Healthcare resource utilization and costs among nonvalvular atrial fibrillation patients initiating rivaroxaban or warfarin in skilled nursing facilities: a retrospective cohort study

Figures & data

Table 1. Demographic and baseline clinical characteristics.

	Unweighted			IPTW Adjusted
	Rivaroxaban (n = 519)	Warfarin (n = 1129)	Std diff^a	Rivaroxaban	Warfarin	Std diff^a
Demographics
Age, years, mean (SD)	77.1 (9.1)	77.3 (8.8)	1.7%	78.0 (8.4)	77.1 (9.0)	10.4%
Female, n (%)	307 (59.2%)	627 (55.5%)	7.3%	926 (58.0%)	939 (56.5%)	3.0%
Region, n (%)
Northeast	113 (21.8%)	301 (26.7%)	11.4%	391 (24.5%)	410 (24.7%)	0.5%
Midwest	125 (24.1%)	328 (29.1%)	11.3%	468 (29.3%)	499 (30.0%)	1.5%
South	221 (42.6%)	321 (28.4%)	29.9%	526 (32.9%)	522 (31.4%)	3.2%
West	57 (11.0%)	175 (15.5%)	13.4%	207 (13.0%)	225 (13.5%)	1.6%
Unknown	3 (0.6%)	4 (0.4%)	3.3%	4 (0.3%)	5 (0.3%)	0.9%
Plan type, n (%)
HMO	103 (19.8%)	303 (26.8%)	16.6%	366 (22.9%)	400 (24.1%)	2.7%
PPO	49 (9.4%)	135 (12.0%)	8.1%	208 (13.0%)	184 (11.1%)	6.0%
POS	32 (6.2%)	20 (1.8%)	22.7%	47 (2.9%)	72 (4.3%)	7.6%
Others	335 (64.5%)	671 (59.4%)	10.6%	976 (61.1%)	1005 (60.5%)	1.2%
Insurance type, n (%)
Commercial	49 (9.4%)	33 (2.9%)	27.3%	71 (4.4%)	95 (5.7%)	5.9%
Medicare Advantage	470 (90.6%)	1096 (97.1%)	27.3%	1526 (95.6%)	1566 (94.3%)	5.9%
Index year, n (%)
2013	33 (6.4%)	149 (13.2%)	23.2%	154 (9.7%)	183 (11.0%)	4.4%
2014	129 (24.9%)	274 (24.3%)	1.4%	427 (26.7%)	409 (24.6%)	4.8%
2015	121 (23.3%)	313 (27.7%)	10.1%	399 (25.0%)	455 (27.4%)	5.4%
2016	118 (22.7%)	219 (19.4%)	8.2%	335 (21.0%)	331 (20.0%)	2.5%
2017	118 (22.7%)	174 (15.4%)	18.7%	282 (17.6%)	283 (17.0%)	1.6%
6-month baseline comorbid conditions
Comorbidity/risk scores, mean (SD)
QCI	3.0 (2.2)	3.7 (2.3)	30.6%	3.4 (2.2)	3.4 (2.3)	1.3%
CHA2DS2-VASc score	5.1 (1.9)	5.4 (1.8)	14.6%	5.3 (1.8)	5.3 (1.8)	1.7%
HAS-BLED score	3.4 (1.5)	3.8 (1.5)	29.4%	3.7 (1.5)	3.6 (1.5)	2.7%
Comorbid conditions, n (%)
Anemia	226 (43.5%)	597 (52.9%)	18.8%	771 (48.3%)	819 (49.3%)	2.0%
Anxiety	128 (24.7%)	215 (19.0%)	13.6%	325 (20.3%)	327 (19.7%)	1.6%
Asthma	63 (12.1%)	134 (11.9%)	0.8%	195 (12.2%)	191 (11.5%)	2.1%
Cancer	47 (9.1%)	103 (9.1%)	0.2%	154 (9.6%)	146 (8.8%)	2.8%
Cardiovascular disease	371 (71.5%)	918 (81.3%)	23.3%	1268 (79.4%)	1288 (77.5%)	4.6%
Acute myocardial infarction	55 (10.6%)	173 (15.3%)	14.1%	211 (13.2%)	229 (13.8%)	1.7%
Old myocardial infarction	51 (9.8%)	143 (12.7%)	9.0%	156 (9.8%)	192 (11.5%)	5.6%
Stable angina	21 (4.0%)	55 (4.9%)	4.0%	72 (4.5%)	75 (4.5%)	0.1%
Unstable angina	21 (4.0%)	62 (5.5%)	6.8%	72 (4.5%)	83 (5.0%)	2.3%
Peripheral vascular disease	142 (27.4%)	423 (37.5%)	21.7%	571 (35.8%)	573 (34.5%)	2.7%
