Atrial fibrillation and comorbidities: Clinical characteristics and antithrombotic treatment in GLORIA-AF.

BACKGROUND: Patients with AF often have multimorbidity (the presence of ≥2 concomitant chronic conditions).
OBJECTIVE: To describe baseline characteristics, patterns of antithrombotic therapy, and factors associated with oral anticoagulant (OAC) prescription in patients with AF and ≥2 concomitant, chronic, comorbid conditions.
METHODS: Phase III of the GLORIA-AF Registry enrolled consecutive patients from January 2014 through December 2016 with recently diagnosed AF and CHA2DS2-VASc score ≥1 to assess the safety and effectiveness of antithrombotic treatment.
RESULTS: Of 21,241 eligible patients, 15,119 (71.2%) had ≥2 concomitant, chronic, comorbid conditions. The proportions of patients with multimorbidity receiving non-vitamin K antagonist oral anticoagulants (NOACs) and vitamin K antagonists (VKA) were 60.2% and 23.6%, respectively. The proportion with paroxysmal AF was 57.0% in the NOAC group and 45.4% in the VKA group. Multivariable log-binomial regression analysis found the following factors were associated with no OAC prescription: pattern of AF (paroxysmal, persistent, or permanent), coronary artery disease, myocardial infarction, prior bleeding, smoking status, and region (Asia, North America, or Europe). Factors associated with OAC prescriptions were age, body mass index, renal function, hypertension, history of cerebral ischemic symptoms, and AF ablation.
CONCLUSION: Multimorbid AF patients prescribed NOACs have fewer comorbidities than those prescribed VKAs. Age, AF pattern, comorbidities, and renal function are associated with OAC prescription.

Introduction

Atrial fibrillation (AF) affects approximately 3% of adults and its prevalence and incidence are rising [1] with the aging of the population [2]. Older patients with AF often have other chronic conditions that affect their clinical course [3]. Multimorbidity (the presence of ≥2 concomitant chronic conditions) demands a holistic and integrated approach to patient care [4] since these patients face higher risks of stroke and bleeding than those without comorbidities [5, 6]. The interplay between comorbidity, AF, and optimal thromboprophylaxis has both medical and economic implications [7]

The aim of this analysis of the GLORIA-AF dataset is to describe baseline characteristics and antithrombotic therapy prescription patterns in patients with AF and multimorbidity and to identify factors associated with the selection of an oral anticoagulant (OAC) type for these complex patients.

Materials and methods

The design of the GLORIA-AF registry (https://clinicaltrials.gov/ct2/home; trial registration numbers NCT01468701, NCT01671007, NCT01937377) has been reported [8]. The study protocol is concordant with the ethical guidelines of the 1975 Declaration of Helsinki, and informed consent was obtained from each patient before enrollment.

The registry collected routine clinical practice data regarding patients with newly diagnosed AF to evaluate patient characteristics influencing the selection, safety, and effectiveness of antithrombotic therapy. Phase I was conducted before non-vitamin K antagonist oral anticoagulants (NOACs) were available for stroke prevention in AF. Phase II began when dabigatran was approved in countries with participating clinical centers. Baseline characteristics were collected and those prescribed dabigatran were followed up for 2 years in Phase II. Phase III, which started when dabigatran had been more widely adopted, gathered data for up to 3 years, regardless of antithrombotic management [8].

Consecutive patients from 38 countries were enrolled between 2014 and 2016. Adult patients with recently diagnosed nonvalvular AF (<3 months before the baseline visit; Latin America <4.5 months) at risk of stroke (CHA2DS2-VASc score ≥1) achieved by any of the following: heart failure or left ventricular systolic dysfunction, hypertension, diabetes, prior stroke, transient ischemic attack (TIA) or systemic embolism, myocardial infarction (MI), peripheral artery disease, age ≥65 years, or female sex, were enrolled [9]. The risks of stroke and bleeding were assessed using the CHA2DS2-VASc and HAS-BLED (1 point is achieved by any of the following: hypertension, abnormal renal or hepatic function, prior stroke, bleeding or predisposition, labile International Normalised Ratio, elderly [>65 years], or concomitant use of alcohol or anti-inflammatory medications) [10]. Antithrombotic therapy was prescribed by the treating physicians according to local standards. This report is focused on baseline data obtained from patients in Phase III, collected using electronic case report forms.

Statistical analysis

Baseline characteristics are summarized descriptively. Categorical variables are reported as absolute frequencies and percentages, and continuous variables are summarized by median (Quartile 1, Quartile 3). Baseline characteristics included stratification of patients with AF and multimorbidity according to stroke prevention strategies (OAC vs antiplatelet vs no antithrombotic therapy, NOAC vs vitamin K antagonists [VKAs], and NOACs once daily [QD] vs twice daily [BID]). Standardized differences were used to compare baseline characteristics across various stroke prevention strategies, focusing on variables with the highest standardized differences; differences ≤10% in absolute value were considered as balanced between groups [11].

Factors associated with antithrombotic treatment choice were analyzed by log-binomial, multivariable regression models, providing relative probability ratios for prescription (OAC vs no OAC use, NOAC vs VKA; and by region). Missing data were handled using multiple imputation, replacing missing data with multiple simulated values based on regression models to provide comparatively unbiased estimates under the missing-at-random assumption. The procedure introduces random error to compensate for the added, imputed information. The imputation regression models used 56 predictors to impute the missing data, and were repeated 20 times to give 20 datasets with imputed data [12].

Confidence intervals were calculated based on likelihood ratios and Rubin’s method to combine results across imputations. Both univariate and multivariable log-binomial regression analyses were performed to evaluate crude as well as the adjusted probability ratios together with 95% confidence intervals. The term “probability ratio” was used rather than “risk ratio”, as our measure describes treatment selections rather than adverse outcomes.

All data were calculated using SAS version 9.4 (SAS Institute, Inc., Cary, NC).

Results

Of 21,241 eligible patients in this subanalysis, 15,119 (71.2%) had ≥2 concomitant, chronic conditions (Table 1).

Table 1

Proportion of AF patients according to number of comorbid diseases.

Number of Comorbid Diseases	Number of Patients (n = 21,241)
0	1434 (6.8)
1	4688 (22.1)
2	5559 (26.2)
3	4286 (20.2)
4	2664 (12.5)
5	1463 (6.9)
6	695 (3.3)
7	332 (1.6)
8	88 (0.4)
9	22 (0.1)
10	8 (0.0)
11	2 (0.0)

aAF = atrial fibrillation.

