
Richardson
K
Maidment
ID
, et al.  . 
Anticholinergic medication use and cognitive impairment in the older population: the medical research council cognitive function and ageing study
J Am Geriatr Soc
 , 
2011
, vol. 
59
 (pg. 
1477
-
83
)
6
Cai
X
Campbell
N
Khan
B
Callahan
C
Boustani
M
Long-term anticholinergic use and the aging brain
Alzheimers Dement
 , 
2012
7
Campbell
NL
Boustani
MA
Lane
KA
, et al.  . 
Use of anticholinergics and the risk of cognitive impairment in an African American population
Neurology
 , 
2010
, vol. 
75
 (pg. 
152
-
9
)
8
Petersen
RC
Roberts
RO
Knopman
DS
, et al.  . 
The Mayo clinical study of aging: incidence of mild cognitive impairment
Alzheimers Dement
 , 
2008
, vol. 
4
 
9
Bell
J
Mezrani
C
Blacker
N
, et al.  . 
Anticholinergic and sedative medicines – prescribing considerations for people with dementia
Aust Fam Physician
 , 
2012
, vol. 
41
 (pg. 
45
-
9
)
10
Barber
SE
Clegg
AP
Young
JB
Is there a role for physical activity in preventing cognitive decline in people with mild cognitive impairment?
Age Ageing
 , 
2012
, vol. 
41
 (pg. 
5
-
8
)
11
Wilson
NM
Hilmer
SN
March
LM
, et al.  . 
Associations between drug burden index and mortality in older people in residential aged care facilities
Drugs Aging
 , 
2012
, vol. 
29
 (pg. 
157
-
65
)
12
Moher
D
Liberati
A
Tetzlaff
J
Altman
DG
PRISMA Group
Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement
BMJ
 , 
2009
, vol. 
339
 pg. 
b2535
 
13
Campbell
NL
Maidment
I
Fox
C
Kahn
B
Boustani
M
The 2012 update to the Anticholinergic Cognitive Burden Scale
J Am Geriatr Soc
 , 
2013
, vol. 
61
 
Suppl. 1
(pg. 
S142
-
3
)
14
Dyer
C
Ashton
C
Teasdale
TA
Postoperative delirium: a review of 80 primary data-collection studies
Arch Intern Med
 , 
1995
, vol. 
155
 (pg. 
461
-
5
)
15
Wells
GA
Shea
B
O'Connell
D
, et al.  . 
 
