England's National Health Service (NHS England) is planning to modify how it assesses the effectiveness of the controversial Palantir Technologies Inc. (NYSE: PLTR) data platform, following internal employee critiques that the analytical methods were flawed. The adjustments will target two of the most frequently cited core performance indicators: the statistical basis for claims that hospitals using the platform performed an "additional 110,000 surgeries" and experienced a "15% reduction in discharge delays."

According to internal emails, NHS England officials raised questions about the evaluation baseline, arguing that comparing the six months of surgical activity after the platform's launch to the six months prior could introduce significant bias due to seasonal fluctuations. Another official identified similar issues when reviewing the methodology for calculating discharge delays. On June 11, NHS England confirmed that the updated method description page had not yet incorporated the revised baseline, but pledged to include it in the next release. After the page was updated on June 12, the reference to the additional surgery baseline was removed. The next update is expected in September, at least 16 months after the page was first archived in May 2025.

A former head of operational research and evaluation at NHS England stated that the incident raises serious questions about the "analytical culture" within the health service system. To maintain public trust in NHS England's decisions and statements, it is essential to ensure its analyses comply with government guidelines. Previous reports had already identified a series of errors in the key underlying data sets, and research from the Health Foundation found "no significant improvement" in discharge performance attributable to the patient tracking tool.

In June of this year, NHS England quietly added a disclaimer to its methodology page, acknowledging that the data could not prove the effectiveness of Palantir Technologies Inc. (NYSE: PLTR). The UK's statistical regulator subsequently launched an investigation, and NHS England further revised its statements in late July. Internal emails reveal that the disclaimer was added after an official raised concerns in an email on May 28, emphasizing the need to state clearly that "this cannot definitively establish causality because not all variables have been controlled."

A Labour Party member of the parliamentary Health and Social Care Committee expressed shock that NHS England would publish claims about Palantir's positive impact on patients when its own employees had already questioned the methodology. He noted that the Palantir contract expires in March of next year, unless extended by the government, and questioned whether an extension is justified given the growing number of concerns. A Liberal Democrat member of the parliamentary Technology Committee called for an independent verification of the relevant data and questioned whether Palantir should remain within the NHS and other parts of the UK government.

Last week, an official government advisor criticized NHS England for failing to disclose that Palantir employees could access identifiable patient information, warning that this oversight could erode public trust. In response, NHS England stated that exchanging opinions and continuously reviewing and adjusting analyses is a standard part of the NHS's analytical practice, and that adjusting the baseline statistical calculations for institutions using the platform product for less than 12 months is a routine operation.