Obesity indices, fat distribution and the risk of ...  · Web viewTable 2: Risk of mortality and incident cardiovascular event according to total anticholinergic burden score (0,

Total anticholinergic burden and risk of mortality and cardiovascular disease over 10 years in 21,636 middle and older aged men and women of EPIC-Norfolk prospective population study

Concise title: Total anti-cholinergic burden, mortality and CVD

Word count: 2,489

Keywords (3-5): Anticholinergic burden; Mortality; Cardiovascular diseases; Epidemiology

Key points (3-5):

1. People with higher total anticholinergic burden (ACB) from medications had increased risk of mortality and cardiovascular events.

2. There was a linear dose response relationship, and an additive effect of combination of drugs with ACB.

3. Future research should examine the relationship between ACB and adverse outcomes and possibly minimize the ACB load.

4. It would be prudent to minimize the ACB load where feasible.

1

1

2

3

45

678

9

10

11121314151617181920

12

Abstract

Background: Studies have raised concerns that medications with anti-cholinergic property

have potential adverse effects on health outcomes.

Objectives: The objective of this study is to examine the prospective relationships between

total anticholinergic burden (ACB) from medications and mortality, and cardiovascular

disease (CVD) in a general population.

Design: Observational study.

Setting: Community cohort.

Subjects: We examined data collected from 21,636 men and women without cancer at the

baseline who participated in a baseline survey 1993-1997 in the European Prospective

Investigation into Cancer (EPIC)-Norfolk. They were followed until 2009/2011.

Methods: We performed Cox-proportional hazards models to determine the associations

between total ACB and the subsequent risk of all-cause mortality and incident CVD during

the follow up.

Results: There were a total of 4,342 people who died and 7,328 had an incident CVD during

the study follow up (total person years= 322,321 years for mortality and 244,119 years for

CVD event). Compared to people with no anticholinergic burden (ACB=0), people with total

ACB ≥3 from medications had HRs of 1∙83(1∙53,2∙20) and 2∙17(1∙87,2∙52) for mortality and

CVD incidence outcomes, respectively, after adjusting for potential confounders. Repeating

the analyses after excluding people with prevalent illnesses, and events occurring within the

first 2 years of follow up, only slightly attenuated the results.

Conclusions: There appear to be a class effect as well as dose-response relationship between

the ACB and both outcomes. Future research should focus on understanding the relationship

between ACB and mortality, and cardiovascular disease and possibly minimizing ACB load

where feasible.

2

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

34

Introduction

The potential adverse effect of medications with anti-cholinergic (antimuscarinic) property is

of particular interest in ageing populations as older people are commonly exposed to these

medications [1]. Previous research however was conducted in long-term care facilities [2,3],

or older people with a specific medical conditions [4].

Recent studies have classified drugs with different degree of anticholinergic cognitive burden

as class 1 (score value 1), 2 (score value 2) and 3 (score value 3) drugs based on their central

effect [5]. Using the same scale, Fox et al did not find deterioration of cognition in people

with a diagnosis of Alzheimer’s Dementia (n=224) [6], but in a larger sample of general older

population (n=12,250) participating in the Medical Research Council Cognitive Function and

Ageing Study (MRC-CFAS) they showed increased cumulative risk of cognitive impairment

and mortality [7].

There is also evidence to suggest the relationship between anticholinergic medication use and

cardiovascular disease risk [8]. However, whether the use of anticholinergic medications in a

general population is associated with increased risk of mortality and incidence of CVD has

not been examined previously.

In this study we used anticholinergic burden (ACB) assessed as described in Fox et al [6]

(Appendix 1 in the supplementary data on the journal website

(http://www.ageing.oxfordjournals.org/) to examine the relationships between total ACB at

the study baseline and all cause mortality and incidence of cardiovascular diseases in a UK

population based study, European Prospective Investigation into Cancer (EPIC)-Norfolk.

3

4647

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

56

Methods

Participants

Participants were men and women aged between 40-79 years from general practice age-sex

registers at the study baseline during 1993-1997 in the EPIC-Norfolk, UK. The detailed

study protocol of EPIC-Norfolk have been described previously [9]. Briefly, all eligible

community dwelling adults from 35 participating general practices were invited to

participate. A total of 25,639 participants (99∙6% White British) attended a baseline health

examination during 1993-1997. They provided written consent to participate in the study and

the Norwich Local Research Ethics Committee approved the study.

Measurements

Details of data collection and measurement methods were described in Appendix 2 (please

see Appendix 1 in the supplementary data on the journal website

http://www.ageing.oxfordjournals.org/). Trained nurses measured weight, height, body mass

index (BMI) and blood pressure and non-fasting venous blood samples. At the baseline

participants completed a detailed health and lifestyle questionnaire which collected

information on participant’s educational status, occupational social class, physical activity,

smoking status, alcohol consumption, prevalent illness and medications. Drugs associated

with anti-cholinergic burden (Appendix 1, http://www.ageing.oxfordjournals.org/) were

identified by searching the database for exact and similar entries for both generic and brand

name drugs. Each medication was assigned to the corresponding anti-cholinergic score and

the total anticholinergic burden (ACB) was calculated using the formula: {[number of class 1

anti-cholinergic drugs] + [the number of class 2 anti-cholinergic drugs x 2] + [the number

4

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

78

class 3 anti-cholinergic drugs x 3]}. Classification of drugs with ACB was class 0 (none),

class 1 (mild), 2 and 3 (severe) [5].

Case ascertainment

All participants were identified for death at the Office of National Statistics. Participants were

also linked to NHS hospital information system and ENCORE (East Norfolk COmmission

Record) for admission episodes. Mortality and incident CVD were identified from the death

certificates (Office of National Statistics) or hospital discharge code ICD 9, 401 – 448 or ICD

10, I10 - I79 for CVD incidence. The follow up methods of EPIC-Norfolk had been

previously validated using incident stroke cases [10].

The follow up time started at baseline for this study (date of study enrolment) and ended at

end of March 2009 for CVD events and end of December 2011 for mortality outcome.

Statistical analysis

Statistical analyses were performed using STATA version 10.0 (Texas, USA) (please see

Appendix 3 in the supplementary data on the journal website

http://www.ageing.oxfordjournals.org / for details). We performed Cox-proportional hazards

models to determine the associations between ACB score groups (ACB score 1 group, ACB

score 2-3 group and ACB score >3 group) and the subsequent risk of all-cause mortality and

incident CVD using the ACB score 0 group as the reference group. Multivariable

adjustments were made to examine how far the associations might be explained by other

known lifestyle, socioeconomic and cardiovascular risk factors.

We then performed stratified analyses to examine the relationships between total ACB and

outcomes by age category (<65 yrs and ≥65 yrs), sex (male and female), social class (manual

5

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118119

120

910

and. non-manual), educational attainment (low and. high), physical activity level (low and

high). To examine the impact of higher total ACB score by every 2 points increase, we

constructed Cox regression models using models A, B, C and D described above. Effect of

ACB class was further examined by creating eight groups of ACB use (none, class 1 drug

alone, class 2 drug alone, class 3 drug alone, class 1+2, class 1+3, class 1+2+3, and class 2+3

users).

As a sensitivity analysis, propensity score matching with nearest neighbour matching was

used to control for potentially confounding factors.

