Abstract
Background
Public health information exchanges (HIEs) link real-time surveillance and clinical data and can help to re-engage out-of-care people with HIV (PWH).
Methods
We conducted a retrospective cohort study of out-of-care PWH who generated an HIE alert in the Grady Health System (GHS) Emergency Department (ED) between January 2017 and February 2018. Alerts were generated for PWH who registered in the GHS ED without Georgia Department of Public Health (GDPH) CD4 or HIV-1 RNA in the prior 14 months. The alert triggered a social work (SW)–led re-linkage effort. Multivariate logistic regression analyses used HIE-informed SW re-linkage efforts as the independent variable, and linkage to care and 3- and 6-month viral suppression (HIV-1 RNA < 200 c/mL) as primary outcomes. Patients admitted to the hospital were excluded from primary analysis.
Results
One hundred forty-seven out-of-care patients generated an alert. Ninety-eight were included in the primary analysis (mean age [SD], 41 ± 12 years; 70% male; 93% African American), and 20 received the HIE-informed SW intervention. Sixty percent of patients receiving the intervention linked to care in 6 months, compared with 35% who did not. Patients receiving the intervention were more likely to link to care (adjusted risk ratio [aRR], 1.63; 95% confidence interval [CI], 0.99–2.68) and no more likely to achieve viral suppression (aRR, 1.49; 95% CI, 0.50–4.46) than those who did not receive the intervention.
Conclusions
An HIE-informed, SW-led intervention systematically identified out-of-care PWH and may increase linkage to care for this important population. HIEs create an opportunity to intervene with linkage and retention strategies.
Keywords: implementation science, linkage to care, health information exchange, viral suppression, care continuum, information sharing
Advances in HIV treatment transformed HIV infection from an almost certain death sentence into a manageable chronic condition [1]. To realize these benefits, patients must actively and continuously engage in outpatient HIV treatment for their entire lives. Over 50% of people with HIV (PWH) are not retained in HIV care, increasing the risk of disease progression, death, and virus transmission [2, 3]. Additionally, 22% of individuals newly diagnosed with HIV do not link to care within 1 month [4]. Considering that 60% of new HIV infections occur from those who are previously diagnosed but not retained in care [5, 6], interventions targeting these populations are of critical importance in improving receipt of regular care, achieving viral suppression, and preventing HIV transmission. The population of diagnosed but poorly retained patients may face a number of barriers to (re)-linkage to care, including lack of insurance, poor access to health care, housing instability, stigma, transportation difficulties, denial of diagnosis, or comorbidities such as substance use or mental illness [7, 8]. Due to these barriers, traditional linkage and re-linkage efforts have failed to reach this population. Therefore, novel interventions are needed to engage this important population in care.
Linkage to care is especially important in the Southern United States, which is the epicenter of the domestic HIV epidemic. The Southern region accounts for 52% of new HIV diagnoses and 47% of deaths attributable to HIV/AIDS yet only 38% of the nation's population. Georgia is the state with the highest rate of new HIV diagnoses in the country, at 24.9/100 000 people [9]. Additionally, individuals newly diagnosed in the Southern United States are more likely to delay initiating ART and are more likely to experience complications from HIV [10, 11]. HIV care engagement is worse in the South, and re-linkage and retention interventions are especially important in this region.
Despite poor engagement in HIV-related medical care, PWH continue to access the health care system at large, particularly the emergency department (ED) [12]. Often these visits are not for HIV-related medical conditions, and the medical provider may remain unaware that the patient has HIV infection and is out of care. These visits represent an opportunity for more robust re-engagement efforts. A public health information exchange (HIE) leverages data traditionally collected only for surveillance to impact direct patient care. Applied to HIV care, HIEs can assist health care providers and social workers in identifying PWH who are out of care and accessing the health care system for unrelated reasons [13, 14]. The Louisiana Public Health Information Exchange (LaPHIE) relied on direct provider notification that their patient was living with HIV and out of care. Program analysis found that 85% of patients generating an alert had at least 1 subsequent CD4 or HIV-1 RNA, a surrogate for a high level of linkage to care [15]. In view of this model, the Grady Health System (GHS) implemented a tailored version of the HIE model in partnership with the Georgia Department of Public Health (GDPH) [16, 17]. The Georgia Public Health Information Exchange (GPHIE) utilizes GHS social workers to provide linkage resources and motivation to out-of-care patients who present to the GHS ED. Previous studies have demonstrated that brief case management interventions improve linkage to care, whereas enhanced personal contact improves patient retention once engaged in care [18, 19]. However, no studies have combined HIE-based notification with a social worker–led intervention, and no studies have looked into HIEs implemented exclusively in the ED environment. We report the outcomes from an HIE re-linkage intervention in a large urban health system in the Southern United States.
