Last week, I shared the first part of an article in which I am raising questions about the consequences of knowledge. Today, I’m continuing…
An Excess of Information
One of the first pieces of birth-related research to explore this area was Barbara Katz Rothman’s “Spoiling the Pregnancy”, in which she talked with sixty Dutch midwives about their experiences with prenatal fetal anomaly screening and diagnosis [i]. The findings of this study illustrated the way in which the knowledge that is gained through prenatal screening and diagnosis can have a profound effect on women’s experiences of childbirth as well as on society’s understanding of birth and motherhood. This study highlighted the way in which such technologies carry costs as well as benefits, not least of which are the false negative results that are an inevitable consequence of screening tests and the false positive results that the midwives described as giving a false sense of security and control. We know that the psychological and personal downsides of screening include intense feelings of anxiety and grief and the midwives in Katz Rothman’s study understood that, “while some of these “false alarms” are inherent in the nature of screening itself, others come because a baseline for normal has not yet been established” [ii].
The problem of establishing what is normal is an ongoing one, and is further highlighted by a recent systematic review and meta-analysis that looked at medical imaging in relation to low back pain [iii]. When magnetic resonance imaging (MRI) technology started to become widely available, it was assumed that the detailed images of the interior of the human body that it creates would have a positive impact on improving the care and experiences of people with lower back pain. For the first time, doctors might actually be able to see problems such as trapped nerves and would thus be able to diagnose and treat back pain more effectively. In reality, however, the effects of introducing MRI scanning in this area were nowhere near as positive as had been hoped and in some cases the overall effect was actually negative. After excluding patients whose health histories suggested the possibility of a serious underlying condition (which included, for instance, a history of cancer, unexplained fever or recent trauma), the meta-analysis showed that imaging did not improve the outcomes of patients who had low back-pain compared with usual clinical care.
There are a number of reasons why this might be the case, but one is particularly pertinent here. The images that are produced by MRI are so highly detailed that it can be difficult to differentiate relevant from coincidental information. An earlier study [iv] asked two neuroradiologists to comment upon MRI scans of a number of people who had no back pain or other physical symptoms in this area. Of these basically normal people, 64 per cent were found to have an intervertebral disc anomaly and 38 per cent were found to have an abnormality at more than one level. The relationship between the experience of physical symptoms and the appearance of apparent pathophysiology on MRI is clearly not a direct one. The use of MRI for lower back pain leads to a significant potential for over-diagnosis and some specialists in this area argue that we do not yet know enough about how the normal body appears on MRI for this to be an effective diagnostic tool in relation to lower back pain. This does not mean, of course, that MRI is not useful in diagnosing other problems, but at this stage of our knowledge it is clearly not useful on a routine basis in this area and the detailed knowledge that it provides carries risks as well as benefits.
Many midwives already perceive that intervention may be a more positive addition to our toolkit when it is used appropriately and individually rather than on a population basis. Our view of childbearing as a normal life event during which intervention is needed only occasionally tends to make us wary of the use of routine or universal intervention and more focused on using specific interventions on an individual basis when it is truly warranted within the context of a particular woman’s experience. We understand that it can be useful to ask some questions of screening tests on a population level [i], but we focus our actions on what women want. We talk with women and their families about whether or not they would like to have a particular screening test and thus acknowledge that decisions are personal and contextual. We make women aware of the limitations of knowledge and we discuss with them the potential for false positive and false negative results.
This is not necessarily the approach taken by all health care providers. Within modern medical care, it is generally deemed beneficial to create guidelines which encompass population-level recommendations about screening. Some specialties, however, are arguably more likely to undertake and act upon analyses of the pros and cons of screening tests when constructing those guidelines than others. For example, when the US Preventive Services Task Force reviewed the evidence in relation to screening men over the age of 75 for prostate cancer, they concluded that this was probably not beneficial on a population basis [ii]. Their analysis of the evidence showed that, when the cascade of intervention that results from a positive screening test (which in some cases is inevitably going to be a false positive result) is weighed against the potential benefits of treating a form of cancer that is often slow-growing, the overall benefit is small to none. Furthermore, they were concerned that the issue of labelling again ensures that any individual who is found to have a positive result has to deal with the consequences of this knowledge, which can lead to a high level of anxiety for them and their families.
I’ll share the next part of this article next week.
[i] Katz Rothman B (2001) Spoiling the Pregnancy: Prenatal Diagnosis in the Netherlands. In: De Vries RG, Benoit C, Van Teijlingen E and Wrede S (2001) Birth by Design. New York: Routledge. Chapter 9: pp 180-198
[ii] Ibid, p. 194.
[iii] Chou R, Fu R, Carrino JA and Deyo RA (2009). Imaging strategies for low-back pain: systematic review and meta-analysis. Lancet 2009; 373: 463–72.
[iv] Jensen MC, Brant-Zawadzki MN, Obuchowski N et al (1994). Magnetic Resonance Imaging of the Lumbar Spine in People without Back Pain. New England Journal of Medicine 331(2) 69-73.
[v] See, for instance, Nielsen C and Lang RS (1999). Principles of Screening. Medical Clinics of North America 83(6): 1323-1337.
[vi] U.S. Preventive Services Task Force (2008) Screening for Prostate Cancer: U.S. Preventive Services Task Force Recommendation Statement. Annals of Internal Medicine 149: 185-191
A version of this article was originally published as Wickham S (2009) Screening and the consequences of knowledge. Birthspirit Midwifery Journal 2: 9-12.