Can time-related patterns in diagnosis for hospital admission help identify common root causes for disease expression?

For many years medical admissions to acute hospitals have been increasing at a rate far higher than expected from demographic change. Analysis shows that the admissions tend to increase in a step-like manner at an interval of 3-6 years. This study characterises the specific diagnoses associated with the step-changes and uses the resulting pattern in admissions over time to identify further diagnoses with a far lower incidence which may also conform to this pattern. All of the diagnoses located using this method have a common immune function linkage in the expression of the chronic form of the condition. It is proposed that the wider use of data mining techniques may enable association between diagnoses associated with the need for inpatient care arising from unexpected common causes.