6

121

122

123

124

125

126

127

128

129

1112

Results

Of 25,639 EPIC-Norfolk participants who attended the first health examination, 21,636

(10,135 men and 11,501 women) were eligible to be included in the study, after excluding

participants with any missing values and those with prevalent cancer at the baseline. The

mean follow ups were 14∙9 years (total person years = 322,321 years) for all-cause mortality

and 11∙3 years (total person years = 244,119 years) for incident CVD. During the follow up

there were a total of 4,342 participants who died and 7,328 had incident CVD. The flow

diagram of participants and missing data table is shown in the Appendix 4 and 5 in the

supplementary data on the journal website (http://www.ageing.oxfordjournals.org/).

Table 1 shows the sample characteristics and the crude rates of outcome events according to

the ACB score groups. Significant differences were observed with increasing ACB score

group for all variables aside from age. The participants with the higher ACB score groups (2-

3 or >3) at study baseline were more likely to be older and to be women. People in the higher

ACB score groups were less active, more likely to be on aspirin, or have had a diagnosis of

COPD and asthma, myocardial infarction, stroke and diabetes. There were a substantially

higher proportion of people who smoked (defined as current smoker) in the highest ACB

group. With large sample size, although the significant overall trends were observed between

the ACB score groups, there were few material differences between occupational social class,

educational attainment, level of physical activity, total cholesterol level, and BMI. People

who used medications with anticholinergic activity compared to non-users (ACB ≥1 groups

vs. 0), had a significantly higher level of systolic BP. Higher rates of events for mortality and

cardiovascular disease were observed with higher ACB score group. The overall crude

7

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

1314

mortality rates were 10∙8%, 23∙4%, 27∙8% and 33∙7% for ACB score 0, 1, 2-3 and >3 groups

respectively. The respective crude overall cardiovascular events were 14∙0%, 33∙3%, 40∙1%

and 49∙3% over the entire duration of follow up.

Table 2 presents the Cox-proportional Hazards Ratios and corresponding 95% confidence

intervals (95%CI) for the risk of death and incidence of CVD during the respective study

follow up periods by ACB score group. Consistent results were observed with higher ACB

score groups being associated with a worse outcome for both mortality and CVD incidence.

For both outcomes, higher levels of adjustments were associated with attenuation in risk but

the HRs remained highly significant. Exclusion of people with prevalent conditions, and

exclusion of events occurring within the first two years of follow up did not alter the results.

Appendix 6 shows the adjusted HRs for mortality and incident CVD outcomes in stratified

analyses. In all analyses higher ACB score group was associated with a significantly

increased risk of both mortality and incident CVD. The subgroup analyses demonstrated that

participants with higher ACB score and age less than 65 years lacked overlap between 95%

confidence intervals but there was considerable overlap between the 95% confidence

intervals for each strata of gender, social class, education level and physical activity given the

same total ACB. The adjusted HRs for mortality and incident CVD outcomes after excluding

people with prevalent chronic co-morbidities (asthma, COPD, diabetes, stroke and MI) by

ACB score groups is shown in the Appendix 7 (http://www.ageing.oxfordjournals.org/). In

general similar trends in HRs were observed as those without exclusion of prevalent illnesses.

Appendix 8 A shows the adjusted HRs for selected models as in the table 2 for both mortality

and incident CVD outcome by every 2 points increase in ACB score. The crude event rates

8

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

1516

and data are shown in the Appendix 9 (http://www.ageing.oxfordjournals.org/). In fully

adjusted model (model C), every 2 point increase in ACB was associated with an increase in

29% relative risk of death and an increase in 40% relative risk of incident CVD during

follow-up. Appendix 8 B shows the risk of mortality and incident CVD outcomes with

various combinations of ACB classes. This suggested an ACB class effect with combined use

of higher class ACB drugs associated with a worse outcome. The crude event rates and data

are shown in the Appendix 10 (http://www.ageing.oxfordjournals.org/).

The propensity score matched analyses of the 3 matched cohorts showed similar increased of

risk of death and CVD with ACB score ≥1 groups compared to ACB score 0 group.

(Appendix 11, Appendix 12).

9

180

181

182

183

184

185

186

187

188

189

190

1718

Discussion

We found that people with baseline higher total ACB from medications were at increased risk

of mortality and cardiovascular events compared to those with no or lower total ACB in a UK

general population of middle and older age. There appeared to be a linear dose response

relationship, as well as additive effect of combination of drugs with different anticholinergic

burden. While participants with higher anticholinergic burden were older and more likely to

have prior cardiovascular co-morbidities, similar results are seen even after adjustment for

these variables and other potential confounders as well as repeating the analyses after

excluding those with major prevalent illnesses.

The existing literature on anticholinergic drugs and mortality shows inconsistent results but

they have been conducted on high-risk populations such as participants from elderly

residential or long term care facilities [2,3], geriatric wards and nursing homes [11], among

the elderly hospitalized patients with hip fracture [12,13], and elderly patients with

cardiovascular disease [4]. There are only a few studies which have been conducted among

the community dwelling older adults [4,7,15]. In general, the results of these studies are

inconsistent. Cohorts of hospitalized participants with hip fractures [12, 13] and community

dwelling and institutionalized participants [7] showed that a higher anticholinergic activity

was associated with increased mortality. However other studies of long term or residential

care facility participants [2,3], older community dwellers [4, 14] and geriatric wards or

nursing homes [11] failed to demonstrate this relationship.

A few potential mechanisms may explain why anticholinergic medications may increase

mortality and incidence of CVD. A recent report suggests that anticholinergic medications

10

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

1920

are pro-arrhythmic and pro-ischaemic [15]. It has been suggested the inhibition of

parasympathetic control of the heart may be associated with increased hemodynamic lability,

cardiac ischaemia, and cardiac dysrhythmias in response to cardiac ischaemia [16]. In

addition, studies have found that certain anticholinergic drugs such as imipramine and

clozapine decrease heart rate variability [17] and this may contribute to adverse

cardiovascular events. Another plausible mechanism is via immuno-modulation as the

cholinergic system plays an important role in regulating immune response. Nicotinic

receptor activation causes autonomic and vagal systems to inhibit adaptive and innate

immune response [18], and it is possible that inhibition of these systems may lead to an

inflammatory response and subsequent increased risk of mortality and CVD in people who

already possess risk factors.

Our study has several strengths. The data were prospectively collected which reduces recall

bias. The sample size was large enough to capture a sufficient number of participants with

high anticholinergic burden as well as allow us to test the differences in risk between

individuals with higher and lower degrees of anticholinergic burden. Our sample population

had wide age spectrum, social and demographic variation and we were able to take into

account co-morbidities, other lifestyle factors.

Our study has limitations. Due to the requirement to attend a health examination, the response

rate at the study baseline (1993-1997) was modest at ~ 40% in EPIC-Norfolk introducing a

healthy responder effect from the outset. Nevertheless, baseline characteristics of the study

population are similar to other UK population samples except with a slightly lower

prevalence of smokers [9]. Moreover, this should not affect the associations observed within

the study participants; if anything, truncation of the distribution is likely to reduce power for

11

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

2122

any associations. In addition, ~2600 participants were excluded due to missing data and this

could potentially introduce bias to the regression coefficients. The current analysis was not

part of the pre-registered analysis plan of the EPIC-Norfolk study and this may have

implications on generalisability of the findings as the analysis to some extent is contingent on

the data. There were only single measurements of covariates such as cholesterol, blood

pressure etc. The blood sample taken was non-fasting sample and therefore less standardized

for some of the parameters (e.g. cholesterol level). Nevertheless, random measurement error

is likely only to attenuate any associations observed. Ascertainment of drug exposure was

based on a baseline self-report. We do not know whether participants continued to take their

medication over the follow up period as we were unable to measure the pattern and the

duration of drug usage over time and this could have led to misclassification. Although we

were able to calculate total ACB, we were not able to identify particular drugs which are

potentially linked to adverse outcomes. The validity of the models of analysis is unknown but

the results appear to be robust to different parameterisations of ACB.

