Open Collections

UBC Faculty Research and Publications

The impact of HCV co-infection status on healthcare-related utilization among people living with HIV… Ma, Huiting; Villalobos, Conrado F; St-Jean, Martin; Eyawo, Oghenowede; Lavergne, Miriam R; Ti, Lianping; Hull, Mark W; Yip, Benita; Wu, Lang; Hogg, Robert S; Barrios, Rolando; Shoveller, Jean A; Montaner, Julio S G; Lima, Viviane D May 2, 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12913_2018_Article_3119.pdf [ 843.03kB ]
JSON: 52383-1.0366218.json
JSON-LD: 52383-1.0366218-ld.json
RDF/XML (Pretty): 52383-1.0366218-rdf.xml
RDF/JSON: 52383-1.0366218-rdf.json
Turtle: 52383-1.0366218-turtle.txt
N-Triples: 52383-1.0366218-rdf-ntriples.txt
Original Record: 52383-1.0366218-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessThe impact of HCV co-infection status onhealthcare-related utilization among peopleliving with HIV in British Columbia, Canada:a retrospective cohort studyHuiting Ma1, Conrado Franco Villalobos2, Martin St-Jean2, Oghenowede Eyawo2, Miriam Ruth Lavergne3,Lianping Ti2, Mark W. Hull4, Benita Yip2, Lang Wu5, Robert S. Hogg2, Rolando Barrios2, Jean A. Shoveller6,Julio S. G. Montaner4 and Viviane D. Lima2*AbstractBackground: The burden of HCV among those living with HIV remains a major public health challenge. We aimed tocharacterize trends in healthcare-related visits (HRV) of people living with HIV (PLW-HIV) and those living with HIV andHCV (PLW-HIV/HCV), in British Columbia (BC), and to identify risk factors associated with the highest HRV rates over time.Methods: Eligible individuals, recruited from the BC Seek and Treat for Optimal Prevention of HIV/AIDS population-basedretrospective cohort (N = 3955), were≥ 18 years old, first started combination antiretroviral therapy (ART) between 01/01/2000–31/12/2013, and were followed for ≥6 months until 31/12/2014. The main outcome was HRV rate. The mainexposure was HIV/HCV co-infection status. We built a confounder non-linear mixed effects model, adjusting for severaldemographic and time-dependent factors.Results: HRV rates have decreased since 2000 in both groups. The overall age-sex standardized HRV rate (per person-year)among PLW-HIV and PLW-HIV/HCV was 21.11 (95% CI 20.96–21.25) and 41.69 (95% CI 41.51–41.88), respectively. Theexcess in HRV in the co-infected group was associated with late presentation for ART, history of injection drug use, sub-optimal ART adherence and a higher number of comorbidities. The adjusted HRV rate ratio for PLW-HIV/HCV incomparison to PLW-HIV was 1.18 (95% CI 1.13–1.24).Conclusions: Although HRV rates have decreased over time in both groups, PLW-HIV/HCV had 18% higher HRV thanthose only living with HIV. Our results highlight several modifiable risk factors that could be targeted as potential meansto minimize the disease burden of this population and of the healthcare system.Keywords: HIV, Hepatitis C virus, Healthcare utilization, Administrative data, Risk factorsBackgroundIn high-income settings, HIV infection has become achronic manageable condition [1]. Overall, HIV/AIDSassociated morbidity and mortality has decreased to un-precedented levels, largely due to the widespread use ofcombination antiretroviral therapy (ART) [2]. Peopleliving with HIV/AIDS (PLW-HIV) now have life expect-ancies comparable to those observed in the generalpopulation, although variability between sub-groups re-main [3]. Despite successful treatment-mediated viralsuppression, premature morbidity and mortality due tonon-AIDS related infectious and non-infectious comor-bidities are increasingly prevalent, raising new challengesfor healthcare providers and health systems [4].Globally, the hepatitis C virus (HCV) has become oneof the most prevalent co-infection among PLW-HIV [5].Indeed, HCV mortality has surpassed that of all otherreportable infectious diseases together, including AIDSand tuberculosis, in the United States [6]. According toa recent meta-analysis, the prevalence of both HIV and* Correspondence: vlima@cfenet.ubc.ca2British Columbia Centre for Excellence in HIV/AIDS, 608 - 1081 BurrardStreet, Vancouver, BC V6Z 1Y6, CanadaFull list of author information is available at the end of the article© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (, which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver( applies to the data made available in this article, unless otherwise stated.Ma et al. BMC Health Services Research  (2018) 18:319 is the highest (82%) among people who injectdrugs (PWID) [5]. In Canada, similar to the UnitedStates, 20% to 30% of PLW-HIV are also living withHCV, while the prevalence of both viruses among PWIDranges between 50% and 90% [7–9]. In addition, thepresence of both viruses has been shown to increase therisk for clinical progression of HIV as well as prematuremortality (despite ART), and accelerated progression ofHCV-associated liver diseases [10, 11].Enhanced immunosenescence, resulting from HIV in-fection via persistent inflammatory activity and immuneactivation, is associated with increased morbidity andmortality [12, 13]. It is becoming increasingly evident thatHCV infection also contributes to systemic immunosenes-cence [14–16]. Of note, HCV-related disease burden isnot restricted to the liver; it extends to several conditions(immune-mediated or activated by chronic inflammation)impacting extrahepatic organs/tissues [17]. Additionally,life-style factors, socio-economic constraints, and inad-equate engagement in care tend to exacerbate other co-morbid conditions, further complicating clinical outcomesamong some subgroups of PLW-HIV, particularly thoseco-infected with HCV [18, 19].To date, a large number of people living with HIV andHCV (PLW-HIV/HCV) have no or limited access toHCV treatment. However, very recently, the healthcarelandscape in British Columbia (BC), Canada, has beenrapidly