integrated care increases treatment and improves outcomes of patients with chronic hepatitis c virus...
DESCRIPTION
Integrated care could increase the proportion of patients with chronic HCV infection who receive antiviral treatment and achieve a sustained virologic response (SVR).TRANSCRIPT
Q1
Q38
Q2
Q6
Q7
Q8Q9Q10
Clinical Gastroenterology and Hepatology 2015;-:-–-
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
All studies published in Clinical Gastroenterology and Hepatology are embargoed until 3PM ET of the day they are published as correctedproofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.
59
60
61
62
63
64
65
66
67
Integrated Care Increases Treatment and Improves Outcomesof Patients With Chronic Hepatitis C Virus Infection andPsychiatric Illness or Substance Abuse
68
69
70
71
72
73
74
75
76
Samuel B. Ho,*,‡ Norbert Bräu,§,k Ramsey Cheung,¶,# Lin Liu,*,‡ Courtney Sanchez,*Marisa Sklar,*,‡ Tyler E. Phelps,¶ Sonja G. Marcus,§ Michelene M. Wasil,* Amelia Tisi,§
Lia Huynh,¶ Shannon K. Robinson,*,‡ Allen Gifford,**,‡‡ Steven M. Asch,¶,# andErik J. Groessl*,‡
*VA San Diego Healthcare System, San Diego, California; ‡University of California, San Diego, San Diego, California; §James J.Peters VA Medical Center, Bronx, New York; kIcahn School of Medicine at Mount Sinai, New York, New York; ¶VA Palo AltoHealthcare System, Palo Alto, California; #Stanford University, Stanford, California; **Edith Nourse Rogers Memorial VeteransHospital, Bedford, Massachusetts; ‡‡Boston University, Boston, Massachusetts
77
78
BACKGROUND & AIMS:F
79
80
81
82
83
84
Patients with hepatitis C virus (HCV) infection with psychiatric disorders and/or substanceabuse face significant barriers to antiviral treatment. New strategies might be needed toimprove treatment rates and outcomes. We investigated whether an integrated care (IC) pro-tocol, which includes multidisciplinary care coordination and patient case management, couldincrease the proportion of patients with chronic HCV infection who receive antiviral treatment(a combination of interferon-based and direct-acting antiviral agents) and achieve a sustainedvirologic response (SVR).
85
86
METHODS:87
88
89
90
91
We performed a prospective randomized trial at 3 medical centers in the United States. Par-ticipants (n [ 363 patients attending HCV clinics) had been screened and tested positive fordepression, post-traumatic stress disorder, and/or substance use; they were assigned randomly(1:1) to groups that received IC or usual care (controls) from March 2009 through February2011. A midlevel mental health practitioner was placed at each HCV clinic to provide IC withbrief mental health interventions and case management, according to formal protocol. Theprimary end point was SVR.
92
93
RESULTS:94
95
96
97
98
99
100
Of the study participants, 63% were non-white, 51% were homeless in the past 5 years, 64%had psychiatric illness, 65% were substance abusers within 1 year before enrollment, 57%were at risk for post-traumatic stress disorder, 71% had active depression, 80% were infectedwith HCV genotype 1, and 23% had advanced fibrosis. Over a mean follow-up period of28 months, a greater proportion of patients in the IC group began receiving antiviral therapy(31.9% vs 18.8% for controls; P [ .005) and achieved a SVR (15.9% vs 7.7% of controls; oddsratio, 2.26; 95% confidence interval, 1.15L4.44; P [ .018). There were no differences inserious adverse events between groups.
101
102
CONCLUSIONS:103
104
Integrated care increases the proportion of patients with HCV infection and psychiatric illnessand/or substance abuse who begin antiviral therapy and achieve SVRs, without serious adverseevents. ClinicalTrials.gov # NCT00722423.
105
106
Keywords: Care Integration; Hepatitis C; Substance Use Disorders; PTSD.107
Abbreviations used in this paper: DAA, direct-acting antiviral; HCV, hep-atitis C virus; HIV, human immunodeficiency virus; HR, hazard ratio; IC,integrated care; MHP, mental health provider; OR, odds ratio; SUD, sub-stance use disorder; SVR, sustained virologic response; UC, usual care;VA, Veterans Affairs.
© 2015 by the AGA Institute1542-3565/$36.00
http://dx.doi.org/10.1016/j.cgh.2015.02.022
108
109
110
111
112
113
114
115
116
Approximately 1.8% of the US population haschronic hepatitis C. The current estimated prev-
alence of hepatitis C virus (HCV) infection among Vet-erans Affairs (VA) patients is 8.4%, and the prevalence inthe 1945 to 1965 birth cohort is 13.5%.1 Antiviraltreatment has been shown to eradicate HCV, resulting inreduced complications and mortality from liver dis-ease.2–4 Numerous studies have indicated that HCV
LA 5.2.0 DTD � YJCGH54181_proof � 19 March 2015 � 1:42 pm � ce DVC
Q11
Q12
Q13
Q14
Q15Q16
2 Ho et al Clinical Gastroenterology and Hepatology Vol. -, No. -
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
antiviral treatment is cost effective, even with newdirect-acting antiviral (DAA) medications, for all but afew subsets of HCV patients.5–7 Despite these data, todate, only a minority of HCV patients have receivedantiviral treatment. Cumulative data from the VA HCVRegistry indicate that the percentage of VA patients withHCV who have ever received HCV antiviral therapyincreased from 10.9% in 2004 to 14.4% in 2007 and to23% in 2013.8 In the general US population, an estimated7% to 11% of HCV patients have had antiviral treat-ment.9 Without an expansion in treatment rates, pro-jections suggest an increasing HCV burden from theprogression of cirrhosis and the development of hepa-tocellular carcinoma and liver failure.10
Within this past year, antiviral treatment of HCV hasevolved from pegylated interferon and ribavirin, topegylated interferon and ribavirin with DAAs, to inter-feron-free DAA combinations. This has been accompaniedby greatly improved efficacy and reduced treatment-related side effects. Despite these improvements, a largepercentage of HCV patients may be considered poortreatment candidates because of psychiatric comorbidityand/or substance use disorders (SUD). These comorbid-ities are common among HCV patients and have been themost frequently cited reasons for withholding antiviraltherapy in the past.9,11,12 Recent data from one VAmedicalcenter indicated that 45% of current HCV patients arepoor candidates for interferon-free treatment based onactive psychiatric/SUD comorbidity, and Medicaidcurrently precludes patients with active SUD fromreceiving interferon-free medications in many states.13,14
Integrated care (IC) refers to health care in which awide variety of services are brought together to addressinter-related health problems, and maximize patientcompliance and outcomes. IC models have been effectivein improving process measures and outcomes for treat-ing psychiatric illness and substance use in primary careclinics and for improving treatment in acquired immunedeficiency syndrome clinics.15–17 To date, there are nostudies of IC protocols for increasing HCV treatmentrates or viral outcomes.
