Epidemiology and costs of cervical cancer screening and cervical dysplasia in Italy

Cervical cancer is the second most common cancer in women worldwide and accounts for almost 10% of all cancer-related deaths. In Italy, it is estimated that each year 3,418 new cases and 1,186 deaths occur [1]. Infection with Human Papillomavirus (HPV) has been established as the necessary cause of cervical cancer [2, 3]. As this cancer can be diagnosed at an early stage, screening programs using the Papanicolaou (Pap) smear test have reduced significantly the number of invasive cases, through the diagnosis and treatment of precancerous lesions (cervical intraepithelial neoplasia, [CIN]) [4, 5].

The Italian Ministry of Health establishes health-related objectives at national level, while implementation is the responsibility of regional governments. Some regions started implementing cervical cancer screening programs in the 1970s, while nationwide organized programs, rolled out on a regional basis, were recommended in 1996, based on the European Commission Guidelines on Quality Assurance in Cervical Cancer Screening [6]. By 1999, however, only 34% of Italian women aged 25–64 years were included [7]. In 2004, the Italian Government ordered that each region should plan and implement cervical cancer screening programs [8], to be provided by Local Health Authorities. Such programs are still not fully implemented across the country, although opportunistic screening has increased. Two preventive cervical cancer vaccines (Gardasil^® and Cervarix^®) are now registered in Italy. In March 2008, a mass vaccination campaign targeted at eleven-year-old girls was initiated, with some regions vaccinating a second cohort of girls (14 or 15 year-olds) and one region introducing a multi-cohort vaccination strategy, focused on four cohorts of girls aged 11, 14, 17 and 24 years.

In the absence of national data on the epidemiology of cervical cancer screening in Italy, the present study was undertaken to provide an estimation of the number of women undergoing screening annually, the annual incidence of cervical dysplasia, and direct costs associated with the management of women diagnosed with cervical dysplasia.

Our study suggests that costs associated with the detection and treatment of cervical dysplasia are substantial. Each year the Italian healthcare service is expected to spend €181.5 million (€200.5 million if the higher incidence of CIN2+ lesions is taken into account) to screen approximately 6.4 million women aged 25–69 years, to manage more than 150,000 women with abnormal Pap smears, and to treat an estimated 21,000 women with CIN1 lesions and 7,000–17,000 women with CIN2/3 lesions. Although the majority of these costs are directly related to cervical screening, only a small proportion is attributable to organized screening programs, with opportunistic testing accounting for the greatest share of costs. In contrast, our estimates suggest that only 11% of the total healthcare costs are associated with the management of CIN lesions; this percentage increases to 18% if the higher CIN2/3 incidence rate is applied.

Interestingly, although approximately 6.4 million women have a Pap test each year, which should cover the target population for organized screening, there are still 30% of women who had no Pap test in the last three years. Only 1.2 million women are screened within organised programs, but approximately 5.2 million are screened opportunistically. This suggests that some women are screened more frequently than the recommended 3-year interval, while others are screened despite being older than the recommended age, whereas some women may never be screened. This is consistent with analyses from the ISTAT data, showing that up to 50% of women in Southern Italy and the Italian islands have never been screened, compared with approximately 25% of women in Northern and Central Italy [18].

This study has several limitations. Firstly, extrapolation of GISCi data to the whole Italian female population may result in an overestimation of the number of women screened and in an underestimation of CIN management. Secondly, although the GISCi survey represents the only available data relating to screening programs from different Italian regions, this data collection system neither includes opportunistic screening carried out in the private setting, nor takes into account data from each Italian region. Furthermore, some Italian regions do not yet have formalized screening programs, and data included in the GISCi survey 2004 [11, 12] are, therefore, collected exclusively in regions with well-established programs. Thirdly, the absence of any national data on opportunistic screening, which accounts for the majority of screening in Italy, also limits estimates of the epidemiology of cervical dysplasia. Lastly, we could not estimate the number of cases of CIN2/3+ for women younger than 24 years of age and older than 70 years. Although a small proportion of CIN2/3+ occurs in women under 24 years of age, such lesions are very likely to regress [19] and, according to guidelines, should not be diagnosed [8]. Conversely, very few women over 70 years of age have Pap smears. Our calculations assume that in situ lesions detected in women older than 69 years, as reported by the Emilia Romagna Cancer Registry, can be attributed to Pap smears performed in women aged 65–69 years.

