| Characteristic |
Hypertension Awareness Group
|
p-value2 | ||||
|---|---|---|---|---|---|---|
| Overall N = 1601 |
Absent N = 491 |
Recognized N = 531 |
Under-recognized N = 381 |
Unclear N = 201 |
||
| VACS Index 2.0, mean (SD) | 60.2 (12.6) | 63.1 (11.9) | 60.4 (14.0) | 54.9 (10.4) | 62.7 (12.3) | 0.026 |
| Age (years), mean (SD) | 59.4 (7.6) | 59.9 (7.6) | 60.1 (8.8) | 56.2 (5.0) | 62.8 (5.8) | 0.004 |
| Sex, n (%) | 0.091 | |||||
| Male | 52 (32.5%) | 18 (36.7%) | 21 (39.6%) | 11 (28.9%) | 2 (10.0%) | |
| Female | 108 (67.5%) | 31 (63.3%) | 32 (60.4%) | 27 (71.1%) | 18 (90.0%) | |
| Marital status, n (%) | 0.514 | |||||
| Married/Cohabiting | 46 (28.8%) | 11 (22.4%) | 16 (30.2%) | 13 (34.2%) | 6 (30.0%) | |
| Single | 9 (5.6%) | 2 (4.1%) | 3 (5.7%) | 4 (10.5%) | 0 (0.0%) | |
| Widowed/Separated/Divorced | 105 (65.6%) | 36 (73.5%) | 34 (64.2%) | 21 (55.3%) | 14 (70.0%) | |
| Education level, n (%) | 0.646 | |||||
| No formal education | 34 (21.3%) | 13 (26.5%) | 9 (17.0%) | 10 (26.3%) | 2 (10.0%) | |
| Basic/Middle school | 102 (63.8%) | 28 (57.1%) | 35 (66.0%) | 24 (63.2%) | 15 (75.0%) | |
| Secondary/Technical/Tertiary | 24 (15.0%) | 8 (16.3%) | 9 (17.0%) | 4 (10.5%) | 3 (15.0%) | |
| Occupation, n (%) | 0.097 | |||||
| Full-time | 99 (61.9%) | 33 (67.3%) | 29 (54.7%) | 29 (76.3%) | 8 (40.0%) | |
| Retired | 4 (2.5%) | 2 (4.1%) | 1 (1.9%) | 0 (0.0%) | 1 (5.0%) | |
| Unemployed | 57 (35.6%) | 14 (28.6%) | 23 (43.4%) | 9 (23.7%) | 11 (55.0%) | |
| Below-median daily wage, n (%) | 0.600 | |||||
| Below median | 82 (51.3%) | 24 (49.0%) | 31 (58.5%) | 17 (44.7%) | 10 (50.0%) | |
| Above median | 78 (48.8%) | 25 (51.0%) | 22 (41.5%) | 21 (55.3%) | 10 (50.0%) | |
| BMI (kg/m²), mean (SD) | 26.0 (5.1) | 25.2 (4.3) | 26.4 (5.3) | 26.7 (5.1) | 25.9 (6.1) | 0.574 |
| BMI category, n (%) | 0.964 | |||||
| Normal (18.5–24.9) | 72 (45.0%) | 25 (51.0%) | 23 (43.4%) | 15 (39.5%) | 9 (45.0%) | |
| Underweight (<18.5) | 5 (3.1%) | 2 (4.1%) | 1 (1.9%) | 1 (2.6%) | 1 (5.0%) | |
| Overweight (25–29.9) | 50 (31.3%) | 14 (28.6%) | 17 (32.1%) | 12 (31.6%) | 7 (35.0%) | |
| Obese (≥30) | 33 (20.6%) | 8 (16.3%) | 12 (22.6%) | 10 (26.3%) | 3 (15.0%) | |
| Systolic BP (mmHg), mean (SD) | 142.4 (23.6) | 118.8 (9.1) | 162.8 (16.9) | 153.1 (14.8) | 125.9 (10.5) | <0.001 |
| Diastolic BP (mmHg), mean (SD) | 86.9 (14.0) | 75.1 (6.7) | 94.8 (12.3) | 94.9 (13.1) | 79.2 (6.2) | <0.001 |
| Diabetes mellitus, n (%) | 0.466 | |||||
| No diabetes | 93 (58.1%) | 30 (61.2%) | 30 (56.6%) | 23 (60.5%) | 10 (50.0%) | |
| Prediabetes | 46 (28.8%) | 14 (28.6%) | 14 (26.4%) | 11 (28.9%) | 7 (35.0%) | |
| Controlled diabetes | 4 (2.5%) | 0 (0.0%) | 2 (3.8%) | 0 (0.0%) | 2 (10.0%) | |
| Uncontrolled diabetes | 17 (10.6%) | 5 (10.2%) | 7 (13.2%) | 4 (10.5%) | 1 (5.0%) | |
| Dyslipidemia, n (%) | 70 (43.8%) | 24 (49.0%) | 24 (45.3%) | 15 (39.5%) | 7 (35.0%) | 0.682 |
| Renal disease (eGFR-based), n (%) | 0.552 | |||||
| No/Early CKD (eGFR ≥60) | 109 (68.1%) | 34 (69.4%) | 35 (66.0%) | 28 (73.7%) | 12 (60.0%) | |
| CKD (eGFR 15–59) | 50 (31.3%) | 15 (30.6%) | 18 (34.0%) | 9 (23.7%) | 8 (40.0%) | |
| CKD (eGFR <15) | 1 (0.6%) | 0 (0.0%) | 0 (0.0%) | 1 (2.6%) | 0 (0.0%) | |
| Depression (GDS-SF ≥5), n (%) | 0.718 | |||||
| Normal (0–4) | 98 (61.3%) | 31 (63.3%) | 34 (64.2%) | 23 (60.5%) | 10 (50.0%) | |
| Depression (≥5) | 62 (38.8%) | 18 (36.7%) | 19 (35.8%) | 15 (39.5%) | 10 (50.0%) | |
| HIV-1 RNA, median (IQR) | 19.1 (0.1, 21.5) | 19.1 (0.1, 38.4) | 19.1 (0.1, 20.2) | 0.1 (0.1, 19.1) | 9.6 (0.1, 21.3) | 0.173 |
| Viral suppression (<200 c/mL), n (%) | 0.641 | |||||
| Suppressed (<200 c/mL) | 155 (96.9%) | 47 (95.9%) | 51 (96.2%) | 38 (100.0%) | 19 (95.0%) | |
| Unsuppressed (≥200 c/mL) | 5 (3.1%) | 2 (4.1%) | 2 (3.8%) | 0 (0.0%) | 1 (5.0%) | |
