A Two-Stage Data Mining and Randomized Controlled Trial Approach to Evaluate the Gouteng Xiakucao Formula for Essential Hypertension Comorbid with Anxiety Disorder of Liver-Yang Hyperactivity Pattern

Abstract

Essential hypertension (EH) comorbid with anxiety disorder lacks optimal pharmacological options, and Traditional Chinese Medicine (TCM) formula construction has historically relied on classical theory rather than systematic data-driven evaluation. We addressed this gap through a two-stage workflow. Stage 1 was a retrospective data mining analysis of 114 EH–anxiety patients combining association rule, factor, and hierarchical cluster analysis of TCM prescriptions; it identified Liver-Yang Hyperactivity (LYH) as the predominant syndrome (52.63%) and Gouteng–Xiakucao as the highest-confidence herb pair (support 50.88%, confidence 94.83%), and anchored construction of the Gouteng Xiakucao Formula (GXF). Stage 2 randomized 128 patients with EH–anxiety comorbidity of LYH pattern 1:1 to valsartan plus GXF or valsartan plus oryzanol for 14 days. The GXF arm showed significantly greater post-treatment reductions than control in systolic blood pressure (134.03 ± 8.17 vs 137.18 ± 6.92 mmHg, P = 0.026), diastolic blood pressure (80.93 ± 3.88 vs 83.42 ± 4.31 mmHg, P = 0.001), Hamilton Anxiety Rating Scale score (11.63 ± 2.66 vs 12.76 ± 3.08, P = 0.036), Pittsburgh Sleep Quality Index (P = 0.015), and TCM syndrome score (P = 0.002); the SF-36 Physical Functioning, Social Functioning, and Mental Health subscales also improved more in the GXF arm, with no clinically meaningful adverse changes. This data-driven TCM formula derived from real-world prescription mining produced superior short-term outcomes compared with an oryzanol-based comparator in patients with EH–anxiety comorbidity of LYH pattern, illustrating a feasible paradigm for systematic TCM formula development and prospective validation.