Advertisement

Time Costs and Out-of-Pocket Costs in Patients With Chronic Hepatitis C in a Publicly Funded Health System

Open AccessPublished:September 22, 2021DOI:https://doi.org/10.1016/j.jval.2021.08.006

      Highlights

      • The patient-borne costs associated with hepatitis C infection are poorly understood. Most published cost-of-illness studies estimated only the direct medical costs of chronic hepatitis C.
      • So far, all productivity costs, out-of-pocket costs, and utility studies have been conducted in tertiary care clinics. There are no studies among patients receiving care in the community.
      • Our study reveals that the financial burden on patients with hepatitis C and their caregivers is high.
      • Given that patient-borne costs are critical in the economic evaluation of health interventions, this study is an important contribution to the economic assessment of current and future hepatitis C virus interventions.

      Abstract

      Objectives

      Chronic hepatitis C (CHC) infection affects more than 70 million people worldwide and imposes considerable health and economic burdens on patients and society. This study estimated 2 understudied components of the economic burden, patient out-of-pocket (OOP) costs and time costs, in patients with CHC in a tertiary hospital clinic setting and a community clinic setting.

      Methods

      This was a multicenter, cross-sectional study with hospital-based (n = 174) and community-based (n = 101) cohorts. We used a standardized instrument to collect healthcare resource use, time, and OOP costs. OOP costs included patient-borne costs for medical services, nonprescription drugs, and nonmedical expenses related to healthcare visits. Patient and caregiver time costs were estimated using an hourly wage value derived from patient-reported employment income and, where missing, derived from the Canadian census. Sensitivity analysis explored alternative methods of valuing time. Costs were reported in 2020 Canadian dollars.

      Results

      The mean 3-month OOP cost was $55 (95% confidence interval [CI] $21-$89) and $299 (95% CI $170-$427) for the community and hospital cohorts, respectively. The mean 3-month patient time cost was $743 (95% CI $485-$1002) (community) and $465 (95% CI $248-$682) (hospital). The mean 3-month caregiver time cost was $31 (95% CI $0-$63) (community) and $277 (95% CI $174-$380) (hospital). Patients with decompensated cirrhosis bore the highest costs.

      Conclusions

      OOP costs and patient and caregiver time costs represent a considerable economic burden to patient with CHC, equivalent to 14% and 21% of the reported total 3-month income for the hospital-based and community-based cohorts, respectively.

      Keywords

      Introduction

      Hepatitis C virus (HCV) affects more than 70 million people worldwide
      • Petruzziello A.
      • Marigliano S.
      • Loquercio G.
      • Cozzolino A.
      • Cacciapuoti C.
      Global epidemiology of hepatitis C virus infection: an up-date of the distribution and circulation of hepatitis C virus genotypes.
      ,
      Global hepatitis report, 2017. World Health Organization.
      and imposes a substantial health burden.
      • Kwong J.C.
      • Crowcroft N.S.
      • Campitelli M.S.
      • et al.
      Ontario burden of infectious disease study. Institute for Clinical Evaluative Sciences.
      Approximately 15% to 45% of those infected clear the infection spontaneously. The remainder progress to chronic infection.
      • Hoofnagle J.H.
      Course and outcome of hepatitis C.
      ,
      • Dintsios C.M.
      • Haverkamp A.
      • Wiegand J.
      • et al.
      Economic evaluation of early monotherapy versus delayed monotherapy or combination therapy in patients with acute hepatitis C in Germany.
      As many as 40% progress asymptomatically to cirrhosis and hepatocellular carcinoma (HCC), with an increased risk of premature death.
      • Sherman M.
      • Shafran S.
      • Burak K.
      • et al.
      Management of chronic hepatitis C: consensus guidelines.
      • Thein H.H.
      • Yi Q.
      • Dore G.J.
      • Krahn M.D.
      Estimation of stage-specific fibrosis progression rates in chronic hepatitis C virus infection: a meta-analysis and meta-regression.
      • Myers R.P.
      • Ramji A.
      • Bilodeau M.
      • Wong S.
      • Feld J.J.
      An update on the management of hepatitis C: consensus guidelines from the Canadian Association for the Study of the Liver.
      Fortunately, treatment for chronic hepatitis C (CHC) changed significantly with the advent of interferon-free direct-acting antiviral agents (DAAs) in 2014. The newer generations of DAAs are costly but very effective (>95% sustained virologic response rate) and well tolerated.
      • Burstow N.J.
      • Mohamed Z.
      • Gomaa A.I.
      • et al.
      Hepatitis C treatment: where are we now?.
      Despite their high costs, research has shown that DAAs are highly cost-effective.
      • Najafzadeh M.
      • Andersson K.
      • Shrank W.H.
      • et al.
      Cost-effectiveness of novel regimens for the treatment of hepatitis C virus.
      • Chhatwal J.
      • Kanwal F.
      • Roberts M.S.
      • Dunn M.A.
      Cost-effectiveness and budget impact of hepatitis C virus treatment with sofosbuvir and ledipasvir in the United States.
      • Linas B.P.
      • Barter D.M.
      • Morgan J.R.
      • et al.
      The cost-effectiveness of sofosbuvir-based regimens for treatment of hepatitis C virus genotype 2 or 3 infection.
      • Murphy G.
      • Farah B.
      • Wong W.
      • et al.
      Direct-acting antiviral agents for chronic hepatitis C genotype 1. Canadian Agency for Drugs and Technologies in Health.
      • He T.
      • Lopez-Olivo M.
      • Hur C.
      • Chhatwal J.
      Systematic review: cost-effectiveness of direct-acting antivirals for treatment of hepatitis C genotypes 2-6.
      • Nguyen J.
      • Barritt 4th, A.S.
      • Jhaveri R.
      Cost effectiveness of early treatment with direct-acting antiviral therapy in adolescent patients with hepatitis C virus infection.
      • Aggarwal R.
      • Chen Q.
      • Goel A.
      • et al.
      Cost-effectiveness of hepatitis C treatment using generic direct-acting antivirals available in India.
      • Rosenthal E.S.
      • Graham C.S.
      Price and affordability of direct-acting antiviral regimens for hepatitis C virus in the United States.
      Thus, DAAs have been approved and covered in many jurisdictions worldwide for treating patients with CHC.
      Nevertheless, the patient-borne costs associated with HCV infection are poorly understood. Most published cost-of-illness studies estimated only the direct medical costs of CHC.
      • Krajden M.
      • Kuo M.
      • Zagorski B.
      • Alvarez M.
      • Yu A.
      • Krahn M.
      Health care costs associated with hepatitis C: a longitudinal cohort study.
      • Rein D.B.
      • Borton J.
      • Liffmann D.K.
      • Wittenborn J.S.
      The burden of hepatitis C to the United States Medicare system in 2009: descriptive and economic characteristics.
      • Wong J.B.
      • McQuillan G.M.
      • McHutchison J.G.
      • Poynard T.
      Estimating future hepatitis C morbidity, mortality, and costs in the United States.
      • Xu F.
      • Tong X.
      • Leidner A.J.
      Hospitalizations and costs associated with hepatitis C and advanced liver disease continue to increase.
      The costs that are not covered by public or private insurance represent an important component of the burden of CHC from the perspectives of society, employers, patients, and caregivers. A few studies have evaluated out-of-pocket (OOP), caregiver, and productivity costs associated with CHC,
      • Leigh J.P.
      • Bowlus C.L.
      • Leistikow B.N.
      • Schenker M.
      Costs of hepatitis C.
      ,
      • Federico C.A.
      • Hsu P.C.
      • Krajden M.
      • et al.
      Patient time costs and out-of-pocket costs in hepatitis C.
      but they are outdated because patients were treated with PEGylated interferon. Given that DAAs have transformed care pathways and treatment outcomes, time, and OOP costs need to be re-estimated. These patient-borne costs are critical for economic evaluations assessing current and future HCV interventions. Recent economic evaluation guidelines in Canada and the United States call for a societal perspective, which requires incorporation of patient OOP costs and productivity loss.
      Guidelines for the economic evaluation of health technologies: Canada
      Canadian Agency for Drugs and Technologies in Health.
      ,
      • Sanders G.D.
      • Neumann P.J.
      • Basu A.
      • et al.
      Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: second panel on cost-effectiveness in health and medicine [published correction appears in JAMA. 2016;316(18):1924].
      Even in publicly funded healthcare systems such as those in Canada, patients have significant OOP costs.
      • Martin D.
      • Miller A.P.
      • Quesnel-Vallée A.
      • Caron N.R.
      • Vissandjée B.
      • Marchildon G.P.
      Canada’s universal health-care system: achieving its potential.
      For example, in Ontario, the Ontario Health Insurance Plan (OHIP) covers approximately 98.2% of the population.
      • Martin D.
      • Miller A.P.
      • Quesnel-Vallée A.
      • Caron N.R.
      • Vissandjée B.
      • Marchildon G.P.
      Canada’s universal health-care system: achieving its potential.
      OHIP covers medically necessary healthcare, including services provided by physicians and some other healthcare practitioners, emergency department visits, hospitalizations, and diagnostic and laboratory tests, with no copayments. Nevertheless, it does not cover many services provided by allied healthcare professionals, and it covers outpatient prescription drugs only for individuals aged ≥65 years, in long-term care, or on social assistance. Others must pay for outpatient prescription drugs OOP or through private supplemental insurance. At the time of this study, DAAs were new treatments that had not been included in the Ontario publicly funded drug formulary; patients had to apply for reimbursement through special authorization and would be reimbursed if they are eligible.
      • Tadrous M.
      • Mason K.
      • Dodd Z.
      • et al.
      Prescribing trends in direct-acting antivirals for the treatment of hepatitis C in Ontario, Canada.
      ,
      • Myers S.
      • Khosa G.
      • Kuo I.F.
      • Janzen D.
      • Alessi-Severini S.
      Moving towards universal coverage of direct-acting antiviral therapies for hepatitis C infection in Canada: an environmental scan of Canadian provinces and international jurisdictions.
      The objective of this study was to estimate the OOP and time costs associated with CHC healthcare for patients and their caregivers. We recruited patients with CHC from a tertiary care hospital setting and a community care setting that provided services to individuals who are marginalized by poverty or substance use.