| Polypharmacy (incl. HIV), median (IQR) | 2.0 (1.0, 3.0) | 2.0 (1.0, 2.0) | 2.0 (2.0, 3.0) | 1.5 (1.0, 2.0) | 3.0 (2.0, 3.0) | <0.001 |
| Medication adherence (HB-MAS), mean (SD) | 34.8 (1.9) | 34.8 (1.9) | 34.8 (1.6) | 34.8 (2.1) | 34.9 (1.9) | 0.873 |
| Alcohol use (AUDIT-C), n (%) | 0.460 | |||||
| No alcohol | 109 (68.1%) | 36 (73.5%) | 35 (66.0%) | 24 (63.2%) | 14 (70.0%) | |
| Low | 39 (24.4%) | 10 (20.4%) | 11 (20.8%) | 13 (34.2%) | 5 (25.0%) | |
| Moderate | 7 (4.4%) | 1 (2.0%) | 5 (9.4%) | 0 (0.0%) | 1 (5.0%) | |
| High/Severe | 5 (3.1%) | 2 (4.1%) | 2 (3.8%) | 1 (2.6%) | 0 (0.0%) | |
| Smoking status, n (%) | 0.233 | |||||
| Never | 152 (95.0%) | 47 (95.9%) | 48 (90.6%) | 38 (100.0%) | 19 (95.0%) | |
| Former | 8 (5.0%) | 2 (4.1%) | 5 (9.4%) | 0 (0.0%) | 1 (5.0%) | |
| Current | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | 0 (0.0%) | |
| 1 Mean (SD); n (%); Median (Q1, Q3) | ||||||
| 2 Kruskal-Wallis test for continuous variables; χ² test for categorical variables. p-values for variables with sparse cells (expected count < 5) should be interpreted with caution. | ||||||
Hypertension Awareness and VACS Index 2.0 Among Older People Living with HIV
KAHO Study — Results
1 Results
NoteImputation strategy
Missing data were handled using multiple imputation by chained equations (MICE) with m = 20 imputed datasets and 20 iterations. A root-component strategy was applied: the outcomes were not imputed directly; instead their constituent variables were imputed and the outcomes passively re-derived:
- VACS Index 2.0:
height_cm+weight_kg(pmm) →new_bmi→ full VACS 2.0 algorithm applied over all complete lab inputs (Hb, WBC, albumin, FIB-4, eGFR, HCV, viral load).
- Hypertension awareness:
has_a_doctor_nurse_ever_to(logreg) →htn_statusderived from imputed self-report + complete BP + antihypertensive use.
Eight additional variables with missing data were also imputed: occupation (7.5%), duration_of_diagnosis_yrs (5.6%), number_of_children (3.1%), is_nhis_valid (1.9%), smoking_status (1.2%), education (0.6%), height_cm/weight_kg (0.6%), and daily_wage (58.1% — descriptive only; interpret with caution).
The imputation model used 27 auxiliary predictors aligned with the study’s prior imputation comparison analysis, including three composite variables — any_hearing (right or left hearing problem), any_visual (right or left non-Normal vision), dm_flag (antidiabetic medication or HbA1c ≥ 6.5) — plus liver enzymes (s_ast, s_alt), nutritional status (mna_fs_score), and physical activity (physical_activity_who). Predictor assignments per variable are documented in Table 6.
Results were pooled across all 20 datasets using Rubin’s rules.
1.1 Participant Characteristics
A total of 160 participants were included in the analysis after multiple imputation (20 imputed datasets, 20 iterations). Table 1 presents sociodemographic, clinical, and HIV-related characteristics stratified by hypertension awareness group.
The mean age of participants was 59.4 years (SD = 7.6), and 108 (67.5%) were female. Nearly half (34/160, 21.2%) had no formal education. The mean VACS Index 2.0 score across the full sample was 60.2 points (SD = 12.6), indicating substantial predicted mortality risk burden in this population.
Note
χ² p-values for multi-level categorical variables stratified across four awareness groups should be interpreted with caution given sparse cells (expected count < 5 in some strata) with N = 160. Categories were pre-collapsed to reduce sparsity (marital status: 3 groups; education: 3 groups; depression: binary; renal disease: binary).
1.2 Hypertension Awareness
Four hypertension awareness categories were identified based on self-reported diagnosis, antihypertensive medication use, and measured blood pressure (≥140/90 mmHg):
- Absent (49/160, 30.6%): normal blood pressure, no self-report, no antihypertensive use.
