The actual journal paper seems to be behind a subscription wall. But, here’s a summary…
Since the water quality of domestic wells is not federally regulated or nationally monitored, this study provides a unique, previously nonexistent perspective on the quality of the self-supplied drinking water resources used by 45 million Americans in the United States. This national reconnaissance study is based on a compilation of existing data from a very large number of wells sampled as part of multiple USGS programs.
Well water is not held to the same standards as municipal water, which means it is not normally tested for nasties such as arsenic.
Well (no pun intended!), looky here, but arsenic levels in well water exceed EPA Maximum Contaminant Levels (MCLs) more than 10% of the time. If, and this is a big if, you extrapolate these results to the whole country, as much as 5 million people may be exposed to higher than allowed arsenic levels. Arsenic is a notorious contaminant with an MCL of 0.01 mg/L, down a factor of 5 as of January 2006 due to data that indicates effects at even lower doses.
If I drank well water, I would get it tested for arsenic.
Most of the results are from the North East, which means that outside research circles (and behind subscription walls), groundwater arsenic levels could be a significant problem that not too many people are aware of.
All figures are from the paper.
Focazio, Michael J., Tipton, Deborah, Dunkle Shapiro, Stephanie & Geiger, Linda H. (2006) The Chemical Quality of Self-Supplied Domestic Well Water in the United States. Ground Water Monitoring & Remediation 26 (3), 92-104. doi: 10.1111/j.1745-6592.2006.00089.x