ABSTRACT
Objective.
To describe blood pressure trajectories and hypertension control among patients enrolled in the HEARTS program in south-west Trinidad.
Methods.
This retrospective longitudinal cohort study used routinely collected HEARTS program data from 31 public primary health centers in the South-West Regional Health Authority, Trinidad and Tobago. Eligible patients initiated HEARTS between 1 April and 31 October 2023, with a baseline program record and at least one follow-up blood pressure measurement. Follow-up was categorized as month-1 to month-5. McNemar’s test assessed paired baseline versus month-5 control status; linear and generalized linear mixed-effects models assessed blood pressure trajectories and hypertension control.
Results.
The analytic cohort included 548 patients; 63.3% were female and mean age was 61.4 ± 11.0 years. At baseline, 80.7% had uncontrolled hypertension. Mean baseline systolic blood pressure was 155 ± 22 mmHg and diastolic 86 ± 14 mmHg. Among 502 patients with paired baseline and month-5 data, more patients transitioned from uncontrolled to controlled hypertension than the reverse (70 vs 41; McNemar OR 1.71, 95% CI [1.16, 2.51], p = 0.006). In mixed-effects analyses, month was associated with lower systolic blood pressure. Odds of hypertension control were higher at month-2 and month-5 relative to baseline, while higher body mass index (BMI) was associated with lower odds of control.
Conclusions.
Among patients with available follow-up data, the HEARTS program was associated with improved hypertension control in routine primary care, while higher BMI was associated with lower odds of control. These findings should be interpreted as observational program outcomes rather than evidence of a causal program effect.
Keywords: Hypertension, primary health care, longitudinal studies, Trinidad and Tobago
RESUMEN
Objetivo.
Describir las tendencias de la presión arterial y el control de la hipertensión en pacientes que participaron en la iniciativa HEARTS en el suroeste de Trinidad.
Método.
En este estudio de cohorte longitudinal retrospectivo se utilizaron datos recopilados de forma sistemática a partir de los registros de la iniciativa HEARTS en 31 centros públicos de atención primaria de la autoridad regional de salud del suroeste de Trinidad y Tabago. Los pacientes elegibles comenzaron a participar en la iniciativa HEARTS entre el 1 de abril y el 31 de octubre del 2023, y tenían un registro inicial de la presión arterial y al menos una medición de seguimiento. El seguimiento se clasificó en los meses 1 a 5. Se usó la prueba de McNemar para evaluar el estado de control emparejado entre la línea de base y el mes 5; se evaluaron las tendencias de la presión arterial y del control de la hipertensión mediante modelos lineales y modelos lineales generalizados de efectos mixtos.
Resultados.
La cohorte analítica incluyó a 548 pacientes; el 63,3 % eran mujeres y la edad promedio fue de 61,4 ± 11,0 años. Al inicio del estudio, el 80,7% presentaba hipertensión no controlada. La presión arterial sistólica promedio en la línea de base fue de 155 ± 22 mmHg y la diastólica, de 86 ± 14 mmHg. De los 502 pacientes con datos emparejados de la línea de base y del mes 5, más pacientes pasaron de tener hipertensión no controlada a controlada que al revés (70 frente a 41; OR de McNemar 1,71; IC95%: 1,16-2,51; p = 0,006). En los análisis de efectos mixtos, el mes de seguimiento se asoció con una presión arterial sistólica más baja. Las probabilidades de control de la hipertensión fueron mayores en los meses 2 y 5 en comparación con la línea de base, mientras que un índice de masa corporal (IMC) más alto se asoció con una menor probabilidad de control.
Conclusiones.
En los pacientes con datos de seguimiento disponibles, la iniciativa HEARTS se asoció con un mejor control de la hipertensión en la atención primaria habitual, mientras que un IMC más alto se asoció con una menor probabilidad de control. Estos hallazgos deben interpretarse como resultados observacionales de la iniciativa y no como evidencia de un efecto causal atribuible a ella.
Palabras clave: Hipertensión, atención primaria de salud, estudios longitudinales, Trinidad y Tobago
RESUMO
Objetivo.
Descrever as trajetórias da pressão arterial e o controle da hipertensão arterial entre os pacientes inscritos no programa HEARTS no sudoeste de Trinidad.
Métodos.
Este estudo de coorte longitudinal retrospectivo utilizou dados do programa HEARTS coletados rotineiramente em 31 centros públicos de atenção primária à saúde da Autoridade Regional de Saúde do Sudoeste de Trinidad e Tobago. Pacientes elegíveis iniciaram o programa HEARTS entre 1º de abril e 31 de outubro de 2023, tiveram seus dados da linha de base registrados e pelo menos uma medição de monitoramento da pressão arterial. O monitoramento foi classificado conforme o período, do mês 1 ao mês 5. O teste de McNemar avaliou os dados da linha de base pareados em comparação com os do controle do mês 5; modelos lineares e generalizados de efeitos lineares mistos avaliaram as trajetórias da pressão arterial e o controle da hipertensão.
Resultados.
A coorte analítica incluiu 548 pacientes – 63,3% eram do sexo feminino e a idade média era de 61,4 ± 11,0 anos. Na linha de base, 80,7% apresentavam hipertensão não controlada. A pressão arterial sistólica média inicial era de 155 ± 22 mmHg e a diastólica, de 86 ± 14 mmHg. Entre os 502 pacientes com dados pareados na linha de base e no mês 5, mais pacientes fizeram a transição da hipertensão não controlada para a controlada do que o inverso (70 versus 41; McNemar OR 1,71; IC 95% [1,16–2,51], p = 0,006. Nas análises de efeitos mistos, o mês foi associado a uma pressão arterial sistólica mais baixa. As chances de controle da hipertensão arterial foram maiores no mês 2 e no mês 5 em relação à linha de base, enquanto que um índice de massa corporal (IMC) mais alto foi associado a menores chances de controle.
Conclusões.
Entre os pacientes com dados de acompanhamento disponíveis, o programa HEARTS foi associado a um melhor controle da hipertensão arterial na atenção primária de rotina, ao passo que um IMC mais alto foi associado a menores chances de controle. Esses achados devem ser interpretados como resultados observacionais do programa, e não como evidências de um efeito causal do programa.
Palavras-chave: Hipertensão, atenção primária à saúde, estudos longitudinais, Trinidad e Tobago
Cardiovascular disease (CVD) remains the leading cause of mortality worldwide, accounting for approximately one-third of global deaths (1). Hypertension is the most important modifiable risk factor underlying this burden and contributes substantially to the development of stroke, myocardial infarction, heart failure, and chronic kidney disease (2, 3). Globally, an estimated 1.28 billion adults are living with hypertension, yet nearly half are unaware of their condition, with only 1 in 5 persons having optimum control of their blood pressure (BP) (4).
This public health concern directly threatens the achievement of United Nations Sustainable Development Goal 3 (SDG 3), which aims to ensure healthy lives and promote well-being for all. Specifically, target 3.4 of SDG 3 seeks to reduce premature mortality from noncommunicable diseases (NCDs) by one-third through prevention, treatment, and promotion of mental health and well-being by 2030 (5). Uncontrolled hypertension, a principal driver of CVD, remains a substantial barrier to achieving optimal cardiometabolic health in Trinidad and Tobago and the wider Caribbean (6). Hypertension prevalence is estimated at 34.5% in the Caribbean and 25.8% in Trinidad and Tobago, highlighting the substantial regional burden (7).
