Telemedicine successfully engages marginalized rural hepatitis C patients in curative care

Abstract

Background

Rurally located individuals living with hepatitis C virus (HCV) face barriers to engagement and retention in care. Telemedicine technologies coupled with highly curative direct acting antiviral (DAA) treatments may increase accessibility to HCV care while achieving high sustained virologic response (SVR) rates. We compared clinical and socio-economic characteristics, SVR, and loss to follow-up among telemedicine (TM), mixed delivery (MD), and outpatient clinic (OPC) patients receiving care through The Ottawa Hospital Viral Hepatitis Program (TOHVHP).

Methods

TOHVHP clinical database was used to evaluate patients engaging HCV care between January 1, 2012, and December 31, 2016. SVR rates by HCV care delivery method (TM versus OPC versus MD) were calculated.

Results

Analysis included 1,454 patients who engaged with TOHVHP at least once. Patients were aged almost 50 years on average and were predominately male and Caucasian. A greater proportion of TM patients were rurally based, were Indigenous, had a history of substance use, and had previously been incarcerated. Per-protocol DAA SVR rates for TM, OPC, and MD patients were 100% (26/26), 93% (440/472), and 94% (44/47), respectively. Loss-to-follow-up rates for HCV-treated TM and MD patients were higher (27% [10/37], 95% CI 0.58 to 0.88, and 11% [7/62], 95% CI 0.81 to 0.97, respectively) than for those followed exclusively in the OPC (5% [39/800], 95% CI 0.94 to 0.97).

Conclusions

TM can successfully engage, retain, and cure rurally based HCV patients facing barriers to care. Strategies to improve TM retention of patients initiating HCV antiviral treatment are key to optimizing the impact of this model of care.

High rates of hepatitis C virus (HCV)–related morbidity and mortality, coupled with the increased HCV diagnosis rate and demand for chronic HCV care in Canada, highlight a critical need for new approaches to care for and treat individuals living with chronic HCV in the early stages of the disease (1). The introduction of direct acting antiviral (DAA) therapies has vastly improved tolerability and sustained virologic response (SVR) rates, with upward of 90%–100% of patients clearing their HCV infection compared with a clearance rate of 6%–63% observed with previously used interferon-based therapies (2–6). DAA therapies have also allowed for increased treatment eligibility, including people with active substance use (7) and psychiatric illnesses (2,4,8,9). However, many individuals living with HCV face barriers that prevent their sustained engagement with HCV care. Treatment restrictions based on available insurance coverage and socio-economic barriers resulting from stigma and housing instability, as well as geographic isolation, all make it difficult for many to initiate and adhere to traditionally delivered, in-clinic HCV care and treatment (9–13).

Telemedicine-delivered HCV services have been suggested as a means to increase access for remote populations and to reduce the financial strain on patients seeking HCV care and treatment (14). Telemedicine services have been used in marginalized populations residing in underserviced areas and have achieved similar clinical outcomes as services obtained in traditional clinic settings (15–17). In addition to comparable clinical outcomes, evidence suggests that there are additional benefits to using telemedicine services, including decreased wait time for specialty care, decreased health care expenditures, improved patient follow-up, and improved patient experience as a result of comprehensive continuity of care (1,18–21).

In Canada, the use of telemedicine services is particularly important because of geographic considerations. Many tertiary care centres are located in large, urban areas, requiring many patients living with chronic HCV to travel great distances for care (14). This often causes financial stress and additional travel time and requires time off from work for patients, their family, and caregivers (1). The Ottawa Hospital Viral Hepatitis Program (TOHVHP) has implemented a multidisciplinary HCV telemedicine program to serve Eastern Ontario, Western Quebec, and Nunavut to increase access to HCV care. This program serves rurally situated individuals using real-time telemedicine consultations. To evaluate the effectiveness of TOHVHP, we compared the SVR rates and patient retention of TOHVHP telemedicine patients with those of patients managed in person at TOHVHP outpatient clinic at The Ottawa Hospital–General Campus. We hypothesized that TOHVHP telemedicine program would have patient retention in care, proportion of patients initiating treatment, and rates of achieved HCV cure similar to those of TOHVHP outpatient clinic. For the purposes of this analysis, we defined telemedicine care as any clinic visit conducted using video and teleconference technology to link an HCV patient to the HCV health care team.

