Background

There is a widely accepted consensus based on data from cohort [1], case-control [1,2] and modelling [3] studies that cytological screening followed by treatment of preinvasive cervical neoplasia results in a substantial reduction of the incidence of and mortality from cervical cancer (CC). The burden of CC is geographically heterogeneous throughout the world with highest incidence in low income countries lacking effective screening programmes [4]. Poland has a medium incidence of and mortality from CC among European countries with world-age standardised rates of 9.8/100 000 and 4.8/100 000 respectively in 2011 [5].

Well-organised high-quality cytological screening may reduce CC incidence substantially. The successive steps of successful screening include information and invitation of the eligible target population, performance of the screening test, follow-up and treatment of the screen-detected neoplasia [6-8].

CC prevention in Poland has been present for over 50 years now with opportunistic screening run since the 1950ties and the first active programme (for personal invitation) implemented locally in one district of Warsaw in 1976 [9]. However, active country-wide population-based screening was absent. In 2007, such a programme was set up with full registration of invitations, responses to invitations, results of Pap smears and other procedures in an internet-based electronic registry.

The purpose of this study was to analyse the adherence of the Polish programme to the European Guidelines (EuG) for Quality Assurance in Cervical Cancer Screening [7,8]. We also summarised the course and effects of the first seven years (over two screening rounds) of the programme in Poland.

Methods

The use of data for this publication was approved by National Health Fund (NHF) authorities. The data for this study was drawn collectively from NHF databases and other available sources. All records were coded in such a way that identification of individuals was impossible.

Policy, structure, parameters and performance indicators of the programme

Data on the policy, structure and functioning of the organised cervical screening programme were drawn from published documents of the Director of the NHF in Poland [10], previously published reports [11-17], as well as analysis of the internet-based electronic data base SIMP (abbreviated from Polish: System Informatyczny Monitorowania Profilaktyki; English: Informatic System for Monitoring of Prevention) [18].

Organised CC screening programme in Poland is a public healthcare intervention organised by the Ministry of Health and the NHF. In its current version it was started in 2006/2007. Every three years, all women aged 25-59 identified from the lists of General Practitioners’ (GP) practices (which cover ~95% of population of women) are sent a written invitation via ordinary mail to have a Pap smear taken. The invitations without a set time and date of appointment are sent after a 36-month interval from a previous normal smear performed in the programme. HPV- and HIV-infected women and those taking immunosuppressants are eligible to perform smears every year. No reminder letters are sent to non-responders. A gynaecological clinic in the neighbourhood of woman’s GP practice is suggested on the invitation, but a Pap smear may be taken by a gynaecologist or a certified midwife in any of the ambulatory clinics in the country which provide gynaecological and obstetrical care reimbursed by the NHF. Since 2014, certified family midwives are also eligible to collect smears at GPs’ practices. The smears are processed by cytotechnicians and pathologists in selected cytological laboratories. The laboratories evaluating smears within the programme must fulfill explicit criteria and are expected to perform internal quality control. This includes full reviewing of 10% of negative slides or rapid reviewing of all negative slides and control of all positive slides by a specialist pathologist. All labs undergo quality audits annually by external pathologists according to a protocol elaborated by the Central Coordinating Office. Underperforming labs are excluded from the programme. The smears are interpreted according to a modification of the Bethesda system [19] (See Additional file 1). Women with abnormal Pap smear results are supposed to undergo triage within the programme via: 1) repeated Pap smear (ASC-US - Atypical Squamous Cells of Undetermined Significance, LSIL – Low Grade Squamous Intraepithelial Lesion), or 2) colposcopy/colposcopically directed biopsy (ASC-US, LSIL, ASC-H - Atypical Squamous Cells - Cannot Exclude High Grade Squamous Intraepithelial Lesion, HSIL – High Grade Squamous Intraepithelial Lesion, AGC – Atypical Glandular Cells, SCC - Squamous Cell Carcinoma). If triage procedures are performed within the programme, the results are recorded in the SIMP. Human Papillomavirus (HPV) testing recommended for triage of ASC-US and LSIL [20] is not available in the programme but can be performed within NHF-reimbursed gynaecological services. Medical procedures in opportunistic screening and management of screen-positive women are reimbursed by the NHF within ambulatory and hospital gynaecological care outside the screening programme but results are not recorded in SIMP. Despite recommendations [21], there are no obligatory certification or quality requirements for cytological laboratories operating outside the screening programme.