Baseline characteristics of AF multimorbid patients

Baseline characteristics of patients are summarized based on antithrombotic therapy (Table 2). Among multimorbid AF patients, 83.8% were prescribed OACs, 11.0% were prescribed antiplatelet therapy, and 5.2% were prescribed no antithrombotic therapy. The median (66.0, 79.0) age was 73.0 years in the OAC group, 71.0 (63.0–79.0) years in the antiplatelet therapy group, and 72.0 (64.0–80.0) years in the no antithrombotic therapy group. The proportions of females in these groups were 44.5%, 41.7%, and 45.5%, respectively. The median CHA2DS2-VASc and HAS-BLED scores were similar across the 3 groups.

Table 2

Baseline characteristics of AF multimorbid patients prescribed OAC or antiplatelets or no antithrombotic therapy.

	OAC (n = 12,677)	Antiplatelets (n = 1658)	No Antithrombotic Therapy (n = 784)
Age (y), median (Q1, Q3)	73.0 (66.0–79.0)	71.0 (63.0–79.0)	72.0 (64.0–80.0)
Females, n (%)	5645 (44.5)	691 (41.7)	357 (45.5)
BMI (kg/m2), median (Q1, Q3)	28.0 (24.8–32.0)	26.1 (23.5–30.0)	26.1 (23.4–29.6)
Missing	123 (1.0)	17 (1.0)	8 (1.0)
Current smoker	1145 (9.0)	223 (13.4)	100 (12.8)
Alcohol abuse, ≥8 units/ week	866 (6.8)	85 (5.1)	54 (6.9)
Type of AF, n (%)
Paroxysmal	6810 (53.7)	1166 (70.3)	496 (63.3)
Persistent	4478 (35.3)	401 (24.2)	242 (30.9)
Permanent	1389 (11.0)	91 (5.5)	46 (5.9)
Categorization of AF, n (%)
EHRA I	4686 (37.0)	550 (33.2)	273 (34.8)
EHRA II	4025 (31.8)	563 (34.0)	270 (34.4)
EHRA III	3063 (24.2)	431 (26.0)	183 (23.3)
EHRA IV	903 (7.1)	114 (6.9)	58 (7.4)
Creatinine clearance (mL/min) (measured), median (Q1, Q3)	70.6 (52.5–95.3)	69.5 (50.9–92.4)	67.8 (49.7–90.3)
Creatinine clearance (mL/min), n (%)
<15	100 (0.8)	18 (1.1)	10 (1.3)
15–29	305 (2.4)	62 (3.7)	23 (2.9)
30–49	1848 (14.6)	252 (15.2)	136 (17.3)
50–79	4152 (32.8)	526 (31.7)	253 (32.3)
≥80	4080 (32.2)	520 (31.4)	243 (31.0)
Missing	2192 (17.3)	280 (16.9)	119 (15.2)
CHA2DS2-VASc score, median (Q1, Q3)	4.0 (3.0–5.0)	4.0 (2.0–5.0)	3.0 (2.0–4.0)
HAS-BLED score, median (Q1, Q3)	1.0 (1.0–2.0)	2.0 (2.0–3.0)	1.0 (1.0–2.0)
Missing (HAS-BLED), n (%)	1234 (9.7)	134 (8.1)	69 (8.8)
Medical history, n (%)
Congestive heart failure	3509 (27.7)	487 (29.4)	215 (27.4)
Hypertension	10,989 (86.7)	1370 (82.6)	638 (81.4)
Diabetes mellitus	4021 (31.7)	510 (30.8)	226 (28.8)
Previous stroke or TIA	2347 (18.5)	336 (20.3)	159 (20.3)
Myocardial infarction	1580 (12.5)	384 (23.2)	58 (7.4)
Coronary artery disease	3017 (23.8)	745 (44.9)	149 (19.0)
Peripheral artery disease	503 (4.0)	79 (4.8)	21 (2.7)
Cancer	1671 (13.2)	167 (10.1)	115 (14.7)
Dementia	101 (0.8)	18 (1.1)	1 (0.1)
Gastric ulcer	145 (1.1)	20 (1.2)	13 (1.7)
Gastritis or duodenitis	455 (3.6)	70 (4.2)	50 (6.4)
Chronic kidney disease	3881 (30.6)	526 (31.7)	271 (34.6)
COPD	1045 (8.2)	120 (7.2)	59 (7.5)
Bleeding (after diagnosis of AF), n (%)	182 (1.4)	32 (1.9)	33 (4.2)
Bleeding on OAC, n (%)	159 (87.4)	27 (84.4)	18 (54.5)
Location of bleeding (after diagnosis of AF), n (%)*
Intracranial hemorrhage	12 (6.6)	6 (18.8)	8 (24.2)
Upper GI bleed	12 (6.6)	4 (12.5)	3 (9.1)
Lower GI bleed	25 (13.7)	6 (18.8)	5 (15.2)
GI bleed not further specified	11 (6.0)	4 (12.5)	4 (12.1)
Urogenital hemorrhage	31 (17.0)	3 (9.4)	3 (9.1)
Bleeding at other location	81 (44.5)	7 (21.9)	8 (24.2)
Bleeding with unknown location	10 (5.5)	2 (6.3)	2 (6.1)
Region, n (%)
Asia	1739 (13.7)	719 (43.4)	325 (41.5)
Europe	6514 (51.4)	443 (26.7)	266 (33.9)
North America	3429 (27.0)	415 (25.0)	144 (18.4)
Latin America	995 (7.8)	81 (4.9)	49 (6.3)
Type of site, n (%)
GP/primary care	686 (5.4)	171 (10.3)	77 (9.8)
Specialist office	3902 (30.8)	512 (30.9)	191 (24.4)
Community hospital	3757 (29.6)	350 (21.1)	175 (22.3)
University hospital	3878 (30.6)	543 (32.8)	326 (41.6)
Outpatient health care centre	222 (1.8)	51 (3.1)	6 (0.8)
Anticoagulation clinics	82 (0.6)	6 (0.4)	4 (0.5)
Other	150 (1.2)	25 (1.5)	5 (0.6)

aAF = atrial fibrillation; BMI = body mass index; CHA2DS2-VASc = congestive heart failure/left ventricular dysfunction, hypertension, age ≥75 years, diabetes, stroke/transient ischemic attack/systemic embolism, vascular disease, age 65–74 years, sex category (female); COPD = chronic obstructive pulmonary disease; EHRA = European Heart Rhythm Association; GI = gastrointestinal; GP = general practitioner; HAS-BLED = hypertension, abnormal renal /liver function, stroke, bleeding history or predisposition, labile International Normalised Ratio, elderly (>65 years), drugs or alcohol concomitantly; OAC = oral anticoagulant; Q = quartile; TIA = transient ischemic attack; y = years.

