It also indicates that the conclusion is based on controlling for the major social confounding factors that are typically also viewed as key intermediary mechanisms by which genetic factors influence the outcomes of interest, including both own occupation and educational attainment and parental educational attainment.

Even if the underlying reality is that growing up around books is an effect of a shared hidden (genetic or otherwise) cause, the fact that it is a predictor even after for controlling for the identified alternative factors is quite a significant result.

(Though, as other comments point out, self-reporting may be a problem, though “people that are more successful than expected for their occupation, education, and parental education tend to overreport the number of books they grew up with compared to less successful individuals” would be an interesting result in its own way.)