Congestive heart failure	257 (49.5%)	642 (56.9%)	14.8%	867 (54.3%)	903 (54.4%)	0.2%
Coronary artery disease	103 (19.8%)	327 (29.0%)	21.3%	382 (23.9%)	418 (25.1%)	2.8%
Venous thromboembolism	242 (46.6%)	553 (49.0%)	4.7%	764 (47.8%)	795 (47.9%)	0.1%
Chronic kidney disease	136 (26.2%)	471 (41.7%)	33.2%	604 (37.8%)	605 (36.4%)	2.9%
COPD	182 (35.1%)	428 (37.9%)	5.9%	584 (36.6%)	647 (38.9%)	4.9%
Depression	126 (24.3%)	294 (26.0%)	4.1%	394 (24.7%)	409 (24.6%)	0.2%
Diabetes	204 (39.3%)	550 (48.7%)	19.0%	723 (45.3%)	748 (45.0%)	0.5%
Hypertension	470 (90.6%)	1043 (92.4%)	6.5%	1449 (90.7%)	1529 (92.1%)	4.8%
Hyperlipidemia	344 (66.3%)	794 (70.3%)	8.7%	1121 (70.2%)	1129 (68.0%)	4.8%
Nonalcoholic fatty liver disease	54 (10.4%)	195 (17.3%)	20.0%	266 (16.7%)	248 (15.0%)	4.7%
Obesity	139 (26.8%)	285 (25.2%)	3.5%	430 (26.9%)	414 (24.9%)	4.6%
Osteoarthritis	227 (43.7%)	424 (37.6%)	12.6%	653 (40.9%)	651 (39.2%)	3.4%
Osteoporosis	61 (11.8%)	131 (11.6%)	0.5%	171 (10.7%)	186 (11.2%)	1.6%
Prior stroke/TIA	135 (26.0%)	348 (30.8%)	10.7%	448 (28.0%)	489 (29.4%)	3.1%
Sleep apnea	78 (15.0%)	152 (13.5%)	4.5%	267 (16.7%)	225 (13.6%)	8.8%
Thyroid disease	159 (30.6%)	334 (29.6%)	2.3%	508 (31.8%)	490 (29.5%)	5.1%
6-month baseline medication use
Number of different pharmacy prescriptions, mean (SD)	10.9 (6.5)	10.3 (6.8)	8.0%	10.2 (6.2)	10.6 (6.6)	6.5%
Medication classes, n (%)
Unfractionated heparin	0 (0.0%)	5 (0.4%)	9.4%	0 (0.0%)	5 (0.3%)	7.8%
Low-molecular weighted heparin	6 (1.2%)	31 (2.7%)	11.5%	29 (1.8%)	36 (2.1%)	2.3%
Antianxiety agents	94 (18.1%)	153 (13.6%)	12.5%	225 (14.1%)	232 (14.0%)	0.3%
Antiarrhythmics	69 (13.3%)	88 (7.8%)	18.0%	168 (10.5%)	156 (9.4%)	3.8%
Antibiotics	260 (50.1%)	503 (44.6%)	11.1%	721 (45.1%)	783 (47.1%)	4.0%
Antidepressants	161 (31.0%)	348 (30.8%)	0.4%	468 (29.3%)	499 (30.1%)	1.6%
Antidiabetics	130 (25.0%)	329 (29.1%)	9.2%	469 (29.3%)	451 (27.2%)	4.8%
Antihyperlipidemics	279 (53.8%)	593 (52.5%)	2.5%	831 (52.0%)	899 (54.1%)	4.2%
Antihypertensives	292 (56.3%)	615 (54.5%)	3.6%	837 (52.4%)	885 (53.3%)	1.7%
Antiplatelet agents	68 (13.1%)	152 (13.5%)	1.1%	223 (13.9%)	219 (13.2%)	2.2%
Antineoplastic agents	28 (5.4%)	35 (3.1%)	11.4%	70 (4.4%)	68 (4.1%)	1.3%
Anti-inflammatory agents	67 (12.9%)	161 (14.3%)	3.9%	249 (15.6%)	260 (15.7%)	0.2%
Beta-blockers	323 (62.2%)	641 (56.8%)	11.1%	944 (59.1%)	993 (59.8%)	1.4%
Calcium channel blockers	200 (38.5%)	370 (32.8%)	12.1%	573 (35.9%)	572 (34.5%)	3.0%
Cardiotonic agents	40 (7.7%)	91 (8.1%)	1.3%	125 (7.8%)	130 (7.8%)	0.1%
Corticosteroids	127 (24.5%)	205 (18.2%)	15.5%	301 (18.9%)	358 (21.5%)	6.7%
Diuretics	289 (55.7%)	604 (53.5%)	4.4%	864 (54.1%)	923 (55.5%)	2.9%
Opioids	235 (45.3%)	420 (37.2%)	16.5%	598 (37.5%)	677 (40.8%)	6.8%