- Recognized (53/160, 33.1%): elevated BP or antihypertensive use with self-reported diagnosis.
- Under-recognized (38/160, 23.8%): elevated BP without self-report or antihypertensive use.
- Unclear (20/160, 12.5%): on antihypertensives or self-reported diagnosis but with BP below threshold.
Prior to imputation, 12 participants (7.5%) could not be classified due to missing self-reported hypertension diagnosis (has_a_doctor_nurse_ever_to). This variable was imputed using logistic regression with predictors: age, sex, duration of HIV diagnosis, non-HIV polypharmacy, eGFR, systolic and diastolic BP, antihypertensive use, HbA1c, and viral load. Following imputation, all 160 participants were successfully classified.
1.3 VACS Index 2.0 by Hypertension Awareness
Figure 1 presents the distribution of VACS Index 2.0 scores across the four hypertension awareness groups. A Kruskal-Wallis test indicated a statistically significant overall difference in VACS scores across groups (H(3) = 9.22, p = 0.0264). Post-hoc Dunn pairwise comparisons with Bonferroni correction identified a significant difference between the Absent and Under-recognized groups (pBonferroni = 0.03).
Median VACS scores were: Absent 61 (IQR 55.3–70.6), Recognized 60.7 (IQR 49.1–69.8), Under-recognized 54.8 (IQR 47.5–62.7), and Unclear 64 (IQR 52.6–72.8).
1.4 Association Between Hypertension Awareness and VACS Index 2.0
Table 2 presents the results of five hierarchical linear regression models examining the association between hypertension awareness and VACS Index 2.0. All estimates are pooled across 20 imputed datasets using Rubin’s rules.
In the unadjusted model (Model 1), participants in the Under-recognized group had significantly lower VACS scores compared with the Absent group (β = -2.95, 95% CI: -7.2424 to 1.3424, p = 0.181). This association attenuated progressively with adjustment for sociodemographic factors (Model 2), behavioural and cardiometabolic covariates (Model 3), and anthropometric and renal variables (Model 4). The Recognized and Unclear groups did not differ significantly from the Absent group in any model.
| Characteristic1 |
Model 1 Unadjusted
|
Model 2 Sociodemographic
|
Model 3 Behavioural/Cardiometabolic
|
Model 4 Full DAG
|
Model 5 Healthcare access
|
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | |
| htn_awareness | |||||||||||||||
| Absent | — | — | — | — | — | — | — | — | — | — | |||||
| Recognized | -2.48 | -7.34, 2.37 | 0.314 | -2.14 | -6.57, 2.30 | 0.342 | -0.30 | -4.54, 3.95 | 0.891 | 0.65 | -3.25, 4.55 | 0.741 | 0.84 | -3.05, 4.73 | 0.670 |
| Under-recognized | -8.10 | -13.46, -2.74 | 0.003 | -4.40 | -9.34, 0.54 | 0.080 | -3.66 | -8.34, 1.02 | 0.124 | -2.95 | -7.28, 1.38 | 0.181 | -2.85 | -7.17, 1.46 | 0.194 |
| Unclear | 0.12 | -6.36, 6.60 | 0.971 | -0.36 | -6.54, 5.82 | 0.909 | 1.57 | -4.41, 7.54 | 0.605 | 1.96 | -3.50, 7.43 | 0.478 | 2.13 | -3.30, 7.56 | 0.440 |
| Age (years) | 0.68 | 0.42, 0.94 | <0.001 | 0.75 | 0.50, 1.00 | <0.001 | 0.59 | 0.35, 0.83 | <0.001 | 0.58 | 0.35, 0.82 | <0.001 | |||
| Sex | |||||||||||||||
| Male | — | — | — | — | — | — | — | — | |||||||
| Female | -5.34 | -9.90, -0.78 | 0.022 | -6.54 | -10.95, -2.13 | 0.004 | -3.50 | -7.73, 0.72 | 0.103 | -3.09 | -7.34, 1.17 | 0.154 | |||
| education_3cat | |||||||||||||||
| No formal education | — | — | — | — | — | — | — | — | |||||||
| Basic/Middle school | -1.93 | -6.46, 2.60 | 0.402 | -3.30 | -7.58, 0.97 | 0.129 | -4.16 | -8.08, -0.23 | 0.038 | -4.48 | -8.42, -0.54 | 0.026 | |||
| Secondary/Technical/Tertiary | -6.52 | -13.01, -0.02 | 0.049 | -9.47 | -15.85, -3.10 | 0.004 | -8.60 | -14.64, -2.57 | 0.006 | -8.22 | -14.25, -2.19 | 0.008 | |||
| occupation | |||||||||||||||
| Full-time | — | — | — | — | — | — | — | — | |||||||
| Retired | 4.61 | -7.26, 16.49 | 0.443 | 4.71 | -7.30, 16.72 | 0.438 | 2.91 | -8.02, 13.85 | 0.598 | 3.36 | -7.54, 14.26 | 0.542 | |||
| Unemployed | 0.66 | -3.77, 5.08 | 0.769 | -0.47 | -4.69, 3.75 | 0.826 | -1.30 | -5.15, 2.56 | 0.507 | -1.14 | -4.98, 2.69 | 0.556 | |||
| marital_status_3cat | |||||||||||||||
| Married/Cohabiting | — | — | — | — | — | — | — | — | |||||||
