Abstract
BACKGROUND
Many challenges remain in successfully engaging people with HIV (PWH) into lifelong HIV care. Living in non-urban or rural areas has been associated with worse outcomes. Uncertainties remain regarding how to provide optimal HIV care in non-urban areas.
METHODS
Using a retrospective descriptive analysis framework, we compared multiple measurable HIV care metrics over time on the basis of urban versus non-urban residency, under a centralized HIV care model. We examined rates of HIV diagnosis, access to and retention in HIV care, and longitudinal outcomes for all newly diagnosed PWH between January 1, 2008, and January 1, 2020, categorized by their home location at the time of HIV diagnosis in southern Alberta.
RESULTS
Of 719 newly diagnosed PWH, 619 (86%) lived in urban areas and 100 (14%) lived in non-urban areas. At HIV diagnosis, the groups had no significant differences in initial CD4 count or clinical characteristics (p = 0.73). Non-urban PWH, however, had slightly longer times to accessing HIV care and initiating antiretroviral therapy (ART) (p < 0.01). Non-urban PWH showed trends toward slightly lower retention in care and lower sustained ART use, with higher rates of unsuppressed viral loads at 12, 24, and 36 months after diagnosis (p < 0.01). However, by 2020 both cohorts had suppression rates above 90%.
CONCLUSIONS
Sustained retention in care was more challenging for non-urban PWH; however, adherence to ART and viral suppression rates were more than 90%. Although encouraging, challenges remain in identifying and reducing unique barriers for optimal care of PWH living in non-urban areas.
Keywords: ARV, HIV/AIDS, non-urban, retention in care, rural, urban, viral suppression
Abstract
HISTORIQUE
Il reste de nombreux défis à relever pour mobiliser les personnes atteintes du VIH (PAV) à adhérer aux soins du VIH jusqu’à la fin de leurs jours. La résidence en région non urbaine ou rurale est associée à des résultats cliniques plus négatifs. Il reste des incertitudes quant aux moyens d’offrir des soins optimaux du VIH hors des régions urbaines.
MÉTHODOLOGIE
Au moyen d’une analyse descriptive rétrospective, les chercheurs ont comparé de multiples mesures des soins du VIH au fil du temps d’après le lieu de résidence en milieu urbain ou non urbain, en fonction d’un modèle de soins du VIH centralisé. Ils ont examiné les taux de diagnostic du VIH, l’accès aux soins du VIH et la rétention de ces soins, ainsi que les résultats cliniques longitudinaux de toutes les PAV nouvellement diagnostiquées entre le 1er janvier 2008 et le 1er janvier 2020, classées d’après leur lieu de résidence au moment du diagnostic de VIH dans le sud de l’Alberta.
RÉSULTATS
Sur les 719 PAV nouvellement diagnostiquées, 619 (86 %) vivaient en région urbaine et 100 (14 %), en région non urbaine. Au diagnostic du VIH, les groupes ne présentaient pas de différence significative pour ce qui est de la numération initiale des CD4 ou des caractéristiques cliniques (p = 0,73). Cependant, il fallait légèrement plus de temps aux PAV de milieu non urbain pour accéder aux soins du VIH et entreprendre une thérapie antirétrovirale (TAR) (p < 0,01). Ainsi, 12, 24 et 36 mois après le diagnostic, les PAV des milieux non urbains affichaient des tendances vers une rétention légèrement plus faible des soins et une utilisation soutenue légèrement plus faible du traitement antirétroviral (TAR), de même que des charges virales non supprimées plus élevées (p < 0,01). Cependant, en 2020, les deux cohortes présentaient des taux de suppression supérieurs à 90 %.
CONCLUSIONS
Il était plus difficile de maintenir une rétention soutenue des soins pour les PAV hors des milieux urbains, mais l’adhésion au TAR et les taux de suppression virale étaient supérieurs à 90 %. C’était encourageant, mais il reste des défis pour déterminer et réduire les obstacles à des soins optimaux chez les PAV qui habitent hors des milieux urbains.
Mots-clés : non urbain, rétention des soins, rural, suppression virale, TAR, urbain, VIH/sida
Introduction
Antiretroviral therapy (ART) has dramatically reduced HIV-associated morbidity and mortality for persons with HIV (PWH), transforming HIV infection into a manageable chronic condition (1–3). Yet gaps in care remain and need to be addressed (4). In 2018, only 70% of Canadian PWH achieved an undetectable VL (5). Although achieving viral suppression in any population is the result of multiple factors, a potential barrier is location of residence (ie, living in an urban, semi-urban, or rural area) (6–9). One study examining the continuum of HIV care in rural communities in the United States and Canada noted that although the HIV epidemic was initially centred in large urban centres, more PWH are now residing in non-urban and rural areas in which far fewer HIV care resources are easily available (8). This study proposes that greater attention is required to understand the rural HIV epidemic (8).
