While an inherited contribution is widely recognized for Alzheimer’s disease (AD), most risk prediction is based on close family history only. For a study published in Neurology, my colleagues and I integrated family history in both close and distant relatives to improve accuracy and utility of risk prediction.

We used a different approach to estimate an individual’s risk from disease from large groups of probands with the same extended family history by utilizing a genealogy database representing the population of Utah from its founders in the 1800s to modern day that is linked to multiple statewide disease data sources.

The analysis shows that the number of affected first-degree relatives (FDR), rather than solely the “presence or absence of any affected FDRs,” is necessary for appropriate risk prediction. Relative risk (RR) increased from 1.73 for one or more affected FDRs to 14.77 for four or more affected FDRs. A significant contribution based on affected second-degree relatives (SDR) was also noted; RR was 2.46 with at least three affected SDRs, even with no affected FDRs and was 21.29 with two SDRs affected in the presence of one affected FDR. Even if the closest affected relative was a third-degree relative (TDR) (eg, first cousin) risk was elevated; RR was 1.43 with three or more affected TDRs. Intriguing evidence suggested that risk for a male was higher than risk for a female with equivalent family history, and for maternal, versus paternal, family history.

These analyses demonstrate the significant role of affected relatives in risk prediction for AD and indicate the importance of completion of a full family history that extends beyond immediate family members. With knowledge of an individual’s family history, care providers are in an optimal position to educate on risk and recommend appropriate lifestyle changes for potential prevention.