      Methods

       Study Design

      We conducted a multicenter cross-sectional study with hospital-based and community-based cohorts of patients with CHC. The hospital-based cohort was recruited from the Toronto Centre for Liver Disease at the University Health Network, a tertiary care hospital in Toronto, Ontario, Canada. The community-based cohort was recruited from 3 community clinics of the Toronto Community Hepatitis C Program (Regent Park Community Health Centre, South Riverdale Community Health Centre, and Sherbourne Health). The Toronto Community Hepatitis C Program
      Toronto Community HEPC Program (TCHCP)
      South Riverdale Community Health Centre.
      provides low-threshold access to HCV treatment for underserved populations and support for people living in poverty or who use drugs or alcohol.
      • Mason K.
      • Dodd Z.
      • Guyton M.
      • et al.
      Understanding real-world adherence in the directly acting antiviral era: a prospective evaluation of adherence among people with a history of drug use at a community-based program in Toronto, Canada.
      Eligible patients had to have a confirmed diagnosis of CHC infection, with or without late-stage complications (eg, cirrhosis or HCC) and with or without previous treatment. We excluded patients who were unable or unwilling to provide consent or who had significant cognitive impairment (as judged by patient’s physician).

       Data Collection

      We used a modified version of a standardized resource use and cost instrument
      • Federico C.A.
      • Hsu P.C.
      • Krajden M.
      • et al.
      Patient time costs and out-of-pocket costs in hepatitis C.
      ,
      • de Oliveira C.
      • Bremner K.E.
      • Ni A.
      • Alibhai S.M.
      • Laporte A.
      • Krahn M.D.
      Patient time and out-of-pocket costs for long-term prostate cancer survivors in Ontario, Canada.
      to collect information about OOP costs and time spent obtaining healthcare for patients and their accompanying caregivers. The instrument covered the 3-month period before the survey and included patient-borne costs for medical services, procedures and tests, nonprescription CHC-related medications, home care and community services, and expenses for transportation, gas, parking, and childcare. Prescription drugs costs or DAAs costs were not part of our study. It also asked about time spent for inpatient hospitalizations and visits to emergency departments, ambulatory care clinics, and healthcare professionals, including associated travel, for patients and accompanying caregivers (see Appendix 1 in Supplemental Materials found at https://doi.org/10.1016/j.jval.2021.08.006). Patients also completed questionnaires on demographic characteristics and comorbid illness.
      Patients were recruited and completed the questionnaires once between February 2015 and March 2017. Most patients completed the questionnaires with a study coordinator, either at the clinic or over the phone. A study coordinator also reviewed medical charts to obtain clinical information and reports of laboratory and diagnostic tests. Charlson comorbidity index (CCI)
      • Charlson M.E.
      • Charlson R.E.
      • Peterson J.C.
      • Marinopoulos S.S.
      • Briggs W.M.
      • Hollenberg J.P.
      The Charlson Comorbidity Index is adapted to predict costs of chronic disease in primary care patients.
      ,
      • Charlson M.E.
      • Pompei P.
      • Ales K.L.
      • MacKenzie C.R.
      A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.
      and the index of coexistent disease (ICED)
      • Miskulin D.C.
      • Athienites N.V.
      • Yan G.
      • et al.
      Comorbidity assessment using the Index of Coexistent Diseases in a multicenter clinical trial.
      ,
      • Cleary P.D.
      • Greenfield S.
      • Mulley A.G.
      • et al.
      Variations in length of stay and outcomes for six medical and surgical conditions in Massachusetts and California.
      scores were calculated for each patient. Both indices were adjusted to exclude HCV-related conditions.

       Outcome Variables

      Our outcomes of interest were OOP costs and time costs associated with CHC care during the preceding 3 months, based on information collected in the questionnaire. The OOP costs included traveling and parking expenses related to healthcare visits, healthcare services not covered by OHIP, and nonprescription drug costs (Mina Tadrous, personal communication, 2020). The number of healthcare visits in the preceding 3 months was multiplied by the OOP cost of each visit reported by each patient, to estimate the 3-month cost for each healthcare professional for each patient. Because a previous study found that CHC-infected patients lost significantly workdays than noninfected individuals,
      • Su J.
      • Brook R.A.
      • Kleinman N.L.
      • Corey-Lisle P.
      The impact of hepatitis C virus infection on work absence, productivity, and healthcare benefit costs.
      it is important to measure the time costs associated with CHC care. Nevertheless, there is no universally accepted method for valuing time loss because of obtaining healthcare. One approach values time loss according to current self-reported income. Another approach values time using current average age- and sex-adjusted wage statistics for employed persons.
      • Federico C.A.
      • Hsu P.C.
      • Krajden M.
      • et al.
      Patient time costs and out-of-pocket costs in hepatitis C.
      This approach adjusts for potential inequity associated with market-based evaluations of time for the unemployed and retired and those working in unpaid positions.
      • van den Berg B.
      • Gafni A.
      • Portrait F.
      Attributing a monetary value to patients’ time: a contingent valuation approach.

       Analysis

      The sociodemographic and clinical characteristics of the study population were summarized using proportions, means, and standard deviations (SDs). The 3-month estimates of OOP costs, patient time costs, and caregiver time costs were summarized using sample means, sample SDs, and 95% confidence intervals (CIs) estimated using bootstrapping with resampling 500 times. All costs were adjusted to 2020 Canadian dollars using the health and personal care Consumer Price Index for Canada.
      Table 18-10-0005-01 Consumer Price Index, annual average, not seasonally adjusted. Statistics Canada.
      We used 1-way analysis of variance tests to compare OOP costs, time costs, work activity (working vs not working), hourly time values, time spent obtaining healthcare, nonprescription drug costs, and comorbidity indices among different levels of demographic variables and CHC disease severity.
      In our base case analysis, we estimated patient time cost by multiplying the patient-reported time (in hours) spent per healthcare visit (including round-trip travel, waiting, consultation, and procedures) by the reported number of visits, times an hourly wage value derived from the patient-reported gross employment income. For patients who reported “not working” status and did not report any employment income, the hourly value was estimated based on the age- and sex-stratified average employment income under the “Did not work” category reported by Statistics Canada from the 2016 census.
      Income Statistics (17), Occupation - National Occupational Classification (NOC) 2016 (13A), Work Activity During the Reference Year (9), Age (9) and Sex (3) for the Population Aged 15 Years and Over in Private Households of Canada, Provinces and Territories, Census Metropolitan Areas and Census Agglomerations, 2016 Census - 25% Sample Data. Statistics Canada.
      Caregiver time was valued in the same manner. In our sensitivity analysis, we explored alternative ways of valuing time.
      We used multivariable regression models using the backward stepwise variable selection method to identify individual significant predictors of OOP costs and time costs. The final model was selected based on the Bayesian information criterion. All analyses were performed using R (version 3.4.4).

       Sensitivity Analysis

      We also conducted 2 sensitivity analyses. In one sensitivity analysis, we used patient-reported gross employment income to derive the hourly time value and replaced missing values with the age- and sex-stratified median employment income distribution from the 2016 Canadian census.
      Income Statistics (17), Occupation - National Occupational Classification (NOC) 2016 (13A), Work Activity During the Reference Year (9), Age (9) and Sex (3) for the Population Aged 15 Years and Over in Private Households of Canada, Provinces and Territories, Census Metropolitan Areas and Census Agglomerations, 2016 Census - 25% Sample Data. Statistics Canada.
      In the second sensitivity analysis, we estimated the hourly time value for all patients based on the age- and sex-stratified average employment income distribution from the 2016 Canadian census.
      Income Statistics (17), Occupation - National Occupational Classification (NOC) 2016 (13A), Work Activity During the Reference Year (9), Age (9) and Sex (3) for the Population Aged 15 Years and Over in Private Households of Canada, Provinces and Territories, Census Metropolitan Areas and Census Agglomerations, 2016 Census - 25% Sample Data. Statistics Canada.

       Ethics Approval

      This study was approved by the Research Ethics Boards of the University of Toronto and the University Health Network, Toronto, Ontario, Canada.

      Results

       Patient Characteristics

      We recruited a total of 295 patients. A total of 20 were excluded because they did not complete the cost questionnaire. Therefore, 275 participants were included in the analyses, 101 (37%) from community clinics and 174 (63%) from the hospital clinics (Table 1). The mean age of the 275 participants was 56 years (range 18-86), and 63% were male. Only 37% were married or had a common-law partner. A total of 63% had completed high school, 70% were not working, and 70% reported an annual income of <$20 000.
      Table 1Characteristics of study population.
      CharacteristicsTotalCommunity clinic cohortHospital clinic cohortP
      P values in chi-square test for categorical variables; in 1-way ANOVA test for continuous variables.
      N = 275n = 101n = 174
      Age, mean ± SD55.8 ± 10.851.3 ± 9.558.4 ± 10.6<.001
      Male, n (%)172 (63)68 (67)104 (60).263
      White, n (%)186 (69)77 (77)109 (65).045
      Married or common-law, n (%)99 (37)8 (8)91 (55)<.001
      With caregivers, n (%)78 (28)5 (5)73 (48)<.001
      Completed high school, n (%)169 (63)55 (55)114 (69).035
      Not working, n (%)194 (70)84 (83)110 (63).001
      Annual income from all sources, 2020 Can$, mean ± SD23 916 ± 51 10615 439 ± 13 69028 836 ± 62 944.036
      Annual income, 2020 Can$, n (%)<.001
       0-19 999193 (70)90 (89)103 (59)
       20 000-39 99948 (17.5)8 (8)40(23)
       40 000-59 99915 (5.5)0 (0)15 (9)
       ≥60 00019 (7)3 (3)16(9)
      Monthly government support, mean ± SD$809 ± 770$982 ± 455$708 ± 889.004
      Clinical severity states, n (%)<.001
       No cirrhosis149 (54)76 (75)73 (42)
       Compensated cirrhosis55 (20)22 (22)33 (19)
       Decompensated cirrhosis33 (12)3 (3)30 (17)
       Hepatocellular carcinoma38 (14)0 (0)38 (22)
      Clinical variables
       ALT IU/liter, mean ± SD63.8 ± 57.371.7 ± 69.759.6 ± 49.0.100
       CTP scores,
      Only calculated for patients with cirrhosis (N = 126).
      mean ± SD
      6.36 ± 1.655.46 ± 0.836.58 ± 1.73.003
       MELD scores,
      Only calculated for patients with cirrhosis (N = 126).
      mean ± SD
      7.49 ± 5.283.78 ± 2.718.34 ± 5.36<.001
      Previous HCV treatment, n (%)94 (36)8 (7)86 (50)<.001
      Probable HCV transmission mode, n (%)<.001
       Blood transfusion33 (12)8 (8)25 (14)
       Injection drug use92 (34)57 (56)35 (20)
       Needle related injury26 (9)6 (6)20 (11)
       Sex22 (8)10 (10)12 (7)
       Other47 (17)12 (12)35 (20)
       Do not know55 (20)8 (8)47 (27)
      Problematic substance use (alcohol or drugs), n (%)145 (53)95 (94)50 (29)<.001
      Mental health diagnosis, n (%)125 (45)68 (67)57 (33)<.001
      ICED, mean ± SD2.12 ± 1.072.41 ± 0.981.86 ± 1.09<.001
      ICED categories, n (%)<.001
       030 (14)11 (11)19 (16)
       123 (11)3 (3)20 (17)
       256 (26)21 (21)35 (30)
       3108 (50)66 (65)42 (36)
      CCI, mean ± SD1.08 ± 1.610.85 ± 1.491.21 ± 1.66.072
      CCI categories, n (%).021
       0126 (46)51 (52)75 (43)
       184 (31)35 (36)49 (28)
       233 (12)6 (6)27 (16)
       3+29 (11)6 (6)23 (13)
      ALT indicates alanine aminotransferase; ANOVA, analysis of variance; CCI, Charlson comorbidity index; CTP, Child-Turcotte-Pugh; HCV, hepatitis C virus; ICED, index of coexistent disease; MELD, Model for End-Stage Liver Disease.
      P values in chi-square test for categorical variables; in 1-way ANOVA test for continuous variables.
      Only calculated for patients with cirrhosis (N = 126).
      Using the chart review data, we classified patients into 4 liver disease states: no cirrhosis (54%), compensated cirrhosis (20%), decompensated cirrhosis (DC) (12%), and HCC (14%). A total of 36% of patients were previously treated with PEGylated interferon. No patient was on HCV drug treatment at the time of the survey. A total of 12% of participants thought that they acquired HCV through blood transfusion and 34% ascribed HCV exposure to injection drug use. A total of 53% reported past or present problematic drug (22%), alcohol use (4%), or both (27%). We defined problematic use as continuous use on a daily basis for at least 4 weeks. A total of 45% reported having past or current psychiatric diagnosis requiring specialist treatment or hospitalization. The overall mean ICED score was 2.12 ± 1.07 (SD), and the overall mean CCI score was 1.08 ± 1.61 (SD).
      Compared with the hospital cohort, the community cohort was younger (51 ± 10 [SD] vs 58 ± 11 [SD] years, P<.001), had less formal education, was less frequently working, and was less likely to be married or have a common-law partner or have a caregiver. The community cohort also had a lower mean income, and more of them received social assistance and had problematic drug/alcohol use (94% vs 29%) and mental health diagnoses (67% vs 33%) (Table 1). Nevertheless, their liver disease was less severe. Among patients with cirrhosis, the community cohort had lower Child-Turcotte-Pugh scores and Model for End-Stage Liver Disease scores than the hospital cohort. The treatment guideline for the community hepatitis C clinics was to refer people with advanced liver disease to tertiary care center. Nevertheless, the patients in the community clinic cohort had a higher mean ICED scores (2.41 ± 0.98 [SD] vs 1.86 ± 1.09 [SD], P < .001) and slightly lower CCI (0.85 ± 1.49 [SD] vs 1.21 ± 1.66 [SD]).

       OOP Costs

      The mean 3-month OOP cost (Table 2) of the hospital clinic cohort was 5 times higher than that of the community clinic cohort ($299 [95% CI $170-$427] vs $55 [95% CI $21-$89]). Approximately 60% of the OOP costs were for traveling and parking related to healthcare visits, 20% for healthcare services not covered by OHIP, and 20% for nonprescription drugs. Only 17% of patients reported costs for nonprescription medications. The community cohort did not incur any traveling costs to the community hepatitis C clinic because they were borne by the community hepatitis C program.
      Table 2Mean 3-month out-of-pocket and time costs for patients with chronic hepatitis C and their caregivers by disease states.
      Clinical severity statesnOut-of-pocket cost
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare (Statistics Canada—Table 18-10-0004-08).
      Patient time cost
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare (Statistics Canada—Table 18-10-0004-08).
      ,
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      (base case: patient-reported employment income)
      Caregiver time cost
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare (Statistics Canada—Table 18-10-0004-08).
      ,
      Time value assigned to 78 caregivers according to the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302). A total of 22 patients did not answer questions on caregivers and were considered as missing data.
      Estimate ± SD95% CIEstimate ± SD95% CIEstimate ± SD95% CI
      Community clinic cohort10155 ± 17221-89743 ± 1340485-100231 ± 1660-63
       No cirrhosis7643 ± 11218-68786 ± 1521431-113928 ± 172–12 to 68
       Compensated cirrhosis2297 ± 308–23 to 217583 ± 491382-78347 ± 158–22 to 115
       Decompensated cirrhosis340 ± 367-72844 ± 308552-1131
       Hepatocellular carcinoma0
      Hospital clinic cohort174299 ± 875170-427465 ± 1504248-682277 ± 657174-380
       No cirrhosis73404 ± 1270109-700390 ± 152255-724148 ± 58918-276
       Compensated cirrhosis33119 ± 20451-188360 ± 89654-666156 ± 34835-277
       Decompensated cirrhosis30310 ± 518136-482972 ± 249991-1853672 ± 1035275-1069
       Hepatocellular carcinoma38243 ± 374118-368300 ± 473152-447348 ± 483179-517
      Note. 95% CI estimated based on the bootstrapping method with resampling B = 500 times.
      CI indicates confidence interval; no., number.
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare (Statistics Canada—Table 18-10-0004-08).
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      Time value assigned to 78 caregivers according to the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302). A total of 22 patients did not answer questions on caregivers and were considered as missing data.