transforming with the advent of highly effica-cious and tolerated HCV direct-acting antiviral therap-ies, resulting in a steady increase in the number ofindividuals accessing this life-saving therapy [20, 21].Still, there are barriers for accessing healthcare servicesand programs for those afflicted by these diseases thatneed to be identified, particularly among those who aremarginalized and vulnerable in the population [20].Thus, the main objective of this study was tocharacterize the trends in healthcare-related visits (HRV)of PLW-HIV and PLW-HIV/HCV, in BC. Additionally, weaimed at identifying modifiable risk factors, associatedwith the highest HRV rates over time that could be tar-geted as potential means to minimize the disease burdenof this population and on the healthcare system.MethodsStudy settingThe province of BC established the BC Centre for Excel-lence in HIV/AIDS Drug Treatment Program (DTP) in1992; it has since been responsible for the distribution ofantiretrovirals [22]. The DTP, funded by the provincialgovernment, provides HIV medical care and laboratorymonitoring (e.g., CD4 cell counts and viral load) for alldiagnosed PLWH residing in BC at no cost, in accord-ance with BC’s HIV therapeutic guidelines, which havelargely remained consistent with those of the Inter-national Antiviral Society-USA since 1996 [23, 24].Study design and dataThis retrospective study was carried out using data fromthe British Columbia Seek and Treat for Optimal Pre-vention of HIV/AIDS (STOP HIV/AIDS) population-based cohort, which is derived from various linkagesamong provincial databases [22, 25–30]. This cohort isbriefly described in the Additional file 1. Our inclusioncriteria for individuals were as follows: (i) ART-naïve in-dividuals aged ≥18 years, (ii) enrolled in the (DTP) be-tween January 1, 2000 and December 31, 2013, (iii)initiated ART consisting of two nucleoside/nucleotide re-verse transcriptase inhibitors (NRTIs) as backbone, pluseither a non-nucleoside reverse transcriptase inhibitor(NNRTI) or a ritonavir-boosted protease inhibitor (bPI),(iv) had a CD4 count and a viral load measurementwithin 6 months of ART treatment initiation, and (v)had at least 6 months of follow-up. Note that we decidedto exclude other initial ART regimens due to the smallnumber of individuals who have initiated on them. Eli-gible individuals were followed until December 31, 2014,the last contact date (i.e., the last available laboratorytest date, the last filled ART prescription refill date orthe date of last encounter with the healthcare systemidentified in any of the STOP HIV/AIDS databases), orthe date of death (all-causes).All viral load tests and the majority of CD4 cell counttests were performed by the St. Paul’s Hospital laborator-ies in Vancouver, BC, and were subsequently transferredto the DTP via electronic linkage. CD4 cell count testscompleted at other laboratories throughout BC weremanually entered into the DTP; altogether resulting in ap-proximately 85% data capture of all CD4 cell count testsdone in the province. For analytical purposes, all viral loadmeasurements were transformed to range from < 50(coded as 49) to > 100,000 (coded as 100,010) copies/mL.This process was necessary to account for advances intesting methodology, as previously described elsewhere[31]. CD4 cell counts were measured by flow cytometry(Beckman Coulter, Inc., Mississauga, Ontario).Main outcomeThe main outcome was the rate of HRV per individual forevery 6-month interval. HRV was calculated based on re-cords from the Medical Services Plan (MSP) billing pro-vincial database linked to STOP HIV/AIDS cohort. MSPcaptures HIV and non-HIV-related inpatient and out-patient services provided by physicians and supplementaryhealth care practitioners, as well as diagnostic procedures.The unit of analysis for the crude HRV rate was person-year, which was calculated by dividing the number of HRVby the number of person-years of follow-up in eachMa et al. BMC Health Services Research  (2018) 18:319 Page 2 of 12calendar year. The corresponding 95% confidence inter-vals (CI) for these rates were based on the Fisher’s exacttest [32]. Note that we also presented the age-sex stan-dardized HRV rates for the overall follow-up utilizing BC’spopulation estimates as the reference [33].The number of HRV was derived from the MSP phys-ician billing records, which contain the date, type and lo-cation of service, physician number, practitionerspeciality and costs. Since multiple MSP database re-cords could be associated with one unique HRV, a rec-ord was considered to be a unique HRV if it satisfiedone of the following conditions:i. If the service date, which is the date on which theservice was rendered by a practitioner, wasdifferent; orii. If the speciality number, which is a numberassigned to identify the practitioner’s specialty, wasdifferent; oriii. If the location of the service was different.Once the rules outlined above were adopted, we iden-tified the unique HRV for these individuals. Subse-quently, we calculated the number of HRV in each 6-month interval for all individuals. Visits were classifiedas general practitioners, other healthcare practitioners,and laboratory services.Main exposure variableThe main exposure variable of interest was HIV/HCVco-infection status, derived from the DTP database,which indicates whether PLW-HIV has ever had evi-dence of HCV infection (i.e., a HCV antibody positive orHCV RNA detected, as indicated by laboratory resultdata or physician reported status). Note that upon suc-cessful HCV treatment, individuals can become re-infected with HCV. Thus, we were only able to incorp-orate HCV ever status for these individuals.Potential confoundersThe confounders measured at ART initiation included:sex (male, female), risk for HIV acquisition (gay, bisexualand other men who have sex