In response to this global crisis, the World Health Organization (WHO) launched the HEARTS technical package in 2016 as a strategic initiative to strengthen the prevention and control of CVD in primary health care, representing a comprehensive, systems-based approach to hypertension management which comprises six interrelated components: healthy-lifestyle counseling; evidence-based treatment protocols; access to essential medicines and technology; risk-based CVD management; team-based care; and systems for monitoring (8–10). The Ministry of Health of Trinidad and Tobago, in partnership with the Pan American Health Organization, introduced the HEARTS technical package for hypertension control in 2019 (11). The program was introduced through a phased rollout, beginning with five primary health centers (PHCs) in Trinidad and Tobago and subsequently expanded to all PHCs nationwide in 2021. Concurrent efforts focused on implementation of a standardized monitoring tool and strengthening human resource capacity. The HEARTS treatment algorithm, as shown in Table 1, was displayed throughout all 34 PHCs of the South-West Regional Health Authority (SWRHA) in Trinidad and Tobago.
TABLE 1. HEARTS medication algorithm used in the primary care facilities of the SWRHA, Trinidad and Tobago.
|
If BP ≥140/90 with no comorbidities | |||
|---|---|---|---|
|
If uncontrolled (SBP ≥140 mmHg or DBP ≥90 mmHg) after one month, move to the next step | |||
|
Algorithm 1 – Acceptable |
Algorithm 2 – Preferred |
||
|
Step 1 |
ACE - lisinopril 10 mg, plus CCB - amlodipine 5 mg |
Step 1 |
ARB - irbesartan 150 mg or candesartan 16 mg, plus CCB - amlodipine 5 mg |
|
Step 2 |
ACE - lisinopril 20 mg, plus CCB - amlodipine 5 mg |
Step 2 |
ARB - irbesartan 300 mg or candesartan 32 mg, plus CCB - amlodipine 5 mg |
|
Step 3 |
ACE - lisinopril 20 mg, plus CCB - amlodipine 10 mg |
Step 3 |
ARB - irbesartan 300 mg or candesartan 32 mg, plus CCB - amlodipine 10 mg |
|
Step 4 |
ACE - lisinopril 40 mg, plus CCB - amlodipine 10 mg |
Step 4 |
Add chlorthalidone 12.5 mg |
|
Step 5 |
Add bendrofluazide 2.5 mg |
Step 5 |
Increase chlorthalidone 25 mg |
|
Step 6: If uncontrolled refer | |||
ACE, angiotensin-converting enzyme inhibitor; CCB, calcium channel blocker; ARB, angiotensin II receptor blockers – candesartan may be substituted if irbesartan is unavailable. ACE and ARB are avoided in pregnancy.
Source: Prepared by the authors.
One aspect of the HEARTS program is to convert a patient from a state of uncontrolled hypertension to a state of controlled hypertension via a standardized medication algorithm with incorporation of a combination therapy and stepwise escalation framework for patients with persistent uncontrolled hypertension (defined as systolic BP ≥140 mmHg and/or diastolic BP ≥90 mmHg) (12, 13). A key objective of the HEARTS program is to support hypertensive patients in achieving and maintaining BP control.
Regional studies have documented HEARTS implementation and hypertension outcomes across Caribbean and Latin American primary care settings. A community-based hypertension-control program in Matanzas, Cuba, incorporating interventions recommended within HEARTS reported improvements in hypertension treatment and control following implementation (14). A multicountry quality-improvement evaluation that included several Caribbean countries also reported improvements in the quality of standardized hypertension clinical pathways following the HEARTS intervention (15). However, Caribbean evidence remains predominantly focused on implementation and program-level or cross-sectional outcomes, with limited longitudinal patient-level analyses of BP trajectories under routine primary care conditions.
This study examined BP trajectories among patients with a first recorded HEARTS program visit in SWRHA primary care facilities during April to October 2023. Specifically, it assessed transitions in hypertension control and changes in systolic blood pressure (SBP) and diastolic blood pressure (DBP) over a follow-up period of up to five months. Using longitudinal analytical approaches, this study aimed to provide observational evidence from a Caribbean routine-care setting on BP outcomes among patients enrolled in the HEARTS program. Findings may inform program refinement, quality improvement initiatives, and future prospective evaluations.
METHODS
Study design and setting
A retrospective longitudinal cohort study was conducted using routinely collected program data extracted from HEARTS patient monitoring spreadsheets implemented across 34 PHCs within the SWRHA, Trinidad and Tobago. The spreadsheets recorded visit dates and demographic and clinical information collected during routine follow-up of patients enrolled in the HEARTS program.
Study population
The monitoring spreadsheets contained records for 5 001 patients across 34 PHCs from April to October 2023. Eligible patients had a first recorded HEARTS program visit and at least one follow-up BP measurement within five months of baseline. Patients were excluded if they had any recorded HEARTS visit before April 2023 or no eligible follow-up visit. Records were excluded from the relevant analyses because of invalid or inconsistent visit dates or sequences, missing or erroneous BP measurements, and duplicate entries. Implausible or inconsistent values for ethnicity, mass, height, body mass index (BMI), recorded hypertension status, prescribed treatment step, CVD, diabetes, and smoking status were coded as missing for the affected visit. The patient and all other valid data were retained when the eligibility criteria were met. A complete comparison of baseline characteristics between included and excluded patients was not possible because valid, comparable baseline information was unavailable for many excluded patients. Following the procedures shown in Figure 1, the final analytic cohort consisted of 548 patients from 31 health centers.
FIGURE 1. Selection process for retrospective cohort.
BP, blood pressure; PHC, primary health center; SWRHA, South-West Regional Health Authority; T&T, Trinidad and Tobago.
Source: Prepared by the authors based on the study data.
The HEARTS program enrolled patients with elevated BP regardless of prior antihypertensive treatment. Because pre-enrolment treatment status was not recorded separately, participants could not be reliably classified as treatment-naive or previously treated. Patients with documented allergies to medications in the HEARTS treatment algorithm were excluded.
HEARTS treatment protocol
At baseline, patients entered the recorded treatment pathway at varying stages of prior hypertension management and were expected to receive protocol-based pharmacological therapy following clinical review. The treatment protocol consisted of five sequential treatment steps, each representing an escalation in the intensity or combination of antihypertensive medications, as outlined in Table 1.
Patients who remained uncontrolled after completion of the five treatment steps were referred for specialist evaluation. Uncontrolled hypertension was defined as an SBP ≥140 mmHg or DBP ≥90 mmHg.
Follow-up structure
Baseline corresponded to each patient’s first recorded HEARTS program visit. Subsequent routine-care records were reviewed for up to five months after baseline, and eligible follow-up visits were categorized as month-1 through month-5.
Cohort structure
All included patients contributed baseline data. A total of 502 patients had an eligible month-5 follow-up, irrespective of interim attendance. Nested subcohorts included baseline plus two (n = 52), three (n = 11), and four (n = 5) consecutive monthly follow-up visits.
Variables
Demographic variables included age, sex, and ethnicity. Clinical variables extracted from the monitoring system included SBP, DBP, BMI, smoking status, presence of CVD, presence of diabetes mellitus (DM), and prescribed HEARTS treatment step.
Outcomes
The primary outcomes were SBP and DBP measured during program visits. The secondary outcome was hypertension control status.
Statistical analysis
Baseline characteristics were described using mean values with standard deviations for continuous variables and frequencies with percentages for categorical variables.
Changes in hypertension control status between baseline and month-5 were evaluated using McNemar’s test for paired categorical data irrespective of visits in between. McNemar’s odds ratio (ORMc) was calculated to quantify the direction and magnitude of change between the two time points.
Repeated-measures analysis of variance (ANOVA) was used to evaluate within-patient changes in SBP and DBP from baseline among subcohorts with two or more consecutive monthly follow-up visits.