Methods

Patient Population

A retrospective cohort analysis was conducted for patients aged 18 years or older who were engaged with TOHVHP at least once for HCV care between January 1, 2012, and December 31, 2016. Patient data were collected from both patient charts and electronic patient files in The Ottawa Hospital Viral Hepatitis Clinical dataset (Ottawa Health Science Network Research Ethics Board REB 2004-196). Treatment information was restricted to the latest HCV treatment regimen on file. Analysis excluded patients who were HCV RNA negative, those mono-infected with hepatitis B virus, and those who did not attend at least a single scheduled appointment.

Definition of HCV Care Delivery Method

Telemedicine patients were those who received all HCV care using The Ontario Telemedicine Health Network visual and audio equipment at geographically separated medical clinics. Briefly, our telemedicine program serves more than 70 sites across a vast territory including Ottawa, Eastern Ontario, Western Quebec, and Nunavut. It employs a full-time nurse who has access to social work, psychology, and pharmacy supports as required. Key activities include a once-weekly formal telemedicine clinic and onsite education and transient elastography clinics conducted several times per month in a rotating fashion at sites throughout Eastern Ontario.

Outpatient clinic patients were those who received all of their HCV care exclusively in the outpatient clinic at The Ottawa Hospital–General Campus. Patients allocated to the mixed-delivery group were those who had both outpatient clinic and telemedicine visits on their hospital record. Patients are able to choose which model of care they would prefer before their initial visit. However, in some instances (i.e., advanced liver disease), it is necessary to assess patients in the outpatient clinic. Our program includes infectious diseases, hepatology, endocrine, psychiatry, and primary care psychiatrists. Our multidisciplinary team provides support for people with substance use and coordinates linkage to regional substance abuse programs. All patients, irrespective of model of care, have access to these services.

Study Variables

Demographic and socio-economic information, including sex, race, education level, housing suitability, and history of mental illness, was collected. Patients reported risk factors for HCV exposure, demographic, socio-economic, and mental health information at the time of entry into TOHVHP. Laboratory results, including pre-treatment fibrosis scores by transient elastography, HCV genotype, HIV, and hepatitis B virus co-infection, HCV viral load, and liver studies, were collected. Clinical outcomes included treatment initiation and completion dates, HCV antiviral regimen, and SVR status. SVR was defined as HCV RNA negativity of 12 weeks or more after completion of HCV antiviral therapy. For individuals who had received multiple courses of HCV antiviral therapies, SVR was determined on the basis of their most recent treatment.

Social and Material Deprivation Index

Geographic-level census information for TOHVHP patient population was determined through linkage of extracted TOHVHP patient postal codes collected during patient intake with available 2016 Statistics Canada postal codes. These codes represented geographic areas used in the 2016 Statistics Canada Census data and were used to measure geographic-level social and material deprivation (22). Six socio-economic indicators were selected on the basis of their association with material deprivation (proportion of individuals without a high school diploma, employed-to-population ratio, and mean income) and social deprivation (proportion of individuals living alone, proportion of single-parent families, and proportion of individuals who have been divorced, widowed, or separated). The lowest quintile represented patients experiencing the greatest level of deprivation. Geographic areas of residence for patients were allocated into two regions: rural or urban centre.

Statistical Analysis

Comparisons among telemedicine, outpatient clinic, and mixed-delivery patients were made by means of χ² or Fisher’s Exact Test for categorical variables and analysis of variance for continuous variables. The proportions of patients initiating HCV antiviral therapy were calculated as a function of model of care. SVR proportions by HCV care delivery method were calculated using both intention-to-treat (ITT) and per-protocol analyses.

ITT SVR included all individuals who initiated a HCV treatment regimen; it was selected as the primary outcome given the importance of HCV cure among those initiating HCV antiviral therapy. Per-protocol SVR included only individuals who successfully completed their treatment regimen and had post-treatment SVR bloodwork. SVR was determined for all treatment types, including interferon-based regimens, and restricted to interferon-free DAA regimens given its relevancy to the current HCV treatment standard of practice.