METHODS
Study Design
Health Information Exchange Intervention
The GPHIE is a bidirectional health information–sharing platform between the GDPH and the GHS that utilizes data traditionally collected for surveillance in the delivery of patient care. All CD4 and HIV-1 RNA results for Georgia residents are reportable by laboratories to GDPH and uploaded to the Georgia HIV Surveillance database. An out-of-care watch list that includes all persons diagnosed with HIV and reported to GDPH who had no CD4 or HIV-1 RNA tests during the last 14 months is updated regularly and matched against patient registrations in GHS in real time. The 14-month interval was selected to minimize the number of patients falsely identified as being out of care due to delays in data reporting, maximizing the specificity of the alert. An HIE alert is automatically generated when a match occurs between a patient registration message and the out-of-care watch list. Based on Georgia Law, the GPHIE alert must be sent to a physician provider. At GHS, the alert is sent to 2 clinicians, who then notify 1 of 2 ED social workers (SWs). These SWs are specifically trained in motivational interviewing and have significant experience working with PWH. The SWs approach the patient, explain the GPHIE system, and ascertain if the patient is engaged in HIV care. If not, the SW educates the patient, assesses barriers to care, and offers re-linkage support, either to the prior site of care or the Grady HIV clinic, known as the Infectious Disease Program (IDP). Patients are able to walk into the IDP to initiate enrollment and see a prescribing physician within 72 hours without an appointment. The SW then facilitates a warm handoff to the IDP health educator or another clinic. Depending on patient preferences, the SW will follow up with reminder calls. SWs only respond to HIE alerts Monday through Friday, 9:00 am to 5:00 pm. Patients who generate alerts overnight or on weekends do not receive the SW intervention and cannot be initially contacted after they leave the ED for institutional restrictions. However if the electronic medical record (EMR) indicates the patient was previously seen at IDP, the SW will notify the IDP Health Educator who has jurisdiction to then pursue re-linkage interventions. This IDP health educator intervention was not standardized but generally involved calling the patient, discussing their recent ED visit and reasons for being out of care, encouraging them to return to clinic and attempting to solve any barriers such as transportation or paperwork. Importantly this intervention differed from the HIE SW intervention in that it was not in person and did not occur at the time of contact with the health care system.
Setting
The study was conducted within GHS in Atlanta, Georgia. Patients newly diagnosed with HIV at GHS are 77% male, 72% black/non-Hispanic, 77% uninsured, and 66% have AIDS at the time of diagnosis [20]. The majority of ambulatory HIV care within the GHS is provided at IDP, a Ryan White HIV/AIDS Program–funded clinic. The IDP is the largest HIV care provider in Georgia with over 6200 active patients.
Patients
All patients ≥18 years of age who generated an HIE alert between January 1, 2017, and January 31, 2018, were assessed for study inclusion. Patients who self-reported as currently in care were excluded as they did not represent the study population of out-of-care patients. Patients who were admitted to the hospital were excluded from the primary analysis because these patients received an intensive inpatient SW linkage intervention during admission and upon discharge regardless of the HIE alert, obscuring the effect of the relatively brief ED SW contact.
Data Collection and Definitions
Patient demographics, sociobehavioral characteristics, HIV diagnosis and treatment history, hospitalizations, clinic visits, and laboratory data (CD4 and HIV-1 RNA) were collected from the GHS electronic health record through July 31, 2018. The clinical encounters, notes, and labs from the prior 3 years were individually reviewed, and the EMR search feature was used to search for key terms such as “homeless,” “unstable housing,” “shelter,” “substance use,” and “employment.” In addition, diagnosis date, pre-alert HIV-1 RNA and CD4 values, and post-alert HIV-1 RNA and CD4 values were collected from the GDPH surveillance database accounting for patients who received treatment or linked to care outside GHS.
Unstable housing was defined as any indication in the EMR of homelessness or living in a shelter within the last 2 years based on provider or SW notes. Substance use was defined as any documented use of cocaine, amphetamines, or opiates in the notes or on urine drug screens or alcohol use >2 drinks per day for men or >1 drink per for women in the last 2 years in the provider notes. Unemployment was defined as any documentation of unemployment within the Grady EMR in the last 2 years. Previous HIV care was defined as >1 visit with an HIV provider at GHS at any point before HIE alert generation.
The control group was defined as the group of patients who generated an HIE alert but who were not contacted by the HIE SW in the ED or admitted to the hospital. Linkage to care was defined as any visit with an HIV provider within GHS and/or any CD4 or HIV RNA result within the GDPH database during the 6-month follow-up period. GHS linkage, specifically, was defined as a known visit with a provider at IDP. Viral suppression was defined as achieving an HIV RNA <200 copies/mL during the 6-month follow-up window in the GHS or GDPH database.
Statistical Analysis
Descriptive statistics were used to characterize the out-of-care and not admitted population overall and by SW intervention status (intervention, control). Distributions of continuous variables by SW intervention status were compared by t test or Wilcoxon rank-sum test as appropriate. Distributions of categorical variables by SW intervention status were compared by chi-square test or Fisher exact test as appropriate. Separate logistic regression models assessed achievement of the primary outcomes (6-month linkage to care and viral suppression) by SW intervention status, adjusting for age, gender, race/ethnicity, homelessness, substance use, and history of antiretroviral treatment. Model fit was assessed with Hosmer-Lemeshow and c-statistic. A secondary analysis among all out-of-care patients evaluated the linear association of degree of SW intervention with each of the following outcomes: linkage to care in 6 months, viral suppression in 3 months, and viral suppression in 6 months. A decision was made a priori for the primary analyses to exclude hospitalized patients. A post hoc secondary analysis included the hospitalized population in a dose–response (SW) evaluation. The Emory Institutional Review Board (IRB 00087151), GDPH Institutional Review Board, and Grady Research Oversight Committee approved the study.
RESULTS
Patient Characteristics
Between 1/1/17 and 1/31/18, 166 patients generated an HIE alert in the GHS ED. Eighteen patients (11%) self-reported being engaged in care. One patient was known to be HIV negative and thus represented an erroneous alert. Thus, 147 out-of-care PWH generated an alert; these individuals represent our study population. Forty-four (27%) patients were admitted from the ED and received an intensive inpatient SW intervention, and 3 (2%) had a subsequent hospital admission within the 6-month follow-up period. The remaining 98 patients were included in the primary analysis (Figure 1). Of these 98 patients, 20 (20%) received the ED SW re-linkage intervention. The remaining 78 patients represent the control group of patients who generated an alert but did not interact with an ED SW. Of these 78 patients, 18 had their information sent to a Health Educator at IDP to attempt re-linkage.
Figure 1.
Exclusion criteria. Abbreviation: SW, social worker.
Table 1 describes the population characteristics overall and by SW intervention. The majority of patients were male (70%) and African American (93%). Thirty percent were unstably housed, 58% unemployed, and 54% uninsured. Twenty-nine percent of patients had been incarcerated within the last year, and 37% had a history of substance use. Twenty-five percent had a psychiatric diagnosis documented within the EMR. Thirty-five patients (35%) had subsequent ER visits within 6 months. There was no difference in these variables between the intervention and control groups.
Table 1.
Patient Demographics (n = 98)
| Total (n = 98) | SW Intervention | P Value | ||
|---|---|---|---|---|
| Characteristic | No (n = 78) | Yes (n = 20) | ||
| Basic demographic | ||||
| Age, mean ± SD, y | 41.0 ± 12.3 | 40.8 ± 12.4 | 41.9 ± 12.2 | .7185 |
| Birth sex, male, No. (%) | 69 (70) | 55 (71) | 14 (70) | .9642 |
| Race/ethnicity, No. (%) | ||||
| African American | 91 (93) | 72 (92) | 19 (95) | 1.0000 |
| Non–African American | 6 (6) | 5 (6) | 1 (5) | |
| Socioeconomic education, No. (%) | ||||
| Less than HS/GED | 5 (5) | 4 (5) | 1 (5) | 1.0000 |
| HS/GED or higher | 15 (15) | 13 (17) | 2 (10) | |
| Unstably housed, No. (%) | 29 (30) | 23 (29) | 6 (30) | .9818 |
| Unemployed, No. (%) | 57 (58) | 48 (62) | 9 (45) | .1413 |
| Substance use, No. (%) | 36 (37) | 28 (36) | 8 (40) | .6723 |
| Payer source, No. (%) | ||||
| Private | 8 (8) | 7 (9) | 1 (5) | .8272 |
| Medicaid/Medicare | 36 (37) | 28 (36) | 8 (40) | |
| None | 53 (54) | 42 (54) | 11 (55) | |
Abbreviations: GED, general education development; HS, high school; SW, social worker.
As outlined in Table 2, the average time since diagnosis with HIV was 9.1 ± 6.7 years. Thirty percent of patients had no prior documentation of their HIV status within the GHS EMR, whereas 31% were previously engaged in care within the GHS. Fifty-one percent of patients were engaged in HIV treatment within the GHS in the past, and 31% achieved viral suppression before disengaging from care, as documented in the GHS electronic health record (EHR) or GDPH records.
Table 2.
HIV Infection History
| Total (n = 98) | SW Intervention | P Value | ||
|---|---|---|---|---|
| Characteristic | No (n = 78) | Yes (n = 20) | ||
| Diagnosis | ||||
| Years since HIV diagnosis, mean ± SD | 9.1 ± 6.7 | 8.9 ± 7.0 | 10.1 ± 5.5 | .4674 |
| HIV status | ||||
| Documentation of HIV status in GHS chart before alert, No. (%) | 69 (70) | 53 (68) | 16 (80) | .2922 |
| Treatment history | ||||
| Ever in care within GHS, No. (%) | 50 (51) | 36 (46) | 14 (70) | .1353 |
| Ever virologically suppressed before alert, No. (%) | 30 (31) | 23 (30) | 7 (35) | .6332 |
| CD4 nadir in the 2 y before alert, median (IQR), cells/µL | 185 (325) | 232 (334) | 116 (254) | .1756 |
| CD4 nadir <200 cells/µL, No. (%) | 36 (37) | 24 (31) | 12 (60) | .1066 |
| Condition before alert | ||||
| Last HIV-1 RNA before alert, median (IQR), copies/mL | 89.1 (4571) | 303.7 (4231) | 39.8 (58 884) | .8293 |
| Last HIV-1 RNA before alert, median (IQR), log10 copies/mL | 2.0 (3.7) | 2.7 (3.7) | 1.6 (4.8) | .8020 |
| Last CD4 before alert, mean (SD), cells/µL | 395.8 ± 236.6 | 408.1 ± 231.3 | 347.5 ± 262.3 | .4533 |
| Years between last HIV-1 RNA/CD4 and alert, median (IQR) | 2.3 (1.7) | 2.3 (2.0) | 1.7 (1.2) | .1560 |
| Condition at alert | ||||
| HIV-1 RNA, median (IQR), copies/mL | 23 947.5 (87 313) | 89 555.4 (222 348) | 5624.1 (17 235) | .0927 |
| HIV-1 RNA, median (IQR), log10 copies/mL | 4.4 (1.4) | 4.9 (0.9) | 3.7 (1.0) | .0927 |
| CD4, median (IQR), cells/µL | 257.0 (332.0) | 82.0 (335.0) | 294.0 (177.0) | .1394 |
Data are from GHS and GDPH unless specifically stated to be only from GHS.
Abbreviations: GDPH, Georgia Department of Public Health; GHS, Grady Health System; IQR, interquartile range; SW, social worker.