A major limitation in assessing the association between medications and health outcomes is

the difficulty in evaluating the possible effect of confounding and reverse causality.

Nevertheless, the associations remained after adjustment for known risk factors for

cardiovascular disease and mortality and even after excluding individuals with known

prevalent illnesses and those with events in the first few years who may have had preclinical

conditions. Though we cannot exclude residual confounding, the limited data from

randomized controlled trials of anticholinergics are also consistent with a causal relationship

[8,19].

12

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

2324

In summary, our study indicates a potential negative impact of medications with

anticholinergic properties on mortality and CVD incidence in middle and older age

population. This has implications in clinical practice as anticholinergic drugs are commonly

prescribed, especially among the older people with long term conditions. While the

relationships were prospective, it remains unclear whether there was a causal relationship.

Nonetheless, the potential benefits of drug use must be weighed against adverse effects so it

is recommended that patients should undergo regular medication review and discontinuation

of unnecessary anticholinergic drugs should be considered. Future studies should explore

whether systematic attempts to reduce the anticholinergic burden may improve health

outcomes.

13

266

267

268

269

270

271

272

273

274

275

2526

Contributors

KTK and NJW are the Principal Investigators of EPIC-Norfolk cohort. PKM and CF

conceptualized and designed the study. RNL was responsible for data management and CSK

analyzed the data. PKM and CSK drafted the manuscript. All authors contributed to the

study design and writing of the paper. PKM is the guarantor.

Acknowledgements

The authors would like to thank the participants of the EPIC-Norfolk cohort and the funders.

The EPIC-Norfolk study was supported by grants from the Medical Research Council and

Cancer Research UK. Funders had no role in study design or interpretation of the findings.

Disclosures

The authors have no conflicts of interest to declare.

14

276277

278

279

280

281

282

283

284

285

286

287

288

289

2728

References

1. Chew ML, Mulsant BH, Pollock BG, et al. Anticholinergic activity of 107 medications commonly used by older adults. J Am Geriatr Soc 2008;56:1333-1341.

2. Kumpula E-K, 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;51:256.

3. 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;29:157-65.

4. Uusvaara J, Pitkala KH, Kautiainen H, Tilvis RS, Strandberg TE. Association of anticholinergic drugs with hospitalization and mortality among older cardiovascular patients. Drugs Aging 2011;28:131-138.

5. Boustani M, Campbell N, Munger S, Maidment ID, Fox C. The Impact of Anticholinergics on the Aging Brain: A Review and Practical Application. Aging Health 2008;4:311-20.

6. Fox C, Livingston G, Maidment ID, et al. The impact of anticholinergic burden in Alzheimer's dementia-the LASER-AD study. Age Ageing 2011;40:730-5.

7. Fox C, Richardson, K Maidment I, et al on behalf of the Medical Research Council Cognitive Function and Ageing Study (MRC CFAS). Anticholinergic Medication Use and Cognitive Impairment in the Older Population: The Medical Research Council Cognitive Function and Ageing Study. J Am Geriatr Soc 2011;59:1477-83.

8. Singh S, Loke YK, Furberg CD. Inhaled anticholinergics and risk of major adverse cardiovascular events in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis. JAMA 2008;300:1439-1450.

9. Day N, Oakes S, Luben R, et al. EPIC-Norfolk: study design and characteristics of the cohort. European Prospective Investigation of Cancer. Br J Cancer 1999;80:95-103.

10. Sinha S, Myint PK, Luben RN, Khaw KT. Accuracy of death certification and hospital record linkage for identification of incident stroke. BMC Med Res Methodol 2008;8:74.

11. Luukkanen MJ, Uusvaara J, Laurila JV, et al. Anticholinergic drugs and their effects on delirium and mortality in the elderly. Dement Geriatr Cogn Disord Extra 2011;1:43-50.

12. Mangoni A, van Munster BC, Woodman RJ, de Rooij SE. Measures of anticholinergic drug exposure, serum anticholinergic activity, and all-cause post discharge mortality in older hospitalized patients with hip fractures. Am J Geriatr Psychiatry 2013;21:785-93.

15

290

291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338

2930

13. Panula J, Puustinen J, Jaatinen P, Vahlberg T, Aarnio P, Kivela SL. Effects of potent anticholinergics, sedatives and antipsychotics on postoperative mortality in elderly patients with hip fracture: a retrospective, population-based study. Drug Aging 2009;26:963-971.

14. Boudreau DM, Yu O, Gray SL, Raebel MA, Johnson J, Larson EB. Concomitant use of cholinesterase inhibitors and anticholinergics: prevalence and outcomes. JAGS 2011;59:2069-2076.

15. Singh S, Loke YK, Enright P, Furberg CD. Pro-arrhythmic and pro-ischaemic effects of inhaled anticholinergic medications. Thorax 2013;68:114-116.

16. Parlow, JL, van Vlymen JM, Odell MJ. The Duration of Impairment of Autonomic Control After Anticholinergic Drug Administration in Humans. Anesth Analg 1997;84:155-9

17. O'Brien P, Oyebode F. Psychotropic medication and the heart. APT 2003, 9:414-423.

18. Razani-Boroujerdi S, Behl M, Hahn FF, Pena-Philippides JC, Hutt J, Sopori ML. Role of muscarinic receptors in the regulation of immune and inflammatory responses. J Neuroimmunol 2008;194:83–88.

19. Daumit GL, Goff DC, Meyer JM, et al. Antipsychotic effects on estimated 10-year coronary heart disease risk in the CATIE schizophrenia study. Schizophr Res 2008;105:175-87.

16

339340341342343344345346347348349350351352353354355356357358359360361362363364365

3132

List of Tables and Figures

Table 1: Sample characteristics of 21,636 men and women of the EPIC-Norfolk (1993/1997-2009/2011) according to the total anticholinergic burden (ACB) score

Table 2: Risk of mortality and incident cardiovascular event according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow up (1993-2011) in EPIC-Norfolk

Appendix 1: Anticholinergic Cognitive Burden scoring of drugs

Appendix 2: Methods of data collection

Appendix 3: Methods of data analysis

Appendix 4: Flow diagram of participants

Appendix 5: Missing data table Appendix 6: Subgroup multivariable adjusted analysis of the risk of mortality and cardiovascular events according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow-up (1993-2011)

Appendix 7: Subgroup multivariable analysis of the risk of mortality and cardiovascular events according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow-up (1993-2011) after excluding prevalent illnesses

Appendix 8: Hazard ratios and corresponding 95% CI for risk of mortality and cardiovascular disease incidence during follow up (1993/97-2009/11) in EPIC-Norfolk by every two points increase in total anticholinergic burden score according to various models of adjustment and by combination of class of drugs which contribute to total anticholinergic burden score

* Caption for Appendix 8 A): Model A: adjusted for age and sex; Model B: adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure and body mass index; Model C: adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level, BMI, prevalent conditions including asthma, COPD, diabetes, stroke and myocardial infarction; Model D: adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level, BMI and excluded participants with asthma, COPD, diabetes, stroke and myocardial infarction. Numbers of events were 4,342/21,636 for models A, B and C and 3,029/17,242 for model D for mortality outcome and 7,328/21,636 for models A, B and C and 5,270/17,242 for model D for incident CVD outcome. The tabular form of this figure is presented in the appendix as Appendix Table 9.