with men (MSM), PWID,MSM/PWID, Other, Unknown), initial ART regimen(NNRTI, bPI) and period of ART initiation (2000–2003,2004–2007, 2008–2011, 2012–2013). Several time-varying confounders, measured every year, were consid-ered in this study, including: age (< 30, 30–39, 40–49,≥50 years), CD4 cell count (< 50, 50–199, 200–349, ≥350cells/mm3), viral load (log10 transformed), adherencelevel (< 40%, 40–79%, 80–94%, ≥95%), cumulativenumber of comorbid diseases (0, 1, 2, ≥3) and person-years of follow-up time. Adherence level was determinedon the basis of a validated measure assessing refillcompliance, which was calculated by dividing theamount of days of dispensed ART medication by theamount of days of study follow-up, for each period (pre-sented as percentage). ART adherence calculations werederived from distinct regimen exposures for each indi-vidual [34]. PWID included individuals with past andcurrent exposure to injection drug use. Individual co-morbidities were derived from the Charlson Comorbid-ity Index (Additional file 1: Table S1) [35]. Theseincluded 16 conditions, other than HIV/AIDS (e.g.,renal, liver, heart and lung diseases and cancer). Theseconditions were identified using International Classifica-tion of Diseases (Ninth and Tenth Revisions, ClinicalModification) diagnosis codes obtained from the STOPHIV/AIDS cohort databases based on a validated case-finding algorithm [36].Statistical analysisCategorical variables were compared using the Fisher’sexact test or the X2 test, and continuous variables werecompared using the Kruskal-Wallis test [37]. Based onexploratory data analyses to determine the best distribu-tion for modeling HRV rates, non-linear mixed effectsmodels were used assuming a Poisson distribution, theperson-years of follow-up time as the offset, a log linkfunction, a random intercept term and an autoregressiveof order one working correlation matrix [38]. We havechosen these models since they are flexible in taking intoaccount the inter- and intra-individual sources of vari-ation, they can handle imbalanced longitudinal data, andzero-inflated models did not show any gain over thefinal fitted model [38, 39]. Potential confounders wereselected for inclusion in the final model using abackward-selection approach, published by our groupbased on the work by Maldonado and Greenland [40],that considered the magnitude of change in the coeffi-cient of the HIV/HCV co-infection status variable. Spe-cifically, starting with a fixed model, which consideredall available variables, potential confounders weredropped one at a time, using the relative change in thecoefficient for the variable related to the HIV/HCV co-infection status as a criterion, until the maximumchange from the full model exceeded 5% [41]. Note thatin the multivariable model, we did not adjust for risk forHIV acquisition given its high collinearity with the mainstudy exposure. All analyses were performed using eitherR© version 3.3.2 (The R foundation for statistical com-puting, Vienna, Austria) or SAS version 9.4 (SAS, Cary,North Carolina, USA).ResultsCohort characteristicsOverall, 4217 ART-naïve adults with a total of 615,776HRV were initially eligible to participate in this study.Ma et al. BMC Health Services Research  (2018) 18:319 Page 3 of 12Table 1 Study population characteristics by inclusion statusVariables Overall Included Excluded P-valueN = 4217 N = 3955 N = 262Sex, n(%)Female 805 (19) 735 (91) 70 (9) 0.0016Male 3412 (81) 3220 (94) 192 (6)StatusHIV mono-infected 2333 (55) 2333 (100) 0 (0) NAHIV/HCV co-infected 1622 (39) 1622 (100) 0 (0)Unknown 262 (6) 0 (0) 262 (100)Age at ART initiation (years), n (%)< 30 516 (12) 481 (93) 35 (7) 0.172430–39 1294 (31) 1216 (94) 78 (6)40–49 1523 (36) 1441 (95) 82 (5)≥ 50 884 (21) 817 (92) 67 (8)Risk, n(%)MSM 1376 (33) 1294 (94) 82 (6) < 0.0001PWID 1335 (32) 1298 (97) 37 (3)MSM/ PWID 314 (7) 309 (98) 5 (2)Other 652 (15) 581 (89) 71 (11)Unknown 540 (13) 473 (88) 67 (12)ART era, n(%)2000–2003 859 (20) 797 (93) 62 (7) 0.10372004–2007 1214 (29) 1153 (95) 61 (5)2008–2011 1704 (40) 1599 (94) 105 (6)2012–2013 440 (10) 406 (92) 34 (8)Baseline CD4 cell count(cells/mm3), n (%)< 50 498 (12) 474 (95) 24 (5) 0.413650–199 1366 (32) 1285 (94) 81 (6)200–349 1329 (32) 1238 (93) 91 (7)≥ 350 1024 (24) 958 (94) 66 (6)ART Adherence (first sixmonths), n (%)≥ 95% 3291 (78) 3084 (94) 207 (6) 0.756380–94% 218 (5) 208 (95) 10 (5)40–79% 474 (11) 445 (94) 29 (6)< 40% 234 (6) 218 (93) 16 (7)Number of comorbiditiesat baseline, n(%)0 1451 (34) 1357 (94) 94 (6) 0.81221 1397 (33) 1307 (94) 90 (6)2 768 (18) 725 (94) 43 (6)≥3 601 (14) 566 (94) 35 (6)Initial ART regimen, n(%)NNRTI 2033 (48) 1895 (93) 138 (7) 0.1531Ma et al. BMC Health Services Research  (2018) 18:319 Page 4 of 12Among these individuals, 81% were males, 67% wereaged between 30 and 49 years, 72% were either MSM,PWID or both, 40% started ART between 2008 and2011, 76% initiated ART with a CD4 cell count < 350cells/mm3 and had a median viral load 4.90 log10 copies/mL (25th–75th percentile (Q1-Q3): 4.38–5.00 log10copies/mL), 52% started on a bPI-based ART, 78% hadadherence ≥95% during the first six months on ART,and 34% had no comorbidities while 14% had 3 or more.The median follow-up time was 4.99 (Q1-Q3: 2.50–7.98)years, in which the median number of HRV was 97 (Q1-Q3: 48–187) (Table 1). This cohort comprised of 2333(55%) PLW-HIV, 1622 (39%) PLW-HIV/HCV, and 262(6%) whose HCV status was unknown. The distributionof study variables among those included in this analysis(N = 3955) was very similar to the original 4217 individ-uals described above. For the purposes of this study,those with unknown HCV status, who contributed25,393 HRV (4% of total visits), were excluded from thesubsequent analyses. As noted in Table 1, those excludedwere more likely to have risk for HIV acquisition otherthan MSM or PWID, a lower number