Outcome Data
Of 98 patients generating an alert, 20 received the SW intervention. Of those 20 patients, 12 (60%) linked to care within 6 months (5 [25%] within 30 days) and 4 (20%) achieved viral suppression, whereas of those who did not receive the SW intervention, only 27 (35%) linked to care (8 [10%] within 30 days) and 10 (13%) achieved viral suppression.
After controlling for age, gender, race, drug use, homelessness, and prior HIV treatment, patients who received the SW intervention had an adjusted risk ratio for linkage to care within 6 months of 1.63 (95% CI, 0.99–2.68) and an adjusted risk ratio for viral suppression within 6 months of 1.49 (95% CI, 0.50–4.46) compared with those who did not receive the intervention (Table 3).
Table 3.
Adjusted Relative Risks of Successful Linkage to HIV Care Within 6 Months and Viral Suppression Within 6 Months by SW Intervention and Selected Characteristics
| Linkage to Care | Viral Suppression | |||
|---|---|---|---|---|
| Adjusted Relative Risk (95% Confidence Interval) | P Value | Adjusted Relative Risk (95% Confidence Interval) | P Value | |
| HIE SW contact | ||||
| Yes | 1.63 (0.99–2.68) | .0545 | 1.49 (0.50–4.46) | .4719 |
| No | Ref | |||
| Age, y | ||||
| 18–25 | 1.39 (0.58– 3.34) | .4598 | 1.38 (0.30–6.38) | .6833 |
| 26–39 | 1.03 (0.56–1.88) | .9213 | 0.55 (0.17–1.78) | .3212 |
| ≥40 | Ref | Ref | ||
| Gender | ||||
| Male | 1.00 (0.56–1.78) | .9906 | 0.65 (0.23– 1.87) | .4256 |
| Female | Ref | Ref | ||
| Race/ethnicity | ||||
| Non–African American | 0.87 (0.27–2.77) | .809 | NI | |
| African American | Ref | |||
| Drug use | ||||
| Yes | 0.85 (0.45–1.62) | .6187 | 0.97 (0.34–2.75) | .9498 |
| No | Ref | Ref | ||
| Homeless | ||||
| Yes | 1.65 (0.91–2.96) | .0966 | 0.62 (0.20–1.98) | .4228 |
| No | Ref | Ref | ||
| History of HIV treatment | ||||
| Yes | 1.42 (0.78–2.57) | .2491 | 0.93 (0.32–2.73) | .9023 |
| No | Ref | Ref | ||
Abbreviations: HIE, health information exchange; NI, not included; SW, social worker.
Of the 98 patients generating an alert, 40% linked to care and 14% achieved viral suppression within 6 months. When stratified by degree of intervention, 67% of admitted patients, 60% of those who received the inpatient SW intervention, 50% of those whose information was sent to IDP, and 30% of patients receiving no intervention were linked to care within 6 months of alert generation. Likewise, 39% of admitted patients, 20% of patients receiving the SW intervention, 11% of patients whose information was sent to IDP, and 13% of patients receiving no intervention achieved viral suppression within 6 months (Figure 2).
Figure 2.
Rates of linkage and viral suppression by degree of intervention. Abbreviations: HIE, health information exchange; IDP, Infectious Disease Program; SW, social worker.
DISCUSSION
Over the course of 13 months, 166 HIE alerts were generated at registration in an urban ED, and 147 of those alerts represented out-of-care PWH. Twenty of these individuals were approached in the ED and received a tailored linkage intervention by a social worker. After multivariate analyses, patients who received the SW intervention were 63% more likely to be linked to HIV care within 6 months of visiting the ED. Although bordering on statistical significance at the P < .05 cutoff, the sample is small, and the potential effect size of this finding warrants further investigation in appropriately powered studies. The GHS-GPDH HIE represents a novel intervention that successfully identified and linked out-of-care individuals living with HIV to care in the urban Southern United States. Our program is the first to report that HIEs can successfully increase linkage to care when implemented in combination with an SW-led intervention in an urban ED. This differs from the LaPHIE program, where alerts were sent directly to the treating clinician, which was not feasible for technical reasons in the GPHIE [11–13].