* Caption for Appendix 8 B): Adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure,

17

366

367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414

3334

cholesterol level, body mass index, prevalent medical conditions including asthma, COPD, diabetes, stroke and myocardial infarction. Numbers of events were 4,342/21,636 for mortality outcome and 7,328/21,636 for incident CVD. The tabular form of this figure is presented in the appendix as Appendix Table 10.

Appendix 9: Hazard ratios (HR) and corresponding 95% confidence intervals (95%CI) for risk of mortality and cardiovascular events during follow up (1993/97-2009/11) in EPIC-Norfolk by every two points increase in anticholinergic burden score according to various models of adjustment

Appendix 10: Hazard ratios (HR) and corresponding 95% confidence intervals (95%CI) for risk of mortality and cardiovascular events during follow up (1993/97-2009/11) in EPIC-Norfolk by combinations of drugs with contribute to anticholinergic burden score

Appendix 11: Sample characteristics of propensity matched men and women of the EPIC-Norfolk (1993/1997-2009/2011) according to the total anticholinergic burden (ACB) score

Appendix 12: Propensity matched risk of mortality and incident cardiovascular event according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow up (1993-2011) in EPIC-Norfolk

18

415416417418419420421422423424425426427428429430431432433434435

3536

Table 1: Sample characteristics of 21,636 men and women of the EPIC-Norfolk (1993/1997-2009/2011) according to the total anticholinergic burden (ACB) score

ACB score 0 group(n=17,317)

ACB score 1 group(n=2,704)

ACB score 2-3 group (n=1,324)

ACB score >3 group (n=291)

P

Age (years) 57.9 (9.1) 62.9 (8.8) 62.2 (9.2) 63.1 (8.9) 0.07Sex

MenWomen

8,068 (47)9,249 (53)

1,348 (50)1,356 (50)

593 (45)731 (55)

126 (43)165 (57)

0.003

Social classProfessional

ManagerSkilled non-manual

Skilled manualSemi-skilledNon-skilled

1,268 (7)6,475 (37)2,806 (16)3,978 (23)2,239 (13)551 (3)

164 (6)927 (34)464 (17)637 (24)403 (15)109 (4)

68 (5)446 (34)234 (18)321 (24)193 (15)62 (5)

19 (7)80 (27)58 (20)78 (27)43 (15)13 (4)

<0.0001*

Smoking Current-smoker

Ex-smokerNever smoker

2,038 (12)7,153 (41)8,126 (47)

256 (9)1,324 (49)1,124 (42)

155 (12)626 (47)543 (41)

51 (18)122 (42)118 (41)

<0.0001*

Alcohol use (units/week)

7.5 (9.6) 6.8 (9.6) 6.0 (8.3) 5.2 (7.9) <0.0001

Education levelNo qualification

0-LevelA-Level

Higher degree

5,827 (34)1,886 (11)7,202 (42)2,402 (14)

1,152 (43)255 (9)1039 (38)258 (10)

598 (45)111 (8)504 (38)113 (9)

132 (45)25 (9)108 (37)26 (9)

<0.0001*

Physical activity Inactive

Moderately inactiveModerately active

Active

4,608 (27)5,017 (29)4,187 (24)3,505 (20)

1,037 (38)760 (28)520 (19)387 (14)

552 (42)361 (27)234 (18)177 (13)

142 (49)86 (30)40 (14)23 (8)

<0.0001*

Cholesterol (mmol/L)

6.1 (1.1) 6.3 (1.2) 6.3 (1.2) 6.4 (1.1) <0.0001

Systolic BP (mmHg) 134 (18) 140 (19) 137 (19) 138 (19) 0.001BMI (kg/m2) 26.1 (3.7) 27.0 (4.2) 26.9 (4.1) 27.2 (4.4) <0.0001COPD 1,425 (8) 346 (13) 184 (14) 40 (14) <0.0001Asthma 896 (5) 630 (23) 204 (15) 58 (20) <0.0001Previous MI 253 (1) 222 (8) 153 (12) 31 (11) <0.0001Previous stroke 131 (1) 78 (3) 52 (4) 13 (4) <0.0001Diabetes 286 (2) 102 (4) 59 (4) 17 (6) <0.0001Aspirin use 1,234 (7) 479 (18) 268 (20) 65 (22) <0.0001New CVD events 4,939 (29) 1,459 (54) 751 (57) 179 (62) <0.0001Deaths 2,833 (16) 887 (33) 498 (38) 124 (43) <0.0001Values presented are mean (sd) for continuous and number (%) for categorical data. *overall P value. BP=blood pressure, BMI = body mass index, COPD= chronic obstructive pulmonary disease; MI=myocardial infarction, CVD= cardiovascular diseases. Total anticholinergic burden (ACB) calculated as a score which is the sum of the [number of class 1 anticholinergic drugs, the number of class 2 anticholinergic drugs x2 and the number class 3 anticholinergic drugs x3]. Classification of drugs with ACB class 1, 2 and 3 based on criteria of Anticholinergic Cognitive Burden Scale (Boustani MA, et al 2008;4:311–320).

19

436437

438439440441442443444

3738

Table 2: Risk of mortality and incident cardiovascular event according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow up (1993-2011) in EPIC-Norfolk

Models Mortality

(Events (n)/Total N=4,342/21,636)ACB score 0 group

ACB score 1 group

ACB score 2-3 group

ACB score >3 group

p-value

A 1.00 1.42 (1.32-1.54) 1.90 (1.73-2.10) 2.20 (1.84-2.64) <0.00001

B 1.00 1.39 (1.29-1.50) 1.85 (1.68-2.04) 2.07 (1.73-2.48) <0.00001

C 1.00 1.28 (1.18-1.39) 1.65 (1.49-1.82) 1.83 (1.53-2.20) <0.00001

D* 1.00 1.34 (1.22-1.48) 1.58 (1.38-1.80) 2.08 (1.63-2.66) <0.00001

E# 1.00 1.25 (1.16-1.36) 1.63 (1.47-1.81) 1.68 (1.38-2.04) <0.00001

F 1.00 1.27(1.18-1.38) 1.63(1.48-1.81) 1.82(1.51-2.18) <0.00001

Models Cardiovascular disease incidence

(Events (n)/Total N=7,328/21,636)ACB score 0 group

ACB score 1 group

ACB score 2-3 group

ACB score >3 group

p-value

A 1.00 1.77 (1.66-1.87) 2.18 (2.02-2.36) 2.48 (2.14-2.88) <0.00001

B 1.00 1.65 (1.57-1.75) 2.09 (1.93-2.26) 2.40 (2.06-2.78) <0.00001

C 1.00 1.51 (1.42-1.61) 1.86 (1.72-2.02) 2.17 (1.87-2.52) <0.00001

D* 1.00 1.70 (1.58-1.83) 1.82 (1.64-2.02) 2.26 (1.84-2.77) <0.00001

E# 1.00 1.50 (1.41-1.60) 1.85 (1.71-2.01) 2.05 (1.75-2.40) <0.00001

F 1.00 1.48 (1.39-1.57) 1.81(1.68-1.96) 2.10(1.80-2.44) <0.00001

ACB = Anticholinergic burden score. Model A: adjusted for age and sex. Model B: Model A plus smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level and body mass index. Model C: Model B plus prevalent conditions asthma, COPD, diabetes, stroke and myocardial infarction. Model D: as in Model B excluding people with prevalent asthma, COPD, diabetes, stroke and myocardial infarction. Model E: as in Model C excluding all events occurring within first two years of follow up. Model F: Model C plus aspirin use. *Model D=n/N=3,029/17,242 for mortality analysis, n/N=5,270/17,242 for CV events analysis; #Model E= n/N=4,141/21,435 for mortality analysis, n/N=7,208/21,435 for CV events analysis.

