When data becomes a weapon: How manipulating evidence can downplay drug-related harm

Data is intended to settle debates, providing insights into the seriousness of issues, affected populations, and necessary government actions. However, when data itself becomes detrimental, it becomes a weapon. One notable example involves the tobacco industry. Rather than proving smoking's safety, the industry aimed to generate enough doubt about the science, making uncertainty almost as effective.

In Kenya, the passage of the Tobacco Control (Amendment) Bill has ignited significant public and political controversy. The bill aims to tighten regulations on modern non-traditional tobacco products, including flavour bans, nicotine limits, and restrictions on digital marketing to safeguard young people. An investigation revealed a tobacco company lobbied for lesser health-warning requirements on nicotine pouches.

This case highlights the broader issue: manipulating what policymakers see, what consumers are informed, and shaping the risk perception. Research from the Tobacco Control Data Initiative and health organizations indicates that tobacco use incurs substantial economic losses in Kenya. For every dollar earned from tobacco, the economy suffers between KES 297 and KES 405 in healthcare expenses and lost productivity due to tobacco-related illnesses.

Examinations of confidential tobacco-industry files showed that companies nurtured connections with favorable scientists, funded seemingly independent organizations, promoted non-peer-reviewed research, and concealed unfavorable findings. Internal documents also disclosed that the industry was aware of nicotine's addictive nature, yet resisted accepting this fact publicly.

The broader lesson extends beyond tobacco: data can be manipulated without the need to fabricate numbers. Researchers might be chosen based on their favorable findings, while undesirable results could be omitted. Data ranges could be altered, short-term outcomes could be highlighted while long-term consequences ignored, and statistically significant associations could be dismissed, while favorable ones could be presented as conclusive.

Deceptively, uncertainty could be exaggerated until the public concludes "we simply do not know." The intention may not always be to deny harm directly; sometimes it's to make regulation appear harmful. If an industry can convince policymakers that taxation will result in a massive increase in illicit trade, the debate shifts from "How do we protect health?" to "Can we afford to regulate?"

The Kenya case underscores the importance of distinguishing industry data from independent evidence, even when the numbers are presented professionally. Moreover, tactics are evolving. In 2024, a major global tobacco company concealed funding arrangements to promote research and advocacy around heated tobacco products in Japan. The company denied the allegations, labeling its actions as legitimate regulatory engagement.

To combat misinformation, governments, researchers, and media must question: Who sponsored the study? Who designed it? What data were omitted? Can the methodology be independently replicated? Were alternative findings considered? Does the conclusion logically follow from the evidence? The tobacco story teaches us that misinformation isn't always a blatant lie; it can appear as a sophisticated spreadsheet, a credible expert, a carefully chosen statistic, or an authentic-sounding report.

This is why manipulated data is so dangerous: it does not merely distort what we know; it can distort the decisions governments make based on what we know.