Breast cancer mortality trends in Italy by region and screening programme,1980

Abstract

Objectives

To describe trends in breast cancer mortality in 14 Italian regions representing 90% of the Italian population, and in the country overall, in relation to the diffusion of organized or opportunistic breast cancer screening programmes.

Settings: Breast cancer mortality figures from 1980–2008 in Italy and 14 Italian regions were retrieved from the National Statistics Institute.

Methods

Estimated annual percent change (EAPC) was calculated for three time-periods (1980–1989, 1990–1999, 2000–2008). To estimate the age-adjusted and age-specific EAPC, trends in mortality based on annual data were examined using a log-linear regression model, and the year of death as a regressor variable.

Results

In the period 1980–1989 an increasing trend was observed in Italy overall (EAPC = +1.1%; 95% confidence interval [95%CI]: + 0.9%; + 1.3%) and in the 14 regions. In the second decade a decreasing trend was recorded both overall in Italy (EAPC = −1.6%; 95CI:−1.8%;−1.4%) and in all the regions. Between 2000–2008 a slightly decreasing trend was observed in Italy overall (EAPC = −0.8%; 95%CI:−1.0%;−0.5%). A steeper and earlier decrease compared with the rest of Italy was observed in 1990–1999 in Liguria (EAPC = −3.1%; 95%CI:−4.0%,−2.1%), Lombardy (EAPC = −2.3%; 95%CI:−2.8%;−1.9%), and Tuscany (EAPC = −3.0%; 95%CI:−3.7%;−2.2%), mainly due to the 55–64 age−group. There were significant EAPCs between −4.1% and −3.7% in these three regions, 2.5–3.0 times greater than in the rest of Italy (EAPC = −1.4%).

Conclusions

This earlier decline in three Italian regions could be partially attributable to local organized screening programmes implemented earlier than in the rest of Italy, and to opportunistic screening that became more common in 1990s.

Keywords

Breast cancer mortality Italy mammography screening

Introduction

Breast cancer is the leading cause of death and the most common cancer among women, in both economically developed and developing countries,¹ and represents 29% of all cancers in Italian women.² In recent years breast cancer mortality rates have levelled or begun to decline in many Western countries. In Europe the age-adjusted mortality rates declined by 6.9% between 2002 and 2006, with the largest falls in northern European countries.³ We aimed to describe breast cancer mortality trends in Italy as a whole and also in 14 Italian regions, that represent 90% of the Italian population, in relation to the diffusion of organized or opportunistic breast cancer screening programmes.

Methods

Data on breast cancer mortality for Italy overall and in 14 regions for the period 1980–2008 were retrieved from the National Institute of Statistics, which collects death certificates of Italian residents.⁴ We selected the nine most populated Italian regions (Piedmont, Lombardy, Veneto, Emilia-Romagna, Tuscany, Lazio, Apulia, Campania, and Sicily), plus five regions (Liguria, Friuli Venezia-Giulia, Umbria, Marche, Basilicata) with the highest mortality rates amongst the 11 smallest regions (Table 1). Because data for 2004 and 2005 were not available, we estimated figures for these years with a linear regression from the preceding and subsequent available years (2003 and 2006). Adjusted mortality rates using direct standardization to the European Standard Population, and age-specific (35–44, 45–54, 55–64, 65–74, 75–84, and >= 85 years) mortality rates with 95% confidence intervals (CI) were calculated.

Table 1.

Number of deaths for breast cancer, person-years, and crude rates (per 100,000) women, 1980–2008 in Italy overall and by region.