of HRV and aslight shorter follow-up time.Characteristics by HIV/HCV co-infection statusBivariable analysis exploring associations between studycharacteristics and HCV co-infection status (shown inTable 2) revealed that PLW-HIV and PLW-HIV/HCVdiffered significantly in all study characteristics, exceptfor initial ART regimen. PLW-HIV/HCV were morelikely to be younger, female, PWID, initiate ART prior to2008, have a lower CD4 cell count and viral load meas-urement at baseline, and maintain adherence < 40% dur-ing follow-up. Additionally, these individuals were alsomore likely to present with and develop a higher numberof comorbidities compared to PLW-HIV. The number ofPLW-HIV/HCV with moderate or severe liver diseaseincreased substantially during follow-up (75 (5%) to 124(8%); p-value 0.0004). Apart from liver disease, chronicpulmonary disease and cancer (all causes) were the mostprevalent comorbidities (results not shown).Trends in healthcare-related visitsThe overall age-sex standardized HRV rate for PLW-HIV and PLW-HIV/HCV was 21.11 per person-year(95% CI 20.96–21.25) and 41.69 per person-year (95%CI 41.51–41.88), respectively. For PLW-HIV/HCV, visitswere mainly related to methadone or buprenorphine/na-loxone treatment, and not to the same extent, to labora-tory tests done to monitor liver and kidney function andthe hematology profile. For PLW-HIV, most visits wereconcerned with the latter. As demonstrated in Fig. 1a,the annual crude HRV rates in both groups have beensteadily decreasing from 2000 to 2013. The decrease inHRV rates was more prominent for PLW-HIV (34.28per person-year to 17.98 per person-year; 48% decrease;p-value < 0.0001) compared to PLW-HIV/HCV (48.57per person-year to 33.92 per person-year; 30% decrease;p-value < 0.0001). The trends in HRV rates stratified byHIV/HCV co-infection status and by the type of HRVare illustrated in Fig. 1b. As shown, except for HRV ratesrelated to laboratory services, PLW-HIV/HCV consist-ently maintained higher rates, especially those related togeneral practitioner visits (> 2 times higher). Note thatthe rates for other healthcare practitioners (which in-cludes visits to specialists), although much lower thanthe HRV rates for general practitioner visits, they havebeen more stable over time and the difference betweenthese groups was not as pronounced. More detailed infor-mation on these trends can be found in Additional file 1:Table S2. The result of the multivariable confoundermodel showed that the adjusted HRV rate ratio for PLW-HIV/HCV in comparison to PLW-HIV was 1.18 (95% CI1.13–1.24), after controlling for sex, age, ART era, time-varying CD4, adherence to ART, viral load and the num-ber of comorbidities (Table 3).DiscussionThis population-based cohort study adds to the growingbody of evidence indicating that PLW-HIV/HCV incursignificantly greater HRV rates relative to those only livingwith HIV [42–44]. It is worth noting that although PLW-HIV/HCV experienced an 18% higher rate relative toTable 1 Study population characteristics by inclusion status (Continued)Variables Overall Included Excluded P-valueN = 4217 N = 3955 N = 262bPI 2184 (52) 2060 (94) 124 (6)Total healthcare-related visits,median (Q1-Q3)97 (48–187) 99 (49–191) 71 (31–125) < 0.0001Baseline viral load (log10 copies/mL),median (Q1-Q3)4.90 (4.38–5.00) 4.90 (4.38–5.00) 4.85 (4.28–5.00) 0.4339Follow-up time (years), median (Q1-Q3) 4.99 (2.50–7.98) 4.99 (2.81–8.00) 3.50 (1.76–5.99) < 0.0001Q1-Q3: 25th - 75th percentiles; MSM: Gay, bisexual and other men who have sex with men; PWID: people who have ever injected drugs; ART: combination antiretroviraltherapy; NNRTI: non-nucleoside reverse transcriptase inhibitor; bPI: ritonavir-boosted protease inhibitor; NA: not applicable. Note that Overall column shows columnpercent, while Included/Excluded columns show row percentMa et al. BMC Health Services Research  (2018) 18:319 Page 5 of 12Table 2 Study population characteristics by hepatitis C (HCV) co-infection statusVariables HIV mono-infected HIV/HCV co-infected P-valueN = 2333 N = 1622Sex, n(%)Female 269 (37) 466 (63) < 0.0001Male 2064 (64) 1156 (36)Age at ART initiation (years), n(%)< 30 304 (63) 177 (37) < 0.000130–39 726 (60) 490 (40)40–49 780 (54%) 661 (46%)≥ 50 523 (64%) 294 (36%)Risk, n(%)MSM 1144 (88) 150 (12) < 0.0001IDU 122 (9) 1176 (91)MSM/IDU 122 (39) 187 (61)Other 514 (88) 67 (12)Unknown 431 (91) 42 (9)ART era, n(%)2000–2003 403 (51) 394 (49) < 0.00012004–2007 628 (54) 525 (46)2008–2011 1001 (63) 598 (37)2012–2013 301 (74) 105 (26)Baseline CD4 cell count (cells/mm3), n(%)< 50 276 (58) 198 (42) < 0.000150–199 640 (50) 645 (50)200–349 746 (60) 492 (40)≥ 350 671 (70) 287 (30)Last CD4 cell count (cells/mm3), n(%)< 50 40 (33) 81 (67) < 0.000150–199 101 (31) 228 (69)200–349 232 (46) 277 (54)≥ 350 1732 (67) 864 (33)Unknown 228 (57) 172 (43)ART Adherence (first six months), n(%)≥ 95% 1987 (64) 1097 (36) < 0.000180–94% 100 (48) 108 (52)40–79% 182 (41) 263 (59)< 40% 64 (29) 154 (71)ART Adherence (last six months), n(%)≥ 95% 1821 (65) 960 (35) < 0.000180–94% 122 (47) 140 (53)40–79% 224 (45) 277 (55)< 40% 166 (40) 245 (60)Number of comorbidities at baseline, n(%)0 1067 (79) 290 (21) < 0.00011 745 (57) 562 (43)Ma et al. BMC Health Services Research  (2018) 18:319 Page 6 of 12PLW-HIV, we observed a decrease in HRV rates over timeamong both groups, even after controlling for several con-founders including disease severity and the cohort effect.Apart from HCV infection, the excess in HRV ratesamong PLW-HIV/HCV were at least partially attributableto the fact that these individuals had a history of injectiondrug use, presented later for HIV treatment, had sub-optimal adherence to ART and had higher prevalence ofcomorbidities.To understand the reason behind the decreasing ratesover time, the reader should be aware that the studyperiod encompasses three phases