| Single | -2.36 | -10.49, 5.77 | 0.566 | -2.21 | -10.11, 5.68 | 0.580 | -3.40 | -10.65, 3.85 | 0.355 | -3.26 | -10.56, 4.04 | 0.379 | |||
| Widowed/Separated/Divorced | 3.86 | -0.45, 8.16 | 0.079 | 3.58 | -0.58, 7.74 | 0.091 | 2.55 | -1.29, 6.39 | 0.191 | 2.76 | -1.14, 6.66 | 0.163 | |||
| audit_c_cat | |||||||||||||||
| No alcohol | — | — | — | — | — | — | |||||||||
| Low | -6.88 | -11.11, -2.65 | 0.002 | -5.44 | -9.36, -1.51 | 0.007 | -5.56 | -9.46, -1.65 | 0.006 | ||||||
| Moderate | -15.63 | -23.95, -7.31 | <0.001 | -12.61 | -20.31, -4.92 | 0.001 | -12.10 | -19.78, -4.41 | 0.002 | ||||||
| High/Severe | 1.62 | -8.28, 11.51 | 0.747 | 0.54 | -8.56, 9.64 | 0.907 | -0.89 | -10.20, 8.43 | 0.851 | ||||||
| diabetes_cat | |||||||||||||||
| No diabetes | — | — | — | — | — | — | |||||||||
| Prediabetes | -1.63 | -5.50, 2.23 | 0.405 | 1.24 | -2.46, 4.94 | 0.510 | 1.62 | -2.12, 5.35 | 0.393 | ||||||
| Controlled diabetes | -8.03 | -19.10, 3.04 | 0.154 | -8.16 | -18.32, 2.00 | 0.114 | -7.31 | -17.54, 2.93 | 0.160 | ||||||
| Uncontrolled diabetes | -7.85 | -13.38, -2.32 | 0.006 | -4.15 | -9.38, 1.08 | 0.119 | -3.64 | -8.88, 1.60 | 0.172 | ||||||
| gds_binary | |||||||||||||||
| Normal (0–4) | — | — | — | — | — | — | |||||||||
| Depression (≥5) | -1.86 | -5.38, 1.66 | 0.298 | -1.63 | -4.88, 1.61 | 0.321 | -2.04 | -5.30, 1.22 | 0.218 | ||||||
| dyslipidemia_any | |||||||||||||||
| No | — | — | — | — | — | — | |||||||||
| Yes | -1.90 | -5.51, 1.71 | 0.299 | 0.06 | -3.35, 3.47 | 0.972 | 0.32 | -3.10, 3.74 | 0.854 | ||||||
| new_bmi | -0.93 | -1.27, -0.58 | <0.001 | -0.97 | -1.32, -0.62 | <0.001 | |||||||||
| renal_binary | |||||||||||||||
| No/Early CKD (eGFR ≥60) | — | — | — | — | |||||||||||
| CKD (eGFR 15–59) | 3.24 | -0.22, 6.71 | 0.066 | 3.24 | -0.23, 6.72 | 0.067 | |||||||||
| CKD (eGFR <15) | 1.77 | -18.24, 21.78 | 0.861 | 1.21 | -18.78, 21.19 | 0.905 | |||||||||
| nhis_valid | |||||||||||||||
| Not valid | — | — | |||||||||||||
| Valid | -6.14 | -15.31, 3.04 | 0.188 | ||||||||||||
| Medication adherence (HB-MAS) | -0.55 | -1.41, 0.32 | 0.212 | ||||||||||||
| 1 β = unstandardised regression coefficient (VACS index points). Reference group: Absent (no hypertension, no antihypertensive, BP <140/90 mmHg). Model 2: adjusted for age, sex, education, occupation, marital status. Model 3: Model 2 + AUDIT-C, diabetes, depression (GDS-SF), dyslipidemia. Model 4: Model 3 + BMI + renal disease (eGFR-based). Model 5: Model 4 + NHIS registration + medication adherence (HB-MAS). | |||||||||||||||
| Abbreviation: CI = Confidence Interval | |||||||||||||||
| Characteristic1 |
M1 Unadjusted
|
M2 Sociodemographic
|
M3 Behavioural
|
M4 Full
|
M5 + Healthcare
|
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | Beta1 | 95% CI1 | p-value1 | |
| htn_awareness | |||||||||||||||
| Absent | — | — | — | — | — | — | — | — | — | — | |||||
| Recognized | -0.20 | -0.58, 0.19 | 0.314 | -0.17 | -0.52, 0.18 | 0.342 | -0.02 | -0.36, 0.31 | 0.891 | 0.05 | -0.26, 0.36 | 0.741 | 0.07 | -0.24, 0.37 | 0.670 |
| Under-recognized | -0.64 | -1.07, -0.22 | 0.003 | -0.35 | -0.74, 0.04 | 0.080 | -0.29 | -0.66, 0.08 | 0.124 | -0.23 | -0.58, 0.11 | 0.181 | -0.23 | -0.57, 0.12 | 0.194 |
| Unclear | 0.01 | -0.50, 0.52 | 0.971 | -0.03 | -0.52, 0.46 | 0.909 | 0.12 | -0.35, 0.60 | 0.605 | 0.16 | -0.28, 0.59 | 0.478 | 0.17 | -0.26, 0.60 | 0.440 |
| age | 0.40 | 0.25, 0.56 | <0.001 | 0.45 | 0.30, 0.60 | <0.001 | 0.35 | 0.21, 0.49 | <0.001 | 0.35 | 0.21, 0.49 | <0.001 | |||
| Sex | |||||||||||||||
| Male | — | — | — | — | — | — | — | — | |||||||
| Female | -0.42 | -0.78, -0.06 | 0.022 | -0.52 | -0.87, -0.17 | 0.004 | -0.28 | -0.61, 0.06 | 0.103 | -0.24 | -0.58, 0.09 | 0.154 | |||
| education_3cat | |||||||||||||||
| No formal education | — | — | — | — | — | — | — | — | |||||||
| Basic/Middle school | -0.15 | -0.51, 0.21 | 0.402 | -0.26 | -0.60, 0.08 | 0.129 | -0.33 | -0.64, -0.02 | 0.038 | -0.35 | -0.67, -0.04 | 0.026 | |||
| Secondary/Technical/Tertiary | -0.52 | -1.03, 0.00 | 0.049 | -0.75 | -1.25, -0.25 | 0.004 | -0.68 | -1.16, -0.20 | 0.006 | -0.65 | -1.13, -0.17 | 0.008 | |||
