Survey of organized cervical cancer screening programs in Italy

Guglielmo Ronco*, Anna Iossa°, Carlo Naldoni+, Silvia Pilutti*, Emanuela Anghinoni§, Marco Zappa°, Paolo Dalla Palma$, Stefano Ciatto°, Nereo Segnan*, and the GISCi working group on organization and evaluation

* Unit of Cancer Epidemiology, Department of Oncology, ASL TOI. Center for Cancer Epidemiology and Prevention (CPO), Regione Piemonte, Torino;
° Center for the Study and Prevention of Cancer (CSPO), Florence;
+ Department of Preventione, Center for the Cancer Prevention AUSL Ravenna;
§ Pathology Department, Ospedale C. Poma, Mantova;
$Pathology Department Ospedale S. Chiara, Trento, Italy

THE GISCi WORKING GROUP ON ORGANIZATION AND EVALUATION:
DR.A PATRIZIA SCHINCAGLIA, PROGRAMMA DI SCREENING DELLA PROVINCIA DI RAVENNA;
DR. DINO DAVI, DR.A PAOLA GARUTTI, SCREENING POP.PREV.TUMORI COLLO UTERO PROV. FERRARA;
DR.A SONIA PRANDI, SCREENING PREVENZIONE TUMORI C.U. PROV. REGGIO EMILIA;
DR.A DEBORA CANUTI, UN PROGETTO PER LA SALUTE DELLA DONNA - RIMINI;
DR. FABIO FALCINI, PROGETTO PREVENZIONE SALUTE DONNA - FORLI';
DR.A SILVANA BORSARI, SCREENING CITOLOGICO PROVINCIA DI MODENA;
DR. PAOLO CRISTIANI, PROGETTO SCREENING BOLOGNA SUD;
DR. GIANNI FATTORINI, PROGETTO BOLOGNA DONNA (BOLOGNA CITTA');
DR.A NATALINA COLLINA, PROGETTO SCREENING BOLOGNA NORD;
DR. PIER GIORGIO DATARO, PROGRAMMA SCREENING PROVINCIA DI PIACENZA;
DR. ARRIGO BONDI, SCREENING PREVENZIONE CERVICALE - IMOLA;
DR. CLEMENTE VINEIS, PROGRAMMA DI SCREENING CITTA' DI IVREA E DINTORNI;
DR. GUGLIELMO RONCO, PREVENZIONE SERENA CITTA' DI TORINO;
DR. MAURO GIUDICI, DR. ANGELO PENNA, PROGRAMMA SCREENING - BIELLA;
DR. A EMANUELA ANGHINONI, SCREENING PER PREVENZIONE CERVICOCARCINOMA MANTOVA;
DR. RICCARDO PEASSO, SCREENING PREV. CERVICA CARCINOMA - OSTIGLIA;
DR. A. ANTONELLA PINARELLA, DR. GIUSEPPE COLLE, PROGRAMMA NEFERTARI CASTELFRANCO VENETO;
PROF. PAOLO DALLA PALMA, PROGETTO PAP-TEST PROV. TRENTO;
DR. LUCA VANNUCCINI, SCREENING DIAGNOSI PRECOCE TUMORI CERVICE UTERINA AREZZO;
DR.A ALDA MARTINI, PROGRAMMA SCREENING DEL CERVICOCARCIN. AREA PISTOIESE;
DR. GIORGIO NINCI, SCREENING MASSA M.MA - GROSSETO;
DR. MAURO GIRALDI, PROGRAMMA DI SCREENING - PIANA DI LUCCA;
DR. MAURO GIRALDI, PROGRAMMA DI SCREENING - VALLE DEL SERCHIO;
DR. FRANCO BERTI, DR. MAURIZIO BALLANTINI, SCREENING DI MASSA - ZONA CECINA;
DR. FRANCESCO MIRRI, PROGRAMMA DI SCREENING VALDARNO SUPERIORE ARETINO;
DR.A ANNA IOSSA, SCREENING CERVICO VAGINALE FIRENZE E PROVINCIA;
DR. A. MARGHERITA BRANCA, SCEREENING TUMORI FEMMINILI ROMA 1;
DR CARLO DI GIACOMO, SCREENING DEL CERVICO CARCINOMA DI TERAMO;
DR. CESARE PRINCIPE, PROGETTO GINECCEO - POMIGLIANO D'ARCO (NA);
DR.A SILVIA BREZZI, PROGRAMMA DI SCREENING VITERBO;
DR.A FRANCA SABBADINI, PROGRAMMA DI SCREENING CESENA;
DR. A. MARELLA ZATELLI, PROGRAMMA DI SCREENING PARMA.

