In countries under universal health care programs, individuals with HIV infection can move relatively freely and access equivalent health care in different locations within their country. In such programs, HIV-infected individuals can engage, disengage, and reengage with care at anytime for whatever reason. From a health perspective, however, the issue becomes problematic when mobility or disengagement impacts continuity of care leading potentially to negative effects on health. Studies show that continuous monitoring and treatment of HIV infection leads to better health outcomes than interrupted treatment.1-9 Patients who become lost-to-follow-up (LTFU) exhibit higher mortality and morbidities rates.10-13 Much less is known about patients who move (or state they are moving) with transfer of care from one HIV center to another. This demographic group is often overlooked in mobility studies. It is often assumed, although not known, that these patients resume HIV care with a new care-giver shortly after relocating.
The objective of this study was to examine the characteristics and outcomes of patients who “move away” and then return to our HIV care center to assess the impact of moving on disease progression when compared with patients remaining continuously in care. As far as we know, we are the first to assess health outcomes for patients who “move” rather than being exclusively LTFU. We use the term “moved” rather than “transferring care” as the latter term may refer to patients simply transferring care from one center to another without necessitating a change of home address. We use “moved” in a broader sense to signify both geographical relocation and the obligatory transfer of care from our cohort. Physically moving home and transferring care may create significant difficulties impacting continuity of care. Our findings can contribute to developing strategies that encourage the continuity of disease management for those who geographically move.
All HIV-positive individuals living in southern Alberta are referred to the centralized Southern Alberta Clinic (SAC) program for their HIV care and are enrolled in the Southern Alberta Cohort, which represents a regional HIV population. SAC provides exclusive and comprehensive HIV care including antiretroviral therapy (ART), HIV-specific laboratory testing, on an ongoing basis.
All adult (>16 years) patients initiating HIV care at SAC between January 1, 2000, and January 1, 2008, were included and followed until they moved, were LTFU or until July 1, 2009. We excluded patients who died during the study period and patients who were part of the SAC/Corrections Canada HIV Care Program (ie, patients followed in local prisons).
Patients were placed into 3 groups: (1) “continuous care patients”-patients followed continuously at SAC; (2) “discontinuous patients”-patients who ever left SAC either by moving out of the region or being LTFU. After individual chart reviews, all discontinuous patients were categorized as either moved when they informed SAC of their relocation outside the region (and received, if on ART, a 3-month drug supply to facilitate transition to another HIV center), and/or if their medical records were requested by another care centre outside the region, or as LTFU if they had no clinic visits, clinic contacts, local hospitalizations, or HIV-related laboratory tests within the region for 12 months after their last clinic contact; (3) “Returning Patients”-a subset of discontinuous patients who left and then returned to care.
Demographic data (gender, age at HIV diagnosis, self-reported ethnicity, most likely HIV risk factor) at initial clinic visit and clinical data (ie, CD4 count, date of HIV diagnosis, AIDS diagnosis, ART use) at first and subsequent visits were taken directly from the SAC database.
To determine if returning patients accessed HIV care while away from SAC, we examined all transferred medical records of these patients for receipt of ART, and/or CD4 or Viral Load testing while away.
Patients accessing care have signed a voluntary informed consent form allowing administrative data to be used in research projects as approved by the University of Calgary Ethics Committee.
Impact on Health
To assess in all returning patients the impact of moving on disease progression, we used “change in CD4 count” as a proxy measure of health status. We compared the most recent CD4 count immediately before leaving HIV care to the first CD4 count after returning. Change in health status was determined by the difference of the before and after CD4 count divided by the number of days between CD4 dates multiplied by 30 to determine the rate of increase or decrease per month for each patient. We also evaluated AIDS events before and during the absence.
Simple descriptive and statistical analysis was performed to describe the comparative populations with univariate analysis (ie, χ2 testing) used where appropriate with P < 0.05 as the level of significant. We compared the demographic and clinical characteristics between “continuous and discontinuous patients; and discontinuous patients” who left and did not return to “returning patients”; and “returning patients” who moved compared with patients who were LTFU.
