Maureen Williams was in her late 20s and on her way to becoming an opera singer in London when she began to feel exhausted. After a while, she stopped being able to reach the high notes that were her forte. “It just progressed from there until I wasn’t able to sing.”

She lost 13 kilograms and her skin turned dark. She no longer had enough strength to lift her toddler son. So she flew home to Sydney to see a doctor – who told her to rest. That can’t be right, she came away thinking. Another doctor diagnosed her with a virus and sent her off with a script. Over 18 months, eight doctors explained away her symptoms and even offered anti-anxiety medication. “I wasn’t being listened to, so I probably presented like someone who looked a bit anxious.”

As her health continued to spiral, she took herself to Royal Prince Alfred Hospital and asked to see a diagnostician, a doctor whose speciality is identifying complex conditions. He asked if she was usually so suntanned. “No, I normally burn in the sun,” she replied. The doctor looked stunned. Within hours, he’d confirmed she had Addison’s disease, which affects just 100 people a year in Australia. The disease, which stops people’s adrenal glands from producing hormones – and which can cause hyperpigmentation of the skin – can be fatal if untreated.

Williams never did sing again; the steroids she takes affect her vocal cords. Instead, she has worked to give patients a voice by helping them get second opinions about their hard-to-diagnose conditions and options for treatment. “My experience has taught me that patients need to step up and take more responsibility for their health and not abdicate it totally to the medical profession,” she tells us. So, how common are misdiagnoses? What good can come from reviewing your treatment options? When should you ask for a second opinion?

First, what’s the risk of misdiagnosis?

Emergency physician Carmel Crock arrived home from a shift at the Royal Melbourne Hospital one night when she became violently ill. Assuming she had food poisoning, she went back to hospital and admitted herself. “I really had two diagnoses in my head, which was gastro or maybe renal colic,” she recalls. But the doctor who saw her noticed her symptoms weren’t adding up and called for a surgeon, who found she had appendicitis. Crock had been thrown by not being in pain, which is the most common symptom with appendicitis. “It’s made me really good at diagnosing it since.”

Diagnosis, one of the cornerstones of medicine, can be a difficult process, says Crock, now head of emergency at the Royal Victorian Eye and Ear Hospital. “I’ve made diagnostic errors,” she says. “We just have to have humility that this is a difficult task that we’re doing, and we’re not always going to get it right. Sometimes it’s about finding the needle in the haystack and listening for the red flags.”

Diagnostic errors aren’t widely recorded. The Australian Commission on Safety and Quality in Health Care says it doesn’t track data on misdiagnosis or have plans to collect it, but it “remains abreast of developments in the sector”. There are challenges with how you define “error”: some would say it is a delayed, missed or wrong diagnosis, says Crock, where others contend it’s a “failure to establish an accurate, timely and communicated explanation of the patient’s problem – you have to communicate it to the patient”.

Another challenge is the nature of disease itself. Not all conditions “declare” themselves immediately, and a doctor might have only vague symptoms to work with until a clearer pattern allows them to join the dots. On top of this, many patients interact with different doctors and healthcare providers as their condition develops. “Diagnosis is more of a process than a one-point-in-time event,” says Stacy Goergen, a professor in the radiology department at Monash University. “Sometimes signs and symptoms evolve over time and make it easier to pinpoint a diagnosis some time after the initial presentation in the emergency department.” A patient might present with, say, a cough at one hospital and be told they have the flu, then go to a different hospital days later, where they receive a chest X-ray and are told they have pneumonia.

‘We talk about misdiagnosis as the bottom of the iceberg of patient safety and quality – it’s the hidden, very large sink of morbidity and mortality and harm to patients.’

Neurologist David Newman-Toker

David Newman-Toker, a world-leading expert on misdiagnosis from Johns Hopkins University in Baltimore in the US has tried to quantify the problem using data mostly from across the US, and some from Australia, Canada and Western Europe. He notes: “No single study can rigorously ascertain diagnostic errors and harms in a representative sample across all clinical contexts”. The figures he arrives at by extrapolating the data suggest errors occur in between 5 and 10 per cent of cases on average across hospital inpatients, emergency departments and primary health. “That’s probably still an underestimate,” he tells us. “If you really count every single delay in a diagnosis that doesn’t have an impact on a patient’s health, the number is probably a little higher.” Still, he notes, “the fact that doctors are able to correctly diagnose people most of the time is, in some sense, impressive. It’s a testament to the skills and capabilities of physicians that the error rates are this low.”