of BC’s response toHIV/AIDS: the harm reduction and health service scale-up phase (2000–2005); the early Treatment as Preven-tion phase (2006–2009); and the STOP HIV/AIDS phase(2010-present), during which BC’s HIV therapeuticguidelines recommended ART treatment for all adultswith HIV infection, regardless CD4 count [45, 46].Throughout these phases, various HIV care initiativeshave been implemented and may have attenuated thehealthcare-related utilization of this population. Namely,the evolving deployment of biomedical and health ser-vice interventions (e.g., the development of improvedantiretroviral drugs, substance use treatment, and medi-cation adherence support) and structural interventions(e.g., legal and policy), which have been comprehensivedescribed elsewhere [45].We should also note that both groups of individualswere linked to HIV care and receiving treatment. Mostlikely, if people only living with HCV were included inthis analysis, we would see that these trends, in thissame period, did not decrease given that, in BC, most ofthese individuals are not fully engaged into care. Inaddition, given the recent approval for use of safer, moretolerable and efficacious interferon-free direct actingantivirals-based HCV therapy, going forward, thesetrends will likely change, particularly for specialist-related visits as they will be the ones mainly prescribingthese medications and following these individuals. Thus,there is a need for continued monitoring and evaluationof HRV among PLW-HIV/HCV, especially since in 2017,the BC Ministry of Health has announced a province-wide expansion of HCV treatment to all of those livingwith HCV starting in March 2018 [21].The persisting disparity in HRV rates observed amongPLW-HIV and PLW-HIV/HCV clearly indicates thatthere is a critical need for interventions that may attenu-ate the risk of requiring higher resource use for careamong those living with both viruses. Additionally, anysuccessful strategy to attenuate the utilization of PLW-HIV/HCV will require significant levels of treatment up-take and adherence, especially among PWID (includingPLW-HIV/HCV). Our data also suggest that addressingthe underlying substance use disorder may be beneficialin this regard. Furthermore, doing so would also contrib-ute to preventing HCV reinfection, which would greatlyenhance the individual and societal impact of interferon-free direct acting antivirals-based HCV therapy [47–49].On that note, several clinical models have proved to besuccessful in this regard by combining services aimed toaddress viral hepatitis and HIV, substance use detoxifica-tion, opioid substitution, and primary care in lowTable 2 Study population characteristics by hepatitis C (HCV) co-infection status (Continued)Variables HIV mono-infected HIV/HCV co-infected P-valueN = 2333 N = 16222 331 (46) 394 (54)≥3 190 (34) 376 (66)Number of comorbidities at the end of follow-up, n(%)0 961 (82) 215 (18) <0.00011 687 (62) 421 (38)2 333 (47) 382 (53)≥3 352 (37) 604 (63)Initial ART regimen, n(%)NNRTI 1098 (58) 797 (42) 0.2109bPI 1235 (60) 825 (40)Total healthcare-related visits, median (Q1-Q3) 83 (43–150) 143 (68–278) < 0.0001Baseline viral load (log10 copies/mL), median (Q1-Q3) 4.92 (4.42–5.00) 4.87 (4.35–5.00) 0.0048Last viral load (log10 copies/mL), median (Q1-Q3) 1.69 (1.69–1.69) 1.69 (1.69–1.98) < 0.0001Follow-up time (years), median (Q1-Q3) 4.95 (2.50–7.91) 5.30 (2.99–8.31) 0.0038Q1-Q3: 25th - 75th percentiles; MSM: Gay, bisexual and other men who have sex with men; PWID: people who have ever injected drugs; ART: combination antiretroviraltherapy; NNRTI: non-nucleoside reverse transcriptase inhibitor; bPI: ritonavir-boosted protease inhibitorMa et al. BMC Health Services Research  (2018) 18:319 Page 7 of 12threshold environment, coupled with comprehensiveand integrated multidisciplinary teams of health careprofessionals including treaters, nurses, substance useand behavioral health service providers, as well as othersocial support services [48]. In 2016, the BC Centre forExcellence in HIV/AIDS launched a province-wide mon-itoring and evaluation strategy, which will address thehealth needs of those living with or at risk of HCV infec-tion, including those also living with HIV [50, 51]. Thekey aims of this program include the normalization ofHCV testing, especially among those at higher risk; sup-port to facilitate access to HCV and substance usetreatment; extensive deployment of harm reductionstrategies; and strengthening of educational programs totreat and care for this population.The implications of HIV/HCV co-infection in thecontext of a rapidly expanding population of agingPLW-HIV are important, particularly at a time whenmeeting health demands in BC, and in other high-resource settings, is becoming exceedingly challengingamid fiscal constraints [52]. As PLW-HIV live longerand non-AIDS-related comorbidities continue to rise,the impact of HIV/HCV co-infection will be increas-ingly relevant, both from a clinical perspective and abaFig. 