By Katherine Hobson

At the beginning of his new book, “Overdiagnosed,” Dartmouth professor H. Gilbert Welch imagines all the medical diagnoses he might receive if he were given a thorough physical — borderline hypertension, overweight, benign prostate enlargement, degenerative joint disease and a few others — even though he generally feels pretty good.

Welch and co-authors Lisa Schwartz and Steven Woloshin, also of Dartmouth, write about the hazards of looking too hard for conditions or illnesses in healthy people, including additional procedures that carry no benefit but may cause harm, higher health-care costs and the psychological impact of being told you’re sick. We chatted recently with Welch. Here are edited excerpts:

What is overdiagnosis?

Overdiagnosis occurs when physicians make a diagnosis in an individual who would never go on to develop symptoms or die from the condition. it happens when we try to make diagnoses too early, in people who don’t have symptoms. I’m not saying we should never do that, but members of the general public have gotten the message that early diagnosis is always in your best interest, that it’s always good to look harder and find more. but the reality now is that we can find abnormalities in just about everyone and that can start a whole train of harmful events. so we all need to adopt a more balanced approach. Overdiagnosis doesn’t necessarily lead to overtreatment, but it often does.

How do you define health?

Health is much more than not being able to find something wrong. It’s how people feel, it’s a state of mind. and it’s hard to feel good when things are constantly being found wrong with you. but we are moving towards a [notion of] health that means the absence of any abnormalities. That’s not a good definition.

The book discusses overdiagnosis in a range of medical settings, including cancer. Where did the problem start?

I say it starts with hypertension, though some doctors might argue with me. but that’s the first place where we really started looking for abnormalities in a broad way. I’m not saying nobody with high blood pressure should be treated, I’m just saying it was an incredible paradigm shift to start looking for [problems] in people who were apparently well. In hypertension we had really persuasive data about the value of treating people with really high blood pressure. It’s incredibly important to treat them. The question is how we define abnormal and what warrants treatment. and we have progressively gotten interested in treating people with lower and lower blood pressure.

Is this problem acknowledged?

When I first wrote about the problem of overdiagnosis in cancer I got the sense that doctors thought I was crazy. but [the issue of overdiagnosis] in prostate cancer woke physicians up to this general problem.

So when your physician offers you a test, what should you do?

First get a sense of how you feel about this issue. some people prefer to look as hard as they can for anything that might cause problems in the future with the hope of avoiding those outcomes, understanding that the price they’ll pay for this is a vast increase in being told about things that won’t matter. that may mean unnecessary treatment and side effects. Other people may feel they’re well as long as they feel well and won’t [get screened and tested for everything], accepting the fact that perhaps they’ll miss some benefit of finding something early. see where you sit on this spectrum and communicate with your doctor.

When a specific test is being offered or suggested, ask why the physician is recommending it — that may lead to an interesting discussion. Ask whether the physician him- or herself has the test. and ask what randomized trials show about how it helps in the long run, how much evidence there is to support that and how many people are overtreated.