Linear mixed-effects models (LMMs) were used for the continuous SBP and DBP outcomes, and binomial generalized linear mixed-effects models (GLMMs) for the binary hypertension-control outcome. These models accommodated unbalanced repeated observations within patients and clustering within PHCs. Models with different covariate combinations were fitted to assess the robustness of the month estimates and show the effect of missing covariate data on sample size. Month was entered as a categorical fixed effect, with baseline as the reference category. In both the LMMs and GLMMs, prescribed treatment step was entered as a mean-centered numeric fixed effect. BMI and age were entered as mean-centered continuous fixed effects, and DM and CVD as categorical fixed effects. Random intercepts were fitted for patients and PHCs.
All eligible observations were included irrespective of missing interim visits. Follow-up observations were assigned to months 1–5 according to elapsed time from baseline, and consecutive monthly attendance was not required. No values were imputed; observations missing a covariate were excluded only from models containing that covariate. Among 1 333 observations from 548 patients, prescribed treatment step was missing for 39 observations (2.9%), BMI for 407 (30.5%), DM for 468 (35.1%), and CVD for 226 (17.0%). Missingness was quantified by variable but was not formally modeled.
LMMs were estimated using restricted maximum likelihood and GLMMs using maximum likelihood with Laplace approximation. Residual-versus-fitted and quantile–quantile plots were examined for the LMMs, while convergence and overdispersion were assessed for the GLMMs. Analysis of the available follow-up observations relied on the assumption that missing data were conditionally missing at random given the observed attendance and clinical data. This assumption could not be verified and may not hold if missingness depended on unmeasured disease severity or medication adherence. Exclusion of observations with missing covariates may therefore have affected the magnitude and precision of the estimates. No missing-data sensitivity analysis was performed.
All statistical analyses were performed using jamovi (version 2.6.26) and R (version 4.5.0). The threshold for α was established a priori at 0.05. All odds ratios (ORs) were presented with 95% confidence intervals (CIs).
Ethics
The study was approved by the SWRHA’s bioethics committee on 13 May 2024 (reference number 2024–01–17–15–05).
RESULTS
Following cohort derivation procedures highlighted in Figure 1, the final analytic cohort comprised 548 patients receiving care through the HEARTS program across 31 PHCs within the SWRHA, Trinidad and Tobago. Table 2 shows that from baseline, 126 of those patients visited in month-1, 94 in month-2, 24 in month-3, 39 in month-4, and 502 in month-5. Mean age at baseline was 61.4 ± 11.0 years (n = 548), 63.3% (n = 347) were female, and 80.7% (n = 442) had uncontrolled hypertension at baseline. Mean SBP at baseline was 155 ± 22 mmHg (n = 548) and mean DBP was 86 ± 14 mmHg (n = 548).
TABLE 2. Patient characteristics at baseline and subsequent visits (variable follow-up attendance).
|
Parameter |
Baseline |
Month-1 |
Month-2 |
Month-3 |
Month-4 |
Month-5 |
|
|---|---|---|---|---|---|---|---|
|
Age |
Years mean ± SD |
61.4 ± 11.0 (n = 548) |
59.6 ± 11.4 (n = 126) |
59.1 ± 10.8 (n = 94) |
58.8 ± 11.0 (n = 24) |
59.1 ± 12.5 (n = 39) |
62.0 ± 10.8 (n = 502) |
|
Sex |
Female n (%) |
347 (63.3%) |
73 (57.9%) |
53 (56.4%) |
12 (50.0%) |
26 (66.7%) |
324 (64.5%) |
|
Male n (%) |
201 (36.7%) |
53 (42.1%) |
41 (43.6%) |
12 (50.0%) |
13 (33.3%) |
178 (35.5%) |
|
|
Ethnicity |
East Indian n (%) |
202 (49.9%) |
32 (36.0%) |
30 (42.9%) |
7 (41.2%) |
8 (26.7%) |
184 (50.8%) |
|
African n (%) |
102 (25.2%) |
35 (39.3%) |
27 (38.6%) |
7 (41.2%) |
17 (56.7%) |
84 (23.2%) |
|
|
Mixed n (%) |
41 (10.1%) |
7 (7.9%) |
3 (4.3%) |
1 (5.9%) |
0 (0.0%) |
39 (10.8%) |
|
|
Other n (%) |
60 (14.8%) |
15 (16.9%) |
10 (14.3%) |
2 (11.8%) |
5 (16.7%) |
55 (15.2%) |
|
|
Mass |
kg mean ± SD |
79.1 ± 20.0 (n = 439) |
84.1 ± 22.2 (n = 102) |
81.6 ± 19.7 (n = 80) |
85.4 ± 17.3 (n = 22) |
88.2 ± 29.0 (n = 34) |
77.8 ± 18.8 (n = 443) |
|
Height |
m mean ± SD |
1.62 ± 0.10 (n = 345) |
1.64 ± 0.10 (n = 89) |
1.64 ± 0.10 (n = 73) |
1.62 ± 0.12 (n = 18) |
1.66 ± 0.11 (n = 30) |
1.62 ± 0.10 (n = 395) |
|
BMI |
kgm-² mean ± SD |
29.7 ± 6.3 (n = 333) |
31.1 ± 6.4 (n = 86) |
30.3 ± 6.0 (n = 72) |
33.2 ± 5.7 (n = 18) |
31.7 ± 9.2 (n = 30) |
29.7 ± 6.5 (n = 387) |
|
BP |
SBP mmHg mean ± SD |
155 ± 22 (n = 548) |
153 ± 20 (n = 126) |
150 ± 20 (n = 94) |
156 ± 22 (n = 24) |
165 ± 22 (n = 39) |
153 ± 23 (n = 502) |
|
DBP mmHg mean ± SD |
86 ± 14 (n = 548) |
86 ± 16 (n = 126) |
85 ± 11 (n = 94) |
84 ± 14 (n = 24) |
89 ± 12 (n = 39) |
85 ± 14 (n = 502) |
|
|
Hypertensive status |
Uncontrolled n (%) |
442 (80.7%) |
100 (79.4%) |
67 (71.3%) |
19 (79.2%) |
35 (89.7%) |
372 (74.1%) |
|
Controlled n (%) |
106 (19.3%) |
26 (20.6%) |
27 (28.7%) |
5 (20.8%) |
4 (10.3%) |
130 (25.9%) |
|
|
Prescribed HEARTS treatment step at visit |
Step 1 n (%) |
321 (62.6%) |
60 (49.2%) |
35 (37.2%) |
9 (37.5%) |
7 (17.9%) |
249 (50.3%) |
|
Step 2 n (%) |
97 (18.9%) |
31 (25.4%) |
29 (30.9%) |
6 (25.0%) |
14 (35.9%) |
122 (24.6%) |
|
|
Step 3 n (%) |
74 (14.4%) |
22 (18.0%) |
22 (23.4%) |
7 (29.2%) |
12 (30.8%) |
90 (18.2%) |
|
|
Step 4 n (%) |
13 (2.5%) |
7 (5.7%) |
6 (6.4%) |
1 (4.2%) |
4 (10.3%) |
28 (5.7%) |
|
|
Step 5 n (%) |
8 (1.6%) |
2 (1.6%) |
2 (2.1%) |
1 (4.2%) |
2 (5.1%) |
6 (1.2%) |
|
|
CVD |
Yes n (%) |
20 (4.6%) |
4 (3.4%) |
0 (0.0%) |
0 (0.0%) |
1 (2.6%) |
18 (4.3%) |
|
No n (%) |
411 (95.4%) |
112 (96.6%) |
83 (100.0%) |
23 (100.0%) |
37 (97.4%) |
398 (95.7%) |
|
|
DM |
Yes n (%) |
312 (90.2%) |
72 (73.5%) |
53 (67.9%) |
13 (68.4%) |
18 (72.0%) |
291 (97.3%) |
|
No n (%) |
34 (9.8%) |
26 (26.5%) |
25 (32.1%) |
6 (31.6%) |
7 (28.0%) |
8 (2.7%) |
|
|
Smoker |
Current n (%) |
23 (4.3%) |
5 (4.1%) |
5 (5.7%) |
1 (4.3%) |
3 (7.7%) |
15 (3.3%) |
|
Former n (%) |
8 (1.5%) |
2 (1.6%) |
3 (3.4%) |
0 (0.0%) |
0 (0.0%) |
4 (0.9%) |
|
|
Never n (%) |
498 (94.1%) |
115 (94.3%) |
80 (90.9%) |
22 (95.7%) |
36 (92.3%) |
436 (95.8%) |
|
BMI, body mass index; BP, blood pressure; CVD, cardiovascular disease; DBP, diastolic blood pressure; DM, diabetes mellitus; SBP, systolic blood pressure.