In Canada, where large distances may separate PWH from skilled caregivers, residency in semi-urban and rural communities has been associated with lower rates of HIV testing, delays in HIV diagnosis, delays in starting ART, and a higher rate of HIV-related mortality compared with residency in urban centres (6,8–13). Increased stigma, social isolation, long distances to HIV social support and care, transportation, lack of access to providers with HIV expertise, and concerns about privacy and confidentiality in smaller non-urban or rural health care centres have all been identified as potential contributing factors (6–12,14). Some factors associated with inability to achieve viral suppression in non-urban settings are confidentiality concerns, discrimination, stigma, and challenges in reaching HIV care and prevention services (6,13). Significant controversies exist concerning what the optimal HIV care delivery model (ie, centralized, or localized) should be for PWH living non-urban areas (11,14–16). McClarty et al. and Jaworsky et al. identified geographical differences in viral suppression among PWH living in different geographic locations in British Columbia and Manitoba, respectively (17,18).
The HIV epidemic in southern Alberta, as in many other regions, is complex and heterogeneous. All PWH residing in southern Alberta receive free specialized HIV care delivered from the centralized Southern Alberta Clinic (SAC) in Calgary, with a catchment area estimated at 75,500 square kilometres. The majority (>85%) of patients live in the immediate Calgary area, whereas 15% live in semi-urban and rural areas, ranging from 80 to 300 kilometres or more (48–180 miles) away from SAC.
The objective of our study is to characterize HIV outcomes over time for people living in non-urban areas under a universal health care system with a centralized HIV care model. We hypothesized that health care outcomes for PWH in non-urban areas would be clinically similar to those for PWH living in urban settings.
Methods
Study population
All patients aged older than 16 years newly diagnosed with HIV living in southern Alberta and being followed at SAC from January 1, 2008, to January 1, 2020, were included. New PWH are automatically enrolled in the ongoing longitudinal Southern Alberta Cohort study and sign an informed consent form allowing the use of their data in research approved by the University of Calgary Conjoint Health Research Ethics Board. We undertook this study as a quality assurance exercise.
SAC is responsible for all HIV care in southern Alberta, as illustrated in Figure 1. SAC provides free HIV care and free ART to PWH. All PWH included in this study received their HIV care exclusively through the SAC program either directly or indirectly. Patients can have their laboratory investigations done on site at SAC, or they are given a list of facilities where these investigations can be done closer to their home. We defined urban as the area consisting of Calgary (estimated population of 1.4 million in 2020) and immediate areas with close access to Calgary (Figure 2) (19). Non-urban areas are detailed in Figure 2 and include all cities or towns outside the perimeter of the urban area. PWH diagnosed outside southern Alberta but receiving care at SAC were excluded. We conducted a descriptive model analysis classifying urban and non-urban areas initially subdivided into semi-urban (ie, towns) versus rural (ie, hamlets, farms, ranches) versus First Nations communities. However, because of the low number of people living in what we defined as rural and First Nations communities, we incorporated PWH in semi-urban, rural, and First Nations communities into one general category called non-urban. The final analysis reported is of urban versus non-urban areas.
Figure 1:
Catchment area for the Southern Alberta Clinic.
Note: Southern Alberta is delimited by Saskatchewan (eastern border), British Columbia (western border), Red Deer (northern border), and the United States (southern border) Source: Adapted from Google Maps
Figure 2:
Urban area in southern Alberta, consisting of Calgary and areas directly surrounding the city (non-urban areas are outside of this urban core area)
Notes: Urban is defined as Calgary and areas directly surrounding Calgary, including the towns of Cochrane, Airdrie, Chestermere, Okotoks, and Priddis. They are included per the Alberta Health Services definition of commuter communities, meaning easy commute to Calgary for work or business. Non-urban is defined as all other cities, towns, villages, hamlets, agricultural areas, and reserves situated outside this perimeter in Figure 2 and in southern Alberta (Figure 1). Non-urban includes semi-urban areas (meeting the criteria for definition of non-urban, with a population >10,000), rural areas (meeting criteria for our non-urban definition, with a population <10,000, including towns, villages, hamlets, and agricultural areas), and reserves (as defined under the Indian Act and treaties for exclusive use of an Indigenous band) (25). Source: Adapted from Google Maps
Socio-demographic variables
Variables of interest included gender identity (female, male, transgender), age at diagnosis (≤30 y, 31–40 y, 41–50 y, >50 y), self-reported ethnicity (Indigenous; African, Caribbean, or Black; Caucasian; other [including Asian, Hispanic, and self-reported ‘other’]), most likely HIV exposure factor (men having sex with men versus men having sex with women or women having sex with men versus persons who inject drugs versus other versus unknown).
