The world has learned a great deal from the careful evaluation of the cervical cytology screening programs in the Nordic countries.1 The analysis of data from these programs together with those from other countries not only resulted in a general recognition of the effectiveness of screening using cervical cytology but also led to recommendations about the best age at which to initiate screening and the frequency of rescreening after a normal smear, which have been accepted in a number of jurisdictions, especially those where screening is approached from the public-health point of view. It was therefore with a sense of shock that many readers of the International Journal of Cancer learned a little over 2 years ago that, despite a cytology screening program that achieved high coverage of the target population, the incidence of cervical cancer was increasing in younger women in Finland.2 At that time, however, it was felt in Finland that the increase, undoubtedly due to an extension of the sexual revolution in that country, did not justify a reduction in the age of initiating screening or more frequent rescreening; rather, an attempt to ensure that women in the target group (aged 30–60) did not “slip through” the screening net. In this issue, Sigurdsson and Adalsteinnson3 address this point using data from the Icelandic program but this time from the viewpoint of the frequency of detection of high-grade smears rather than invasive cancer. Although they concede at the end of their discussion that decisions on “specific intervals need to be backed up with analyses similar to the one presented here on the occurrence of invasive cancer among those screened”, it is clear that they feel that their analyses support the view that the current policies of initiating screening at age 20 and a maximum of 3-year intervals between screens should be maintained in their country. Is this a valid conclusion, based as it is on the frequency of detection of high-grade smears at the second visit, largely among women under the age of 30? It has been recognized that those at highest risk of developing a high-grade cervical abnormality are largely those previously diagnosed with a low-grade abnormality. This is confirmed by Sigurdsson and Adalsteinnson’s analyses.3 It is largely for this reason that all recommendations for frequency of rescreening are based on women with at least one previous negative smear.1,4 The fact that many high-grade lesions are diagnosed in women in Iceland with previously diagnosed low-grade lesions has no bearing on such recommendations, though it does re-enforce the view that once detected with an abnormality such women should be placed on special surveillance. Should one attempt to detect and treat the low-grade disease at all, especially in young women? We and others have demonstrated that the risk of invasive cancer in such women is remarkably low.5,6 This is because the large majority of such lesions regress spontaneously, an observation that Sigurdsson and Adalsteinnson appear to have confirmed, at least at a 6-month smear.3 Further, the majority of what are now called high-grade lesions also regress, not least because an important component is what used to be called moderate dysplasia (CIN2).5 Many, perhaps particularly in countries where medicolegal issues tend to dominate decision-making, will agree with Sigurdsson and Adalsteinnson3 that screening should begin from the age of 20 and be repeated in those with negative smears no less frequently than every 3 years. However, my purpose in this note is to indicate that another point of view is tenable as the incidence of invasive cancer at young ages is very low and the rapidity of progression of these cancers may be too great for them to be detected by screening. In several European countries, organized screening is initiated above the age of 20:7–15 in 1 country at age 23,8 in 2 countries at age 309,14 and in the remaining countries at age 25; also, in 4 countries, 5-yearly screening is the generally recommended interval.9,12,14,16 I also caution that it is not wise to make decisions based on intermediate end points when the objective is to reduce the incidence of invasive cancer.17 It does no one a service to detect and treat a lesion that is destined to regress. This is a lesson decision-makers in developing countries have to bear in mind as most of them simply cannot afford the efforts involved in detecting and placing under surveillance the large number of young women who develop detectable but temporary cervical abnormalities. We need to bear in mind the ultimate objective of our endeavors, which is to reduce the incidence of invasive cancer, and assess cervical screening programs in terms of how successful we are in that respect.