1 Crude rate of healthcare-related visits (per person-year) for PLW-HIV and PLW-HIV/HCV, from 2000 to 2013. Panel a, corresponds to theoverall healthcare-related visit rate. Panel b corresponds to the healthcare-related visit rate stratified by service typeMa et al. BMC Health Services Research  (2018) 18:319 Page 8 of 12health systems perspective. Chronic HCV infectionamong PLW-HIV may have contributed to the ex-acerbation of progressive immunosenescence, and itmay be associated with premature morbidity andmortality manifested by the development of multiplecomorbidities (as observed in this study) [16, 53], fur-ther contributing to increased financial strain onhealthcare systems.The findings of the present study should be inter-preted in light of several limitations. First, HIV/HCV co-Table 3 Results from the multivariable confounder modelVariables Rate Ratio (95% Confidence Interval)Unadjusted Model Adjusted ModelStatusHIV mono-infected 1 (−) 1 (−)HIV/HCV co-infected 1.65 (1.57–1.73) 1.18 (1.13–1.24)SexMale 1 (−) 1 (−)Female 1.45 (1.36–1.54) 1.19 (1.13–1.25)Age at ART initiation (years)< 30 1 (−) 1 (−)30–39 1.08 (1.07–1.10) Not selected40–49 1.23 (1.21–1.25)≥ 50 1.47 (1.43–1.50)ART era2000–2003 1 (−) 1 (−)2004–2007 0.84 (0.77–0.93) Not selected2008–2011 0.82 (0.76–0.87)2012–2013 0.94 (0.87–1.00)CD4 cell count (cells/mm3) (time-varying)< 50 2.29 (2.25–2.32) 2.27 (2.24–2.31)50–199 1.45 (1.44–1.47) 1.44 (1.43–1.46)200–349 1.22 (1.21–1.23) 1.22 (1.21–1.23)≥ 350 1 (−) 1 (−)Unknown 1.00 (0.99–1.01) 1.00 (0.98–1.01)ART Adherence (time-varying)≥ 95% 1 (−) 1 (−)80–94% 1.19 (1.17–1.20) Not selected40–79% 1.20 (1.19–1.22)< 40% 1.12 (1.11–1.13)Number of comorbidities (time-varying)0 1 (−) 1 (−)1 1.39 (1.32–1.47) 1.30 (1.23–1.37)2 1.87 (1.76–1.99) 1.64 (1.54–1.75)≥3 2.82 (2.67–2.99) 2.34 (2.21–2.48)Initial ART regimenNNRTI 1 (−) 1 (−)bPI 1.12 (1.07–1.18) Not selectedViral load (log10 copies/mL) (time-varying) 1.13 (1.13–1.13) Not selectedART: combination antiretroviral therapy; NNRTI: non-nucleoside reverse transcriptase inhibitor; bPI: ritonavir-boosted protease inhibitor. Note that in the multivariablemodel we did not adjust for risk for HIV acquisition given its high collinearity with the main study exposure. Not selected means that the variable was not a confounderin the modelMa et al. BMC Health Services Research  (2018) 18:319 Page 9 of 12infection status was assigned based on ever having re-corded a positive HCV antibody test or detected HCVRNA. Thus, it is unknown whether these individuals hadactive HCV infection during the study period. Second,several factors such as rates of spontaneous viral clear-ance, HCV treatment and re-infection (among thosesuccessfully treated) were unknown for this cohort, thuslimiting our ability to identify and adjust for these fac-tors in the model. Third, therapy for opioid dependencewas not considered, but may have impacted HRV amongPLW-HIV/HCV with active injection drug use. Fourth,although healthcare administrative data are an importantsource of information for evidence-based clinical andpolicy decision-making as well as medical research, weshould note that these data are susceptible to inaccurateor incomplete coding, potentially leading to missing ormisclassified HRV in both groups. Finally, while thisstudy examined HRV exclusively, there are other formsof healthcare utilization not accounted for in these ana-lyses (e.g., addiction support services).ConclusionsIn conclusion, in this retrospective study, we found thatalthough HRV rates have been decreasing steadily overtime, PLW-HIV/HCV consistently maintained higherHVR rates relative to PLW-HIV. Our results highlightseveral modifiable risk factors (i.e., late presentation forART, injection drug use, sub-optimal ART adherenceand comorbidities) that could be targeted as potentialmeans to minimize the disease burden of this populationand on the healthcare system.Additional fileAdditional file 1: Supplementary information detailing the data linkagewithin the STOP HIV/AIDS cohort, the ascertainment of comorbiditiesbased on the Charlson Comorbidity Index, and the number ofhealthcare-related visits recorded. (DOCX 35 kb)AbbreviationsART: Combination antiretroviral therapy; BC: British Columbia; bPI: Ritonavir-boosted protease inhibitor; CI: Confidence Interval; HCV: Hepatitis C virus;HRV: Healthcare-related visits; MSM: Men who have sex with men;MSP: Medical Services Plan; NNRTI: Non-nucleoside reverse transcriptaseinhibitor; NRTI: Nucleoside/nucleotide reverse transcriptase inhibitor; PLW-HIV: People living with HIV/AIDS; PLW-HIV/HCV: People living with HIV andHCV; PWID: People who inject drugs; STOP HIV/AIDS: Seek and Treat forOptimal Prevention of HIV/AIDSAcknowledgementsWe would like to thank our patients, and the physicians, nurses, socialworkers and volunteers who support them.FundingJSGM is supported with grants paid to his institution by the British ColumbiaMinistry of Health and by the US National Institutes of Health (R01DA036307).VDL is funded by a grant from the Canadian Institutes of Health Research (PJT-148595), by a Scholar Award from the Michael Smith Foundation for HealthResearch and a New Investigator award from the Canadian Institutes of HealthResearch. OE is supported by a Canadian Institutes of Health Research doctoralaward. MH has received grant support from the National Institute on DrugAbuse (NIDA R01DA031043–01). The sponsors had no role in the design, datacollection, data analysis, data interpretation, or writing of the report. Thecorresponding author had full access to all data in the study and had finalresponsibility to submit for publication.Availability of data and materialsThe authors confirm that, for approved reasons, some access restrictionsapply to the data underlying the findings. All data is housed at the BritishColumbia Centre for Excellence in HIV/AIDS Drug Treatment Program. Giventhat we must protect the identity of Drug Treatment Program participantswe cannot provide our study data to the public. We do not allow qualifiedresearchers to obtain our data nor would it be possible to share anonymizeddata that underlie the summary statistics presented in this manuscript. If youhave further questions regarding our data sharing and privacy policy pleasedo not hesitate to contact our Director of Operations, Ms. Irene Day’ contributionsThe author’s contributions were as follows: Initial study concept and design:VDL; Acquisition of data: VDL, BY, JSGM; Analysis and interpretation of data:HM, CFV, VDL, BY; Drafting of the manuscript: VDL, HM, MS; Consultationregarding study design and interpretation of findings: MS, OE, MRL, LT, MH,LW, RSH, RB, JAS, JSGM; Critical revision of the