	1980–89 (10 years)			1990–99 (10 years)			2000–2003 & 2006–2008 (7 years)
	Deaths	Person-years	crude rates ^	Deaths	Person-years	crude rates ^	Deaths	Person-years	crude rates ^
Tuscany	6,881	10,923,511	62.99	7,249	11,527,666	62.88	5,201	8,711,850	59.70
Emilia Romagna	8,281	12,072,037	68.60	9,260	12,914,546	71.70	6,584	9,897,951	66.52
Liguria	4,583	5,844,574	78.41	4,627	5,808,489	79.66	3,139	4,174,119	75.20
Piedmont	9,804	13,144,072	74.59	10,629	13,753,676	77.28	7,309	10,347,376	70.64
Veneto	8,022	11,846,325	67.72	9,268	13,203,253	70.19	6,589	10,564,051	62.37
Lombardy	18,865	25,213,039	74.82	21,399	27,219,471	78.62	15,067	21,403,888	70.39
Friuli	2,905	3,796,616	76.52	3,252	3,950,252	82.32	2,198	2,946,502	74.60
Lazio	8,365	13,680,114	61.15	9,597	15,299,626	62.73	7,011	12,194,009	57.50
Umbria	1,322	2,428,414	54.44	1,546	2,625,390	58.89	1,153	2,032,920	56.72
Marche	2,369	4,146,906	57.13	2,745	4,491,765	61.11	1,923	3,498,885	54.96
Basilicata	577	1,491,823	38.68	654	1,624,001	40.27	535	1,262,612	42.37
Apulia	4,950	9,231,984	53.62	5,852	10,514,425	55.66	4,557	8,398,050	54.26
Campania	5,951	12,468,296	47.73	7,262	14,061,916	51.64	5,907	11,337,172	52.10
Sicily	6,399	12,171,461	52.57	7,696	13,379,008	57.52	5,870	10,454,487	56.15
Italy	96,787	152,853,975	63.32	110,619	167,671,560	65.97	80,282	130,930,540	61.32

^per 100,000

To calculate the age-adjusted and age-specific Estimated Annual Percent Change (EAPC), trends in mortality based on annual data were examined, using a log-linear regression model, and the year of death as a regressor variable.

Results

Mortality rates for Italy as a whole showed an increasing trend up to 1989, after which rates began to decrease, and then level out after 2000 (Figure 1). A similar trend was observed for the 14 selected regions (Figure 1).

Figure 1.

Age-standardized breast cancer mortality rates (x100,000) for the whole Italy and for Lombardy (Northern Italy), Emilia-Romagna, Tuscany (Central Italy), Apulia (Southern Italy).

We analysed the EAPC for these three time-periods (1980–1989, 1990–1999, 2000–2008). In the first decade an increasing trend in mortality rates was observed for the 14 Italian regions, with significant EAPCs ranging between +0.8% and +4.2% for all regions except Liguria, Piedmont, and Marche (Table 2). For the whole of Italy a significant EAPC of +1.1% was observed in this period. In the second decade a decreasing trend in mortality rates was observed for all regions, with a significant EAPC of −1.6% for the whole of Italy, and significant EAPCs for all regions except Umbria, Basilicata, and Campania. In the last decade a slightly decreasing significant trend of −0.8% was observed for Italy overall, with EAPCs ranging between −1.7 and +0.2 in the 14 regions.

Table 2.

Age-adjusted Estimated Annual Percent Change (EAPC) with 95% confidence intervals for the whole of Italy and by region, for the three decades 1980–1989, 1990–1999, 2000–2008.

	1980–1989	1990–1999	2000–2008
	EAPC, % (95% CI)	EAPC, % (95% CI)	EAPC, % (95% CI)
Italy	+1.1 (+0.9,+1.3)	−1.6 (−1.8,−1.4)	−0.8 (−1.0,−0.5)
Tuscany	+0.8 (−0.1,+1.6)	−3.0 (−3.7,−2.2)	−0.1 (−1.1,+0.8)
Emilia Romagna	+0.8 (+0.1,+1.6)	−1.0 (−1.7,−0.3)	−0.8 (−1.6,0.0)
Liguria	+0.3 (−0.7,+1.3)	−3.1 (−4.0,−2.1)	−1.7 (−2.9,−0.5)
Piedmont	+0.5 (−0.2,+1.2)	−1.1 (−1.7,−0.4)	−0.9 (−1.7,−0.1)
Veneto	+1.3 (+0.5,+2.1)	−1.4 (−2.1,−0.7)	−0.3 (−1.1,+0.6)
Lombardy	+1.7 (+1.2,+2.2)	−2.3 (−2.8,−1.9)	−0.8 (−1.4,−0.3)
Friuli	+0.1 (−1.2,+1.4)	−0.8 (−2.0,+0.4)	−1.0 (−2.5,0.4)
Lazio	+2.2 (+1.4,+2.9)	−1.7 (−2.4,−1.0)	−1.4 (−2.2,−0.6)
Umbria	+3.2 (+1.3,+5.1)	−1.1 (−2.8,0.6)	−0.5 (−2.5,+1.4)
Marche	+1.2 (−0.2,+2.6)	−1.4 (−2.7,−0.1)	−0.2 (−1.7,+1.4)
Basilicata	+4.2 (+1.3,+7.2)	−0.3 (−3.0,+2.4)	0.2 (−2.7,+3.1)
Apulia	+2.7 (+1.7,+3.7)	−1.7 (−2.5,−0.8)	−0.6 (−1.6,+0.4)
Campania	+1.9 (+1.0,+2.8)	−0.8 (−1.6,0.0)	−0.6 (−1.5,+0.3)
Sicily	+2.6 (+1.7,+3.5)	−1.1 (−1.9,−0.4)	−0.1 (−1.0,+0.7)