       Time Costs

      In the base case analysis, the mean hourly value for patient time was $15.02 and the mean 3-month time cost were $743 (95% CI $485-$1002) and $465 (95% CI $248-$682) for the hospital clinic cohort and the community clinic cohort, respectively (Table 2). The highest patient mean time cost was in the DC group (hospital clinic cohort $972 [95% CI $91-$1853], community clinic cohort $844 [95% CI $552-$1131]).
      A total of 78 patients (28%) reported having caregivers, 5 (5%) of the community clinic cohort and 73 (42%) of the hospital clinic cohort. Only 30 patients (all in the hospital clinic cohort) reported that their caregivers were working. The caregiver time cost of the hospital clinic cohort was almost 9 times that of the community clinic cohort ($277 [95% CI $174-$380] vs $31 [95% CI $0-$63]).
      Relationships among cohort, sociodemographic and clinical characteristics, and OOP and time costs are presented in Table 3.
      Table 3Mean 3-month costs by sociodemographic and clinical characteristics (1-way ANOVA tests).
      CharacteristicsnOOP cost

      $
      All $ values adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08).
      (SD)
      Patient time cost

      $
      All $ values adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08).
      (SD)
      Caregiver time cost

      $
      All $ values adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08).
      (SD)
      Not working (%)Hourly time value
      The time value for 64 patients who reported employment income was based on their reported income. The time value of 211 patients without employment income data was estimated from the average employment income distribution from the 2016 Canadian census. (Source: Statistics Canada, Table 11-10-0240-01 Distribution of employment income of individuals by sex and work activity, Canada, provinces and selected census metropolitan areas. DOI: https://doi.org/10.25318/1110024001-eng).
      $
      All $ values adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08).
      (SD)
      Hours spent in healthcare per 3 months (SD)Number of visits per 3 months (SD)Nonprescription drug cost
      Nonprescription (over-the-counter) drug price based on the Canadian wholesale price list accessed on March 24, 2020.


      $
      All $ values adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08).
      (SD)
      Cohort
       Community clinic10155 (172)743 (1340)31 (166)8312 (8)68 (107)18 (15)13 (84)
       Hospital clinic174299 (875)
      P < .01, statistical values in ANOVA.
      465 (1504)277 (657)
      P < .001, statistical values in ANOVA.
      63
      P < .01, statistical values in ANOVA.
      17 (34)33 (122)
      P < .05, statistical values in ANOVA.
      5 (5)
      P < .001, statistical values in ANOVA.
      55 (332)
      Age categories, years
       <4020152 (248)396 (621)182 (443)8010 (5)39 (49)10 (6)41 (175)
       40-4940238 (576)792 (1761)49 (205)7016 (8)52 (99)11 (10)6 (16)
       50-59113271 (1011)530 (1162)143 (463)7115 (11)37 (58)11 (14)83 (411)
       60+101142 (398)541 (1709)279 (693)6816 (44)54 (172)7 (10)
      P < .05, statistical values in ANOVA.
      5 (20)
      Sex
       Male172203 (805)663 (1611)167 (480)7017 (34)45 (90)9 (11)43 (284)
       Female103219 (528)407 (1119)200 (614)7211 (9)48 (154)10 (12)35 (245)
      Ethnicity
       Non-white83146 (343)312 (420)167 (334)7612 (9)31 (43)8 (8)35 (260)
       White186239 (834)690 (1728)171 (565)6916 (33)53 (140)10 (12)43 (278)
      Marital status
       Single/divorced/widowed167189 (829)582 (1389)107 (426)7812 (9)55 (144)12 (12)35 (228)
       Married or common-law99252 (499)571 (1611)278 (612)
      P < .05, statistical values in ANOVA.
      61
      P < .01, statistical values in ANOVA.
      21 (44)
      P < .05, statistical values in ANOVA.
      33 (58)5 (6)
      P < .001, statistical values in ANOVA.
      52 (338)
      Highest education
       Attended high school or less97150 (431)533 (1386)145 (353)8012 (11)42 (71)11 (14)60 (360)
       Completed high school52125 (352)547 (986)210 (470)8312 (10)40 (69)9 (12)5 (25)
       Attended postsecondary40329 (1502)769 (1992)206 (743)6527 (67)36 (34)12 (13)14 (55)
       Completed postsecondary77289 (573)559 (1555)153 (548)57
      P < .01, statistical values in ANOVA.
      14 (12)
      P < .05, statistical values in ANOVA.
      63 (198)7 (7)
      P < .05, statistical values in ANOVA.
      56 (306)
      Annual gross income, $
       0-19 999193232 (818)568 (1334)173 (504)8311 (8)55 (137)11 (13)40 (263)
       20 000-39 99948191 (453)197 (224)80 (253)5811 (7)22 (32)6 (6)57 (371)
       40 000-59 99915110 (158)400 (668)342 (684)3316 (10)29 (50)4 (3)15 (45)
       ≥60 0001998 (124)1621 (3313)
      P < .01, statistical values in ANOVA.
      367 (1008)0
      P < .001, statistical values in ANOVA.
      66 (89)
      P < .001, statistical values in ANOVA.
      31 (54)6 (10)
      P < .01, statistical values in ANOVA.
      17 (73)
      Work activity status
       Not working194235 (817)558 (1315)162 (478)10 (5)54 (137)11 (13)41 (272)
       Working81148 (354)588 (1740)223 (653)27 (48)
      P < .001, statistical values in ANOVA.
      27 (43)6 (7)
      P < .01, statistical values in ANOVA.
      38 (265)
      Supplemental health insurance
       No214208 (776)566 (1380)128 (420)7814 (30)50 (129)10 (13)49 (305)
       Yes50245 (466)655 (1851)353 (761)
      P < .01, statistical values in ANOVA.
      50
      P < .001, statistical values in ANOVA.
      20 (20)36 (67)6 (7)
      P < .05, statistical values in ANOVA.
      8 (27)
      Problematic substance use
       No130276 (928)326 (1041)266 (626)6416 (38)33 (137)5 (5)37 (264)
       Yes145149 (433)783 (1712)
      P < .01, statistical values in ANOVA.
      112 (441)
      P < .05, statistical values in ANOVA.
      77
      P < .05, statistical values in ANOVA.
      14 (12)57 (96)14 (14)
      P < .001, statistical values in ANOVA.
      43 (275)
      Mental health diagnosis
       No150224 (856)530 (1496)218 (585)6417 (36)48 (151)6 (8)23 (212)
       Yes125192 (491)612 (1398)137 (470)78
      P < .05, statistical values in ANOVA.
      13 (10)44 (56)13 (15)
      P < .001, statistical values in ANOVA.
      60 (325)
      Proximity to liver clinic (community/hospital), km
       <10123112 (366)535 (1220)58 (228)7415 (38)50 (98)13 (13)46 (265)
       10-99121202 (493)514 (1348)221 (553)7114 (12)44 (145)7 (10)42 (306)
       100+28676 (1788)
      P < .01, statistical values in ANOVA.
      954 (2519)466 (945)
      P < .001, statistical values in ANOVA.
      5719 (23)37 (60)4 (4)
      P < .001, statistical values in ANOVA.
      7 (23)
      Liver disease states
       No cirrhosis149220 (907)592 (1530)84 (423)6917 (35)46 (97)11 (14)43 (284)
       Compensated cirrhosis55110 (248)449 (763)111 (288)6915 (17)39 (54)9 (8)4 (21)
       Decompensated cirrhosis33285 (500)960 (2380)603 (1000)7312 (11)78 (260)6 (5)79 (412)
       Hepatocellular carcinoma38243 (374)300 (473)348 (483)
      P < .001, statistical values in ANOVA.
      7910 (9)28 (33)5 (5)
      P < .01, statistical values in ANOVA.
      44 (256)
      ANOVA indicates analysis of variance; OOP, out-of-pocket.
      All $ values adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08).
      The time value for 64 patients who reported employment income was based on their reported income. The time value of 211 patients without employment income data was estimated from the average employment income distribution from the 2016 Canadian census. (Source: Statistics Canada, Table 11-10-0240-01 Distribution of employment income of individuals by sex and work activity, Canada, provinces and selected census metropolitan areas. DOI: https://doi.org/10.25318/1110024001-eng).
      Nonprescription (over-the-counter) drug price based on the Canadian wholesale price list accessed on March 24, 2020.
      § P < .01, statistical values in ANOVA.
      P < .001, statistical values in ANOVA.
      P < .05, statistical values in ANOVA.
      Compared with the hospital clinic cohort, the mean caregiver time cost was significantly lower for the community clinic cohort ($31 ± 166 [SD] vs $277 ± 657 [SD]), as was the mean OOP cost ($55 ± 172 [SD] vs $299 ± 875 [SD]). Nevertheless, the community cohort had more than one and a half times mean patient time cost ($743 ± 1340 [SD] vs $465 ± 1504 [SD]), had many more healthcare visits (18 ± 15 [SD] vs 5 ± 5 [SD]), and spent more time obtaining healthcare (68 ± 107 [SD] hours vs 33 ± 122 [SD] hours). By design, the community clinic encouraged patients to return more often for psychosocial support in addition to medical care.
      The mean time cost for patients with problematic substance use was more than double that of those without ($783 ± 1712 [SD] vs $326 ± 1041 [SD]).
      The mean caregiver time cost was significantly affected by cohort, marital status, supplemental health insurance, problematic substance use, proximity to liver specialists, and severity of liver disease. Patients with a marital or common-law partner (37%) had a higher mean caregiver cost than single patients ($278 ±6127 [SD] vs $107 ± 426 [SD]) and fewer healthcare visits (5 ± 6 [SD] vs 12 ± 12 [SD]). Patients with problematic substance use (53%) had a much lower mean caregiver cost ($112 ± 441 [SD] vs $266 ± 626 [SD]), a higher rate of not working (77% vs 64%), and a higher number of visits (14 ± 14 [SD] vs 5 ± 5 [SD]). Of the 3 categories of patient-borne costs, only the mean caregiver time cost showed a direct relationship with the severity of the liver disease (no cirrhosis:compensated cirrhosis:DC:HCC = $84:$111:$603:$348).
      All costs were lower for patients who lived within 10 km of the hospital or community liver clinic than for those who had to travel ≥100 km to the clinic (OOP cost $112 ± 366 [SD] vs $676 ± 1788 [SD], patient time cost $535 ± 12 201 [SD] vs $954 ± 2519 [SD], caregiver time cost $58 ± 228 [SD] vs $466 ± 945 [SD]).