The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analysis www.ohri.ca/programs/clinical_epidemiology/oxford.asp
16
Higgins
JPT
Altman
DG
Higgins
JPT
Green
S
Assessing risk of bias in included studies
Cochrane Handbooks for Systematic Reviews of Interventions
 , 
2011
Chichester, UK
Wiley-Blackwell
(pg. 
187
-
242
)
17
Agar
M
To
T
Plummer
J
Abernethy
A
Currow
D
Anti-cholinergic load, health care utilization, and survival in people with advanced cancer: a pilot study
J Palliative Med
 , 
2010
, vol. 
13
 (pg. 
745
-
52
)
18
Ancelin
ML
Artero
S
Portet
F
Dupuy
A-M
Touchon
J
Ritchie
K
Non-degenerative mild cognitive impairment in elderly people and use of anticholinergic drugs: longitudinal cohort study
BMJ
 , 
2006
, vol. 
332
 (pg. 
455
-
8
)
19
Boustani
M
Hall
KS
Lane
KA
, et al.  . 
The association between cognition and histamine-2 receptor antagonists in African Americans
J Am Geriatr Soc
 , 
2007
, vol. 
55
 (pg. 
1248
-
53
)
20
Caeiro
L
Ferro
JM
Claro
MI
Coelho
J
Albuquerque
R
Figueira
ML
Delirium in acute stroke: a preliminary study of the role of anticholinergic medications
Eur J Neurol
 , 
2004
, vol. 
11
 (pg. 
699
-
704
)
21
Campbell
N
Perkins
A
Hui
S
Khan
B
Boustani
M
Association between prescribing of anticholinergic medications and incident delirium: a cohort study
J Am Geriatr Soc
 , 
2011
, vol. 
59
 (pg. 
277
-
81
)
22
Cancelli
I
Gigli
GL
Piani
A
, et al.  . 
Drugs with anticholinergic properties as a risk factor for cognitive impairment in elderly people: a population-based study
J Clin Psychopharmacol
 , 
2008
, vol. 
28
 (pg. 
654
-
9
)
23
Cao
Y-J
Mager
DE
Simonsick
EM
, et al.  . 
Physical and cognitive performance and burden of anticholinergics, sedatives, and ACE inhibitors in older women
Clin Pharmacol Ther
 , 
2008
, vol. 
83
 (pg. 
422
-
9
)
24
Carriere
I
Fourrier-Reglat
A
Dartigues
J-F
, et al.  . 
Drugs with anticholinergic properties, cognitive decline, and dementia in an elderly general population: the 3-city study
Arch Intern Med
 , 
2009
, vol. 
169
 (pg. 
1317
-
24
)
25
Cruce
R
Vosoughi
R
Freedman
MS
Cognitive impact of anticholinergic medication in MS: adding insult to injury?
Mult Scler Relat Disord
 , 
2012
, vol. 
1
 (pg. 
156
-
61
)
26
De Luise
C
Lanes
SF
Jacobsen
J
Pedersen
L
Sorensen
HT
Cardiovascular and respiratory hospitalizations and mortality among users of tiotropium in Denmark
Eur J Epidemiol
 , 
2007
, vol. 
22
 (pg. 
267
-
72
)
27
Drag
LL
Wright
SL
Bieliauskas
LA
Prescribing practices of anticholinergic medications and their association with cognition in an extended care setting
J Appl Gerontol
 , 
2012
, vol. 
31
 (pg. 
239
-
59
)
28
Fox
C
Livingston
G
Maidment
ID
, et al.  . 
The impact of anticholinergic burden in Alzheimer's dementia-the LASER-AD study
Age Ageing
 , 
2011
, vol. 
40
 (pg. 
730
-
5
)
29
Gaudreau
JD
Gagnon
P
Harel
F
Roy
MA
Tremblay
A
Psychoactive medications and risk of delirium in hospitalized cancer patients
J Clin Oncol
 , 
2005
, vol. 
23
 (pg. 
6712
-
8
)
30
Geller
EJ
Crane
AK
Wells
EC
, et al.  . 
Effect of anticholinergic use for the treatment of overactive bladder on cognitive function in postmenopausal women
Clin Drug Invest
 , 
2012
, vol. 
32
 (pg. 
697
-
705
)
31
Gnjidic
D
Le Couteur
DG
Naganathan
V
, et al.  . 
Effects of drug burden index on cognitive function in older men
J Clin Psychopharmacol
 , 
2012
, vol. 
32
 (pg. 
273
-
7
)
32
Gnjidic
D
Cumming
RG
Le Couteur
DG
, et al.  . 
Drug Burden Index and physical function in older Australian men
Br J Clin Pharmacol
 , 
2009
, vol. 
68
 (pg. 
97
-
105
)
33
Han
L
Agostini
JV
Allore
HG
Cumulative anticholinergic exposure is associated with poor memory and executive function in older men
J Am Geriatr Soc
 , 
2008
, vol. 
56
 (pg. 
2203
-
10
)
34
Harvey
PD
Green
MF
McGurk
SR
Meltzer
HY
Changes in cognitive functioning with risperidone and olanzapine treatment: a large-scale, double-blind, randomized study
Psychopharmacology
 , 
2003
, vol. 
169
 (pg. 
404
-
11
)
35
Hilmer
SN
Mager
DE
Simonsick
EM
, et al.  . 
Drug Burden Index Score and functional decline in older people
Am J Med
 , 
2009
, vol. 
122
 (pg. 
1142
-
9
)
36
Kay
G
Crook
T
Rekeda
L
, et al.  . 
Differential effects of the antimuscarinic agents darifenacin and oxybutynin ER on memory in older subjects
Eur Urol
 , 
2006
, vol. 
50
 (pg. 
317
-
26
)
37
Kersten
H
Molden
E
Tolo
IK
Skovlund
E
Engedal
K
Wyller
TB
Cognitive effects of reducing anticholinergic drug burden in a frail elderly population: a randomized controlled trial
J Gerontol A Biol Sci Med Sci
 , 
2013
, vol. 
68
 (pg. 
271
-
8
)
38
Kersten
H
Molden
E
Willumsen
T
Engedal
K
Wyller
TB
Higher anticholinergic drug scale (ADS) scores are associated with peripheral but not cognitive markers of cholinergic blockade. Cross sectional data from 21 Norwegian nursing homes
Br J Clin Pharmacol
 , 
2013
, vol. 
75
 (pg. 
842
-
9
)
39
Kolanowski
A
Fick
DM
Campbell
J
Litaker
M
Boustani
M
A preliminary study of anticholinergic burden and relationship to a quality of life indicator, engagement in activities, in nursing home residents with dementia
J Am Med Dir Assoc
 , 
2009
, vol. 
10
 (pg. 
252
-
7
)
40
Koyama
A
Steinman
M
Ensrud
K
Hillier
TA
Yaffe
K
Ten-year trajectory of potentially inappropriate medications in very old women: importance of cognitive status
J Am Geriatr Soc
 , 
2013
, vol. 
61
 (pg. 
258
-
63
)
41
Kumpula
EK
Bell
JS
Soini
H
Pitkala
KH
Anticholinergic drug use and mortality among residents of long-term care facilities: a prospective cohort study
J Clin Pharmacol
 , 
2011
, vol. 
51
 (pg. 
256
-
63
)
42
Lampela
P
Lavikainen
P
Garcia-Horsman
JA
Bell
JS
Huupponen
R
Hartikainen
S
Anticholinergic drug use, serum anticholinergic activity, and adverse drug events among older people: a population-based study
Drugs Aging
 , 
2013
, vol. 
30
 (pg. 
321
-
30
)
43
Lipton
RB
Kolodner
K
Wesnes
K
Assessment of cognitive function of the elderly population: effects of darifenacin
J Urol
 , 
2005
, vol. 
173
 (pg. 
493
-
8
)
44
Luukkanen
MJ
Uusvaara
J
Laurila
JV
, et al.  . 
Anticholinergic drugs and their effects on delirium and mortality in the elderly
Dement Geriatr Cogn Dis Extra
 , 
2011
, vol. 
1
 (pg. 
43
-
50
)
45
Mangoni
AA
Munster
BCV
Woodman
RJ
Rooij
SED
Measures of anticholinergic drug exposure, serum anticholinergic activity, and all-cause postdischarge mortality in older hospitalized patients with hip fractures
Am J Geriatr Psychiatry
 , 
2013
, vol. 
21
 (pg. 
785
-
93
)
46
Merchant
RA
Li
B
Yap
KB
Ng
TP
Use of drugs with anticholinergic effects and cognitive impairment in community-living older persons
Age Ageing
 , 
2009
, vol. 
38
 (pg. 
105
-
8
)
47
Pandharipande
P
Shintani
A
Peterson
J
, et al.  . 
Lorazepam is an independent risk factor for transitioning to delirium in intensive care unit patients
Anesthesiology
 , 
2006
, vol. 
104
 (pg. 
21
-
6
)
48
Pasina
L
Djade
CD
Lucca
U
, et al.  . 
Association of anticholinergic burden with cognitive and functional status in a cohort of hospitalized elderly: comparison of the anticholinergic cognitive burden scale and anticholinergic risk scale: results from the REPOSI study
Drugs Aging
 , 
2013
, vol. 
30
 (pg. 
103
-
12
)
49
Shah
RC
Janos
AL
Kline
JE
, et al.  . 
Cognitive decline in older persons initiating anticholinergic medications
PLoS ONE
 , 
2013
, vol. 
8
 pg. 
5
 
