Evaluation of patient care cascade for HIV-positive patients diagnosed in La Romana,Dominican Republic in 2011: a retrospective cohort study

Abstract

The Caribbean has the highest adult HIV prevalence in the world after sub-Saharan Africa (2011). One sub-population in the Dominican Republic is the migratory Batey community primarily comprised of Haitian immigrants with limited access to healthcare and HIV prevalence rates of between 3.0% and 9.0%, compared to 0.7% nationally. This retrospective cohort study describes the cumulative retention from diagnosis to virological suppression for newly-diagnosed HIV-infected adults presenting to the Clínica de Familia and its Batey programme in La Romana, during 2011. Of the patients diagnosed with HIV, 65% entered into care, 59% completed immunologic testing, 53% were eligible for antiretroviral therapy (ART) initiation, 36% initiated ART within three months of eligibility and 27% were retained in care. Seventeen per cent of those retained on ART with a 12-month viral load measure had undetectable viral load. Attrition primarily occurred before ART initiation. The Batey programme had a first step lost-to-follow-up of 88% compared to 20% at the clinic (p < 0.001). This retrospective study details the continuum of care and indicates where structural changes must occur to increase continuity between steps. The manuscript results are important to help implement programmes to enhance engagement and retention in HIV primary care.

Keywords

Caribbean treatment antiretroviral therapy care cascade HAART HIV AIDS epidemiology

Introduction

The Caribbean has one of the highest adult HIV prevalence rates in the world after sub-Saharan Africa, at about 1% (0.9%–1.1%) in 2011.¹ Approximately 250,000 persons are living with HIV/AIDS in the Caribbean, with the majority residing on Hispaniola, the island shared by the upper-middle and low-income countries: Dominican Republic (DR) and Haiti.^2,3 The adult prevalence rate of HIV in the DR is 0.7%.¹ The main mode of infection is heterosexual transmission. High tourism regions, such as the province of La Romana, are the areas most afflicted by the epidemic, where commercial sex work is common.⁴ Another high-risk population is the migratory Batey community, where sugar cane cultivation occurs. The Batey community is primarily comprised of Haitian immigrants with limited access to education and healthcare services, with HIV prevalence rates ranging from 3.0% to 9.0%.^5–8

While improvements have been made in response to the HIV epidemic, 90% of HIV-related deaths occur in resource-poor settings.^9,10 A need for improvements regarding access to antiretroviral therapy (ART), adherence to ART, retention in ART programmes and management of HIV-associated illnesses and complications is of the utmost importance.^11,12 With recent evidence identifying HIV treatment as an important HIV transmission prevention intervention, reducing attrition in the cascade of care is paramount.¹³ The HIV care cascade incorporates HIV testing, linkage to and retention in care, initiation of ART, and adherence to ART, with the ultimate goal of virological suppression.^12,14 To increase virological suppression, improvements in every step of the care cascade must be achieved simultaneously.¹⁵ Currently, a gap still exists regarding knowledge of the cascade of care in the DR.

Clínica de Familia La Romana (CFLR) is a large clinic located in the Eastern Dominican province of La Romana that collaborates with the Columbia University IFAP Global Health Program to provide free, comprehensive and HIV-specialised care.¹⁶ This clinic is the second largest in the nation, providing antiretroviral therapy to its HIV-positive patients. The clinic also has a Batey programme for HIV counselling and testing, to identify HIV among the Batey residents and to help link them to HIV care at the clinic. This clinic incorporates a multidisciplinary approach to family-centred HIV care and has been looked to as a model in the DR for comprehensive HIV prevention, care and treatment programmes.¹⁷ The purpose of this study is to describe the HIV care continuum at CLFR and to compare differences in achievement of various steps in the continuum of care between those enrolling at the clinic and Batey programme, and between those of Haitian vs. Dominican nationality. This study is the first in the region to describe each stage of the continuum of care for HIV-positive adult patients and to determine where the greatest amount of attrition occurs in this progression. Specifically, this analysis will serve as a baseline for proposed programmatic changes at CFLR in order to improve identified gaps in the HIV care cascade at this location.