20

445446447

448449450451452453454455456457458459460461

3940

Appendix 1: Anticholinergic Cognitive Burden scoring of drugs

Score 1 Score 2 Score 3Alimemazine Amantadine Amitriptyline

Alverine Belladone alkaloids AmoxapineAlprazolam Carbamazepine Atropine

Atenolol Cyclobenzaprine BenztropineBrompheniramine maleate Cyproheptadine BrompheniramineBupropion hydrochloride Empracet Carbinoxamine

Captopril Loxapine ChlorpheniramineChlorthalidone Meperidine Chlorpromazine

Cimetidine hydrochloride Methotrimeprazine ClemastineRanitidine Molindone Clomipramine

Clorazepate Oxcarbazepine ClozapineCodeine Pethidine hydrochloride Darifenacin

Colchicine Pimozide DesipramineCoumadin DicyclomineDiazepam DimenhydrinateDigoxin Diphenhydramine

Dipyridamole DoxepinDisopyramide phosphate Flavoxate

Fentanyl HydroxyzineFurosemide Hyoscyamine

Fluvoxamine ImipramineHaloperidol MeclizineHydralazine Nortriptyline

Hydrocortisone OlanzapineIsosorbide Orphenadrine

Loperamide OxybutyninMetoprolol ParoxetineMorphine PerphenazineNifedipine ProcyclidinePrednisone PromazineQuinidine Promethazine

Risperidone PropenthelineTheophylline Pyrilamine

Trazodone QuetiapineTriamterene Scopolamine

ThioridazineTolterodine

TrifluoperazineTrihexyphenidyTrimipramine

21

462463464

41

* Adapted from: Boustani MA, Campbell NL, Munger S et al. Impact of anticholinergics on the aging brain: A review and practical application. Aging Health 2008;4:311–320.Appendix 2: Methods of data collection

Measurements

Trained nurses examined individuals at clinic visit. Weight was measured with

participants wearing light clothing without shoes. Height was measured up to the

nearest 0∙1 cm using a stadiometer with shoes removed. Body mass index (BMI) was

calculated as weight (kilogramme) divided by height in metres squared (m2). Blood

pressure (BP) was measured with an Accutorr monitor (Datascope, Huntingdon, UK)

after the participant had been seated for 5 min. We used the mean of two BP

measurements for analysis. Non-fasting venous blood samples were taken into plain

and citrate bottles. We measured serum total cholesterol with the RA 1000 (Bayer

Diagnostics, Basingstoke, UK).

At the baseline participants completed a detailed health and lifestyle questionnaire.

Participant’s educational status, occupational social class, and physical activity were

obtained from the baseline health and lifestyle questionnaire. Educational status was

recorded as no qualification, O- level, A-level, degree or higher qualification. Social

class was classified according to the Registrar General’s occupation-based

classification scheme. A four-level physical activity index was derived from the

validated EPIC short physical activity questionnaire designed to assess combined

work and leisure activity. For stratified analyses, social class was re-categorised into

manual (III-manual, IV and V) and non manual (III-non-manual, II and I), educational

attainment was re-categorised as low educational attainment (no or O level) and high

educational attainment (at least A level) and physical activity was re-categorised as

22

465466467468469470

471

472

473

474

475

476

477

478

479

480

481

482

483

484

485

486

487

488

489

490

491

42

high (active and moderately active) and low (inactive and moderately inactive)

physical activity categories.

Smoking status was categorised as current smoker, ex-smoker and those who have

never-smoked. “Current smokers” were defined as those who answered “yes” to the

question “Do you smoke cigarettes now?”. “Never smokers” were defined as those

who answered “no” to the question “Have you ever smoked as much as one cigarette

a day for as long as a year?” All others were classed as “former smokers”. Average

alcohol consumption (units/week) was derived from a food frequency questionnaire

(FFQ) completed at the baseline. Prevalent illnesses were determined by a positive

response to the question “Has a doctor ever told you that you have any of the

following?” followed by a list of options including asthma, COPD, cancer, stroke,

heart attack, and diabetes.

Aspirin, steroid tablets or injections and diuretics use was ascertained by a question

“Have you taken (aspirin, steroid tablets or injections and diuretics) continuously for

three months or more?”. Other medications were identified by participant’s response

to the question “In the last week have you taken any drugs or medicines either

prescribed by your doctors or bought from the chemist? If YES, please name them.”

The medication name or brand, dose and frequency of administration were recorded

and each medication was coded exactly as written in the baseline survey into a

database. Drugs associated with anti-cholinergic burden (Appendix Table 1) were

identified by searching the database for exact and similar entries for both generic and

brand name drugs. Each medication was assigned to the corresponding anti-

cholinergic score and the total anticholinergic burden (ACB) was calculated using the

23

492

493

494

495

496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

512

513

514

515

516

43

formula: {[number of class 1 anti-cholinergic drugs] + [the number of class 2 anti-

cholinergic drugs x 2] + [the number class 3 anti-cholinergic drugs x 3]}.

The development of the anti-cholinergic burden (ACB) scale used in this study has

been previously reported. Classification of drugs with ACB was class 0 (none), class

1 (mild), 2 and 3 (severe). Examples of drugs with include atenolol, ranitidine,

codeine (class 1), amantadine, carbamazepine, pethidine (class 2) and amitriptyline,

oxybutynin, olanzapine (class 3). The score’s predictive validity in cognitive decline

has been shown in three large scale studies and a score of 2 or more was associated

with increased mortality in an older population.

Case ascertainment

All participants were identified for death at the Office of National Statistics.

Participants were also linked to NHS hospital information system so that admission

anywhere in the UK was notified to EPIC-Norfolk. They were also linked to

ENCORE (East Norfolk COmmission Record) for admission episodes. Mortality and

incident CVD were identified from the death certificates (Office of National

Statistics) or hospital discharge code ICD 9, 401 – 448 or ICD 10, I10 - I79 for CVD

incidence. The follow up methods of EPIC-Norfolk had been previously validated

using incident stroke cases.

The follow up time started at baseline for this study (date of study enrolment) and

ended at end of March 2009 for CVD events and end of December 2011 for mortality

outcome.

24

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

44

Appendix 3: Methods of data analysis

Statistical analyses were carried out using STATA version 10.0 (Texas, USA). We

performed Cox-proportional hazards models to determine the associations between

total ACB and the subsequent risk of all-cause mortality and incident CVD using the

ACB score 0 group as the reference group

Multivariable adjustments were made to examine how far the associations might be

explained by other known lifestyle, socioeconomic and cardiovascular risk factors.