Costs of Pap smear collection, evaluation as well as colposcopy and biopsies performed within the programme are fully covered by the NHF. Costs of administration, coordination of the programme as well as mailing of invitations are covered by the Polish Ministry of Health from the funds of the National Programme for Fight Against Cancer Act [22]. The screening programme is coordinated by the Ministry of Health, a Central and 16 Regional Coordinating Offices (RCO). The RCOs send invitations to women and are supposed to monitor, evaluate and perform quality control of the programme. They are also responsible for training of personnel involved in the program and for setting up interventions to increase the programme coverage. These interventions include, but are not limited to: local and countrywide media campaigns, cooperation with local authorities, governmental and non-governmental organisations, direct meetings with programme participants and healthcare providers, promotion of the programme via countrywide and local cultural and social events.

Cancer registration

The National Cancer Registry (NCR) collects data via its 16 regional offices on cancer incidence, mortality, morbidity and survival in Poland [5]. Data on incidence come from electronic or traditional paper reports [5,23]. 87% of diagnoses were confirmed by pathology codes in 2011 [24]. Data on cancer mortality come from Central Statistical Office and are based on death certificates.

Since adherence of the programme to some points of the EuG [7,8] cannot be measured in a quantitative manner, a descriptive comparison of the valid recommendations, legislative acts and real-life clinical practice to the major points of the EuG is given. The adherence of the programme was qualified as full, partial or absent according to the level of agreement between the EuG and the Polish programme. Data on the available parameters and outcome measures of the CC screening programme in Poland were drawn from SIMP and other NHF electronic databases. We analysed which of the performance parameters in CC screening recommended by the EuG [7] can be calculated for Poland.

Burden of invasive cervical cancer

World age-standardised incidence and mortality rates of CC in Poland were extracted from the NCR database in Poland [5]. Rates were aggregated by calendar year and 10-year age groups (except for the oldest women categorised as ≥ 80 years). Joinpoint regression was used to analyse time trends (Joinpoint Regression Software) [25]. Joinpoint regression identifies periods with distinct linear slopes that can be separated by joinpoints, where the slope of the trends changes significantly [26]. For each linear segment, the annual percent change (APC) and corresponding 95% confidence intervals (95% CI) were calculated. Analysis was performed for a 13-year period from 1999 to 2011 which encompassed years of implementation of the screening programme (2006/2007).

Results

Policy and structure of the programme

The adherence of the most important constituents of the Polish programme to EuG is summarised in Table 1. The Polish programme is fully adherent to the EuG in terms of: type of screening test; interval between tests with normal results; and the age to start testing (Table 1). Partial adherence is noted for: the type of screening and screening policy; the age to stop testing; the issue of older women who have never attended screening; discouragement for opportunistic screening; information system; publication of programme indicators and new screening technologies (Table 1).

Table 1 Adherence of the organised cervical screening programme in Poland to European Guidelines for Quality Assurance in Cervical Cancer Screening [7,8] – screening policy, organisation, monitoring, evaluation and new screening technologies
Full size table

The adherence of organised screening in Poland to EuG for cytology laboratories, histopathology, management of screen-positive women is presented in Table 2. The Polish programme is fully adherent to EuG in the aspect of: collection, preparation, handling, staining, screening of samples and reporting of the results; and partially adherent regarding: grading of cytological abnormalities; histopathology as the gold standard and its terminology; availability of cytological results for pathologists, accuracy of histological diagnosis, correlation with cytological results; and management of screen-positive women (Table 2) [27-32]. No major points in the Polish screening programme have been identified to be completely non-adherent to EuG.

Table 2 Adherence of the organised cervical screening programme in Poland to European Guidelines for Quality Assurance in Cervical Cancer Screening [7,8,27-32] – guidelines for cytology laboratories, histopathology, management of screen-positive women
Full size table

Parameters and performance indicators of the programme

Selected available parameters of the organised and reimbursed opportunistic screening are presented in Table 3. The data collected in SIMP and other systems enable calculation of the following performance indicators [7]: programme extension; coverage of the target population by invitation; coverage of the target population by smear tests; compliance to invitation; distribution of screened women by the results of cytology; referral rate for repeat cytology; compliance to referral for repeat cytology; referral rate to colposcopy; positive predictive value of referral to colposcopy; compliance to referral to colposcopy.