Data at each time point arose from an unbalanced longitudinal cohort in which some individuals contributed repeated observations, but overall sample composition varied across visits. Accordingly, differences across time points did not represent purely within-patient change.
Source: Prepared by the authors based on the study data.
At baseline mean BMI was 29.7 ± 6.3 kgm-² (n = 333). DM was present in 90.2% (n = 312) of patients and CVD in 4.6% (n = 20). At the initial visit, 62.6% (n = 321) were prescribed treatment step 1, 18.9% (n = 97) step 2, 14.4% (n = 74) step 3, 2.5% (n = 13) step 4, and 1.6% (n = 8) step 5. A recorded treatment step at baseline was unavailable for 6.4% of patients (n = 35).
Among 502 patients with paired baseline and month-5 hypertension status, 70 transitioned from uncontrolled to controlled hypertension, whereas 41 transitioned from controlled to uncontrolled hypertension (ORMc 1.71, 95% CI [1.16, 2.51], p = 0.006; n = 502).
In complete-case subcohorts, repeated-measures ANOVA showed significant changes in SBP and DBP across three consecutive observations, including baseline (n = 52; p < 0.001 and p = 0.004, respectively). Mean SBP declined from 164 ± 21 mmHg at baseline to 151 ± 20 mmHg at month-2, while mean DBP declined from 90 ± 14 mmHg to 85 ± 11 mmHg. Neither SBP nor DBP changed significantly in the four-observation cohort (n = 11). In the five-observation cohort (n = 5), SBP changed significantly (p = 0.009) whereas DBP did not (p = 0.226).
Estimated marginal mean SBP and DBP trajectories based on LMM are shown in Figure 2. Unadjusted LMM demonstrated an overall effect of follow-up month on SBP (p < 0.001; n = 538). After inclusion of prescribed treatment step, the temporal effect on SBP remained significant (p < 0.001; n = 538), and prescribed treatment step remained significant in the SBP model (p = 0.02; n = 538) but was not significant in the DBP model (p = 0.08; n = 538). In models additionally adjusted for age and BMI, the month effect for SBP attenuated (p = 0.09; n = 436), whereas BMI (p = 0.03; n = 436) and age (p < 0.001; n = 436) remained significant predictors. In the corresponding DBP LMM adjusted for prescribed treatment, age, and BMI, follow-up month remained significant (p = 0.007; n = 436), and age was also a significant predictor (p < 0.001; n = 436).
FIGURE 2. Estimated marginal means of systolic and diastolic blood pressure based on linear mixed-effects models.
Source: Prepared by the authors based on the study data.
GLMM analyses in Table 3 demonstrated time-dependent improvements in the odds of hypertension control, particularly at month-2 and month-5 relative to baseline. At month-2, the odds of hypertension control were approximately 2.8 times those at baseline in the unadjusted model (OR 2.78, 95% CI [1.42, 5.40]; n = 548). This corresponded to an absolute increase of 14.7 percentage points in the predicted probability of hypertension control (AME 0.147, 95% CI [0.038, 0.255]; n = 548).
TABLE 3. GLMM estimates of hypertension control across follow-up visits (OR and AME, 95% CI).
|
Month (n = 548) Observations = 1 333 |
Month & treatment (n = 548) Observations = 1 294 |
Month, treatment, CVD (n = 462) Observations = 1 077 |
Month, treatment, BMI (n = 436) Observations = 894 |
Month, treatment, BMI, age, DM (n = 273) Observations = 576 |
Month, treatment, BMI, age, DM, CVD (n = 238) Observations = 514 |
|
|---|---|---|---|---|---|---|
|
Baseline odds of hypertension control |
OR 0.12 (0.08, 0.18) |
OR 0.10 (0.06, 0.17) |
OR 0.09 (0.05, 0.15) |
OR 0.11 (0.06, 0.20) |
OR 0.28 (0.10, 0.75) |
OR 0.32 (0.12, 0.90) |
|
Month-1 compared to baseline |
OR 1.43 (0.78, 2.65) |
OR 1.79 (0.86, 3.73) |
OR 2.05 (0.96, 4.39) |
OR 1.97 (0.85, 4.56) |
OR 1.53 (0.58, 4.01) |
OR 1.48 (0.55, 3.98) |
|
AME 0.045 (–0.036, 0.126) p = 0.271 |
AME 0.072 (–0.024, 0.168) p = 0.139 |
AME 0.087 (–0.011, 0.185) p = 0.082 |
AME 0.087 (–0.028, 0.205) p = 0.137 |
AME 0.055 (–0.075, 0.185) p = 0.403 |
AME 0.052 (–0.082, 0.186) p = 0.447 |
|
|
Month-2 compared to baseline |
OR 2.78 (1.42, 5.40) |
OR 3.49 (1.56, 7.81) |
OR 4.32 (1.83, 10.16) |
OR 3.34 (1.35, 8.29) |
OR 3.94 (1.44, 10.79) |
OR 3.29 (1.13, 9.53) |
|
AME 0.147 (0.038, 0.255) p = 0.008 |
AME 0.178 (0.051, 0.304) p = 0.006 |
AME 0.208 (0.073, 0.342) p = 0.003 |
AME 0.177 (0.032, 0.322) p = 0.016 |
AME 0.220 (0.049, 0.390) p = 0.012 |
AME 0.187 (0.012, 0.362) p = 0.036 |
|
|
Month-3 compared to baseline |
OR 1.72 (0.47, 6.23) |
OR 2.25 (0.57, 8.81) |
OR 2.73 (0.69, 10.85) |
OR 1.59 (0.31, 8.12) |
OR 1.18 (0.18, 7.68) |
OR 1.07 (0.16, 7.10) |
|
AME 0.071 (–0.115, 0.257) p = 0.457 |
AME 0.105 (–0.097, 0.308) p = 0.307 |
AME 0.130 (–0.079, 0.340) p = 0.224 |
AME 0.057 (–0.164, 0.278) p = 0.612 |
AME 0.020 (–0.216, 0.257) p = 0.868 |
AME 0.008 (–0.225, 0.241) p = 0.946 |
|
|
Month-4 compared to baseline |
OR 0.84 (0.25, 2.88) |
OR 1.13 (0.30, 4.21) |
OR 1.29 (0.34, 4.92) |
OR 0.93 (0.20, 4.26) |
OR 0.93 (0.14, 5.97) |
OR 0.84 (0.13, 5.60) |
|
AME –0.019 (–0.150, 0.112) p = 0.775 |
AME 0.013 (–0.136, 0.163) p = 0.863 |
AME 0.028 (–0.123, 0.179) p = 0.717 |
AME –0.008 (–0.171, 0.154) p = 0.920 |
AME –0.009 (–0.220, 0.202) p = 0.934 |
AME –0.020 (–0.230, 0.190) p = 0.851 |
|
|
Month-5 compared to baseline |
OR 1.67 (1.17, 2.38) |
OR 2.17 (1.25, 3.78) |
OR 2.38 (1.30, 4.35) |
OR 2.01 (1.02, 3.95) |
OR 2.54 (1.17, 5.50) |
OR 2.46 (1.08, 5.59) |
|
AME 0.066 (0.020, 0.113) p = 0.005 |
AME 0.100 (0.028, 0.171) p = 0.006 |
AME 0.108 (0.033, 0.184) p = 0.005 |
AME 0.081 (0.002, 0.160) p = 0.044 |
AME 0.137 (0.026, 0.247) p = 0.015 |
AME 0.133 (0.015, 0.252) p = 0.027 |
|
|
Treatment |
– |
OR 0.87 (0.67, 1.13) |
OR 0.83 (0.63, 1.10) |
OR 0.93 (0.69, 1.25) |
OR 0.85 (0.59, 1.21) |
OR 0.87 (0.60, 1.26) |
|
BMI |
– |
– |
– |
OR 0.93 (0.89, 0.97) |
OR 0.91 (0.86, 0.96) |
OR 0.92 (0.87, 0.97) |
|
Age |
– |
– |
– |
– |
OR 0.99 (0.96, 1.02) |
OR 0.99 (0.96, 1.01) |
|
CVD (Yes vs No) |
– |
– |
OR 4.18 (1.39, 12.56) |
– |
– |
OR 4.02 (1.09, 15.51) |
|
DM (Yes vs No) |
– |
– |
– |
– |
OR 0.38 (0.12, 1.16) |
OR 0.43 (0.18, 1.04) |
AME, average marginal effect; BMI, body mass index; CVD, cardiovascular disease; DM, diabetes mellitus; GLMM, generalized linear mixed models; OR, odds ratio.