HIV health outcomes
We evaluated three primary outcomes of interest: (1) immune status at the time of HIV diagnosis; (2) cumulative proportion of PWH who had a suppressed VL at 12, 24, and 36 months after diagnosis and in 2020; and (3) mortality. Immune status was defined by CD4 count at diagnosis, with delayed or late diagnosis being classified as a CD4 of less than 350 per cubic millimetre (20). Viral suppression was defined as less than 200 copies per millilitre and was determined by the last VL measurement at 12 months, 24 months, and 36 months after HIV diagnosis and the last VL in 2020 if still followed at SAC (21). Mortality was determined from the SAC database, which is updated monthly from the vital statistics registry, and was categorized as HIV or non-HIV related. Outcomes were initially compared between urban and non-urban areas and then among semi-urban, rural, and First Nations communities’ subgroups in the non-urban category.
Access to and retention in HIV care
Outcomes examined were (1) linkage to HIV care (ie, time from HIV diagnostic test to first HIV clinic visit), (2) retention in care (measured at 12 mo, 24 mo, and 36 mo after diagnosis and in 2020), and (3) use of ART (ever, then currently on ART measured at 12 mo, 24 mo, and 36 mo after diagnosis and in 2020). Retention in care was defined as having at least one clinical contact in the previous 12 months; at 12, 24, and 36 months after diagnosis; and in 2020. Clinical contact included a clinic visit or a telehealth visit. Telehealth visits are offered to clients who live outside of Calgary, have stable or controlled HIV, and have their routine laboratory work performed closer to their home. Telehealth visits must take place in an Alberta Health Services facility with telehealth capacity and a dedicated room.
We then identified patients who were lost to follow-up (LTFU), died, or moved out of southern Alberta at 12, 24, and 36 months after diagnosis and in 2020. LTFU was defined as not having any clinical contact for at least 12 months. For 2020 values, we included the outcomes of interest from 4 months before January 1, 2020. These outcomes were compared across the various populations, as with the primary outcomes discussed earlier.
Statistical analysis
We used standard descriptive statistics to summarize characteristics for our study population. Next, we calculated descriptive statistics using frequencies and percentages for our outcomes of interest both for the cohort overall and stratified by our specific population subgroups (urban versus non-urban).
Results
Baseline characteristics
Of 719 PWH newly diagnosed in southern Alberta during the study period, 619 (86%) were diagnosed in urban and 100 (14%) in non-urban areas. Only 18 of those PWH diagnosed in non-urban areas were diagnosed in rural areas, including 4 in First Nations communities. Because of these low numbers, PWH living in rural or First Nations communities were subsequently incorporated into the non-urban group for analysis. There were no statistical differences between urban and non-urban groups on baseline socio-demographic characteristics (Table 1).
Table 1:
Baseline socio-demographic characteristics of PWH diagnosed in southern Alberta from 2008 to 2020
| Characteristic | No. (%)* | p-value | ||
|---|---|---|---|---|
| Overall cohort; N = 719 | Urban; n = 619 | Non-urban; n = 100) | ||
| Gender identity | NS | |||
| Male | 567 (79) | 485 (78) | 82 (82) | |
| Female | 149 (21) | 133 (22) | 16 (16) | |
| Transgender | 3 (0.4) | 1 (0.2) | 2 (2) | |
| Age at HIV diagnosis, y | NS | |||
| ≤30 | 189 (26) | 164 (26) | 25 (25) | |
| 31–40 | 220 (31) | 186 (30) | 34 (34) | |
| 41–50 | 197 (27) | 173 (28) | 24 (24) | |
| >50 | 113 (16) | 96 (16) | 17 (17) | |
| Median (IQR) | 38 (30–47) | 36 (30–48) | 37 (28–45) | |
| Self-reported ethnicity | NS | |||
| Caucasian | 470 (65) | 400 (65) | 70 (70) | |
| Indigenous | 80 (11) | 70 (11) | 10 (10) | |
| ACB | 81 (11) | 72 (12) | 9 (9) | |
| Other | 80 (11) | 70 (11) | 10 (10) | |
| Not reported | 8 (1) | 7 (1) | 1 (1) | |
| Most likely HIV risk factor | NS | |||
| MSM | 355 (49) | 303 (49) | 52 (52) | |
| MSW–WSM | 282 (39) | 247 (40) | 35 (35) | |
| PWID | 65 (9) | 52 (8) | 13 (13) | |
| Other | 2 (0.3) | 2 (0.3) | 0 | |
| Unknown | 15 (2) | 15 (2) | 0 | |
* Unless otherwise indicated
PWH = Persons with HIV; NS = Not significant; IQR = Inter-quartile range; ACB = African, Caribbean, or Black; MSM = Men having sex with men; MSW/WSM = Men having sex with women–women having sex with men; PWID = Persons who inject drugs
Primary outcomes of interest
Immune status at diagnosis
We found no significant differences in immune status between the cohorts, with the majority of all PWH diagnosed with a CD4 count of less than 350 per cubic millimetre; the overall median CD4 was 298 per cubic millimetre [inter-quartile range 130 to 447/mm3] (Table 2).