*Proportion calculated out of Bleeding (after diagnosis of AF).

Baseline characteristics of patients prescribed NOACs or VKAs are shown in Table 3. The median age was 73.0 (66.0–79.0) years, and the proportion of females was 44% in both treatment groups. There were no differences in CHA2DS2-VASc and HAS-BLED scores between these 2 groups. The prevalence of paroxysmal AF in patients with multimorbidity on NOACs and VKAs was 57.0% and 45.4%, respectively. Among patients on NOACs, 38.4% had a European Heart Rhythm Association symptom score of I, compared with 33.3% for patients on VKAs. A lower proportion (1.6%) of patients on NOACs had a glomerular filtration rate of 15–29 mL/min, compared with 4.4% of those on VKAs.

Table 3

Baseline characteristics of AF multimorbid patients prescribed NOACs or VKAs.

	NOAC (n = 9105)	VKA (n = 3572)	Standardized Difference
Age (y), median (Q1, Q3)	73.0 (66.0–79.0)	73.0 (66.0–79.0)	0.005
Females, n (%)	4072 (44.7)	1573 (44.0)	–0.014
BMI (kg/m2), median (Q1, Q3)	28.0 (24.8–32.2)	27.8 (24.6–31.6)	–0.066
Missing	37 (1.2)	60 (1.0)	0.020
Current smoker	812 (8.9)	333 (9.3)	0.014
Alcohol abuse, ≥8 units/ week	651 (7.1)	215 (6.0)	–0.046
Type of AF, n (%)
Paroxysmal	5187 (57.0)	1623 (45.4)	–0.232
Persistent	3052 (33.5)	1426 (39.9)	0.133
Permanent	866 (9.5)	523 (14.6)	0.158
Categorization of AF, n (%)
EHRA I	3496 (38.4)	1190 (33.3)	–0.106
EHRA II	2886 (31.7)	1139 (31.9)	0.004
EHRA III	2131 (23.4)	932 (26.1)	0.062
EHRA IV	592 (6.5)	311 (8.7)	0.083
Creatinine clearance (mL/min) (measured), median (Q1, Q3)	72.1 (53.7–97.0)	66.8 (48.9–91.0)	–0.078
Creatinine clearance (mL/min) n (%)
<15	50 (0.5)	50 (1.4)	0.087
15–29	148 (1.6)	157 (4.4)	0.163
30–49	1280 (14.1)	568 (15.9)	0.052
50–79	3046 (33.5)	1106 (31.0)	–0.053
≥80	3053 (33.5)	1027 (28.8)	–0.103
Missing	1528 (16.8)	664 (18.6)	0.047
CHA2DS2-VASc score, median (Q1, Q3)	4.0 (3.0–5.0)	4.0 (3.0–5.0)	0.080
HAS-BLED score, median (Q1, Q3)	1.0 (1.0–2.0)	1.0 (1.0–2.0)	0.016
Missing (HAS-BLED), n (%)	858 (9.4)	376 (10.5)	0.037
Medical history, n (%)
Congestive heart failure	2232 (24.5)	1277 (35.8)	0.247
Hypertension	7907 (86.8)	3082 (86.3)	–0.016
Diabetes mellitus	2839 (31.2)	1182 (33.1)	0.041
Previous stroke or TIA	1741 (19.1)	606 (17.0)	–0.056
Myocardial infarction	1039 (11.4)	541 (15.1)	0.110
Coronary artery disease	2104 (23.1)	913 (25.6)	0.057
Peripheral artery disease	355 (3.9)	148 (4.1)	0.012
Cancer	1223 (13.4)	448 (12.5)	–0.027
Dementia	76 (0.8)	25 (0.7)	–0.016
Gastric ulcer	111 (1.2)	34 (1.0)	–0.026
Gastritis or duodenitis	317 (3.5)	138 (3.9)	0.020
Chronic kidney disease	2663 (29.2)	1218 (34.1)	0.104
COPD	743 (8.2)	302 (8.5)	0.011
Bleeding (after diagnosis of AF), n (%)	130 (1.4)	52 (1.5)	0.002
Bleeding on OAC, n (%)	112 (86.2)	47 (90.4)	0.132
Location of bleeding (after diagnosis of AF), n (%)*
Intracranial hemorrhage	11 (8.5)	1 (1.9)	–0.298
Upper GI bleed	8 (6.2)	4 (7.7)	0.061
Lower GI bleed	20 (15.4)	5 (9.6)	–0.175
GI bleed not further specified	9 (6.9)	2 (3.8)	–0.137
Urogenital hemorrhage	20 (15.4)	11 (21.2)	0.150
Bleeding at other location	56 (43.1)	25 (48.1)	0.101
Bleeding with unknown location	6 (4.6)	4 (7.7)	0.128
AF cardioversion	1814 (19.9)	521 (14.6)	–0.142
Region, n (%)
Asia	1222 (13.4)	517 (14.5)	0.030
Europe	4498 (49.4)	2016 (56.4)	0.141
North America	2808 (30.8)	621 (17.4)	–0.319
Latin America	577 (6.3)	418 (11.7)	0.188
Type of site, n (%)
GP/primary care	502 (5.5)	184 (5.2)	–0.016
Specialist office	3053 (33.5)	849 (23.8)	–0.217
Community hospital	2880 (31.6)	877 (24.6)	–0.158
University hospital	2454 (27.0)	1424 (39.9)	0.276
Outpatient health care centre	72 (0.8)	150 (4.2)	0.220
Anticoagulation clinics	37 (0.4)	45 (1.3)	0.094
Other	107 (1.2)	43 (1.2)	0.003

aAF = atrial fibrillation; BMI = body mass index; CHA2DS2-VASc = congestive heart failure/left ventricular dysfunction, hypertension, age ≥75 years, diabetes, stroke/transient ischemic attack/systemic embolism, vascular disease, age 65–74 years, sex category (female); COPD = chronic obstructive pulmonary disease; EHRA = European Heart Rhythm Association; GI = gastrointestinal; GP = general practitioner; HAS-BLED = hypertension, abnormal renal /liver function, stroke, bleeding history or predisposition, labile International Normalised Ratio, elderly (>65 years), drugs or alcohol concomitantly; NOAC = nonvitamin K antagonist oral anticoagulants; OAC = oral anticoagulant; Q = quartile; TIA = transient ischemic attack; VKA = vitamin K antagonists; y = years.