Materials and methods

Setting

This study was conducted in La Romana, DR, located in the southeastern part of the island of Hispaniola. La Romana has a high prevalence of HIV, driven primarily by heterosexual transmission.⁴ HIV testing is free of cost at CFLR, along with medications for those eligible to start ART as described in 2009 World Health Organization (WHO) guidelines.^16,18 To be eligible, the patient must have had at least one of the following: CD4 cell count below 350 cells/mm,³ WHO stage III or IV, an opportunistic infection or pregnancy. HIV-infected patients enrolled at CFLR completed two adherence consultations, a week apart after ART eligibility was determined and before starting ART. After ART initiation, patients were scheduled for monthly appointments to monitor health status.

Data collection

This is a retrospective cohort study of newly-diagnosed HIV-infected individuals presenting to CFLR or its Batey programme from January 2011 to December 2011. We included all newly-diagnosed HIV-positive patients 18 years or older at the time of diagnosis. Patients from the clinic and the Batey programme completed voluntary HIV antibody testing as a result of the personal decision to undergo testing. Study population information, including basic demographics and ART regimen, were collected from a chart review from the records of HIV-positive patients in 2013 to allow for all patients diagnosed in 2011 to be in care for at least one year. The chart review was conducted by a team of doctors from the clinic and later transferred to an electronic database by the principal investigator. Lab results collected included baseline CD4 cell counts and HIV viral load, along with CD4 cell count and viral load at six and 12 months after ART initiation, when available. The Institutional Review Boards at Columbia University (IRB-AAAM3250) as well as the Instituto Dermatológico de Cirugía y Piel in the DR (RNC: 430-10256-5) approved this study.

Study population

The study population included all adults who underwent HIV testing at CFLR or as a part of the clinic’s Batey programme between January and December 2011. Twelve patients testing HIV positive at the CFLR were excluded from this study because they were less than 18 years of age at the time of their HIV test.

Cascade definitions

Step one in the cascade was receipt of a positive HIV test. Step two in the cascade was enrollment in clinic services, measured through proof of receipt of a record number and HIV care charts integrated into the clinic’s system. Step three was documentation of at least one recorded laboratory result (CD4 cell count and/or viral load). Step four was determination of eligibility for ART. Initiation of ART was step five and was measured as those initiating ART within three months of eligibility. Step six included identifying those who were retained on ART. Retention was determined through the number of patients initiating ART who came to the clinic at least once in the past three months. Step seven was identifying patients with viral load suppression, defined as less than 20 copies per mL six and/or at 12 months after the initiation of treatment.

Statistical analysis

Descriptive statistics were used to describe baseline demographic factors, baseline CD4 cell counts and HIV viral load results in groups of interest. The proportion of all patients testing HIV-positive (the cumulative proportion) who achieved each step was reported, as well as the proportion of patients who met the prior step who also met the current step (conditional proportion). HIV care cascades are presented for the entire population, as well as separately stratified by point of HIV test (clinic vs. Batey) and nationality (Dominican vs. Haitian). Assessments of difference in proportions (both cumulative and conditional) between these strata were compared using Chi-square tests and Fisher’s exact tests at a level of significance of 0.05. Statistical analysis was performed using SAS 9.3 (SAS Institute Inc., Cary, NC, USA).

