{
  "abstract": "Background and Importance Dolutegravir is a core component of modern HIV therapy, but long-term renal safety remains a clinical concern. Conventional monitoring based solely on estimated glomerular filtration rate (eGFR) may not reliably identify patients at high risk of clinically significant decline. Early risk stratification is critical for optimising surveillance and preventing irreversible kidney injury. A validated, multi-factor prediction tool integrated into a clinical application may enhance decision support in routine care.Aim and Objectives The aim was to develop and validate a multi-factor model for predicting renal function decline in patients initiating dolutegravir, and to design a prototype application for real-time patient risk classification.Material and Methods A retrospective cohort of adults initiating dolutegravir-based therapy between 2021 and 2023 was analysed. Patients were randomly divided into a training dataset (n=880) and a validation dataset (n=220). Logistic regression was used to identify predictors of a ≥25% decline in eGFR, calculated by the CKD-EPI equation. Model performance was compared with a model based on baseline eGFR alone using area under the receiver operating characteristic curve (AUC). A prototype application was developed to embed the validated model for clinical use.Results In total, 1,100 patients were included; 17.8% experienced a ≥25% reduction in eGFR. Five independent predictors were identified: age >40 years, body mass index >23 kg/m 2, CD4 count <400 cells/µL, baseline eGFR <90 mL/min/1.73m2, and alanine aminotransferase >40 U/L. The five-factor model demonstrated superior predictive performance compared with baseline eGFR alone (AUC 0.780; 95% CI 0.716–0.844 vs AUC 0.651; 95% CI 0.571–0.730). The application prototype enabled rapid patient-level classification and displayed tailored monitoring recommendations.Conclusion and Relevance The validated five-factor model improved renal risk prediction compared with eGFR alone. Its translation into a prototype application demonstrates a feasible approach to support hospital pharmacists and clinicians in proactive surveillance of patients receiving dolutegravir. Prospective validation and workflow integration will be required to confirm clinical impact and promote adoption in practice.References and/or Acknowledgements The authors thank the leadership team, Department of Pharmacy, HIV/AIDS clinic, and medical staff of Pakchongnana Hospital for their support, as well as the Faculty of Pharmacy, Thammasat University.Conflict of Interest No conflict of interest",
  "authors": [
    {
      "affiliations": [
        "Faculty of Pharmacy- Thammasat University- Thailand, Department of Pharmaceutical Care, Pathum Thani, Thailand"
      ],
      "name": "S Suriya"
    },
    {
      "affiliations": [
        "Pakchongnana Hospital, Pharmacy Department, Nakhon Ratchasima, Thailand"
      ],
      "name": "S Wanitchakorn"
    }
  ],
  "title": "4CPS-040 Multi-factor model for renal risk prediction in patients initiating dolutegravir: development, validation and application prototype",
  "uid": "791cd8db-913c-562e-8a2f-df79056b87ad"
}