*Proportion calculated out of Bleeding (after diagnosis of AF).

Cardioversion was performed in 19.9% of patients on NOACs vs 14.6% of those on VKAs. Treatment in specialist offices was more prevalent for patients on NOACs (33.5% vs 23.8% in the VKA group), while comorbidities such as heart failure (HF) and MI were less prevalent among patients given NOACs.

Patient demographics, cardiovascular risk factors, comorbid diseases, AF categorization, stroke and bleeding risks, and concomitant treatments of patients on NOACs QD vs BID are summarized in Table 4 There were generally small differences between patients taking NOACs QD vs BID. Previous TIA or stroke were present in 14.9% of the patients on NOACs QD vs 21.3% of the patients on NOACs BID (Table 4).

Table 4

Baseline characteristics of AF multimorbid patients prescribed NOACs QD or NOACs BID.

	NOAC QD (n = 3071)	NOAC BID (n = 6034)	Standardized Difference
Age (y), median (Q1, Q3)	72.0 (65.0–79.0)	73.0 (66.0–79.0)	–0.098
Females, n (%)	1306 (42.5)	2766 (45.8)	–0.067
BMI (kg/m2), median (Q1, Q3)	28.3 (25.0–32.8)	27.9 (24.8–32.0)	0.089
Current smoker	250 (8.1)	562 (9.3)	–0.042
Alcohol abuse, ≥8 units/ week	242 (7.9)	409 (6.8)	0.042
Type of AF, n (%)
Paroxysmal	1767 (57.5)	3420 (56.7)	0.017
Persistent	1045 (34.0)	2007 (33.3)	0.016
Permanent	259 (8.4)	607 (10.1)	–0.056
Categorization of AF, n (%)
EHRA I	1138 (37.1)	2358 (39.1)	–0.042
EHRA II	983 (32.0)	1903 (31.5)	0.010
EHRA III	775 (25.2)	1356 (22.5)	0.065
EHRA IV	175 (5.7)	417 (6.9)	–0.050
Creatinine clearance (mL/min), (measured), median (Q1, Q3)	74.4 (55.3–101.8)	70.5 (53.1–94.3)	0.041
Creatinine clearance, n (%)
<15	18 (0.6)	32 (0.5)	0.008
15–29	40 (1.3)	108 (1.8)	–0.040
30–49	401 (13.1)	879 (14.6)	–0.044
50–79	1018 (33.1)	2028 (33.6)	–0.010
≥80	1125 (36.6)	1928 (32.0)	0.099
Missing	469 (15.3)	1059 (17.6)	–0.062
CHA2DS2-VASc score, median (Q1, Q3)	3.0 (2.0–4.0)	4.0 (3.0–5.0)	–0.127
HAS-BLED score, median (Q1, Q3)	1.0 (1.0–2.0)	1.0 (1.0–2.0)	–0.066
Missing (HAS-BLED), n (%)	302 (9.8)	556 (9.2)	0.021
Medical history, n (%)
Congestive heart failure	772 (25.1)	1460 (24.2)	0.022
Hypertension	2672 (87.0)	5235 (86.8)	0.007
Diabetes mellitus	1021 (33.2)	1818 (30.1)	0.067
Previous stroke or TIA	457 (14.9)	1284 (21.3)	–0.167
Myocardial infarction	366 (11.9)	673 (11.2)	0.024
Coronary artery disease	746 (24.3)	1358 (22.5)	0.042
Peripheral artery disease	119 (3.9)	236 (3.9)	–0.002
Cancer	407 (13.3)	816 (13.5)	–0.008
Dementia	24 (0.8)	52 (0.9)	–0.009
Gastric ulcer	40 (1.3)	71 (1.2)	0.011
Gastritis or duodenitis	116 (3.8)	201 (3.3)	0.024
Chronic kidney disease	839 (27.3)	1824 (30.2)	–0.064
COPD	258 (8.4)	485 (8.0)	0.013
Bleeding (after diagnosis of AF), n (%)	57 (1.9)	73 (1.2)	0.053
Bleeding on OAC, n (%)	52 (91.2)	60 (82.2)	0.269
Location of bleeding (after diagnosis of AF), n (%)
Intracranial hemorrhage	2 (3.5)	9 (12.3)	–0.331
Upper GI bleed	4 (7.0)	4 (5.5)	0.064
Lower GI bleed	10 (17.5)	10 (13.7)	0.106
GI bleed not further specified	5 (8.8)	4 (5.5)	0.128
Urogenital hemorrhage	6 (10.5)	14 (19.2)	–0.245
Bleeding at other location	24 (42.1)	32 (43.8)	–0.035
Bleeding with unknown location	6 (10.5)	0 (0.0)	0.438
AF cardioversion	710 (23.1)	1104 (18.3)	0.119
Region, n (%)
Asia	356 (11.6)	866 (14.4)	–0.082
Europe	1465 (47.7)	3033 (50.3)	–0.051
North America	1056 (34.4)	1752 (29.0)	0.115
Latin America	194 (6.3)	383 (6.3)	–0.001
Type of site, n (%)
GP/primary care	184 (6.0)	318 (5.3)	0.031
Specialist office	1110 (36.1)	1943 (32.2)	0.083
Community hospital	921 (30.0)	1959 (32.5)	–0.053
University hospital	773 (25.2)	1681 (27.9)	–0.061
Outpatient health care center	19 (0.6)	53 (0.9)	–0.030
Anticoagulation clinics	18 (0.6)	19 (0.3)	0.041
Other	46 (1.5)	61 (1.0)	0.044

aAF = atrial fibrillation; BID = twice daily; BMI = body mass index; CHA2DS2-VASc = congestive heart failure/left ventricular dysfunction, hypertension, age ≥75 years, diabetes, stroke/transient ischemic attack/systemic embolism, vascular disease, age 65–74 years, sex category (female); COPD = chronic obstructive pulmonary disease; EHRA = European Heart Rhythm Association; GI = gastrointestinal; GP = general practitioner; HAS-BLED = hypertension, abnormal renal /liver function, stroke, bleeding history or predisposition, labile International Normalised Ratio, elderly (>65 years), drugs or alcohol concomitantly; NOAC = nonvitamin K antagonist oral anticoagulants; OAC = oral anticoagulant; Q = quartile; QD = once daily; TIA = transient ischemic attack.