This HIE-informed intervention represents a novel strategy to re-link PWH to care. Traditionally, linkage interventions have targeted patients identified by an HIV test. Among newly diagnosed patients, 78% link within 1 month [4]. Systematic attempts at linkage for those who have disengaged from care or do not link initially are often lacking. Although PLWH may have frequent interaction with health care systems, without a mechanism in place to identify those with an HIV diagnosis, acute care clinicians often do not know the HIV status of the patient unless testing is offered or the patient volunteers the information. In the population studied at GHS, 30% of patients had no documentation of HIV status in the EMR and an additional 31% were previously engaged in care at GHS, but even prior care in the same system can be missed in the busy acute care setting. By sharing information traditionally collected for surveillance with health care providers, the HIE intervention successfully targets a population that may be overlooked by traditional linkage or re-linkage interventions.
It is important to note that linkage-to-care interventions exist on a spectrum. Although we separated patients who received the SW intervention from those who did not, the patients could be further broken down into 4 categories representing progressively increased linkage support. The first group received no intervention, the second group had their information sent to a patient navigator at IDP (outpatient HIV clinic) by the ED SW for intervention, a third group received the intervention from the ED SW, and a fourth group was admitted to the hospital and received intensive, multiday inpatient SW support, including working with a linkage coordinator affiliated with the outpatient HIV clinic. Although the groups were too small for analysis, there was a trend suggesting increased linkage to care and viral suppression as the intensity of the intervention increased. Of note, 35% of patients generating an HIE alert were admitted to the hospital, compared with an average admission rate of 22% at Grady in 2017. This higher rate of hospital admission likely reflects the clinical acuity of patients generating HIE alerts, and this too has been associated in the past with linkage to care [21]. As a whole, this trend suggests that more intensive linkage interventions may lead to improved linkage outcomes.
Disparities in linkage, retention, and viral suppression outcomes persist for African Americans and individuals with a history of unstable housing or substance use [6, 22–24]. PWH who are experiencing homelessness or unstable housing are known to access acute care settings such as EDs at a higher rate [25]. Thus, the population identified through HIE is a group in need of targeted interventions to improve care continuum outcomes. Specific interventions targeting these populations are particularly important to improving HIV outcomes as a whole [26]. By identifying these patients for linkage interventions, the HIE provides a platform on which to implement targeted interventions for linkage to HIV care or other social services.
Models such as the LaPHIE, which directly inform providers through the EMR, offer logistical benefits; however, some providers may feel uncomfortable discussing engagement with HIV treatment or may defer the discussion due to competing clinical responsibilities. Models such as the GPHIE, which rely on SWs trained in HIV disclosure and counseling, may offer a more effective intervention; however, they are limited by SW availability and by relying on the physician conduit. These barriers could be overcome by directly routing alerts to SWs and by increasing funding for SW support.
Strengths and Limitations
The strengths of this study include evaluation of a novel intervention and its direct comparison with a similar group of out-of-care patients, all of whom generated an HIE alert. Additionally, the design accounts for linkage outside of the GHS by matching of databases with GPDH, and indeed, 70% of patients who linked to care within 6 months linked outside of the GHS.
The primary limitation of this study is the small sample size. This is especially evident with the wide confidence interval for viral suppression given few occurrences. Additionally, the retrospective, nonrandomized nature of the study only allows for conclusions around association. Because SWs were only available during business hours, it is possible that unmeasured variables affected both whether a patient received the SW intervention and whether they linked to care. Outcome data were further affected by the definition of linkage, which included patients with any CD4 HIV-1 RNA labs in the DPH system. It is possible that patients had HIV labs drawn outside of an HIV clinic visit (eg, emergency room or an outside hospital admission), and these labs therefore did not represent linkage to outpatient HIV care. Finally, patients who had their clinical information sent to a health educator at IDP were included in the control group because they did not have a face-to-face discussion with the HIE SW. Therefore, any increase in linkage that these patients experienced would have served to dampen the effect size of the intervention. The demographic data may underestimate certain factors such as homelessness and substance use because many patients only visited the ED and never linked to care within the GHS, and thus some charts did not include complete sociodemographic information.