Table 2:
Comparison of Initial CD4 count, time (days) to accessing HIV care and initiating antiretroviral therapy, missed clinical appointments and mortality between PWH diagnosed in either urban or non-urban settings in southern Alberta, Canada (2008–2020)
| Variable of interest | No. (%)* | p-value | ||
|---|---|---|---|---|
| Total; N = 719 | Urban; n = 619 | Non-urban; n = 100 | ||
| CD4 count at HIV diagnosis, mm3 | NS | |||
| ≤200 | 258 (36) | 223 (36) | 35 (35) | |
| 201–350 | 158 (22) | 139 (22) | 19 (19) | |
| 351–500 | 153 (21) | 136 (22) | 17 (17) | |
| >500 | 149 (21) | 121 (20) | 28 (28) | |
| Median (IQR) | 298 (30–47) | 294 (130–459) | 305 (78–459) | |
| Time from HIV diagnosis to 1st HIV clinic visit, d | <0.001 | |||
| ≤30 | 349 (49) | 324 (52) | 25 (25) | |
| 31–60 | 191 (27) | 161 (26) | 30 (30) | |
| 61–90 | 55 (8) | 43 (7) | 12 (12) | |
| >90 | 107 (15) | 81 (13) | 26 (26) | |
| No visits | 17 (2) | 10 (2) | 7 (7) | |
| Median (IQR) | 30 (20–54) | 29 (19–49) | 41 (28–103) | |
| Time from HIV diagnosis to ART start, d | <0.01 | |||
| ≤30 | 83 (12) | 74 (12) | 9 (9) | |
| 31–60 | 166 (23) | 151 (24) | 15 (15) | |
| 61–90 | 101 (14) | 93 (15) | 8 (8) | |
| >90 | 299 (42) | 245 (40) | 54 (54) | |
| Unknown | 70 (10) | 56 (9) | 20 (20) | |
| Median (IQR) | 81 (45–288) | 76 (44–258) | 127 (65–555) | |
| Missed regular clinic visits† | <0.001 | |||
| Total visits | 9,148 | 8,238 | 910 | |
| Attended | 7,717 (84) | 7,013 (85) | 703 (77) | |
| Missed | 1,432 (16) | 1,224 (15) | 207 (23) | |
| Mortality | NS | |||
| Total deaths | 69 (100) | 55 (79) | 14 (20) | |
| HIV- related | 29 (42) | 24 (44) | 5 (36) | |
| Non-HIV related | 36 (52) | 28 (51) | 8 (57) | |
| Unknown etiology | 4 (6) | 3 (5) | 1 (7) | |
* Unless otherwise indicated
† Missed regular clinic visits in total during the study period from January 1, 2008, to January 1, 2020
PWH = Persons with HIV; NS = Not significant; ARV = Anti-retroviral; IQR = Inter-quartile range
Rates of viral suppression
For the overall cohort, among patients with a recent (ie, between 4 mo before or 1 mo after) VL completed, 83% were suppressed at 12 months, 88% at 24 months, and 91% at 36 months after HIV diagnosis (Table 3). By 2020, 95% of active patients in care had a suppressed VL. The proportion of non-urban PWH with a suppressed VL on a recent test was lower at 12, 24, and 36 months and in 2020 than in urban PWH (73%, 81%, 83%, and 91% versus 84%, 88%, 92%, and 96%, respectively; p < 0.01). However, a higher proportion of non-urban patients compared with urban patients had not completed a recent VL (19%, 24%, 21%, and 9% versus 11%, 12%, 9%, and 4%, respectively; p < 0.001). By 2020, the proportions of PWH who completed their VL testing had increased in both groups (91% non-urban versus 96% urban). Table 3 shows in detail the differences between patients with VL completed or no VL completed, the VL percentage of all active patients versus the percentage of patients with VL completed within 4 months of the 12-, 24-, and 36-month follow-ups and in 2020.