Factors associated with OAC non-prescription in multimorbid AF patients globally

Results from univariate analyses are presented in the S1 File. In the multivariable log-binomial regression analysis, factors associated with prescriptions for no OAC use in multimorbid AF patients were: type of AF (paroxysmal/persistent vs permanent), coronary artery disease (CAD), MI, history of bleeding, smoking status (current vs nonsmoker), and region (Asia, North America vs Europe). Factors associated with increased OAC use were: age 65–74 vs ≥75 years, body mass index (BMI) class (≥25 vs 18.5–24 kg/m2), creatinine clearance (30–59 vs ≥80 mL/min), hypertension, prior TIA or stroke, and AF ablation (Table 5).

Table 5

Multivariable log-binomial analysis for factors associated with prescription of OAC therapy (no OAC vs OAC),.

Factor	Relative Risk (95% CI) For Prescription of No OAC Globally
Age
<65	1.05 (0.95–1.16)
65–74	0.90 (0.83–0.99)
≥75	1.0 (ref)
BMI class
<18.5	0.98 (0.77–1.24)
18.5–24	1.0 (ref)
25–29	0.85 (0.79–0.91)
30–34	0.77 (0.69–0.87)
≥35	0.70 (0.60–0.81)
Gender
Male	1.0 (ref)
Female	1.05 (0.97–1.13)
Current smoker	1.14 (1.03–1.25)
Past smoker	0.91 (0.84–0.99)
Categorization of AF
EHRA I	1.0 (ref)
EHRA II	1.04 (0.96–1.12)
EHRA III	0.99 (0.91–1.07)
EHRA IV	1.07 (0.95–1.20)
Type of AF
Paroxysmal	1.67 (1.42–1.97)
Persistent	1.20 (1.02–1.43)
Permanent	1.0 (ref)
Hypertension	0.89 (0.83–0.97)
Coronary artery disease	1.42 (1.31–1.53)
Myocardial infarction	1.18 (1.08–1.28)
Congestive heart failure	1.01 (0.94–1.08)
Diabetes mellitus	0.95 (0.88–1.02)
Previous TIA or stroke	0.81 (0.68–0.97)
Bleeding after diagnosis of AF	1.60 (1.42–1.79)
Peripheral artery disease	1.13 (0.96–1.34)
Cancer	1.00 (0.90–1.12)
Functional dyspepsia	0.85 (0.56–1.27)
Gastric ulcer	0.91 (0.69–1.21)
Gastritis or duodenitis	0.95 (0.82–1.10)
COPD	1.03 (0.90–1.19)
Hyperthyroidism	0.96 (0.79–1.17)
Hepatic disease	1.05 (0.87–1.27)
Dementia	1.09 (0.76–1.56)
AF cardioversion	0.96 (0.89–1.04)
Creatinine clearance (mL/min)
<30	1.09 (0.94–1.26)
30–59	0.88 (0.79–0.97)
60–79	0.91 (0.83–1.00)
≥80	1.0 (ref)
AF ablation	0.30 (0.20–0.45)
Region
Asia	3.17 (2.88–3.49)
Europe	1.0 (ref)
North America	1.24 (1.11–1.39)
Latin America	1.14 (0.96–1.37)
Medical treatment reimbursed by
Self-pay/no coverage	0.82 (0.69–0.96)
Not self-pay	1.0 (ref)
Type of site
Specialist office	1.26 (1.14–1.39)
Community hospital	1.0 (ref)
University hospital	1.28 (1.17–1.40)

aA few other variables (alcohol abuse, psychosocial factors, biological heart valve implant, valve repair, and peptic ulcer) are included in the multivariable log-binomial regression analysis model and are presented in the S1 File.

bAF = atrial fibrillation; BMI = body mass index; CI = confidence interval; COPD = chronic obstructive pulmonary disease; EHRA = European Heart Rhythm Association; OAC = oral anticoagulant; ref = reference; TIA = transient ischemic attack.

Factors associated with OACs non-prescription in multimorbid AF patients in Asia, Europe, and North America

Factors associated with prescriptions for no OAC use in multimorbid AF patients in Asia, Europe, and North America are presented in S1 Table in S2 File. Factors associated with increased OAC use are included in S1 Table in S2 File.

Factors associated with type of OAC use in multimorbid AF patients globally

Factors associated with prescriptions for VKA use globally in multimorbid AF patients were: age <75 vs ≥75 years, MI, congestive HF, diabetes mellitus, creatinine clearance (<60 vs ≥80 mL/min), S2 Table in S2 File.

Factors associated with decreased VKA use globally were: type of AF (paroxysmal/persistent vs permanent), previous TIA or stroke, medical treatment reimbursement (self-pay/no coverage vs not self-pay), S2 Table in S2 File.

Factors associated with OAC use in multimorbid AF patients in Asia, Europe, North America, and Latin America

Factors associated with prescriptions for VKA use in multimorbid AF patients in Asia, Europe, North America, and Latin America are presented in S3 Table in S2 File. Factors associated with decreased prescriptions for VKA use in multimorbid AF patients in Asia, Europe, North America, and Latin America are presented in S3 Table in S2 File.

Discussion

There are still knowledge gaps in how OACs are used in clinical practice in patients with AF and multiple comorbidities and which factors influence OAC prescription in such patients. Our study shows that, despite a median CHA2DS2-VASc score >3, approximately 16% of patients with multimorbidity and AF are not anticoagulated. The baseline characteristics in these complex patients differ in relation to antithrombotic therapy selection, suggesting that comorbidities may influence antithrombotic therapy prescription patterns for patients with AF. For example, prescription of OACs globally in patients with AF and multimorbidity was associated with age, BMI, cardiovascular risk factors (smoking status), AF pattern, concomitant diseases (ie, hypertension, CAD, MI, previous TIA or stroke), history of bleeding, renal function, rhythm control strategy (AF ablation and AF cardioversion), and region (Asia and North America). Prescriptions patterns were also subject to regional differences in clinical practice.