       Utilization and Costs Related to Healthcare Visits

      Both hepatitis C-related and hepatitis C-unrelated healthcare visits were included (Table 4). A total of 55% of the patients visited a family physician, and 37% visited medical specialists other than liver specialists. Patients traveled farther and incurred higher OOP costs to visit liver specialists and other medical specialists than to visit family physicians. Only 2% to 3% of patients visited physiotherapists, occupational therapists, naturopaths, herbalists, acupuncturists, or chiropractors, but these visits incurred the highest mean OOP costs, because they were not covered by OHIP. Inpatient hospitalization (11% of patients) accounted for the highest mean time cost of $2264 ± 2998 (SD) per patient admitted. By design, the community cohort included less severe liver disease and patients were not seen by liver specialists. The hospital clinic cohort used more services from medical specialists, ophthalmologists, physical and occupational therapists, and day surgery. More of the community clinic patients used emergency room services (23% vs 14%).
      Table 4Three-month healthcare utilization and costs (2020 Canadian dollars).
      Medical careCommunity liver clinicLiver specialistOther medical specialistFamily doctorEye doctorPsychiatrist/clinical psychologistsNurse/nurse practitioner
      Number (%) of patients receiving care101 (37)176 (64)102 (37)151 (55)16 (6)15 (5)15 (5)
      Community cohort (n = 101):hospital cohort (n = 174)101:02:17421:8155:963:139:611:4
      Total number of visits9001772624732293106
      Average number of visits per patient who received care9133167
      Mean time spent, hours (SD)28 (9)5 (4)11 (37)6 (9)3 (2)38 (101)6 (9)
      Mean time cost
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      for those receiving care (SD), $
      328 (232)74 (120)118 (390)60 (87)31 (39)626 (1831)41 (35)
      Mean OOP cost
      Healthcare visit-related OOP cost.
      for those receiving care (SD), $
      0 (0)62 (159)154 (817)28 (85)121 (368)41 (93)18 (45)
      Mean distance traveled, km (SD)109 (152)188 (634)181 (1000)61 (166)18 (22)156 (341)28 (51)
      Allied/ alterative healthcarePhysiotherapist/occupational therapistSocial workerMethadone clinicAcupuncturist/chiropractor/naturopath
      Number (%) of patients receiving care6 (2)13 (5)11 (4)9 (3)
      Community cohort (n = 101):hospital cohort (n = 174)0:68:510:12:7
      Total number of visits261128865
      Average number of visits per patient receiving care4987
      Mean time spent, hours (SD)15 (20)6 (8)12 (7)12 (20)
      Mean time cost
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      for those receiving care (SD), $
      165 (246)50 (71)122 (117)106 (187)
      Mean OOP cost
      Healthcare visit-related OOP cost.
      for those receiving care (SD), $
      438 (837)8 (16)20 (30)574 (721)
      Mean distance traveled, km (SD)83 (111)51 (113)139 (175)77 (90)
      Healthcare institutionsDay surgeryEmergency roomHospitalization
      Number (%) of patients receiving care17 (6)48 (17)31 (11)
      Community cohort (n = 101):hospital cohort (n = 174)3:1423:2511:20
      Total number of visits205633
      Average number of visits per patient receiving care111
      Mean time spent, hours (SD)10 (11)8 (10)182 (251)
      Mean time cost
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      for those receiving care (SD), $
      159 (230)100 (108)2264 (2998)
      Mean OOP cost
      Healthcare visit-related OOP cost.
      for those receiving care (SD), $
      59 (115)20 (34)95 (141)
      Mean distance traveled, km (SD)43 (100)9 (11)17 (26)
      no. indicates number; OOP out-of-pocket.
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      Healthcare visit-related OOP cost.

