Healthcare and treatment experiences among people diagnosed with HIV before and after a province-wide treatment as prevention initiative in British Columbia, Canada

Eligibility for the SHAPE study includes people living with HIV in BC who were 19 or older as verified in the BC-CfE Drug Treatment Program (DTP) registry, able to complete the survey in English, and able to provide informed consent. Recruitment targets were developed based on the proportion of key sociodemographic characteristics of PLWH in BC, using estimates derived from DTP in order to ensure representativeness of the study population. The DTP provides longitudinal clinical (i.e. viral load, CD4 cell count), laboratory (i.e. resistance testing, phylogenetic monitoring), therapeutic (i.e. ART regimen), and adherence data (i.e. pharmacy refill) for all individuals who receive ART in the province. SHAPE study participants were recruited using purposive sampling in order to ensure representation based on age, gender, HIV risk group, and health authority of residence. Participants were recruited through posting advertisements on Facebook, Craigslist, Grindr, and Scruff as well as through posters and postcards via existing and new partnerships with clinics, pharmacies, and AIDS Service Organizations. This analysis concerns the baseline survey collected from January 1, 2016 to September 1, 2018. The one-hour surveys were conducted either independently online or were administered by a Peer Research Associate (PRA), who are trained research team members with living experience with HIV either in-person or over the phone. A $30 cash honorarium was given to each participant for participation in the baseline study. As part of the enrolment procedures, participants were asked to provide consent to link their survey responses to longitudinal clinical, laboratory, and pharmacy refill data from the provincial DTP, which was verified prior to the enrolment in order to ensure linkage to clinical data for all participants. This linkage was completed through matching personal health numbers, name and date of birth. Inclusion criteria for this analysis included participants who completed the baseline survey, were linked to the DTP clinical registry, and had an HIV diagnosis date of January 1, 2000 or later, and were ART-naïve at the start of first ART dispensation.

The protocol was approved by the Research Ethics Board of the University of BC/Providence Health Care (ID Number H15-01807). Written informed consent was obtained from each participant for in-person and phone interviews, and electronically for online surveys. All identifying information was removed from the analysis and cells with counts of less than 5 were suppressed.

We used Chi-square and Wilcoxon Rank Sum tests to compare the distribution of sociodemographic, experiential, and engagement in care variables with the exposure of prior to and during the provincial STOP HIV/AIDS program. Univariable and multivariable Cox proportional hazards regression models were used to assess both the time to ART initiation from HIV diagnosis date and the time to virological suppression from ART initiation date between participants diagnosed with HIV prior to and during the STOP HIV/AIDS era. The multivariable models adjusted for confounding variables of age, gender, history of ever using injection drugs, and sexual orientation. Kaplan–Meier curves displayed both the difference in time to ART initiation and time to virological suppression between prior to and during STOP HIV/AIDS era groups.

Of the 325 (50.5% of 644) participants from the SHAPE baseline survey eligible for inclusion in this analysis, 198 (61%) were diagnosed with HIV in the pre-STOP HIV/AIDS era and 127 (39%) were diagnosed during the post-STOP HIV/AIDS era. The median age of individuals at diagnosis in the analytic sample was 37 years old (Q1-Q3: 30–46), 26.2% were women, 72.3% were men, < 2% were transgender or non-binary individuals, 38.8% identified as gay, 48.3% lived within the Vancouver Coastal Health Authority, and 64.2% had insufficient food security (see Table 1).

Participants diagnosed with HIV during the STOP HIV/AIDS era were more frequently men (81.9% vs. 66.2%; p = 0.004), identified as gay (45.7% vs. 34.3%; p = 0.039), and had completed high-school or greater (79.5% vs. 61.6%; p = 0.001) in comparison to those diagnosed prior to the implementation of the provincial program. There was no significant difference between participants in the pre- vs post-intervention eras (median age of 37 and 38, respectively, p = 0.258). PLWH diagnosed during the post-intervention era were less likely to report a history of incarceration as an adult versus PLWH diagnosed prior to the provincial initiative (30.7% vs 48.0%; p = 0.002). History of homelessness was prevalent overall, with 59.6% and 50.4% reported being homeless currently or previously, prior to and during the post-intervention era, respectively, although there was no statistical difference between the two groups (p = 0.103).

