The incidence of osteopetrosis, as implied by Schinke et al., is extremely rare — 1 in 250,000 according to this orphanet estimate: http://www.ojrd.com/content/4/1/5
With an allele frequency of 5% in whites, about 1 in 400 individuals would be homozygous for this variant — this is several orders of magnitude higher than the incidence of the disease. In addition, OMIM does not mention this variant — if it played a causal role in the disease it would almost certainly have been observed and well-reported on by now.
To establish significance for the observations of this variant in samples from the general population, we test it against a hypothetical variant with moderate penetrance (5%) for causing osteopetrosis in a recessive manner and responsible for 50% of cases (1 in 25,000 homozygous, 1 in 500,000 affected). Such a hypothetical variant would be expected to have an allele frequency 0.63%. Using the HapMap data: the chances of seeing this 30 times or more in a sample of 1,140 alleles is p=1.45 * 10^-10.