       Multivariable Models of Patient-Borne Costs

      Predictors were selected for our multivariable regression models using the backward stepwise variable selection method (Table 5). These variables were age, sex, education, liver disease severity states, cohort, marital status, natural log of reported gross monthly income, supplemental health insurance, problematic substance use, work activity status, and proximity to liver clinic. Only DC was significantly associated with OOP cost. Hospital cohort, marital status, and natural log income were significantly associated with patient time cost. DC and natural log income were significantly associated with caregiver time cost.
      Table 5Final multivariable regression models of indirect annual costs based on backward stepwise variable selection method.
      Out-of-pocket costPatient time costCaregiver time cost
      PredictorsEstimateSEP valuePredictorsEstimateSEP valuePredictorsEstimateSEP value
      (Intercept)23994.0142
      P < .05.
      (intercept)2447596.0001
      P < .001.
      (intercept)274118.025
      P < .05.
      Age
      Age was the age at the time of the survey and is a continuous variable.
      Age
      Age was the age at the time of the survey and is a continuous variable.
      −1310.172Age
      Age was the age at the time of the survey and is a continuous variable.
      Education
      Education is a binary variable with 0 = not completed high school, 1 = completed high school or more.
      −194104.068Education
      Education is a binary variable with 0 = not completed high school, 1 = completed high school or more.
      Education
      Education is a binary variable with 0 = not completed high school, 1 = completed high school or more.
      Decompensated cirrhosis
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      292124.022
      P < .05.
      Decompensated cirrhosis
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      Decompensated cirrhosis
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      17772.017
      P < .05.
      Hepatocellular carcinoma
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      123148.407Hepatocellular carcinoma
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      Hepatocellular carcinoma
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      −7686.377
      Hospital cohort
      Cohort is a binary variable with 0 = community clinic, 1 = hospital clinic.
      Hospital cohort
      Cohort is a binary variable with 0 = community clinic, 1 = hospital clinic.
      –372166.029
      P < .05.
      Hospital cohort
      Cohort is a binary variable with 0 = community clinic, 1 = hospital clinic.
      Marital status
      Marital status is a binary variable with 0 = single/separated/divorced, 1 = married or common-law.
      −166113.147Marital status
      Marital status is a binary variable with 0 = single/separated/divorced, 1 = married or common-law.
      321146.032
      P < .05.
      Marital status
      Marital status is a binary variable with 0 = single/separated/divorced, 1 = married or common-law.
      11763.070
      ln_income
      Ln income is the natural log of gross monthly income and is a continuous variable.
      ln_income
      Ln income is the natural log of gross monthly income and is a continuous variable.
      −11437.003
      P < .01.
      ln_income
      Ln income is the natural log of gross monthly income and is a continuous variable.
      −3716.024
      P < .05.
      Supplemental Insurance
      Supplemental Insurance is a binary variable with 0 = without private supplemental insurance, 1 = with private supplemental insurance.
      Supplemental Insurance
      Supplemental Insurance is a binary variable with 0 = without private supplemental insurance, 1 = with private supplemental insurance.
      Supplemental Insurance
      Supplemental Insurance is a binary variable with 0 = without private supplemental insurance, 1 = with private supplemental insurance.
      10970.122
      Working status
      Working status is a binary variable with 0 = not working, 1 = working.
      195124.121Working status
      Working status is a binary variable with 0 = not working, 1 = working.
      Working status
      Working status is a binary variable with 0 = not working, 1 = working.
      Model statisticsModel statisticsModel statistics
      Multiple R-squared = 0.1823Multiple R-squared = 0.2598Multiple R-squared = 0.301
      Adjusted R-squared = 0.1065Adjusted R-squared = 0.206Adjusted R-squared = 0.2363
      F-statistic (df) = 2.407 on 5 and 54 dfF-statistic (df) = 4.826 on 4 and 55 dfF-statistic (df) = 4.653 on 5 and 54 df
      P = .048
      P < .05.
      P = .0021
      P < .01.
      P = .0013
      P < .01.
      SE indicates standard error.
      P < .05.
      P < .001.
      Age was the age at the time of the survey and is a continuous variable.
      § Education is a binary variable with 0 = not completed high school, 1 = completed high school or more.
      Disease severity state is a four-factor variable with 1 = no cirrhosis, 2 = compensated cirrhosis, 3 = decompensated cirrhosis, 4 = hepatocellular carcinoma.
      Cohort is a binary variable with 0 = community clinic, 1 = hospital clinic.
      £ Marital status is a binary variable with 0 = single/separated/divorced, 1 = married or common-law.
      ∗∗ P < .01.
      †† Ln income is the natural log of gross monthly income and is a continuous variable.
      ‡‡ Supplemental Insurance is a binary variable with 0 = without private supplemental insurance, 1 = with private supplemental insurance.
      §§ Working status is a binary variable with 0 = not working, 1 = working.

       Sensitivity Analyses

      Applying different methods of valuing patient time (Table 6) resulted in the estimates of the mean hourly value ranging from $10.16 to $15.39, and the mean 3-month patient time cost for the community clinic cohort ranged from $345 (95% CI $193-$498) to $847 (95% CI $573-$1120). The estimates for the hospital clinic cohort ranged from $303 (95% CI $124-$482) to $465 (95% CI $248-$682).
      Table 6Sensitivity analyses: effect of different methods of valuing time on 3-month patient time costs.
      Methods of valuing timeAverage hourly value
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08)
      ($)
      Community clinic cohort (n = 101)Hospital clinic cohort (n = 174)
      Mean cost estimate
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08)
      ($)
      95% CIMean cost estimate
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08)
      ($)
      95% CI
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      Base case: Patient-reported employment income (n = 64) and average employment income estimated for patients with no reported employment income (n = 211)
      15.02743485-1002465248-682
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the median employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      Patient-reported employment income (n = 64) and median employment income estimated for patients with no reported employment income (n = 211)
      10.16345193-498303124-482
      Time value was estimated for all patients (N = 275) from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      Average employment income estimated for all patients (N = 275)
      15.39847573-1120458275-640
      Note. 95% CI estimated based on the bootstrapping method with resampling B = 500.
      CI indicates confidence interval; no., number.
      All cost estimates adjusted to 2020 Canadian dollar value using Consumer Price Index for healthcare for Canada (Statistics Canada—Table 18-10-0004-08)
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      Time value based on patient-reported employment income for 64 patients who reported employment income. The time value of the 211 patients without employment income data was estimated from the median employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).
      § Time value was estimated for all patients (N = 275) from the average employment income distribution by age, sex, and work activity from the 2016 Statistics Canada census (Source: Statistics Canada, 2016 Census of Population, Statistics Canada catalog no. 98-400-X2016302).

      Discussion

       Patient-Borne Costs Related to HCV Are High

      The 2010 Ontario burden of infectious disease study suggested that HCV had the highest health burden of any infectious disease in Ontario in terms of health-adjusted life-years.
      • Kwong J.C.
      • Crowcroft N.S.
      • Campitelli M.S.
      • et al.
      Ontario burden of infectious disease study. Institute for Clinical Evaluative Sciences.
      Another study suggested that the financial burden of HCV to the health system is also high.
      • Wong W.W.L.
      • Haines A.
      • Bremner K.E.
      • et al.
      Health care costs associated with chronic hepatitis C virus infection in Ontario, Canada: a population-based analysis.
      Our study reveals that the financial burden on patients is also high: 3-month OOP costs and patient time costs represented 10.6% and 20.7% of the patient-reported total income for the hospital-based and community-based cohorts, respectively. The burden on caregivers was also considerable. Overall, 28% of patients had caregivers, and 38% of caregivers worked and reported missing from 3 to 120 hours of work, with an estimated 3-month time cost of $277 and $31 for the hospital clinic cohort and the community clinic cohort, respectively.

       Severity of Liver Disease

      It was not surprising to find that patients with DC incurred higher costs than patients in other severity states. The small sample of patients in our study with HCC had lower than expected patient time cost, likely because 82% of them had early-stage HCC (66% Barcelona Clinic Liver Cancer stage A and 16% stage B).

       Hospital-Based Cohort Versus Community-Based Cohort

      The estimated patient time costs for our community cohort ($743) was 1.6 times that of the hospital cohort ($465). The community cohort had more healthcare visits (18 ± 15 [SD] vs 5 ± 5 [SD]) and spent more time on healthcare (68 ± 107 [SD] hours vs 33 ± 122 [SD] hours). This was partly because the design of the community model was to hold weekly clinics and support groups to keep patients engaged and provide additional health education and social and medical support. Circumstances such as problematic substance use and mental health issues also contributed to higher healthcare use for the community cohort. For example, there were 9 community clinic patients (8.9%) who spent 511 hours in total in 3 months on mental health services, compared with 6 hospital clinic patients (3.4%) who spent 59 hours. There were 10 community clinic patients (9.9%) who spent 123 hours in total in 3 months attending methadone clinic compared with only 1 hospital clinic patient (0.6%) who spent 6 hours. Finally, it is worth noting that only 5% of the community cohort reported having caregivers, compared with 42% of the hospital cohort, reflecting the lack of social support for this cohort. Like all infectious diseases, the impact on the community cohort was disproportionally higher than on the hospital cohort.

