Primum Non Nocere

Over the last month or so, I've been doing a little blogging about the experience of going through chemotherapy. My doctor says the treatments have been working and I have just one more cycle of chemo before the stem cell transplants. So my doctor's very happy and I know I should be happy too. But what I'm dealing with on a day to day basis is different from what she's addressing. She's battling the cancer itself. I'm facing the side effects of the treatments. My world is a world of unrelenting pain and anguish. The side effects, as I've written about before, are a universe unto themselves.

When my doctor decided to add a drug called bortezomib (velcade) to the chemo cocktail, she told me there was a 3% chance I could wind up with nerve damage (peripheral neuropathy). 3%... good odds. I mean that's like 97% chance you won't get neuropathy. But I was one of the 3%-- and the neuropathy is far more horrible than I could have ever imagined reading about it. A good friend of mine just happens to be a specialist in neuropathy. He's written the scholarly papers about it that other doctors read. And he's been advising me. His advice doesn't make me very hopeful. The most commonly used commercial drugs that are used to treat it have poor track records. Each of them has about an equal chance to lessen the pain by 50% as it does to make the condition significantly worse. Those are bad odds. Other drugs haven't been studied thoroughly enough, primarily because it isn't profitable for Big Pharma. Yesterday he sent me a NYTimes article by Austin Frakt and Aaron Carroll, If Patients Only Knew How Often Treatments Could Harm Them. It's certainly a lesson this patient has been learning-- and fast.

If we knew more, would we opt for different kinds and amounts of health care? Despite the existence of metrics to help patients appreciate benefits and harms, a new systematic review suggests that our expectations are not consistent with the facts. Most patients overestimate the benefits of medical treatments, and underestimate the harms; because of that, they use more care.

The study, published in JAMA Internal Medicine and written by Tammy Hoffmann and Chris Del Mar, is the first to systematically review the literature on the accuracy of patients’ expectations of benefits and harms of treatment. They examined over 30 studies that assessed whether patients understood the upsides or downsides of certain treatments. To a great extent, patients didn’t.

In the 34 studies that assessed understanding of benefits, patients overestimated their potential gain in 22 of them, or 65 percent. For instance, a 2002 study published in the Journal of the National Cancer Institute asked women who had undergone prophylactic bilateral (double) mastectomy to estimate how much the procedure reduced their risk of breast cancer. On average, the women thought they had reduced that risk from 76 percent to 11 percent, an absolute risk reduction of 65 percentage points.

For the more than 80 percent of the women in the study who did not have a BRCA genetic mutation-- which drastically increases the risk of breast cancer-- the real risk before surgery of developing breast cancer was 17 percent, meaning they greatly overestimated their risk reduction. Even the women with a BRCA mutation overestimated their risk reduction, but to a lesser extent.

Another 2012 study published in the Annals of Family Medicine asked patients to estimate the benefits of screening for bowel and breast cancer, and the use of medications to prevent hip fracture and cardiovascular disease. More than two-thirds of patients overestimated the benefits of medications to prevent cardiovascular disease, and more than 80 percent overestimated the benefits of medications to prevent hip fractures.

Further, 90 percent of patients overestimated the benefits of breast cancer screening, and 94 percent overestimated the benefits of bowel cancer screening. The researchers also asked the patients to estimate the minimum reduction in bad outcomes (like fractures or deaths) they would need to achieve to find the treatment worthwhile. For three of the four studied interventions, the minimum benefit patients would accept was higher than the actual benefit.

In the 15 studies examined in the systematic review for which harms were a focus, patients underestimated potential downsides in 10 of them (67 percent). For example, a study published in 2012 in the Journal of Medical Imaging and Radiation Oncology asked patients to estimate the risks associated with a CT scan. A single CT scan exposes a patient to the same amount of radiation as 300 chest X-rays, and carries with it a 1-in-2,000 chance of inducing a fatal cancer. More than 40 percent of patients underestimated a CT’s radiation dose, and more than 60 percent of patients underestimated the risk of cancer from a CT scan.

Why do patients err in assessments of risks and benefits? One reason could be that what they know is driven by the messages they hear. Doctors, direct-to-consumer ads and the media can skew our perceptions. They tend to focus on the benefits, but rarely quantify them. Health care centers, screening advocacy programs and pharmaceutical ads all push us to talk to our doctors about getting treatment without talking about actual gains.

Doctors also aren’t always good at communicating risks. A 2013 study published in JAMA Internal Medicine found that fewer than 10 percent of patients were told about overdiagnosis and overtreatment associated with cancer screening, even though 80 percent of patients wanted to know about harms.

This study, and others, indicate that patients would opt for less care if they had more information about what they may gain or risk with treatment. Shared decision-making in which there is an open patient-physician dialogue about benefits and harms, often augmented with use of treatment decision aids, like videos, would help patients get that information. However, a majority of patients still report that they prefer to leave medical decision-making to their doctors.

It might also be the case that some patients would use more of certain types of care if they had more information. Many chronic conditions remain undermanaged and undertreated in the United States. It’s possible that people with these conditions who had more information would use more care, which could raise spending for these patients but make them better off.

There’s also an argument to be made that people who overestimate the benefits of medicine to treat some conditions are more likely to take it regularly, which might lead to better outcomes, in some cases, than would occur if these patients were better informed.

Regardless, even though some patients may benefit somewhat from being ill informed, it seems wrong to argue that we should keep them in the dark. Many of the studies in the systematic review show that people report that they would opt for less care if they better understood benefits and harms. Improved communication could better serve patients and might improve the efficiency of our health system if patients focus on getting the types of care for which the benefit outweighs risk of harm.

It’s also possible that unrealistic expectations of care help patients cope with disease or provide them with some sense of control. Feeling hopeful about one’s future is not to be dismissed. But those unrealistic expectations don’t come cheap. We should at least consider the price that we pay for being uninformed.