After a series of studies found that the Bureau of Labor Statistics (BLS) Survey of Occupational Injury and Illness (SOII) was undercounting injuries and illnesses at work, the BLS funded further research to explore these findings. These additional studies were conducted in Massachusetts, California and Washington and used varied methodologies. The overall conclusion, across the three jurisdictions and across methodologies, was that the SOII significantly undercounts - and therefore underestimates - the number of injuries, even when looking only at objectively verifiable and often serious injuries such as amputations. These findings were published together in the American Journal of Industrial Medicine in October 2014. This article, written as the introduction to this set of BLS-funded papers, explores the possible reasons for under reporting and the negative consequences that flow from unreliable data in this field. The authors note that the SOII relies on unaudited employer-reported data, and there are significant disincentives to report injuries and illnesses. Inappropriate pressure or fear of retaliation (particularly for workers), ignorance (on the part of health care providers and workers), disincentives for employers (including the possibility of increased regulatory oversight) all contribute to the problem. For a wide range of reasons, few occupational diseases are reported. Under reporting means that the magnitude of the public health and safety issues as well as the real economic and human costs of occupational morbidity and mortality are hidden from regulators, health care providers, employers, unions, workers and the public at large. The authors conclude by presenting several actions the BLS could take to ameliorate the problem of undercounting.