Liz Wager: Are we making too much fuss about patient confidentiality?

Liz WagerIn my last blog, I addressed calls for raw research data to be made available. Like most other discussions about publishing data I started from the assumption that individual information must be kept confidential at all costs. That’s a helpful stance when considering the classic doctor-patient relationship but I wonder if it is always necessary when it comes to clinical trials.

Confidentiality in the surgery has various benefits, such as making people more honest and open when discussing their problems. Patients might be coy about their sex lives or drinking habits if they thought intimate details would be broadcast to the world. But how many people would object to the fact that, if a trial database were published, somebody might find out that they had migraine, acne or high blood pressure? In these cases the supposed harms of loss of confidentiality seem meagre compared to the potential benefits of being able to spot fraudulent research or, more likely, biased reporting.

I admit my radical solution wouldn’t work for trials in stigmatising conditions: people probably wouldn’t want their neighbours to know they’ve been treated for gonorrhea. But I just wonder if we could try, when obtaining patient’s consent to enter a study, to ask whether they would consent to the anonymised dataset being made public. People who didn’t consent would have to be excluded (admittedly a potential problem) as it doesn’t make sense only to publish a partial dataset – it has to be all or nothing or it can’t be checked or reanalysed. And it would be important to explain exactly how the data would be displayed and what efforts would be made to anonymise it.

But if somebody told me that a publicly available website describing me as [female, age 48, weight 59 kg, non-smoker, in Buckinghamshire] would also reveal that I had received placebo for a certain time in a trial for (let’s say) dyspepsia, I don’t think it would put me off entering the trial. And if it was explained that, by taking this apparently small risk, I could be helping ensure research integrity, I think I’d definitely go for it. What about you? Is patient confidentiality really sacrosanct or is it, perhaps, a smokescreen behind which those who don’t want their data scrutinised are hoping to hide?

Liz Wager PhD is a freelance medical writer, editor, and trainer. She is the current chair of the Committee on Publication Ethics (COPE).

  • I quite like your solution, Liz. Most people probably wouldn't object. Some people would, so the informed consent process would have to make it really clear what would happen, and that would be a challenge.

    I also think it should be possible in some circumstances to make data available without compromising confidentiality. Why would the dataset need to say that you were in Buckinghamshire? If you could take that sort of information off, and identify sites in a multicentre trial simply by their site numbers, then perhaps, particularly for common conditions, the data are so hard to link to individuals then for all practical purposes they could be considered completely anonymised anyway.

    Obviously for single centre trials, or trials of rare conditions, that would be more difficult, and someone (and ethics committee perhaps?) would need to make a judgement on a case-by-case basis whether there was any problem here.

    Either way, I think it would still be wise to get patient consent, but if the data were sufficiently anonymised, then that would be easier.

    Clearly there are problems here, but given the huge advantages of making datasets available, it's worth putting quite a bit of effort into solving them.

  • Avril

    As a patient with a rare medical condition I would be only to pleased to help out with any medical research. It would be great to improve research integrity as so much of it is being called into question making it difficult for Doctors to rely on.

  • Liz Wager

    Thanks — that's really interesting and confirmed my suspicion that some patients (even those with a rare condition who, in theory, might be more likely to be able to be identified even from anonymised records) are prepared to sacrifice confidentiality in the interests of research integrity … let's see if there are more like you!

  • Liz Wager

    I think patient consent is essential — but I'm encouraged by the response from one patient with a rare condition that she would not consider this a barrier to entering a study. You raise a good point that multicentre trials would probably make it easier to ensure anonymity than single centre.

  • Liz-
    Here in the States we have a law- HIPPA- that places all sorts of strictures on patient confidentiality. I am unsure if we could and how we would go about getting patients to waive their rights under HIPPA.

  • roseylisa

    I do think confidentiality has gone to far in the medical world. Especially when it comes to family, especially a husband or mom or dad grown children should know when their is something serious or a person that is in a comma or knocked out or annerysm etc. I heard even me as a wife would have to get a noterized paper from my husband done up ahead of time if something like that were to happen! It's gone to far!

  • Liz Wager

    Interesting point — I've come across the HIPPA legislation — but I do think patients might consider waiving their rights if the benefits were explained to them