| occupation | |||||||||||||||
| Full-time | — | — | — | — | — | — | — | — | |||||||
| Retired | 0.36 | -0.57, 1.30 | 0.443 | 0.37 | -0.58, 1.32 | 0.438 | 0.23 | -0.63, 1.10 | 0.598 | 0.27 | -0.60, 1.13 | 0.542 | |||
| Unemployed | 0.05 | -0.30, 0.40 | 0.769 | -0.04 | -0.37, 0.30 | 0.826 | -0.10 | -0.41, 0.20 | 0.507 | -0.09 | -0.39, 0.21 | 0.556 | |||
| marital_status_3cat | |||||||||||||||
| Married/Cohabiting | — | — | — | — | — | — | — | — | |||||||
| Single | -0.19 | -0.83, 0.46 | 0.566 | -0.18 | -0.80, 0.45 | 0.580 | -0.27 | -0.84, 0.30 | 0.355 | -0.26 | -0.84, 0.32 | 0.379 | |||
| Widowed/Separated/Divorced | 0.31 | -0.04, 0.65 | 0.079 | 0.28 | -0.05, 0.61 | 0.091 | 0.20 | -0.10, 0.51 | 0.191 | 0.22 | -0.09, 0.53 | 0.163 | |||
| audit_c_cat | |||||||||||||||
| No alcohol | — | — | — | — | — | — | |||||||||
| Low | -0.54 | -0.88, -0.21 | 0.002 | -0.43 | -0.74, -0.12 | 0.007 | -0.44 | -0.75, -0.13 | 0.006 | ||||||
| Moderate | -1.24 | -1.89, -0.58 | <0.001 | -1.00 | -1.61, -0.39 | 0.001 | -0.96 | -1.57, -0.35 | 0.002 | ||||||
| High/Severe | 0.13 | -0.65, 0.91 | 0.747 | 0.04 | -0.68, 0.76 | 0.907 | -0.07 | -0.81, 0.67 | 0.851 | ||||||
| diabetes_cat | |||||||||||||||
| No diabetes | — | — | — | — | — | — | |||||||||
| Prediabetes | -0.13 | -0.44, 0.18 | 0.405 | 0.10 | -0.19, 0.39 | 0.510 | 0.13 | -0.17, 0.42 | 0.393 | ||||||
| Controlled diabetes | -0.64 | -1.51, 0.24 | 0.154 | -0.65 | -1.45, 0.16 | 0.114 | -0.58 | -1.39, 0.23 | 0.160 | ||||||
| Uncontrolled diabetes | -0.62 | -1.06, -0.18 | 0.006 | -0.33 | -0.74, 0.09 | 0.119 | -0.29 | -0.70, 0.13 | 0.172 | ||||||
| gds_binary | |||||||||||||||
| Normal (0–4) | — | — | — | — | — | — | |||||||||
| Depression (≥5) | -0.15 | -0.43, 0.13 | 0.298 | -0.13 | -0.39, 0.13 | 0.321 | -0.16 | -0.42, 0.10 | 0.218 | ||||||
| dyslipidemia_any | |||||||||||||||
| No | — | — | — | — | — | — | |||||||||
| Yes | -0.15 | -0.44, 0.14 | 0.299 | 0.00 | -0.26, 0.27 | 0.972 | 0.03 | -0.25, 0.30 | 0.854 | ||||||
| new_bmi | -0.37 | -0.51, -0.23 | <0.001 | -0.39 | -0.53, -0.25 | <0.001 | |||||||||
| renal_binary | |||||||||||||||
| No/Early CKD (eGFR ≥60) | — | — | — | — | |||||||||||
| CKD (eGFR 15–59) | 0.26 | -0.02, 0.53 | 0.066 | 0.26 | -0.02, 0.53 | 0.067 | |||||||||
| CKD (eGFR <15) | 0.14 | -1.44, 1.72 | 0.861 | 0.10 | -1.49, 1.68 | 0.905 | |||||||||
| nhis_valid | |||||||||||||||
| Not valid | — | — | |||||||||||||
| Valid | -0.49 | -1.21, 0.24 | 0.188 | ||||||||||||
| hbmas | -0.08 | -0.21, 0.05 | 0.212 | ||||||||||||
| 1 β_std = standardised coefficient. Continuous predictors (age, BMI, HB-MAS) z-scored within each imputed dataset. Categorical predictors on natural scale. | |||||||||||||||
| Abbreviation: CI = Confidence Interval | |||||||||||||||
| Category | β (95% CI) | β_std (95% CI) | p | |
|---|---|---|---|---|
| Unadjusted | ||||
| Recognized | -2.48 (-7.30, 2.33) | -0.20 (-0.58, 0.18) | 0.31383 | |
| Under-recognized | -8.10 (-13.42, -2.78) | -0.64 (-1.06, -0.22) | 0.00331 | ** |
| Unclear | 0.12 (-6.31, 6.55) | 0.01 (-0.50, 0.52) | 0.97114 | |
| + Sociodemographic | ||||
| Recognized | -2.14 (-6.53, 2.26) | -0.17 (-0.52, 0.18) | 0.3422 | |
| Under-recognized | -4.40 (-9.30, 0.50) | -0.35 (-0.74, 0.04) | 0.0803 | |
| Unclear | -0.36 (-6.49, 5.77) | -0.03 (-0.51, 0.46) | 0.9089 | |
| + Behavioural/Cardiometabolic | ||||
| Recognized | -0.30 (-4.51, 3.91) | -0.02 (-0.36, 0.31) | 0.89071 | |
| Under-recognized | -3.66 (-8.30, 0.98) | -0.29 (-0.66, 0.08) | 0.12434 | |
| Unclear | 1.57 (-4.35, 7.48) | 0.12 (-0.34, 0.59) | 0.60467 | |
| Full DAG | ||||
| Recognized | 0.65 (-3.21, 4.52) | 0.05 (-0.25, 0.36) | 0.74095 | |
| Under-recognized | -2.95 (-7.24, 1.34) | -0.23 (-0.57, 0.11) | 0.18053 | |
| Unclear | 1.96 (-3.45, 7.37) | 0.16 (-0.27, 0.58) | 0.47836 | |
| + Healthcare access | ||||
| Recognized | 0.84 (-3.01, 4.69) | 0.07 (-0.24, 0.37) | 0.66982 | |
| Under-recognized | -2.85 (-7.13, 1.42) | -0.23 (-0.56, 0.11) | 0.19351 | |
| Unclear | 2.13 (-3.25, 7.51) | 0.17 (-0.26, 0.59) | 0.43992 | |
| β = unstandardised (VACS points). β_std = standardised (SD units of VACS). *** p<0.001, ** p<0.01, * p<0.05. | ||||
