Hype, Spin or Accurate Reporting: The Quality of Reporting on Oncology News Website

Introduction: Oncology media websites such as Oncology Live (OncLive) and Targeted Oncology (TargetedOnc) play an important role in the dissemination of oncology news to patients and clinicians; however, the quality of the content on these websites has not been assessed. Our study aimed to systematically analyze content from these websites using previously validated criteria and assess financial conflicts of interest (FCOI) amongst speakers interviewed on these websites.

Methods: To establish a representative sample, all articles published on OncLive and TargetedOnc from October 1-31, 2021, were prospectively captured and analyzed. We included any news articles that discussed cancer treatments, including reviews, case discussions, and expert interviews, and included all cancers and treatment modalities. We did not include FDA press release articles, articles that narrated findings of a clinical trial without additional interpretation or analysis, abstracts from major meetings, or news articles not related to cancer treatment. The primary endpoint of our study was the quality of oncology news reporting in OncLive and TargetedOnc. Using a modified version of previously published scoring criteria, we analyzed the quality of news articles published in OncLive and TargetedOnc. We further assessed the financial conflicts of interest of authors and interviewees by capturing general payments from the Open Payments database for the years 2019 and 2020.

Results: 308 articles were reported on these websites during the study duration, of which 196 were included in our final analysis (OncLive 108, TargetedOnc 88). Table 1 lists characteristics of included studies, and Figure 1 highlights the flowchart for study selection.

FCOI disclosure was invariably absent in all the articles on both the oncology news websites. Independent experts (not associated with primary trial) were quoted in 47% (51/108) and 51% (44/86) of the articles discussed in OncLive and TargetedOnc. A direct link to the primary literature was provided in 15 % (16/108) of articles in OncLive and 17 % (13/75) in TargetedOnc. Where a link was provided, it was frequently directed towards a similar previous article on the respective websites. Sufficient information to trace the primary source of literature, accessible in < 5 minutes was provided in 93 % (100/107) and 61 % (54/88) of OncLive and TargetedOnc articles.

Study limitations, including study design, generalizability, and side effects were reported in only 7 % (7/105) of OncLive news articles and zero TargetedOnc news articles. Reporting of benefit/risks in absolute numbers was low at 28 % (28/99) for OncLive and 16 % (7/45) for TargetedOnc. Misleading headlines were seen in a small number of articles at 3% (3/100) and 3% (3/88). Emphasis was maintained throughout the headline and body in 45 % (49/108) and 57% (50/88) of articles in OncLive and TargetedOnc. Overgeneralization (either in headline or body) was seen in 57% (62/108) and 52% (46/88) articles in OncLive and TargetedOnc respectively.

The mean general payment received from industry by included United States physicians was found to be much higher ($63,861 and $39,639 for 2019 and 2020 respectively) than the average general payment for hematology-oncology physicians nationally ($8,326 and $6,007 respectively). A total of 75% (87/115) and 66% (76/115) of individuals for whom payment information was available on Open Payments received payments greater than $10,000 and 15% (18/115) and 8% (9/115) received payments more than $100,000 during 2019 or 2020.

Conclusion: In our analysis of oncology news websites, we demonstrate low quality of the content reported, bias and spin in how the results are presented, and a lack of disclosure of financial conflicts of interests amongst those interviewed. As these websites may serve an important role in bringing news to patients and health care providers, further steps are needed to improve the quality of the content reported.

No relevant conflicts of interest to declare.