I think the vast majority of schizophrenics have relatively mild symptoms (just as most diseases have more sufferers of a mild version vs. more severe version)

I think the vast majority of schizophrenics have relatively mild symptoms (just as most diseases have more sufferers of a mild version vs. more severe version) and their diagnosis is made primarily because they have a disease that makes it hard for them to be 100% self sufficient, and there needs to be a paper trail in place that documents this. Although the majority of blind people have essentially 100% self sufficiency (maybe with some occasional help from family/friends on some edge cases) I suspect it is relatively easy for a congenitally blind person to get financial and other assistance in the US, since as a society we are fairly supportive of people with obvious congenital disabilities. I could therefore imagine that if there was mild schizophrenia in a congenitally blind person, there'd be far less incentive for the affected person or their family to establish a formal diagnosis to this effect. I wonder if this could be a partial explanation for this surprising statistic (and the lack of evidence for severe schizophrenia could maybe just be due to inadequate sample sizes in the OP study to detect this smaller set of people.)