Newman-Toker, a neurologist, became interested in misdiagnosis when he was a hospital registrar in Boston. In one case, an aspiring Olympic ice-skater was left paralysed on one side by a stroke after doctors missed a damaged blood vessel in their neck. Another time, an elderly patient went blind when doctors failed to diagnose an autoimmune disease called giant cell arteritis, which could have been treated with steroids. “We talk about misdiagnosis as the bottom of the iceberg of patient safety and quality – it’s the hidden, very large sink of morbidity and mortality and harm to patients,” says Newman-Toker. Diagnostic error is “by far” the largest single category of medical errors, he says, “and probably bigger than all other medical errors combined. They’re terrible when they happen.”

Misdiagnosis is most likely in emergency departments and general practices, where doctors encounter the full spectrum of injury and disease. Estimates of how many diseases there are in the world vary between thousands and tens of thousands. It’s not entirely surprising, then, that rare conditions are more likely to be misdiagnosed. An aortic dissection – in which a tear in the aorta can cut off blood to vital organs – affects about two in 100,000 people a year but can be missed in an emergency room up to 30 per cent of the time, says Newman-Toker. But some quite common conditions can also be tricky to diagnose, too. Stroke, for example, is missed about 10 times more often than heart attack because patients don’t always have the classic symptom of numbness down one side of the body. They might just be dizzy, which is a symptom of any number of conditions, from dehydration to balance disorders.

‘You have to keep an open mind, be non-judgmental and be open to other people’s perspectives.’

Emergency physician Carmel Crock

Diagnostic error is far less likely in specialist medicine, he says, as “we operate at the edge of scientific knowledge all the time for this one problem – but that’s because we eat, sleep and breathe it”. Specialists also have the benefit of the frontline healthcare workers, such as GPs, making provisional diagnoses. Radiologists, for example, have an error rate of about 3 per cent worldwide, says Stacy Goergen. “The person who first sees the patient, often under stressful circumstances, has the job of synthesising the information that the radiologist [later] receives so, in a sense, there has already been a filtration process.” Radiologists can electronically transfer scan images to colleagues at other hospitals or interstate. “This gives us a big advantage over specialties who cannot easily send the patient interstate for a second opinion.” When they do err, it’s often because of high workload, or when they are tired or hungry, or repeatedly interrupted, says Goergen.

Radiologists might also be asked to revisit an image, she says, “when new information comes to light or the non-radiologist doctor wonders about a diagnosis we may not have considered”. In fact, collaboration over a diagnosis is widespread in the health system. Carmel Crock says that at the Eye and Ear Hospital, second opinions are “always happening” behind the scenes. Patients admitted overnight have their diagnoses cross-checked by a morning team to mitigate the fall in cognition we all experience during nighttime hours, and certain symptoms are considered red flags. “More and more, we think of diagnosis as this fuzzy, inexact science,” she says. “You have to keep an open mind, be non-judgmental and be open to other people’s perspectives.” The clues could come from anyone, she says, “from nurses, pharmacists, allied health and, of course, the patient and their loved ones”.

Indeed, some doctors see their work as akin to solving a mystery. Lisa Sanders, a physician and teacher at Yale Medical School, has authored a column in The New York Times for more than 20 years about medical mystery cases. Her writing inspired the TV drama House, starring Hugh Laurie, which she describes as “my column in the brain of a Sherlock Holmes-type of guy”. “Different diseases are going to have different rates of diagnostic errors,” she says from Connecticut. Some require a longer and more roundabout route to diagnosis than others. “So it takes a different number of wrong diagnoses to get to the right answer. But doctors can be extremely reluctant to buy into that.”

Most recently, Sanders has been the star of the Netflix show Diagnosis, which dramatises the outcomes of crowdsourcing difficult-to-diagnose patients. In one episode, the family of a six-year-old who suffers seizures seeks alternatives to surgery on the child’s brain. Information from people, including medical experts, from around the world eventually helps the family find a specialist who rules out the need for the surgery and implants a device instead. “In making the television show, I was [initially] reluctant to do crowdsourcing because I thought, that’s not really how medicine works,” Sanders tells us. “But the producers talked me into it, and I think it really shows something about medicine that I’m glad is out there. We, the doctors, certainly are more likely to make a diagnosis but by far are not the only people who make a diagnosis.”

So, when should you get a second opinion?

This brings us to the modern habit of consulting Dr Google and, now, AI bots. On the one hand, this can heighten anxiety about something benign or explain away what is actually a serious symptom. On the other hand, online research can prompt patients to ask doctors more nuanced questions. Patients have always sought informal second opinions after seeing doctors, notes Carmel Crock. “We ask family, we ask friends, we ask Aunt Betsy, who had it, ‘What was it like?’” So, her advice about the internet? “Make the most of all available information but have a trusted relationship with your clinician, and discuss it with them. And always try to keep to that middle path – it might be good information, or it might be misinformation you’re reading.”

‘How much do I push this? What if they’re wrong? What if I’m wrong? It’s not always clear until after all the dust settles.’