Patient retention rates (defined as the number of HCV antiviral treatment recipients completing SVR testing divided by the number of those initiating HCV antiviral therapy) were compared by HCV care delivery method by means of χ² analysis. Patients treated with HCV antivirals were deemed lost to retention if they (a) did not have a SVR bloodwork result on record 12 or more weeks after completing HCV treatment or (b) were lost from HCV care while on treatment.

Univariate and multivariate logistic regression models were developed using per-protocol data to assess the association among HCV care delivery method, patient characteristics, and clinical outcomes on both patient retention and SVR among telemedicine, outpatient clinic, and mixed-delivery patients. Model variables were selected a priori. Variables determined to be significant by univariate analysis (p < .10) were included in the multivariate analysis. Estimates are presented as odds ratios (ORs) and 95% CIs. Analyses were conducted using SAS software (Version 9.4; SAS Institute Inc., Cary, NC).

Results

A total of 1,454 patients engaged with TOHVHP at least once for HCV care between January 2012 and December 2016 (Figure 1). Among these patients, 106 received all care by telemedicine, 1,267 received exclusively outpatient clinic care, and 81 received mixed-delivery care. These three groups were comparable in age, sex, and proportion of individuals experiencing the greatest quintile of material and social deprivation (Tables 1 and 2). With a few notable exceptions described later, these characteristics were generally similar between rural-based patients and those residing in urban locations (data not shown). Telemedicine and mixed-delivery patients were far less likely to reside in urban areas than those receiving care exclusively in the outpatient clinic (2% versus 5% versus 75%). Telemedicine patients were more likely than outpatient clinic and mixed-delivery patients to be Indigenous (10% versus 2% versus 4%) and to not have graduated high school (17% versus 11% versus 14%). A lifetime history of mental illness was most common in the telemedicine group. Risk factors for HCV exposure, including injection drug use and tattooing, were more common among telemedicine patients. This group was less likely to have received blood products leading to HCV infection.

Figure 1:

Description of The Ottawa Hospital Viral Hepatitis Program patient population by HCV care delivery method

HCV = Hepatitis C virus

Table 1:

Population characteristics for telemedicine, outpatient clinic, and mixed-delivery patients: categorical socio-economic factors, patient characteristics, and engagement in HCV-related risk factors.

	Telemedicine (n = 106)		Clinic (n = 1,267)		Mixed delivery (n = 81)
Characteristic	No. (n)	%	No. (n)	%	No. (n)	%	p value
Socio-economic factors
Male	72 (106)	67.9	804 (1,261)	63.8	53 (81)	65.4	0.7
Race							<0.0001
Caucasian	74 (106)	69.8	877 (1,267)	69.2	66 (81)	81.5
Black	2 (106)	1.9	96 (1,267)	7.6	1 (81)	1.2
Asian	0 (106)	0	90 (1,267)	7.1	1(81)	1.2
Indigenous	11 (106)	10.4	27 (1,267)	2.1	3 (81)	3.7
Hispanic	0 (106)	0	1 (1,267)	0.08	0 (81)	0
Other	1 (106)	0.9	20 (1,267)	1.6	1 (81)	1.2
No high school diploma*	18 (104)	17.3	132 (1,238)	10.7	11 (78)	14.1	<0.0001
Unsuitable housing†	3 (104)	2.9	69 (1,238)	5.6	3 (78)	3.8	<0.0001
Area of residence							<0.0001
Rural	42 (106)	39.6	164 (1,267)	12.9	36 (81)	44.4
Intermediate	59 (106)	55.7	91 (1,267)	7.2	37 (81)	45.7
Urban	2 (106)	1.9	951 (1,267)	75.1	4 (81)	4.9
Greatest material deprivation	33 (106)	31.1	275 (1,276)	21.7	17 (81)	21.0	0.2
Greatest social deprivation	19 (106)	17.9	282 (1,267)	22.3	20 (81)	24.7	0.3
Mental illness‡	PTSD	1 (106)	0.9	29 (1,267)	2.3	0 (81)	0	0.3
Schizophrenia	6 (106)	5.7	17 (1,267)	1.3	0 (81)	0	0.001
Depression	43 (106)	40.6	322 (1,267)	25.4	12 (81)	14.8	0.0002
Bipolar disorder	4 (106)	3.8	47 (1,267)	3.7	0 (81)	0	0.2
Anxiety disorder	31 (106)	29.2	201 (1,267)	15.9	8 (81)	9.9	0.0004
HCV-related risk factors
IDU history	77 (106)	72.6	703 (1,267)	55.5	49 (81)	60.5	<0.0001
High-risk sex§	4 (106)	3.8	105 (1,267)	8.3	11 (81)	13.6	0.05
Received blood products	15 (106)	14.2	192 (1,267)	15.1	19 (81)	23.5	0.0001
Cocaine snorting	52 (106)	49.1	595 (1,267)	47.0	38 (81)	46.9	<0.0001
Tattoos	58 (106)	54.7	525 (1,267)	41.4	38 (81)	46.9	<0.0001
Piercings	25 (106)	23.6	416 (1,267)	32.8	21 (81)	25.9	<0.0001
History of incarceration	49 (106)	46.2	474 (1,267)	37.4	34 (81)	42.0	<0.0001

