OBJECTIVES: The US Panel on Cost-Effectiveness recommends excluding productivity losses from numerators of cost-effectiveness ratios. This critical literature review evaluated evidence as to whether individuals consider work loss when completing utility assessments reflected in the denominator. METHODS: PubMed/MEDLINE, Cochrane, and EMBASE databases were searched using terms such as “quality of life”, “productivity”, “utility”, and “EQ-5D”. Titles/abstracts of 1,114 identified studies were double screened using pre-defined criteria.  References from key articles were manually reviewed. Sixteen relevant publications were identified. RESULTS: Nine studies representing 6 countries (Australia (1), Japan (1), Netherlands (4), Sweden (1), Switzerland (1), United States (1)) asked respondents if they spontaneously considered income effects or believed study health states would impact income.  Between 31% and 64% of respondents in 6 studies indicated they considered lost income effects in health state valuations in situations where they had been given no instruction whether to consider income.  Two studies reported significantly lower proportions (Sweden: 4%; United States: 15%), while 1 study a higher proportion (Japan: 77%). Differences in study population (students versus general), health state severity, and utility elicitation method (time trade off versus visual analogue scale) may explain conflicting results.  The impact of cultural attitudes towards work and the social security net (e.g. wage replacement rates) is not clear.   No published studies shed light on whether respondents consider income loss when assigning utilities to the EQ-5D or other widely-used generic health tariffs. CONCLUSIONS: In many studies reviewed, a large proportion of respondents stated that they did not consider income loss during utility elicitation, suggesting income loss is not fully captured in health state valuations. Further research is needed to determine whether productivity losses are fully captured in QoL measurements for health scales such as the EQ-5D, or if the economic value of work loss should be included explicitly in cost-effectiveness analyses.