Coinfection of HCV (18.1% vs. 48.2%; p < 0.001), HBV (6.3% vs. 16.2%; p = 0.008), self-reported ever using heroin (26.8% vs. 39.9%; p = 0.015), and lifetime injection drug use (36.2% vs. 53.0%; p = 0.003) were less frequent among post-intervention participants. Indication of significant depressive symptoms (CES-D 10 scale scores of ≥ 10 [18, 19]) remained highly prevalent across eras (pre-STOP HIV/AIDS: 50.6%; post-STOP HIV/AIDS: 55.4%; p = 0.415).

There were important differences in the clinical experiences around HIV diagnoses reported in this analysis, with participants in the post-STOP era reporting fewer HIV diagnoses with a family doctor (20.5% vs 32.3%) and a greater proportion of HIV diagnoses with walk-in clinics (44.9% vs 39.4%) and hospitals (20.5% vs 11.1%) compared to PLWH diagnosed in the pre-STOP HIV/AIDS era (p = 0.042) (see Table 1). Participants diagnosed in the post-STOP HIV/AIDS era reported higher use of clinical services such as follow-up appointments with their diagnosing physician (46.5% vs 32.8%,%, p = 0.014), nurse consultations (41.7% vs 17.7%, p < 0.001), appointments with another medical professional (33.9% vs 22.7, p = 0.028), and visits with specialists (70.1% vs 57.1%, p = 0.018), than pre-STOP HIV/AIDS participants (see Fig. 1). Furthermore, the resources allocated in the regional health authorities as part of the STOP HIV/AIDS program for psycho-social supports led to increased utilization of nutritionist appointments (25.2% vs 13.1%, p = 0.006), social workers (28.3% vs 17.7%, p = 0.008), and peer support (26.0% vs 16.2%, p = 0.031), in the post-STOP HIV/AIDS era. Desire for HIV care and supportive services following HIV diagnosis did not differ significantly (p > 0.05) between participants in the two groups (see Fig. 2). However, participants diagnosed in the post-STOP HIV/AIDS era were less likely to report a desire for counselling (post-intervention 14% vs pre-intervention 17%), peer support (14% vs 21%), substance use counselling (11% vs 10%), medication adherence assistance programs (6% vs 9%), a nutritionist (10% vs 15%), a social worker (11% vs 14%), and housing assistance (13% vs 19%). This is a reflection of increased provision and availability of nursing, outreach, and psycho-social services, addressing much needed gaps in care identified in the cohort diagnosed with HIV prior to the STOP HIV/AIDS implementation.

Median CD4 count at time of ART initiation was significantly higher among post-STOP HIV/AIDS participants with 410 cells/ul (Q1-Q3: 220–620) compared with 270 (170–430) (p = 0.001). Additionally, there was a significantly lower proportion of participants with treatment interruptions reported within five years following ART initiation in the post-STOP HIV/AIDS era (during 17.3% vs prior 48.0%; p < 0.001) (see Table 1).

When adjusted for age, gender, sexual orientation, and history of injection drug use, STOP HIV/AIDS era participants were 5.97 times more likely to initiate ART than pre-STOP HIV/AIDS participants (adjusted hazards ratio [aHR] 5.96, 95% CI 4.47–7.97; p < 0.001) (see Table 2). Median time to ART initiation for STOP HIV/AIDS era participants was within 1.5 months of HIV diagnosis, compared to 27.5 months for participants prior to the provincial initiative (see Fig. 3A).

In the adjusted multivariable model, the STOP HIV/AIDS era participants were significantly more likely to reach viral suppression than prior to STOP HIV/AIDS era participants (aHR 2.03, 95% CI 1.58–2.60; p < 0.001) (see Table 3). Median time to virological suppression for STOP HIV/AIDS era participants was 2.3 months versus 4 months for prior to STOP HIV/AIDS participants (see Fig. 3B).