^* The proportion of individuals without a high school diploma was obtained by linking TOHVHP postal codes with 2016 Statistics Canada Census information through the use of the Statistics Canada 2016 Postal Code Conversion file. Average proportion for each clinic was used

^† The proportion of individuals with unsuitable housing was obtained by linking TOHVHP postal codes with 2016 Statistics Canada Census Information through the use of the Statistics Canada 2016 Postal Code Conversion file. Average proportion for each clinic was used

^‡ All mental illness is self-reported history of any of the reported mental illnesses for any duration in the patient’s life

^§ High-risk sex encompassed both individuals who engage in high-risk heterosexual intercourse and men who have sex with men.

HCV = Hepatitis C virus; PTSD = Post-traumatic stress disorder; IDU = Injection drug use; TOHVHP = The Ottawa Hospital Viral Hepatitis Program

Table 2:

Population characteristics for telemedicine, outpatient clinic, and mixed-delivery patients: categorical clinical characteristics

	No. (%)
Clinical characteristic	Telemedicine (n = 106)	Clinic (n = 1,267)	Mixed delivery (n = 81)	p value
HCV Genotype*				0.03
1	68 (64.2)	824 (65.0)	61 (75.3)
2	6 (5.7)	101 (8.0)	5 (6.2)
3	26 (24.5)	205 (16.2)	13 (16.0)
4	2 (1.9)	74 (5.8)	0 (0)
Mixed	0 (0)	20 (1.6)	0 (0)
HIV co-infection	1 (0.9)	76 (6.0)	0 (0)	0.008
HBV co-infection	2 (1.9)	13 (1.0)	0 (0)	0.09
Underwent a liver biopsy	12 (11.3)	438 (38.1)	24 (29.6)	<0.0001

^* Excludes individuals without HCV genotyping and those with a viral load too low to determine the infecting HCV genotype

HCV = Hepatitis C virus; HBV = Hepatitis B virus

Telemedicine patients were more likely to be genotype 3 infected (23% versus 16%). They were less likely than clinic patients to be HIV co-infected (1% versus 6%) and to have undergone a liver biopsy (11% versus 38%; Table 2), and they were less likely than both outpatient clinic and mixed-delivery patients to have cirrhosis (23% versus 30% versus 44%, respectively) (Table 3). Telemedicine patients had slightly lower baseline median transient elastography scores (8.0 kPa versus 8.3 kPa versus 11.0 kPa), but this result was not statistically significant (Table 3).

