The burden of undertreatment and non-treatment among patients with non-valvular atrial fibrillation and elevated stroke risk: a systematic review

Figures & data

Figure 1. Attrition chart using PRISMA criteria. Abbreviations. AF, atrial fibrillation; CADTH, Canadian Agency for Drugs and Technologies in Health; HTA, health technology assessment; ICER, Institute for Clinical and Economic Review; NICE, National Institute for Health and Care Excellence; PRISMA, Preferred Reporting Items for Systematic Review and Meta-Analyses.

Table 1. Summary table of studies of the burden of nontreatment and undertreatment in NVAF.

Study	Country	Study type and data source	Study population size	Study follow-up (months)	Baseline patient characteristics		Anticoagulation treatment rates	Impact of treatment on clinical outcomes	Impact of treatment on healthcare resource utilization	Impact of treatment on healthcare costs
Apenteng 2018	United Kingdom	Study of GARFIELD-AF registry, 2011–2016	3482	Mean = 24	CHF	7.2%	High-risk population: 47.9% treated with warfarin, 20.3% treated with DOAC, 20.7% undertreated, 11.1% untreated	Not reported	Not reported	Not reported
					Hypertension	69.6%
					Diabetes	17.6%
					Vascular disease	13.6%
					Stroke or TIA	12.3%
Casciano 2013	United States	Retrospective cohort study of Thomson Reuters MarketScan, 2003–2007	13,289	Mean = 15.6	CHF	23.8%	High-risk population: 52.9% treated with anticoagulants, 47.1% untreated	Incidence of ischemic stroke events per 100 person-years: Untreated: 4.41 Low adherence to warfarin: 1.87 High adherence to warfarin: 0.72 Adherence to  warfarin was defined by  proportion of days covered (PDC) by warfarin: >0.8 (high) and ≤0.8 (low)	High adherence warfarin treatment was associated with 7% fewer inpatient encounters (p < .001), 16% fewer ER visits (p = .019), and nearly a 40% reduction in the number of hospital days (p < .001) compared with patients who did not receive warfarin Inpatient and ER services utilization were not significantly different between patients with low warfarin adherence and patients who did not receive warfarin	Total costs by warfarin adherence (costs presented for inpatient, outpatient, medical cost, etc.): No warfarin = $1344 Low adherence = $1384 High adherence = $1165 Adherence to  warfarin was defined by proportion of days covered (PDC) by warfarin: > 0.8 (high) and ≤ 0.8 (low)
					Hypertension	75.1%
					Diabetes	34.8%
					Vascular disease (PAD)	11.1%
					Stroke or TIA	10.9%
de Andrés-Nogales 2015	Spain	Retrospective cohort study of the Real Evidence of Anti Coagulation Treatment in AF (REACT-AF) cohort, 2010–2012	12,514	Not reported	CHF	20.0%	Overall population: 60.1% treated with warfarin, 1.3% treated with DOACs, 30.9% undertreated, 7.7% untreated.	Bleeding incidence rates per 100 person-years: DOAC: 0.6 (95% CI: 0.0–3.4) Antiplatelets: 2.4 (95% CI: 2.1–2.7) VKAs: 2.4 (95% CI: 2.0–2.8) Untreated: 1.7 (95% CI: 1.3–2.3) Ischemic stroke  incidence rates  per 100  person-years: Untreated : 3.1; 95% CI: 2.67–3.55 Antiplatelets: 0.6; 95% CI: 0.49–0.77 DOACs: 0.0; 95% CI: 0.0–10.4 VKAs: 0.3; 95%CI: 0.22–0.35	Not reported	Not reported
					Hypertension	Not reported
					Diabetes	26.1%
					Vascular disease	7.5%
					Stroke or TIA	6.9%
Frewen 2013	Ireland	Prospective cohort study of Irish Longitudinal Study on Ageing, 2009–2011	118	Not reported	Not reported		High-risk population: 40.7% treated with anticoagulants, 34.6% undertreated, 24.7% untreated	Not reported	Not reported	Not reported
Hess 2017	United States	Study of ORBIT-AF registry, 2010–2011	9553	Median = 30	CHF	35.6%	High-risk population: 78.7% treated with warfarin, 21.3% untreated	- Patients not receiving OAC with CHADS-VASc ≥ 2 were associated with higher adjusted rates of death (OR = 1.22), stroke/non-central nervous system embolism (OR = 1.14), stroke/non-central nervous system embolism/TIA (OR = 1.18), and composite outcome of stroke and non-central nervous system embolism (OR = 1.19), and lower bleeding rates (OR = 0.35) [ORs adjusted for propensity to receive OAC]	Not reported	Not reported
					Hypertension	87.3%
					Diabetes	32.2%
					Vascular disease (MI + CAD)	56.9%
					Stroke or TIA	16.4%
Holt 2013	United Kingdom	Retrospective survey study of QResearch database, 2007–2010	59,804	N/A	Not reported		High-risk population: 50.7% treated with warfarin, 35.9% undertreated, 13.4% untreated	Not reported	Not reported	Not reported
