Screening doesn't save lives.

I received a memo recently from an institution, extolling the virtues of its screening programs. It was entitled:

Screening Saves Lives. 

It was in large block capitals. I call bullshit on this.  Screening does not save lives. FULL. STOP.

This simplistic and self-serving public relations material is a typical example of common screening fallacy, that screening is an action that saves lives. This is worth deconstructing.

First, screening is not an action, something that stands alone. It is not a test or a measurement. It is a complex process, a veritable cascade of separate but connected actions, and, as with any complex process, there are ample opportunities to err. 

Here are the components of ethical and effective screening:

  • Identify the disease for which we will screen.  Not everything warrants screening. The utility of screening depends on many things, including the burden of the disease, how common it is, whether or not there is a safe and accurate screening test, whether or not the problem is treatable with better results if found early than late, and whether or not the treatment is safe, affordable and available.
  • Education and informed consent of the screened population.  It is unethical to screen unless the patient or population understands and accepts the screening.
  • Performing the screening test accurately, and reporting it in a usable manner to the patient and to the health care provider(s) working with the patient.
  • A defined and validated cascade of actions for both (+) and  (-) screening results. For example, screening for high fall risk does no good if the at-risk patients are not further evaluated and treated to mitigate their risk.
  • Identifying and mitigating or avoiding harms involved in screening and subsequent actions. This includes education, understanding and minimizing the impact of false (+) and (-) results, avoiding pseudo disease and over treatment, providing education and support to avoid anxiety, minimizing the social and financial harms of screening and subsequent actions (cost, time off work, stigma of diagnosis).
  • Tracking results and outcomes and using these for self-assessment, modifying the screening process when necessary.
  • Making sure there are adequate resources, not just to screen, but to address results.

Second, as should be obvious from the above discussion, screening itself does not save lives. Screening may identify people who might benefit if the screening and follow-up is done right, but the screening itself doesn’t provide any benefit.  Screening provides information. Unless that information is accurate,  available, understood, and then acted upon appropriately, screening is at best a waste of time and money. 

 (Note: in the current rush to transition from a volume-based to a value-based healthcare system, we are seeing incentives for screening, with no mechanism to ensure that the screening results in appropriate care. The implications are frightening. The result is the medical version of ‘teaching to the the test’ where institutions get rewarded for screening and don’t bother to use the results to improve patient outcomes.) 



Links to more on this topic::