Odds ratios (OR) were derived from generalized linear mixed-effects models with random intercepts for patients and primary health centers. Average marginal effects for month represent the average difference in predicted probability of hypertension control between each follow-up month and baseline. Models account for repeated measurements within individuals; follow-up observations were unbalanced, with patients not necessarily observed at every time point.
Source: Prepared by the authors based on the study data.
After adjustment for prescribed treatment step and CVD, the odds of hypertension control were approximately 4.3 times those at baseline (OR 4.32, 95% CI [1.83, 10.16]; n = 462), corresponding to a 20.8 percentage-point increase in predicted probability of hypertension control (AME 0.208, 95% CI [0.073, 0.342]; n = 462). This estimate should be interpreted cautiously because the CVD subgroup was small.
In the model additionally adjusted for prescribed treatment, BMI, age, and DM, while excluding CVD, the odds of hypertension control were approximately 3.9 times those at baseline (OR 3.94, 95% CI [1.44, 10.79]; n = 273), corresponding to a 22.0 percentage-point increase in predicted probability of hypertension control (AME 0.220, 95% CI [0.049, 0.390]; n = 273). These were the largest month-2 OR and AME among models excluding CVD. Overall, month-2 improvements remained consistent across the series of models.
At month-5, the odds of hypertension control were approximately 1.7 times those at baseline in the unadjusted model (OR 1.67, 95% CI [1.17, 2.38]; n = 548), corresponding to a 6.6 percentage-point increase in predicted probability of hypertension control (AME 0.066, 95% CI [0.020, 0.113]; n = 548). After adjustment for prescribed treatment, BMI, age, and DM, while excluding CVD, the odds of hypertension control were approximately 2.5 times those at baseline (OR 2.54, 95% CI [1.17, 5.50]; n = 273), corresponding to a 13.7 percentage-point increase in predicted probability of hypertension control (AME 0.137, 95% CI [0.026, 0.247]; n = 273). Overall, month-5 improvements remained consistent across the series of models.
BMI was consistently associated with reduced odds of hypertension control. For each 1 kgm-² increase in BMI, patients had approximately 7% lower odds of hypertension control in the model adjusted for prescribed treatment step and BMI (OR 0.93, 95% CI [0.89, 0.97]; n = 436). In the model adjusted for prescribed treatment step, BMI, age, and DM, each 1 kgm-² increase in BMI was associated with approximately 9% lower odds of hypertension control (OR 0.91, 95% CI [0.86, 0.96]; n = 273). This inverse association also remained statistically significant in all models that included BMI.
DISCUSSION
This retrospective longitudinal cohort study described short-term BP outcomes among patients receiving care within the HEARTS program at public primary care facilities in the SWRHA, Trinidad and Tobago. Higher odds of hypertension control were observed at month-2 and month-5 relative to baseline, while higher BMI was consistently associated with lower odds of control. The mixed-effects analyses also identified differences in continuous BP outcomes across follow-up months. These findings provide patient-level observational evidence from a Caribbean routine-care setting and are consistent with favorable BP outcomes among patients receiving care within the HEARTS program; however, the study design does not permit definitive causal attribution to the program (16, 17).
The paired transition analysis provided directly interpretable within-patient evidence of improvement. Among 502 patients with paired baseline and month-5 observations, 70 transitioned from uncontrolled to controlled hypertension compared with 41 who changed from controlled to uncontrolled status, indicating that the odds of improvement were 1.71 times those of deterioration during follow-up. These findings represent an observed association rather than proof of a causal program effect, and regression to the mean, treatment initiation effects, and selective retention cannot be excluded (18).
The complete-case ANOVA findings were consistent with early BP improvement in the three-visit cohort, but the small four- and five-visit cohorts limited inference; therefore, primary emphasis was placed on the mixed-effects analyses.
Linear mixed-effects analyses of continuous BP outcomes also suggested favorable change over time. In unadjusted models, follow-up month was significantly associated with SBP. After adjustment for prescribed treatment step, the month effect remained significant for SBP, and prescribed treatment step was also significantly associated with SBP, whereas treatment step was not significantly associated with DBP. In models additionally adjusted for age and BMI, the month effect for SBP was no longer statistically significant, while both age and BMI remained significant predictors. For DBP, follow-up month remained statistically significant after adjustment, age remained a significant predictor, and BMI was not significantly associated with outcome. These findings indicate that SBP and DBP trajectories were influenced by different combinations of follow-up time and patient-level characteristics.
Generalized linear mixed-effects analyses supported a pattern of favorable associations during follow-up. In unadjusted models, odds of hypertension control were higher at month-2 and month-5 than at baseline. Across models adjusted for treatment step, CVD, BMI, age, and DM, improvements remained evident. At month-2, the odds of control were three to four times those at baseline across adjusted models (OR 3.29–4.32). By month-5, improvement persisted but was attenuated (OR 2.01–2.54). Early gains may reflect treatment initiation, dose optimization, reinforcement of adherence, and heightened patient engagement. Later attenuation may reflect declining adherence, therapeutic inertia, intermittent attendance, or persistence of patients with more treatment-resistant hypertension in follow-up (19–21).
Some adjusted estimates nevertheless require careful interpretation. The largest month-2 OR (4.32), displayed in Table 3, arose in the model including CVD and prescribed treatment step. Although the overall findings remained directionally consistent, this model contained a very small subgroup of patients with CVD, which may have contributed to statistical instability, sparse-data effects, or residual confounding. The estimate should therefore be interpreted as imprecise rather than definitive, and greater emphasis should be placed on the consistency of improvement observed across models than on the single highest OR.
Higher BMI was one of the most consistent adverse predictors of hypertension control. Across BMI-adjusted models, each 1 kgm-² increase in BMI was associated with approximately 7–9% lower odds of hypertension control. This finding is concordant with established evidence that obesity complicates BP management through sympathetic nervous system activation, renin–angiotensin–aldosterone system overactivity, insulin resistance, sodium retention, endothelial dysfunction, and sleep-disordered breathing (22–24). The observed association may have implications for program design, as structured weight management, dietary counseling, physical activity support, and broader cardiometabolic risk reduction may warrant consideration alongside medication algorithms (25, 26). Recent Trinidadian HEARTS data reporting lower control among overweight patients are consistent with this observational association (12). The present study did not directly evaluate obesity-management interventions; therefore, the extent to which such interventions may influence BP outcomes within HEARTS remains uncertain.