Results

Figure 1 presents the HIV care cascade at CFLR. The total number of persons tested for HIV in 2011 in the CFLR or its Batey programme was 7489, with 298 positive confirmations (4%). Of the 298 HIV-positive patients included in this study, 193 (65%) were enrolled in clinic services and 176 (59%) completed either CD4 or viral load testing. Of those who tested HIV-positive, 158 (53%) patients were eligible for ART according to the 2009 WHO guidelines and 107 (36%) initiated ART within three months of eligibility. The median time to initiate ART after diagnosis was 33 days (IQR: 15, 123). Seventy-nine (27%) were recorded as still on treatment at the time of the chart review. Of these 79, 70 (89%) had a recorded viral load result 12 months after ART initiation; 12 (17%) of these had an undetectable viral load (4% of those testing HIV-positive). When stratified by point of entry (Figure 1), either diagnosed in the clinic or through the Batey programme, notable differences occur in the first step with 80% of patients diagnosed in the clinic enrolling in clinic services, compared to only 22% of those diagnosed in the Batey programme (p < 0.01). Conditional percentages in subsequent steps decrease at similar rates for all remaining steps, with the least number of patients completing step five from step four (69% and 50%).

Figure 1.

HIV care cascade: All patients, by point of entry.

Demographic information shown in Table 1 was obtained during enrollment into HIV care (step 2) and showed that the majority of patients were female and Dominican. During 2011, there were nine pregnancies of newly diagnosed HIV-positive women, 18 cases of tuberculosis, eight cases of hepatitis B and one case of Kaposi’s sarcoma. Among the 105 patients not enrolling in clinic services, nine died, eight transferred to another clinic to receive care and 88 did not have any follow-up information recorded. Among the 17 patients enrolling in clinic services but not completing CD4 or viral load testing, one died, one began ART because of pregnancy, one began ART because of a tuberculosis co-infection, five completed other testing but dropped out before beginning treatment and nine did not have any follow-up information recorded. Among the 18 patients completing CD4 or viral load testing but not documented as ART eligible, one died, and 17 did not have any follow-up information recorded. For the 176 patients with a recorded CD4-cell count and viral load at enrollment, the median CD4 was 272 (IQR: 117.5, 459.95) cells/mm³ and median viral load was 13,110 (IQR: 275, 127778) copies per mL. Among the 51 patients eligible for ART but not initiating ART within three months, 42 initiated later than 13 months, two transferred to another clinic and seven did not have any follow-up information recorded. Among the 28 patients initiating ART but no longer on treatment at the time of chart review, six died, six transferred to another clinic and 16 did not have any follow-up information recorded.

Table 1.

Patient characteristics recorded at time of enrolment.

	All, n (%)	Clinic, n (%)	Batey, n (%)	p	% Missing
Tested (n)	7489	3732	4117		NA
HIV diagnosis n (%)	298 (4)	221 (6)	77 (2)		NA
Adults enrolled into care, n (%)	193 (65)	176 (80)	17 (22)	<0.001	NA
Gender, n (% male)	81 (42)	73 (42)	8 (47)		6 (3)
Age 18–24, n (%)	28 (15)	27 (15)	1 (6)		22 (11)
Age 25–39, n (%)	79 (41)	71 (40)	8 (47)
Age 40–49, n (%)	40 (21)	35 (20)	5 (29)
Age 50+, n (%)	24 (12)	22 (13)	2 (12)
Nationality, n (% Dominican)	158 (82)	154 (88)	4 (24)		6 (3)
Nationality, n (% Haitian)	34 (18)	21 (12)	13 (76)
Relationship status, n (% in a relationship)	96 (50)	87 (49)	9 (53)		5 (3)
Tests done	176 (59)	162 (73)	14 (18)	0.177	NA
Baseline CD4 median (IQR) cells/mm³	272 (118, 460)	262 (105, 452)	350.95 (218, 626)		22 (11)
Baseline viral load median (IQR)	13,110 (275, 127,778)	15,222 (283, 130,166)	1986.5 (112, 4138)		78 (39)
Eligible for ART	158 (53)	146 (66)	12 (16)	0.639	NA
Initiated ART (in three months)	107 (36)	101 (46)	6 (8)	0.204	NA
Retained on ART	79 (27)	74 (33)	5 (6)	1	NA
Virological suppression after one year of treatment	12 (4)	11 (5)	1 (1)	<0.571	NA

ART, antiretroviral therapy; IQR, interquartile range; NA, not applicable.