Table 3 Selected available parameters on cervical cancer screening in Poland after implementation of organised screening programme (2007-2013)
Full size table

The following indicators cannot be automatically calculated based on the available systems: smear consumption; incidence of invasive cancer in unscreened or underscreened women in a given interval; test specificity; detection rate by histological diagnosis; treatment of high-grade intraepithelial lesions; proportion of women hysterectomised on screen-detected intraepithelial lesions; proportion of women treated for Cervical Intraepithelial Neoplasia grade 1 (CIN1); incidence of invasive cancer after normal and after abnormal cytology; proportion of women with cytology negative for squamous intraepithelial lesions, 6 months after treatment.

At this stage, smear consumption can be calculated by merging data from SIMP with data on opportunistic NHF-reimbursed Pap smears but not automatically on regular basis. Pap smear consumption in private care and private insurance plans is not recorded. Indicators such as incidence of cervical cancer in unscreened or underscreened women in a given interval, cancer incidence after normal or after abnormal cytology, treatment of high-grade intraepithelial lesions, proportion of women hysterectomised on screen-detected lesions theoretically can be calculated based on data from NHF treatment databases containing International Classification of Diseases version 10 (ICD10) codes and types of procedures with International Classification System for Surgical, Diagnostic and Therapeutic Procedures (ICD-9-CM) codes. However at present it would require special searches in the electronic systems and the quality of these data is undetermined since they contain no histological reports. NCR database and SIMP have not been connected electronically so far and computation of certain indicators is therefore still impossible at present.

Burden of invasive cervical cancer

Both age-standardised CC incidence and mortality rates have been decreasing steadily for the last decade in Poland (Figure 1). Figure 2 shows the age-standardised incidence rates (ASIR), and mortality rates (ASMR), as well as the annual percentage change (APC) with 95% CIs for CC in Poland (1999-2011). The declines in both standardised rates decreased significantly in the period 1999-2011 with the APC for incidence: -2.6, 95% CI: -3.1 to - 2.1 and the APC for mortality: -2.2, 95% CIN: -2.8 to -1.7 (Figure 2). The decreases in incidence were significant in all 10-year age groups apart from women aged 60-69 years (Figure 2). The age-specific mortality rates dropped significantly in all age groups apart from women aged 50-59 and 60-69 years (Figure 2). The linear slopes were constant in all age groups and no significant trend changes were identified over the analysed period (1999-2011) encompassing implementation of the programme (2006/2007) (Figure 2).

Figure 1
figure 1

Cervical cancer burden in Poland between 1999 and 2011.

Full size image
Figure 2
figure 2

World age-standardised rates of incidence of and mortality from cervical cancer in Poland (1999-2011). Figure legend: ASIR - Age-Standardised Incidence Rate; ASMR - Age-Standardised Mortality Rate; ^ - The Annual Percent Change (APC) is significantly different from zero; 95% CI - 95% confidence interval.

Full size image

Discussion

The implementation of organised cervical screening programmes in the member states of the European Union was recommended by the European Council in 2003, in agreement with EuG [7]. After an initial phase between 2004-2006, an organised programme was rolled out in its current state in 2007 in Poland. The trend of the burden of CC may be indicative of the impact of the enrolment of preventive programmes. Our initial analysis of the age-specific trends in CC incidence and mortality indicates no evidence of a significant period effect attributable to the introduction of organised screening (Figure 2). We consider these results with caution being aware that data from a simple age-period trend analysis might not be a sufficient proof of programme ineffectiveness. Additional more complex analyses utilising linkage between screening and cancer registries and incorporating data on changes in exposure to risk factors, changes in coverage and quality of non-monitored screening will be required to assess the impact of the implemented screening on the CC burden in Poland over a longer time perspective.

Our analysis reveals that organised CC screening in Poland is only partially adherent to EuG and there are a number of issues which require further insight and actions. The coverage and quality of screening seem to be most important. No registration of Pap smears collected outside the organised screening hampers comprehensive assessment of the coverage. In 2012 the coverage rate of the NHF-reimbursed (organised and opportunistic) screening reached 36% (Table 3). However according to survey-based data from Central Statistical Office in Poland, 86% of women aged 20-69 at least once in their lifetime have undergone pap testing and 73% of women at this age underwent a pap test within the preceding three-year interval [33]. Although, these questionnaire-based results probably overestimate the real coverage [34], they indicate that many women participate in the private-based opportunistic screening.