We adjusted for age and sex in model A and age, sex, lifestyle behaviours (smoking

status, alcohol consumption and physical activity), social class, education level,

systolic blood pressure, serum cholesterol level and BMI in model B. To account for

illness driving the higher ACB score as well as contributing as confounder for

outcomes examined, prevalent medical conditions including asthma, COPD, diabetes,

stroke and myocardial infarction are additionally included in the model C. Further

sensitivity analyses were conducted. Model D was constructed as in model B after

excluding people with prevalent asthma, COPD, diabetes, stroke and myocardial

infarction (MI). Model E excluded all events occurring within the first two years of

follow up for both outcomes and adjusted as in model C. Model F was constructed as

in model C and additionally adjusted for aspirin use.

We then performed stratified analyses to examine the relationships between total ACB

and outcomes by age category (<65 yrs and ≥65 yrs), sex (male vs. female), social

class (manual vs. non-manual), educational attainment (low vs. high), physical

activity level (low vs. high). In all analyses adjustments were made for age, smoking

status, alcohol consumption, systolic blood pressure, cholesterol level, BMI, prevalent

conditions including asthma, COPD, diabetes, stroke and myocardial infarction. In

25

542

543

544

545

546

547

548

549

550

551

552

553

554

555

556

557

558

559

560

561562

563

564

565

566

567

45

stratified analyses we also adjusted for other factors e.g. in age stratified analysis also

adjusted for sex, social class, education level and physical activity level. The analyses

were repeated after excluding those who reported prevalent illnesses at the baseline

including asthma, COPD, diabetes, stroke and MI.

To examine the impact of higher total ACB by every 2 points increase, we constructed

Cox regression models using models A, B, C and D described above. Effect of ACB

class was further examined by creating eight groups of ACB use (none, class 1 drug

alone, class 2 drug alone, class 3 drug alone, class 1+2, class 1+3, class 1+2+3, and

class 2+3 users). The hazards ratios were calculated using the non-ACB users as the

reference group for both mortality and CVD end points using fully adjusted model

(Model C).

As a sensitivity analysis, propensity score matching with nearest neighbour matching

was used to control for potentially confounding factors. This was used to account for

differences in clinical characteristics based on anticholinergic burden score. A

propensity score was estimated for anticholinergic burden score considering the

following variables: age, sex, social class, smoking status, alcohol consumption,

education level, physical activity, cholesterol, systolic blood pressure, body mass

index, asthma, previous myocardial infarction, previous stroke and previous diabetes.

A 1-to-1 matched cohort was generated by using Mahalanobis distance matching with

the propensity score as the distance matrix. Three groups were matched: ACB score 1

group vs. ACB score 0 group, ACB score 2-3 group vs. ACB score 0 group, and ACB

score >3 group vs. ACB score 0 group. The balance achieved by matching was

assessed by descriptive statistics for the variables used to generate the propensity

26

568

569

570

571

572

573

574

575

576

577

578

579

580581

582

583

584

585

586

587

588

589

590

591

592

46

score. We estimated adjusted odds ratios and Cox proportional hazards for matched

groups.

27

593

594

47

Appendix 4: Flow diagram of participants

28

30,445 participated and complete baseline questionnaire during 1993-1997.

21,636 participants included in the analysis.

4,806 participants excluded from analysis because they did not attend the first health check

25,639 participants attended first health check.

4,003 participants excluded from analysis because of prevalent Cancer and missing data (see table below).

595596597598599600601602603604605606607608609610611612613614615

48

Appendix 5: Missing data table

Variable Missing dataDid not attend 1st health check 4,806Cancer at baseline 1,395Social class 570Smoking 220Alcohol use 268Education level 18Physical activity 1Cholesterol level 1,768Systolic blood pressure 60COPD 41Asthma 37

29

616

619

49

Appendix 6: Subgroup multivariable adjusted analysis of the risk of mortality and cardiovascular events according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow-up (1993-2011)

Mortality Cardiovascular disease incidence

ACB score 0 group

ACB score 1 group

ACB score 2-3 group

ACB score >3 group

P-for trend ACB score 0 group

ACB score 1 group

ACB score 2-3 group

ACB score >3 group

P for trend

Age<65

≥65

1.00(n/N=1,420/15,151)

1.00(n/N=2,922/6,504)

1.51 (1.30-1.75)

1.20 (1.10-1.32)

1.58 (1.30-1.92)

1.65 (1.47-1.85)

2.93 (2.16-3.99)

1.48 (1.18-1.86)

<0.0001

<0.0001

1.00(n/N=3,555/1,5132)

1.00(n/N=3,773/6,504)

1.66 (1.51-1.83)

1.40 (1.30-1.52)

1.85 (1.63-2.09)

1.86 (1.67-2.06)

2.88 (2.32-3.57)

1.68 (1.36-2.08)

<0.0001

<0.0001

SexMen

Women

1.00(n/N=2,521/1,0135)

1.00(n/N=1,821/11,501)

1.32(1.19-1.46)

1.21 (1.07-1.37)

1.56 (1.36-1.79)

1.74 (1.50-2.02)

2.05 (1.61-2.62)

1.61 (1.22-2.13)

<0.0001

<0.0001

1.00 (n/N=3,960/10,135)

1.00 (n/N =3,368/11,501)

1.56 (1.44-1.70)

1.44 (1.31-1.58)

1.81 (1.61-2.02)

1.93 (1.72-2.16)

2.60 (2.10-3.22)

1.83 (1.48-2.27)

<0.0001

<0.0001

Social classNon-manual

Manual

1.00(n/N=2,520/13,009)

1.00(n/N =1,822/8,627)

1.27 (1.15-1.41)

1.30 (1.15-1.47)

1.67 (1.46-1.91)

1.63 (1.41-1.89)

2.14 (1.68-2.73)

1.55 (1.18-2.04)

<0.0001

<0.0001

1.00 (n/N =4,134/13,009)

1.00 (n/N =3,194/8,627)

1.52 (1.40-1.65)

1.50 (1.37-1.65)

1.90 (1.70-2.11)

1.81 (1.61-2.04)

2.23 (1.81-2.76)

2.06 (1.65-2.56)

<0.0001

<0.0001

Education levelLow

High

1.00(n/N =2,359/9,986)

1.00(n/N =1,983/11,650)

1.21 (1.09-1.35)

1.38 (1.23-1.56)

1.64 (1.44-1.86)

1.63 (1.39-1.91)

1.58 (1.23-2.04)

2.15 (1.65-2.81)

<0.0001

<0.0001

1.00 (n/N =3,882/9,986)

1.00 (n/N =3,446/11,650)

1.49 (1.37-1.62)

1.55 (1.41-1.70)

1.88 (1.69-2.09)

1.84 (1.63-2.08)

1.94 (1.57-2.39)

2.49 (2.00-3.10)

<0.0001

<0.0001

Physical activity levelLow 1.00 1.33 1.69 1.93 <0.0001 1.00 1.51 1.89 2.10 <0.0001

620621622

50

High

(n/N =3,046/12,563)

1.00(n/N =1,296/9,073)

(1.21-1.45)

1.21(1.03-1.41)

(1.51-1.89)

1.56 (1.27-1.92)

(1.58-2.36)

1.56 (1.01-2.41)

<0.0001

(n/N =4,757/12,563)

1.00 (n/N =2,571/9,073)

(1.40-1.62)

1.52(1.36-1.70)

(1.72-2.08)

1.81 (1.55-2.10)

(1.77-2.49)

2.49 (1.79-3.45)