This study was partially supported by EC Contract SOC 97 200284 05F02

Introduction

In Italy, where no national screening program for cervical cancer exists, organized programs have developed on a local basis. We performed the first survey of existing organized programs by mailing a standard questionnaire to a large network of possibly involved services. For the present survey, a program was defined as organized if personal invitations were sent. We identified 29 already active organized programs and 4 others in a starting phase. The target population of active programs included 2,074,820 women in the age range 25-64 years, corresponding to 13.5% of the Italian female population of the same age. The situation Is rapidly evolving since many regional programs are being implemented. Most programs followed Italian and European recommendations as regards the age limits, interval between screening rounds, presence of a fail-safe system for women referred for colposcopy, presence of protocols for diagnostic workup and treatment, and presence of referral centers for such phases. However, many programs did not meet national guidelines as regards the size of Laboratories interpreting smears, which were frequently small. Second-level referral centers also frequently had a very small activity. The average (weighted for size of the invited population) compliance to invitation and coverage (proportion of women with at least one test in the last 3 years) was 32.6% and 66.0%, respectively, therefore needing to be improved. Compliance to colposcopy (weighted for number of referred women) was 81.4%. We found a very high variability in the proportion of women referred for colposcopy that could only be partly explained by different referral protocols and could depend on different criteria of smear interpretation: the average (weighted for number of tested women) was 2.01%. A need for improvement in the process of evaluation and for homogenization of criteria of cytology interpretation was identified: work in this regard is on-going.
Effectiveness of screening for cervical cancer is generally recognized, and such activity is recommended as a public health measure (8). Organized screening programs are considered to reach the best results and are therefore reconunended (9). Screening for cervical cancer has been largely practiced in Italy as a spontaneous activity. However, no organized national program exists at the moment: organized screening programs have developed on a local basis and as local initiatives. Recently, a consensus document was prepared by the Italian National Research Council (CNR) and the Italian Association for Research on Cancer (AIRC) (6), and national guidelines for organized screening programs have been published (4). In September 1996, an Italian Group for Cervical Cancer Screening (GISCi - Gruppo Italiano Screening Citologico) was founded, as a spontaneous work aggregation of professionals active in organized screening programs and of experts of related fields (pathologists, gynecologists, epidemiologists) even if not directly involved at present in organized screening. One of the first activities of the group was a survey of the organized programs for cervical cancer screening existing in Italy. The purpose was to collect in a standardized manner data on the main features of the programs and some indicators for process evaluation.

Materials and methods

A standardized questionnaire was prepared and mailed in January 1997 to a number of different subjects. They were asked to complete the questionnaire if they were directly involved in organized screening programs already active or being implemented. They were also asked to contact any organized program they were aware of and ask them to complete the questionnaire. The mailing list included members of GISCi, a large number of services (Departments of Pathology, Gynecology, Cancer Prevention, family clinics etc.) potentially involved in organized programs or reported by members of the Group, and the Ministries of Health of all Italian Regions. In addition, questionnaires were mailed by the request of screening centers informed of the initiative by GISCi members. Overall, 123 questionnaires were mailed. We asked that only one questionnaire be Sent for each existing program and checked for duplicates.
A program was defined as organized if some kind of personal invitation was sent. This was considered the main feature of "organization". In addition, programs that invited women needed to have defined a target population and in general have some kind of evaluation. Completed questionnaires were received by May 1997 with a few exceptions. Each was checked for completeness and consistency, and the person who filled in the questionnaire was contacted, sometimes repeatedly, in order to obtain integration and clarification if needed. Only four process indicators were included: (a) compliance to invitation, (b) coverage (proportion of the target population having had at least a smear within 3 years either in the organized program or spontaneously), (c) proportion of women referred to colposcopy, and (d) compliance to colposcopy. Indicators did not need to be related to a fixed time period, but we asked to specify the time period to which each indicator was related.