Stratified by nationality, we found that 76% of patients enrolling in HIV care after diagnosis in the Batey programme were Haitian vs. only 12% among those at the clinic (p < 0.001). After enrollment, the percentage of Dominican and Haitian patients decreased by different increments for the different steps (Figure 2), with marginally more Haitians starting ART than Dominicans (62% compared to 54%, [p = 0.064]). Additionally, only 50% of Haitians and 39% of Dominicans who enrolled in care were recorded as still on treatment at the time of the chart review (p = 0.58).

Figure 2.

HIV care cascade: All patients, by nationality.

For patients diagnosed at the clinic who enrolled in care (Figure 3(a)), stratification by nationality showed large differences, with a higher percentage of Haitians initiating ART than Dominicans (76% vs. 55%, p = 0.064). For patients diagnosed through the Batey programme then enrolling in care (Figure 3(b)), differences by nationality are apparent in the later steps, with no Dominicans being retained in clinic services compared to 38% of Haitians.

Figure 3.

(a) HIV care cascade: patients diagnosed in the clinic, by nationality. (b) HIV care cascade: patients diagnosed in the Bateyes, by nationality. (c) HIV care cascade: Dominican patients, by point of entry. (d) HIV care cascade: Haitian patients, by point of entry.

Stratification of Dominican patients by point of entry (Figure 3(c)) shows that those diagnosed in the Batey programme and enrolling in care showed consistently less retention in each step than those diagnosed in the clinic. Similar results occurred for Haitian patients (Figure 3(d)).

Discussion

This study found that less than 30% of all patients diagnosed with HIV in 2011 at the CFLR or its Batey programme were retained on treatment into 2013. Major differences were observed between patient cohorts throughout the cascade based on entering through the clinic vs. the Batey programme. The largest gap in the HIV care cascade was observed in the first step, with fewer enrollments in clinic services from patients diagnosed in the Batey programme compared to the clinic. Substantial attrition occurred for both Dominican and Haitian patients from the Batey programme after being determined to be ART eligible but before treatment initiation, suggesting that this programme is not optimally ensuring that eligible patients are initiating ART in a timely manner. Only 70 of the 79 patients retained on ART after one year had a viral load test recorded, with only 12 (17%) having achieved an undetectable viral load (92% from clinic, 8% from Batey). This suggests that adherence to ART is suboptimal in this setting.

The HIV care cascade described here is similar to data from the United States and sub-Saharan Africa, thus supporting the generalisability of these results. In the US, about 50% of HIV patients are linked to care, compared to 65% of those in this cohort.¹⁹ A sub-Saharan African review estimated an attrition proportion of 22% by ART initiation of HIV-positive patients.^19,20 However, in the US of those that are linked to care, 75% received ART and 80% had an undetectable viral load.¹⁹ This is higher than the rates in the clinic’s cohort with 68% (107 out of 158) of those eligible initiating ART and 15% (12 out of 79) with an undetectable viral load after one year.

Failure to enrol in clinical services was extreme in the Batey programme, with only 22% of those testing HIV-positive enrolling. The Batey community is a migratory community with limited access to education and transportation, as well as high levels of stigma concerning HIV. Difficulties in maintaining retention in HIV care services among transient populations are well documented.^7,21–24 While the clinic provides counselling services on the importance of prompt enrollment in HIV care and offers to assist with transportation costs, our findings suggest that these services are not resulting in large-scale engagement in HIV care services. In contrast, the population testing positive at the clinic is less transitory than the population recruited through the Batey programme, which suggests that those seeking care at this clinic are more likely to reside permanently near the clinic, enabling easier engagement in care. It is not likely that the patients who did not enrol in clinic services enrolled at another location, thereby supporting the notion that this cascade is representative of the patient experience, as well as representing the clinic’s performance overall.