Table 3:

Population characteristics for telemedicine, outpatient clinic, and mixed-delivery patients: baseline HCV clinical measurements

	Telemedicine (n = 106)		Clinic (n = 1,267)		Mixed delivery (n = 81)
Baseline variables	No.	Mean (SD)*	No.	Mean (SD)*	No.	Mean (SD)*	p value
Age	106	48.2 (11.4)	1,261	49.3 (11.6)	81	50.9 (10.6)	0.2
ALT (IU/L)†	100	95 (90)	1,246	97 (87)	79	120 (113)	0.07
HCV viral load (IU/mL)	97	3.2e+6 (5.4e+6)	1,125	4.8e+6 (1.9e +7)	77	4.2e+6 (7.8e+6)	0.6
Median fibroscan score (kPa)	70	8.0 (3.4–58)	757	8.3 (2.4–75)	62	11.0 (3.2–63)	0.1
Proportion with cirrhosis, no. (%)‡	70	16 (23)	757	227 (30)	62	27 (44)	0.03

* Unless otherwise indicated

† Baseline ALT values

‡ Cirrhosis was defined as having a transient elastography score by Fibroscan of 12.5 kPa or more

HCV = Hepatitis C virus; ALT = Alanine aminotransferase

The proportion of patients initiating HCV antiviral therapy differed by clinic modality: telemedicine, 35% (37/106); outpatient clinic, 64% (815/1,267); and mixed delivery, 77% (62/81). After controlling for the difference in treatment and program availability, treatment uptake did not statistically differ by HCV care delivery method (p = 0.24). SVR by ITT for all HCV antiviral regimens (i.e., both interferon-containing and interferon-free) was 75% (28/37) for telemedicine patients compared with 80% (655/815) for outpatient clinic patients and 83% (51/62) for mixed-delivery patients. The likelihood of achieving ITT SVR did not differ by HCV care delivery method (p = 0.72). Per-protocol SVR rates were 100% (27/27) for telemedicine patients compared with 85% (655/768) for outpatient clinic patients and 93% (51/55) for mixed-delivery patients (p = 0.03). After controlling for the difference in treatment efficacy between interferon-based treatments and DAA treatments, the likelihood of achieving per-protocol SVR did not differ by HCV care delivery method (p = 0.43) (data not shown).

Given their relevance to current HCV treatment standard of practice, interferon-free, we determined DAA-specific ITT and per-protocol SVR rates by HCV care delivery method (Figure 2). ITT SVR was 79% (27/34) for telemedicine patients, 88% (440/500) for outpatient clinic patients, and 85% (44/52) for mixed-delivery patients. Per-protocol SVR rates were 100% (26/26) for telemedicine patients, 93% (440/472) for outpatient clinic patients, and 94% (44/47) for mixed-delivery patients. Restricting multivariate analysis to recipients of DAA therapies, genotype 3 infection was found to be associated with decreased odds of achieving SVR (OR = 0.37; 95% CI 0.16 to 0.85; p = 0.02). Because all telemedicine patients were prescribed DAAs and achieved per-protocol SVR, HCV care delivery could not be included in the multivariate logistic regression analysis because doing so created model instability (Table 4).

Figure 2:

Overall percentage of HCV cure by per-protocol SVR definition for telemedicine, outpatient clinic, and mixed delivery patients represented by treatment regimen

Note: No telemedicine patients included in this analysis received INF-based therapy

HCV = Hepatitis C virus; SVR = Sustained virologic response; IFN = Interferon; DAAs = Direct acting antivirals

Table 4:

Odds ratios (ORs) of predictors of SVR among DAA recipients

Variable	(a/c)/(b/d)*	OR (95% CI)	p value
Female	(183/327)/(12/23)	1.19 (0.57 to 2.49)	0.64
Age	—	1.02 (0.99 to 1.05)	0.24
HCV genotype (Ref.: genotype 1)
2	(38/472)/(1/34)	2.04 (0.27 to 15.59)	0.49
3	(55/455)/(9/26)	0.37 (0.16 to 0.85)	0.02
4	(26/484)/(3/32)	0.47 (0.13 to 1.69)	0.25

Note: Patients achieved SVR and received DAAs using per-protocol analysis.