Results

Overall, we received 42 completed questionnaires. On that basis we identified 29 already active organized programs (Figure 1) and 4 others in a starting phase (Cesena, Parma, Ravenna and Pavia). In addition, one pilot program had been run in Genoa from 1993 to 1997 and another organized program had been active in an area within Naples during 1994 but had to stop because of organizational problems. There was a large variability in the size of the target population of each program (Table 1), with only two (Turin and Florence) exceeding 250,000 women and 10 (34%) with at least 75,000. The recommendation of the Italian National Guidelines (4) is of a catchment area of at least 250,000 inhabitants. Overall, the sum of the target populations of the 29 active organized programs in the age range 25-64 years was 2,074,820 women. This corresponds to 13.5% of the 1991 Italian census population 25 to 65 years old. Programs were mainly concentrated in northern (1,472,571 women) and central (471,600 women) Italy: the proportion of women aged 25 to 64 years included in the target population of active organized programs was 20.7%, 16% and 3% in northern, central and southern Italy, respectively. Organized programs, active or in a starting phase, covered the entire population of the Emilia-Romagna Region and the Province of Trento. A large number of programs started very recently (in particular, all those from Emilia-Romagna): 16 (55% of those already active) started after October 1995.

Table 1 - Organized screening programs for cervical cancer in Italy by size of the target population 

Percentages on column totals.

Main characteristics

Most of the 29 active programs substantially respected the age limits (25 to 64 years) for the target population reported by the national (4) and European (3) guidelines, with the main exception of 7 programs starting invitation before 25 years of age (6 at 20 and 1 at 18) (Table 2). However, the latter included one program which at the moment invites only women in the age range 31-40. The overall target population before age 25 was 67,122 women.

Table 2 - Active organized programs by age at first invitation and age at stop

Although 27/29 programs used 3-year intervals between negative tests as recommended (3) (4) (8) (9) (of the remaining, one program invited subjects every 2 years and one used "personalized" intervals between 1 and 3 years), in 6 programs the first two tests were planned at a one-year interval. Concerning integration with spontaneous activity, we could identify three main models. In 12 programs all women in the target population were invited independently of their spontaneous screening experience (Model 1). In another 6 programs, only women who had not had spontaneous tests within 3 years received a reminder recommending a test (Model 2). In another 6 programs, women not spontaneously covered were first invited but the program aimed at "taking in charge" all women in the target population (Model 3). With models 1 and 3, the program will generally re-invite women at due intervals and advise them not to perform other tests except in the presence of suspect symptoms. Invitations were based on lists of the resident population in 13 programs, on lists of subjects in charge of the National Health Service in 10, and on both in 6. Programs following models 2 and 3 usually based the selection of uncovered women on computerized pathology records. All organized programs mailed personal invitation letters; such letters were signed by the general practitioner in 5 programs (17%); 24 programs (83%) mailed a reminder to women who had not complied with the first invitation. As regards smear taking (Table 3), we found programs with a scattered activity (more than one smear taking unit every 5000 women in the target population) and others with relatively centralized smear-taking centers (less than one every 20,000 women). In the former, smears were usually taken by a larger variety of staff (not only midwives but also National Health Service and private gynecologists) and in a larger variety of locations. In some screening programs, more than one (up to 3) cytology units was present. Overall, in the 29 active programs smears were interpreted in 43 centers. It is noteworthy that among them, even including activity outside the organized program, only 8 interpreted more than 25,000 tests per year, as recommended by the Italian National Guidelines (4), and 8 interpreted less than 10,000 tests per year (Table 4). The Bethesda classification was used by 24 active programs (7 of them used a modified classification), whereas 3 used CIN and 2 Papanicolau. A protocol for smear repetition was reported to be present in 27/29 active programs and one for referral to colposcopy also in 27/29. There was total agreement in referring to colposcopy women with a cytology of HSIL or invasive cancer and almost total in referring those with LSIL (Table 5).

Table 3 - Smear taking activity

(*) Each program can have different types of smear-taking units and smear takers. Therefore, headings are not exclusive and the sum of percentages is over 100.

Table 4 - Cytology Laboratories involved in active organized prngrams by number of smears interpreted per year and number of laboratories in the program

(*) Includes smears interpreted for spontaneous activity outside the organized program
(**) In parenthesis, percentage.