       Comparison With Other Studies

      Costing studies are often difficult to compare because of differences in healthcare systems, treatment patterns, costing periods, services, and populations studied. Our previous study in a hospital clinic cohort in Vancouver, Canada, in 2007
      • Federico C.A.
      • Hsu P.C.
      • Krajden M.
      • et al.
      Patient time costs and out-of-pocket costs in hepatitis C.
      estimated the 3-month costs to be $332 for OOP, $534 for patient time, and $93 for caregiver time. After adjusting for inflation,
      Table 18-10-0005-01 Consumer Price Index, annual average, not seasonally adjusted. Statistics Canada.
      the present estimates for our hospital cohort represent 76% of the OOP costs ($299 vs $393), 73% of the patient time costs ($465 vs $634), and 250% of the caregiver time costs ($277 vs $111) that were reported in that study. The cost differences may be because of different care patterns across provinces, and updated treatments and care pathways after DAAs became available. In contrast, our findings of high rates of unemployment and mental health issues were congruent with other recent studies. A US study confirmed a lower employment rate among patients with CHC than the national average (54% vs 62%).
      • Stepanova M.
      • Younossi Z.M.
      Economic burden of hepatitis C infection.
      Only 35.5% of patients with HCV had a paid job in Italy.
      • Ciampichini R.
      • Scalone L.
      • Fagiuoli S.
      • et al.
      Societal burden in hepatitis C patients: the COME study results.
      Other studies confirmed the high rates of mental and emotional health issues
      • Younossi I.
      • Weinstein A.
      • Stepanova M.
      • Hunt S.
      • Younossi Z.M.
      Mental and emotional impairment in patients with hepatitis C is related to lower work productivity.
      ,
      • Stepanova M.
      • De Avila L.
      • Afendy M.
      • et al.
      Direct and indirect economic burden of chronic liver disease in the United States.
      and low work productivity among patients with CHC.
      • Liu G.G.
      • DiBonaventura Md
      • Yuan Y.
      • et al.
      The burden of illness for patients with viral hepatitis C: evidence from a national survey in Japan.
      ,
      • Younossi Z.M.
      • Stepanova M.
      • Henry L.
      • et al.
      Association of work productivity with clinical and patient-reported factors in patients infected with hepatitis C virus.
      Compared with other diseases, the mean 3-month patient time cost in our hospital-based cohort was roughly twice that reported in prostate cancer survivors ($210 in 2006, equivalent to $261 in 2020),
      • de Oliveira C.
      • Bremner K.E.
      • Ni A.
      • Alibhai S.M.
      • Laporte A.
      • Krahn M.D.
      Patient time and out-of-pocket costs for long-term prostate cancer survivors in Ontario, Canada.
      ,
      Table 18-10-0005-01 Consumer Price Index, annual average, not seasonally adjusted. Statistics Canada.
      and the mean 3-month OOP cost was 4 times higher than reported for prostate cancer survivors ($50 in 2006, equivalent to $62 in 2020).
      • de Oliveira C.
      • Bremner K.E.
      • Ni A.
      • Alibhai S.M.
      • Laporte A.
      • Krahn M.D.
      Patient time and out-of-pocket costs for long-term prostate cancer survivors in Ontario, Canada.
      ,
      Table 18-10-0005-01 Consumer Price Index, annual average, not seasonally adjusted. Statistics Canada.
      Our estimated patient time cost was comparable with that associated with breast cancer care, which was estimated to be $252 in 2002 US dollar,
      • Yabroff K.R.
      • Davis W.W.
      • Lamont E.B.
      • et al.
      Patient time costs associated with cancer care.
      equivalent to Can$508 in 2020.

      Strengths and Limitations

      Our study is unique in recruiting patients from the vulnerable community-based CHC population that is often under-represented in research studies. Nevertheless, in interpreting our results, we need to consider that our sample sizes for patients with DC and HCC were relatively small and most patients had mild disease. There was a high refusal rate from patients with advanced states of liver disease, who were not willing or unable to participate when they were very sick. Similar to all studies that used convenience sampling in ambulatory/outpatient clinics, our results may have underestimated or overestimated the costs of CHC disease states because of this unintended selection bias. In addition, patients may have acquired care or health services outside our research institutions; external validation of the reported cost and time was not possible. Thus, this self-reported retrospective survey design could potentially introduce recall and reporting biases. Furthermore, our questionnaire may not have covered all the time related to self-management, for example, the time associated with special food preparation or exercise that was recommended by health professionals. Finally, our cost data are most relevant to Canada in a particular community care setting; it may be difficult to generalize to other jurisdictions because of differences in healthcare systems. Nevertheless, the comprehensiveness of our study should allow analysts in other settings to better understand the implications of age, income, and particularly disease stage as predictors of patient-borne costs.

      Conclusions

      Our study showed that even in a jurisdiction with publicly paid universal health insurance like Canada, patients with CHC incur high OOP and time costs. In settings without publicly paid health insurance, these patient-borne costs would exacerbate because patients with CHC must also cover the high costs of DAAs and medically necessary healthcare.
      Irrespective of the care setting, OOP and time costs are considerable burdens to patients with CHC. Caregiver time is an additional cost, especially for patients with advanced disease. Early diagnosis of HCV and equitable and timely access to DAA treatment are essential to mitigate these disease burdens.

      Article and Author Information

      Author Contributions: Concept and design: W. Wong, Krahn
      Acquisition of data: J. Wong, Saeed, Mason, Phoon, Feld, Powis
      Analysis and interpretation of data: W. Wong, J. Wong, Bremner, Saeed, Mason, Feng, Feld, Mitsakakis, Powis, Krahn
      Drafting of the manuscript: W. Wong, J. Wong, Saeed, Bremner, Krahn
      Critical revision of the paper for important intellectual content: W. Wong, J. Wong, Bremner, Saeed, Mason, Phoon, Feng, Feld, Mitsakakis, Powis, Krahn
      Statistical analysis: Feng, Mitsakakis
      Provision of study materials or patients: Mason, Feld, Powis
      Obtaining funding: W. Wong, Krahn
      Administrative, technical, or logistic support: J. Wong, Phoon
      Supervision: W. Wong, Krahn
      Conflict of Interest Disclosures: Dr W. Wong reported receiving a grant from the Canadian Liver Foundation; a grant from the Ontario Ministry of Research, Innovation, and Science Early Researcher Award; and a grant from the Natural Sciences and Engineering Research Council. Dr Feld reported receiving grants and personal fees from AbbVie and grants and personal fees from Gilead. Dr Krahn reported receiving a grant from the Tier 1 Canada Research Chair in Health Technology Assessment and a grant from the Canadian Liver Foundation. No other disclosures were reported.
      Funding Support: This work was supported by grants 137490 and 148970 from the Canadian Institutes of Health Research .
      Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the article for publication.

      Acknowledgment

      The authors thank Dr Mina Tadrous, the staff of the Liver Clinic (University Health Network) and the Toronto Community Hepatitis C Program for their help and contribution to the study.

