Summary

Download PDF

Screening has been defined as “the systematic application of a test or inquiry, to identify individuals at sufficient risk of a specific disorder to benefit from further investigation or direct preventive action, among persons who have not sought medical attention on account of symptoms of that disorder”.1 This definition highlights the aim of screening, which is to benefit at least some participants.

The potential benefits of cancer screening include: extra years of life for some people with early cancer diagnosed by screening, simpler treatment for some people with early cancer, and reassurance for those with negative screen results. The potential harms are: earlier diagnosis but no extra years of life for some people, over-treatment for some people with early disease, false reassurance for those with false-negative results, and anxiety and unnecessary investigations for those with false-positive results.2,3

Most of us, when we consider screening for cancer, focus on the potential for screening to extend life, even though the small proportion with the cancer being screened for means most participants cannot benefit in this way. It is documented that members of the public and health professionals commonly overestimate the benefits of screening.49 Contributing to this is the difficulty of assessing the efficacy of cancer screening, due to the following biases,2,10 all of which can cause the benefits of cancer screening to be overestimated.

Lead time bias:

Screening advances the date of diagnosis, thus extending the interval between diagnosis and death even if the date of death is unchanged. Because survival is measured from the date of diagnosis to the date of death, people with cancer detected by screening will appear to have longer survival than people diagnosed in the usual way, thus overestimating the benefit of screening.

Length bias:

Fast-growing cancers will progress quickly through the presymptomatic screen-detectable phase, so will be less likely to be detected by screening. Screening will therefore detect a disproportionate number of people with slow-growing cancers with a good prognosis, again tending to overestimate the benefit of screening.

Selection bias:

People who accept the offer of screening may differ in their underlying risk of developing or dying from cancer so that their prognosis would differ from non participants even in the absence of screening. If people at increased risk of cancer are more likely to participate, the benefit of screening could be underestimated. Conversely, if people at low risk of cancer are more likely to participate, the benefit of screening could be overestimated.

Overdiagnosis bias:

Screening may detect cancers that would not have been diagnosed in the absence of screening, and would never have affected an individual’s life. If overdiagnosis of cancer occurs, screening participants detected with cancer will include a disproportionate number with a good prognosis, again overestimating the benefit of screening.

The only research design that can avoid these biases is a randomised controlled trial of cancer screening, with mortality as the outcome measure. It is important that evidence of benefit from such trials is available because if a screening programme is established without adequate evidence of benefit, the harms of screening will still exist; leading to the aphorism “All screening programmes cause harm; some do good as well”.11 Evidence of benefit is an important issue, which has been recognised for over 40 years:

“We believe that there is an ethical difference between everyday medical practice and screening. If a patient asks a medical practitioner for help, the doctor does the best he can. He is not responsible for defects in medical knowledge. If, however, the practitioner initiates screening procedures he is in a very different situation. He should, in our view, have conclusive evidence that screening can alter the natural history of the disease in a significant proportion of those screened.”

“We believe that there is an ethical difference between everyday medical practice and screening. If a patient asks a medical practitioner for help, the doctor does the best he can. He is not responsible for defects in medical knowledge. If, however, the practitioner initiates screening procedures he is in a very different situation. He should, in our view, have conclusive evidence that screening can alter the natural history of the disease in a significant proportion of those screened.”12

CC-BY

There is an ethical obligation for those who establish and maintain a screening programme to ensure that the programme can deliver the benefit claimed. This benefit cannot be assumed, as a poorly run programme may not deliver the expected benefit. Additionally, a poorly-planned cancer screening programme has the potential to cause harm even to those outside the programme. For instance if the flow-on from screening is not anticipated, the increased need for investigation and treatment will cause longer waiting times for all those requiring these investigations or treatments, including those outside the screening programme.

A screening programme is more than a test; it is a pathway that includes identifying the eligible population, sending invitations, performing the screening test, ensuring appropriate and timely investigation of positive tests, provision of treatment, follow-up and re-invitation at the appropriate screening interval. The best way to minimise the potential harm from screening and still achieve a benefit is to ensure that the programme is properly organised and appropriately monitored at each step in the screening pathway (quality control). This will keep the possibility of harm to those taking part as low as possible.

Further complicating the picture for cancer screening is that the balance of benefits and harms differs according to: the type of cancer being screened for (for instance colorectal cancer screening compared with prostate cancer screening);13 the type of screening test (for instance FOBT compared with flexible sigmoidoscopy);14 the eligible age-range for screening (for instance breast cancer screening for women aged under 50 compared with screening for women aged 50 and over);15 the screening interval;16 the potential for overdiagnosis;17 and the effect of advances in treatment, which may cause screening to become less effective over time.18

Decision-making about the benefits and harms of cancer screening occurs at two levels; the population level and the individual level. At the population level, because screening has harms as well as benefits, and because of the resulting ethical implications, principles of screening were developed by the World Health Organization19 and some countries have adapted and extended these principles to develop their own criteria for screening.20,21 At the individual level there is a tendency to overestimate the benefits of screening, compounded by difficulties in estimating the efficacy of screening without overestimating benefit.4,9 Thus, it is important to present both the potential harms and benefits of screening when inviting people to participate in cancer screening programmes; hopefully moving from “uninformed compliance to informed choice”.22,23 Some cancer screening programmes, such as the NHS Breast Cancer Screening Programme, now use leaflets which include estimates of potential benefits and harms of screening.24

The challenge for all cancer screening programmes is to provide accurate information on the potential benefits and harms, to allow eligible people to make an informed decision about whether to participate or not.