Table 5 - Criteria for smear repetition and for referral to colposcopy according to reported protocols

(a) Inflammation, "limited by" absence of endocervicai cells
(b) Alter oestrogenic therapy
(c) Presence of red blood cells in menopause, "limited hy" red blood cells, reactive changes
(d) ASCUS in menopause, squamous metaplasia, reactive changes according to interpreter (2 progrs)
(e) According to intenpreter's judgement
(f) Women repeat after short interval. If ASCUS/AGCUS is confirmed they are referred for colposcopy
(g) According to interpreter's judgement (2 progrs), if plausibly reactive, if plausibly condyloma, if severe inflammation or plausibly HPV signs
(h) According tu interpreter's judgement
(i) If koilocytosis without CIN or rare dysplastic cells
(I) Immature metaplasia, "Iimited by", atypical metaplasia or severe dystrophy

(a) Women repeat after shont interval. If ASCUS/AGCUS is confirmed they are referred for colposcopy
(b) According to interpreter's judgement
(c) Except some cases of koilocytosis only, except koilocytosis without CIN
(d) Herpes virus infection signs or radiotherapy reactive changes, hyperkeratosis (2 progrs), "more severe than reactive changes although less severe than ASCUS"
(e) Atypical metaplasia, immature metaplasia or hyperkeratosis, squamous metaplasia, reactive changes according to interpreter2

However, there was some variation in the management of women with ASCUS, although referral for colposcopy was largely the most common choice. It is remarkable that in 9 programs women with other cytologic diagnoses were, at least in some cases, also referred for colposcopy.
Twenty-four of 29 active programs reported the presence of a fail-safe system to remind women who did not show up after referral for colposcopy. Twenty-eight of 29 programs had referral centers for colposcopy. Their number was however very high (overall women were referred to 69 colposcopy units), resulting frequently in a very scattered practice: even including activity outside the program, only 9 colposcopy units performed more than 1000 colposcopies per year and 24 units performed <300 colposcopies per year (Table 6).

(*) lncludes colposcopies performed for spontaneous activity outside the organized program
(**) In parenthesis, percentage

The presence of a treatment protocol was reported by 23 programs: CIN Il was always treated by most programs (n = 19) and only in some cases by the remaining 4; CIN I was not treated as a first approach by 14 programs and in some case by the other 9. Referral centers for treatment were reported by 27/29 active programs, again with a remarkable dispersion in some cases (up to 6, although 12 programs had 1 referral center only).

Process Indicators

Table 7 shows figures reported by each program. Of the 29 active programs, compliance was provided by 26, coverage by only 16, the proportion of women referred for colposcopy by 22, and compliance to colposocopy by 20. All 6 programs using model 2 for invitation (see above) provided data on coverage compared to 10/23 of those using the other models (P <0.05, Fisher's exact test).
The average (weighted for size of the invited population) compliance to invitation was 32.6%. Most programs were in the range 20-40% (Table 8A). In all the programs, coverage (proportion of women with at least one test in the last 3 years) was much higher than compliance to invitation. The average (weighted for size of the invited population) value was 66.0%. Coverage was over 70% in 44% of programs (Table 8B).
The average (weighted for number of tested women) proportion of women referred for colposcopy was 2.01%. Data were extremely dispersed (Tables 7 and 8C), ranging from 0.14% to 9.2%. The most extreme values represented programs started very recently or with a very small activity, therefore unstable. However, even including only programs started before October 1995, values ranged from 0.4% to 5.3%. Referral rates were significantly high in programs referring ASCUS to colposcopy but, surprisingly, significantly low in those also referring cases other than CIN and ASCUS. Reciprocally adjusted (by a logistic regression, always with weights proportional to the number of tested women) odds ratios were 1.8 and 0.8, respectively.
The average compliance to colposcopy (weighted for number of women referred for colposcopy) was 81.4%.
In 60% of the programs it was over 90% but in 3 it was below 60% (Table 8D). Values did not include, in many cases, colposcopies perfortmed outside reference centers of the organized program. Many of the most extreme values were from very recently started programs and were based on very few cases.

(*) The protocol was changed during 1977. With the new protocol II level referral was 1.5 %. Tha last value was used in all other analyses