Real-world reduction in healthcare resource utilization following treatment of opioid use disorder with reSET-O, a novel prescription digital therapeutic

Figures & data

Figure 1. Study design

Figure 2. Patient enrollment

Table 1. Demographic characteristics of patients who initiated re-SET-O

Demographic/characteristic	N = 351
Age on index date
Mean (SD)	37.0 (8.64)
Median	36.0
IQR	31.0–41.0
Range	20–67
Age on index date, n (%)
18–24	15 (4.3%)
25–34	134 (38.2%)
35–44	145 (41.3%)
45–54	39 (11.1%)
55–64	16 (4.6%)
65–74	2 (0.6%)
Sex, n (%)
Male	142 (40.5%)
Female	209 (59.5%)
Payer on index date, n (%)
Commercial	45 (12.8%)
Medicaid	290 (82.6%)
Medicare Advantage	8 (2.3%)
Unknown	8 (2.3%)
Opioid use disorder in the pre-index period^a, n (%)
F11.10 Opioid abuse, uncomplicated	19 (5.4%)
F11.11 Opioid abuse, in remission	5 (1.4%)
F11.14 Opioid abuse with opioid-induced mood disorder	1 (0.3%)
F11.19 with unspecified opioid-induced disorder	1 (0.3%)
F11.20 Opioid dependence, uncomplicated	290 (82.6%)
F11.21 Opioid dependence, in remission	38 (10.8%)
F11.23 Opioid dependence with withdrawal	16 (4.6%)
F11.288 Opioid dependence with other opioid-induced disorder	1 (0.3%)
F11.29 Opioid dependence with unspecified opioid-induced disorder	1 (0.3%)
F11.90 Opioid use, unspecified, uncomplicated	8 (2.3%)
F11.988 Opioid use, unspecified with other opioid-induced disorder	1 (0.3%)
F11.99 Opioid use, unspecified with unspecified opioid-induced disorder	1 (0.3%)
No opioid use diagnosis code, n (%)	46 (13.1%)
Buprenorphine treatment
Pre-Index, n (%)	256 (76.7%)
Post-Index, n (%)	243 (72.8%)
Both Pre and post-index, n (%)	240 (71.9%)

Abbreviations: IQR, interquartile range; SD, standard deviation.

Source data: HealthVerity Private Source 20 claims data from 1 October 2018 through 31 October 2019.

^aPercentages may add to more than 100% as patients claims may be indicated for different opioid use disorder indications.

Note: Age is based on index date and year of birth (using July 1st for the month and day). Index dates: 1 January 2019 through 4 October 2019.

Table 2. Buprenorphine adherence among patients with buprenorphine prescriptions pre- and post- re-SET-O initiation

Demographic/characteristic	Pre-index period N = 238	Post-index period N = 238	p-value^a
Medication possession ratio (patients treated with buprenorphine pre-and post-index)
Mean (SD)	0.757 (0.2960)	0.792 (0.2637)
Adjusted Mean (SE)	0.729 (0.0214)	0.820 (0.0214)	0.004
Median	0.9	0.9
IQR	0.6–1.0	0.7–1.0
Range	0–1	0–1
Medication possession ratio level
Level, n (%)
0.00–0.19	22 (9.2%)	13 (5.4%)	0.095
0.20–0.39	16 (6.7%)	14 (5.8%)	0.706
0.40–0.59	21 (8.8%)	24 (10.0%)	0.639
0.60–0.79	35 (14.6%)	36 (15.0%)	0.895
≥0.60–1.00	181 (75.4%)	189 (78.8%)	0.302
≥0.80–1.00	146 (60.8%)	153 (63.8%)	0.482

IQR, interquartile range; MPR, medication possession ratio; n, number of patients; SD, standard deviation.

^ap-values comparing percentages between the pre-index and post-index periods are derived from McNemar tests for categorical variables and by fitting a repeated measures model of MPR on period and follow-up time for continuous variables. The adjusted mean MPR for each period is estimated at the overall mean follow-up time across both periods. The p-value for the period parameter is reported. However, a longer follow-up period may be correlated with a lower MPR, so results should be interpreted with caution.