Our objective was to determine if an IC protocol couldincrease sustained virologic response (SVR) and treat-ment rates among chronic HCV patients at risk for psy-chiatric and substance use comorbidities at 3 VA MedicalCenters.
221
222
223
224
225
226
227
228
229
230
231
232
Materials and Methods
Design Overview
A detailed description of the study methods waspublished in 2013.18 The study was conducted at 3diverse VA medical centers with established HCV clinicsstaffed by experienced physicians (VA San Diego, VA PaloAlto, Bronx, VA). Patients attending these HCV clinicswere screened and recruited from March 2009 through
FLA 5.2.0 DTD � YJCGH54181_proof �
February 2011 (Figure 1). Consented patients wererandomized at each site 1:1 using random assignmentsoftware administered by the central site. The blocked,stratified, randomization sequence was concealed fromresearch staff.
A data safety and monitoring board oversaw trialprogress including enrollment, study outcomes, andserious adverse events. The study protocol is registeredwith ClinicalTrials.gov (#NCT00722423). The study wasapproved by the institutional review boards and eachinstitution and all co-authors had access to the studydata and reviewed and approved the final manuscript.
Study Participants
Study participants were VA patients with confirmedactive HCV infection (HCV polymerase chain reactionpositive) with substance use or psychiatric risk factors forantiviral treatment. All patients attending VA HCV clinicsroutinely received a standardized screening form as partof their clinical care, consisting of a Beck Depression In-ventory, screening questionnaires for drug use, alcoholuse (AUDIT-C), and post-traumatic stress disorder,as described previously18 and as outlined in theSupplementary Materials and Methods section. Humanimmunodeficiency virus (HIV)/HCV co-infected patientswere eligible at the Bronx site only to provide preliminarydata on the benefits of the IC model for these patients.Exclusion criteria included non–HCV-related liver disease(except co-existing alcoholic liver disease), decom-pensated cirrhosis, or other significant life-threateningdiseases (known malignancies and any incapacitatinglung, cardiac, renal, or autoimmune medical disease).Ineligible patients also included treatment-experiencedpatients who were not considered re-treatment candi-dates (eg, previous treatment nonresponders or withsignificant adverse events). The definition of “homeless”refers to patients who were considered homeless within5 years of the recruitment date; specific dates of beinghomeless without accessing available temporary serviceswere not collected.
Intervention
Usual care. Patients randomized to usual care (UC)received the standard of care required for HCV patientsconsistent with current VA treatment guidelines. EachHCV clinic had gastroenterology or infectious diseasephysicians working with clinical nursing or midlevelproviders and a clinic psychiatrist or psychologist. Pa-tients were either managed within the HCV clinic orreferred to standard mental health and substance useclinics for further assessment and treatment as indicatedby the severity of the risk factors. Mental health careprovided by the HCV and non-HCV clinics did not followa specific protocol and varied in accordance to thestandard of care at those clinics.
19 March 2015 � 1:42 pm � ce DVC
Q17
Figure 1. Patient enroll-ment and randomization.All HCV clinic patientswere screened as part ofstandard clinical care.PCLQ26 , __________________.
- 2015 Integrated Care for Hepatitis C 3
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
Integrated care. The IC intervention was deliveredaccording to a protocol manual by a midlevel mentalhealth provider (MHP) locatedwithin each HCV clinic. Theprotocol included brief psychological interventions andcase management provided in collaboration with clinicphysicians, nurses, and othermental health providers. TheMHP evaluated study participants and provided ongoinginterventions designed to treat specific mental healthproblems. The MHP also facilitated a complete treatmentevaluation, encouraged the initiation of antiviral treat-ment, and served as a regular contact and case manager.Further details of the IC case management protocol areavailable in the Supplementary Materials and Methodssection and as described previously.18
Antiviral treatment. Physicians offered antiviraltreatment to all patients in both study arms following
FLA 5.2.0 DTD � YJCGH54181_proof �
recommendations and criteria in published professionalorganization and VA hepatitis C treatment guidelines.19
These guidelines were applied at each site, and speci-fied that patients should show stable psychiatric disease,compliance with treatment recommendations, and so-briety from substance use for a period of time asestablished in each clinic. Patients initiating antiviraltreatment were monitored using standard protocols. Topromote inclusiveness and generalizability, the specifictype of antiviral treatment was not specified in the studyprotocol, and was left to the discretion of the HCV clinicalteam. The standard of care for HCV treatment at studyinitiation was pegylated interferon alfa and ribavirin, buttreatment was limited at all sites in late 2010 in antici-pation of new therapies. DAA therapies, adding boce-previr or telaprevir to pegylated interferon alfa and
19 March 2015 � 1:42 pm � ce DVC
Q18
Q19
Q20
4 Ho et al Clinical Gastroenterology and Hepatology Vol. -, No. -
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
ribavirin, were approved by the Food and Drug Admin-istration and available in mid- to late 2011. Patients weremonitored for significant adverse events that resulted inearly termination of treatment.
Outcomes and Follow-Up Evaluation
The primary outcome for the study was SVR, deter-mined by viral tests completed either 12 or 24 weeksafter the termination of therapy, because either of thesetime frames currently are accepted as standard ofcare.20,21 The main secondary outcomes were interferon-based treatment initiation and completion of prescribedtreatment (range, 0%–100%). Treatment data wereabstracted from medical records and included type anddoses of medications initiated, planned treatment dura-tion, and final treatment duration attained. Abstractionwas conducted by trained research staff at each site, andthen reviewed and audited by the data manager at thecentral site.
Other secondary outcomes included serious adverseevents and health care utilization. Serious adverse eventswere defined as any hospitalization, emergency roomvisit, and/or death. All patients were followed up fromtreatment initiation through July 2012, at which time theintervention ended. Treatment completion outcomeswere followed up through May 2013, and the primaryoutcome of SVR was followed up until August 2013.
Statistical Analysis
Sample size determinationwas performed as indicatedin the Supplementary Materials and Methods section, andan enrollment of 360 patients was targeted to account forattrition. An intent-to-treat analysis was performed for allclinical outcomes. Descriptive statistics were used tosummarize baseline characteristics. Univariate andmultivariate analyseswere used to assess the primary andsecondary outcomes. See the Supplementary Materialsand Methods section for further details of the statisticalanalysis.
Results
Study Participants
A total of 1627 patients attending 3 HCV clinics wereevaluated; 966 patients were eligible for the study andscreened for psychiatric and/or substance use risk fac-tors as part of standard clinic care. Of these, 755 (78%)had a positive screen and 209 (22%) screened negativefor risk factors. Of the screen-positive patients, 378 pa-tients provided informed consent and 364 patientscompleted a baseline evaluation and were randomized(Figure 1). One patient was enrolled in error and with-drawn, leaving 182 patients in the IC and 181 patients inthe UC arm. Patients were enrolled over 22 months, and
FLA 5.2.0 DTD � YJCGH54181_proof �
the mean patient follow-up period across all sites was28.1 months (SD, 5.53 mo).