All patients enrolled in clinic services should have undergone CD4 or viral load testing. This was not the case in 2011 for two reasons. The first was a shortage of reagents for several months at the National Laboratory, the only location for viral load testing in the DR. This also contributed to the underrepresented proportion of patients who initiated treatment with viral load testing after one year of treatment (46%). However, 89% of those retained on treatment had a viral load test after one year. The second reason is that while the clinic has a CD4 analyser, it was not in service for a few months and therefore samples were sent to the National Laboratory, delaying the results. This reduced the amount of patients who completed testing and delayed the initiation of treatment for those eligible.

This analysis can be used to ensure that existing treatment programmes at the clinic are tailored to the needs of the city and that additional treatment efforts are targeted to the appropriate populations. Furthermore, as a principal HIV clinic in the nation, the results of this study will be useful at the national level to help provide a regional baseline for an expanded, national analysis of the country’s HIV care cascade. The results will also serve as a baseline for subsequent interventions necessary to boost adherence and increase viral suppression in the larger HIV care system in the DR. The clinic implemented new programmes in 2011 such as ART adherence counselling, the development of a quality improvement committee, and the hiring of employees to aid specifically in the Batey programme. Because these changes were implemented late in 2011, their impact is minimal for our study population, but are expected to have a large impact in future years. Additionally, the clinic’s quality improvement committee has already developed a strategy to improve each step of the cascade based on the results of this study. Future studies are planned to determine the effectiveness of these programmatic interventions, which focus heavily on retention once patients are diagnosed as positive, as well as on keeping patients adherent to scheduled appointments and their prophylactic treatment.

Strengths and limitations

This is the first study documenting steps along the HIV care cascade in the DR, and one of the only able to evaluate steps in the continuum of care from receipt of a positive HIV test through retention on ART and viral suppression. While other cascades in resource-limited settings have ended at ART initiation, or started at enrollment in HIV care, our analysis provides an assessment of the complete continuum of care.^15,20

This study has important limitations. First, demographic information such as country of origin was not collected until the time of enrollment in HIV care, prohibiting investigation into differences in retention between receipt of HIV diagnosis and enrollment into care between Haitian and Dominican patients. Second, information was based on chart review and no independent assessment of patient outcomes was conducted. Third, this analysis was done in a single clinic representing a unique population in the DR, so caution is warranted when generalising results to different populations. Finally, the relatively small sample size limited power to conduct additional sub-group analyses. While these limitations are critical to consider in the dissemination of these results, this study will provide the clinic with a baseline against which to compare the effectiveness of planned interventions to reduce attrition from the HIV care cascade. Additionally, these results can be disseminated and expanded upon to provide a starting point from a major clinic for national HIV care strategies in the DR.

Conclusion

As the first descriptive study of the HIV care cascade in the DR, this investigation identified barriers to access in HIV care. This study found that only 65% of HIV-positive patients diagnosed at the clinic or the Batey programme were enrolled into clinic services, only 36% started ART within the first three months of eligibility, and only 4% had a suppressed viral load after one year of treatment. Key sources of discontinuation in the HIV care cascade occurred between HIV testing and enrollment in care, particularly for the Batey programme, and between ART eligibility and initiation. These results will act as a baseline to compare the effectiveness of planned programmatic interventions, to improve each step of the care cascade at the clinic. While more research must be conducted to ensure reliability and generalisability of these results, they will act as a strong starting point to help form national guidelines for monitoring and evaluation of programmes that promote access to HIV care.

Footnotes

Acknowledgements

The authors would like to thank Columbia University IFAP Global Health Program as well as the Columbia University Mailman School of Public Health and the Clínica de Familia La Romana.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

UNAIDS. Regional fact sheet 2012: Latin America and the Caribbean, www.UNAIDS.org (2012, accessed 10 April 2015).

Rojas

Malow

Ruffin

Rothe

et al.

The HIV/AIDS epidemic in the Dominican Republic: key contributing factors. J Int Assoc Physicians AIDS Care (Chic) 2011; 10: 306–335.