1.4.1 Forest Plot
Figure 2 presents the unstandardised (left) and standardised (right) β coefficients for the three hypertension awareness categories (vs Absent) across all five regression models. The Under-recognized group consistently showed a negative association with VACS scores in Models 1–3, though this attenuated after adjustment for BMI and renal disease in Model 4.
2 Supplementary Material
2.1 S1. Missing Data and Imputation
2.1.1 Missing Data Pattern
| Variable | N missing | % missing | Method1 | Derived outcome |
|---|---|---|---|---|
| daily_wage2 | 93 | 58.1 | logreg | below_median_wage (descriptive) |
| has_a_doctor_nurse_ever_to | 18 | 11.2 | logreg | htn_status (HTN awareness) |
| occupation | 12 | 7.5 | polyreg | — |
| duration_of_diagnosis_yrs | 9 | 5.6 | pmm | — |
| number_of_children | 5 | 3.1 | pmm | — |
| is_nhis_valid | 3 | 1.9 | logreg | nhis_valid (Model 5) |
| smoking_binary | 2 | 1.2 | logreg | smoking_3cat (descriptive) |
| height_cm | 1 | 0.6 | pmm | new_bmi → VACS Index 2.0 |
| weight_kg | 1 | 0.6 | pmm | new_bmi → VACS Index 2.0 |
| education | 1 | 0.6 | polyreg | — |
| 1 pmm = predictive mean matching; logreg = logistic regression; polyreg = polytomous logistic regression. | ||||
| 2 ⚠ daily_wage has 58.1% missing — likely non-response rather than MAR. Imputed values used in Table 1 only; interpret with caution. | ||||
age gender marital_status b_p_systolic b_p_diastolic antihypertensive hba1c
50 1 1 1 1 1 1 1
68 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
9 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
6 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
0 0 0 0 0 0 0
egfrcr new_polypharmacy new_hiv_polypharmacy audit_c_scores new_viral_load
50 1 1 1 1 1
68 1 1 1 1 1
3 1 1 1 1 1
9 1 1 1 1 1
3 1 1 1 1 1
6 1 1 1 1 1
1 1 1 1 1 1
3 1 1 1 1 1
3 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
2 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
0 0 0 0 0
s_cholesterol s_hdl_cholesterol s_ldl_cholesterol s_triglyceride any_hearing
50 1 1 1 1 1
68 1 1 1 1 1
3 1 1 1 1 1
9 1 1 1 1 1
3 1 1 1 1 1
6 1 1 1 1 1
1 1 1 1 1 1
3 1 1 1 1 1
3 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
2 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
0 0 0 0 0
any_visual dm_flag s_ast s_alt mna_fs_score physical_activity_who height_cm
50 1 1 1 1 1 1 1
68 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
9 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
6 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
3 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
2 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 0
0 0 0 0 0 0 1
weight_kg education smoking_binary is_nhis_valid number_of_children
50 1 1 1 1 1
68 1 1 1 1 1
3 1 1 1 1 1
9 1 1 1 1 1
3 1 1 1 1 1
6 1 1 1 1 1
1 1 1 1 1 1
3 1 1 1 1 1
3 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
1 1 1 1 1 1
2 1 1 1 1 0
1 1 1 1 1 0
1 1 1 1 1 0
1 1 1 1 0 1
1 1 1 1 0 1
1 1 1 1 0 1
1 1 1 0 1 1
1 1 1 0 1 1
1 1 0 1 1 1
1 0 1 1 1 0
1 1 2 3 5
duration_of_diagnosis_yrs occupation has_a_doctor_nurse_ever_to daily_wage
50 1 1 1 1
68 1 1 1 0
3 1 1 0 1
9 1 1 0 0
3 1 0 1 1
6 1 0 1 0
1 1 0 0 0
3 0 1 1 1
3 0 1 1 0
1 0 1 0 1
1 0 1 0 0
1 0 0 1 0
2 1 1 1 1
1 1 1 1 0
1 1 0 1 0
1 1 1 1 1
1 1 1 1 0
1 1 1 0 1
1 1 1 1 1
1 1 1 0 1
1 1 1 1 0
1 1 1 0 1
9 12 18 93
50 0
68 1
3 1
9 2
3 1
6 2
1 3
3 1
3 2
1 2
1 3
1 3
2 1
1 2
1 3
1 1
1 2
1 2
1 1
1 2
1 2
1 4
145
2.1.2 Composite Predictor Variables
Three composite predictor variables were derived prior to imputation, following the specification in the study’s prior imputation comparison analysis:
| Composite variable | Definition | Source variables |
|---|---|---|
any_hearing |
1 if right OR left hearing problem present | rt_hearing_problem, lt_hearing_problem |
any_visual |
1 if right OR left vision non-Normal | rt_visual_impairment_category, lt_visual_impairment_category |
dm_flag |
1 if on antidiabetic medication OR HbA1c ≥ 6.5% | antidiabetic_med, hba1c |
All composite variables were complete (0 missing) and used as predictors only, not as imputation targets.
