Figure 1
HEARTS package of strategies adapted to the Guatemalan Ministry of Health.
The pilot evaluated a package of five HEARTS-aligned implementation strategies in Guatemala to improve the primary care treatment of hypertension and diabetes (the ‘evidence-based intervention’). This figure shows how we adapted HEARTS to the Guatemalan Ministry of Health system based on work in our prior projects.
Figure 2
Flow diagram of study sites and patient participants.
Abbreviations: MOH, Ministry of Health.
Table 1
Baseline characteristics of patient participants.
| CHARACTERISTIC | VALUE |
|---|---|
| Patient participants enrolled, n | 964 |
| Age in years, median (IQR) | 55.2 (44.4 to 66.1) |
| Sex | |
| Women, n (%) | 760 (78.8) |
| Men, n (%) | 204 (21.2) |
| Ethnic group | |
| Maya Indigenous, n (%) | 486 (50.4) |
| Ladino/a or Mestizo/a, n (%) | 474 (49.2) |
| Unknown or other, n (%) | 4 (0.4) |
| Mayan linguistic community | 464 (48.1) |
| Health district | |
| Chiquimula, n (%) | 472 (49.0) |
| Sololá, n (%) | 492 (51.0) |
| Type of health facility | |
| Health center, n (%) | 837 (86.8) |
| Health post, n (%) | 127 (13.2) |
| Condition | |
| Hypertension only, n (%) | 567 (58.8) |
| Diabetes only, n (%) | 293 (30.4) |
| Hypertension and diabetes, n (%) | 104 (10.8) |
| Medication use among those treated for hypertension (n = 588) | |
| Number of antihypertensive medications, mean | 1.4 |
| Modified Therapeutic Intensity Score, meana | 0.8 |
| Enalapril, n (%) | 330 (56.1) |
| Losartan, n (%) | 199 (33.8) |
| Hydrochlorothiazide, n (%) | 299 (50.9) |
| Medication use among those treated for diabetes (n = 352) | |
| Number of glucose-lowering medications, mean | 1.6 |
| Medication Effect Score, meanb | 1.2 |
| Metformin, n (%) | 330 (93.4) |
| Glimepiride or glibenclamide, n (%) | 230 (65.3%) |
[i] Abbreviations: IQR, interquartile range.
aThis score is a combined metric of the number of antihypertension medications and dose intensity (43).
bThis score is a combined metric of the number of glucose-lowering medications, dose intensity, and expected HbA1c reduction for each medication (44, 45).
Table 2
Primary outcomes.
| MEASURE | DATA SOURCE AND SAMPLE | VALUE | PRESPECIFIED BENCHMARK |
|---|---|---|---|
| Feasibility (FIM) score, median (IQR)a | Surveys among n = 20 MOH participants | 5.0 (5.0 to 5.0) | ≥3.5 |
| Acceptability (AIM), median (IQR)a | Surveys among n = 20 MOH participants | 5.0 (4.8 to 5.0) | ≥3.5 |
| Percent of districts meeting enrollment goalb | Routine MOH data from n = 2 districts | 100% | 100% |
| Percent of patients with follow-up visit within 3 monthsc | Routine MOH data from n = 483 patients | 36% | ≥75% |
[i] aFeasibility of Intervention Measure (FIM) and Acceptability of Intervention Measure (AIM) scores are assessed on a scale of 1 to 5 with higher scores representing a greater degree of feasibility or acceptability, respectively. The use of median scores was prespecified in our protocol; mean FIM and AIM scores were 4.8 and 4.9, respectively.
bThe enrollment goal was ≥25 hypertension patients and ≥25 diabetes patients in each district.
cThis calculation was conducted among patients who were enrolled with ≥3 months remaining in pilot period.
