"Poverty in material terms is inextricably linked with poverty of [health] data" argues Peter Byass (Umeå Centre for Global Health Research, Sweden) this week in the Magazine section of open access journal, PLoS Medicine. He argues that there is a substantial lack of quantity and sometimes questionable quality of heath data from resource-poor settings, and warns of the double danger of "either publishing substandard research from poor settings because nothing better is available, or excluding results from poor settings from the literature because nothing is offered of sufficient quality." Byass also discusses several alternative strategies to improve the global representative span of health data.