2.1.3 Predictor Specification
| Imputed variable | Method | Predictors |
|---|---|---|
| education | polyreg | age, gender, occupation |
| occupation | polyreg | age, gender, marital_status, number_of_children, education, duration_of_diagnosis_yrs, any_hearing, any_visual |
| duration_of_diagnosis_yrs | pmm | age, gender, education, occupation, new_viral_load, new_hiv_polypharmacy, egfrcr |
| number_of_children | pmm | age, gender, marital_status, education, occupation, duration_of_diagnosis_yrs |
| smoking_binary | logreg | age, gender, education, occupation, audit_c_scores |
| is_nhis_valid | logreg | age, gender, education, occupation, b_p_systolic, egfrcr, new_viral_load, new_hiv_polypharmacy |
| daily_wage | logreg | age, gender, education, marital_status, number_of_children, is_nhis_valid [occupation excluded — perfect separation with Retired] |
| height_cm | pmm | age, gender, weight_kg, mna_fs_score, b_p_systolic, audit_c_scores, s_cholesterol |
| weight_kg | pmm | age, gender, height_cm, mna_fs_score, physical_activity_who, b_p_systolic, b_p_diastolic, hba1c, dm_flag, s_ast, s_alt, s_cholesterol, s_ldl, s_hdl, s_triglyceride, audit_c_scores |
| has_a_doctor_nurse_ever_to | logreg | age, gender, duration_of_diagnosis_yrs, new_polypharmacy, egfrcr, b_p_systolic, b_p_diastolic, antihypertensive, hba1c, new_viral_load |
| Highlighted rows: root-component targets whose imputed values drive passive re-derivation of VACS Index 2.0 and hypertension awareness. | ||
2.1.4 MICE Convergence
2.1.5 Imputed Distributions
2.1.6 Sensitivity: MICE vs Complete-Case Analysis
| Term |
MICE (n = 160, m = 20)
|
Complete-case (n = 147)
|
||||
|---|---|---|---|---|---|---|
| MICE β | MICE SE | MICE p | CC β | CC SE | CC p | |
| Recognized | 0.65 | 1.97 | 0.741 | 1.06 | 2.06 | 0.606 |
| Under-recognized | -2.95 | 2.19 | 0.181 | -1.19 | 2.28 | 0.604 |
| Unclear | 1.96 | 2.76 | 0.478 | 4.80 | 2.96 | 0.107 |
| β = unstandardised coefficient vs Absent. CC = complete-case analysis. | ||||||
2.2 S2. Model Diagnostics
Residual diagnostics were performed on Model 4 fitted to imputed dataset m = 1 as a representative check. Formal inference uses pooled estimates across all 20 datasets.