50
Sakakibara
R
Tateno
F
Yano
M
, et al.  . 
Imidafenacin on bladder and cognitive function in neurologic OAB patients
Clin Auton Res
 , 
2013
, vol. 
23
 (pg. 
189
-
95
)
51
Uusvaara
J
Pitkala
KH
Kautiainen
H
Tilvis
RS
Strandberg
TE
Association of anticholinergic drugs with hospitalization and mortality among older cardiovascular patients: a prospective study
Drugs Aging
 , 
2011
, vol. 
28
 (pg. 
131
-
8
)
52
Uusvaara
J
Pitkala
KH
Kautiainen
H
Tilvis
RS
Strandberg
TE
Detailed cognitive function and use of drugs with anticholinergic properties in older people: a community-based cross-sectional study
Drugs Aging
 , 
2013
, vol. 
30
 (pg. 
177
-
82
)
53
Wagg
AS
Dale
M
Tretter
R
Stowe
B
Compion
G
Solifenacin and cognitive function in elderly people with mild cognitive impairment: the SENIOR study
Eur Urol Suppl
 , 
2012
, vol. 
11
 (pg. 
e689
-
e689a
)
54
Wesnes
KA
Edgar
C
Tretter
RN
Bolodeoku
J
Exploratory pilot study assessing the risk of cognitive impairment or sedation in the elderly following single doses of solifenacin 10 mg
Expert Opin Drug Saf
 , 
2009
, vol. 
8
 (pg. 
615
-
26
)
55
Whalley
LJ
Sharma
S
Fox
HC
, et al.  . 
Anticholinergic drugs in late life: adverse effects on cognition but not on progress to dementia
J Alzheimer's Dis.
 , 
2012
, vol. 
30
 (pg. 
253
-
61
)
56
Wilson
NM
Hilmer
SN
March
LM
, et al.  . 
Associations between drug burden index and falls in older people in residential aged care
J Am Geriatr Soc
 , 
2011
, vol. 
59
 (pg. 
875
-
80
)
57
Yeh
Y-C
Liu
C-L
Peng
L-N
Lin
M-H
Chen
L-K
Potential benefits of reducing medication-related anticholinergic burden for demented older adults: a prospective cohort study
Geriatr Gerontol Int
 , 
2013
, vol. 
13
 (pg. 
694
-
700
)
58
Low
LF
Anstey
KJ
Sachdev
P
Use of medications with anticholinergic properties and cognitive function in a young-old community sample
Int J Geriatr Psychiatry
 , 
2009
, vol. 
24
 (pg. 
578
-
84
)
59
Torjesen
I
Anticholinergic effects of common drugs are associated with increased mortality in over 65s
BMJ
 , 
2011
, vol. 
342
 pg. 
d4037
 

Comments

0 Comments