Health policy as a fuzzy concept: methodological problems encountered when evaluating health policy reforms in an international perspective.

Investigating health policy reforms at a national level is a troublesome task, since it is difficult to establish exactly when a certain policy change took place and it is also difficult to determine the content of the reform. In this paper three main causes are distinguished that contribute to the 'fuzziness' surrounding reforms. Firstly, ordinary mistakes occur, even experts err. Secondly, in some countries responsibility for (part of the) health care system is delegated to the regional level, causing regional variations which lead to uninterpretable effects at country-level health care system descriptions. Thirdly policy reforms are often not a straightforward process. Implementations can be characterized as a gradual process which can be disturbed at any point in time. For example, the use of framework legislation, effects of public discussions and law enforcement problems contribute to the fuzziness. To strengthen the validity of information from secondary sources (often the best available and linguistically accessible sources) the use of primary sources and expert consultation highly recommended, respectively giving insight in formal rules and initial intentions of policy measures and in the impact of reforms at performance level. The fuzziness can be captured into an index (reform implementation index), containing the gathered information and thus facilitating statistical analyses controlling for process-information.