The baseline characteristics for study participants arelisted in Table 1, and were similar in the IC and UCgroups. Participants were 63.5% non-white and had ahigh frequency of known barriers to access (88.7% wereunemployed or disabled, 51.1% were homeless withinthe prior 5 years, 63.9% had a psychiatric illness, and64.5% had active drug use within 1 year and/or activealcohol abuse based on a positive AUDIT-C score Themean Beck Depression Inventory score was 15.5, and70.7% met the criteria for depression at enrollment. TheUC group has a higher percentage of married and sepa-rated patients. There were no significant differences inany other baseline characteristic.
Sustained Viral Response
The number of patients with SVR was 2-fold greaterin the IC group (29 patients; 15.9%) compared with theUC group (14 patients; 7.7%) and patients receiving ICwere more likely to have an SVR (odds ratio [OR], 2.26;P ¼ .018) in univariate analysis. The simple logisticregression between each baseline characteristic and SVRis shown in Table 2. The multivariate model showed thatpatients receiving IC were more likely to have an SVRthan the UC group independent of the effects of genotypeand study site (OR, 2.26; P ¼ .022) (Table 3). Primarygenotype (OR, 2.20; P ¼ .033 for genotypes 2, 3, 4 vsgenotype 1), prior psychiatric disorder (OR, 0.44;P ¼ .017 yes vs no), and active drug use (OR, 0.47;P ¼ .034 for yes vs no) also were associated significantlywith SVR. By adding site to the fitted model, the inter-vention effect stayed similar and site was not associatedsignificantly with the SVR. Of the 42 patients with HIV/HCV co-infection, 6 patients (25%) initiated treatmentand 3 patients (12.5%) achieved SVR in the IC arm,compared with 1 patient (5.6%) who initiated treatmentand 0 patients with SVR in the UC arm (P ¼ .21 and .25,respectively). An evaluation of the subgroups of patientswith and without active drug and alcohol abuse atbaseline is presented in the Supplementary Materials andMethods section.
Time to Treatment Initiation
Patients in the IC arm were more likely to initiatetreatment over time (Figure 2A). The overall treatmentinitiation rate in the IC arm was 58 of 182 (31.9%)compared with 34 of 181 (18.8%) in the UC arm (P ¼.0054). The log-rank test showed that the time to treat-ment initiation was significantly different between the ICand UC groups (P ¼ .003) (Figure 2B).
The multivariate Cox regression model showed thatpatients treated with IC started treatment earlier thanpatients treated with UC (hazard ratio [HR], 2.01; P ¼.002), and the rate of starting treatment was increased
19 March 2015 � 1:42 pm � ce DVC
Table 1. Patient Characteristics
Total (n ¼ 363),n (%)
Integrated care (n ¼ 182),n (%)
Usual care (n ¼ 181),n (%) P value
DemographicsMean age, y (SD) 55.4 (5.65) 55.3 (5.51) 55.5 (5.79) .47Mean BMI (SD) 27.7 (4.83) 27.5 (5.11) 27.8 (4.54) .43SexMale 355 (97.8) 178 (97.8) 177 (97.8) 1.00Female 8 (2.2) 4 (2.2) 4 (2.2)
Race/ethnicityAfrican American or black 141 (39.3) 66 (36.7) 75 (41.9) .65White (non-Hispanic) 131 (36.5) 69 (38.3) 62 (34.6)Hispanic 65 (18.1) 32 (17.8) 33 (18.4)Others 22 (6.1) 13 (7.2) 9 (5)
Marital statusSingle 82 (22.8) 47 (26.1) 35 (19.4) .045* Q29
Married 60 (16.7) 26 (14.4) 34 (18.9)Separated 47 (13.1) 16 (8.9) 31 (17.2)Divorced 153 (42.5) 79 (43.9) 74 (41.1)Widowed 18 (5) 12 (6.7) 6 (3.3)
EducationGrades 1–11 38 (10.5) 21 (11.5) 17 (9.4) .81High school/GED 119 (32.9) 56 (30.8) 63 (35)Some college 167 (46.1) 86 (47.3) 81 (45)College/postgraduate 38 (10.5) 19 (10.4) 19 (10.6)
EmploymentFull and part time 41 (11.3) 21 (11.6) 20 (11) .87Unemployed 141 (39) 69 (38.1) 72 (39.8)Disabled 138 (38.1) 72 (39.8) 66 (36.5)Retired and others 42 (11.6) 19 (10.5) 23 (12.7)
Homeless in past 5 yearsPositive 179 (51.1) 86 (48.9) 93 (53.4) .40
Clinical characteristicsPrimary genotypeType 1 281 (79.6) 148 (83.1) 133 (76) .11Types 2, 3, and 4 72 (20.4) 30 (16.9) 42 (24)
Prior liver biopsy 116 (32) 60 (33) 56 (30.9) .74Biopsy after randomization 104 (28.7) 59 (32.4) 45 (24.9) .13PTSD riskPositive 201 (57.6) 97 (56.1) 104 (59.1) .59
HIV/HCV co infection 42 (11.6) 24 (13.3) 18 (10) .41Prior HCV antiviral treatment 44 (12.1) 19 (10.4) 25 (13.8) .34Prior psychiatric illness 232 (63.9) 113 (62.1) 119 (65.7) .51Prior substance abuse 240 (66.1) 118 (64.8) 122 (67.4) .66Number of prior medical illnessMean (SD) 1.04 (1.06) 0.99 (1.01) 1.1 (1.11) .36
Prior medical illness 237 (65.3) 115 (63.2) 122 (67.4) .44Screen AUDIT-C scoreMean (SD) 2.27 (3.2) 2.09 (2.97) 2.46 (3.41) .37
Screen AUDIT-C (�4 as positive)a
Positive 96 (27.1) 46 (26.1) 50 (28.1) .72Screen BDI scoreMean (SD) 15.5 (9.82) 15.4 (10.1) 15.5 (9.57) .65
Screen BDI (�10 positive)b
Positive 248 (70.7) 123 (71.1) 125 (70.2) .91Alcohol drinks at baseline month, nMean (SD) 11.4 (33.7) 12.8 (36.4) 10.1 (30.8) .95
Active drug use at baseline (within 1 year)c 172 (47.4) 81 (44.5) 91 (50.3) .29Fibrosis levelMean (SD) 2.26 (1.68) 2.3 (1.75) 2.22 (1.61) .93
FLA 5.2.0 DTD � YJCGH54181_proof � 19 March 2015 � 1:42 pm � ce DVC
- 2015 Integrated Care for Hepatitis C 5
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
Q21
Table 1. Continued
Total (n ¼ 363),n (%)
Integrated care (n ¼ 182),n (%)
Usual care (n ¼ 181),n (%) P value
Advanced fibrosis,d (%) 28/124 (22.6) 15/60 (25.0) 13/64 (20.3) .67SiteSan Diego, n 157 78 79Bronx, n 124 63 61Palo Alto, n 82 41 41
BDI, Beck Depression Inventory; PTSD, post-traumatic stress disorder.aScreen AUDIT-C–positive score of 4 or higher at baseline. Patients with a positive AUDIT-C included 18% with a score of 4 to 7 and 9% with a score of 8 orhigher.bScreen BDI-positive score includes scores of 10 or higher. Overall this included 35% of patients with mild depression (scores, 10–18), 27% with moderatedepression (scores, 19–29), and 9% with severe depression (score, >30).cSelf-report active drug use and/or positive urine toxicology within 1 year of baseline (not including marijuana use).dAdvanced fibrosis: Metavir fibrosis scores of 3 to 4 or Ishak fibrosis scores of 4, 5, and 6.