Table 3
Secondary outcomes.
| MEASURE | DATA SOURCE AND SAMPLE | VALUE | CHANGE FROM BASELINE (95% CI) |
|---|---|---|---|
| Treatment rates | |||
| Hypertension treatment rate in month 6, n per month | Routine MOH data | 197 patients | See Figure 3 |
| Diabetes treatment rate in month 6, n per month | Routine MOH data | 108 patients | See Figure 3 |
| Disease controla | |||
| Blood pressure control (<130/80 mmHg) among patients with hypertension, % | Routine MOH data merged with electronic monitoring tool data | 50.6% | n/a |
| Glycemic control (FBG <115 mg/dl or RBG <160 mg/dl) among patients with diabetes, % | Routine MOH data merged with electronic monitoring tool data | 41.0% | n/a |
| Adoption | |||
| Facilities enrolling ≥1 patient, % | Routine MOH data from 10 facilities | 100% | n/a |
| Fidelity | |||
| Health worker training strategy: Health workers attending all trainings, n per district | Training attendance records | 20 health workers per district | n/a |
| Team-based care strategy: Facilities conducting monthly coordination meetings, %a | Monthly assessments in n = 10 facilities | 10% | –10% (–26% to 15%) |
| Team-based care strategy: Prescriptions by non-physician health worker, % | Routine MOH data from 1,341 total visits during pilot | 57% | n/a |
| Strategy to improve access: Availability of core medications, %b | Monthly assessments in n = 10 facilities | 81% | +21% (2% to 40%) |
| Strategy to improve access: Availability of core diagnostics, %b | Monthly assessments in n = 10 facilities | 82% | –5.0% (–25% to 15%) |
| Fidelity to electronic monitoring tool strategy: Visits captured in electronic monitoring tools (either DHIS2 or digitized chart data), % | Comparing routine MOH data with study data from electronic monitoring tools | 7.6% | n/a |
| Sustainabilityc | |||
| Leadership support, mean | Surveys among n = 20 MOH participants | 5.0 | n/a |
| Adequate staff to achieve goals, mean | Surveys among n = 20 MOH participants | 5.1 | n/a |
| Protocol easy for clinicians to use, mean | Surveys among n = 20 MOH participants | 5.5 | n/a |
| Integrated into MOH operations, mean | Surveys among n = 20 MOH participants | 5.4 | n/a |
| Defined roles and responsibilities, mean | Surveys among n = 20 MOH participants | 5.7 | n/a |
| Ongoing support, feedback, and training, mean | Surveys among n = 20 MOH participants | 4.7 | n/a |
| Usabilityd | |||
| DHIS2 system | Surveys among n = 10 MOH participants | 67.7 | n/a |
| Paper-based digitization system | Surveys among n = 8 MOH participants | 80.6 | n/a |
[i] Abbreviations: DHIS2, District Health Information System. FBG, Fasting blood glucose. n/a, not applicable. RBG, Random blood glucose.
aDisease control was only calculated among the subsample of 7.6% of patients whose visits were captured in the electronic monitoring tools (either DHIS2 or digitized chart data).
bCalculated as the mean monthly proportion across clinics over the 6-month pilot period. Core medications include: enalapril, losartan, hydrochlorothiazide, metformin, and glimepiride/glibenclamide. Core diagnostics include functioning glucometer, glucose test strips, and digital blood pressure monitor.
cSelect questions from the Program Sustainability Assessment Tool (PSAT) (38) and Clinical Sustainability Assessment Tools (CSAT) (39) are assessed on a scale of 1 to 7 with higher scores representing a greater degree of agreement.
dThe System Usability Scale (41, 42) is assessed on a scale of 0 to 100 with higher scores representing a greater degree of usability.
Figure 3
Monthly treatment rates.
Data underlying these figures were obtained from the Guatemala Ministry of Health. Lines reflect the single-group interrupted time series approach with segmented linear regression as described in the methods. The pre-post change in slope for the hypertension (panel A) and diabetes (panel B) treatment rates were 22.3 (95% CI: 16.2 to 28.4; P < 0.001) and 3.5 (95% CI: –1.6 to 8.7; P = 0.17) patients per month, respectively. Full results from models are provided in Appendix 8.