One goal, variable policies – cervical cancer screening case study

As I am following an online course on cancer (Cancer in the 21st century – the genomics revolution – it’s free by the way, and put together very well), here is another case where the goal is pretty uniform, but the policies put in place to reach that goal vary. It’s a different angle to the Tay Sachs Disease case, because this is about national policies, and one could argue that the policies are actually quite similar but just vary in their details. That’s true,  but these details have significant consequences – financially for example, and emotionally too.

The topic of cervical cancer screening came up in the course as an example of how cancer is a multi-step process. This means that it takes several ‘hits’ of cell mutation before cancer develops. These mutations are made more likely by exposure to DNA-damaging agents (think smoking for lung cancer). In cases of individuals with inherited mutations, it may take less ‘hits’ since their DNA is already affected, which may increase the speed of the process eventually leading to actual cancer.

Cervical cancer screening therefore aims to identify “changes in the cells that have the potential to develop into cancer if left untreated”. If such changes in cells are identified, we cannot be sure that they will necessarily develop into cancer (as it takes multiple ‘hits’ for this to happen, and we don’t know whether the cells will experience such a path). However, when identified, we can try to remove those affected cells to be more certain that they don’t end up going down that path at all (this “usually involves removing an area of the cervix about the size of a finger tip” as described by NHS UK). That is essentially the goal of cervical cancer screening (which in fact screens for pre-cancer characteristics, following the above logic).

This topic is a great example of the interaction between scientific evidence and society, the effects of which I have personally experienced. I am a 27 year-old woman, my nationality is Belgian, and I live in the Netherlands. From my grand-mother and my mother who grew up in the Belgian healthcare system, and from a good friend who grew up in the German healthcare system, my expectation has been to visit a gynecologist and get screened for cervical cancer from my early twenties. Indeed, in Belgium screening is currently reimbursed from age 25, every 3 years (women are not invited personally as in other screening campaigns, but there are information campaigns about it). In Germany, screening was available and reimbursed from age 20, on a yearly basis, in 2005.* It also seems like regular/annual visits to the gynecologist are routine and expected in Germany and Belgium.**

However I was surprised and did not understand why I had never been referred to such screening in the Netherlands. I now know that the Netherlands also has an organized screening program, but it starts from age 30. Recurrence is also every 3-5 years. Altogether, this makes the Netherlands one of the European countries with the lowest total amount of tests over the course of a woman’s life.***

Why do different countries (of similar socio-economic status i.e. which are able to spend a similar amount of resources on screening) have different approaches to cervical cancer screening ? 

Age group beginning: in a 2009 overview the target age group varied from as early on as 15 years old in Luxemburg, while the current UK recommendation is 25, and Dutch one 30. Cancer Research UK  outlines that “younger women are more likely to receive an abnormal test result”, “cell changes found in younger women are less likely to develop into cancer”, and  “younger women are therefore more likely to have unnecessary further investigations and treatments.”

Popular debate including a government petition arose about whether to change the UK recommendation to 20 years old UK following the death of some younger women, including Sophie Jones (19), and celebrity Jane Goody (27). Such public debate affected by the sad and shocking news of young deaths is an example of societal perception and emotion interacting with evidence from science. In this case, the government determined that following the existing evidence and considering the above facts about younger women, it would not be cost-effective to lower the age of screening. To illustrate this, Cancer Research UK states that such a move would prevent between 3 and 9 cancers, at the cost of 8,000 unnecessary referrals for further tests, and 3,000 women receiving unnecessary treatment for abnormal cell changes (with possible negative health effects).

In addition to the interaction between public perception and evidence from science affecting public policy, another point of interaction is that of public behavior – in this case sexual behavior. Having sex earlier on in life increases the risk of cervical cancer (this makes sense considering the multi-step process I mentioned above – HPV is transmitted through sexual contact, and is the biggest risk of a ‘hit’ to cell DNA, so if this ‘hit’ occurs earlier on, the potential result of the process of hits may result in cancer earlier on).**** One could then argue that public behavior and/or the perception of public behavior can influence the beginning of the target age group. For example in a country where it is known – or believed/desired – that women only start having sex once they are married, screening may only be recommended or made available to married women. I have heard of this anecdotally but was not yet able to verify it.*****

Age group end: the recommended end for cancer screening also varies. The same 2009 overview indicates that Letonia continues to support screening until age 70;  Cancer Research UK states 64 years old for England, Wales and Northern Ireland, and 60 for Scotland; and the Dutch Ministry of Health states 60 years old. The appropriate age group end may be affected by how long women actually live – a 2014 study concludes that “stopping screening between ages 60 and 69 in women with adequate negative screening seems sensible, but further screening may be justifiable as life expectancy increases”. So in countries where women live longer, it may be relevant to continue screening until a later age. As socio-economic factors often affect life expectancy, this can be an example of socio-economic factors being relevant for public health policy.

Method used for screening: the screening method varies between two approaches. A cyotology ‘pap’ test (pap-smear or liquid-based) looks for abnormalities in these cells, as a sign of the first possible mutations which could, following further mutations, lead to cancer. HPV testing looks for DNA from the human papillomavirus that causes cervical cancer. So both approaches require a small sample of cervical cells, but they are looking for different things. Different screening programs do not use one or the other, but rather vary the sequence: cyotology test first, HPV test first, or both at the same time.