6 Ho et al Clinical Gastroenterology and Hepatology Vol. -, No. -
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
by 101% for subjects treated with IC (Table 3). We alsofound that homelessness in the past 5 years (HR, 1.88;P ¼ .005 for yes vs no), active drug use (HR, 0.59; P ¼.019 for yes vs no), and primary genotype (HR, 1.72; P ¼.024 for genotypes 2, 3, or 4 vs genotype 1) were asso-ciated significantly with time to treatment initiation. Byadding site to the fitted model, the intervention effectstayed similar and site showed a significant associationwith the time to treatment initiation (Palo Alto: HR, 0.43;P ¼ .008; Bronx: HR, 0.57; P ¼ .037). There was nosignificant interaction between site and intervention.
Q22
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
Treatment Adherence
Patients in the IC arm tended to show greateradherence to the planned therapy duration (Figure 3A).The mean percentage of treatment completion of plan-ned duration was 70.3% (SD, 33.1%) in the IC arm and61.7% (SD, 36.5%) in the UC arm. The proportion ofpatients completing at least 80% of planned treatmentwas 52% in the IC arm and 44% in the UC arm. The largemajority of patients within the more than 80% adher-ence group completed 100% of the planned treatmentduration in each arm (Figure 3A). Neither of thesecompletion rates was significantly different between the2 treatment groups. Of the patients completing less than80% of the planned treatment duration in the IC and UCgroups, reasons included adverse event (39% and 44%,respectively), viral nonresponse (46% and 56%,respectively), and nonadherence (15% and 0%, respec-tively). Patients in the IC group tended to have higherrates of on-treatment virologic response at week 12, atthe end of treatment, and in the follow-up time period,with final SVR rates of 50.0% in the IC and 41% in the UCarms (Figure 3B), however, these differences were notstatistically significant.
Q23
691
692
693
694
695
696
Adverse Events
There was no significant difference in the number ofserious adverse events between the IC and UC groups,
FLA 5.2.0 DTD � YJCGH54181_proof �
although there was a trend for numerically fewer hos-pitalizations, emergency room visits, and deaths in the ICgroup compared with the UC group (SupplementaryTable 1).
Protocol Adherence
Patients randomized to the IC arm had frequentcontact with the midlevel mental health practitioner(Supplementary Table 2). There was no evidence ofcross-contamination of mental health practitioner withpatients randomized to the UC group at any site. Patientsin the IC group had a greater number of visits to thehepatitis C clinic at each site compared with patients inthe UC group.
Discussion
A large percentage of HCV patients have psychiatricand SUD comorbidities. A nationwide VA database anal-ysis indicated that 85.4% of HCV patients had a psychi-atric or SUD comorbidity, and 31% had an inpatientpsychiatric or SUD hospitalization in the past year.22 Innon-VA clinics these comorbidities were cited as contra-indications for antiviral therapy in 28% to 38% of pa-tients.11 Interferon-free regimens have greatly simplifiedtreatment; however, high costs, limited access to care, andconcerns about compliance continue to represent barriersto treatment for patients with these comorbidities.13,14
The data presented indicate that an IC protocol usingmidlevel mental health providers for patients with hepa-titis C and substance use and psychiatric comorbidities iseffective, resulting in higher antiviral treatment rates anda 2-fold increase in the numbers of patients with a SVR.The intervention was safe, with no significant differencesin serious adverse events of death, hospitalization, andemergency room visits. This study was a randomized trialthat showed the effectiveness of any interventiondesigned to increase the number of patients to receiveeffective antiviral therapy who had hepatitis C and psy-chiatric and substance use comorbidities.
19 March 2015 � 1:42 pm � ce DVC
Table 2. Simple Logistic Regression for the AssociationBetween SVR and Baseline Characteristics
Baseline characteristicsP
valueOddsratio
95% confidenceinterval
Age .90 1.004 0.95–1.06BMI .93 0.997 0.93–1.07Race ethnicity
African American 1.00White .018 2.52 1.17–5.40Hispanic .73 1.20 0.42–3.40Others .84 1.18 0.24–5.73
Marital statusSingle 1.00Married .77 1.19 0.38–3.74Separated .996 0.997 0.28–3.60Divorced .13 1.99 0.82–4.85Widowed .73 1.34 0.25–7.06
EducationGrades 1–11 1.00High school .96 1.03 0.35–3.01Some college .84 0.90 0.31–2.57College or postgraduate .25 0.37 0.07–2.02
EmploymentEmployed 1.00Unemployed .11 0.51 0.22–1.18
Homeless in past 5 yearsNo 1.00Yes .74 1.11 0.59–2.11
GenotypeType 1 1.00Types 2, 3, and 4 .014 2.38 1.19–4.74
PTSD riskNo 1.00Yes .95 0.98 0.51–1.88
HCV/HIV co-infectionNo 1.00Yes .32 0.54 0.16–1.82
Screen AUDIT-CNegative 1.00Positive .61 0.82 0.39–1.74
Screen BDINegative 1.00Positive .52 1.28 0.60–2.73
Previous substance abuseNo 1.00Yes .24 0.68 0.35–1.30
Prior medical illnessNo 1.00Yes .48 0.79 0.41–1.52
Prior liver biopsyNo 1.00Yes .013 2.26 1.19–4.3
Prior psychiatric disorderNo 1.00Yes .013 0.44 0.23–0.84
Alcohol drinks at baselinemonth, n
.60 1.002 0.99–1.01
Active drug use at baselineNo 1.00Yes .04 0.50 0.25–0.97
SiteSan Diego 1.00Palo Alto .06 0.43 0.18–1.03Bronx .006 0.32 0.14–0.73
BDI, Beck Depression Inventory; PTSD, post-traumatic stress disorder.