Prescribed HEARTS treatment steps should be interpreted cautiously. This variable reflects intended prescribing intensity rather than confirmed medication exposure, pharmacy refill continuity, or adherence. Lifestyle interventions – including sodium reduction, weight management, physical activity, smoking cessation support, and counseling – are also central components of the HEARTS package and may have contributed materially to observed improvements (8–10). Adherence and implementation-fidelity data were unavailable, so the relative contribution of pharmacological versus behavioral mechanisms cannot be determined from the present study.
The present findings are broadly concordant with the wider HEARTS literature. Previous regional reports have described HEARTS as a patient-centered model built on standardized treatment pathways, validated automated BP measuring devices, reliable medicine supply, team-based care, and monitoring systems (27). Published evidence has consisted of program descriptions, implementation reports, or cross-sectional evaluations. Longitudinal patient-level mixed-effects analyses from Caribbean public primary care settings remain limited. The present study therefore adds regionally relevant evidence by demonstrating measurable improvement under routine service conditions.
Several strengths support the credibility of these findings. First, the study used complementary analytical approaches suited to different outcome structures, including paired categorical analysis and mixed-effects modeling. Second, mixed-effects methods were appropriate for irregular and unbalanced follow-up data and likely reduced inefficiency relative to complete-case methods. Third, the cohort was drawn from 31 public primary care facilities, improving practical relevance across routine service settings. Fourth, the study evaluated routine implementation rather than idealized trial conditions, thereby increasing policy relevance.
Several limitations should also be considered. First, this observational study did not include a comparison group, counterfactual or quasi-experimental framework; therefore, the study design does not permit definitive causal attribution of the observed changes to the HEARTS program. Alternative explanations include treatment initiation or intensification, regression to the mean, changes in medication adherence, lifestyle modifications, secular trends, selective retention in care, and other unmeasured clinical or contextual factors. The routinely collected data available for this study could not disentangle the respective contributions of these factors. Second, the analytic cohort represented a selected subset of the program source population with available follow-up data, creating potential selection and survivorship bias. Third, missing covariate data reduced sample sizes in adjusted models; no imputation or sensitivity analysis was performed, and the missing-at-random assumption could not be verified. Informative missingness may therefore have affected the magnitude and precision of the estimates. Fourth, BP measurements were obtained in routine clinical practice across multiple centers, so variation in measurement technique or devices may have introduced non-differential measurement error. Fifth, month-5 was the final predefined follow-up timepoint within the observation period. Interim monthly visits were not consistently observed for all participants; therefore, this timepoint should not be interpreted as evidence of continuous monthly retention in care. Finally, the study was conducted within one regional health authority, limiting generalizability to other health systems.
The study population nevertheless reflects an important routine-care group: predominantly older adults with a high burden of uncontrolled hypertension receiving care in public primary care clinics. As such, the findings might be particularly relevant to similar low- and middle-income settings where hypertension burden is high, specialist access is limited, and primary care systems must manage chronic disease at scale.
In this selected analytic cohort, favorable hypertension-control outcomes were observed during routine primary care follow-up. Although the observational design does not permit conclusions about program effectiveness or the necessity of specialist-centered care, the findings are consistent with the potential importance of core system components, including standardized treatment pathways, validated BP measuring devices, reliable medicine supply, team-based care, continuity systems, and routine performance monitoring. The observed association between BMI and hypertension control identifies integrated obesity-management approaches as an area for future program evaluation rather than demonstrating the effectiveness of such interventions.
Future research should extend follow-up beyond five months, include direct measures of medication adherence and implementation-fidelity, compare included versus excluded patients where feasible, assess center-level heterogeneity, and evaluate longer-term cardiovascular outcomes and cost-effectiveness. Prospective pragmatic studies comparing different recall systems, lifestyle support models, or treatment-intensification strategies would also be valuable.
In conclusion, among patients with available follow-up data, favorable BP trends and higher observed hypertension control were seen over a five-month follow-up period after enrolment in the HEARTS program. Follow-up attendance was variable, and findings should be interpreted as observational program outcomes rather than definitive estimates of program effectiveness. Improvement was greatest early after baseline assessment and remained detectable at month-5 across multiple adjusted models. Higher BMI remained an important barrier to control. Taken together with existing regional evidence, these findings may help inform future efforts to strengthen structured hypertension programs in Trinidad and Tobago and the wider Caribbean, including further evaluation of retention in care, weight-management approaches, implementation quality, and long-term cardiovascular prevention.
Acknowledgments.
The authors thank the South-West Regional Health Authority of Trinidad and Tobago for its support. We also thank Abigail Nurse, Amanda Roopchand, Amelia Edwards, Antonia Elder, Cedellar Samuels, Cristina Lutchman, Dayna Ali, Delisa McLean, Farida Dolly, Keeran Manohar, Dr. Shevanand Gopeesingh, Stephon King, and Vishal Seunarine for their contributions.
Funding Statement
No specific funding was received for this study.
Footnotes
Funding.
No specific funding was received for this study.
Data availability.
Data available upon request. Data from the study may be requested from the Policy, Planning and Research Department of the South-West Regional Health Authority, Trinidad and Tobago.
Declaration of artificial intelligence (AI) use.
The authors declare that no AI tool was used during research or manuscript preparation.
Disclaimer.
Authors hold sole responsibility for the views expressed in the manuscript, which may not necessarily reflect the opinion or policy of the RPSP/PAJPH or the Pan American Health Organization (PAHO).
REFERENCES
- 1.Vaduganathan M, Mensah GA, Turco JV, Fuster V, Roth GA. The global burden of cardiovascular diseases and risk: a compass for future health. J Am Coll Cardiol. 2022;80(25):2361–2371. doi: 10.1016/j.jacc.2022.11.005. [DOI] [PubMed] [Google Scholar]; Vaduganathan M, Mensah GA, Turco JV, Fuster V, Roth GA. The global burden of cardiovascular diseases and risk: a compass for future health. J Am Coll Cardiol. 2022;80(25):2361–2371. https://doi.org/10.1016/j.jacc.2022.11.005. [DOI] [PubMed]
- 2.Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, et al. Global burden of cardiovascular diseases and risk factors, 1990–2019: update from the GBD 2019 Study. J Am Coll Cardiol. 2020;76(25):2982–3021. doi: 10.1016/j.jacc.2020.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]; Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, et al. Global burden of cardiovascular diseases and risk factors, 1990–2019: update from the GBD 2019 Study. J Am Coll Cardiol. 2020;76(25):2982–3021. https://doi.org/10.1016/j.jacc.2020.11.010. [DOI] [PMC free article] [PubMed]
- 3.Di Cesare M, Perel P, Taylor S, Kabudula C, Bixby H, Gaziano TA, et al. The heart of the world. Glob Heart. 2024;19(1):11. doi: 10.5334/gh.1288. [DOI] [PMC free article] [PubMed] [Google Scholar]; Di Cesare M, Perel P, Taylor S, Kabudula C, Bixby H, Gaziano TA, et al. The heart of the world. Glob Heart. 2024;19(1):11. https://doi.org/10.5334/gh.1288. [DOI] [PMC free article] [PubMed]
- 4.World Health Organization . Global report on hypertension: the race against a silent killer. Geneva: WHO; 2023. https://iris.who.int/handle/10665/372896 Available from: [Google Scholar]; World Health Organization. Global report on hypertension: the race against a silent killer. Geneva: WHO; 2023. Available from: https://iris.who.int/handle/10665/372896.