Huisman 2017	Multiple	Study of GLORIA-AF registry, 2011–2014	15,641	Mean = 24	CHF	24.2%	High-risk population: 33.3% treated with warfarin, 48.9% treated with DOAC, 10.0% undertreated, 7.8% untreated	Not reported	Not reported	Not reported
					Hypertension	74.6%
					Diabetes	23.1%
					Vascular disease	20.3%
					Stroke or TIA	10.5%
Inoue 2014	Japan	Study of J-RHYTHM registry, 2009–2011	7516	Mean = 24	CHF	32.2%	Overall population: 86% treated with warfarin, 8% undertreated, 6% untreated	Warfarin and AP use were associated with lower rates of thromboembolism (ORs = 0.36 and 0.92) and higher rates of major hemorrhaging (ORs = 2.35 and 1.34) and all-cause mortality (ORs = 0.50 and 0.94) as compared to untreated patients	Not reported	Not reported
					Hypertension	60.4%
					Diabetes	18.3%
					Vascular disease	10.5%
					Stroke or TIA	13.8%
Jolliffe 2016	New Zealand	Retrospective cohort study of Wellington Regional Hospital, 2012–2012	751	Not reported	CHF	40.3%	High-risk population: 31.6% treated with warfarin, 45.1% undertreated, 23.3% untreated	Not reported	Not reported	Not reported
					Hypertension	68.4%
					Diabetes	22.3%
					Vascular disease	32.3%
					Stroke or TIA	24.9%
LaMori 2013	United States	Retrospective survey study of National Health and Wellness Survey, 2009–2009	1297	Not reported	CHF	19.0%	High-risk population: 48.9% treated with warfarin, 51.1% untreated	Not reported	Not reported	Not reported
					Hypertension	66.0%
					Diabetes	29.0%
					Vascular disease	Not reported
					Stroke or TIA	13.0%
Piccini 2013	United States	Retrospective cohort study of Medicare 5% database, 2006–2007	45,711	Not reported	CHF	52.4%	High-risk population: 57.5% treated with warfarin, 42.5% untreated	Not reported	Not reported	Not reported
					Hypertension	87.9%
					Diabetes	63.5%
					Vascular disease	Not reported
					Stroke or TIA	Not reported
Proietti 2018	Europe	Study of EURObservational Research Programme-Atrial Fibrillation (EORP-AF) Pilot registry, 2012–2014	1990	Mean = 24	Not reported		High-risk population: 69.4% treated with warfarin, 13.2% treated with DOAC, 16.4% untreated	Not reported	-Proportion of patients experiencing readmissions and for what reason are stratified by AF subtype -AF (12.0%), other CV events (8.2%) and non-CV events (9.7%) were the most common reasons for readmission	Not reported
Son 2017	South Korea	Retrospective cohort study of National Health Insurance Services database, 2007–2013	10,654	Mean = 38	CHF	25.5%	High-risk population: 13.6% treated with warfarin, 43.5% undertreated, 43.0% untreated	Not reported	Not reported	Not reported
					Hypertension	69.9%
					Diabetes	36.9%
					Vascular disease	10.1%
					Stroke or TIA	Not reported
Hsu 2016	United States	Study of PINNACLE registry, 2008–2012	294,642	12	CHF	28.6%	High-risk population: 54.2% treated with warfarin, 5.6% treated with DOAC, 40.2% untreated	Not reported	Not reported	Not reported
					Hypertension	84.4%
					Diabetes	26.1%
					Vascular disease (PAD + CAD)	66.2%
					Stroke or TIA	15.8%
Hsu 2016	United States	Study of PINNACLE registry, 2008–2012	429,417	Not reported	CHF	24.9%	High-risk population: 42.3% treated with warfarin, 4.3% treated with DOAC, 31.2% undertreated, 22.2% untreated	Not reported	Not reported	Not reported
					Hypertension	76.8%
					Diabetes	22.9%
					Vascular disease (PAD + MI)	26.4%
					Stroke or TIA	14.3%
Marzec 2017	United States	Study of PINNACLE registry, 2008–2014	655,000	Not reported	CHF	32.4%	High-risk population: 38.1% treated with warfarin, 16.1% treated with DOAC, 6.9% undertreated, 39.0% untreated	Not reported	Not reported	Not reported
					Hypertension	86.6%
					Diabetes	24.8%
					Vascular disease	28.8%
					Stroke or TIA	8.5%

Abbreviations. CHF, congestive heart failure; CI, confidence interval; DOAC, direct oral anticoagulant; MI, myocardial infarction; NVAF, non-valvular atrial fibrillation; OAC, oral anticoagulant; OR, odds ratio; PAD, peripheral arterial disease; TIA, transient ischemic attack; VKA, vitamin K antagonist. Percentages of treatment shares may not add up to 100% due to rounding of values.

Figure 2. Treatment rates by class among high-risk NVAF patients, pre-DOAC era. Abbreviations. DOAC, direct oral anticoagulants; NVAF, non-valvular atrial fibrillation.

Figure 3. Treatment rates by class among high-risk NVAF patients, DOAC era. Abbreviations. DOAC, direct oral anticoagulants; NVAF, non-valvular atrial fibrillation.