Table 3:
Longitudinal trends after HIV diagnosis (baseline) through January 1, 2020
| Outcomes of interest | 12 mo* | 24 mo* | 36 mo* | 2020† | p-value |
|---|---|---|---|---|---|
| Retention in care‡ | <0.001 | ||||
| Total, n (% of baseline) | 671 (93) | 623 (87) | 535 (74) | 461 (64) | |
| Urban | 583 (94) | 541 (87) | 518 (84) | 411 (66) | |
| Non-urban | 88 (88) | 82 (82) | 68 (68)§ | 50 (50)§ | |
| Died | 26 (4) | 34 (5) | 41 (6) | 69 (10) | 0.71 |
| Urban | 18 (3) | 25 (4) | 32 (5) | 55 (9) | |
| Non-urban | 8 (8)* | 9 (9)* | 9 (9) | 14 (14) | |
| Moved | 22 (3) | 48 (7) | 60 (8) | 118 (16) | 0.1 |
| Urban | 18 (3) | 40 (6) | 49 (8) | 96 (16) | |
| Non-urban | 4 (4) | 8 (8) | 11 (11) | 21 (21)¶ | |
| LTFU | 0 | 13 (2) | 32(5) | 71 (10) | 0.4 |
| Urban | 0 | 12 (2) | 27 (4) | 56 (9) | |
| Non-urban | 0 | 1 (1) | 5 (5) | 15 (15) | |
| Retained patients on ART | 0.9 | ||||
| Total, no. (%) | 452 (67) | 427 (69) | 455 (85) | 427 (92) | |
| Urban | 400 (69) | 380 (70) | 403 (86) | 378 (92) | |
| Non-urban | 52 (60) | 47 (57)¶ | 54 (78)** | 53 (93) | |
| VL | |||||
| Total no. of active patients† † | 671 | 623 | 535 | 475 | <0.001 |
| VL completed† † | 591 (88) | 550 (86) | 525 (90) | 436 (5) | |
| VL not completed† † | 81 (12) | 83 (14) | 61 (10)¶ | 25 (5) | |
| Urban | 0.001 | ||||
| VL completed | 516 (89) | 479 (88) | 471 (91)¶ | 391 (95) | |
| VL not completed | 67 (12) | 63 (12) | 47 (9) | 20 (4) | |
| Non-urban | 0.2 | ||||
| VL completed | 71 (81) | 62 (76) | 54 (79) | 45 (90) | |
| VL not completed | 17 (19)¶ | 20 (24)** | 14 (21)§ | 5 (9) | |
| Of active patients who completed VL‡‡ | <0.001 | ||||
| Suppressed‡‡ | 485 (83) | 474 (88) | 478 (91) | 414 (95) | |
| Not suppressed‡‡ | 102 (17) | 67 (12) | 47 (9) | 22 (5) | |
| Urban | <0.001 | ||||
| Suppressed | 433 (84) | 423 (88) | 433 (92) | 376 (96) | |
| Not suppressed | 83 (16) | 56 (12) | 38 (8) | 15 (4) | |
| Non-urban | 0.1 | ||||
| Suppressed | 52 (73) | 50 (81) | 45 (83) | 41 (91) | |
| Non-suppressed | 19 (27)¶ | 12 (19)¶ | 9 (17)** | 4 (9) |
Notes: p-values were reported for each outcome of interest (for, eg, retention in care). The p-values between each time point of interest were only reported if significant.
Urban = Patients residing in Calgary and surrounding communities within 30 km (18 miles); non-urban = Patients residing outside of Calgary in either communities of <100,000 or in rural areas within southern Alberta, Canada LTFU = Lost to follow-up; ART = Anti-retroviral therapy; VL = Viral load suppression
* After HIV diagnosis
† As of January 1, 2020, if active
‡ N includes retention in care for all PWH without penalizing for people who died, moved away, or were LTFU
§ p < 0.001
¶ p < 0.05
**p < 0.01
† † Active = Followed in care and not moved, died, or LTFU
‡‡ ≤4 mo of 12-, 24-, or 36-mo follow-up or as of January 1, 2020, if active
Mortality
Of 69 PWH (10%) who died during the study, 29 (42%) died from HIV-related conditions, 36 (52%) from non-HIV-related conditions, and 4 (6%) from unknown causes (Table 2). No statistically significant differences in all-cause mortality were seen (14% non-urban, 13.8% urban), except non-urban patients were slightly more likely to die from non-HIV-related conditions (57% versus 51%). At each point in time, mortality was higher in the non-urban group than in the urban group (p < 0.05) (Table 3).
Secondary outcomes of interest
Linkage to HIV care
Small differences in time from HIV diagnosis to first clinic visit and to ART initiation were noted (Table 2). Fifty-two percent of urban patients attended their first HIV clinic visit within 30 days of HIV diagnosis compared with 25% in the non-urban cohort (p < 0.001). The median time until the first visit was only 12 days longer in the non-urban cohort (41 versus 29 d; p < 0.01). The difference in median time from diagnosis to starting ART was longer for non-urban PWH (127 versus 76 d, respectively; p < 0.01). Of non-urban patients, 26% had not had an initial HIV visit at SAC within 90 days compared with 13% of urban patients (p < 0.001).