The rank of regions in Figure 1 remained much the same from 1980 to 2008: ie. high mortality regions in 1980 were still high mortality regions in 2008, but the magnitude of the difference had decreased. Most of the convergence appeared to have occurred since the late 1990s.

All the selected regions showed similar mortality trends, except for Tuscany, Liguria, and Lombardy, which, in the second decade, showed significantly steeper decreases than those observed in other regions: EAPCs in Tuscany, Liguria, and Lombardy were −3.0% (95% Confidence Interval [95%CI]: −3.7,−2.2), −3.1% (95%CI: −4.0,−2.1), and −2.3% (95%CI: −2.8, −1.9), respectively, in comparison with the rest of Italy (EAPC: −1.0, 95%CI:−1.5,−1.2).

To explain these differences, we compared the age-specific EAPCs of Tuscany, Liguria, and Lombardy, with the EAPC for the rest of Italy in the decade 1990–1999 (including all the other regions, not only the selected sample). A significant difference was observed for the age group 55–64; in Tuscany, Liguria, and Lombardy, decreases were 2.5–3.0 times greater than in the rest of Italy, with EAPC between −4.1% and −3.7% in the three regions, and −1.4% in the rest of Italy (Table 3). A non-significant difference in mortality rates was also observed in the age group 45–54, with EAPCs between −3.8% and −4.4%, in the three regions, and −2.1% in the rest of Italy.

Table 3.

Age-specific (35–74) Estimated Annual Percent Change (EAPC) with 95% confidence intervals for the regions Tuscany, Liguria, and Lombardy, and for the other Italian regions in the decade 1990–1999.

Age-class	Tuscany	Liguria	Lombardy	All other Italian regions
Age-class	EAPC (95% CI)	EAPC (95% CI)	EAPC (95% CI)	EAPC (95% CI)
35–44	−3.9 (−7.5,−0.3)	−5.3 (−10.6,−0.1)	−3.9 (−6.0,−1.9)	−3.3 (−4.3,−2.3)
45–54	−4.4 (−6.5,−2.3)	−4.2 (−7.2,−1.3)	−3.8 (−5.0,−2.6)	−2.1 (−2.7,−1.4)
55–64	−3.7 (−5.4,−1.9)	−4.1 (−6.4,−1.9)	−3.5 (−4.5,−2.5)	−1.4 (−1.9,−0.9)
65–74	−2.6 (−4.1,−1.0)	−2.1 (−4.1,−0.2)	−2.1 (−3.1,−1.2)	−1.5 (−2.0,−1.0)
75–84	−2.1 (−3.7,−0.5)	−2.7 (−4.6,−0.8)	−0.5 (−1.5,+0.4)	−0.5 (−1.0,0.0)
85+	−2.5 (−4.6,−0.4)	−2.6 (−5.1,−0.1)	−1.6 (−2.8,−0.3)	0.2 (−0.5,1.0)

Discussion

Trends in breast cancer mortality rates for the whole of Italy and in the Italian regions were in accord with those observed for Europe, with a peak in mortality in the late 1980s and a decrease thereafter.³ In the Liguria, Lombardy, and Tuscany regions a steeper and earlier decrease in mortality was observed in the decade from 1990–1999 compared with the rest of Italy, mainly due to the 55–64 age group.