* Female sex = (a = # females on DAA who achieved SVR/c = # males on DAA who achieved SVR)/(b = # females on DAA who failed/d = # males on DAA who failed); Genotypes = (a = # of specific genotype on DAA who achieved SVR/c = # of all other genotypes combined on DAA who achieved SVR)/(b = # desired genotype on DAA who failed /d = # all other genotypes combined on DAA who failed)

SVR = Sustained virologic response; DAA = Direct acting antiviral; HCV = Hepatitis C virus

Patient retention in HCV antiviral treatment recipients was evaluated (Table 5). Twenty-seven percent (10/37) of telemedicine patients were lost to follow-up compared with 5% (39/815) of outpatient clinic patients and 11% (7/62) of mixed-delivery patients (p < 0.0001). Multivariate analysis showed that HCV care delivery by TOHVHP telemedicine and mixed delivery was associated with patient retention challenges. Older age at assessment and experiencing the greatest quintile of social deprivation were positively associated with patient retention. Patient retention was similar by sex.

Table 5:

Odds ratios (ORs) of HCV care delivery method and potential covariates among those retained in HCV care

Variable	(a/c)/(b/d)*	Unadjusted OR (95% CI)	Adjusted OR (95% CI)	p value
Care delivery method (Ref.: clinic)		0.14 (0.06 to 0.31)	0.13 (0.06 to 0.31)	<0.0001
TM	(27/816)/(10/46)	0.40 (0.17 to 0.94)	0.31 (0.13 to 0.74)	0.009
Mixed delivery	(55/788)/(7/49)
Female	(284/559)/(17/39)		1.30 (0.68 to 2.48)	0.4
Older age	—		1.05 (1.02 to 1.08)	0.003
Greatest social deprivation (Ref: norm)	(202/547)/(6/43)		2.57 (1.06 to 6.26)	0.04

Note: Patients achieved SVR and received DAAs using per-protocol analysis

* Telemedicine (TM) = (a = # TM retained in care/c = # Mixed Delivery (MD) + Outpatient Clinic (OPC) retained in care)/(b = # TM lost-to-follow-up (LTFU)/d = # MD and OPC LTFU); MD = (a = # MD retained in care/c = #TM and OPC retained in care)/(b = # MD LTFU/d = # TM and OPC LTFU); Female Sex = (a = # females retained in care/c = # males retained in care)/(b = # females LTFU/d = # males LTFU); Greatest Social Deprivation = (a = # individuals experiencing greatest social deprivation retained in care/c = # individuals experiencing normal range social deprivation retained in care)/(b = # individuals experiencing greatest social deprivation LTFU/d = # individuals experiencing normal range social deprivation LTFU)

HCV = Hepatitis C virus; TM = Telemedicine; MD = Mixed delivery; OPC = Outpatient clinic; LTFU = Lost to follow-up

We restricted our analysis to rural-based patients (n = 160). Telemedicine patients were less likely to initiate therapy (36%) than outpatient clinic (72%) and mixed-delivery (75%) patients (p < 0.0001). For telemedicine, clinic, and mixed-delivery patients, ITT SVR (86%, 91%, and 91%, respectively; p = 0.7) and per-protocol SVR (100%, 96%, 100%, respectively; p = 1.0) were similar to that of the overall population. Loss to follow-up among recipients of HCV antiviral treatment was poorer among telemedicine patients (21%) versus 4% for outpatient clinic and 9% for mixed-delivery patients, respectively; p = 0.06).

Discussion

Our findings demonstrate that, assuming an appropriate level of resourcing and sustained attention to implementation, telemedicine programs can successfully engage highly vulnerable groups of patients. Patients located in rural settings, with low education levels, mental health challenges, recreational substance use, and a history of incarceration were all successfully engaged, treated, and cured by TOHVHP telemedicine program (3,10,11,23–25). This is particularly relevant for rural settings, where primary care expertise in HCV management is scarce. One benefit of using telemedicine in the context of HCV is that it can develop and maintain high-level primary care provider competence in the realm of HCV treatment (26,27).

Indigenous Canadian populations are disproportionately at risk for blood-borne infections, including HCV (28–31). Our telemedicine program demonstrated a measure of success in engaging Indigenous patients and in achieving HCV cure. Cultural factors may affect how Indigenous patients use health services. We believe that further improvements may be achieved by incorporating cultural competency training among staff (32–34) and by exploring models of HCV care that embody traditional medicine approaches to health (35).

