Doctors are often criticized for prescribing unneeded tests and procedures that harm more than they help and add to medical costs that could otherwise be avoided. Diagnostic tests are better than ever but they pick up conditions that might be perfectly harmless, forcing us to rethink when things are best left alone.

The ability of medicine to detect, heal and prevent ill health is undoubtedly among the crowning achievements of humanity. And those with access to state-of-the-art medicine are lucky indeed. In some respects, though, our faith in the increasingly sophisticated healthcare systems can blind us to a potential downside.

Unnecessary tests and occasionally overzealous healthcare systems in the West have created a crisis of overdiagnosis, according to H. Gilbert Welch, a professor of Medicine at the Dartmouth Institute in New Hampshire. Too much medical care is creating unnecessary risk, anxiety and treatment. What might be driving this? Widely held assumptions, Welch argues, that doctors and patients alike need to challenge to keep medicine in a healthy balance. Here are five of them.

Medical Myth 1: Earlier Diagnosis Is Always Better
When cancer-screening campaigns say early detection of cancer saves lives, they rarely emphasise the trade-off. "There is the potential to help some but you will hurt others," says Welch.

For some cancers, screening and prompt treatment are essential. But with others, you might not help at all. A major trial involving 400,000 men published in March found that a single screening for prostate cancer -- using a blood test for prostate-specific antigens -- didn't prolong lives, but did lead to additional worry and treatment.

Read more: How medicine got too good for its own good
As tests improve, medicine is increasingly finding non-progressing cancers, which wouldn't go on to cause harm. But doctors don't know for certain whether a cancer will spread, so treatment often follows. At best, cancer screening might reduce your risk of dying of, say, colon cancer, from 3 to 2 per cent, Welch says. At worst, it will lead to more tests, worry, and unnecessary surgery, chemotherapy or radiation treatment.

Medical Myth 2: CT Scans For Minor Head Injuries 
Approximately 50 percent of children who visit hospital emergency departments with head injuries are given a CT scan. CT scanning is associated with radiation exposure that may escalate future cancer risk. CT technology exposes patients to approximately 100 times the radiation of a standard chest X-ray which itself increases the risk of cancer.

The recommendation calls for clinical observation prior to making a decision about needing a CT. (American Academy of Pediatrics)

Medical Myth 3: Action is Always Better Than None
One thing about humans, says Welch, is that "we default towards doing something".

While prompt action often saves lives, sometimes no intervention may be the best thing. Lower-back pain is a classic example. Doctors have been increasingly ordering scans as a result of patient complaints of back pain, and finding signs of degeneration in the spine. But the same signs are found in older adults without back pain. Despite little evidence that operations improve quality of life in these cases, back surgeries have risen exponentially, and they come with risks of infection and post-operative stroke.

Why do many doctors go the intervention route? As Welch notes in his book Less Medicine More Health, when doctors do something, they'll either get credit for curing a patient, or credit for trying. If they do nothing, patients often complain that their concerns aren't being taken seriously. There's a money motive as well, particularly in the US, where doctors are paid more when they order more procedures.

Medical Myth 4: Mammograms Save Lives
Mammograms and breast screening have had no impact on breast cancer deaths and have actually been found to increase breast cancer mortality. With toxic radiation, mammogram testing compresses sensitive breast tissue causing pain and possible tissue damage. To make matters worse, the false negative and false positive rates of mammography are a troubling 30% and 89% respectively. 

Another concern is that many breast cancers occur below the armpits; however, mammography completely misses this axiliary region, viewing only the breast tissue compressed between two plates of glass. Considering these drawbacks, breast thermography should be given closer consideration. 

Thermography is a non-invasive and non-toxic technique which can detect abnormalities before the onset of a malignancy, and as early as ten years before being recognized by mammography. This makes it much safer and potentially life-saving health test for women who are unknowingly developing abnormalities, as it can take several years for a cancerous tumor to develop and be detected by a mammogram. 

Medical Myth 5: Newer Treatments Are Always Better
In the 1950s and 1960s, an estimated 2000 babies were born with birth defects in the UK, due to exposure to the drug thalidomide, prescribed for morning sickness. Many think regulation would prevent such a tragedy today, but regulators can only approve or restrict drugs on the basis of the evidence they have in front of them. And drugs are tested on a relatively small number of people, usually people without any health issues, and more often men. Studies might only run for a few months, but risks associated with new drugs might not be become apparent for years.

In the US, new drugs are often aggressively marketed to both doctors and the general public. A study published last year in the Journal of the American Medical Association found that of the 222 new drugs approved by the US Food and Drug Administration between 2001 and 2010, one-third resulted in a "safety event" after going on the market. Of those, the majority needed to include or improve communication about a risk, but three were recalled. The arthritis drug Vioxx (rofecoxib) was pulled in the US and the UK in 2004, after it was linked to cardiovascular issues. The drug's maker, Merck, paid settlements in about 27,000 lawsuits after people taking the drug had heart attacks and strokes.

Medical Myth 7: Inducing Labor or C-Section Before 39 Weeks
Delivery prior to 39 weeks is associated with increased risk of learning disabilities, respiratory problems and other potential risks. While sometimes induction prior to 39 weeks is medically necessary, the recommendation is clear that simply having a mature fetal lung test, in the absence of appropriate clinical criteria, is not an indication for delivery. (American College of Obstetricians and Gynecologists; American Academy of Family Physicians) 

Medical Myth 8: It Never Hurts To Get More Information
Finding out more about what ails you is a good thing -- if it's useful knowledge. At a time of big-data hype, people are confusing more information for useful knowledge, says Welch. Take genome-sequencing services that test for a range of disease risks. Most of the disease-linked genes we know about aren't that influential: they might increase the chance of getting a disease from 4 to 6 cent, for example. What is the use of such information?

People typically want tests to put themselves at ease, says Welch. But even as a test rules out one concern, it may spot another thing to worry about. MRIs and CT scans are increasingly picking up abnormalities that doctors weren't looking for -- they often don't matter, but do cause worry. The worry cycle doesn't need to start: estimates vary widely, but a significant proportion of high-tech medical imaging in the US is unnecessary.

Medical Myth 9: Lipid Profile Tests Prevent Cardiovascular Problems
Lipid Profile test checks various parameters of blood, such as cholesterol (good or high density lipoprotein as well as bad or low density lipoprotein) and triglyceride levels. Several scientific papers have proven that people with high so-called "bad" LDL cholesterol live the longest and there is noow a large number of findings that contradict the lipid hypothesis that cholesterol has to be lowered at all.

Medical Myth 10: It's All About Avoiding Death
Chemotherapy might prolong life in advanced cancers by a few months, but patients don't always want the additional time if it means dozens of extra medical appointments, nausea, fatigue and weakness. Still, chemotherapy is common at the end of life, particularly in the US. A 2016 study found that almost 40 per cent of people with cancer were given chemotherapy in the last six months of their lives, compared with less than 20 per cent in the Netherlands.

In some cases, medical care focused on avoiding death can even have the opposite effect. Welch points to a study of 151 people with advanced lung cancer -- those who received palliative care to ease their symptoms actually lived longer than those who received more aggressive medical care.