The decline in mortality rates has been attributed both to an increasing frequency of early diagnosis through mammography screening and access to more efficient treatments, including adjuvant chemotherapy or tamoxifen, and improved radiotherapy and surgery.^5–11 It has been estimated that less than one-third of the 21.3% decline in mortality of women aged 50–69 (those involved in screening programme) in the United Kingdom in the period 1990–1998 was due to screening, the remainder being due mainly to improved treatment.¹² Another estimate reported that of the 47% reduction in breast cancer mortality recorded in East Anglia (England) after the introduction of service screening, 60% was due to earlier diagnosis, and 40% to innovations in therapy.¹³ In the period 1990–2008 in Catalonia, Spain, mortality reduction attributable to screening was 12.7%, to adjuvant treatments 12.4%, and to both interventions 23.7%, showing that 50% of the reduction can be attributed to screening and adjuvant therapy.¹⁴ The estimates of the impact of screening mammography on breast cancer mortality ranged from 10% from regional mortality data¹⁵ to 20% from trial results.¹⁶

In most Italian regions, organized screening programmes in the age group 50–69 began early in the 2000s or sooner. In 23 rural municipalities covering around 20% of the female population in the province of Florence, organized screening started at the beginning of 1970s.¹⁷ Brescia, Lombardy, followed in 1988, and the remainder of the Florence province (about 27% of the Tuscan female population) in 1990. Screening commenced in Turin, Piedmont, in 1992, in Reggio Emilia in 1994 and in the other Emilia-Romagna provinces in 1996–1998. In the Tuscan provinces other than Florence and in most Veneto provinces it started in 1998–2000.^17,18

Our results do not exclude the potential contribution of opportunistic screening to mortality reduction. A recent analysis of breast cancer incidence and mortality trends in Europe¹⁹ suggested that mammography screening was widespread even in the absence of organized screening programmes. In some Italian areas during the 1990s opportunistic mammography²⁰ covered less than 20% of the target population. In 1991 in Emilia-Romagna about 87,660 mammograms were performed within the target population (about 767,553 women aged 45–74), covering 11.4% of the target population [Naldoni C, personal communication], and in Veneto in 1993, about 112,000 mammograms were performed covering around 17% of women aged 40–49 and 5% of women aged 50–69.²¹

Most of the convergence of breast cancer mortality rates amongst regions seemed to have happened since the late 1990s (Figure 1), a period that was crucial for the development of organized screening and the introduction of new treatments.

One limitation in our study is that figures are not available for the 1990s in Italy on the use of adjuvant chemotherapy or tamoxifen, or on opportunistic screening. It is, therefore, not possible to draw strong conclusions regarding the relationship between screening onset and decline of breast cancer mortality by region.

Another limitation could be due to the codes of causes of death, as between 1980 and 2008 at least two different revisions of the International Classification of Disease were used to code cause of death. Although coding disparities could cause some variations in rates, they are not likely to compromise the comparability of data over time, because differences between revisions are minor for breast cancer.

Conclusions

Our study showed that in three Italian regions an earlier and steeper decline in breast cancer mortality could be partially attributable to local organized screening programmes implemented earlier than in the rest of Italy, and to opportunistic screening that became more common in 1990s.

Footnotes

Declaration of conflicting interests

None to declare.

Funding

None to declare.

Acknowledgements

We thank Dr. Carlo Naldoni, Department of Public Health, Emilia-Romagna Region, and Dr. Manuel Zorzi, Venetian Tumour Registry for having found figures for the 1990s regarding opportunistic mammography in their regions.

References

Jemal

Bray

Center

Ferlay

Ward

Forman

. Global cancer statistics. CA Cancer J Clin 2011; 61: 69–90. doi: 10.3322/caac.20107.

Associazione Italiana Oncologi Medici (AIOM) - Associazione Italiana Registri Tumori (AIRTUM). I numeri del cancro in Italia 2013. Intermedia, 2013: Brescia, Italy <http://www.aiom.it/area+pubblica/area+medica/pubblicazioni/1,332,1,>last access 2014 March, 2.