Note 1: Prescriptions are determined from the presence of an adjudicated claim for the medication, which may or may not have been paid.

Note 2: The mean number of days in the pre-index and post-index periods is 180.0 and 102.7 days, respectively. Index dates: 1 January 2019 through 4 October 2019.

Table 3. Incidence of facility and clinical service encounters with the largest difference 6 months pre- and post- re-SET-O initiation (ranked by magnitude of the difference)

	Pre-index period (N = 351)			Post-index period (N = 351)
	Patients n (%)	Incidence rate (95% CI)	Encounters^a n	Patients n (%)	Incidence rate (95% CI)		Encounters^a n	Difference^a n	IRR^b (95% CI)	p-value
Facility Encounters
All settings	104 (29.6%)	0.651 (0.481, 0.881)	229	48 (13.7%)		0.437 (0.318, 0.600)	153	−76	0.67 (0.45, 0.99)	0.047
Inpatient	29 (8.3%)	0.204 (0.106, 0.392)	72	13 (3.7%)		0.077 (0.042, 0.140)	27	−45	0.38 (0.16, 0.88)	0.024
Intensive care unit	3 (0.9%)	0.011 (0.004, 0.030)^c	4	0 (0.0%)		0.000 (0.000, 0.009)^c	0	−4	NA	NA
Emergency department	84 (23.9%)	0.388 (0.293, 0.514)	136	38 (10.8%)		0.310 (0.220, 0.437)	109	−27	0.80 (0.55, 1.17)	0.247
HOPD surgical	5 (1.4%)	0.014 (0.006, 0.034)^c	5	0 (0.0%)		0.000 (0.000, 0.009)^c	0	−5	NA	NA
Partial hospitalization	3 (0.9%)	0.046 (0.012, 0.172)	16	3 (0.9%)		0.057 (0.017, 0.191)	20	4	1.24 (0.20, 7.55)	0.813
Clinical Services^d
Pathology and laboratory: Drug testing	281 (80.1%)	9.470 (8.493, 10.560)	3324	239 (68.1%)		7.652 (6.775, 8.642)	2686	−638	0.81 (0.73, 0.89)	<0.001
Medicine: psychiatry	198 (56.4%)	7.595 (6.579, 8.767)	2666	189 (53.8%)		6.6 (5.697, 7.646)	2317	−349	0.87 (0.76, 0.99)	0.036
E&M: Case management services	122 (34.8%)	6.087 (4.236, 8.746)	2137	109 (31.1%)		6.589 (4.597, 9.443)	2313	176	1.08 (0.81, 1.44)	0.588
Pathology and laboratory: Other	224 (63.8%)	3.421 (2.841, 4.118)	1201	151 (43.0%)		2.948 (2.402, 3.619)	1035	−166	0.86 (0.75, 0.99)	0.039
E&M: Office/other outpatient services	310 (88.3%)	11.779 (10.652, 13.025)	4134	289 (82.3%)		11.34 (10.338, 12.439)	3980	−154	0.96 (0.90, 1.03)	0.302
Rehabilitative services: Behavioral health	54 (15.4%)	0.538 (0.304, 0.952)	189	20 (5.7%)		0.854 (0.455, 1.604)	300	111	1.59 (0.88, 2.87)	0.124
Rehabilitative services: Alcohol & substance	86 (24.5%)	1.288 (0.783, 2.120)	452	38 (10.8%)		1.015 (0.533, 1.932)	356	−96	0.79 (0.48, 1.30)	0.348
Rehabilitative services: Other	92 (26.2%)	3.255 (2.577, 4.112)	1143	70 (19.9%)		3.068 (2.399, 3.924)	1077	−66	0.94 (0.82, 1.08)	0.387
Rehabilitative services: Mental health	43 (12.3%)	0.247 (0.173, 0.354)	87	18 (5.1%)		0.42 (0.213, 0.828)	147	61	1.7 (0.91, 3.18)	0.097
Surgery	139 (39.6%)	0.78 (0.647, 0.940)	274	75 (21.4%)		0.608 (0.481, 0.769)	213	−60	0.78 (0.60, 1.02)	0.07
Medicine: physical medicine and rehabilitation	31 (8.8%)	0.469 (0.244, 0.901)	165	13 (3.7%)		0.622 (0.270, 1.435)	218	54	1.33 (0.56, 3.14)	0.521
Medicine: cardiovascular	58 (16.5%)	0.326 (0.228, 0.465)	114	24 (6.8%)		0.177 (0.117, 0.269)	62	−52	0.54 (0.36, 0.82)	0.004
E&M: Domiciliary rest home	31 (8.8%)	0.334 (0.224, 0.496)	117	26 (7.4%)		0.445 (0.304, 0.650)	156	39	1.33 (0.99, 1.79)	0.058
Medicine: chiropractic manipulative treatment	9 (2.6%)	0.124 (0.037, 0.418)	44	8 (2.3%)		0.227 (0.071, 0.727)	80	36	1.82 (0.34, 9.74)	0.481
Radiology	108 (30.8%)	0.594 (0.484, 0.730)	208	60 (17.1%)		0.493 (0.380, 0.640)	173	−35	0.83 (0.63, 1.09)	0.182
Transport services	26 (7.4%)	0.212 (0.120, 0.375)	74	10 (2.8%)		0.313 (0.125, 0.783)	110	35	1.48 (0.63, 3.46)	0.371
E&M: Emergency department services	93 (26.5%)	0.446 (0.339, 0.588)	157	43 (12.3%)		0.368 (0.266, 0.509)	129	−27	0.82 (0.57, 1.20)	0.310