Table 3.Multivariate Regression Analysis
Variables Odds ratio
95%confidenceinterval
Pvalue
Association between SVR and intervention groupIntervention group, IC vs UC 2.26 1.13–4.52 .022Genotype, types 2, 3, 4 vs
type 12.20 1.07–4.54 .033
Prior psychiatric disorder,yes vs no
0.44 0.22–0.86 .017
Active drug use, yes vs no 0.47 0.23–0.95 .034
Hazard ratio
Association between time to treatment initiation and interventiongroupIntervention group, IC vs UC 2.01 1.30–3.11 .002Homeless in past 5 years,
yes vs no1.88 1.21–2.92 .005
Genotype, types 2, 3, and 4 vstype 1
1.72 1.07–2.75 .024
Active drug use, yes vs no 0.59 0.38–0.92 .019
FLA 5.2.0 DTD � YJCGH54181_proof �
- 2015 Integrated Care for Hepatitis C 7
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
Multivariate logistic regression analysis of factorssignificantly associated with treatment initiation and SVRindicated that the IC intervention was highly significant(Table 3). We observed that a history of a prior psychi-atric disorder and active drug use was associatedsignificantly with less likelihood of having achieved anSVR on multivariate logistic regression. Subgroup anal-ysis in patients with active drug or alcohol abuse atbaseline, and patients without active substance abusebut with a risk for active psychiatric disease at baseline,showed that the IC intervention had positive effects ontreatment initiation and SVR, respectively, in thesegroups (see the Supplementary Materials and Methodssection). Interestingly, in contrast to non-VA studies,being homeless in the past 5 years at baseline was anindependent predictor of initiating antiviral treatment(HR, 1.88; 95% confidence interval, 1.21–2.92; P ¼ .005).This likely was owing to the existence of robust homelessoutreach programs in the VA that bundle housing, socialservices, and medical care. Approximately 70% ofhomeless veterans were receiving services from long-term homeless shelters, including on-site individual,peer-based, and group counseling; individual case man-agement; vocational rehabilitation; classes and schoolopportunities; and transportation for clinic visits. Theseinclude services that nonhomeless veterans do notreceive. This additional support is hypothesized asenhancing antiviral treatment initiation and completion.
Although IC has been well studied in primary caresettings related to SUD, depression, and HIV, the ICmodels are not well studied in specialty care and fewhave focused on impactful clinical outcomes.15,23,24 Pre-vious studies of patients with chronic hepatitis C andsubstance use and/or psychiatric comorbidities havebeen descriptive, and suggested that multidisciplinary
19 March 2015 � 1:42 pm � ce DVC
Figure 3. Adherence to planned therapy and virologic out-comes in IC vs UC. (A) Amount of adherence to plannedduration of therapy. The percentage of treated patients whoadhered to the indicated percentage of planned therapy. (B)On-treatment virologic response and SVR at 12 or 24 weeksfor Q28patients initiating therapy. LLOQ, lower limit of quantifica-tion. (C) Overall percentage of patients with SVR in each group.
Figure 2. (A) Antiviral treatment initiation over time in the IC vsthe UC group. Antiviral treatment periods are indicated forPEG þ RBV and DAA þ PEG þ RBV. The y-axis shows thecumulative number of patients initiating treatment. Finaltreatment initiation rates were as follows: IC, 58 of 182(31.9%); UC, 34 of 181 (18.8%) (P ¼ .005). (B) Cumulativeprobability of treatment initiation by treatment group overtime (months). Patients were censored at the end of the studyfollow-up period. ne, number of events (treatment initiation);PEGQ27 , pegylated; RBV, ribavirin.
8 Ho et al Clinical Gastroenterology and Hepatology Vol. -, No. -
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
care is feasible and safe,25–28 or may lead to increasedtreatment candidacy.29
The mechanism of the increased antiviral treatmentrates and SVR in the integrated arm could not be speci-fied in this study owing to the multiple components thatwere included in IC. These included elements of casemanagement and linkage to care, self-management,symptom control, substance use treatment, educationand motivation, side-effect management, and access is-sues and co-located care. Multicomponent interventionsevolved out of the recognition that single-componentinterventions often were ineffective. As a result, therehave been few studies that rigorously examined eachcomponent of an integrated intervention.24 A recentmodeling study of hypothetical integrated care programsfor HCV care found that multicomponent interventionsprovided better outcomes and more value for the money
FLA 5.2.0 DTD � YJCGH54181_proof �
than less costly interventions targeting single compo-nents.30 We did observe a trend toward greaterengagement of care at multiple levels in IC patients. Thisincluded an increased number of visits to the hepatitis
19 March 2015 � 1:42 pm � ce DVC
Q24
Q25
- 2015 Integrated Care for Hepatitis C 9
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
clinics, an increased number of liver biopsies afterenrollment, higher adherence to planned duration oftherapy once started on antiviral treatment, and gener-ally lower rates of all adverse events. None of theseobserved differences reached statistical significance, butwere all in the direction favoring IC. It is possible thatphysicians simply were more comfortable treating thehigher-risk HCV patients because they knew they werereceiving integrated care including case management.
Our study had a number of limitations, including thefact that the patients andproviderswere not blinded to theintervention using the IC practitioner. Cross-contamina-tion between treatment arms was possible, however, theIC practitioner never interacted with patients randomizedto the UC comparison group. If cross-contamination hadoccurred and physician involvement with the IC midlevelpractitioner influenced his/her care of patients assignedto the UC arm, this likely would have increased antiviraltreatment processes in the UC group, biasing the studytoward the null hypothesis, and strengthening our con-clusions. The study was limited to VA patients, who pre-dominantly are male, and therefore the results are lessapplicable to community practices. Finally, the study wasconducted during a time when antiviral treatments forHCV were changing to include DAA treatments, whichslowed treatment rates over the period of transition.
To optimize the public health impact of antiviraltreatments for HCV, the number of patients who are ableto receive these treatments must be expanded. Newinterferon-free regimens have fewer side effects and areexpected to expand treatment populations to include abroader range of patients, many with very significantpsychiatric and substance abuse disorders. These datasuggest that integrated care for hepatitis C patients isone tool to maximize the access and success of antiviraltreatment across a broad patient population.
1023
1024
1025
1026
1027
1028
1029
1030
Supplementary Material
Note: To access the supplementary material accom-panying this article, visit the online version of ClinicalGastroenterology and Hepatology at www.cghjournal.org,and at http://dx.doi.org/10.1016/j.cgh.2015.02.022.
1031
References 10321033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1. Backus LI, Belperio PS, Loomis TP, et al. Hepatitis C virusscreening and prevalence among US veterans in Department ofVeterans Affairs care. JAMA Intern Med 2013;173:1549–1552.
2. Veldt BJ, Heathcote EJ, Wedemeyer H, et al. Sustained virologicresponse and clinical outcomes in patients with chronic hepa-titis C and advanced fibrosis. Ann Intern Med 2007;147:677–684.
3. Morgan TR, Ghany MG, Kim HY, et al. Outcome of sustainedvirological responders with histologically advanced chronichepatitis C. Hepatology 2010;52:833–844.
4. Dieperink E, Pocha C, Thuras P, et al. All-cause mortality andliver-related outcomes following successful antiviral treatmentfor chronic hepatitis C. Dig Dis Sci 2014;59:872–880.
FLA 5.2.0 DTD � YJCGH54181_proof �
5. Liu S, Cipriano LE, Holodniy M, et al. New protease inhibitors forthe treatment of chronic hepatitis C: a cost-effectiveness anal-ysis. Ann Intern Med 2012;156:279–290.
