Finding SNV's associated with common diseases

I’m a primary care physician trying to integrate genomics into my clinical practice. A patient with early Alzheimer’s disease (so-called “Mild Cognitive Impairment”) with <1st percentile hippocampal volume and elevated serum amyloid submitted to whole genome sequencing. Every time I look up her non-synomous amino acid variants in OpenEvidence, I’m told heterozygotes don’t count (because the related disease is autosomal recessive), and homozygotes don’t count because the allele is common. But allele frequencies for common diseases must be >1-2% by definition! How can one solve a common disease (ie find the associated SNPs) if the system has been set up by pediatric geneticists to solve rare diseases?