We found that our TOHVHP telemedicine-delivered HCV care was capable of achieving SVR rates similar to those of our outpatient clinic and mixed-delivery patient populations. Our results are consistent with those of previous research that evaluated the use of telemedicine services among rural and remotely situated populations in Project Extension for Community Healthcare Outcomes (ECHO) and Veterans Affairs–ECHO. This research found that telemedicine-delivered HCV care achieved clinical outcomes similar to face-to-face, in-clinic care (15,16,36). The demonstrated success of our telemedicine-delivered HCV care among TOHVHP patients is particularly important during a period of economic stress associated with high costs of health care utilization. Our results may suggest that incorporating HCV care into existing telemedicine services may be a viable means to decrease HCV-related health care expenditure by reducing both clinic operating costs and travel costs incurred by patients for attending HCV appointments.

Loss-to-follow-up rates among those initiating DAA treatment were higher among telemedicine and mixed-delivery patients. Although it appears loss-to-follow-up rates were disproportionately high among telemedicine patients, rates were consistent with those observed among DAA recipients in a comparable population and health care setting (7). Increased loss-to-follow-up rates in the telemedicine program is likely the result of the high barriers to HCV care engagement observed in our cohort. These results demonstrate the need to further enrich telemedicine programs in services targeting vulnerable populations (37), including mobile harm reduction services for people who inject drugs, needle exchange programs, and opioid replacement therapies to improve patient retention (38).

Experiencing the greatest amount of social deprivation was associated with increased likelihood of being retained in TOHVHP HCV care. This finding was unexpected; in previous literature, experiencing the greatest level of social deprivation, an indicator of low socio-economic status (39), was not associated with increased patient retention. This may be a reflection of a health resource allocation bias in our program because additional supports are routinely offered to patients with unstable housing or limited social supports at the time of patient intake (40).

Using verified, high-quality data, our analysis was able to disentangle the effects of multiple and sometimes overlapping models used to deliver HCV care, a feature not considered in previous publications (36,41). Canadian health care is publicly funded. Therefore, personal finances were less likely to influence patient engagement in HCV treatment, irrespective of model of care. However, there were limitations that may have affected the findings of this retrospective cohort analysis, including small sample size in some subgroups and our methodology for calculating social and material deprivation. Using population data to assess individual social and material deprivation indices may have underestimated the effects of health inequity and the impact of socio-economic factors, particularly among our non-urban-based populations (42).

Patients were not randomized to outpatient clinic, mixed-delivery, or telemedicine care. We acknowledge that channelling biases influenced which patients ended up in each clinical stream. Nonetheless, we believe that there is merit in describing the similarities and differences in the populations served by each of these models of care. Reimbursement policy for HCV antiviral therapy was not consistent by jurisdiction or over the period of time evaluated. This factor likely contributed to the high mean fibrosis scores calculated in this patient cohort. We were not able to control for HCV antiviral reimbursement policy in our analysis, but it is plausible that it did influence treatment start decision and may have affected retention in care.

Conclusion

Our analyses suggest that telemedicine-delivered HCV care is capable of engaging patients facing multiple barriers to care and of achieving SVR rates comparable to those obtained with traditional outpatient HCV care. It is vital to build in strategies to minimize loss to follow-up in HCV populations that are more challenging to retain.

Competing Interests:

The authors report unrestricted program support from Merck, Gilead, and Abbvie.

Ethics Approval:

Patient data were collected from both patient charts and electronic patient files in The Ottawa Hospital Viral Hepatitis Clinical dataset (Ottawa Health Science Network Research Ethics Board REB 2004-196).

Informed Consent:

N/A

Registry and the Registration No. of the Study/Trial:

N/A

Animal Studies:

N/A

Funding:

This work was supported by Merck, Gilead, and Abbvie. They are acknowledged for unrestricted program support without any additional role in the conduct of our program and this analysis.

Peer Review:

This article has been peer reviewed.