Between January 1, 2000, and January 1, 2008, 920 HIV-positive adults initiated care at SAC. Of these, 8 died, 64 patients were followed through the SAC/Corrections Canada HIV Care Program, and 12 patients had incomplete medical records. Of the 836 study participants, 465 (56%) remained in continuous HIV care and 371 (44%) left care at some point; 229 (62%) of these 371 “discontinuous patients” left (177 moved and 52 LTFU) and did not return, whereas 142 (38%) who had left (57 moved, 85 LTFU) subsequently returned. Transfer of care between HIV centers was verified for 63% of patients who stated they were moving and did not return, but only 33% for patients who moved and subsequently returned. The sociodemographic and clinical characteristics of all 3 groups are presented in Table 1.
“Returning patients” had similar sociodemographic and clinical characteristics to “continuous patients”. There were no sociodemographic or clinical differences between returning patients who had formally moved (ie, transferring HIV care) and patients who were temporarily LTFU. “Returning patients” who had “moved” returned after a median absence of 194 (IQR: 83-434) days although LTFU returned after a median of 574 (444-802) days.
Impact on Disease Progression
Ninety-one percent of “returning patients” had CD4 counts taken before and after returning to SAC (Table 2). Sixty-six percent of the moved patients and 41% of the LTFU patients were on ART when they left local care. Median CD4 count before leaving was 405 cells per cubic millimeter (IQR: −252 to 565); median CD4 count upon returning was 270 cells per cubic millimeter (175-459). Patients who moved showed a decrease from a median of 424 cells per cubic millimeter (276-574) to 339 cells per cubic millimeter (174-534), with a median rate of decline of −7.2 cells per cubic millimeter per 30 patient-days followed. Patients LTFU had a decrease from a median CD4 of 382 cells per cubic millimeter (230-524) to 240 cells per cubic millimeter (161-409) with a median rate of decline of −8.1 cells per cubic millimeter per 30 patient-days. This contrasts to the median CD4 count of those staying in local care where the median rate increased by +4.3 cells per cubic millimeter per 30 patient-days of follow-up.
Although 13% of “continuous patients” had an AIDS diagnoses at initial visit compared with 9% of “discontinuous patients” and 3% of “returning patients,” only 7.4% of “continuous patients” experienced a subsequent new AIDS event compared with 15% of “returning patients” who were diagnosed with AIDS during or soon after their absence from local care.
Only 2 of 31 (6.5%) antiretroviral (ARV)-experienced patients who moved and none of the ARV-experienced LTFU patients reported receiving antiretroviral drugs although away from SAC beyond the 3-months supply of ARVs given at last SAC visit reflecting a significant disconnection in staying on ART even in those who formally moved.
The churn rate14 (ie, the rate of in-and-out migration of HIV patients) is high in our population with 44% of all patients leaving during the past 8 years. The overall incidence rate of patients who moved was 4.76 per 100 patient years of follow-up (95% confidence interval: 4.12 to 5.81) and was much higher than LTFU patients at 2.8 per 100 patient years of follow-up (95% confidence interval: 2.67 to 3.12), and also higher than reported by Mocroft et al12 (3.72 per 100 patient years of follow-up) and Ndiaye et al11 (3.5 per 100 patient years of follow-up) for their LTFU patients. These differences may be partially explained by LFTU definitions and the new use of term “moved” alongside LFTU.
Overall, 39% of all patients who left returned, however, only 1 in 4 of all patients who “moved” returned to local care. Most patients who moved informed SAC in advance of when they planned to relocate, and if on therapy, received a transition 3-month supply of ART. However, the absence of a request for previous records indicating a successful clinical transfer and relocation in 2 of 3 of patients who move and returned to care is suggestive that despite their original intentions many never connected for HIV care in their new location. It is possible, but against standard practice, that the new provider relied on patient recall of previous laboratory results and interventions rather than requesting records.
The adverse effect on HIV health is apparent in those who returned; the decline in CD4 counts of both those who “moved” and those LFTU is very similar and likely driven by lack of ART. These data suggest that some patients who move (like the LFTU patients) may fail or delay seeking HIV care in new environments if they lack the knowledge on how to access HIV care or if circumstances in their lives prevent them from accessing services in a timely fashion. As health care is universal and access is essentially equivalent across Canada (as specified in the Canada Health Act), there should not be barriers to accessing equivalent HIV care. Our results show that this may not be the case.
