Last week, two high profile COVID-19 related observational studies published in renowned journals – The New England Journal of Medicine (NEJM) and the Lancet were retracted. The NEJM paper released on 1 May claimed that Angiotensin-converting-enzyme (ACE) inhibitors and angiotensin-receptor blockers (ARBs), both of which commonly prescribed to control one’s blood pressure, do not increase one’s risks for coronavirus and patients should carry on with their respective medications as advised.
Another published in the Lancet on 22 May dismissed the benefits of anti-malarial drugs – chloroquine and hydroxychloroquine, whether used alone or together with antibiotics – macrolide on patients with Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV-2), the causative agent of COVID-19. The retractions came at the request of the authors who said they had no direct participation in data collection and analyses. They are concerned about the veracity of these information as they were solely provided by health analytics company – Surgisphere.
A company in doubt
In a statement issued on the Lancet on 5 June, researchers who were involved in both studies – Dr. Mandeep Mehra of Brigham and Women’s Hospital; Dr. Frank Ruschizka of University Hospital Zurich, and Dr. Amit Patel of University of Utah wrote that Dr. Sapan Desai, Chief Executive Officer (CEO) of Surgisphere, refused to transfer “the full dataset, client contracts and the full ISO audit report” to their independent peer reviewers. Dr. Desai claimed the act would violate “client agreements and confidentiality requirements”, thus, peer reviewers were “not able to conduct an independent and private peer review” and a withdrawal has to take place.
The two retracted papers came under scrutiny shortly after their debut. Scientists and medical professionals were puzzled at the number of patients that were involved and the kind of demographics and clinical details that were obtained. The retracted Lancet study said it had reviewed a record of 96,032 COVID-19 patients from 671 hospitals in six continents who were admitted between 20 December 2019 and 14 April 2020. Of which, more than 4000 patients came from Africa but as some scientists pointed out, it appears that the continent may not have such comprehensive electronic health record systems to collate data from the number of indicated patients.
Likewise, earlier on, British newspaper The Guardian also highlighted an error in the number of reported deaths in Australia as the retracted paper declared a higher number of deaths in Australian hospitals than the country’s official statistics. With that, the Lancet had issued a correction saying that number should be attributed to “Australasia” group as data from one of the Asian hospitals was accidentally added to the Australian pool. Nevertheless, there was no change in the overall findings.
On 28 May, James Watson, a statistician at Mahidol University, an autonomous research institution in Thailand, wrote an open letter to the Lancet and the study’s co-authors, to call for the release of hospital level data held and processed by Surgisphere. The letter was signed by over 200 clinicians and researchers. Many of these researchers also signed a separate open letter addressed to NEJM on 2 June; making a similar request. Even though Surgisphere was founded more than a decade ago but it has a sparse presence. It did not list any of its partner medical institutions or scientific advisory board. Physician and entrepreneur James Todaro of Blocktown Capital, an investment fund questioned in his blog post if Surgisphere has such a large database, why it hasn’t appear in any other peer-review papers before May 2020?
Besides, as Data Scientist Peter Ellis noted in his blog post, he found five Surgisphere employees on LinkedIn and all of them, apart from Dr. Desai, lacks a medical, scientific or data background. A company which relies on software to garner and maintain such large amount of sensitive information from healthcare institutions would require a lot more manpower. Besides, how can a small company arrive at data sharing agreements with so many hospitals in different regions that speak different languages, utilize different data recording systems and observe different data regulations?
Damage has been done
Both papers in dispute had attracted a significant amount of public attention because they fueled two continuous COVID-19 debates, especially the recent hype around chloroquine and its less toxic counterpart hydroxychloroquine. Although the duo is widely used in malarial treatments and their anti-viral effects are well studied, there is no systematic examinations to find out their effectiveness in combating COVID-19. Yet, US President Donald Trump has repeatedly endorsed the medication in various public appearances. He even said that he has been hydroxychloroquine on a daily basis as a “preventive measure”.
Moreover, Brazil’s Health Ministry had issued new guidelines last Wednesday to use chloroquine and hydroxychloroquine in treating mild COVID-19 cases. UK also ignored scientists warning and bulk purchased 16 million hydroxychloroquine tablets in case the medication proves to be an effective treatment. Once this retracted Lancet study dismissing chloroquine and hydroxychloroquine’s impact on COVID-19 treatment was published, some of the large randomized trials, including SOLIDARITY, backed by the World Health Organization (WHO) were put to a halt.
Scientists believe the incident is a “wake-up call” as COVID-19 related studies are rushed to be published globally. Early on, the scientific community has been warned the hype around chloroquine and hydroxychloroquine may derail the search for effective treatments. Obviously, the warning has not been treated with care. Journals like the Lancet and NEJM often have the power to change clinical practice and certain research decisions. Having two papers with unreliable data appearing on these journals not only damage their reputations but also undermined the integrity and our trust in science.