6. Camma C, Petta S, Cabibbo G, et al. Cost-effectiveness ofboceprevir or telaprevir for previously treated patients with ge-notype 1 chronic hepatitis C. J Hepatol 2013;59:658–666.
7. Chan K, Lai MN, Groessl EJ, et al. Cost effectiveness of direct-acting antiviral therapy for treatment-naive patients with chronicHCV genotype 1 infection in the veterans health administration.Clin Gastroenterol Hepatol 2013;11:1503–1510.
8. Veterans Affairs. State of care for veterans with hepatitis C. USDepartment of Veterans Affairs, Public Health Strategic Health-care Group, Center for Quality Management in Public Health,2010. updated Nov 2014. Available: http://www.hepatitis.va.gov/index.asp.
9. Holmberg SD, Spradling PR, Moorman AC, et al. Hepatitis C inthe United States. N Engl J Med 2013;368:1859–1861.
10. Davis GL, Alter MJ, El-Serag H, et al. Aging of hepatitis C virus(HCV)-infected persons in the United States: a multiple cohortmodel of HCV prevalence and disease progression. Gastroen-terology 2010;138:513–521.
11. Ho SB, Groessl E, Dollarhide A, et al. Management of chronichepatitis C in veterans: the potential of integrated care models.Am J Gastroenterol 2008;103:1810–1823.
12. North CS, Hong BA, Adewuyi SA, et al. Hepatitis C treatmentand SVR: the gap between clinical trials and real-world treat-ment aspirations. Gen Hosp Psychiatry 2013;35:122–128.
13. Judd S, Liubakka AM, Payle A, et al. Assessing key barriers totreatment of chronic hepatitis C virus (HCV) with next generationagents in a veteran population. Gastroenterology 2014;146:S-737.
14. Medicaid restricts eligibility for new HCV treatments, 2014.
15. Willenbring ML. Integrating care for patients with infectious,psychiatric, and substance use disorders: concepts and ap-proaches. AIDS 2005;19(Suppl 3):S227–S237.
16. Gardner LI, Metsch LR, Anderson-Mahoney P, et al. Efficacy of abrief case management intervention to link recently diagnosedHIV-infected persons to care. AIDS 2005;19:423–431.
17. Bruggmann P, Litwin AH. Models of care for the management ofhepatitis C virus among people who inject drugs: one size doesnot fit all. Clin Infect Dis 2013;57(Suppl 2):S56–S61.
18. Groessl EJ, Sklar M, Cheung RC, et al. Increasing antiviraltreatment through integrated hepatitis C care: a randomizedmulticenter trial. Contemp Clin Trials 2013;35:97–107.
19. Yee HS, Currie SL, Darling JM, et al. Management and treatmentof hepatitis C viral infection: recommendations from theDepartment of Veterans Affairs Hepatitis C Resource Centerprogram and the National Hepatitis C Program office. Am JGastroenterol 2006;101:2360–2378.
20. Martinot-Peignoux M, Stern C, Maylin S, et al. Twelve weeksposttreatment follow-up is as relevant as 24 weeks to determinethe sustained virologic response in patients with hepatitis Cvirus receiving pegylated interferon and ribavirin. Hepatology2010;51:1122–1126.
21. Lawitz E, Mangia A, Wyles D, et al. Sofosbuvir for previouslyuntreated chronic hepatitis C infection. N Engl J Med 2013;368:1878–1887.
22. El-Serag HB, Kunik M, Richardson P, et al. Psychiatric disordersamong veterans with hepatitis C infection. Gastroenterology2002;123:476–482.
23. Soto TA, Bell J, Pillen MB. Literature on integrated HIV care: areview. AIDS Care 2004;16(Suppl 1):S43–S55.
19 March 2015 � 1:42 pm � ce DVC
1044
Q3
Q4
Q5
10 Ho et al Clinical Gastroenterology and Hepatology Vol. -, No. -
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
24. Woltmann E, Grogan-Kaylor A, Perron B, et al. Comparativeeffectiveness of collaborative chronic care models for mentalhealth conditions across primary, specialty, and behavioralhealth care settings: systematic review and meta-analysis. Am JPsychiatry 2012;169:790–804.
25. Knott A, Dieperink E, Willenbring ML, et al. Integrated psychi-atric/medical care in a chronic hepatitis C clinic: effect on anti-viral treatment evaluation and outcomes. Am J Gastroenterol2006;101:2254–2262.
26. Schaefer M, Hinzpeter A, Mohmand A, et al. Hepatitis C treat-ment in “difficult-to-treat” psychiatric patients with pegylatedinterferon-alpha and ribavirin: response and psychiatric sideeffects. Hepatology 2007;46:991–998.
27. Sylvestre DL, Clements BJ. Adherence to hepatitis C treatmentin recovering heroin users maintained on methadone. Eur JGastroenterol Hepatol 2007;19:741–747.
28. Brunner N, Senn O, Rosemann T, et al. Hepatitis C treatment formultimorbid patients with substance use disorder in a primarycare-based integrated treatment centre: a retrospective anal-ysis. Eur J Gastroenterol Hepatol 2013;25:1300–1307.
29. Evon DM, Simpson K, Kixmiller S, et al. A randomized controlledtrial of an integrated care intervention to increase eligibility for
FLA 5.2.0 DTD � YJCGH54181_proof �
chronic hepatitis C treatment. Am J Gastroenterol 2011;106:1777–1786.
30. Linas BP, Barter DM, Leff JA, et al. The hepatitis C cascade ofcare: identifying priorities to improve clinical outcomes. PLoSOne 2014;9:e97317.
Reprint requestsAddress requests for reprints to: Samuel B. Ho, MD, VA San Diego HealthcareSystem, 3350 La Jolla Village Drive, San Diego, California 92161. e-mail:[email protected]; fax: (858) 552-4327.
Conflicts of interestThese authors disclose the following: Samuel Ho has received research andgrant support from Genetech, Inc, and Gilead Sciences, has served on theadvisory board for Janssen Pharmaceuticals, Inc, and has served on thespeakers bureau for Prime Education, Inc; Norbert Brau has received researchand grant support from Gilead Sciences, BMS, Vertex, and AbbVie, has servedon the advisory boards for Janssen Pharmaceuticals, Inc, and Gilead Sciences,and has served on the speakers bureau for Prime Education, Inc; RamseyCheung has received research and grant support from Gilead Sciences, andhas served on the advisory board for Janssen Pharmaceuticals, Inc. Theremaining authors disclose no conflicts.
FundingSupported by VA Health Services Research and Development grant IIR-07-101-3.