- 5.United Nations Department of Economic and Social Affairs . New York: United Nations; 2025. The Sustainable Development Goals report 2025.https://unstats.un.org/sdgs/report/2025/The-Sustainable-Development-Goals-Report-2025.pdf Available from: [Google Scholar]; United Nations Department of Economic and Social Affairs. The Sustainable Development Goals report 2025. New York: United Nations; 2025. Available from: https://unstats.un.org/sdgs/report/2025/The-Sustainable-Development-Goals-Report-2025.pdf.
- 6.GBD 2019 Diseases and Injuries Collaborators Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1204–1222. doi: 10.1016/S0140-6736(20)30925-9. [DOI] [PMC free article] [PubMed] [Google Scholar]; GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1204–1222. https://doi.org/10.1016/S0140-6736(20)30925-9. [DOI] [PMC free article] [PubMed]
- 7.NCD Risk Factor Collaboration (NCD-RisC) Worldwide trends in hypertension prevalence and progress in treatment and control from 1990 to 2019: a pooled analysis of 1201 population-representative studies with 104 million participants. Lancet. 2021;398(10304):957–980. doi: 10.1016/S0140-6736(21)01330-1. [DOI] [PMC free article] [PubMed] [Google Scholar]; NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in hypertension prevalence and progress in treatment and control from 1990 to 2019: a pooled analysis of 1201 population-representative studies with 104 million participants. Lancet. 2021;398(10304):957–980. https://doi.org/10.1016/S0140-6736(21)01330-1. [DOI] [PMC free article] [PubMed]
- 8.World Health Organization . Geneva: WHO; 2018. Technical package for cardiovascular disease management in primary health care: implementation guide.https://iris.who.int/handle/10665/275728 Available from: [Google Scholar]; World Health Organization. Technical package for cardiovascular disease management in primary health care: implementation guide. Geneva: WHO; 2018. Available from: https://iris.who.int/handle/10665/275728.
- 9.Ordunez P, Campbell NRC, Giraldo Arcila GP, Angell SY, Lombardi C, Brettler JW, et al. HEARTS in the Americas: innovations for improving hypertension and cardiovascular disease risk management in primary care. Rev Panam Salud Publica. 2022;46:e96. doi: 10.26633/RPSP.2022.96. [DOI] [PMC free article] [PubMed] [Google Scholar]; Ordunez P, Campbell NRC, Giraldo Arcila GP, Angell SY, Lombardi C, Brettler JW, et al. HEARTS in the Americas: innovations for improving hypertension and cardiovascular disease risk management in primary care. Rev Panam Salud Publica. 2022;46:e96. https://doi.org/10.26633/RPSP.2022.96. [DOI] [PMC free article] [PubMed]
- 10.Ordunez P, Campbell NRC, DiPette DJ, Jaffe MG, Rosende A, Martinez R, et al. HEARTS in the Americas: targeting health system change to improve population hypertension control. Curr Hypertens Rep. 2024;26(4):141–156. doi: 10.1007/s11906-023-01286-w. [DOI] [PMC free article] [PubMed] [Google Scholar]; Ordunez P, Campbell NRC, DiPette DJ, Jaffe MG, Rosende A, Martinez R, et al. HEARTS in the Americas: targeting health system change to improve population hypertension control. Curr Hypertens Rep. 2024;26(4):141–156. https://doi.org/10.1007/s11906-023-01286-w. [DOI] [PMC free article] [PubMed]
- 11.Doon R, Malcolm T, Lewis Y, Holder L, Gulston L, Hamid A, et al. Improving cardiovascular health with the patient-centered, integrated primary care HEARTS model in Trinidad and Tobago. Rev Panam Salud Publica. 2022;46:e169. doi: 10.26633/RPSP.2022.169. [DOI] [PMC free article] [PubMed] [Google Scholar]; Doon R, Malcolm T, Lewis Y, Holder L, Gulston L, Hamid A, et al. Improving cardiovascular health with the patient-centered, integrated primary care HEARTS model in Trinidad and Tobago. Rev Panam Salud Publica. 2022;46:e169. https://doi.org/10.26633/RPSP.2022.169. [DOI] [PMC free article] [PubMed]
- 12.Mendoza Z, Khan R, Maharaj RG, Motilal MS. Hypertension control applying the HEARTS in the Americas protocol in Trinidad: A cross-sectional study. Rev Panam Salud Publica. 2025;49:e123. doi: 10.26633/RPSP.2025.123. [DOI] [PMC free article] [PubMed] [Google Scholar]; Mendoza Z, Khan R, Maharaj RG, Motilal MS. Hypertension control applying the HEARTS in the Americas protocol in Trinidad: A cross-sectional study. Rev Panam Salud Publica. 2025;49:e123. https://doi.org/10.26633/RPSP.2025.123. [DOI] [PMC free article] [PubMed]
- 13.Londoño E, Gupta R, Van der Stuyft P, Heine M, Giraldo G, Ku GM, et al. HEARTS quality: a policy framework to strengthen hypertension and cardiovascular risk management in primary healthcare—insights from HEARTS in the Americas. Lancet Reg Health Am. 2025;53:101311. doi: 10.1016/j.lana.2025.101311. [DOI] [PMC free article] [PubMed] [Google Scholar]; Londoño E, Gupta R, Van der Stuyft P, Heine M, Giraldo G, Ku GM, et al. HEARTS quality: a policy framework to strengthen hypertension and cardiovascular risk management in primary healthcare—insights from HEARTS in the Americas. Lancet Reg Health Am. 2025;53:101311. https://doi.org/10.1016/j.lana.2025.101311. [DOI] [PMC free article] [PubMed]
- 14.Valdés González Y, Campbell NRC, Pons Barrera E, Calderón Martínez M, Pérez Carrera A, Morales Rigau JM, et al. Implementation of a community‐based hypertension control program in Matanzas, Cuba. J Clin Hypertens (Greenwich) 2020;22(2):142–149. doi: 10.1111/jch.13814. [DOI] [PMC free article] [PubMed] [Google Scholar]; Valdés González Y, Campbell NRC, Pons Barrera E, Calderón Martínez M, Pérez Carrera A, Morales Rigau JM, et al. Implementation of a community‐based hypertension control program in Matanzas, Cuba. J Clin Hypertens (Greenwich). 2020;22(2):142–149. https://doi.org/10.1111/jch.13814. [DOI] [PMC free article] [PubMed]
- 15.Rosende A, DiPette DJ, Martinez R, Brettler JW, Rodriguez G, Zuniga E, et al. HEARTS in the Americas clinical pathway. Strengthening the decision support system to improve hypertension and cardiovascular disease risk management in primary care settings. Front Cardiovasc Med. 2023;10:1102482. doi: 10.3389/fcvm.2023.1102482. [DOI] [PMC free article] [PubMed] [Google Scholar]; Rosende A, DiPette DJ, Martinez R, Brettler JW, Rodriguez G, Zuniga E, et al. HEARTS in the Americas clinical pathway. Strengthening the decision support system to improve hypertension and cardiovascular disease risk management in primary care settings. Front Cardiovasc Med. 2023;10:1102482. https://doi.org/10.3389/fcvm.2023.1102482. [DOI] [PMC free article] [PubMed]
- 16.Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39(33):3021–3104. doi: 10.1093/eurheartj/ehy339. [DOI] [PubMed] [Google Scholar]; Williams B, Mancia G, Spiering W, Agabiti Rosei E, Azizi M, Burnier M, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39(33):3021–3104. https://doi.org/10.1093/eurheartj/ehy339. [DOI] [PubMed]