manuscript for importantintellectual content: HM, CFV, MS, OE, MRL, LT, MH, BY, LW, RSH, RB, JAS,JSGM; Final approval of the manuscript to be published: HM, CFV, MS, OE,MRL, LT, MH, BY, LW, RSH, RB, JAS, JSGM; Statistical analysis: HM, CFV, VDL;Obtained funding: VDL, JSGM; Administrative, technical, or material support:VDL; Study supervision: VDL.Ethics approval and consent to participateAdministrative database linkage and use was approved and performed bydata stewards in each collaborating agency and facilitated by the BC Ministryof Health. The University of British Columbia Ethics Review Committee at theSt. Paul’s Hospital site provided ethics approval for this study (H08–02095).The study complies with BC’s Freedom of Information and Protection ofPrivacy Act. As the study was conducted using anonymized administrativedatabases, informed consent was not obtained.Competing interestsWe have the following competing interests: JSGM has received limitedunrestricted funding, paid to his institution, from Abbvie, Bristol-MyersSquibb, Gilead Sciences, Janssen, Merck, ViiV Healthcare. MH has receivedhonoraria for speaking engagements and/or consultancy meetings from thefollowing: Bristol Myers Squibb, Gilead, Merck, Ortho-Janssen, and ViiV. Theremaining authors do not have conflicts to declare.Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.Author details1Department of Statistics, University of British Columbia, 3182 Earth SciencesBuilding, 2207 Main Mall, Vancouver, BC V6T 1Z4, Canada. 2British ColumbiaCentre for Excellence in HIV/AIDS, 608 - 1081 Burrard Street, Vancouver, BCV6Z 1Y6, Canada. 3Faculty of Health Sciences, Simon Fraser University,Blusson Hall, Room 10502, Burnaby, BC V5A 1S6, Canada. 4British ColumbiaCentre for Excellence in HIV/AIDS, 667 - 1081 Burrard Street, Vancouver, BCV6Z 1Y6, Canada. 5Department of Statistics, University of British Columbia,3182 Earth Sciences Building room ESB 3126, 2207 Main Mall, Vancouver, BCV6T 1Z4, Canada. 6School of Population & Public Health, University of BritishColumbia, 2206 East Mall, Rm 414, Vancouver, BC V6T 1Z3, Canada.Received: 9 June 2017 Accepted: 15 April 2018References1. Deeks SG, Lewin SR, Havlir DV. The end of AIDS: HIV infection as a chronicdisease. Lancet. 2013;382(9903):1525–33.Ma et al. BMC Health Services Research  (2018) 18:319 Page 10 of 122. Lima VD, Lourenco L, Yip B, Hogg RS, Phillips P, Montaner JS. AIDSincidence and AIDS-related mortality in British Columbia, Canada, between1981 and 2013: a retrospective study. Lancet HIV. 2015;2(3):e92–7.3. Lima VD, Eyawo O, Ma H, Lourenco L, Chau W, Hogg RS, Montaner JS. Theimpact of scaling-up combination antiretroviral therapy on patterns ofmortality among HIV-positive persons in British Columbia, Canada. J IntAIDS Soc. 2015;18:20261.4. Deeks SG, Phillips AN. HIV infection, antiretroviral treatment, ageing, andnon-AIDS related morbidity. BMJ. 2009;338:a3172.5. Platt L, Easterbrook P, Gower E, McDonald B, Sabin K, McGowan C, Yanny I,Razavi H, Vickerman P. Prevalence and burden of HCV co-infection inpeople living with HIV: a global systematic review and meta-analysis. LancetInfect Dis. 2016;16(7):797–808.6. Ly KN, Hughes EM, Jiles RB, Holmberg SD. Rising mortality associated withhepatitis C virus in the United States, 2003-2013. Clin Infect Dis. 2016;62(10):1287–8.7. Hepatitis C in Canada: 2005–2010 Surveillance Report [].8. HIV and Viral Hepatitis [].9. Amin J, Kaye M, Skidmore S, Pillay D, Cooper DA, Dore GJ. HIV and hepatitisC coinfection within the CAESAR study. HIV Med. 2004;5(3):174–9.10. Graham CS, Baden LR, Yu E, Mrus JM, Carnie J, Heeren T, Koziel MJ.Influence of human immunodeficiency virus infection on the course ofhepatitis C virus infection: a meta-analysis. Clin Infect Dis. 2001;33(4):562–9.11. Greub G, Ledergerber B, Battegay M, Grob P, Perrin L, Furrer H, Burgisser P,Erb P, Boggian K, Piffaretti JC, et al. Clinical progression, survival, andimmune recovery during antiretroviral therapy in patients with HIV-1 andhepatitis C virus coinfection: the Swiss HIV cohort study. Lancet. 2000;356(9244):1800–5.12. Fernandez S, Price P, McKinnon EJ, Nolan RC, French MA. Low CD4+ T-cellcounts in HIV patients receiving effective antiretroviral therapy areassociated with CD4+ T-cell activation and senescence but not with lowereffector memory T-cell function. Clin Immunol. 2006;120(2):163–70.13. Deeks SG, Verdin E, McCune JM. Immunosenescence and HIV. Curr OpinImmunol. 2012;24(4):501–6.14. Grady BP, Nanlohy NM, van Baarle D. HCV monoinfection and HIV/HCVcoinfection enhance T-cell immune senescence in injecting drug users earlyduring infection. Immun Ageing. 2016;13:10.15. Gruener NH, Lechner F, Jung MC, Diepolder H, Gerlach T, Lauer G, Walker B,Sullivan J, Phillips R, Pape GR, et al. Sustained dysfunction of antiviral CD8(+) Tlymphocytes after infection with hepatitis C virus. J Virol. 2001;75(12):5550–8.16. Rallon N, Garcia M, Garcia-Samaniego J, Rodriguez N, Cabello A, Restrepo C,Alvarez B, Garcia R, Gorgolas M, Benito JM. HCV coinfection contributes toHIV pathogenesis by increasing immune exhaustion in CD8 T-cells. PLoSOne. 2017;12(3):e0173943.17. Cacoub P, Comarmond C, Domont F, Savey L, Desbois AC, Saadoun D.Extrahepatic manifestations of chronic hepatitis C virus infection. Ther AdvInfect Dis. 2016;3(1):3–14.18. Louie KS, St Laurent S, Forssen UM, Mundy LM, Pimenta JM. The highcomorbidity burden of the hepatitis C virus infected population in theUnited States. BMC Infect Dis. 2012;12:86.19. Goulet JL, Fultz SL, Rimland D, Butt A, Gibert C, Rodriguez-Barradas M,Bryant K, Justice AC. Aging and infectious diseases: do patterns ofcomorbidity vary by HIV status, age, and HIV severity? Clin Infect Dis. 2007;45(12):1593–601.20. Janjua NZ, Islam N, Wong J, Yoshida EM, Ramji A, Samji H, Butt ZA, ChongM, Cook D, Alvarez M, et al. Shift in disparities in hepatitis C treatment frominterferon to DAA era: a population-based cohort study. J Viral Hepat. 