Bollettino epidemiologico nazionaleReport from the Field Cesarean sections in Campania
Renato Pizzuti, Enrico de Campora, Sergio Lodato (Analysis and Monitoring Section – Regional Health Agency of Campania, Naples, Italy)
Cesarean sections (C-sections) are considered appropriate in well-defined clinical circumstances. According to recommendations made by the World Health Organization in 1985, the frequency of C-sections should not exceed 10-15% of all deliveries (1,2). In the last twenty years, the frequency of C-section deliveries in Italy has dramatically increased and is now the highest in Europe (32.9% in 1999) (3). The Campania region in Southern Italy has the highest rate within Italy, with 51.0% of deliveries performed by C-section in 1999.
The Regional Health Agency of Campania recently initiated a project designed to modify the current levels through the determination of the frequency and risk factors for C-section and the development of clinical and administrative guidelines. To date, the project has concentrated on a descriptive study of deliveries in the region, using hospital discharge records for 1996-1999 as the principal source of data. Six diagnosis-related groups (DRG) concerning method of delivery were selected (370 and 371 for C-sections and 372 to 375 for vaginal deliveries). Using information contained in the hospital discharge records, it was possible to determine the number of C-section and vaginal deliveries by year, type of health facility, age and nationality of the mother and to calculate mortality rates. In addition, the economic costs for the two types of deliveries was determined. For purposes of the analysis, place of delivery was classified according to whether it was a public or private structure and by the number of annual deliveries performed by the facility (<500, 500-800 and >800).
In the four years included in this analysis, an average of 65,000 deliveries occurred annually in either public facilities or private facilities accredited by the regional health authorities. The percentages of deliveries taking place in the private and public sectors remained similar over the four years, with 39% occurring in private facilities in 1999. The mean age of the women remained similar during all four years of the study, with a four year mean value of 28.3 years. Women undergoing C-sections were on average slightly older than those undergoing vaginal delivery 28.8 years versus 28.0 years).
The percentage of deliveries performed via C-section underwent a linear increase between 1996 and 1999 (Figure 1), with a 19.4% increase over the four-year period. Increases were observed in both the private and public sectors, although consistently the percentage of C-sections in the private sector was 1.3 times that observed in the public sector.
When the rates of C-section were compared between non-residents of the region for the four-year study period, the rate was 28.5% among the 1006 foreign-born women and 47.6% among the 5388 Italian women whose legal residence was outside Campania. When C-section rates were examined for the 4736 legal residents of Campania who delivered in other regions, the corresponding value was considerably lower than that observed among Campania residents delivering in their home region (34.8% versus 47.0%).
In addition, an inverse relationship in both the private and public sectors was observed between the number of annual deliveries in a facility and the frequency of C-section (Table 1). In private facilities with 800 or fewer deliveries per year, the C-section rate exceeded 55%.
The cost analysis demonstrated that in 1999, C-sections accounted for 61.7% of costs related to deliveries [133.7 billion lira (US $62 million) of a total expenditure of 216.7 billion lira (US$100.8 million)].
Review of records of women whose DRG was delivery-related and who were reported to have died during hospitalization or who were transferred to another facility revealed 6 deaths (9.1/100,000 deliveries), with 5 of the 6 occurring among the 33,467 women undergoing C-section (14.9/100,000 deliveries) and 1 death among the 23,169 delivering vaginally (3.1/100,000 deliveries; relative risk = 4.8, 95% confidence interval = 0.6-41.1).
As a result of this initial evaluation, the Regional Health Agency is planning an intervention, which will focus on the criteria for accrediting health structures and on the development of guidelines, with the creation of a group of physicians who will be responsible for disseminating and evaluating these guidelines.
References
1. Basevi V, Cerrone L, Gori G, Epid. Prev. 1994; 18: 194-199
2. Signorelli C, Cattaruzza MS, Osborn JF, Result from a study in three italian hospital. Milano: Kailash Editore ; 1995
3. S.I.S. – Ministero della Sanità - D.G. Programmazione Sanitaria, Rapporto annuale sull’attività di ricovero ospedaliero – dati SDO 1999
4. Stafford RS. JAMA 1990; 263:683-687
5. www.asplazio.it/index_1.htm
Editorial Note
Domenico Di Lallo (Agency for Public Health, Lazio Region)
The progress that has occurred in prenatal and intrapartum care represents an important factor in the improvement in the health of the maternal and infant population. The greatest benefits have been seen in the estimated 10% of pregnancies that are at the greatest risk of maternal and perinatal morbidity and mortality. However, such findings have contributed to the conviction, unsupported by demonstrated clinical evidence of efficacy, that increasingly complex and invasive diagnostic and therapeutic procedures need to be applied to the remaining 90% of uncomplicated pregnancies. In this context, the steady increase in C-section rates represent only one of many examples of the inappropriate use of an efficacious health care procedure.
The data presented by the Regional Health Agency of Campania serve to confirm this phenomenon and appropriately identify it as a high-priority health problem. The simple analysis of Regional data make two findings particularly evident: 1) that Italy, especially central and southern Italy, have rates of C-section notably higher than that of other countries in which values of 20-25% are considered to be optimal for maternal and perinatal health and2) that the wide variations observed between regions of the countries is not justified by differing distribution of clinical risk factors. Excluding the regions with few deliveries, in 1999, the C-section rate in Italy varied from 24% in Lombardy and 26% in Veneto, both in northern Italy, to 38% in Sicily and 51% in Campania in southern Italy. The importance of “non-clinical” factors emerges clearly from the analysis of discharge records in Campania. The rates of C-section are higher in private facilities than in public facilities performing similar number of deliveries (a proxy for the complexity of services provided by the maternity unit), and, for both public and private facilities, the C-section rate was inversely proportional to the number of deliveries preformed. These two findings suggest that both type of facility as well as complexity of services offered (services available and the previous experiences of single practitioners) play an important role in affecting the choice of delivery method.
What strategies are most effective to reduce the number of inappropriate C-sections? The experiences reported in the literature can be summarized in the following types of interventions: 1) education and peer evaluation; 2) external audit and review of practices; 3) public dissemination of information on the “performance” of single hospitals; 4) changes in physician reimbusement; 5) changing hospital reimbursement based on their performance; and 6) medical malpractice reform (4). Some of these strategies have been demonstrated to be ineffective while others have indeed proven useful, but results have not been consistent. The most promising of these methods has been peer audits. To test the usefulness of this strategy in the Italian setting, the Lazio Region is conducting an intervention study involving 17 maternity services based on the dissemination of regional guidelines (5) combined with internal discussions within the participating services regarding the statistics of individual physicians.
The use of hospital discharge records presented in this study by the Health Agency of Campania, even with all the limitations resulting from their primary function as an administrative tool, represents nonetheless a rich potential source for epidemiologic information about C-sections and their determinants. The production and dissemination to physicians of regional “ratings” for individual maternity services, adjusted for clinical factors, may represent the natural next step in the efforts of the Campania Region.