There were many challenges to implementation of the HIE in our clinical context. Because the HIE was initially implemented across multiple health care systems with different EMRs, the alerts were not sent directly through the EMR, differentiating the GPHIE from LaPHIE. Additionally, Georgia law requires health information to be sent to health care providers, preventing alerts from being sent to SWs directly and thus delaying interventions. SWs were only available during normal business hours, and private face-to-face meetings were difficult to arrange around patient care in a busy ED, resulting in only 20% of alerts being acted upon. Finally, institutional rules restrict SWs from contacting patients after leaving the ED if the SW did not meet the patient in person, limiting the opportunity to engage the 80% of patients not contacted before leaving the ED.
CONCLUSIONS
In conclusion, HIEs successfully identify out-of-care PWH and provide an opportunity for real-time, (re)linkage to care interventions and programs. These results further suggest that an HIE-informed, SW-driven intervention may improve outcomes for out of-care patients who receive linkage support in the ED. Further research with controlled trials is needed to directly assess the impact of an HIE intervention, while implementation science will be critical to ensure uptake of successful interventions built around the HIE model.
Acknowledgments
The authors would like to thank the patients who participated in the intervention. The authors would also like to thank the staff of the Grady ED for facilitating conversation between the SW and the patients.
Financial support. This work was supported by the Center for AIDS Research (CFAR) at National Institutes of Health via Emory University (P30AI050409) through a CFAR data award to support statistical analysis of the study.
Potential conflicts of interest. All authors: no reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Data availability. Data are not publicly available.
Author contributions. J.S. was responsible for conducting the chart review, building and maintaining the database, and writing the manuscript. Statistical analysis was completed by J.S., C.A., and C.M. J.S., T.M., W.A., B.S., and J.C. contributed to the plan of the analysis and interpretation of the data. T.M., P.W., E.P., L.R., B.S., and J.C. developed and implemented the HIE program at Grady. All authors commented on drafts of the manuscript and approved the final version.
References
- 1. Samji H, Cescon A, Hogg RS, et al. ; North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) of IeDEA Closing the gap: increases in life expectancy among treated HIV-positive individuals in the United States and Canada. PLoS One 2013; 8:e81355. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Centers for Disease Control and Prevention. Monitoring selected national HIV prevention and care objectives by using HIV surveillance data—United States and 6 dependent areas, 2016. HIV Surveillance Supplemental Report.2018. Available at: http://www.cdc.gov/hiv/library/reports/ hiv-surveillance.html. Accessed 20 January 2019.
- 3. Skarbinski J, Rosenberg E, Paz-Bailey G, et al. Human immunodeficiency virus transmission at each step of the care continuum in the United States. JAMA Intern Med 2015; 175:588–96. [DOI] [PubMed] [Google Scholar]
- 4. Centers for Disease Control and Prevention. Understanding the HIV care continuum. 2019. Available at: https://www.cdc.gov/hiv/pdf/library/factsheets/cdc-hiv-care-continuum.pdf. Accessed 13 August 2019.