Patient characteristics according to antithrombotic therapy use

The results suggest that patients with AF and multimorbidity prescribed NOACs are more likely to have paroxysmal AF, and have fewer comorbidities than those prescribed VKAs, consistent with other reports [13-15]. Declining renal function may influence the choice of VKA in those with chronic kidney disease. Healthcare system-related factors (such as center type) also influence treatment strategies. Patients with AF and multimorbidity treated in specialist offices and community hospitals are more often prescribed NOACs than VKAs.

The patients in this cohort prescribed antiplatelet agents had a higher risk of bleeding according to HAS-BLED score than those who were prescribed OACs. They also more often had paroxysmal AF compared to those prescribed OACs. Patients with AF and CAD were more often prescribed antiplatelets than OACs despite the fact that antiplatelet therapy does not prevent stroke or reduce mortality, elevates the risk of bleeding, and is not recommended for prevention of AF-related thromboembolism [16]. Unfortunately, antiplatelet monotherapy is still a frequent choice of prescribing physicians based on several European reports [17, 18].

Factors associated with OAC prescription in multimorbid AF patients globally

The majority of multimorbid AF patients had a high risk of stroke (CHA2DS2-VASc score ≥2) and oral anticoagulation therapy is recommended for these patients [19]. Hypertension and HF were the most prevalent risk factors for thromboembolic complications [20] and these factors and previous stroke or TIA are associated with a greater frequency of OAC prescription. Prescription of OACs was inversely associated with comorbidities that are strongly associated with elevated thromboembolic risk (eg, MI, CAD), just as conditions associated with an increased risk of bleeding (eg, previous hemorrhagic events) were associated with less frequent prescription of OACs. This is also consistent with prior reports [13] although current clinical practice guidelines recommend that patients with AF at a high risk of bleeding should generally continue anticoagulation with frequent visits and close monitoring [21]. A history of AF ablation in multimorbid AF patients was associated with more frequent OAC prescription as per guidelines [21] and consistent with other studies [22].

Younger age (≤75 years) was associated with greater OAC prescription and more frequent selection of VKAs compared to practice patterns for older patients. Several studies have suggested that increasing age is a barrier to implementing OAC use [23, 24]. Importantly, stroke risk increases with age, and the absolute benefit of OACs is clearly increased for older patients with AF [25]. In one report, when adjusted for comorbidity, age was not an important determinant of anticoagulation [26].

Multimorbid AF patients with paroxysmal or persistent AF were less often prescribed OACs in particular VKAs than those with permanent AF. NOACs should be preferred in patients with multimorbidity and polypharmacy given their lower number of drug–drug interactions compared with VKAs [27]. Ischemic stroke may occur as frequently in paroxysmal AF as in permanent AF, especially with multiple risk factors [28]. Moreover, the use of OACs should be based on stroke risk assessment according to the CHA2DS2-VASc risk score [21]. The pattern of AF seems to be related to patient profiles characterized by age, concomitant diseases, symptoms, and risk factors for stroke and bleeding [13]. Patients with higher European Heart Rhythm Association symptom scores were more often prescribed VKAs than those who were asymptomatic.

Multimorbid AF patients with a history of cardioversion were less often prescribed VKAs than those without prior cardioversion. NOACs were preferred in multimorbid AF patients after cardioversion. A similar pattern was found in another study where rhythm control strategy was associated with selection of NOAC [14].

OAC prescription in multimorbid AF patients regionally

In this study, multimorbidity influenced ATT use within particular regions. In Europe, younger patients (age <65 years) were less likely to be prescribed OACs than older patients (age ≥75 years). Multimorbid AF patients with congestive HF were more likely to be anticoagulated due to an increased risk of thromboembolism. In Europe, bleeding risk of a patient as perceived by physicians may be the reason for decreased use of anticoagulation. Patients with gastritis or duodenitis or hepatic disease are less likely to be prescribed OACs, probably because of the elevated risk of bleeding. This association has been previously noted [26]. In Asia, younger patients (age <75 years) were more likely to be prescribed OACs than older patients (age ≥75 years). Interestingly, patients with gastritis or duodenitis or a history of cancer were more likely to receive OAC than those without those diseases. In North America, younger multimorbid AF patients (age <65 years) were less likely to be prescribed OACs than older patients (age ≥75 years). Multimorbid AF patients with diabetes were more likely to receive OACs, due to their association with higher thromboembolic risk, as well as higher all-cause, cardiovascular, and noncardiovascular mortality [29]. AF patients with multimorbidity and cancer in North America were less likely to receive OAC.

Asia and North America were associated with decreased OAC prescription. In Asia, OACs are less commonly prescribed in nonvalvular AF patients than in Europe, possibly because of suspicion of the risk of bleeding during treatment [30]. Also, NOACs are not reimbursed in some Asian countries.

Strengths

It is one the largest prospective global cohort of consecutive AF patients receiving different antithrombotic treatments. Initiation of Phase III was region-specific, once relevant baseline characteristics of patients initiating dabigatran and VKA therapy in Phase II overlapped based on propensity score comparisons. After the baseline visit, all patients in this Phase III were managed according to local clinical practice and were followed for 3 years, regardless of prescribed antithrombotic therapy. This study had regular follow-up with physicians, alongside on-site monitoring, multiple standards for data quality assurance and review.

Limitations

Although the GLORIA-AF study was designed to capture all outcome events, this analysis did not consider follow-up data. The following limitations exist in our study: we have no data on patient and prescriber treatment preferences; similarly, reasons for OAC nonprescription were not reported. Furthermore, this study reflects single, initial-treatment decisions during a period when prescribing patterns may have been changing, and the analysis was based on prescription pattern shortly after entry into the registry (baseline). Neither have we accounted for quality of anticoagulation or changes in clinical practice patterns over time.

Conclusion

AF patients with multimorbidity who were prescribed NOACs were relatively healthier, more likely to have paroxysmal AF, and had fewer prevalent comorbidities than AF multimorbid patients on VKAs. Multimorbidity may determine the antithrombotic therapy prescription pattern within AF patients. Several factors are related to increased OAC prescription in multimorbid AF patients, including younger age, hypertension, prior TIA or stroke, and AF ablation. Pattern of AF (paroxysmal and persistent AF), CAD, MI, history of bleeding, and region (Asia, North America) were inversely associated with OAC prescription.

(PDF)

Click here for additional data file.

(ODT)

Click here for additional data file.