Bosetti

Bertuccio

Levi

Chatenoud

Negri

La Vecchia

. The decline in breast cancer mortality in Europe: an update (to 2009). Breast 2012; 21: 77–82. doi: 10.1016/j.breast.2011.08.001.

I.Stat. Direct Acces to statistical data, 2014 [http://dati.istat.it/] last access 2014 March, 12.

Berry

Cronin

Plevritis

Fryback

Clarke

Zelen

. Effect of screening and adjuvant therapy on mortality from breast cancer. N Engl J Med 2005; 353: 1784–1792.

Early Breast Cancer Trialists’ Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet 1998; 351: 1451–1467.

Cole

Gelber

Coates

Goldhirsch

() . Polychemotherapy for early breast cancer: an overview of the randomised clinical trials with qualityadjusted survival analysis. Lancet 2001; 358: 277–286.

Jatoi

Miller

. Why is breast-cancer mortality declining? Lancet Oncol 2003; 4: 251e4–251e4.

Clarke

Collins

Darby

Davies

Elphinstone

Evans

. Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet 2005; 366: 2087–2106.

10.

Dowsett

Cuzick

Ingle

Coates

Forbes

Bliss

. Meta-analysis of breast cancer outcomes in adjuvant trials of aromatase inhibitors versus tamoxifen. J Clin Oncol 2010; 28: 509–518. doi: 10.1200/JCO.2009.23.1274.

11.

Early Breast Cancer Trialists' Collaborative Group, Correa C, McGale P, Taylor C, Wang Y, Clarke M, et al. Overview of the randomized trials of radiotherapy in ductal carcinoma in situ of the breast. J Natl Cancer Inst Monogr 2010; 2010: 162–177. doi: 10.1093/jncimonographs/lgq039.

12.

Blanks

Moss

McGahan

Quinn

Babb

. Effect of NHS breast screening programme on mortality from breast cancer in England and Wales, 1990–1998: comparison of observed with predicted mortality. BMJ 2000; 321: 665–659.

13.

Duffy SW1 McCann

Godward

Gabe

Warwick

. Some issues in screening for breast and other cancers. J Med Screen 2006; 13(Suppl 1): S28–S34.

14.

Vilaprinyo

Puig

Rue

. Contribution of early detection and adjuvant treatments to breast cancer mortality reduction in Catalonia, Spain. PLoS One 2012; 7: e30157doi: 10.1371/journal.pone.0030157.

15.

Kalager

Zelen

Langmark

Adami

. Effect of screening mammography on breast-cancer mortality in Norway. N Engl J Med 2010; 363(13): 1203–1210. doi: 10.1056/NEJMoa1000727.

16.

Independent UK Panel on Breast Cancer Screening. The benefits and harms of breast cancer screening: an independent review. Lancet 2012; 380: 1778–1786. doi: 10.1016/S0140-6736(12)61611-0.

17.

Gorini

Zappa

Miccinesi

Paci

Seniori Costantini

. Breast cancer mortality trends in two areas of the Province of Florence, Italy, where screening programmes started in the 70s and 90s. Br J Cancer 2004; 90: 1780–1783.

18.

Osservatorio Nazionale Screening. I programmi di screening in Italia. Ministry of Health, Rome, 2014.

19.

Amaro

Severo

Vilela

Fonseca

Fontes

La Vecchia

Lunet

. Patterns of breast cancer mortality trends in Europe. Breast 2013; 22(3): 244–253. doi: 10.1016/j.breast.2013.02.007. Epub 2013 Mar 9.

20.

Holland W, Stewart S, Masseria C. Policy in brief: screening in Europe. World Health Organization, 2006: Copenhagen.

21.

Delibera n. 2284/1996. Finanziamento di progetti pilota di screening di popolazione per il cancro della mammella e del collo dell’utero. Regione Veneto. Venice, 1996.

Breast cancer mortality trends in Italy by region and screening programme,1980–2008

Abstract

Objectives

Methods

Results

Conclusions

Keywords

Introduction

Methods

Results

Discussion

Conclusions

Footnotes

Declaration of conflicting interests

Funding

Acknowledgements

References