The latest research recommends that screening first start with HPV testing – if the test is positive, a woman would be referred for a cyotology test, instead of the other way around. Note that the chosen method also impacts the total number of tests performed. For example as HPV testing has been shown to be more effective, and due to the nature of the test (it looks for HPV, which usually takes several years before it would cause actual cancer), the time interval between tests can be increased to up to 5 years, instead of yearly or every 3 years.

This HPV-test first method has been adopted by some countries (e.g. The Netherlands in 2009), and is being piloted by others (e.g. the UK) – why? In this case, the experience of society and women in particular seems to be less relevant (as the experience of the test itself is similar since the way the sample is collected does not change). However, as always, the processes of medical actors and regulators remains relevant. In the US for example, although the FDA did approve “the use of one HPV DNA test (cobas HPV test, Roche Molecular Systems, Inc.) as a first-line primary screening test for use alone”, “first-line HPV testing has not yet been incorporated into the current professional cervical cancer screening guidelines. Professional societies are developing interim guidance documents, and some medical practices might incorporate primary HPV screening.”

We can note two points here. First, so far only one HPV test has been approved by the FDA for primary use – this may affect the availability and/or affordability of the test (I don’t know if it is the case for this test, this is a general statement of possibility considering the pharmaceutical industry). Second, even though a test has been approved, it takes time for professional guidelines and/or governmental regulations such as reimbursement rules to adapt. This is a known, general issue, with varying impact depending on the case. 


In addition to these three considerations around screening, the evidence for and availability of vaccination against HPV has added further layers of interaction between society, scientific evidence, industry, and public policy. I won’t go into detail here since this brings up a host of further questions – for example, since HPV is sexually transmitted, public emotions around such topics as the appropriate age for girls and women to have sex affect the debates around whether or not to administer the HPV vaccine. Furthermore, perceptions of vaccination in general also come into play. I will just note here for the record that the HPV vaccine does not prevent against all HPV viruses, so the screening recommendations also apply to vaccinated women.

Conclusion/ summary

In this post I used cervical cancer screening as an illustration of how public opinion, public (perception of) behavior, socio-economic factors, professional medical bodies, the pharmaceutical industry, governmental regulators and policy-makers can all interact with scientific evidence, leading to differences in public policy even when there is scientific evidence which could be the basis for common policies. The fact that scientific evidence  itself evolves over time is of course relevant in this dynamic as well.

There are many, many possible other examples of such dynamics, some may be better examples or highlight different aspects – I just chose this topic as it came up in my class and brought up these questions. I hope to continue these reflections with other examples as well.

Note: I did not go into the questions surrounding HPV vaccination, nor around (socio-economic-cultural) factors at the community and individual level – rather than the governmental public policy level – that also greatly affect policy and outcomes. See footnote *****


Provided throughout the text.


*German figure is from 2005 as I was able to check the current (2016) status of screening programs in Belgium, but not the German one. Due to the above points, some countries have changed their policies in the past ~15 years, others are debating it, so it may be that this has changed. My statement with regards to Germany is from the WHO Cervix Cancer Screening handbook from 2005. An update with reference is welcome (did not take the time to look for German-language sources as I would need to play around with Google Translate..).

**I looked for some references on national guidance and/or public perceptions of routine visits to gynecologists that would confirm that such visits are recommended and/or seen as normal in some countries (e.g. Belgium, Germany) and not in others (e.g. The Netherlands). I found some casual sources that seem to confirm my perception, however would like to find some stronger evidence. What I did find was the ‘strong’ recommendation “against performing screening pelvic examination in asymptomatic, nonpregnant, adult women” issued by the American College of Physicians in 2014 (+good summary here). Will investigate further for a later post.

***This doesn’t surprise me considering the general tendency of the Dutch healthcare system to be comparatively more conservative in its economic and societal/emotional cost-benefit evaluations of such medical acts like gynecologic visits, medicine prescription, hospital/medicalised birth, and more. Also a possible subject to explore in another post.

****Having multiple sex partners, or a partner with multiple sex partners, also increases the risk – this because the more sex partners the more possibility that one of them carried HPV and transmitted it to you. Note that this does not mean that those with only one sex partner are not at risk – the probability is just lower all other things equal. Other risks factors are relevant as well (e.g. smoking, poor diet, weakened immune system, and maybe genetic predisposition). It also not known for sure if being a virgin fully protects from cervical cancer.

*****To write this post I focused on Western European countries and the US and on public policy rather than individual behavior, because I wanted to see differences between countries who can definitely afford putting screening programs in place – i.e. their different policies would not be affected by very big variations in financial capabilities, so presumably it is other factors that influence their public policies. These countries issue guidelines that are mostly age-based (although they do also communicate on other risk factors), but that doesn’t mean that at community or family level (perceptions of) sexual behavior are not relevant. Other socio-economic-cultural factors may be awareness and knowledge; access to services including diagnostic testing and potential treatment; etc. There is definitely relevant and important research available on community/individual-level aspects, and this for a much wider variety of countries.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s