- 17.Abreu APD. Hypertension in Latin America and the Caribbean: an analysis of recent progress and remaining challenges. J Bras Nefrol. 2025;47(3):e20240245. doi: 10.1590/2175-8239-jbn-2024-0245en. [DOI] [PMC free article] [PubMed] [Google Scholar]; Abreu APD. Hypertension in Latin America and the Caribbean: an analysis of recent progress and remaining challenges. J Bras Nefrol. 2025;47(3):e20240245. https://doi.org/10.1590/2175-8239-jbn-2024-0245en. [DOI] [PMC free article] [PubMed]
- 18.Downes M, O’Connor M, Olsson CA, Burgner D, Goldfeld S, Spry EA, et al. Causal inference in multi-cohort studies using the target trial framework to identify and minimize sources of bias. Am J Epidemiol. 2025;194(9):2685–2697. doi: 10.1093/aje/kwae405. [DOI] [PMC free article] [PubMed] [Google Scholar]; Downes M, O’Connor M, Olsson CA, Burgner D, Goldfeld S, Spry EA, et al. Causal inference in multi-cohort studies using the target trial framework to identify and minimize sources of bias. Am J Epidemiol. 2025;194(9):2685–2697. https://doi.org/10.1093/aje/kwae405. [DOI] [PMC free article] [PubMed]
- 19.Hamrahian SM, Maarouf OH, Fülöp T. A critical review of medication adherence in hypertension: barriers and facilitators clinicians should consider. Patient Prefer Adherence. 2022;16:2749–2757. doi: 10.2147/PPA.S368784. [DOI] [PMC free article] [PubMed] [Google Scholar]; Hamrahian SM, Maarouf OH, Fülöp T. A critical review of medication adherence in hypertension: barriers and facilitators clinicians should consider. Patient Prefer Adherence. 2022;16:2749–2757. https://doi.org/10.2147/PPA.S368784. [DOI] [PMC free article] [PubMed]
- 20.McEvoy JW, McCarthy CP, Bruno RM, Brouwers S, Canavan MD, Ceconi C, et al. 2024 ESC guidelines for the management of elevated blood pressure and hypertension. Eur Heart J. 2024;45(38):3912–4018. doi: 10.1093/eurheartj/ehae178. [DOI] [PubMed] [Google Scholar]; McEvoy JW, McCarthy CP, Bruno RM, Brouwers S, Canavan MD, Ceconi C, et al. 2024 ESC guidelines for the management of elevated blood pressure and hypertension. Eur Heart J. 2024;45(38):3912–4018. https://doi.org/10.1093/eurheartj/ehae178. [DOI] [PubMed]
- 21.Mohl JT, Moreno CA, Sadik K, Singhal M, Rooney A, Ciemins EL. Evaluation of blood pressure control, medication adherence, and therapeutic inertia in US patients with hypertension prescribed multiple antihypertensives. J Am Heart Assoc. 2025;14(12):e034787. doi: 10.1161/JAHA.124.034787. [DOI] [PMC free article] [PubMed] [Google Scholar]; Mohl JT, Moreno CA, Sadik K, Singhal M, Rooney A, Ciemins EL. Evaluation of blood pressure control, medication adherence, and therapeutic inertia in US patients with hypertension prescribed multiple antihypertensives. J Am Heart Assoc. 2025;14(12):e034787. https://doi.org/10.1161/JAHA.124.034787. [DOI] [PMC free article] [PubMed]
- 22.Parvanova A, Reseghetti E, Abbate M, Ruggenenti P. Mechanisms and treatment of obesity-related hypertension—Part 1: Mechanisms. Clin Kidney J. 2024;17(1):sfad282. doi: 10.1093/ckj/sfad282. [DOI] [PMC free article] [PubMed] [Google Scholar]; Parvanova A, Reseghetti E, Abbate M, Ruggenenti P. Mechanisms and treatment of obesity-related hypertension—Part 1: Mechanisms. Clin Kidney J. 2024;17(1):sfad282. https://doi.org/10.1093/ckj/sfad282. [DOI] [PMC free article] [PubMed]
- 23.Grassi G, Drager LF. Sympathetic overactivity, hypertension and cardiovascular disease: state of the art. Curr Med Res Opin. 2024;40(sup1):5–13. doi: 10.1080/03007995.2024.2305248. [DOI] [PubMed] [Google Scholar]; Grassi G, Drager LF. Sympathetic overactivity, hypertension and cardiovascular disease: state of the art. Curr Med Res Opin. 2024;40(sup1):5–13. https://doi.org/10.1080/03007995.2024.2305248. [DOI] [PubMed]
- 24.Loh HH, Tay SP, Koa AJ, Yong MC, Said A, Chai CS, et al. Hypertension in obstructive sleep apnea: the hidden role of renin–angiotensin–aldosterone system dysregulation. Trop Med Health. 2025;53(1):62. doi: 10.1186/s41182-025-00742-4. [DOI] [PMC free article] [PubMed] [Google Scholar]; Loh HH, Tay SP, Koa AJ, Yong MC, Said A, Chai CS, et al. Hypertension in obstructive sleep apnea: the hidden role of renin–angiotensin–aldosterone system dysregulation. Trop Med Health. 2025;53(1):62. https://doi.org/10.1186/s41182-025-00742-4. [DOI] [PMC free article] [PubMed]
- 25.Charchar FJ, Prestes PR, Mills C, Ching SM, Neupane D, Marques FZ, et al. Lifestyle management of hypertension: International Society of Hypertension position paper endorsed by the World Hypertension League and European Society of Hypertension. J Hypertens. 2024;42(1):23–49. doi: 10.1097/HJH.0000000000003563. [DOI] [PMC free article] [PubMed] [Google Scholar]; Charchar FJ, Prestes PR, Mills C, Ching SM, Neupane D, Marques FZ, et al. Lifestyle management of hypertension: International Society of Hypertension position paper endorsed by the World Hypertension League and European Society of Hypertension. J Hypertens. 2024;42(1):23–49. https://doi.org/10.1097/HJH.0000000000003563. [DOI] [PMC free article] [PubMed]
- 26.Yao T, Hu Q, Miao D, Xia Z, Li M, Wang X, et al. The effects of multimodal lifestyle interventions on blood pressure, body weight, and waist circumference in hypertensive patients: a network meta-analysis. Front Public Health. 2026;14:1720695. doi: 10.3389/fpubh.2026.1720695. [DOI] [PMC free article] [PubMed] [Google Scholar]; Yao T, Hu Q, Miao D, Xia Z, Li M, Wang X, et al. The effects of multimodal lifestyle interventions on blood pressure, body weight, and waist circumference in hypertensive patients: a network meta-analysis. Front Public Health. 2026;14:1720695. https://doi.org/10.3389/fpubh.2026.1720695. [DOI] [PMC free article] [PubMed]
- 27.Ordunez P, Rosende A, Brettler J, Londono E, Van Der Stuyft P, Martinez-Piedra R, et al. Readiness to deliver integrated cardiovascular, kidney and metabolic care in primary healthcare: phase II of HEARTS 2.0 in 26 countries in the Americas. BMJ Glob Health. 2026;11(1):e021298. doi: 10.1136/bmjgh-2025-021298. [DOI] [PMC free article] [PubMed] [Google Scholar]; Ordunez P, Rosende A, Brettler J, Londono E, Van Der Stuyft P, Martinez-Piedra R, et al. Readiness to deliver integrated cardiovascular, kidney and metabolic care in primary healthcare: phase II of HEARTS 2.0 in 26 countries in the Americas. BMJ Glob Health. 2026;11(1):e021298. https://doi.org/10.1136/bmjgh-2025-021298. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data available upon request. Data from the study may be requested from the Policy, Planning and Research Department of the South-West Regional Health Authority, Trinidad and Tobago.