| Statistic | p-value | Conclusion |
|---|---|---|
| 0.9901 | 0.327 | No significant departure from normality |
| Term | VIF_M4 | VIF_M5 | ||
|---|---|---|---|---|
| htn_awareness | 1.15 | 1.15 | ✓ | ✓ |
| age | 1.48 | 1.49 | ✓ | ✓ |
| gender | 1.82 | 1.86 | ✓ | ✓ |
| education_3cat | 1.29 | 1.32 | ✓ | ✓ |
| occupation | 1.30 | 1.30 | ✓ | ✓ |
| marital_status_3cat | 1.21 | 1.24 | ✓ | ✓ |
| audit_c_cat | 1.18 | 1.19 | ✓ | ✓ |
| diabetes_cat | 1.14 | 1.17 | ✓ | ✓ |
| gds_binary | 1.15 | 1.18 | ✓ | ✓ |
| dyslipidemia_any | 1.31 | 1.33 | ✓ | ✓ |
| new_bmi | 1.38 | 1.45 | ✓ | ✓ |
| renal_binary | 1.16 | 1.18 | ✓ | ✓ |
| nhis_valid | NA | 1.16 | ✓ | ✓ |
| hbmas | NA | 1.19 | ✓ | ✓ |
| VIF > 5: moderate; VIF > 10: severe. Model 5 adds nhis_valid + hbmas. | ||||
| Model | R² | Adj. R² | Residual SE | F statistic | df | p (F-test) |
|---|---|---|---|---|---|---|
| Model 1: Unadjusted | 0.063 | 0.045 | 12.35 | 3.50 | 3, 156 | 0.0169 |
| Model 4: Full DAG | 0.523 | 0.446 | 9.40 | 6.83 | 22, 137 | <0.001 |
| Model 5: + Healthcare | 0.534 | 0.451 | 9.36 | 6.45 | 24, 135 | <0.001 |
| Single imputed dataset only. Not for formal inference. | ||||||
2.3 S3. Pooled SDs Used for Standardisation
| Variable | Pooled SD |
|---|---|
| score_v2 | 12.64 |
| age | 7.55 |
| new_bmi | 5.06 |
| hbmas | 1.86 |
| new_hiv_polypharmacy | 0.76 |
| audit_c_scores | 1.58 |
| SD pooled as the mean SD across 20 imputed datasets. | |
2.4 Session Information
sessionInfo()R version 4.5.2 (2025-10-31)
Platform: x86_64-apple-darwin20
Running under: macOS Ventura 13.7.8
Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/4.5-x86_64/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/4.5-x86_64/Resources/lib/libRlapack.dylib; LAPACK version 3.12.1
locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
time zone: Africa/Accra
tzcode source: internal
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] officer_0.7.5 readstata13_0.11.0 janitor_2.2.1 haven_2.5.5
[5] labelled_2.16.0 flextable_0.9.11 car_3.1-5 carData_3.0-6
[9] patchwork_1.3.2 ggstatsplot_0.13.5 gt_1.3.0 gtsummary_2.5.0
[13] miceadds_3.20-10 mice_3.19.0 lubridate_1.9.5 forcats_1.0.1
[17] stringr_1.6.0 dplyr_1.2.0 purrr_1.2.1 readr_2.2.0
[21] tidyr_1.3.2 tibble_3.3.1 ggplot2_4.0.2 tidyverse_2.0.0
loaded via a namespace (and not attached):
[1] RColorBrewer_1.1-3 rstudioapi_0.18.0 jsonlite_2.0.0
[4] shape_1.4.6.1 datawizard_1.3.0 correlation_0.8.8
[7] magrittr_2.0.5 SuppDists_1.1-9.9 TH.data_1.1-5
[10] estimability_1.5.1 jomo_2.7-6 cardx_0.3.2
[13] farver_2.1.2 nloptr_2.2.1 rmarkdown_2.31
[16] fs_2.0.1 ragg_1.5.1 vctrs_0.7.3
[19] memoise_2.0.1 minqa_1.2.8 paletteer_1.7.0
[22] base64enc_0.1-6 askpass_1.2.1 effectsize_1.0.2
[25] htmltools_0.5.9 broom_1.0.12 BWStest_0.2.3
[28] Formula_1.2-5 mitml_0.4-5 sass_0.4.10
[31] parallelly_1.46.1 htmlwidgets_1.6.4 sandwich_3.1-1
[34] cachem_1.1.0 emmeans_2.0.2 zoo_1.8-15
[37] uuid_1.2-2 commonmark_2.0.0 lifecycle_1.0.5
[40] iterators_1.0.14 pkgconfig_2.0.3 Matrix_1.7-4
[43] R6_2.6.1 fastmap_1.2.0 PMCMRplus_1.9.12
[46] rbibutils_2.4.1 future_1.69.0 snakecase_0.11.1
[49] digest_0.6.39 furrr_0.3.1 rematch2_2.1.2
[52] prismatic_1.1.2 textshaping_1.0.5 labeling_0.4.3
[55] timechange_0.4.0 abind_1.4-8 compiler_4.5.2
[58] fontquiver_0.2.1 withr_3.0.2 S7_0.2.1
[61] backports_1.5.0 DBI_1.3.0 ggsignif_0.6.4
[64] broom.mixed_0.2.9.7 pan_1.9 MASS_7.3-65
[67] openssl_2.3.5 tools_4.5.2 otel_0.2.0
[70] zip_2.3.3 statsExpressions_1.7.3 nnet_7.3-20
[73] glue_1.8.1 nlme_3.1-168 grid_4.5.2
[76] generics_0.1.4 gtable_0.3.6 tzdb_0.5.0
[79] data.table_1.18.2.1 hms_1.1.4 utf8_1.2.6
[82] xml2_1.5.2 ggrepel_0.9.7 markdown_2.0
[85] foreach_1.5.2 pillar_1.11.1 mitools_2.4
[88] splines_4.5.2 lattice_0.22-7 gmp_0.7-5.1
[91] survival_3.8-3 tidyselect_1.2.1 fontLiberation_0.1.0
[94] knitr_1.51 fontBitstreamVera_0.1.1 reformulas_0.4.4
[97] litedown_0.9 xfun_0.57 stringi_1.8.7
[100] yaml_2.3.12 boot_1.3-32 kSamples_1.2-12
[103] evaluate_1.0.5 codetools_0.2-20 gdtools_0.5.0
[106] multcompView_0.1-11 cli_3.6.6 RcppParallel_5.1.11-2
[109] rpart_4.1.24 xtable_1.8-8 parameters_0.28.3
[112] systemfonts_1.3.2 Rdpack_2.6.6 Rcpp_1.1.1-1.1
[115] globals_0.19.1 zeallot_0.2.0 coda_0.19-4.1
[118] parallel_4.5.2 rstantools_2.6.0 bayestestR_0.17.0
[121] Rmpfr_1.1-2 lme4_2.0-1 listenv_0.10.1
[124] glmnet_4.1-10 broom.helpers_1.22.0 mvtnorm_1.3-5
[127] scales_1.4.0 insight_1.4.6 rlang_1.2.0
[130] multcomp_1.4-30 cards_0.7.1