Continuing from the previous post: I’d noticed in the People Also Ask data that adoptees and people without known biological family history seemed to be searching for BRCA testing in a noticeably different vocabulary than the rest of the population. This post is the cluster.
The query family
Once I started looking, the cluster was easy to find:
- BRCA test no family history
- BRCA test for adoptees
- can adoptees get BRCA tested
- BRCA panel without family history
- BRCA without medical records
- donor-conceived BRCA testing
These queries are doing something specific: they are pre-screening the system. The person searching has already encountered, somewhere, the standard family-history triage — the question that asks them to enumerate first-degree relatives and their cancer histories — and they have correctly intuited that the standard pathway might not apply to them. They’re searching for permission to proceed without the triage’s preconditions.
What the standard triage assumes
The standard pathway, broadly, looks like:
- Patient mentions a family history concern or has a personal cancer diagnosis.
- A genetic counselor takes a three-generation pedigree.
- The pedigree determines whether testing is indicated and which panel.
- Insurance coverage is largely contingent on (3).
This pathway works extremely well when the pedigree is available. It is the pathway that most clinical sites — major cancer centers, FORCE, the genetic counseling associations — describe in their public-facing pages.
The pathway fails silently when the pedigree isn’t available. Not catastrophically. Not in a way that leaves an obvious data trail in published clinical literature. It fails in the form of a person closing the browser tab and not coming back.
What helps
A few things, based on what existed when I went looking and what didn’t:
Explicit language. Pages that use the literal phrase “without family history” rank for the cluster. Pages that talk about “limited” or “uncertain” family history don’t. Adoptees do not search for “limited family history.” They search for “no family history” or “adopted.” Plain language matches plain searches.
The order of information matters. A page that leads with the family-history triage and then, fourteen paragraphs in, mentions that there are pathways for adoptees, is functionally invisible to this cluster. The information has to be the first thing the search-result snippet shows.
Explaining what insurance actually does. This is where I think most sites are weakest. The clinical guidelines on whether to test in the absence of family history are clearer than the insurance reality. People in this search cluster are already worried they will be told no. A page that explains which insurers cover testing without a pedigree, and what the typical out-of-pocket is otherwise, removes the bottleneck more effectively than any clinical reassurance.
What I’m building toward
This is going into the BRCA resource I’ve been working on. The page is going to live near the top of the resource, in the order someone arriving cold from a search engine would encounter it — not buried under a generic “who should test” tree.
The next post will probably be about the matrescence keyword cluster, which has its own very different shape. Or possibly about the schema markup decisions, which are less interesting but matter.