Retention in care
Lower retention in care was seen among non-urban PWH (Table 3). At study conclusion in 2020, 66% of all urban PWH were retained in care compared with 50% of non-urban PWH (p < 0.001). By the end of the study, 69 patients had died, 118 moved, and 71 became LTFU. The percentages of those moving out of the region were comparable for the urban and non-urban patients. Non-urban PWH were more likely to miss scheduled clinical appointments (Table 2). Overall, 16% of visits were missed, with 23% missed by non-urban PWH compared with 15% missed by urban PWH (p < 0.001).
ART use
Of all PWH, 90% initiated ART at or shortly after their first HIV visit, with only slightly fewer PWH from non-urban areas (86% versus 91%, respectively; data not shown). Of all patients retained in care at 12 months, 67% were on ART—60% of PWH from non-urban areas compared with 69% of urban PWH (p < 0.05) (Table 3). At 24 months, the difference was 57% versus 70%, respectively (p < 0.02). At 36 months, however, those who were retained in care had much higher rates of being on ART (85% overall), although there were fewer non-urban patients than urban patients (78% versus 86%; p < 0.05). By 2020, both cohorts had more than 92% of retained patients on ART.
Discussion
We present the results of a 12-year longitudinal analysis of PWH followed at an HIV clinic under a centralized care model, comparing HIV outcomes for PWH living in non-urban areas versus those living in urban areas. Our results, although showing some differences between non-urban and urban PWH, were overall encouraging because the non-urban population is moving toward meeting the UNAIDS 2020 goals for rates of viral suppression (4). At baseline, we identified no socio-demographic differences between PWH with urban residency and PWH with non-urban residency. We also found no significant differences in CD4 count at HIV diagnosis, indicating that access (or no access) to HIV testing was similar for both groups. This is counter to previously reported data (8,10,11,13,14) and suggests that access to HIV diagnosis and initial HIV care was not necessarily, or significantly, more problematic among the non-urban population in our region. However, 36% of the overall cohort had a CD4 count of less than 200 per cubic millimetre at the time of the diagnosis, which suggests that although there were no differences between urban and non-urban groups, our overall cohort contained a large percentage who were diagnosed with advanced HIV disease. We also found small differences between both groups in rates of VL suppression, with increasing differences over time, and with the difference reduced to 92% versus 96% by 2020. However, for those who were retained in care, the UNAIDS goals for viral suppression were met (4). Last, although our overall results for primary outcomes were more positive than previously reported for PWH living in non-urban area (8,9–12), challenges remain.
We found small clinically significant differences in access to HIV care after diagnosis, with the median time until the first visit longer in the non-urban cohort. We identified lower retention in HIV care and lower ART use for the non-urban population at 12-, 24-, and 36-month follow-ups and in 2020. The non-urban PWH had higher LTFU rates, higher missed clinical appointments, and lower rates of VL completion than the urban PWH. Other studies have noted difficulties in transportation to clinics and lack of care providers with HIV expertise as factors associated with poorer HIV outcomes in non-urban areas (13,23,24). Transportation issues may also affect PWH’s timely travel to Calgary, especially during the winter and with limited commercial transportation. Lack of confidentiality and stigma in smaller non-urban settings have also been cited as impediments to some PWH’s access to local care, including lab tests and even telemedicine (14). Such concerns have led to some electing to receive all HIV care (including HIV-specific laboratory and ART dispensing) in the central clinic although this necessitates significant travel. Last, our study shows that mortality was higher in the non-urban group at each studied point in time. These findings are concerning, and although no statistically significant differences in all-cause mortality were seen, non-urban patients were more likely die from non-HIV-related conditions (57% versus 51%). These findings are similar to those of another Canadian study that found that PWH in rural areas had increased mortality from noncommunicable disease (22). Further attention is needed to understand the factors associated with increased mortality for non-urban PWH.
Controversy surrounds what constitutes the best delivery of care to the non-urban population living with HIV in high-income countries—a centralized or a localized model of care (6,8,10,11,14–16). SAC delivers centralized HIV care under a universal health care system across a large geographical area. Over the past 20 years, SAC has developed multiple strategies to increase retention in care for PWH living in the region outside of urban Calgary. Telehealth programs were implemented to improve non-urban HIV care more than 10 years ago (ie, 2010) for PWH for whom transportation and confidentiality were limitations to accessing HIV care (17). PWH who are eligible for telehealth can have access to full SAC services, which include a visit with an HIV doctor, a registered nurse, a social worker, a pharmacist, and a registered dietician. In addition, clinic social workers monitor patients who have missed their appointments more than five times over the past 2–3 years. They investigate the barriers that contribute to missed appointments and discuss strategies with both patient and HIV care providers. SAC also has alliances with a rural AIDS service organization to improve HIV awareness, testing, transportation, ART adherence, and linkage to care. Finally, SAC offers free delivery of ART to the patient’s home or community pharmacy if the patient is unable to travel to the clinic, which may improve adherence. Despite these multiple strategies, differences remain. Our study highlights the ongoing challenges in optimizing the HIV care model and difficulties in finding a perfect fit for all patients (5,23).
