Contextual Variability in Health-Related Quality of Life Measurement: Psychometric Validation and Localized Factor Structure of the Tigrigna SF- 36 in Eritrean Hypertensive Patients

Abstract Background High-quality chronic disease management requires accurate patient-reported outcome measures, yet the structural transportability of conventionally-derived health-related quality of life (HRQoL) instruments remains unverified in many Sub-Saharan African contexts. This study aimed to linguistically and psychometrically validate the Tigrigna version of the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36) and investigate health related quality of life as well as the clinical and sociodemographic determinants of HRQoL among hypertensive patients in Asmara, Eritrea. Methods An analytical cross-sectional study was conducted between May and August 2025 among adult hypertensive patients (diagnosed > 1 year) attending four public health facilities in Asmara. Data were collected via face-to-face interviews using an electronic platform. To evaluate and construct validity, confirmatory factor analysis (CFA) was conducted on the standard US two-factor model. Because of significant fit divergence, exploratory latent structure analysis via principal axis factoring with oblique Promax rotation was performed on the eight subscales to identify a contextually derived latent architecture. Predictors of the extracted composite domains were evaluated using multivariate analysis of covariance (MANCOVA) followed by step-down multivariable linear regressions. Statistical analysis was performed in SPSS (v. 27) and R Studio (v. 4.4.3). Results Of 424 sampled individuals, 389 participated (91.7% response rate; median age: 66 years; 69.2% female). The overall Tigrigna SF-36 demonstrated good internal consistency (α = 0.928). Ceiling effects were observed in the Role Emotional (76.6%) and Social Functioning (64.3%) domains. The US-derived two-factor framework exhibited a suboptimal structural fit (CFI = 0.890, TLI = 0.837, RMSEA = 0.124). Conversely, a localized, data-derived two-factor solution emerged, classifying HRQoL into Emotional Well-Being (EWB) and General Health Perceptions (GHP), explaining 59.8% of the cumulative variance and revealing significant cross-loading behavior for Physical Functioning and Bodily Pain. MANCOVA revealed that age, presence and number of adverse drug events (ADE), duration of hypertension, and alcohol use significantly impacted joint outcomes ( p  < 0.05). In multivariable linear regression models, each additional reported adverse drug event (ADE) significantly reduced Emotional Well-Being (EWB) scores (B = -2.396, 95% CI: -3.332 to -1.460, p < 0.001), observed on General Health Perceptions (GHP) (B = -1.009, 95% CI: -1.935 to -0.083, p = 0.033). Conversely, longer duration since hypertension diagnosis significantly degraded GHP (B = -0.169, 95% CI: -0.277 to -0.061, p = 0.002) but had no effect on EWB (B = 0.003, p = 0.953). Conclusion While the Tigrigna SF-36 possesses good reliability, its latent covariance structure partially diverges from the conventional US model within this chronic disease cohort. The heavy psychological toll inflicted by ADE indicates that hypertension management paradigms must evolve beyond standard blood pressure targets to systematically monitor patient-reported adverse drug events and implement localized scoring frameworks.