Health economic evaluations, developed for the purposes of informing technology adoption decisions in publicly-funded health care systems, should strive to make use of all relevant evidence. Any failure to meet this objective will result in a partial representation of the evidence base and, consequently, there is a potential risk of obtaining misleading results. Unfortunately, a lack of comparability amongst the alternative measures of health-related quality of life (HRQoL) complicates the synthesis of this type of evidence. One solution to this problem is to specify a reference case measurement to promote comparability, although this may provide an incomplete representation of HRQoL effects or, in some cases, no evidence at all. The application of mapping functions - statistical algorithms that link HRQoL measures - might provide a means to incorporate a broader range of heterogeneous outcome measures for evidence synthesis. One method in particular, known as the common factor model (CFM), has been proposed in this regard due to its coherent mapping properties. Research involving the CFM has been conceptual to date and only a handful of case studies have ever been conducted. However, this method can be formulated as a structural equation model (SEM), an approach that has benefited from extensive application in other areas of research. The primary aim of this thesis is to investigate the plausibility of SEM methods serving as a generalised framework for the handling of HRQoL evidence. SEM methods are tested across scenarios involving aggregate data, individual patient data and a combination of both; in each case, a comprehensive synthesis of heterogeneous HRQoL outcomes using the SEM approach is compared against a restrictive synthesis involving a reference case measurement. In addition, the implications of these alternative approaches are explored from a decision-making viewpoint.