The World Bank. Country and lending groups. Available at: data.worldbank.org/about/country-and-lending-groups (2013, accessed 10 April 2015).

Teva

Bermudez

Ramiro

et al.

Situación epidemiológica actual del VIH/SIDA en Latinoamérica en la primera década del siglo XXI. Análisis de las diferencias entre países. Rev Med Chil 2012; 140: 50–58.

Abreu LM. Análisis del contexto sociodemográfico de la base poblacional Batey. COPRESIDA, p. 57.

(CESDEM) CdESyD. Encuesta Sociodemográfica y sobre VIH/SIDA en los Bateyes Estatales de la República Dominicana 2007, Dominican Republic: CESDEM, 2007, pp. 270–283.

Brewer

Hasbun

Ryan

et al.

Migration, ethnicity and environment: HIV risk factors for women on the sugar cane plantations of the Dominican Republic. AIDS 1998; 12: 1879–1887.

(CESDEM) CdESyD. Encuesta Demográfica y de Salud 2007, Dominican Republic: República Dominicana (ENDESA) Prevelencia de VIH, 2007, pp. 2–2.

Farmer

Léandre

Mukherjee

et al.

Community-based approaches to HIV treatment in resource-poor settings. Lancet 2001; 358: 5–5.

10.

UNAIDS. Global Report: UNAIDS report on the global AIDS epidemic, Geneva: UNAIDS, 2012, pp. 110–110.

11.

Duerr

Altice

Harries

et al.

Antiretroviral treatment in resource-limited settings 2012. AIDS Res Treat 2012; 2012: 346708–346708.

12.

Kilmarx

Mutasa-Apollo

. Patching a leaky pipe: the cascade of HIV care. Curr Opin HIV AIDS 2013; 8: 59–64.

13.

El-Sadr

Gamble

Cohen

. Linkage from HIV testing to care: a positive test often leads nowhere. Sex Transm Dis 2013; 40: 26–27.

14.

Alvarez-Uria

Pakam

Midde

et al.

Entry, retention, and virological suppression in an HIV cohort study in India: description of the cascade of care and implications for reducing HIV-related mortality in low- and middle-income countries. Interdiscip Perspect Infect Dis 2013; 2013: 384805–384805.

15.

Margaret

McNairy

Batya Elul et al.

The HIV care continuum: no partial credit given. AIDS 2012; 26: 1735–1738.

16.

Halpern M. Reporte Anual 2011. Dominican Republic: Clínica de Familia La Romana. 2011, p. 19.

17.

Scholl E. HIV in the land of baseball and Bachata. AIDSTAR-One: case study series. Available at: www.aidstar-one.com (2012, accessed 10 April 2015).

18.

World Health Organization. WHO ART guidelines for adults and adolescents, Geneva: WHO, 2009.

19.

Gardner

McLees

Steiner

et al.

The spectrum of engagement in HIV care and its relevance to test-and-treat strategies for prevention of HIV infection. Clin Infect Dis 2011; 52: 793–800.

20.

Mugglin

Estill

Wandeler

et al.

Loss to programme between HIV diagnosis and initiation of antiretroviral therapy in sub-Saharan Africa: systematic review and meta-analysis. Trop Med Int Health 2012; 17: 1509–1520.

21.

Andrews

Wood

Bekker

et al.

Projecting the benefits of antiretroviral therapy for HIV prevention: the impact of population mobility and linkage to care. J Infect Dis 2012; 206: 543–551.

22.

Decosas

Adrien

. Migration and HIV. AIDS 1997; 11(Suppl A): S77–S84.

23.

Lurie

Williams

Zuma

et al.

The impact of migration on HIV-1 transmission in South Africa: a study of migrant and nonmigrant men and their partners. Sex Transm Dis 2003; 30: 149–156.

24.

Soskolne

Shtarkshall

. Migration and HIV prevention programmes: linking structural factors, culture, and individual behaviour – an Israeli experience. Social Sci Med 2002; 55: 1297–1307.