Newspaper reporting on schizophrenia: a content analysis of five national newspapers at two time points.

AIM: This study aimed to assess change in the quality of reporting of schizophrenia in UK national daily newspapers, comparing 1996 with 2005.
METHODS: Five newspapers were searched using the PROQUEST database for articles published in 1996 or 2005 which contained the term 'schizo...'. 1196 articles were identified and rated against indicators of poor quality reporting. Derived from guidelines for media reporting on mental health the indicators were: metaphoric use of schizophrenia terms; use of stigmatising descriptors; use of equating descriptors; not including information putting the risk of violence into perspective in articles about violence; and use of the term 'release(d)' in articles about discharge from psychiatric hospital.
RESULTS: More recent year of publication was associated with a decrease in the odds of an article using a schizophrenia term metaphorically by a factor of 0.103 (95% CI 0.014-0.776), but this masked a pattern of decreasing use in broadsheets and increasing use in tabloids. The use of equating descriptors was significantly lower in 2005 in the univariate analysis (p<0.001), but this was no longer significant (p=0.558) when confounding variables were controlled for. There was no significant change in the use of stigmatising descriptors; the non-inclusion of information putting risk of violence into perspective, nor in the use of the term 'release(d)'.
CONCLUSIONS: There is little evidence that the quality of reporting of schizophrenia has changed over time. This suggests a need for the implementation of effective measures to bring newspaper reporting in line with current guidelines.