Validating online approaches for rare disease research using latent class mixture modeling.

Area of Science:

• Endocrinology
• Genetics
• Psychology

Background:

• Rare disease research faces challenges due to patient geographic dispersion.
• Web-based recruitment and patient organization partnerships can overcome these barriers.
• Critics question the homogeneity and representativeness of online-recruited samples.

Purpose of the Study:

• To define patient clusters in congenital hypogonadotropic hypogonadism (CHH) using latent class mixture modeling (LCMM).
• To test the critique that online-recruited patients are a homogenous group with high needs.
• To demonstrate the validity and transferability of findings from online recruitment methods.

Main Methods:

• Latent class mixture modeling (LCMM) applied to online-recruited CHH patient data.
• Included patient demographics, clinical information, Revised Illness Perception Questionnaire (IPQ-R), and Zung Self-Rating Depression Scale (SDS).
• Analysis aimed to identify distinct patient subgroups based on these characteristics.

Main Results:

• LCMM identified three distinct patient subgroups (Classes I, II, III) in 154 CHH patients.
• Subgroups differed significantly in age, education, disease/emotional consequences, illness coherence, depression symptoms, and age at diagnosis.
• Later diagnosis was significantly associated with worse psychological adaptation and coping.

Conclusions:

• Three distinct patient classes were identified from online recruitment, refuting prior critiques.
• Findings support the validity of patient partnership and web-based recruitment for rare disease research.
• This study provides the first empirical data on negative psychosocial sequelae of a delayed diagnosis in CHH.