CI, confidence interval; E&M, evaluation and management; IRR, incidence rate ratio; n, number of patients; NA, not applicable.

^aThe number of encounters are estimated from the incidence rates (i.e. rate x 351). Encounters difference is the estimated encounters in the post-index period minus the estimated encounters in the pre-index period. Hospital encounters (inpatient, emergency department, HOPD surgical, and partial hospitalization) are mutually exclusive.

^bIRR is the incidence in the post-index period relative to the incidence in the pre-index period (e.g. an IRR < 1 indicates lower healthcare resource use in the post-index period). Incidence and IRR are evaluated from a repeated-measures (i.e. pre and post for each patient) negative binomial model of count of encounters, with an offset for the number of days in each period.

^cBased on the crude estimate and Poisson 95% CIs. Where incidence is 0 (i.e. no events), the upper 95% CI is based on the rule of 3 (i.e. 3 ÷ 351 = 0.009).

^dClinical encounters represents groups of procedures (e.g. Current Procedure Terminology categories).

Note: The mean number of days in the pre-index and post-index periods was 180.0 and 104.0 days, respectively. The number and percent of patients are provided for description only and should not be compared given the difference in the number of days between the pre-index and post-index periods. Index dates: 1 January 2019 through 4 October 2019.

Table 4. Scenario analysis of costs related to facility and clinical encounters resource utilization

	Number of Episodes	Projected Cost of Episodes	Cost Change Per Patient
Utilization Reduced
Facility encounters	−81	-$627,868.20	-$1,788.80
Clinician encounters	−1760	-$172,779.99	-$492.25
Subtotal	−1841	-$800,648.19	-$2,281.05
Utilization Increased
Facility encounters	25	$23,903.10	$68.10
Clinician encounters	537	$22,088.37	$62.93
Subtotal	562	$45,991.47	$131.03
Net Impact	−1279	-$754,656.71	-$2,150.02

Note: Estimated six-month cost impact across all facility and clinician service categories among 351 patients.

Data availability statement

The data that support the findings of this study are available from the authors upon reasonable request and with permission from HealthVerity.