1123
19 March 2015 � 1:42 pm � ce DVC
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
Supplementary Materials and MethodsQ30
Screening Tests
The tests used to screen patients for eligibility in thestudy were described previously.1 These inclusioncriteria included a Beck Depression Inventory score of10 or greater, an AUDIT-C score of 4 or higher, a positivepost-traumatic stress disorder-VA Primary Care screen(endorsement of �3 items), and/or drug use consistingof self-reported illicit drug use (illicit drugs other thanmarijuana and prescription drug abuse) within the pre-vious 6 months. Patients scoring higher than the vali-dated cut-off levels for any of these conditions werecandidates for the study. For depression screening theoriginal Beck Depression Inventory was used as updatedin 1979.2 Studies have indicated a high correlation of theBeck Depression Inventory and the later Beck Depres-sion Inventory II test.3 The diagnostic accuracy of theBeck Depression Inventory for clinical depression is82% and higher as expressed by the area under thereceiver operating characteristics curve, with sensitiv-ities and specificities generally exceeding 80%.4,5 Forpost-traumatic stress disorder we used the primary carepost-traumatic stress disorder screen, a 4-question toolthat at a cut-off score of 3 or higher has been shown tohave a sensitivity and specificity of 78% and 87%,respectively, for the diagnosis of post-traumatic stressdisorder.6 Screening for illicit drug use (not includingmarijuana) included a drug use questionnaire thatscreened for drug use in the previous 6 months. PatientsmeetingQ31 this screening test completed questionnairesthat evaluated drug use within the proceeding 1 yearand charts were audited for positive urine toxicologyscreens. The AUDIT-C was used to identify patients withhigh-risk alcohol use, with a cut-off score of greater than4, as indicated by national VA guidelines (www.hepatitis.va.gov/provider/tools/audit-c.asp). This score willidentify 86% of patients with heavy drinking and/oractive alcohol abuse or dependency (sensitivity) with aspecificity of 72%.7
Intervention
Integrated care protocol. The mental health providersin the study included 1 marriage and family therapistand 2 psychologists. All received uniform training inperson at the beginning of the study and a writtenprotocol and therapy manual. They had no prior expe-rience with hepatitis C patients and they remained ineach clinic for the duration of the study. Ongoing trainingand monitoring during the study consisted of monthlyconference calls and patient discussions designed tomaintain uniformity of the protocol and approach. Weused a concept of integrated care that included practi-tioners working together as a team across specialtiesand service lines. The addition of a mental health
provider to the usual clinic and under the collaborativedirection of the hepatitis C providers (rather than a su-pervisor not involved in the clinic), represents anexample of this type of integrated care. Other aspects ofintegrated care include using a common protocol, havingfrequent communication and meetings, having collabo-rative and common goals for patient care (initiatingsuccessful antiviral therapy), all of which were facilitatedby the MH provider. Descriptions of the IC protocol werepublished previously.1,8 Parameters for antiviral treat-ment initiation for both IC and UC were the inclusion andexclusion criteria provided in the VA treatment guide-lines in effect at the time of the study.9 These guidelineswere uniform for the VA system and served as guidancefor clinicians at each site. Treatment initiation guidelinescalled for substance use, depression, and other psychi-atric conditions to be stable, which was determined bythe practitioners as per standard medical criteria.
Statistical Analysis
Sample size determination. By using preliminary data,we estimated that 15% of patients in the UC arm wouldreceive treatment and 30% would achieve an SVR,resulting in an overall SVR of 4.5%. For the IC arm, weestimated that 35% of patients would initiate treatmentand that 40% of those initiating treatment would achievean SVR, resulting in an overall SVR of 14%. With theearlier-described assumptions and a 2-sided type I errorof 0.05, there was at least 80% power to detect thatdifference with a total of 330 patients (110 per site). Anenrollment of 360 subjects was targeted to account forattrition.
Baseline characteristics. Descriptive statistics wereused to summarize baseline characteristics. Baselinecharacteristics were compared between the interventiongroups using the Wilcoxon rank sum test, the chi-squaretest and the Fisher exact test were used for comparison Q32.
Primary and secondary outcomes. Intent-to-treatanalysis was performed for all clinical outcomes.Initially, the proportion of patients with SVR (primarystudy outcome) was compared between the UC and ICgroups using a univariate Fisher exact test. Multivariatelogistic regression was used to assess the difference inSVR between the 2 groups with adjustment for baselinecharacteristics. The association between baseline char-acteristics and SVR was assessed with univariate logisticregression first. The method of purposeful selection wasused for the selection of covariates, with a P value lessthan .10 being kept in the final model.10 The likelihoodratio test was used for model comparison. Influentialobservations were assessed using Cook statistics andleverages. The final model was fitted by excludinginfluential observations and compared with the originalmodel.
The main secondary outcome of time to treatmentinitiation was analyzed using the log-rank test and
FLA 5.2.0 DTD � YJCGH54181_proof � 19 March 2015 � 1:42 pm � ce DVC
- 2015 Integrated Care for Hepatitis C 10.e1
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
visualized with a Kaplan–Meier curve. Cox proportionalhazard modeling was used for multivariate analysis toadjust for baseline characteristics. The association be-tween baseline characteristics and time to treatmentinitiation was assessed with univariate Cox regression aspotential covariates in a multivariate model. Covariatesfor the final model were identified by purposeful selec-tion. The partial likelihood ratio test was used for modelcomparison. The effects of influential observations onestimated parameters were assessed by score re-siduals11 and the proportional hazards assumption wasassessed using the test by Grambsch and Therneau.12
The final multivariable Cox regression model was strat-ified by post-traumatic stress disorder risk owing to theviolation of proportional hazards assumption. The influ-ence of the study site and the interaction of the study sitewith the intervention arm was assessed in the fittedmultivariate models for both SVR and treatment initia-tion. The percentage of treatment completion and theproportion of subjects completing 80% of treatmentwere compared between the 2 intervention groups usingthe Wilcoxon rank-sum test and the Fisher exact test.Adverse events were summarized by treatment groupand compared using the Wilcoxon rank-sum test and theFisher exact test. For subjects with and without activedrug use at baseline and/or active alcohol abuse basedon the AUDIT-C score at baseline, the differences in SVR,treatment initiation, and treatment completion betweenthe UC and IC arms were compared using descriptivestatistics and the appropriate univariate test such as theFisher exact test and the Wilcoxon rank-sum test. Allanalyses were performed by SPSSQ33 and R, and a P valueless than .05 was interpreted as statistically significant.
Subgroup Analyses: Subjects With and WithoutActive Drug and Alcohol Abuse
Among 234 subjects with active drug use and alcoholabuse at baseline, the IC intervention was associatedsignificantly with treatment initiation (IC, 30.4%; UC,16.4%; P ¼ .013), but did not significantly affect theoverall SVR rate in this group (IC, 10.7%; UC, 7.4%; P ¼.49). The mean adherence to the planned duration oftreatment was 63.3% (SD, 34.3Q34 ) for IC and 56.5% (SD,39.5) for UC, and the proportion of patient completing atleast 80% of the planned treatment was 41.2% in the ICand 45% in the UC arms. Neither of these completion rateswas significantly different between the 2 interventiongroups. Among 129 subjects with positive post-traumaticstress disorder risk and depression (Beck DepressionInventory, �10) but without active drug or alcohol abuseat baseline, the IC intervention was not associatedsignificantly with treatment initiation (IC, 34.3%; UC,23.7%; P ¼ .25); but was associated with overall SVR (IC,24.3% vs UC, 8.5%; P¼ .020). The mean adherence to theplanned duration of treatment was 81.0% (SD, 27.5%) forthe IC and 69.2% (SD, 31.5%) for the UC arm, and the
proportion of patients completing at least 80% of theplanned treatment was 62.5% in the IC and 42.9% in theUC arm. Neither of these completion rates was signifi-cantly different between the 2 intervention groups.