Abbreviations. COPD, chronic obstructive pulmonary disease; HMO, health maintenance organization; IPTW, inverse probability of treatment weighting; POS, point of service; PPO, preferred provider organization; QCI, Quan-Charlson Comorbidity Index; Std diff, standardized difference; SD, standard deviation; TIA, transient ischemic attack.

a Standardized difference ≥10% considered significant.

Table 2. Healthcare resource utilization during the index skilled nursing facility stay.

	IPTW Adjusted			95% CI
	Rivaroxaban	Warfarin	Estimate^a	Lower	Upper	p Value
Length of the index SNF stay, mean (SD)	32.07 (41.60)	37.44 (52.20)	–5.36	–5.73	–4.98	<.0001
Index drug use during the index SNF stay
Number of fills, mean (SD)	1.62 (1.31)	2.72 (4.38)	–1.10	–1.18	–1.02	<.0001
PDC, mean (SD)	0.84 (0.22)	0.86 (0.21)	–0.02	–0.08	0.05	.6171
Healthcare resource utilization during the index SNF stay
≥1 event of interest, n (%)
Hospitalization	408 (25.6%)	531 (31.9%)	0.73	0.63	0.85	<.0001
ER visit	141 (8.8%)	164 (9.9%)	0.88	0.70	1.12	.2920
INR test	117 (7.3%)	1088 (65.5%)	0.04	0.03	0.05	<.0001
Pathology and lab encounter	902 (56.5%)	1273 (76.6%)	0.40	0.34	0.46	<.0001
PPPM number of event, mean (SD)
Hospitalization	0.67 (1.42)	0.85 (1.59)	–0.17	–0.18	–0.16	<.0001
ER visit	0.10 (0.45)	0.10 (0.44)	0.00	0.00	0.00	.9526
INR test	0.08 (0.38)	2.76 (3.04)	–2.68	–2.68	–2.67	<.0001
Pathology and lab encounter	1.70 (3.08)	3.96 (3.72)	–2.26	–2.28	–2.25	<.0001

Abbreviations. CI, confidence interval; ER, emergency room; INR, international normalized ratio; IPTW, inverse probability of treatment weighting; PDC, proportion of days covered; PPPM, per-patient-per-month; SD, standard deviation; SNF, skilled nursing facility.

a Odds ratio was reported for categorical variables, and mean difference was reported for continuous variables. Statistical comparisons are comparing rivaroxaban to warfarin (reference group).

Figure 1. Total all-cause PPPM costs during the index skilled nursing facility stay. All differences between rivaroxaban and warfarin costs were significant with p < .0001.

Table 3. Healthcare resource utilization and costs during the 30-day and 100-day post-index periods.

