Why, when we spend such money on health care, do we not do better when we measure ourselves against other countries regarding preventing cancer deaths?  My theory as to how we came to this place is three-fold. First, the public has difficulty understanding small probabilities. Sometimes, when presented with a scenario, they see a 1% risk as very high (my 40 year old friend got breast cancer and you are telling me not to worry? I have a ONE PERCENT risk!). Sometimes they see it as very low (sure I smoke, but of all the people who smoke NINETY NINE out of 100 don’t get lung cancer). As a consequence, Americans tend to participate in risky behaviors, spend personal money ineffectively on preventive services and don’t demand that we as a country spend money effectively in aggregate. Secondly, Americans want value for their health care dollar and believe they are getting it (but believe that others are being wasteful):

The mismatch between reality and voters’ perceptions is most acute when voters consider their own health care:
■ 65% of voters think that most or nearly all of the health care they receive is backed up by scientific evidence, with 26% thinking that half or less is backed up by science;
■ However, confidence diminishes when voters are asked about the health care that “most people” receive. Only a slim majority, 51%, believes that nearly all or most care is backed up by good science.

So, Americans believe their OWN health care decisions are good but every one else is being wasteful. Lastly, physicians have incentive to suggest the use of expensive technology when a less expensive one might do. From a study about prostate cancer:

We found that in the early period of IMRT [a very expensive treatment for aggressive prostate cancer] adoption (2001–03) men with high-risk disease were more likely to receive IMRT, whereas after IMRT’s initial dissemination (2004–07) men with low-risk disease [more folks but much less likely to benefit from such aggressive care] had fairly similar likelihoods of receiving IMRT as men with high-risk disease. This raises concerns about over treatment, as well as considerable health care costs, because treatment with IMRT costs $15,000–$20,000 more than other standard therapies.

So, changes in cancer detection and treatment leading to reducing “wasteful” care may not occur until the incentives change. Even to a doctor, $20,000 is a lot of money.

The US Preventive Services Task Force has for the last 20 years been making  sense out of small number probabilities and demonstrating where the value lies in for those of us caring for people who would prefer not to get cancer and  who are not yet sick. They have posted evidence on mammography for early breast cancer detection and PSA for prostate cancer detection that seem to be the opposite of the statement we all want hear about our cancer: “You are lucky, we found it just in time.”

Screening for cancer, it turns out, is an exact science when measured in aggregate but confusing on an individual level. I will try to clear up some of the confusion.

All screening discussions must include these truths:

  • The first is the fact that ultimately we all die. Dead with a cancer, whether we know about it or not, is dead.
  • The second is that some cancers will kill us or require aggressive therapy to stave off death ultimately no matter how early in the course it is detected. Living with a cancer that we otherwise wouldn’t have known about for 5 years does us no good if the treatment is the same, it just gives us 5 years longer to fret about it.
  • The third is that just because a cancer starts in a body part like the breast, doesn’t make it the same as your neighbor’s cancer with the same name. Some people with have cancer in their breasts for 20 years and die of old age but others with go from no detectable cancer to metastatic disease in 3 months.
  • The fourth is that all of our expensive testing has made it impossible to compare results today with results from 10 years ago. If we find a whole lot of non-aggressive types of cancer that start in the breast, it makes us look like we are better at treatment when all we are doing is counting cancers that wouldn’t have been counted 10 years ago.
  • The fifth is that there are behaviors (smoking, obesity) and some non-behaviors (family history) that make us prone to getting cancers, and if that is the case we ought to be watched a little more closely. Also we all ought to work on being healthier, as not getting the cancer in the first place is better than trying to beat it.
  • The sixth is that there are certain folks (older and sicker) who ought to worry about things other than having a cancer detected and we as doctors need to make them feel good about focusing on things other than early cancer detection

Upshot is that for folks OF AVERAGE RISK there is clear evidence that for certain cancers (breast, colon, cervical) at certain ages (over 50 and under 75 for breast and colon, over 21 and under 65 for cervical) early detection will make people live longer and feel better but the detection efforts don’t need to be very frequent (every other year for breast, every 3-5 years for cervical, and every 10 for colon screening with a colonoscopy). Outside of these age ranges there is some evidence for overtreatment (finding tumors when early treatment would not matter). There is also clear evidence for other cancers (ovarian, prostate) that early detection leads to overtreatment without people living longer or feeling better. Folks not of average risk include those with a strong family history (who may benefit from more intensive surveillance), folks who smoke (please quit!), and folks who are obese (exercise and weight loss are protective). In addition, if you have made it to 75, worry about something else.