Organized screening is more effective and more cost-effective [7], however in many countries opportunistic screening is the only or dominating way of secondary prevention [35,36]. In Poland a direct shift from opportunistic to organised screening may be difficult to achieve in a short time due to habits of women and healthcare providers. Therefore integration of both screening modalities should be considered, as in other countries [35]. Analysis of systems integrating both types of screening in countries with longer experience such as Finland [37], the Netherlands [38], Denmark [39] and France [40] should be performed to select a model which fits best into Polish conditions. Nevertheless, more research on the reasons for non-attendance to the organised programme is required in order to find targeted solutions. Very recent findings from Finland indicate that carefully managed invitation/reminder letters with scheduled appointments and self-sampling options offered to non-attending women can increase organised programme effectiveness [41]. On the other hand, Belgian experience shows that sending invitations hardly influences screening coverage in a country with a long tradition of opportunistic screening [42]. Appropriate trials run in the real-life conditions are required to demonstrate the effectiveness of actions planned in Poland to increase organised screening coverage before costly implementation.

EuG propose discouragement of yearly opportunistic screening but opportunistic smears are still reimbursed by NHF and are recommended in pregnancy by the Ministry of Health [43] and by other actively operating non-governmental organisations in Poland [44]. No reimbursement for opportunistic smears could reduce overscreening in some cohorts, promote screening at recommended intervals and drive shifts towards organised screening [42]. However, stopping reimbursement of opportunistic screening might be controversial both for women and most of gynaecologists in Poland in the initial years. Nonetheless, it may induce better adherence of the real-life practice to the EuG and finally result in increased programme effectiveness. In Poland smears outside the screening programme are used and reimbursed for many clinical indications such as: triage of previously abnormal cytology, follow-up after treatment of CIN and cancer. A reimbursement protocol would have to be developed to reimburse smears for the above mentioned indications but not for screening beyond the recommended target age group and screening interval.

As explained in results, assessment of screening performance as recommended in EuG is currently impossible in Poland since several activities take place outside the organised programme, are not recorded and/or are inaccessible by lack of effective data linkage. Participation in the screening programme requires signing a consent for collection, storage and processing of womens’ personal data. However an additional legal framework is needed for data collection, storage, processing, including linkage of screening, follow-up and cancer register allowing to run and to evaluate the programme [7].

Although cytological labs in the programme undergo external quality audits every year, their results have not been published and assessed to reach conclusive points for programme improvements. Some of the cytological labs working outside the programme in the opportunistic screening are not monitored and their quality is unknown. Data on mortality audits and interval cancers have been published for some countries [45], but no such attempts have been made for Poland yet. Quality of colposcopy, and histopathology should be assured as well [46].

Consideration of new evidence regarding better performance of primary HPV-based compared to cytology-based screening is important for the country [47]. A pilot study on the use of HPV testing as a primary screening co-test has started in two regions. However results of HPV testing are not registered in SIMP which will hamper performance evaluation.

Implementation of a cancer screening programme can be divided into seven phases: (1) pre- planning, (2) planning, (3) feasibility testing, (4) piloting or trial implementation, (5) scaling up from pilot to service, (6) running, and (7) sustaining the full-scale programme [48]. For each phase, a substantial number of specified conditions have to be met. The Polish experience shows that some of the major points such as feasibility testing and pilot studies were omitted and some important detailed conditions such as the incorporation of an IT-system linking registries and a comprehensive system covering all steps in the screening process, development of a quality assurance plan and publication of its results have not been fully implemented thus far.

There may be other factors than the performance of organised screening, which may explain the higher rates of mortality from CC in Poland than in most of Western European countries [49]. The exposure to risk factors associated with cervical cancer as well as the effectiveness of diagnosis and treatment of women with invasive or preinvasive disease are among them and require further insight [50].

Conclusions

Polish organised cervical screening programme is only partially adherent to evidence based EuG. Its implementation has had no impact on the burden of CC in the country so far. Comprehensive research of non-attendance to organised screening and targeted actions are required to increase its coverage. Development of information systems to obtain data on opportunistic smears and histology reports, and linkage with cancer registry data is required to increase programme adherence to EuG and to measure its effectiveness. Our findings may be useful to improve the Polish programme and those implemented or planned in other countries.