From the linked research article:

‘Over-diagnosis’ is observed when the prevalence of diagnoses made in clinical services, referred to as administrative prevalence (based on healthcare databases or insurance claims) exceeds prevalence estimates based on accurate assessments in representative population-based samples. Over-diagnosis may occur when diagnostic criteria are not applied with sufficient rigour, leading to false-positive cases. Over-diagnosis may also happen when people inappropriately self-diagnose. Notably, for individuals with milder or subclinical symptoms, a diagnosis can sometimes do more harm than good, creating stigma or leading to low-benefit treatments with significant side-effects.

So is Admin Prevalence > Prevalence Estimates where the estimates are made based on representative population-based samples?