Emergency physician Carmel Crock

Two former patients we spoke with didn’t need online input to propel them to seek a second opinion. Andrew, in his 70s, went to his GP several years ago and asked for a precautionary heart check. The results led him to get an angiogram at a clinic, which found that a blood vessel in his heart had narrowed. A private cardiologist at the clinic told him he would put a stent in the artery, that most surgeons wouldn’t do the procedure but he was “very good”. “The confidence to do a procedure that was clearly difficult and that other doctors don’t do made me feel uncomfortable,” Andrew says. “I just wanted to check it out; it seemed too dangerous not to.” He asked his GP for a referral to a second private cardiologist who said the vessel wasn’t sufficiently blocked to warrant a surgical stent. Instead, Andrew manages the condition with medication and gets annual ECG tests.

Jeremy Linton-Mann, 78, from the NSW Central Coast, had a sore throat that kept getting worse. Within a few nights, he was struggling to breathe. He took himself to an urgent-care facility. A doctor told him it was a peritonsillar abscess, a type that forms near one of the tonsils, and gave him antibiotics; if they didn’t work, he should go to hospital. But Linton-Mann was worried. He went to see another GP who told him to go to hospital immediately. In the emergency department, a specialist told him the abscess could be obstructing his airway – and surgically removed it. “If I hadn’t got that second opinion, the alternative would not have been good,” Linton-Mann says. “I’d had that brief period the night before [when] I couldn’t breathe, and I just didn’t want to take a chance.”

‘If you really feel like you’re not being heard by your doctor … then I think it’s just logical to get a second opinion.’

Emergency physician Carmel Crock

But for many of us, it can be genuinely hard to know if we should seek a second opinion. “How much do I push this? What if they’re wrong? What if I’m wrong? It’s not always clear until after all the dust settles,” says Crock. “This is why it’s a fraught area.”

Any doctor-patient relationship, whether in primary care or specialist medicine, is built on trust. Newman-Toker says patients should feel comfortable to ask the questions they’re most worried about and expect the doctor to do their best to answer. “If the person gets bristly or upset at you for asking the question, you need a new doctor,” he says. “And if it sounds like they don’t know what they’re talking about, or they say just ‘because’ or something dismissive, then you need another doctor.” Carmel Crock agrees the biggest sign of potential trouble is not being listened to. “If you really feel like you’re not being heard by your doctor – if you’re concerned about something and your concerns are not being heard – then I think it’s just logical to get a second opinion.”

What about in cases of cancer?

Oncologist Ian Olver surveyed 355 people receiving cancer care in Australia between 2013 and 2015 and found a sixth of them sought second opinions – about half of them for reassurance they were on the right track with their diagnosis and treatment. “It’s not even necessarily that they don’t believe the first doctor,” he says. “And sometimes it’s not what you say but the way you say it that gives patients confidence.”

Many of those surveyed wanted another opinion to explore further treatment options. Sometimes patients have a choice between treatments that are equally effective. “If a small tumour can be treated curatively with radiotherapy or curatively with surgery then you genuinely have an option,” says Olver. “But there’s a range of side effects with each that you have to be aware of.” In that case, a patient might seek an opinion from both the radiotherapy and the surgical specialist, or other specialists in the field. “The common cancers, in general, have been more intensely researched particularly in terms of drug therapy. So, there are a large number of drug combinations that can be used in breast cancer,” says Olver, “but if you’ve got a neuroendocrine tumour, or something of the adrenal gland, there are very few options.” A patient’s oncologist can work as an independent voice among specialists involved in your treatment.

‘… if people take the time and have these conversations, their outcomes will often be better because they chose a treatment that is going to be right for them.’

Danielle Spence, Cancer Council Victoria

It can be difficult for cancer patients to seek second opinions early in the process, says Danielle Spence, who is the head of strategy at Cancer Council Victoria. “There’s that fear of death, that fear that says, ‘I’ve got to get cracking, I’ve got to get moving.’ But we know that if people take the time and have these conversations, their outcomes will often be better because they chose a treatment that is going to be right for them,” she says. “A good example is prostate cancer, where the focus might be on removing cancer but what might be important for the man is erectile preservation. If the treatment affects their life, relationship and sexuality afterwards, they might have wished they made a different decision or had been guided through conversations around what is important to them.”

Another reason to seek a second opinion is to understand treatment costs, particularly in the private system. Spence says someone with head and neck cancer might receive radiotherapy in the public system but possibly need dental work afterwards that can only be done in the private system. Or a breast cancer patient might have a mastectomy and face a long wait for a breast reconstruction so opt to have the surgery done in the private system. She says the costs can vary by thousands or even tens thousands of dollars, depending on the specialist. “People in their fear of what this cancer means to them will go with that first opinion and later on, they’ll say, ‘I wish I knew, I talked to someone in my support group and they saw someone down the road and paid out-of-pocket half of what I did.’”