2017;24(8):624–30.21. More patients to benefit from hepatitis C treatments [].22. Patterson S, Cescon A, Samji H, Cui Z, Yip B, Lepik KJ, Moore D, Lima VD,Nosyk B, Harrigan PR, et al. Cohort profile: HAART observational medicalevaluation and research (HOMER) cohort. Int J Epidemiol. 2015;44(1):58–67.23. British Columbia Centre for Excellence in HIV/AIDS: Therapeutic GuidlinesAntiretroviral (ARV) Treatment of Adult HIV Infection In. ; 2015.24. Gunthard HF, Saag MS, Benson CA, del Rio C, Eron JJ, Gallant JE, Hoy JF,Mugavero MJ, Sax PE, Thompson MA, et al. Antiretroviral drugs fortreatment and prevention of HIV infection in adults: 2016 recommendationsof the international antiviral society-USA panel. JAMA. 2016;316(2):191–210.25. HIV/AIDS Information System (HAISYS). Clinical Prevention Services, BritishColumbia Centre for Disease Control, 2016 [].26. Medical Services Plan (MSP) Payment Information File; Consolidation File (MSPRegistration & Premium Billing); Home & Community Care (Continuing Care);Mental Health; PharmaNet. British Columbia Ministry of Health [publisher]. DataExtract. MOH (2016) [].27. Discharge Abstract Database (Hospital Separations). British ColumbiaMinistry of Health [publisher]. Data Extract. MOH (2016) [].28. Vital Statistics. British Columbia Ministry of Health [publisher]. Data Extract.MOH (2016) [].29. Heath K, Samji H, Nosyk B, Colley G, Gilbert M, Hogg RS, Montaner JS, GroupSHAS. Cohort profile: seek and treat for the optimal prevention of HIV/AIDS inBritish Columbia (STOP HIV/AIDS BC). Int J Epidemiol. 2014;43(4):1073–81.30. HIV laboratory testing datasets (tests: ELISA, Western blot, NAAT, p24,culture). Clinical Prevention Services, British Columbia Centre for DiseaseControl, 2016 [].31. Cobb BR, Vaks JE, Do T, Vilchez RA. Evolution in the sensitivity ofquantitative HIV-1 viral load tests. J Clin Virol. 2011;52(Suppl 1):S77–82.32. Grebely J, Raffa JD, Lai C, Krajden M, Conway B, Tyndall MW. Factorsassociated with spontaneous clearance of hepatitis C virus among illicitdrug users. Can J Gastroenterol. 2007;21(7):447–51.33. Population Estimates [].34. Lima VD, Bangsberg DR, Harrigan PR, Deeks SG, Yip B, Hogg RS, MontanerJS. Risk of viral failure declines with duration of suppression on highly activeantiretroviral therapy irrespective of adherence level. J Acquir Immune DeficSyndr. 2010;55(4):460–5.35. Quan H, Li B, Couris CM, Fushimi K, Graham P, Hider P, Januel JM,Sundararajan V. Updating and validating the Charlson comorbidity indexand score for risk adjustment in hospital discharge abstracts using datafrom 6 countries. Am J Epidemiol. 2011;173(6):676–82.36. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, SaundersLD, Beck CA, Feasby TE, Ghali WA. Coding algorithms for definingcomorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130–9.37. McDonald JH. Handbook of biological statistics. Baltimore SHP: 2nd editionEDN; 2009.38. Cameron AC, Trivedi PK. Regression analysis of count data. New York:Cambridge University Press; 1998.39. Pinheiro JCB. D.M.: mixed-effects models in S and S-PLUS. New York:Springer-Verlag; 2000.40. Maldonado G, Greenland S. Simulation study of confounder-selectionstrategies. Am J Epidemiol. 1993;138(11):923–36.41. Lima VD, Geller J, Bangsberg DR, Patterson TL, Daniel M, Kerr T, Montaner J,Hogg RS. The effect of adherence on the association between depressivesymptoms and mortality among HIV-infected individuals first initiatingHAART. AIDS. 2007;21(9):1175–83.42. Katrak S, Park LP, Woods C, Muir A, Hicks C, Naggie S. Patterns of healthcareutilization among veterans infected with hepatitis C virus (HCV) and humanimmunodeficiency virus (HIV) and Coinfected with HIV/HCV: uniqueburdens of disease. Open Forum Infect Dis. 2016;3(3):ofw173.43. Linas BP, Wang B, Smurzynski M, Losina E, Bosch RJ, Schackman BR, Rong J,Sax PE, Walensky RP, Schouten J, et al. The impact of HIV/HCV co-infectionon health care utilization and disability: results of the ACTG longitudinallinked randomized trials (ALLRT) cohort. J Viral Hepat. 2011;18(7):506–12.44. Crowell TA, Berry SA, Fleishman JA, LaRue RW, Korthuis PT, Nijhawan AE,Moore RD, Gebo KA, Network HIVR. Impact of hepatitis coinfection onhealthcare utilization among persons living with HIV. J Acquir Immune DeficSyndr. 2015;68(4):425–31.45. Olding M, Enns B, Panagiotoglou D, Shoveller J, Harrigan PR, Barrios R, Kerr T,Montaner JSG, Nosyk B. A historical review of HIV prevention and care initiativesin British Columbia, Canada: 1996-2015. J Int AIDS Soc. 2017;20(1):21941.Ma et al. BMC Health Services Research  (2018) 18:319 Page 11 of 1246. Theapeutic Guidelines for Antiretroviral Treatment (ARV) of Adult HIVInfection [].47. Lima VD, Rozada I, Grebely J, Hull M, Lourenco L, Nosyk B, Krajden M,Yoshida E, Wood E, Montaner JS. Are interferon-free direct-acting antiviralsfor the treatment of HCV enough to control the epidemic among peoplewho inject drugs? PLoS One. 2015;10(12):e0143836.48. Smyth D, Webster D. Hepatitis C virus infection: accessing drug treatment.CMAJ. 2015;187(15):1113–4.49. Islam N, Krajden M, Shoveller J, Gustafson P, Gilbert M, Buxton JA, Wong J,Tyndall M, Janjua NZ. Incidence, risk factors, and prevention of hepatitis Creinfection: a population-based cohort study. The Lancet Gastroenterology& Hepatology. 2017;2(3):200–10.50. Hepatitis C Research Program [].51. British Columbia Centre on Substance Use [].52. Aging with HIV: Long, healthy lives are possible with treatment [].53. Operskalski EA, Kovacs A. HIV/HCV co-infection: pathogenesis, clinicalcomplications, treatment, and new therapeutic technologies. Curr HIV/AIDSRep. 2011;8(1):12–22.Ma et al. BMC Health Services Research  (2018) 18:319 Page 12 of 12


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items