These snippets are extracts from a monthly service called the Journal Article Summary Service. It is a service that summarises all that is new in obstetrics and gynaecology over the preceding month. If you would like to know the details of how to subscribe, please email the editor Athol Kent at firstname.lastname@example.org or visit the website http://www.jassonline.com.
What’s new in the other journals?
Article first published online: 10 MAY 2012
© 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology © 2012 RCOG
BJOG: An International Journal of Obstetrics & Gynaecology
Volume 119, Issue 7, pages 898–899, June 2012
How to Cite
Kent, A. (2012), What’s new in the other journals?. BJOG: An International Journal of Obstetrics & Gynaecology, 119: 898–899. doi: 10.1111/j.1471-0528.2012.03386.x
- Issue published online: 10 MAY 2012
- Article first published online: 10 MAY 2012
Screening is being questioned on both sides of the Atlantic. No fewer than five editorials have appeared in recent American journals asking the profession and the public to re-evaluate screening and routine investigations (New York Times 9 April 2012; 20).
In the UK the debate on mammography screening has been referred for an independent assessment and budget cuts will undoubtedly call for routine tests to be justified. As advisors whose opinions are valued by our patients, our own feelings sometimes need review.
Intuition about screening
Screening for cancer is intuitively a good idea. However, many of our intuitions about screening do not actually stand up to careful scrutiny. We all need to re-examine the facts from time to time, so here are some thoughts that may cause reflection. We are being asked ever more searching questions on the topic so let us re-examine some of the ‘facts’.
Catch it early—improve the prognosis.
All the evidence supports detection of a malignancy at the earliest or even precancerous stage but does screening actually detect the disease?
If the screening test did indeed diagnose the cancer—not demonstrate a precursor or a surrogate marker—then it would be of immense value. None of the large screening programmes claim to diagnose the malignancy for which they are screening. They alert the individual in the screened population to their increased risk of harbouring the disease, which still requires definitive diagnosis. The best screening tests have 90% sensitivity and 95% specificity, so there will always be more incorrect positive results than correct positive results if the prevalence of the condition being screened for is low in the population. In other words there will be more false or erroneous positives than true or correct positives.
This would not matter if the false positives were of no consequence. If they carried with them no harm then all would be well. Unfortunately that is not true and these harms are becoming more and more apparent. They are medical, psychological, financial and epidemiological. People feel that they have cancer until proven otherwise, which carries anxiety and concern. Further investigations are costly and time-consuming. Having had a false-positive result, it is difficult for the person to be re-assigned to a low-risk category or to opt out of a screening programme.
Early or precancerous lesions progress.
If an early lesion is found by screening, is it going to progress or might it regress?
In cervical cytology, the likelihood with which precursors are going to evolve into invasive cancers depends on factors such as re-infection by causative viruses, hormones or host responses. As it is not known which tumours advance and which do not, there is a tendency to treat all as potentially malign and progressive. Apart from lesions that may regress, there are some that are so slow-growing (indolent lesions) that they never cause symptoms, let alone threaten the longevity of the host. This is the world of over-diagnosis, which is thought to occur in one-fifth of the commonest cancers.
Screening prevents the disease.
Screening does not protect against a disease. It is not a preventative process. It should not be confused with actively promoting health. One’s intuition is to clump healthy habits or sensible precautions with screening despite the fact that they are conceptually different.
Wearing a seat belt or insuring your house are not preventative measures—nor is screening. Prevention and screening must be thought about differently.
Screening is cost effective.
The cost of screening is a minefield. The financial costs to the individual may or may not be covered by insurance schemes or national health services. Some screening programmes are included in insurance policies, others are not.
There are vast industries built around screening, from employment in national programmes to equipment sales and the processing of slides and radiographs, the measurement of serum levels and the management of results. Billions are spent or saved depending on national guidelines, which vary considerably from country to country and time to time (Woolf and Harris JAMA 2012;307:565–6). Those favouring screening are portrayed as heroes or exploiters depending on the commentators’ point of view, while those sounding caution are being accused of being over-frugal when lives are at stake.
The personal costs of acquiring a disease or detecting it early are theoretically calculable by Quality-Adjusted Life Years (QALYs) but fears, concerns, worries and financial circumstances are much more difficult to calculate for individuals and their families.
It is clear when screening should start and stop.
Everyone has their own ideas about when to start and stop screening based on the likelihood of the disease in the population. Epidemiologists draw up guidelines and august bodies recommend them but it is patently unscientific to lump people together irrespective of their age and risk. Also an understanding of the natural history of the disease is crucial, as gynaecologists have found out in relation to cervical cancer; improved understanding of the aetiology has radically altered the age at which screening is started.