- 5. Cohen MS, Chen YQ, McCauley M, et al. ; HPTN 052 Study Team Antiretroviral therapy for the prevention of HIV-1 transmission. N Engl J Med 2016; 375:830–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Skarbinski J, Rosenberg E, Paz-Bailey G, et al. Human immunodeficiency virus transmission at each step of the care continuum in the United States. JAMA Intern Med 2015; 175:588–96. [DOI] [PubMed] [Google Scholar]
- 7. Colasanti J, Stahl N, Farber EW, et al. An exploratory study to assess individual and structural level barriers associated with poor retention and re-engagement in care among persons living with HIV/AIDS. J Acquir Immune Defic Syndr 2017; 74(Suppl 2):113–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Pollini RA, Blanco E, Crump C, Zúñiga ML. A community-based study of barriers to HIV care initiation. AIDS Patient Care STDS 2011; 25:601–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Centers for Disease Control and Prevention. HIV surveillance report, 2017 Vol. 29 2018. Avaliable at: http://www.cdc.gov/hiv/library/reports/hiv-surveillance.html. Accessed 20 December 2018. [Google Scholar]
- 10. Meditz AL, MaWhinney S, Allshouse A, et al. Sex, and race, and geographic region influence clinical outcomes following primary HIV-1 infection. J Infect Dis 2010; 203:442–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Philbin MM, Feaster DJ, Gooden L, et al. The North-South divide: substance use risk, care engagement, and viral suppression among hospitalized human immunodeficiency virus-infected patients in 11 US cities. Clin Infect Dis 2019; 68:146–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Mohareb AM, Rothman RE, Hsieh YH. Emergency department (ED) utilization by HIV-infected ED patients in the United States in 2009 and 2010 - a national estimation. HIV Med 2013; 14:605–613. [DOI] [PubMed] [Google Scholar]
- 13. Shade SB, Chakravarty D, Koester KA, et al. Health information exchange interventions can enhance quality and continuity of HIV care. Int J Med Inform 2012; 81:e1–9. [DOI] [PubMed] [Google Scholar]
- 14. Herwehe J, Wilbright W, Abrams A, et al. Implementation of an innovative, integrated electronic medical record (EMR) and public health information exchange for HIV/AIDS. J Am Med Inform Assoc 2012; 19:448–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Magnus M, Herwehe J, Gruber D, et al. Improved HIV-related outcomes associated with implementation of a novel public health information exchange. Int J Med Inform 2012; 81:e30–8. [DOI] [PubMed] [Google Scholar]
- 16. Georgia Department of Public Health. Health information exchange (HIE). Available at: https://dph.georgia.gov/hiv-health-information-exchange-hie. Accessed 10 November 2018.
- 17. Bresnick J. Georgia HIE grows with addition of Grady, Emory Health. EHR Intelligence. Available at: https://ehrintelligence.com/news/georgia-hie-grows-with-addition-of-grady-emory-health/. Accessed 10 November 2017.
- 18. Gardner LI, Metsch LR, Anderson-Mahoney P, et al. ; Antiretroviral Treatment and Access Study Study Group Efficacy of a brief case management intervention to link recently diagnosed HIV-infected persons to care. AIDS 2005; 19:423–31. [DOI] [PubMed] [Google Scholar]
- 19. Gardner LI, Giordano TP, Marks G, et al. ; Retention in Care Study Group Enhanced personal contact with HIV patients improves retention in primary care: a randomized trial in 6 US HIV clinics. Clin Infect Dis 2014; 59:725–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Colasanti J, Goswami ND, Khoubian JJ, et al. The perilous road from HIV diagnosis in the hospital to viral suppression in the outpatient clinic. AIDS Res Hum Retroviruses 2016; 32:729–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Croxford S, Burns F, Copas A, et al. ; OptTEST by HIV in Europe Factors associated with delayed linkage to care following HIV diagnosis in the WHO European region. HIV Med 2018; 19(Suppl 1):40–6. [DOI] [PubMed] [Google Scholar]
- 22. Eaton LA, Driffin DD, Kegler C, et al. The role of stigma and medical mistrust in the routine health care engagement of black men who have sex with men. Am J Public Health 2015; 105:e75–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Magnus M, Jones K, Phillips G 2nd, et al. ; YMSM of color Special Projects of National Significance Initiative Study Group Characteristics associated with retention among African American and Latino adolescent HIV-positive men: results from the outreach, care, and prevention to engage HIV-seropositive young MSM of color special project of national significance initiative. J Acquir Immune Defic Syndr 2010; 53:529–36. [DOI] [PubMed] [Google Scholar]
- 24. Taylor BS, Fornos L, Tarbutton J, et al. Improving HIV care engagement in the south from the patient and provider perspective: the role of stigma, social support, and shared decision-making. AIDS Pt Care and STDs 2018; 32(9):368–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Clemenzi-Allen A, Neuhaus J, Geng E, et al. Housing instability results in increased acute care utilization in an urban HIV clinic cohort. Open Forum Infect Dis 2019;XXX(X):XXX–XX. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Fauci AS, Eisinger RW. PEPFAR - 15 years and counting the lives saved. N Engl J Med 2018; 378:314–6. [DOI] [PubMed] [Google Scholar]