Exploiting Real-World Data for Early Detection of X-Linked Hypophosphatemia

OBJECTIVES: Rare diseases are often chronic, degenerative, and take years to be accurately identified, hindering access to therapies and proper allocation of healthcare resources. A paradigmatic example is X-linked hypophosphatemia (XLH), a rare and progressive disease characterized by renal phosphate loss and bone mineralization defects. Early diagnosis of XLH is crucial for improving patients' quality of life and delaying debilitating clinical outcomes. However, diagnosis is delayed due to the rarity of XLH and the heterogeneity of clinical manifestations. This study aims to apply a predictive algorithm to identify individuals potentially affected by XLH.

METHODS: The algorithm uses three variables based on frequent medications, hospitalizations, and laboratory tests prescribed to individuals with XLH. In a previous retrospective analysis using administrative data from Italian Local Health Authorities (LHAs), covering 13 million beneficiaries, these variables were significantly associated with the disease. In the present study, a sample of LHAs comprising over 1.2 million beneficiaries was used to evaluate the applicability of the algorithm. Patients with XLH were identified using exception codes, while potentially affected individuals were identified using the algorithm.

RESULTS: Among the selected LHAs, 16 patients (mean age 33.6±22.4 years; 43.8% male) were confirmed with XLH (1.3/100,000 inhabitants), and 102 individuals (mean age 23.2±9.8 years; 50% males) met the algorithm's selection criteria and were considered as potentially affected by XLH. Among the 102 individuals, the most frequent hospitalizations were related to musculoskeletal and connective tissue diseases (91%), followed by neurological and digestive conditions (23%). 17.6% were hospitalized for fractures, and 52% received reimbursement for X-ray imaging of the knees, feet, or lower limbs.

CONCLUSIONS: In rare diseases, real-world data are essential for better understanding patient journeys and identifying patients. The results of this analysis highlight the potential of applying an algorithm to reduce diagnosis times and accelerate patient care to ensure timely access to treatments.