Self-diagnosis on the internet - a cure for all ills?
Apparently I'll be out of work soon - all my patients will be too busy diagnosing themselves over the internet to visit me. New statistics suggest that one in five of us is self-diagnosing rather than seeing a doctor - and while some of that is down to well publicised shortages of GPs and therefore GP appointments, it's not the only reason. Apparently nearly half of those surveyed said they would rather check the internet than miss a lie-in, and 38% wouldn't rearrange a shopping trip if they were offered a GP appointment that clashed with their social engagement.
The internet is like a curate's egg - good in parts. I use it hundreds of times a day - in fact, I rarely end a consultation without having directed a patient to at least one reliable website (almost always Patient, of course) so they can find out more about the diagnosis I've explained to them. But at the other end of the spectrum it can leave my patients totally fixated on an idea, and it's usually the worst-case scenario. These days, my patients often greet me with 'I'm sure I've got a brain tumour/multiple sclerosis/cancer because this website said…'. Just occasionally they're quite right - but far more often they've suffering agonies of picturing their imminent demise, when in fact they have a simple cold, or they've slept heavily leaning on a nerve etc.
Of course, it works both ways. Checking out bleeding from the bottom on the internet might tell you that you have nothing more to worry about than piles - when a rectal tumour is a rare but disastrous cause to miss.
I remember headlines in the British Medical Journal about a study showing that putting symptoms into a web search was more likely to come up with the right diagnosis than seeing a doctor. But this is where the adage 'garbage in, garbage out' applies. Firstly, you need to know which symptoms might be connected. For instance, if you were feeling breathless when you exercised, would you think to check if your ankles were swollen? Heart failure (where your heart doesn't pump blood efficiently enough round your body) can cause both. If the skin around your midriff became red, sore and itchy, would you know to tell Dr Google that you'd been wearing a belt against your bare skin a few days earlier? Contact dermatitis (a skin allergy) to nickel can keep getting worse after the culprit has been removed.
I decided to try out this self-diagnosis lark for myself, trying to put my medical knowledge to one side. I went onto a website I trust completely, for very good reason - it's written entirely by GPs for GPs and their patients, so it's exactly the information you'd get from a visit to your doctor. Using the Symptom Checker tool on Patient, I entered my name, gender, and a single symptom, headache - and pressed 'search'.
Migraine and sinusitis (common) came up first. But below them were eight other possible causes including three with 'red flags' indicating that they were medical emergencies - brain tumour, giant cell arteritis and intracranial haemorrhage. When I entered two other symptoms of the migraines I suffer from - sensitivity to light and to sound - migraine was the top diagnosis but others included rabies and delirium tremens (acute alcohol withdrawal). The website's answers were accurate but frankly terrifying - even as a doctor I found myself wondering briefly if maybe, just maybe, my next migraine could be something more sinister. The trouble is, they can't risk missing the bad news out because for one in 100 or maybe one in 1000 people it might be true. As a result, the other 99.9% of searchers get the same bad news.
As I would expect from a responsible website, the Patient symptom checker has a proviso at the top which reminded me that 'You may be alarmed by some of the suggested diagnoses. Based on the symptoms you entered they are possible, but most are very unlikely to apply to you. It's important that you carefully read about each result and speak to your doctor before reaching any conclusions'. But that's the problem with internet diagnosis - my eye kept being drawn back to the emergencies.
When I see a patient with headache, the worst case scenarios are all in my mind - meningitis, brain tumour, bleed into the brain. But I take a full history and do a full examination, and the first the patient hears of these diagnosis is when I reassure them that they have no signs or symptoms which suggest anything worse than tension headache etc.
So should we be advising patients to steer clear of the internet? Absolutely not - doctors see you for a few minutes, while you live with your condition 24-hours-a-day. For people with long-term conditions like asthma, heart disease, diabetes etc, it's essential that you take control of your own care, working with your healthcare team. Having access to reliable online information is invaluable.
But where self-diagnosis is concerned, be warned - Dr Google is entirely unpoliced. Even if the information is accurate, it doesn't come with a sugar coating and doom and gloom prevails. A study in the British Medical Journal this year concluded that 'Symptom checkers had deficits in both triage and diagnosis. Triage advice from symptom checkers is generally risk averse, encouraging users to seek care for conditions where self-care is reasonable.' In other words, you're more likely to end up scared witless than lulled into a false sense of security. Maybe I'm not going to be out of a job - but it's certainly going to be bad for my waiting times.