And something to note: while national screening programs for cancer are “excellent and save lives”, says Spence, they only go so far for people with certain risk profiles. A mammogram, for example, might be less effective for people with dense breast tissue. Currently, the national breast screening program alerts people only in Western Australia and South Australia that they have dense breast tissue, although this is due to become the case in Victoria and NSW this year. Spence says people in this situation should discuss it with their doctor and consider getting more detailed scan such as MRIs.

So how do you go about getting a second opinion?

Lisa Sanders had a colonoscopy in her early 50s that showed up a tumour. The news was a “sidecar” to her life, not the main event. So she was shocked when a surgeon told her the cancer couldn’t be removed without her needing a colostomy bag. “That was really not appealing to me,” she says. She found another doctor at a leading cancer centre in New York via a friend. He told her he could do the surgery without the bag. “He said, ‘I’ve done it several times with very good outcomes.’ ‘Several’ is not necessarily the adjective, or the quantifier, you want to hear, but he said, ‘Even if it doesn’t work out for the immediate I can make it so it’s only temporary.’” Sanders, now 68, went with the second surgeon and today describes the brush with illness as “ancient history”.

‘It’s not that you want outside the box because outside the box is, like, craziness. What you want is a bigger box.’

Physician and columnist Lisa Sanders

The most important thing in seeking a second opinion, she says, is to find the people with deep expertise. In the real world, this looks nothing like the misanthropic “genius” doctor in House. “The people who are most likely to be good at a procedure are the people who’ve seen it before,” says Sanders. “It’s not that you want outside the box because outside the box is, like, craziness. What you want is a bigger box.”

If the second surgeon had given her the same advice as the first, she would have accepted it. “You have to be willing to go with what you hear.” She also says expertise can come from a primary care doctor who knows you best. “Just showing up in an office with a stack full of records seems to me to be a very tough way to get a diagnosis unless you’re going to a place that just does [difficult to diagnose diseases].”

“Patients seek second opinions on health care for a variety of reasons,” says Dr Ramya Raman, who is vice president of the Royal Australian College of General Practitioners. The group’s view is that “patients absolutely have a right to a second opinion, and requests are respected,” she says. “GPs will often sense when patients’ have doubt or concerns as to a diagnosis and management, and will suggest and initiate the second opinion. We will also often offer to provide a referral to a provider who is well-placed to give that opinion.”

Sanders didn’t tell the first surgeon she was getting a second opinion – “I got the news, he made his proposal, I was stunned” – but she recommends patients tell their doctor when they seek another view, especially if they have a trusting relationship with them. “It’s going to be OK. His feelings might be hurt, but that is the very definition of ‘not your problem’,” she says. “If the person comes up with a different opinion, he will say, ‘Well, thank god you did that.’ If they agree, he’ll say, ‘See, I told you.’” (Ian Harris, an orthopaedic surgeon and University of NSW professor, assures us doctors don’t feel slighted by a patient wanting a second opinion. “If the second opinion agrees with it, then it reinforces to the patient that this is the right path, and if the second opinion differs, that’s fine as well because then it’s up to the patient to decide which they want to go with. That’s great. I’d much rather that patient go and see someone they’re happy with.” )

When asking for a second opinion, a patient can sometimes use the referral they have to speak to another doctor at the same facility, or they can ask their GP to refer them to another specialist, or they can ask the specialist they’ve already seen for a referral to another specialist.This applies in both the public and the private health systems. In either, there can be delays, depending on the doctor and the hospital. “Some people think that the public system must be worse than the private system,” says Ian Olver, who was clinical director at the Royal Adelaide Hospital Cancer Centre from 1993 to 2006. “I don’t think that’s true, of course. In fact, I found it a lot easier to treat patients in the public section. I had everyone there.”

Maureen Williams felt dread when she was first diagnosed with Addison’s disease. “It’s scary to be told you have something which cannot be cured.” Then she felt relief that she finally had an answer. I was right, she recalls thinking. “Once you’ve got a definitive diagnosis, it’s really over to the doctors to give you the information for how you’re going to deal with it.” Today, she describes herself as a “frequent flyer” in emergency rooms. She’s become something of a lay expert on her condition, founding a support group and helping other patients with similar complex diseases. “Once you’re diagnosed with something as complicated as Addison’s disease, or MS or Parkinson’s or whatever it is, that’s the beginning of the journey. And I think it depends on the patient, how much they want to know, and how empowered they want to be to look after it themselves.”

Get fascinating insights and explanations on the world’s most perplexing topics. Sign up for our weekly Explainer newsletter.