      Supplemental Material

      References

        • Petruzziello A.
        • Marigliano S.
        • Loquercio G.
        • Cozzolino A.
        • Cacciapuoti C.
        Global epidemiology of hepatitis C virus infection: an up-date of the distribution and circulation of hepatitis C virus genotypes.
        World J Gastroenterol. 2016; 22: 7824-7840
      1. Global hepatitis report, 2017. World Health Organization.
        • Kwong J.C.
        • Crowcroft N.S.
        • Campitelli M.S.
        • et al.
        Ontario burden of infectious disease study. Institute for Clinical Evaluative Sciences.
        • Hoofnagle J.H.
        Course and outcome of hepatitis C.
        Hepatology. 2002; 36: S21-S29
        • Dintsios C.M.
        • Haverkamp A.
        • Wiegand J.
        • et al.
        Economic evaluation of early monotherapy versus delayed monotherapy or combination therapy in patients with acute hepatitis C in Germany.
        Eur J Gastroenterol Hepatol. 2010; 22: 278-288
        • Sherman M.
        • Shafran S.
        • Burak K.
        • et al.
        Management of chronic hepatitis C: consensus guidelines.
        Can J Gastroenterol. 2007; 21: 25C-34C
        • Thein H.H.
        • Yi Q.
        • Dore G.J.
        • Krahn M.D.
        Estimation of stage-specific fibrosis progression rates in chronic hepatitis C virus infection: a meta-analysis and meta-regression.
        Hepatology. 2008; 48: 418-431
        • Myers R.P.
        • Ramji A.
        • Bilodeau M.
        • Wong S.
        • Feld J.J.
        An update on the management of hepatitis C: consensus guidelines from the Canadian Association for the Study of the Liver.
        Can J Gastroenterol. 2012; 26: 359-375
        • Burstow N.J.
        • Mohamed Z.
        • Gomaa A.I.
        • et al.
        Hepatitis C treatment: where are we now?.
        Int J Gen Med. 2017; 10: 39-52
        • Najafzadeh M.
        • Andersson K.
        • Shrank W.H.
        • et al.
        Cost-effectiveness of novel regimens for the treatment of hepatitis C virus.
        Ann Intern Med. 2015; 162: 407-419
        • Chhatwal J.
        • Kanwal F.
        • Roberts M.S.
        • Dunn M.A.
        Cost-effectiveness and budget impact of hepatitis C virus treatment with sofosbuvir and ledipasvir in the United States.
        Ann Intern Med. 2015; 162: 397-406
        • Linas B.P.
        • Barter D.M.
        • Morgan J.R.
        • et al.
        The cost-effectiveness of sofosbuvir-based regimens for treatment of hepatitis C virus genotype 2 or 3 infection.
        Ann Intern Med. 2015; 162: 619-629
        • Murphy G.
        • Farah B.
        • Wong W.
        • et al.
        Direct-acting antiviral agents for chronic hepatitis C genotype 1. Canadian Agency for Drugs and Technologies in Health.
        • He T.
        • Lopez-Olivo M.
        • Hur C.
        • Chhatwal J.
        Systematic review: cost-effectiveness of direct-acting antivirals for treatment of hepatitis C genotypes 2-6.
        Aliment Pharmacol Ther. 2017; 46: 711-721
        • Nguyen J.
        • Barritt 4th, A.S.
        • Jhaveri R.
        Cost effectiveness of early treatment with direct-acting antiviral therapy in adolescent patients with hepatitis C virus infection.
        J Pediatr. 2019; 207: 90-96
        • Aggarwal R.
        • Chen Q.
        • Goel A.
        • et al.
        Cost-effectiveness of hepatitis C treatment using generic direct-acting antivirals available in India.
        PLoS One. 2017; 12e0176503
        • Rosenthal E.S.
        • Graham C.S.
        Price and affordability of direct-acting antiviral regimens for hepatitis C virus in the United States.
        Infect Agents Cancer. 2016; 11: 24
        • Krajden M.
        • Kuo M.
        • Zagorski B.
        • Alvarez M.
        • Yu A.
        • Krahn M.
        Health care costs associated with hepatitis C: a longitudinal cohort study.
        Can J Gastroenterol. 2010; 24: 717-726
        • Rein D.B.
        • Borton J.
        • Liffmann D.K.
        • Wittenborn J.S.
        The burden of hepatitis C to the United States Medicare system in 2009: descriptive and economic characteristics.
        Hepatology. 2016; 63: 1135-1144
        • Wong J.B.
        • McQuillan G.M.
        • McHutchison J.G.
        • Poynard T.
        Estimating future hepatitis C morbidity, mortality, and costs in the United States.
        Am J Public Health. 2000; 90: 1562-1569
        • Xu F.
        • Tong X.
        • Leidner A.J.
        Hospitalizations and costs associated with hepatitis C and advanced liver disease continue to increase.
        Health Aff (Millwood). 2014; 33: 1728-1735
        • Leigh J.P.
        • Bowlus C.L.
        • Leistikow B.N.
        • Schenker M.
        Costs of hepatitis C.
        Arch Intern Med. 2001; 161: 2231-2237
        • Federico C.A.
        • Hsu P.C.
        • Krajden M.
        • et al.
        Patient time costs and out-of-pocket costs in hepatitis C.
        Liver Int. 2012; 32: 815-825
        • Guidelines for the economic evaluation of health technologies: Canada
        Canadian Agency for Drugs and Technologies in Health.
        • Sanders G.D.
        • Neumann P.J.
        • Basu A.
        • et al.
        Recommendations for conduct, methodological practices, and reporting of cost-effectiveness analyses: second panel on cost-effectiveness in health and medicine [published correction appears in JAMA. 2016;316(18):1924].
        JAMA. 2016; 316: 1093-1103
        • Martin D.
        • Miller A.P.
        • Quesnel-Vallée A.
        • Caron N.R.
        • Vissandjée B.
        • Marchildon G.P.
        Canada’s universal health-care system: achieving its potential.
        Lancet. 2018; 391: 1718-1735
        • Tadrous M.
        • Mason K.
        • Dodd Z.
        • et al.
        Prescribing trends in direct-acting antivirals for the treatment of hepatitis C in Ontario, Canada.
        Can Liver J. 2021; 4: 51-58
        • Myers S.
        • Khosa G.
        • Kuo I.F.
        • Janzen D.
        • Alessi-Severini S.
        Moving towards universal coverage of direct-acting antiviral therapies for hepatitis C infection in Canada: an environmental scan of Canadian provinces and international jurisdictions.
        J Pharm Sci. 2018; 21: 271s-308s
        • Toronto Community HEPC Program (TCHCP)
        South Riverdale Community Health Centre.
        • Mason K.
        • Dodd Z.
        • Guyton M.
        • et al.
        Understanding real-world adherence in the directly acting antiviral era: a prospective evaluation of adherence among people with a history of drug use at a community-based program in Toronto, Canada.
        Int J Drug Policy. 2017; 47: 202-208
        • de Oliveira C.
        • Bremner K.E.
        • Ni A.
        • Alibhai S.M.
        • Laporte A.
        • Krahn M.D.
        Patient time and out-of-pocket costs for long-term prostate cancer survivors in Ontario, Canada.
        J Cancer Surviv. 2014; 8: 9-20
        • Charlson M.E.
        • Charlson R.E.
        • Peterson J.C.
        • Marinopoulos S.S.
        • Briggs W.M.
        • Hollenberg J.P.
        The Charlson Comorbidity Index is adapted to predict costs of chronic disease in primary care patients.
        J Clin Epidemiol. 2008; 61: 1234-1240
        • Charlson M.E.
        • Pompei P.
        • Ales K.L.
        • MacKenzie C.R.
        A new method of classifying prognostic comorbidity in longitudinal studies: development and validation.
        J Chronic Dis. 1987; 40: 373-383
        • Miskulin D.C.
        • Athienites N.V.
        • Yan G.
        • et al.
        Comorbidity assessment using the Index of Coexistent Diseases in a multicenter clinical trial.
        Kidney Int. 2001; 60: 1498-1510
        • Cleary P.D.
        • Greenfield S.
        • Mulley A.G.
        • et al.
        Variations in length of stay and outcomes for six medical and surgical conditions in Massachusetts and California.
        JAMA. 1991; 266: 73-79
        • Su J.
        • Brook R.A.
        • Kleinman N.L.
        • Corey-Lisle P.
        The impact of hepatitis C virus infection on work absence, productivity, and healthcare benefit costs.
        Hepatology. 2010; 52: 436-442
        • van den Berg B.
        • Gafni A.
        • Portrait F.
        Attributing a monetary value to patients’ time: a contingent valuation approach.
        Soc Sci Med. 2017; 179: 182-190
      2. Table 18-10-0005-01 Consumer Price Index, annual average, not seasonally adjusted. Statistics Canada.
      3. Income Statistics (17), Occupation - National Occupational Classification (NOC) 2016 (13A), Work Activity During the Reference Year (9), Age (9) and Sex (3) for the Population Aged 15 Years and Over in Private Households of Canada, Provinces and Territories, Census Metropolitan Areas and Census Agglomerations, 2016 Census - 25% Sample Data. Statistics Canada.
        • Wong W.W.L.
        • Haines A.
        • Bremner K.E.
        • et al.
        Health care costs associated with chronic hepatitis C virus infection in Ontario, Canada: a population-based analysis.
        CMAJ Open. 2021; 9: E167-E174
        • Stepanova M.
        • Younossi Z.M.
        Economic burden of hepatitis C infection.
        Clin Liver Dis. 2017; 21: 579-594
        • Ciampichini R.
        • Scalone L.
        • Fagiuoli S.
        • et al.
        Societal burden in hepatitis C patients: the COME study results.
        Liver. 2012; 11: 3408-3440
        • Younossi I.
        • Weinstein A.
        • Stepanova M.
        • Hunt S.
        • Younossi Z.M.
        Mental and emotional impairment in patients with hepatitis C is related to lower work productivity.
        Psychosomatics. 2016; 57: 82-88
        • Stepanova M.
        • De Avila L.
        • Afendy M.
        • et al.
        Direct and indirect economic burden of chronic liver disease in the United States.
        Clin Gastroenterol Hepatol. 2017; 15: 759-766.e755
        • Liu G.G.
        • DiBonaventura Md
        • Yuan Y.
        • et al.
        The burden of illness for patients with viral hepatitis C: evidence from a national survey in Japan.
        Value Health. 2012; 15: S65-S71
        • Younossi Z.M.
        • Stepanova M.
        • Henry L.
        • et al.
        Association of work productivity with clinical and patient-reported factors in patients infected with hepatitis C virus.
        J Viral Hepat. 2016; 23: 623-630
        • Yabroff K.R.
        • Davis W.W.
        • Lamont E.B.
        • et al.
        Patient time costs associated with cancer care.
        J Natl Cancer Inst. 2007; 99: 14-23