The two methods used to estimate the annual number of CIN2/3 lesions produced markedly different values. Method 1, based on data from the Emilia Romagna Cancer Registry, suggested that there are about 7,000 cases of CIN2/3 diagnosed. However, data drawn from a single registry cannot have 100% sensitivity to accurately project the incidence of these lesions nationwide, especially as this data collection is relatively new and does not have well-established quality indicators, such as for invasive cancer. The value derived using Method 1 can, therefore, be used as a lower limit of the CIN2/3 range. Method 2, which used data from the GISCi survey, suggested that there are more than 17,000 cases of CIN2+ diagnosed each year. This number might overestimate the incidence rate, as the detection rate observed in organized screening programs using a 3-year interval is probably higher than the detection rate of opportunistic screening, often based on a 1-year interval. This estimate can, therefore, be used as the upper limit of the CIN2/3 range. In our sensitivity analysis, this parameter appeared to be one of the most influential, resulting in the highest cost estimate (+ 6.3%).

With reference to the annual expected number of CIN1, it should be noted that only 42.1% (40 out of 95) of centres participating in the GISCi survey systematically provide data. Although these lesions do not generally require treatment, it is commonly agreed in clinical practice, particularly in opportunistic screening, that deviations from best practice are frequent and over-treatment is often performed. Consequently, the impact of CIN1 treatment on costs might have been underestimated.

In the cost calculations, it was assumed that all women with ASCUS would be referred directly for colposcopy, although this is not entirely consistent with current guidelines or clinical practice. However, the introduction of HPV triage does not change substantially the total costs, as shown by the sensitivity analysis. Nevertheless, estimation of the costs of additional diagnostic procedures for LSIL and HSIL and for treatment of CIN was based on practices reported in the GISCi survey, so may be considered representative of current practice in Italy.

The shift of treatment procedures to a more appropriate setting of care, in particular conisation from inpatient to outpatient is currently the most suitable way to reduce total costs of disease. Nevertheless, a change in the setting of surgery from day hospital to outpatient has a strong impact if tariffs are used to measure costs, but this is an overestimation of the real impact on the real costs sustained by the Health Service.

Although it is not possible to make direct cost comparisons between different healthcare systems, it is interesting to note that the costs of cervical screening in Italy are comparable with those for England, where approximately 4 million women are screened each year at a cost of €157 million (approximately €200 million), including the cost of treating abnormalities [20]. However, the estimated cost per case associated with CIN1 treatment in the present study was lower in comparison with the mean cost reported in England and Wales (€419; approximately €532), whereas the estimated unit costs for CIN2 and CIN3 were considerably higher than the mean costs (€572 [approximately €726] and €606 [approximately €770], respectively) [21].

As the focus of this study was the economic burden of cervical dysplasia, our calculations did not take into account societal costs, such as work loss due to screening or for CIN treatment. We also did not capture the psychological burden of receiving a positive Pap smear result.

Until recently, regular screening was the only tool to prevent cervical cancer. The development of HPV vaccines has provided a new opportunity to reduce the incidence of cervical dysplasia [22‐24] and studies have shown that a combination of both vaccination and screening is a cost-effective solution to reduce morbidity and mortality caused by cervical cancer [25‐28].

This is the first study to assess the epidemiology of cervical cancer screening and cervical dysplasia, as well as the associated healthcare costs, for Italy as a whole. Each year, the Italian healthcare service system makes a substantial investment of €158.5 million for the provision of screening services and €22.9 million for the follow-up and treatment of cervical abnormalities.

Higher adhesion to guidelines and protocols would produce substantial savings. Although the number of Pap tests performed each year would be sufficient to cover the whole target population, 30% of women are still not screened.

The economic burden of disease associated with cervical dysplasia identified in this study will assist health authorities with both the planning and allocation of funding for an effective cervical cancer prevention strategy.