Our study has several limitations. Both the geography and weather challenges in Alberta may not match those in other regions. Similarly, the social dynamics of the population of PWH, use of telehealth, payment systems, and access to even minimal medical services such as phlebotomy may be different from those in other areas. The dynamics of the HIV epidemic and model of care delivery (eg, newer ART, fewer annual CD4 or VL tests) has changed over time, which affects any longitudinal study. The definition and configuration of what constitutes urban, semi-urban, and rural communities remains problematic and may have an impact on examinations of health care outcomes. Moreover, a recent study looking at the influence of the definition of rurality on geographic differences in HIV outcomes in British Columbia showed that health outcome findings can vary depending on the definition of the geographic variable (18). We could not do a thorough analysis of people diagnosed in First Nations communities, or in rural areas, because of their overall low number. This could potentially be explained by low overall testing in those areas or people’s choice to get tested in a bigger centre to avoid lack of confidentiality. However, we could not specifically examine HIV health outcomes for PWH in First Nations communities or rural areas. We defined and included PWH on the basis of where their HIV diagnosis was made, which could introduce selection bias and may not reflect the actual location of residence at each point in time during the study. We excluded PWH who moved into the region and may live in non-urban areas. We did not monitor the subsequent movement of PWH who moved from non-urban areas to urban areas or those who moved from urban to non-urban areas after HIV diagnosis. We used a worst-case scenario that PWH without recent VLs were unsuppressed, which may not be true because we did not reconcile reported use of ART with ART dispensing records, and patients may have skipped their bloodwork but still have been taking their ART. Last, we did not specifically look at what proportion of our non-urban patients used telemedicine and how efficacious it was for them.
Conclusion
Our longitudinal study showed that for PWH who can and do access regular care, health care outcomes are very similar between non-urban and urban groups. However, ongoing difficulties in HIV care exist. Time to first clinic visit and time to ARV start were longer for PWH in non-urban areas. Retention in care and regular bloodwork, while meeting UNAIDS standards, were also more challenging for PWH living in non-urban areas and need to be addressed. Identifying and reducing the physical, structural, geographical, and psychological barriers to care remains paramount to improve HIV health outcomes for PWH.
Acknowledgements:
The authors thank Dr Matthieu Parenteau for assisting with the data analysis; Quang Vu for his continued work on data retrieval; and the SAC staff, including nurses, social workers, administrative assistants, and pharmacists, for their insights and knowledge of health care utilization interventions for non-urban patients.
Authorship Confirmation:
This statement is to certify that each author listed has met the definition of authorship by the International Committee of Medical Journal Editors.
Ethics Approval:
The University of Calgary Conjoint Health Research Ethics Board approved this study.
Informed Consent:
Patients signed an informed consent allowing the use of their data in research approved by the University of Calgary Conjoint Health Research Ethics Board.
Registry and the Registration No. of the Study/Trial:
N/A
Funding:
No funding was received for this work.
Disclosures:
MJ Gill reports being an ad hoc advisor on national HIV advisory boards to Gilead, ViiV, and Merck. The other authors have nothing to disclose.
Peer Review:
This manuscript has been peer reviewed.
Animal Studies:
N/A
References
- 1.Cohen MS, Chen YQ, McCauley M, et al. Antiretroviral therapy for the prevention of HIV-1 transmission. N Engl J Med. 2016;375(9):830–9. 10.1056/NEJMoa1600693. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Lohse N, Obel N. Update of survival for persons with HIV infection in Denmark. Ann Intern Med. 2016;165(10):749–50. 10.7326/L16-0091. Medline: [DOI] [PubMed] [Google Scholar]
- 3.Samji H, Cescon A, Hogg RS, et al. Closing the gap: increases in life expectancy among treated HIV-positive individuals in the United States and Canada. PLoS One. 2013;8(12):e81355. 10.1371/journal.pone.0081355. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.UNAIDS. UNAIDS Data 2019 . 2019. https://www.unaids.org/sites/default/files/media_asset/2019-UNAIDS-data_en.pdf (November 11, 2020).
- 5.Canada Go. Summary: estimates of HIV incidence, prevalence and Canada’s progress on meeting the 90-90-90 HIV targets, 2018. https://www.canada.ca/en/public-health/services/publications/diseases-conditions/summary-estimates-hiv-incidence-prevalence-canadas-progress-90-90-90.html (June 7, 2021.
