Vaso-occlusive crises and costs of sickle cell disease in patients with commercial, Medicaid, and Medicare insurance – the perspective of private and public payers

Figures & data

Figure 1. Patient selection flowchart. Abbreviations. SCD, sickle cell disease; SCT, stem cell transplant.

Table 1. Patient baseline characteristics.

	Commercial	Medicaid	Medicare
	(n = 16,092)	(n = 18,287)	(n = 15,431)
Age, mean ± SD [median]	36.7 ± 14.3 [36.0]	28.5 ± 12.0 [25.0]	48.2 ± 17.9 [45.0]
Age group, n (%)
16–17	1,550 (9.6%)	3,423 (18.7%)	0 (0.0%)
18–64	14,403 (89.5%)	14,819 (81.0%)	11,940 (77.4%)
65+	139 (0.9%)	45 (0.2%)	3,491 (22.6%)
Female, n (%)	9,879 (61.4%)	10,635 (58.2%)	9,197 (59.6%)
Year of Index date, n (%)
2001	85 (0.5%)	—	—
2002	191 (1.2%)	—	—
2003	302 (1.9%)	—	—
2004	496 (3.1%)	—	—
2005	624 (3.9%)	—	—
2006	627 (3.9%)	—	—
2007	1,054 (6.5%)	—	—
2008	1,108 (6.9%)	433 (2.4%)	—
2009	1,124 (7.0%)	5,179 (28.3%)	—
2010	1,222 (7.6%)	4,694 (25.7%)	—
2011	1,486 (9.2%)	3,651 (20.0%)	—
2012	1,640 (10.2%)	2,706 (14.8%)	177 (1.1%)
2013	1,421 (8.8%)	1,624 (8.9%)	7,157 (46.4%)
2014	1,287 (8.0%)	—	3,850 (24.9%)
2015	1,193 (7.4%)	—	4,247 (27.5%)
2016	1,306 (8.1%)	—	—
2017	915 (5.7%)	—	—
Race, n (%)
Black	—	14,632 (80.0%)	12,997 (84.2%)
Hispanic	—	734 (4.0%)	243 (1.6%)
White	—	559 (3.1%)	1,683 (10.9%)
Other	—	234 (1.3%)	330 (2.1%)
Unknown	—	2,128 (11.6%)	178 (1.2%)
Treatments,^a n (%)
Disease-modifying treatments	2,355 (14.6%)	7,730 (42.3%)	6,765 (43.8%)
Blood transfusions	1,019 (6.3%)	5,386 (29.5%)	5,289 (34.3%)
Hydroxyurea	1,555 (9.7%)	4,561 (24.9%)	3,487 (22.6%)
Blood transfusion and hydroxyurea	219 (1.4%)	2,217 (12.1%)	2,011 (13.0%)
Opioids	7,312 (45.4%)	12,920 (70.7%)	10,307 (66.8%)
Hydroxyurea adherence
MPR^b (%), mean ± SD [median]	74.9 ± 23.1 [80.6]	71.2 ± 24.9 [74.4]	73.4 ± 24.2 [78.4]
MPR^b ≥ 80%	785 (50.5%)	2,058 (45.1%)	1,701 (48.9%)

Abbreviations. MPR, medication possession ratio; SCD, sickle cell disease; SD, standard deviation.

^a Evaluated during the 12-month baseline period; numbers represent patients who ever received a given treatment during this period.

^b MPR among patients using hydroxyurea was calculated as the total number of days of supply divided by the number of days between the first and the last refill dates during the baseline period (i.e. maximum of 365 days between the first and the last refill dates). The MPR can be > 100% due to the overlapping days between two prescriptions filled; if applicable, the MPR was truncated at 100%.

Figure 2. Occurrence of SCD-related complications during the 12-month follow-up period.

Table 2. Frequency and characteristics of vaso-occlusive crises during the 12-month follow-up period.