<0.0001

In all analyses adjustments were made for age, smoking status, alcohol consumption, systolic blood pressure, cholesterol level, body mass index, prevalent conditions including asthma, COPD, diabetes, stroke and myocardial infarction. Lower occupational social class was defined as skilled manual worker, semi-skilled worker and non-skilled worker. Upper occupational social class was defined as professionals, managerial or skilled non-manual worker. Low education level was defined as no qualifications or O level. High education level was defined as A level or a higher degree. Low physical activity level was defined as inactive or moderately inactive. High physical activity level was defined as moderately active or active

623624625626627628629

51

Appendix 7: Subgroup multivariable analysis of the risk of mortality and cardiovascular events according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow-up (1993-2011) after excluding prevalent illnesses

Mortality Cardiovascular disease incidence

ACB score 0 group

ACB score 1 group

ACB score2-3 group

ACB score >3 group

P-for trend ACB score 0 group

ACB score 1 group

ACB score 2-3 group

ACB score >3 group

P for trend

Age<65

≥65

1.00(n/N=1,037/12,347)

1.00(n/N =1,992/4,895)

1.68 (1.40-2.02)

1.25 (1.11-1.40)

1.42 (1.09-1.85)

1.62 (1.39-1.90)

2.85 (1.86-4.37)

1.82 (1.34-2.45)

<0.0001

<0.0001

1.00(n/N=2,611/12,347)

1.00(n/N =2,659/4,895)

1.90 (1.69-2.14)

1.57 (1.42-1.73)

1.82 (1.55-2.12)

1.81 (1.58-2.08)

2.93 (2.19-3.92)

1.80 (1.35-2.40)

<0.0001

<0.0001

SexMen

Women

1.00 (n/N =1,699/7,903)

1.00(n/N =1,330/9,339)

1.38 (1.21-1.57)

1.30 (1.12-1.51)

1.46 (1.20-1.77)

1.68(1.40-2.03)

2.41 (1.71-3.39)

1.83 (1.28-2.61)

<0.0001

<0.0001

1.00 (n/N =2,733/7,903)

1.00 (n/N =2,537/9,339)

1.78 (1.60-1.97)

1.62 (1.46-1.81)

1.89 (1.62-2.20)

1.77 (1.54-2.04)

2.78 (2.05-3.75)

1.92 (1.45-2.54)

<0.0001

<0.0001

Social classNon-manual

Manual

1.00(n/N =1,753/10,369)

1.00(n/N =1,276/6,873)

1.37 (1.21-1.56)

1.30 (1.11-1.51)

1.53 (1.28-1.83)

1.65 (1.36-2.01)

2.47 (1.74-3.51)

1.78 (1.26-2.50)

<0.0001

<0.0001

1.00(n/N =2,934/10,369)

1.00(n/N =2,336/6,873)

1.69 (1.53-1.87)

1.70 (1.53-1.91)

1.93 (1.68-2.22)

1.69 (1.45-1.98)

2.44 (1.81-3.31)

2.11 (1.59-2.78)

<0.0001

<0.0001

Education levelLow

High

1.00(n/N =1,643/7,971)

1.00 (n/N =1,386/9,271)

1.23 (1.08-1.41)

1.50 (1.30-1.74)

1.55 (1.31-1.84)

1.57 (1.27-1.95)

1.85 (1.34-2.56)

2.47 (1.70-3.59)

<0.0001

<0.0001

1.00(n/N =2,803/7,971)

1.00(n/N =2,467/9,271)

1.62 (1.47-1.79)

1.82 (1.63-2.04)

1.83 (1.60-2.10)

1.79 (1.53-2.11)

1.97 (1.50-2.59)

2.78 (2.04-3.80)

<0.0001

<0.0001

Physical activity level

630631632633

52

Low

High

1.00(n/N =2,070/9,791)

1.00(n/N =959/7451)

1.40(1.25-1.57)

1.26 (1.04-1.53)

1.63 (1.40-1.90)

1.43 (1.08-1.90)

2.28 (1.73-2.99)

1.69 (0.97-2.96)

<0.0001

<0.0001

1.00(n/N =3,339/9,791)

1.00(n/N =1,931/7,451)

1.66 (1.52-1.82)

1.82(1.59-2.07)

1.83 (1.62-2.07)

1.79 (1.48-2.16)

2.21 (1.75-2.79)

2.56 (1.66-3.95)

<0.0001

<0.0001

In all analyses adjustments were made for age, smoking status, alcohol consumption, systolic blood pressure, cholesterol level, body mass index, prevalent conditions including asthma, COPD, diabetes, stroke and myocardial infarction and prevalent illnesses were excluded. In the stratified analyses, sex, social class, education level and physical activity were included in the models apart from the variable used for stratification. Lower occupational social class was defined as skilled manual worker, semi-skilled worker and non-skilled worker. Upper occupational social class was defined as professionals, managerial or skilled non-manual worker. Low education level was defined as no qualifications or O level. High education level was defined as A level or a higher degree. Low physical activity level was defined as inactive or moderately inactive. High physical activity level was defined as moderately active or active.

634635636637638639640

53

Appendix 8: Hazard ratios and corresponding 95% CI for risk of mortality and cardiovascular disease incidence during follow up (1993/97-2009/11) in EPIC-Norfolk by every two points increase in total anticholinergic burden score according to various models of adjustment and by combination of class of drugs which contribute to total anticholinergic burden score

641642643

54

644

55

Appendix 9: Hazard ratios (HR) and corresponding 95% confidence intervals (95%CI) for risk of mortality and cardiovascular events during follow up (1993/97-2009/11) in EPIC-Norfolk by every two points increase in anticholinergic burden score according to various models of adjustment

Models of

adjustment

Mortality Cardiovascular disease incidence

n events/ total

N

HR (95% CI) n events/ total

N

HR (95% CI)

Model A 4,342/21,636 1.40 (1.34-1.46) 7,328/21,636 1.51 (1.46-1.56)

Model B 4,342/21,636 1.37 (1.32-1.43) 7,328/21,636 1.47 (1.42-1.52)

Model C 4,342/21,636 1.29(1.24-1.35) 7,328/21,636 1.40 (1.35-1.45)

Model D 3,029/17,242 1.32 (1.25-1.40) 5,270/17,242 1.43 (1.37-1.49)

Model A: adjusted for age and sex; Model B: adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure and cholesterol level; Model C: adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level, body mass index, prevalent illnesses including asthma, COPD, diabetes, stroke and myocardial infarction; Model D: adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level, body mass index, after excluding participants with prevalent illnesses including asthma, COPD, diabetes, stroke and myocardial infarction.

36

645646647648649

650651652653654655656657658659660661662

56

Appendix 10: Hazard ratios (HR) and corresponding 95% confidence intervals (95%CI) for risk of mortality and cardiovascular events during follow up (1993/97-2009/11) in EPIC-Norfolk by combinations of drugs with contribute to anticholinergic burden score

ACB drug

usage

Mortality Cardiovascular disease incidence

n events/

total N

HR (95% CI) n events/

total N

HR (95% CI)

No ACB drugs 2,833/17,317 1 4,939/17,317 1

Class 1 1,262/3,389 1.40 (1.30-1.50) 1,988/3,389 1.66 (1.57-1.76)

Class 2 18/70 1.89 (1.19-3.01) 27/70 1.46 (1.00-2.13)

Class 3 139/619 1.42 (1.20-1.68) 229/619 1.42 (1.24-1.62)

Class 1 & 2 4/13 1.77 (0.66-4.74) 8/13 1.79 (0.89-3.59)

Class 1 & 3 75/203 1.52 (1.21-1.92) 120/203 1.99 (1.66-2.39)

Class 2 & 3 8/17 2.97 (1.48-5.99) 10/17 2.12 (1.13-3.95)

Class 1& 2 & 3 3/8 1.07 (0.34-3.33) 7/8 6.46 (3.06-13.60)

Adjusted for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level, body mass index, prevalent illnesses including asthma, COPD, diabetes, stroke and myocardial infarction.