It should be noted that the study was not powered todetect significant differences in subgroups. In the sub-group of patients with active drug and alcohol abuse atbaseline, more patients started treatment in IC Q35, althoughthe number of patients was low and a significant increasein SVR was not found. For the subgroup of patients athigh risk for post-traumatic stress disorder and depres-sion without substance abuse, IC Q36was associated with asignificant increase in the total number of patientsachieving SVR Q37, and we observed nonsignificant trends ofincreased treatment initiation and adherence in the ICgroup that may have contributed to the increase in SVRsobserved in this subgroup.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
FLA 5.2.0 DTD � YJCGH54181_proof � 19 M
10.e2 Ho et al Clinical Gastroenterology and Hepatology Vol. -, No. -
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
ReferencesGroessl EJ, Sklar M, Cheung RC, et al. Increasing antiviraltreatment through integrated hepatitis C care: a randomizedmulticenter trial. Contemp Clin Trials 2013;35:97–107.
Beck AT, Rush AJ, Shaw BF, et al. Cognitive therapy ofdepression. New York: Guilford Press, 1979.
Beck AT, Steer RA, Ball R, et al. Comparison of Beck Depres-sion Inventories -IA and -II in psychiatric outpatients. J PersAssess 1996;67:588–597.
Williams JR, Hirsch ES, Anderson K, et al. A comparison of ninescales to detect depression in Parkinson disease: which scale touse? Neurology 2012;78:998–1006.
Wang YP, Gorenstein C. Assessment of depression in medicalpatients: a systematic review of the utility of the Beck Depres-sion Inventory-II. Clinics Sao Paolo 2013;68:1274–1287.
Ouimette P, Wade M, Prins A, et al. Identifying PTSD in primarycare: comparison of the Primary Care-PTSD screen (PC-PTSD)and the General Health Questionnaire-12 (GHQ). J AnxietyDisord 2008;22:337–343.
Bush K, Kivlahan DR, McDonell MB, et al. The AUDIT alcoholconsumption questions (AUDIT-C): an effective brief screeningtest for problem drinking. Ambulatory Care Quality ImprovementProject (ACQUIP). Alcohol Use Disorders Identification Test.Arch Intern Med 1998;158:1789–1795.
Knott A, Dieperink E, Willenbring ML, et al. Integrated psychi-atric/medical care in a chronic hepatitis C clinic: effect on anti-viral treatment evaluation and outcomes. Am J Gastroenterol2006;101:2254–2262.
Yee HS, Currie SL, Darling JM, et al. Management and treatmentof hepatitis C viral infection: recommendations from theDepartment of Veterans Affairs Hepatitis C Resource Centerprogram and the National Hepatitis C Program office. Am JGastroenterol 2006;101:2360–2378.
Hosmer DW, Lemeshow S. Applied logistic regression. NewYork: Wiley, 2000.
Hosmer DW, Lemeshow S, May S. Applied survival analysis:regression modeling of time-to-event data. 2nd ed. Hoboken,NJ: Wiley-Interscience, 2008.
Grambsch PM, Therneau TM. Proportional hazards tests anddiagnostics based on weighted residuals. Biometrika 1994;81:515–526.
arch 2015 � 1:42 pm � ce DVC
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
Supplementary Table 2. Adherence to Protocol by Site for Clinic Visits With IC Mental Health Practitioner and for HCV ClinicVisits (Not Involving Mental Health)
All sites Bronx Palo Alto San Diego
Patients IC, n 182 63 41 78Patients UC, n 181 61 41 79IC mental health practitioner visits
Integrated care (visits per patient) 1821 (10.0) 695 (11.03) 370 (9.02) 756 (9.69)Usual care 0 0 0 0
Total HCV clinic visits (nonmental health)a
Integrated care (visits per patient) 1670 (9.18) 507 (8.05) 208 (5.07) 955 (12.24)Usual care (visits per patient) 1052 (5.82) 404 (6.62) 129 (3.15) 519 (6.57)
IC patients without antiviral treatment, n 124 48 33 43UC patients without antiviral treatment, n 147 52 36 59Total HCV clinic visits (nonmental health) for patients without antiviral treatmentb
Integrated care (visits per patient) 566 (4.57) 312 (6.50) 80 (2.42) 174 (4.05)Usual care (visits per patient) 522 (3.55) 288 (5.54) 80 (2.22) 154 (2.61)
NOTE. The time period was from April 1, 2009, through July 31, 2012. Each site implemented a similar number of visits per patient in the IC arm.aIncludes clinic visits for all patients with or without antiviral treatment.bIncludes clinic visits for patients who never received antiviral treatment.
Supplementary Table 1. Adverse Event Outcomes (Number of Hospitalizations, Emergency Room Visits, and Death From anyCause) for Patients Randomized to the IC Group (n ¼ 182 Unique Patients) or the UC Group (n ¼ 181Unique Patients)
Total (n ¼ 363), mean (SD) IC (n ¼ 182), mean (SD) UC (n ¼ 181), mean (SD) P value
Number of hospitalization events/patientNumber of hospital days/patient
0.89 (1.65) 0.79 (1.55) 0.99 (1.74) .15
Subjects with a hospital eventa 16.5 (37.2) 16.3 (42.5) 16.7 (31.7) .13All subjectsb 0.98 (29.9) 9.38 (33.2) 10.3 (26.2) .103
Number of ER visits/patient 2.99 (4.2) 2.85 (4.1) 3.14 (4.30) .41Number of deaths, n (%)c 22 (6.1) 8 (4.4) 14 (7.7) .20
NOTE. The time period was from July 1, 2009, through August 1, 2012.aThere were 216 subjects (104 IC and 112 UC) with hospitalization events.bHospital day was 0 for subjects without a hospitalization event.cCauses of death in the IC group were as follows: decompensated cirrhosis (2), cardiac arrest (1), medication overdose (1), cancer (1), unknown (3). No patientswere receiving antiviral therapy. Causes of death in the UC group were as follows: decompensated cirrhosis (2), cancer (2), colitis (1), cardiac arrest (2), unknown(7). One of the unknown deaths occurred at 2 months on antiviral therapy.
FLA 5.2.0 DTD � YJCGH54181_proof � 19 March 2015 � 1:42 pm � ce DVC
- 2015 Integrated Care for Hepatitis C 10.e3
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518