Dataset Information

Predictive Analytics for Retention in Care in an Urban HIV Clinic.

ABSTRACT: Consistent medical care among people living with HIV is essential for both individual and public health. HIV-positive individuals who are 'retained in care' are more likely to be prescribed antiretroviral medication and achieve HIV viral suppression, effectively eliminating the risk of transmitting HIV to others. However, in the United States, less than half of HIV-positive individuals are retained in care. Interventions to improve retention in care are resource intensive, and there is currently no systematic way to identify patients at risk for falling out of care who would benefit from these interventions. We developed a machine learning model to identify patients at risk for dropping out of care in an urban HIV care clinic using electronic medical records and geospatial data. The machine learning model has a mean positive predictive value of 34.6% [SD: 0.15] for flagging the top 10% highest risk patients as needing interventions, performing better than the previous state-of-the-art logistic regression model (PPV of 17% [SD: 0.06]) and the baseline rate of 11.1% [SD: 0.02]. Machine learning methods can improve the prediction ability in HIV care clinics to proactively identify patients at risk for not returning to medical care.

SUBMITTER: Ramachandran A

PROVIDER: S-EPMC7156693 | biostudies-literature | 2020 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predictive Analytics for Retention in Care in an Urban HIV Clinic.

Ramachandran Arthi A Kumar Avishek A Koenig Hannes H De Unanue Adolfo A Sung Christina C Walsh Joe J Schneider John J Ghani Rayid R Ridgway Jessica P JP

Scientific reports 20200414 1

Consistent medical care among people living with HIV is essential for both individual and public health. HIV-positive individuals who are 'retained in care' are more likely to be prescribed antiretroviral medication and achieve HIV viral suppression, effectively eliminating the risk of transmitting HIV to others. However, in the United States, less than half of HIV-positive individuals are retained in care. Interventions to improve retention in care are resource intensive, and there is currently ...[more]

PMID: 32286333

Similar Datasets

Project description:BACKGROUND:Mental health (MH) comorbidities reduce retention in care for persons living with HIV (PLWH) and are associated with poor health outcomes. Optimizing retention in primary care is vital, as poor retention is associated with delayed receipt of antiretroviral (ARV) therapy, ARV non-adherence, and poor health outcomes, including failure to suppress viral load, decreased CD4 counts, and clinically significant ARV drug resistance. We hypothesized that MH service utilization would be associated with improved retention in care for patients with HIV and MH comorbidities. METHODS:This is a retrospective analysis of PLWH initiating outpatient HIV health care at a university-affiliated HIV clinic between January 2007 and December 2013. We examined the association between MH service utilization and retention in care, the outcome of interest, using univariate and multivariable logistic regression. RESULTS:Overall, 627 (84.4%) out of 743 patients were retained in care using the Health Resources & Services Administration HIV/AIDS Bureau (HRSA/HAB) metric. A multivariable model adjusted for several sociodemographic factors, MH comorbidities, and MH service utilization. The results suggest that lack of health insurance (public ORadj = 0.3, p < 0.01; no insurance ORadj = 0.4, p < 0.01) and ? 3 MH comorbidities (ORadj = 0.3, P = 0.01) were associated with decreased retention in care. Conversely, older age (> 45 years, ORadj. = 1.6, p = 0.14) and ? 3 MH service utilization visits (ORadj. = 6.8, p < 0.01) were associated with increased retention in care. CONCLUSIONS:Even in the absence of documented MH comorbidities, improved retention in care was observed with increasing MH service utilization. In order to achieve the US-based National HIV/AIDS Strategy goal of 90% retention in care for PLWH, MH service utilization should be considered along with other evidence-based interventions to improve retention for PLWH newly engaged in care.

Project description:Few data exist on the use of text messaging as a tool to promote retention in HIV care and virologic suppression at the clinic level in the United States. We describe the protocol for a study designed to investigate whether a text messaging intervention that supports healthy behaviors, encourages consistent engagement with care, and promotes antiretroviral persistence can improve retention in care and virologic suppression among patients in an urban safety-net HIV clinic in San Francisco.Connect4Care (C4C) is a single-site, randomized year-long study of text message appointment reminders vs. text message appointment reminders plus thrice-weekly supportive, informational, and motivational text messages. Eligible consenting patients are allocated 1:1 to the two arms within strata defined by HIV diagnosis within the past 12 months (i.e. "newly diagnosed") vs. earlier. Study participants must receive primary care at the San Francisco General Hospital HIV clinic, speak English, possess a cell phone and be willing to send/receive up to 25 text messages per month, a have viral load >200 copies/?L, and be either new to the clinic or have a history of poor retention. The primary efficacy outcome is virologic suppression at 12 months and the key secondary outcome, which will also be examined as a mediator of the primary outcome, is retention in HIV care, as operationalized by kept and missed primary care visits. Process outcomes include text message response rate and percent of time in study without cell phone service. Generalized estimating equation log-binomial models will be used for intent to treat, per protocol, and mediation analyses. An assessment of the cost and cost-effectiveness of the intervention is planned along with a qualitative evaluation of the intervention.Findings from this study will provide valuable information about the use of behavioral-theory based text messaging to promote retention in HIV care and virologic suppression, further elucidate the challenges of using texting technology with marginalized urban populations, and help guide the development of new mobile health strategies to improve HIV care cascade outcomes.NCT01917994.

Project description:OBJECTIVE:To determine retention in care and virologic suppression among HIV-infected adolescents and young adults attending an adolescent-friendly clinic compared to those attending the standard pediatric clinic at the same site. DESIGN:Retrospective cohort analysis. SETTING:Government supported, hospital-based antiretroviral clinic in KwaZulu-Natal, South Africa. PARTICIPANTS:Two hundred forty-one perinatally HIV-infected adolescents and young adults aged 13 to 24 years attending an adolescent-friendly clinic or the standard pediatric clinic from April 2007 to November 2015. INTERVENTION:Attendance in an adolescent-friendly clinic compared to a standard pediatric clinic. OUTCOMES MEASURES:Retention in care defined as one clinic visit or pharmacy refill in the prior 6 months; HIV-1 viral suppression defined as < 400 copies/ml. RESULTS:Overall, among 241 adolescents and young adults, retention was 89% (214/241) and viral suppression was 81% (196/241). Retention was higher among those attending adolescent clinic (95%) versus standard pediatric clinic (85%; OR 3.7; 95% confidence interval (CI) 1.2-11.1; p = 0.018). Multivariable logistic regression adjusted for age at ART initiation, gender, pre-ART CD4 count, months on ART, and tuberculosis history indicated higher odds of retention in adolescents and young adults attending adolescent compared to standard clinic (AOR = 8.5; 95% CI 2.3-32.4; p = 0.002). Viral suppression was higher among adolescents and young adults attending adolescent (91%) versus standard pediatric clinic (80%; OR 2.5; 95% CI 1.1-5.8; p = 0.028). A similar multivariable logistic regression model indicated higher odds of viral suppression in adolescents and young adults attending adolescent versus standard pediatric clinic (AOR = 3.8; 95% CI 1.5-9.7; p = 0.005). CONCLUSION:Adolescents and young adults attending an adolescent-friendly clinic had higher retention in care and viral suppression compared to adolescents attending the standard pediatric clinic. Further studies are needed to prospectively assess the impact of adolescent-friendly services on these outcomes.

Dataset Information

Predictive Analytics for Retention in Care in an Urban HIV Clinic.

Publications

Predictive Analytics for Retention in Care in an Urban HIV Clinic.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets