Clinical evidence vs preliminary speculation in newspaper coverage of diabetes innovations: a quantitative analysis. (July 2018)

• Record Type: Journal Article
• Authors: Vehof, H.; Sanders, J.; van Dooren, A.; Heerdink, E.; Das, E.
• Is Part Of: Public health. Volume 160(2018)
• Page Start: 49
• Page End: 51
• Publication Date: 2018-07
• Subjects: Health journalism -- Information behavior -- Diabetes -- Medical innovations -- Evidence-based practice; Public health -- Periodicals; Public health -- Periodicals; Electronic journals; 362.1
• DOI: 10.1016/j.puhe.2018.03.022
• Languages: English
• ISSNs: 0033-3506

Abstract

Abstract: Objectives: Researchers have discussed that journalistic reporting of medical developments is often characterised by exaggeration or lack of context, but additional quantitative evidence to support this claim is needed. This study introduces a quantitative approach to assessing coverage of medical innovations, by aiming at provided references to observed clinical effects. Although observed clinical effects reflect increased chances for future medical applications, it is unknown to which extent newspaper articles refer to it when spreading health information. We aimed to assess, over a 6-year period, newspaper publication characteristics of diabetes innovations, arising from all scientific areas of interest, regarding the total count and the proportion of articles that provide references to demonstrated clinical efficacy. Study design: Quantitative content analysis of newspaper articles covering innovative treatments for diabetes. Methods: We performed a systematic review of newspaper articles between 2011 and 2016 printed in the largest six Dutch newspapers. By assessing in-article references, it was possible to quickly distinguish between (1) articles that referred to actual clinical efficacy demonstrated in a scientific setting and (2) articles that presented either predictions, fundamental research, preclinical research or personal experiences and recommendations. Proportion differences between scientific areas of interest were analysed using the chi-squared test. Results: A total of 613 articles were categorised. Total newspaper publication frequency increased with 9.9 articles per year ( P = .031). In total, 17% of the articles contained a reference to any proven clinical efficacy. Articles about human nutrition science (7%; P = .001) and (neuro)psychology (4.3%; P = .014) less frequently provided a reference to actual clinical efficacy. Conclusions: Our findings show that less than one in five newspaper articles about diabetes research contains a reference to relevant clinical effects, while the publication count is increasing. These statistics may contribute to feelings of false hope and confusion in patients. Highlights: The largest six Dutch newspapers together publish over 100 articles per year that cover innovative research, scientific prognoses and layman tips to treat diabetes. Only 17% of these articles refer to evidence for actual clinical effects in one or more diabetes patients. Articles discussing human nutrition and (neuro)psychology even less frequently refer to proven benefits, when compared with other scientific areas of interest.