25 Jan 2021

PONE-D-20-40256

Manuscript Type: Original article

Atrial Fibrillation and Comorbidities: Clinical Characteristics and Antithrombotic Treatment in GLORIA-AF

PLOS ONE

Dear Dr. Kozieł,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

The study has been revised by two academics, with expertise in the field of the study and its statatistical methods. The methodology employed is partly correct and sound in general, as assessed by the reviewers. However, there are some specific aspects of the statistical methods used that should be revised/clarified. One point is the type of treatment that the individuals are employing. If the treatment changes, the authors should indicate how they account for this change in their model. A reorganization of the manuscript, to a certain extent, is suggested by one reviewer in the attached document, aiming at a more understandable presentation of the data. Further, the reviewer recommends including a further discussion/comparison with recent data on different cohorts.

Please submit your revised manuscript by Mar 07 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Pablo Garcia de Frutos

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for submitting your clinical trial to PLOS ONE and for providing the name of the registry and the registration number. The information in the registry entry suggests that your trial was registered after patient recruitment began. PLOS ONE strongly encourages authors to register all trials before recruiting the first participant in a study.

As per the journal’s editorial policy, please include in the Methods section of your paper:

1) your reasons for your delay in registering this study (after enrolment of participants started);

2) confirmation that all related trials are registered by stating: “The authors confirm that all ongoing and related trials for this drug/intervention are registered”.

3. Thank you for stating the following in the Competing Interests section:

"Dr Kozieł and Professor Rothman declare no conflicts of interest.

Dr Teutsch, Dr Lu, Sabrina Marler, and Venkatesh K. Gurusamy are employees of Boehringer Ingelheim.

Professor Halperin has engaged in consulting activities for Boehringer Ingelheim and advisory activities involving anticoagulants, and he is a member of the Executive Steering Committee of the GLORIA-AF Registry.

Over the past 3 years, Professor Diener received honoraria for participation in clinical trials, contribution to advisory boards, or oral presentations from: Abbott, Bayer Vital, Bristol-Myers Squibb, Boehringer Ingelheim, Daiichi Sankyo, Medtronic, Pfizer, Portola, Sanofi-Aventis, and WebMD Global. Financial support for research projects was provided by Boehringer Ingelheim. He received research grants from the German Research Council (DFG), German Ministry of Education and Research (BMBF), European Union, NIH, Bertelsmann Foundation, and Heinz-Nixdorf Foundation.

Professor Ma received honoraria from Bristol-Myers Squibb, Pfizer, Johnson & Johnson, Boehringer Ingelheim, Bayer, and AstraZeneca for giving lectures.

Professor Huisman reports grants from ZonMW Dutch Healthcare Fund, grants and personal fees from Boehringer Ingelheim, Pfizer/Bristol-Myers Squibb, Bayer Health Care, Aspen, Daiichi Sankyo, outside the submitted work.

Professor Lip has been a consultant for Bayer/Janssen, Bristol-Myers Squibb/Pfizer, Medtronic, Boehringer Ingelheim, Novartis, Verseon, and Daiichi Sankyo. He has been a speaker for Bayer, Bristol-Myers Squibb/Pfizer, Medtronic, Boehringer Ingelheim, and Daiichi Sankyo. No fees directly received personally."

We note that one or more of the authors have an affiliation to the commercial funders of this research study : Boehringer Ingelheim. We also note that one or more of the authors are employed by a commercial company: RTI Health Solutions.

3.1. Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

3.2. Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

5. Please amend either the title on the online submission form (via Edit Submission) or the title in the manuscript so that they are identical.

6. One of the noted authors is a group or consortium [GLORIA-AF Investigators]. In addition to naming the author group, please list the individual authors and affiliations within this group in the acknowledgments section of your manuscript. Please also indicate clearly a lead author for this group along with a contact email address.

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors of the paper “Atrial Fibrillation and Comorbidities: Clinical Characteristics and Antithrombotic Treatment in GLORIA-AF" have been focused on the analysis objective, specifically in baseline characteristics and antithrombotic therapy in patients with AF and more than two concomitant and chronic comorbidities.

For more information, please check the attached file.

Reviewer #2: PONE-D-20-40256: statistical review

SUMMARY. This study describes factors that are associated with oral anticoagulant (OAC) prescription in subjects with atrial fibrillation and comorbidities. I am not sure that the association between patient's characteristics and prescription makes sense as a research question. However, I'm not a medical doctor and I am going to limit my discussion to the statistical methods that are deployed here.

From a statistical viewpoint, the core of the analysis relies on a cross-sectional log-binomial regression model, estimated by likelihood-based multiple imputation methods. These methods could in principle be appropriate, but the paper lacks information about the data structure and the multiple imputation method. I therefore need to ask first some questions (major issues 1 and 2 below) before making a recommendation.

MAJOR ISSUES

1. Little is said about the structure of the response variable: OAC prescriptions. The cross-sectional log-binomial regression model assumes that the response variable is a binary variable. It therefore makes sense if each subject receive only one type of prescription during the whole follow up. In this case, the model chosen provides a correct approach. If instead subjects switch between a type of prescription to another one, then this method is no longer correct and the data must be examined by a longitudinal version of the model that includes subject-specific random effects. Please clarify.

2. The authors correctly use multiple imputation to handle missing data in a regression framework. However, nothing is said about the model used to generate the imputation sample. If all the incomplete covariates are continuous, a multivariate normal distribution is typically used for imputation. However, in this study, incomplete covariates are of mixed type (some are continuous, others are categorical). I therefore wonder what model has been used at the imputation step.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Gloria-AF registry,.pdf

Click here for additional data file.

4 Mar 2021

Reviewer #1

The authors of the paper “Atrial Fibrillation and Comorbidities: Clinical Characteristics and Antithrombotic Treatment in GLORIA-AF" have been focused on the analysis objective, specifically in baseline characteristics and antithrombotic therapy in patients with AF and more than two concomitant and chronic comorbidities. Authors have done, through Gloria-AF registry, an international covered registry, a genuine work in screening of baseline characteristics and medication of an enormous number of patients with AF during a three-year enrolling period. Even though we have the baseline picture, it is of a big interest to know the trend or changes with regards to oral anticoagulation (OAC) therapy during the follow-up (FU), but as mentioned in Limitations the authors have chosen to show the FU data probably later on. General comments: The manuscript is well written with the objective and methodology clearly described. The authors make an important point showing that patients with AF and a high-risk profile for stroke, more than 2 chronic comorbidities, were unfortunately undertreated with NOAC, in a period when at least three NOACs were established in the clinical use. According to the authors, around 60% of those patients were on NOAC regimen, but still 16% of those patients were without OAC, despite CHA2DS2-VASc ≥ 2. This work points the fact that younger patients with AF and less comorbidities were likely to were on NOAC regimen, at the same time that observations and clinical practice today show something different. In this perspective, I think it is very important the future information from the registry if the trends during the follow-up will show the impact of implementation of the current guidelines for the antithrombotic therapy in patients with AF. I have some comments I would like to address to the authors:

1. I will appreciate if the authors can explain or argue what was the reason for choosing of standardized differences in order to compare baseline characteristics between different stroke prevention strategies.