	Commercial	Medicaid	Medicare
	(n = 16,092)	(n = 18,287)	(n = 15,431)
Annual number of VOCs per patient,
mean ± SD [median]
All patients	1.2 ± 2.6 [0.0]	3.1 ± 4.4 [1.0]	3.4 ± 5.2 [1.0]
Patients with 1+ VOCs	3.3 ± 3.4 [2.0]	4.8 ± 4.7 [3.0]	6.2 ± 5.6 [4.0]
Patients with 2+ VOCs	4.8 ± 3.7 [3.0]	6.2 ± 4.8 [5.0]	7.5 ± 5.6 [6.0]
Number of VOCs	18,535	56,400	52,531
VOC complication status, n (%)
Uncomplicated VOC	17,182 (92.7%)	52,713 (93.5%)	48,190 (91.7%)
Complicated VOC	1,353 (7.3%)	3,687 (6.5%)	4,341 (8.3%)
VOC distribution by setting, n (%)
IP	6,929 (37.4%)	25,989 (46.1%)	19,812 (37.7%)
ER	4,897 (26.4%)	19,718 (35.0%)	15,813 (30.1%)
OP	6,709 (36.2%)	10,693 (19.0%)	16,906 (32.2%)
Duration of VOC episodes by setting (days), mean ± SD [median]
IP	11.7 ± 14.4 [8.0]	15.2 ± 26.1 [7.0]	13.1 ± 19.4 [8.0]
ER	2.3 ± 2.7 [1.0]	3.0 ± 3.6 [1.0]	3.1 ± 4.6 [1.0]
OP	1.9 ± 5.4 [1.0]	2.0 ± 2.6 [1.0]	2.0 ± 2.6 [1.0]

Abbreviations. ER, emergency room; IP, inpatient; OP, outpatient; SD, standard deviation; VOC, vaso-occlusive crisis.

Table 3. Annual healthcare resource utilization of patients with SCD stratified by the annual number of VOCs during the 12-month follow-up period.

Annual number of stays/days, mean ± SD [median]	Commercial (n = 16,092)	Medicaid (n = 18,287)	Medicare (n = 15,431)
IP stays
Patients with 0 VOCs	0.2 ± 0.7 [0.0]	2.1 ± 8.8 [0.0]	0.7 ± 1.7 [0.0]
Patients with 1 VOC	0.6 ± 0.8 [0.0]	2.8 ± 7.9 [1.0]	1.0 ± 1.9 [1.0]
Patients with 2+ VOCs	2.3 ± 2.9 [2.0]	6.1 ± 9.4 [3.0]	3.9 ± 4.4 [3.0]
IP days
Patients with 0 VOCs	1.3 ± 6.9 [0.0]	18.9 ± 70.7 [0.0]	5.6 ± 16.7 [0.0]
Patients with 1 VOC	4.0 ± 9.2 [0.0]	28.2 ± 82.1 [3.0]	9.2 ± 27.7 [3.0]
Patients with 2+ VOCs	16.0 ± 25.4 [8.0]	49.1 ± 85.3 [17.0]	29.3 ± 39.3 [15.0]
ER days
Patients with 0 VOCs	0.8 ± 2.3 [0.0]	2.2 ± 5.5 [0.0]	1.5 ± 5.1 [0.0]
Patients with 1 VOC	1.3 ± 3.1 [1.0]	3.9 ± 8.7 [2.0]	1.9 ± 5.9 [1.0]
Patients with 2+ VOCs	3.4 ± 7.5 [2.0]	10.7 ± 17.0 [6.0]	7.5 ± 16.9 [3.0]
Days with OP services
Patients with 0 VOCs	12.5 ± 18.3 [7.0]	20.2 ± 24.0 [16.0]	19.2 ± 20.3 [13.0]
Patients with 1 VOC	13.7 ± 18.6 [8.0]	22.8 ± 23.0 [18.0]	18.8 ± 19.1 [13.0]
Patients with 2+ VOCs	20.3 ± 22.6 [13.0]	27.7 ± 24.1 [22.0]	22.8 ± 20.4 [18.0]

Abbreviations. ER, emergency room; IP, inpatient; OP, outpatient; SD, standard deviation; VOC, vaso-occlusive crisis.

Figure 3. Mean all-cause healthcare costs and inpatient costs per patient stratified by number of VOCs during the 12-month follow-up period. Abbreviations. SCD, sickle cell disease; USD, United States dollars; VOC, vaso-occlusive crisis.

Figure 4. Mean all-cause and SCD-related healthcare costs per patient stratified by number of VOCs during the 12-month follow-up period. Abbreviations. SCD, sickle cell disease; USD, United States dollars; VOC, vaso-occlusive crisis.

Table 4. Annual healthcare costs of patients with SCD stratified by the annual number of VOCs during the 12-month follow-up period.