- 6.Pellowski JA. Barriers to care for rural people living with HIV: a review of domestic research and health care models. J Assoc Nurses AIDS Care. 2013;24(5):422–37. 10.1016/j.jana.2012.08.007. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.UNAIDS. Rural HIV/AIDS. http://data.unaids.org/topics/gender/ruralhivaids_en.pdf (October 30, 2020).
- 8.Schafer KR, Albrecht H, Dillingham R, et al. The continuum of HIV care in rural communities in the United States and Canada: what is known and future research directions. J Acquir Immune Defic Syndr. 2017;75(1):35–44. 10.1097/QAI.0000000000001329. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Schranz AJ, Barrett J, Hurt CB, Malvestutto C, Miller WC. Challenges facing a rural opioid epidemic: treatment and prevention of HIV and hepatitis C. Curr HIV/AIDS Rep. 2018;15(3):245–54. 10.1007/s11904-018-0393-0. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.MacKenzie LJ, Hull MW, Samji H, et al. Is there a rural/urban gap in the quality of HIV care for treatment-naive HIV-positive individuals initiating antiretroviral therapy in British Columbia? AIDS Care. 2017;29(10):1218–26. 10.1080/09540121.2017.1322678. Medline: [DOI] [PubMed] [Google Scholar]
- 11.Rapid Response Service.: HIV services in rural and remote communities. 2013. https://www.ohtn.on.ca/wp-content/uploads/2016/12/RR73-2013-Rural-ASO.pdf (October 30, 2020).
- 12.Schafer KR, Albrecht H, Dillingham R, et al. The continuum of HIV care in rural communities in the United States and Canada: what is known and future research directions. J Acquir Immune Defic Syndr. 2017;75(1):35–44. 10.1097/QAI.0000000000001329. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sullivan PS, Satcher Johnson A, Pembleton ES, et al. Epidemiology of HIV in the USA: epidemic burden, inequities, contexts, and responses. Lancet. 2021;397(10279):1095–106. 10.1016/S0140-6736(21)00395-0. [DOI] [PubMed] [Google Scholar]
- 14.Cyr ME, Etchin AG, Guthrie BJ, Benneyan JC. Access to specialty healthcare in urban versus rural US populations: a systematic literature review. BMC Health Services Research. 2019;19(1):974. 10.1186/s12913-019-4815-5. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ohl ME, Richardson K, Rodriguez-Barradas MC, et al. Impact of availability of telehealth programs on documented HIV viral suppression: a cluster-randomized program evaluation in the Veterans Health Administration. Open Forum Infectious Diseases. 2019;6(6). 10.1093/ofid/ofz206. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Worthington C, Mollison A, Herman T, et al. A qualitative study of community-based HIV/AIDS prevention interventions, programs, and projects for rural and remote regions in Canada: implementation challenges and lessons learned. J Public Health Manag Pract. 2020;26(1):E28–e37. 10.1097/PHH.0000000000000878. Medline: [DOI] [PubMed] [Google Scholar]
- 17.McClarty LM, Blanchard JF, Becker ML. Leaving no one behind? An equity analysis of the HIV care cascade among a cohort of people living with HIV in Manitoba, Canada. BMC Public Health. 2021;21(1):281. 10.1186/s12889-021-10225-w. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jaworsky D, Loutfy M, Lu M, et al. Influence of the definition of rurality on geographic differences in HIV outcomes in British Columbia: a retrospective cohort analysis. CMAJ Open. 2020;8(4):E643–E50. 10.9778/cmajo.20200066. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Government of Alberta. Calgary—Population. 2021. https://regionaldashboard.alberta.ca/region/calgary/population/#/?from=2016&to=2020 (November 22, 2021).
- 20.Althoff KN, Gange SJ, Klein MB, et al. Late presentation for human immunodeficiency virus care in the United States and Canada. Clin Infect Dis. 2010;50(11):1512–20. 10.1086/652650. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Marks G, Gardner LI, Rose CE, et al. Time above 1500 copies: a viral load measure for assessing transmission risk of HIV-positive patients in care. Aids. 2015;29(8):947–54. 10.1097/QAD.0000000000000640. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Burchell AN, Raboud J, Donelle J, et al. Cause-specific mortality among HIV-infected people in Ontario, 1995–2014: a population-based retrospective cohort study. CMAJ Open. 2019;7(1):E1–E7. 10.9778/cmajo.20180159. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Becker ML, Kasper K, Pindera C, et al. Characterizing the HIV epidemic in the Prairie Provinces. Can J Infect Dis Med Microbiol. 2012;23(1):19–22. 10.1155/2012/408561. Medline: [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Hanson E. What are Indian Reserves? 2009. https://indigenousfoundations.arts.ubc.ca/reserves/ (October 10, 2021).