37

663664665666667

668669670671672

57

Appendix 11: Sample characteristics of propensity matched men and women of the EPIC-Norfolk (1993/1997-2009/2011) according to the total anticholinergic burden (ACB) scoreVariable Matched cohort 1 Matched cohort 2 Matched cohort 3

ACB score 0 group (n=2,704)

ACB score 1group (n=2,704)

p-value ACB score 0group (n=1,324)

ACB score 2-3 group(n=1,324)

p-value ACB score 0 group(n=291)

ACB score >3 group(n=291)

p-value

Age (years) 63 (±9) 63 (±9) 0.05 62 (±9) 52 (±9) 0.89 63 (±9) 63 (±9) 0.93Sex

MenWomen

1319 (49%)1385 (51%)

1348 (50%)1356 (50%)

0.43577 (44%)747 (56%)

593 (45%)731 (55%)

0.53115 (40%)176 (60%)

126 (43%)165 (57%)

0.36

Social classProfessional

ManagerSkilled non-manual

Skilled manualSemi-skilledNon-skilled

152 (6%)943 (35%)507 (19%)611 (23%)384 (14%)107 (4%)

164 (6%)927 (34%)464 (17%)637 (24%)403 (15%)109 (4%)

0.5588 (7%)425 (32%)253 (19%)297 (22%)212 (16%)49 (4%)

68 (5%)446 (34%)234 (18%)321 (24%)193 (15%)62 (5%)

0.6022 (8%)69 (24%)57 (20%)70 (24%)58 (20%)15 (5%)

19 (7%)80 (27%)58 (20%)78 (27%)43 (15%)13 (4%)

0.29

Smoking Current-smoker

Ex-smokerNever smoker

280 (10%)1296 (48%)1128 (42%)

256 (9%)1324 (49%)1124 (42%)

0.79155 (12%)609 (46 %)560 (42%)

155 (12%)626 (47%)543 (41%)

0.5840 (14%)139 (48%)112 (38%)

51 (18%)122 (42%)118 (41%)

0.90

Alcohol use (units/week)

6.8 (±9.5) 6.8 (±10) 0.85 5.5 (±7.2) 6.0 (±8.3) 0.08 4.5 (±5.9) 5.2 (±7.9) 0.21

Education levelNo qualification 1147 (42%) 1152 (43%)

0.85617 (47%) 598 (45%)

0.76126 132

1.00

38

673674

58

0-LevelA-Level

Higher degree

276 (10%)997 (37%)284 (11%)

255 (9%)1039 (38%)258 (10%)

113 (9%)456 (34%)138 (10%)

111 (8%)502 (38%)113 (9%)

(43%)33 (11%)111 (38%)21 (7%)

(45%)25 (9%)108 (37%)26 (9%)

Physical activity Inactive

Moderately inactiveModerately active

Active

1068 (40%)785 (29%)490 (18%)361 (13%)

1037 (38%)760 (28%)520 (19%)387 (14%)

0.17526 (39%)391 (30%)252 (19%)155 (12%)

552 (42%)361 (27%)234 (18%)177 (13%)

0.74139 (48%)95 (33%)39 (13%)18 (6%)

142 (49%)86 (30%)40 (14%)23 (8%)

0.92

Cholesterol (mmol/L)

6.4 (±1.2) 6.3 (±1.2) 0.30 6.3 (±1.3) 6.3 (±1.2) 0.69 6.5 (±1.2) 6.4 (±1.2) 0.67

Systolic BP (mmHg) 141 (±19) 140 (±19) 0.16 138 (±19) 137 (±19) 0.17 138 (±18) 138 (±19) 0.92BMI (kg/m2) 27 (±4) 27 (±4) 0.50 27 (±4) 27 (±4) 0.26 28 (±5) 27 (±4) 0.29COPD 346 (13%) 346 (13%) 1.00 201 (15%) 184 (14%) 0.35 36 (12%) 40 (14%) 0.62Asthma 595 (22%) 630 (23%) 0.26 189 (14%) 204 (15%) 0.41 49 (17%) 58 (20%) 0.34Previous MI 210 (8%) 222 (8%) 0.55 157 (12%) 153 (12%) 0.81 37 (9%) 31 (11%) 0.58Previous stroke 71 (3%) 78 (3%) 0.56 50 (4%) 52 (4%) 0.84 17 (6%) 13 (4%) 0.45Diabetes 93 (3%) 102 (4%) 0.51 49 (4%) 59 (4%) 0.33 16 (6%) 17 (6%) 0.86New CVD events 1169 (43%) 1459 (54%) <0.001 552 (42%) 751 (57%) <0.001 136

(47%)179 (62%)

<0.001

Deaths 748 (28%) 887 (33%) <0.001 360 (27%) 498 (38%) <0.001 77 (26%) 124 (43%)

<0.001

39

675

59

Values presented are mean (sd) for continuous and number (%) for categorical data. *overall P value. BP=blood pressure, BMI = body mass index, COPD= chronic obstructive pulmonary disease; MI=myocardial infarction, CVD= cardiovascular diseases. Total anticholinergic burden (ACB) calculated as a score which is the sum of the [number of class 1 anticholinergic drugs, the number of class 2 anticholinergic drugs x2 and the number class 3 anticholinergic drugs x3]. Classification of drugs with ACB class 1, 2 and 3 based on criteria of Anticholinergic Cognitive Burden Scale (Boustani MA, et al 2008;4:311–320).

40

676677678679680681682

60

Appendix 12: Propensity matched risk of mortality and incident cardiovascular event according to total anticholinergic burden score (0, 1, 2-3 or >3) during follow up (1993-2011) in EPIC-Norfolk

Variable Matched cohort 1 Matched cohort 2 Matched cohort 3N OR or HR p-value N OR or HR p-value n OR or HR p-value

Logistic regression for death at full follow up

5408 1.45 (1.26-1.65) <0.001 2648 1.94 (1.59-2.37)

<0.001 582 2.70 (1.77-4.11)

<0.001

Cox proportional hazards for death

5408 1.32 (1.19-1.45) <0.001 2648 1.65 (1.43-1.89)

<0.001 582 2.01 (1.51-2.69)

<0.001

Logistic regression for CVD at full follow up

5408 1.76 (1.56-1.98) <0.001 2648 2.32 (1.94-2.80)

<0.001 582 2.25 (1.53-3.31)

<0.001

Cox proportional hazards for CVD

5408 1.45 (1.34-1.57) <0.001 2648 1.84 (1.64-2.06)

<0.001 582 1.78 (1.42-2.23)

<0.001

Propensity matched for age, sex, smoking, alcohol consumption, physical activity level, education level, occupational social class, systolic blood pressure, cholesterol level, body mass index, prevalent illnesses including asthma, COPD, diabetes, stroke and myocardial infarction

41

683684

685686687

61