Authors’ response:

Thank you. Standardized differences were used because they are independent of sample size and it allows for the comparison of the relative balance of variables measured in different units.

2. Again, when considering the trend aspect, it is of interest if a trend of increased NOAC use was observed during the last year of inclusion as compared with the previous year(s).

Authors’ response:

Thank you. Trends of NOAC use during the last years of inclusion as compared with previous years has been addressed by a different manuscript which has been under separate development for submission.

3. In Discussion, I feel sometimes that authors give some more results rather than discuss the findings and face them with the literature, and specifically in the second paragraph.

Authors’ response:

Thank you. The Discussion has been edited and the findings have been discussed along with other studies (see pages 8-12).

4. I miss the comparison and discussion with some more recent findings regarding to NOAC use, where data from Scandinavia has given a different picture than probably other regions in Europe.

Authors’ response:

Thank you. Our findings have been compared and discussed with Scandinavian registries regarding NOAC (see pages 11 and 12).

‘Of note, younger patients (18-54 years) and ≥75 years were less likely to receive NOAC than those aged 65-74 years in Sweden (31). In Denmark, older age was associated with increased NOAC use (32).’

‘In contrast, HF was associated with decreased OAC initiation in Danish dataset (33).’

‘In Danish nationwide registries, bleeding was also associated with decreased OAC use (33).’

‘In our registry, previous TIA or stroke was the comorbidity associated with decreased VKA use in multimorbid AF patients in Europe. Interestingly, stroke/ thromboembolism or bleeding were associated with increased NOAC initiation in Denmark (32)’.

5. The tables are informative and presented well, but sometimes the perception of overflow is not avoidable. The possibility to cut some information could be considered.

Authors’ response:

Thank you. The tables have been carefully revised and some excessive information has been deleted (see Tables 2-5).

6. I would like to have in the ordinary tables one of those included in supplements, specifically table S3 and would recommend authors to change with another one. I believe that one of strengths with this analysis is the region-specific patients enrolling and comparisons.

Authors’ response:

Thank you. The Table S3 have been changed into separate tables for Asia, Europe, North America and South America (see Table S3, S4, S5 and S6).

Reviewer #2: PONE-D-20-40256: statistical review

SUMMARY. This study describes factors that are associated with oral anticoagulant (OAC) prescription in subjects with atrial fibrillation and comorbidities. I am not sure that the association between patient's characteristics and prescription makes sense as a research question. However, I'm not a medical doctor and I am going to limit my discussion to the statistical methods that are deployed here.

From a statistical viewpoint, the core of the analysis relies on a cross-sectional log-binomial regression model, estimated by likelihood-based multiple imputation methods. These methods could in principle be appropriate, but the paper lacks information about the data structure and the multiple imputation method. I therefore need to ask first some questions (major issues 1 and 2 below) before making a recommendation.

MAJOR ISSUES

1. Little is said about the structure of the response variable: OAC prescriptions. The cross-sectional log-binomial regression model assumes that the response variable is a binary variable. It therefore makes sense if each subject receive only one type of prescription during the whole follow up. In this case, the model chosen provides a correct approach. If instead subjects switch between a type of prescription to another one, then this method is no longer correct and the data must be examined by a longitudinal version of the model that includes subject-specific random effects. Please clarify.

Authors’ response:

Thank you. In this registry, there was no intervention in treatment prescription for patients over time, so patients can have more than one types of treatment prescribed during the study. The cross-sectional association analysis is based on the first prescription of antithrombotic treatment (i.e. index treatment) that was prescribed as long term use at baseline visit. Per inclusion criteria in protocol, the patients enrolled in the study must have been newly diagnosed (<3 months prior to baseline visit) with non-valvular AF. In order to clarify this definition of response variable in the manuscript, we have changed ‘OAC prescription’ to ‘baseline OAC prescription’.

2. The authors correctly use multiple imputation to handle missing data in a regression framework. However, nothing is said about the model used to generate the imputation sample. If all the incomplete covariates are continuous, a multivariate normal distribution is typically used for imputation. However, in this study, incomplete covariates are of mixed type (some are continuous, others are categorical). I therefore wonder what model has been used at the imputation step.

Authors’ response:

Thank you. The technique called multiple imputation by chained equations or fully conditional specification) was used to impute missing continuous and categorical variables. This method was well described in the reference ‘White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30(4):377-399.’. It provides the flexibility of handling different types of variables at the same time by employing different types of models accordingly. For example, ordinal linear regression is used for imputing continuous variable; logistic regression or discriminant function is used for (ordinal or nominal) categorical variables. For each of the total 56 covariates analysed, there is one imputation model specified. For this clinically orientated manuscript, the full details of this statistical analysis were not provided. However, we have added one sentence in the ‘statistical analysis’ section to explain what specific multiple imputation is used that can deal with both continuous and categorical variables (see page 6).

‘Multiple imputation by chained equations was used to impute both missing categorical and continuous values.’

Submitted filename: Rebuttal PLOS one (2)GYHLMK.kjr (1) (3).docx

Click here for additional data file.

22 Mar 2021

Manuscript Type: Original article

Atrial Fibrillation and Comorbidities: Clinical Characteristics and Antithrombotic Treatment in GLORIA-AF

PONE-D-20-40256R1

Dear Dr. Kozieł,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Pablo Garcia de Frutos

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: (No Response)

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: (No Response)

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have answered to my questions and suggestions, and I am satisfied with their comments. They have also improved the discussion and tables.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

5 Apr 2021

PONE-D-20-40256R1

Atrial Fibrillation and Comorbidities: Clinical Characteristics and Antithrombotic Treatment in GLORIA-AF

Dear Dr. Kozieł:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Pablo Garcia de Frutos

Academic Editor

PLOS ONE