	IPTW Adjusted			95% CI
	Rivaroxaban	Warfarin	Estimate^a	Lower	Upper	p Value
30-day all-cause healthcare resource utilization
≥1 event of interest, n (%)
Hospitalization	628 (39.4%)	685 (41.3%)	0.92	0.80	1.06	.2691
ER visit	163 (10.2%)	160 (9.7%)	1.07	0.85	1.34	.5816
PPP30D number of event, mean (SD)
Hospitalization	0.70 (1.17)	0.80 (1.38)	–0.10	–0.11	–0.09	<.0001
ER visit	0.12 (0.38)	0.11 (0.34)	0.01	0.00	0.01	<.0001
Physician visit	2.12 (2.04)	2.57 (2.65)	–0.32	–0.52	–0.09	.0080
Other medical service	3.95 (3.65)	5.55 (4.90)	–1.48	–1.81	–1.13	<.0001
Pharmacy fill	10.07 (8.64)	10.65 (8.16)	–0.58	–0.62	–0.54	<.0001
30-day all-cause healthcare costs, mean (SD)
PPP30D total cost (medical + pharmacy)	$8103 ($11,103)	$10,481 ($18,081)	–$2396	–$3179	–$1529	<.0001
PPP30D total medical cost	$7149 ($10,981)	$9956 ($18,056)	–$2831	–$3588	–$1986	<.0001
Hospitalization	$3883 ($10,352)	$4642 ($15,306)	–$748	–$1632	$394	.1806
ER visit	$280 ($1,078)	$235 ($875)	$45	–$16	$122	.1615
Physician office	$234 ($314)	$267 ($365)	–$32	–$63	$2	.0660
Outpatient service	$1093 ($2665)	$3072 ($8,976)	–$1979	–$2147	–$1780	<.0001
Skilled nursing facility	$1660 ($1850)	$1742 ($1745)	–$82	–$227	$78	.3041
PPP30D total pharmacy cost	$954 ($1,026)	$525 ($784)	$429	$329	$541	<.0001
100-day all-cause healthcare resource utilization
≥1 event of interest, n (%)
Hospitalization	854 (53.5%)	961 (57.9%)	0.84	0.73	0.96	.0126
ER visit	274 (17.2%)	331 (19.9%)	0.83	0.70	0.99	.0427
PPP100D number of event, mean (SD)
Hospitalization	1.96 (3.95)	2.02 (3.40)	–0.06	–0.07	–0.05	<.0001
ER visit	0.27 (0.82)	0.26 (0.59)	0.02	0.01	0.02	<.0001
Physician visit	5.30 (4.35)	6.41 (5.84)	–1.07	–1.52	–0.57	<.0001
Other medical service	9.84 (8.09)	14.70 (12.11)	–4.68	–5.42	–3.88	<.0001
Pharmacy fill	23.24 (18.42)	25.97 (20.36)	–2.73	–2.76	–2.70	<.0001
100-day all-cause healthcare cost, mean (SD)
PPP100D total cost (medical + pharmacy)	$19,246 ($25,105)	$27,992 ($41,053)	–$8746	–$10,591	–$6706	<.0001
PPP100D total medical cost	$17,154 ($24,435)	$26,683 ($40,984)	–$9536	–$11,320	–$7544	<.0001
Hospitalization	$10,037 ($22,279)	$13,790 ($31,305)	–$3759	–$5912	–$1023	.0097
ER visit	$648 ($1919)	$611 ($1557)	$37	–$104	$219	.6360
Physician office	$572 ($714)	$705 ($1124)	–$133	–$197	–$60	.0006
Outpatient service	$3133 ($6625)	$8497 ($23,016)	–$5364	–$5788	–$4875	<.0001
Skilled nursing facility	$2765 ($3405)	$3082 ($3626)	–$322	–$577	–$41	.0255
PPP100D total pharmacy cost	$2092 ($2139)	$1309 ($1834)	$783	$563	$1029	<.0001

Abbreviations. CI, confidence interval; ER, emergency room; IPTW, inverse probability of treatment weighting; PPP30D, per-patient-per-30-days; PPP100D, per-patient-per-100-days; SD, standard deviation.

a Odds ratio was reported for categorical variables, and mean difference was reported for continuous variables. Statistical comparisons are comparing rivaroxaban to warfarin (reference group).

Data availability statement

The data that support the findings of this study are available from Optum but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are, however, available from the authors upon reasonable request and with permission of Optum.