Regardless of the reason for leaving, we found that when compared with remaining patients, the majority of returning patients had experienced a significant and measurable reduction in health as proxied by both a reduced CD4 count and a proportional increase in new AIDS events. Lima et al15 reported lower levels of health among persons living with HIV/AIDS who moved within British Columbia. Ndiaye et al13 found CD4 counts for returning LTFU patients fell from 401 cells per cubic millimeter to 305 cells per cubic millimeter over 19 months and the incidence of AIDS and death significantly higher. Our results concur with their findings. Patients who disengage are, in effect, allowing HIV infection to run its natural course, at least in the short term. The decline in CD4 count of some “movers” suggests that some of this population should also be identified as being at higher risk for disconnecting from HIV care.
Our study does have limitations. Although demographic and clinical characteristics are similar, returning patients may have different reasons and motivation for returning than those who left permanently. We could not measure the outcomes in patients who moved or were LFTU and who never return to our program. Our population, like any HIV population, is unique both demographically and geographically and is situated in an economic environment that may impact on choices of staying or leaving the region. More studies are needed to determine if our population and our results are similar to or different than data obtained elsewhere.
Within our HIV population, mobility remains high. Although disengaging from care (ie, LTFU) may be more damaging, we found that in many, moving also had negative health effects. Our results suggest that our local efforts to transfer care seamlessly to other HIV care centers elsewhere needs to be reinforced to avoid negative health consequences. In the future, easy sharing of pertinent medical information readily, perhaps using electronic medical records, could facilitate better continuity of care for individuals with HIV infection. Recognizing and addressing patient mobility, both planned and unplanned, may be another challenge for the “seek and treat” strategy16 being discussed.
We wish to acknowledge the staff and patients at the SAC for their support especially Jonathan Snider, Ian Cosman, and Danette Mohagen for maintaining the SAC database.
1. The Strategies for Management of Antiretroviral Therapy (SMART) Study Group. CD4+ count-guided interruption of antiretroviral treatment. N. Eng J Med
2. The Antiretroviral Therapy (ART) Cohort Collaboration. Prognosis of HIV-1 infected patients up to 5 years after initiation of HAART: collaborative analysis of prospective studies. AIDS
3. Egger M, May M, Chene G, et al. Prognosis of HIV-1 infected patients starting highly active antiretroviral therapy: a collaborative analysis of prospective studies. Lancet
4. Sterne JA, Hernan MA, Ledergerber B, et al. Long-term effectiveness of potent antiretroviral therapy in preventing AIDS and death: a prospective cohort study. Lancet
5. Kitahata MM, Gange SJ, Abraham AG, et al. Effect of early versus deferred antiretroviral therapy for HIV on survival. N. Eng J Med
6. Chadborn TR, Delpech VC, Sabin CA, et al. The late diagnosis and consequent short-term mortality of HIV-infected heterosexuals (England and Wales, 2000-2004) AIDS
7. Sabin CA, Smith CJ, Gumley H, et al. Late presenters in the era of highly active antiretroviral therapy: update of and responses to highly active antiretroviral therapy. AIDS
8. Girardi E, Sabin CA, d'Armino Monforte A. Late diagnosis of HIV infection: epidemiological features, consequences, and strategies to encourage earlier testing. J Acquir Immune Defic Syndr
9. Adler A, Mounier-Jack S, Coker RJ. Late diagnosis of HIV in Europe: definitional and public health challenges. AIDS Care
10. Lanoy E, Mary-Krause M, Tattevin P, et al. Predictors identified for losses to follow-up among HIV-seropositive patients. J Clin Epidemiol
11. Ndiaye B, Ould-Kaci K, Bataille P, et al. Incidence rate and risk factors for loss to follow-up in five French clinical centres of HIV-infected patient in Northern France - January 1997 to December 2006. Antivir Ther
12. Mocroft A, Kirk O, Aldins P, et al. Loss to follow-up in an international, multicentre observational study. HIV Medicine
13. Ndiaye B, Ould-Kaci K, Salleron J, et al. Characteristics of and outcomes in HIV-infected patients who return to care after loss to follow-up. AIDS
14. Gill MJ, Krentz HB Unappreciated epidemiology: the churn effect in a regional HIV care a programme. Int J STD AIDS
15. Lima V, Fernandes K, Rachlis B, et al. Migration adversely affects antiretroviral adherence in a population-based cohort of HIV/AIDS patients. Soc Sci Med
16. Marks G, Gardner LI, Craw J, et al. Entry and retention in medical care among HIV-diagnosed persons: a meta-analysis. AIDS