Annual healthcare costs per patient	Commercial (n = 16,092)		Medicaid^a (n = 9,172)		Medicare (n = 15,431)
	Mean ± SD	Median [min-max]	Mean ± SD	Median [min-max]	Mean ± SD	Median [min-max]
Total all-cause costs (medical and pharmacy costs)	$27,721 ± $72,891	$5,293 [$0–$2,915,499]	$42,081 ± $76,690	$15,027 [$0–$2,482,826]	$38,639 ± $53,097	$19,032 [$0–$1,002,964]
0 VOCs	$15,747 ± $56,074	$2,437 [$0–$1,954,987]	$16,750 ± $42,961	$2,369 [$0–$653,454]	$21,877 ± $39,877	$6,074 [$0–$382,674]
1 VOC	$27,194 ± $83,792	$9,284 [$0–$2,915,499]	$29,880 ± $96,373	$9,046 [$0–$2,482,826]	$29,250 ± $51,509	$12,763 [$0–$1,002,964]
2+ VOCs	$64,555 ± $94,884	$33,009 [$0–$1,461,301]	$64,566 ± $81,973	$36,381 [$0–$906,907]	$58,308 ± $58,549	$40,161 [$0–$658,844]
IP costs
0 VOCs	$4,883 ± $27,822	$0 [$0–$988,183]	$6,237 ± $29,018	$0 [$0–$601,630]	$10,471 ± $29,227	$0 [$0–$347,314]
1 VOC	$14,442 ± $69,302	$0 [$0–$2,884,538]	$19,202 ± $92,538	$4,242 [$0–$2,482,159]	$16,063 ± $40,564	$6,262 [$0–$877,839]
2+ VOCs	$42,192 ± $74,972	$18,272 [$0–$1,435,751]	$47,097 ± $74,196	$22,790 [$0–$888,456]	$39,331 ± $48,169	$24,729 [$0–$627,687]
ER costs
0 VOCs	$1,049 ± $5,245	$0 [$0–$316,212]	$574 ± $1,994	$0 [$0–$35,929]	$1,323 ± $4,567	$0 [$0–$136,850]
1 VOC	$1,947 ± $6,801	$334 [$0–$194,471]	$1,139 ± $3,230	$448 [$0–$82,540]	$1,601 ± $5,456	$349 [$0–$119,002]
2+ VOCs	$5,084 ± $11,921	$1,746 [$0–$182,725]	$3,921 ± $8,369	$1,430 [$0–$114,278]	$5,780 ± $12,897	$1,974 [$0–$304,220]
OP costs
0 VOCs	$7,349 ± $34,631	$1,169 [$0–$1,721,538]	$3,814 ± $12,621	$894 [$0–$311,579]	$8,107 ± $16,011	$2,248 [$0–$198,101]
1 VOC	$8,687 ± $38,206	$1,515 [$0–$1,010,121]	$4,085 ± $9,640	$1,695 [$0–$174,104]	$8,347 ± $17,408	$2,308 [$0–$194,304]
2+ VOCs	$12,806 ± $35,384	$3,395 [$0–$909,055]	$4,883 ± $9,483	$2,467 [$0–$195,257]	$9,206 ± $15,574	$3,489 [$0–$318,627]
SCD-related costs
0 VOCs	$8,885 ± $42,446	$295 [$0–$1,715,167]	$10,203 ± $33,484	$433 [$0–$621,269]	$16,387 ± $36,323	$1,728 [$0–$379,180]
1 VOC	$21,323 ± $76,847	$6,926 [$0–$2,909,819]	$22,066 ± $69,170	$5,691 [$0–$1,274,534]	$24,685 ± $47,474	$10,214 [$0–$892,460]
2+ VOCs	$60,624 ± $92,403	$29,799 [$0–$1,460,332]	$59,561 ± $79,956	$31,826 [$0–$906,790]	$55,595 ± $57,477	$37,970 [$0–$650,730]

Abbreviations. ER, emergency room; IP, inpatient; OP, outpatient; SCD, sickle cell disease; SD, standard deviation; VOC, vaso-occlusive crisis.

^aHealthcare costs were assessed only among fee-for-service plan members for Medicaid-insured patients.

Data availability statement

Data used for this study are from IBM (used under license for the current study) and Centers for Medicare and Medicaid Services (CMS; administrative permissions were required to access the raw data under a data user agreement requesting HIPAA waiver and IRB exemption). Therefore, restrictions apply to the availability of these data, which are not publicly available.