As populations age it becomes increasingly unclear as to the harms and benefits of screening otherwise healthy older people. In the USA, the ‘home of screening’, more than half of women over the age of 75 years continue to have screening recommended by their physicians despite the paucity of evidence regarding the benefits (Bellizzi et al. Arch Int Med 2011;171:2031–7). Guidelines cannot include all comorbidities, chronic or even terminal conditions that must considerably influence any recommendation. The evidence in the USA is that over-recommendation is common by physicians, possibly because of the fear of litigation if people are not encouraged to have tests for cancers they later turn out to be harbouring (Leach et al. Cancer 2012;118:27–37).
Personal decisions about screening are evidence-based.
It appears that an individual decision about screening is made by processes that are not entirely logical, evidence-based or necessarily weigh the harms and benefits (Woolf and Harris JAMA 2012;307:565–6). Personal experiences, social norms, hopes and fears, advice from friends, beliefs and intuition may be the final arbitrators of decisions and these factors may prevail—despite medical advice.
One of the truly beneficial effects of screening is the reassurance that a negative test can engender. This should never be underestimated (Detsky JAMA 2012;307:1035–6).
If you participate in a screening programme—it is likely to save your life.
Many people are persuaded to join screening programmes in the belief that it is going to ‘save their life’. As the cancers screened for in most populations have a relatively low prevalence, it is unlikely that the screening process will actually save someone’s life but it is one of the most important reasons why people participate in screening. As far as breast screening is concerned, the following is a fair question that every woman has the right to ask of her doctor: ‘If I have breast cancer diagnosed in a mammography programme, what are the chances that screening will have saved my life?’
To answer the question, Welch and Frankel (Arch Int Med 2011;171:2043–6) worked out how often screening mammography actually saved lives. They performed the calculation for a middle-aged woman contemplating 10 years of screening and her chances of dying from breast cancer over the next 20 years. They assumed that the woman was low-risk and that mammography would reduce her risk of death by somewhere between 5% and 20% (the usual estimate of reduction of death risk by screening for breast cancer).
Using the optimistic outcome estimate of screening reducing mortality by 20%, 13 in every 100 women given a positive result from screening would have their lives saved and 87 in every 100 would not have their lives saved.
Using the pessimistic estimate of mortality reduction (5%), then three in every 100 women would have their lives saved and 97 in every 100 would not have their lives saved.
Altering the calculations according to the age of the woman gives different results but none of them show that more than 25% of women would have their lives saved by screening. And these are the women who are given a positive diagnosis—the vast majority are not.
The authors concluded that ‘Most women with screen-detected breast cancer have not had their life saved by screening. They are instead either diagnosed early (with no effect on their mortality) or over-diagnosed’.
These are telling statistics and exemplify how emotive words can be used to persuade people about saving lives. The really important life-saving measures are lifestyle related and cost a great deal less than screening.
Screening is an inexact science.
HPV vaccination for males
The human papillomavirus (HPV) vaccines at present cover types 16 and 18 only (Cervarix® ) or 6 and 11 as well as 16 and 18 (Gardasil®). Both vaccines are highly effective in protecting against cervical abnormalities in women. They do so by protecting the woman from infection through HPV-positive contact, so a case could be made for promoting prophylaxis by vaccinating males so they could not become carriers. This has logistic and cost implications and has not been adopted as a public health policy anywhere.
However, the case can be made for male vaccination to prevent anal, penile and oropharyngeal cancers caused primarily by HPV type 16 and protection against genital warts or condylomata caused by types 6 and 11. All of these conditions are increasing in prevalence and the American Advisory Committee on Immunization Practices has recommended the quadrivalent vaccine for all males in the USA as of October 2011 (JAMA 2012;307:557–9). More data are available from Schlecht (JAMA 2012;307:724–5).
Little is known about the prevalence of oral HPV and its association with cancers of the mouth and throat. Preliminary data from the USA suggests that it is very common with measured HPV infection rates of 7% of the general population with men having a higher incidence than women (Gillison et al. JAMA 2012;307:693–703).
A nocebo is the opposite of placebo. A placebo has positive effects suggested by someone offering the substance or treatment. There is no scientific explanation but the recipient feels better—often measurably so. A nocebo has negative effects that the recipient experiences because they have adverse expectations suggested to them. Both are put down to psychological effects.
Often the nocebo effect is generated by information about an event, drug or procedure being conveyed in a negative fashion (Colloca and Finniss JAMA 2012;307:567–8). By saying that side effects can occur and naming them, such effects can be promoted and compared with when these facts are omitted. As always, clinical judgement is required.