Fax: (011) 49 451 500 5455
An observational study comparing trends in melanoma mortality in regions with and without screening
Article first published online: 19 APR 2012
Copyright © 2012 American Cancer Society
Volume 118, Issue 21, pages 5395–5402, 1 November 2012
How to Cite
Katalinic, A., Waldmann, A., Weinstock, M. A., Geller, A. C., Eisemann, N., Greinert, R., Volkmer, B. and Breitbart, E. (2012), Does skin cancer screening save lives?. Cancer, 118: 5395–5402. doi: 10.1002/cncr.27566
Implementation and conduct of the 12-month SCREEN project was supported by German Cancer Aid (Deutsche Krebshilfe e. V.), the National Association of Statutory Health Insurance Physicians (Kassenaerztliche Bundesvereinigung), and the head associations of health insurance funds (Spitzenverbaende der Krankenkassen).
A.K. planned the study, did the data collection and descriptive data analysis, and drafted the article. A.W. helped to draft the article and provided oversight for the literature review. M.A.W. and A.C.G. critically revised the article, contributed to the literature, and provided substantial input to the discussion. N.E. was responsible for the statistical analysis. R.G. and B.V. critically revised the article. E.B. coordinated the study group, contributed to the design of the study, and helped to interpret and write the article.
- Issue published online: 19 OCT 2012
- Article first published online: 19 APR 2012
- Manuscript Accepted: 29 FEB 2012
- Manuscript Revised: 23 FEB 2012
- Manuscript Received: 23 JAN 2012
- skin cancer melanoma;
- early detection;
From July 1, 2003 to June 30, 2004, a population-based skin cancer screening project was conducted in Schleswig-Holstein, Germany. In total, 360,288 individuals aged ≥20 years were screened by means of a whole-body examination. In this report, the authors compare trends in melanoma mortality in Schleswig-Holstein with those in all adjacent regions, none of which had population-based skin cancer screening.
Trends in melanoma mortality rates for Schleswig-Holstein and the adjacent regions (Denmark and the German federal states of Mecklenburg-Vorpommern, Hamburg, and Lower Saxony) and in Germany excluding Schleswig-Holstein were compared. Log-linear regression was used to assess mortality trends.
In Schleswig-Holstein during the pre skin cancer screening period (1998-1999), the age-standardized melanoma mortality rate (World standard population) was 1.9 per 100,000 for men and 1.4 per 100,000 for women. Melanoma mortality declined by 47% to 1.0 per 100,000 men and by 49% to 0.7 per 100,000 women by 2008/2009. The annual percentage change in the most recent 10-year period (2000-2009) was −7.5% (95% confidence interval, −14.0, −0.5) for men and −7.1% (95% confidence interval, −10.5, −2.9) for women. In each of the 4 adjacent regions and in the rest of Germany, mortality rates were stable, and the decline in Schleswig-Holstein was significantly different from the changes observed in all of the other areas studied.
The current data represent strong evidence, but not absolute proof, that the skin cancer screening program produced a reduction in melanoma mortality in Schleswig-Holstein. Cancer 2012. © 2012 American Cancer Society.
Skin cancer is the most frequent cancer worldwide, and the World Health Organization estimates that there are almost 200,000 incident cases of melanoma and 46,000 deaths worldwide annually.1 In the last 30 years, the incidence of malignant melanoma in United States and Europe has more than tripled.1-3 Unlike most other cancers, such as colorectal cancer, breast cancer, cervical cancer, and prostate cancer, that have experienced substantial declines in mortality, melanoma mortality rates that once were rising have only recently stabilized.1, 2 In Germany, in 2006, an estimated 15,000 individuals were diagnosed with melanoma, and approximately 2500 deaths were recorded.4 Although the relative survival of patients with advanced-stage melanoma is poor, with a 16% 5-year relative survival rate for patients with stage IV disease according to International Union Against Cancer staging criteria, patients have an excellent prognosis when they are diagnosed with early stage disease, with a 99% 5-year relative survival rate for patients with stage I disease.2
In past decades, significant efforts were made to reduce melanoma deaths through health campaigns5 or mass screenings for the early detection of skin cancer.6-9 Because many of those studies included only self-selected participants, few influential decision-making groups interpreted the results as strong evidence that skin cancer screening reduces the burden of skin cancer.10
To reduce skin cancer mortality (focusing on melanoma) and morbidity (focusing on nonmelanoma skin cancer), a pilot skin cancer screening program, the Skin Cancer Research to Provide Evidence for Effectiveness of Screening in Northern Germany (SCREEN) project, was conducted in the northernmost federal state of Germany (Schleswig-Holstein) before the launch of a nationwide skin cancer screening program in the German health care system that began in July 2008. Preliminary SCREEN-associated activities were initiated during the years from 1999 to 2003 in Schleswig-Holstein, which has 2.8 million inhabitants. More than 360,000 men and women ages ≥20 years were screened by full-body skin examination between July 1, 2003 and June 30, 2004 (the SCREEN project period). The first results on participation, tumor findings, and population-based melanoma incidence data have been published elsewhere.11, 12
Sufficient time has now passed to assess the trends in melanoma mortality in Schleswig-Holstein compared with the areas immediately surrounding it to the north, south, east, and west and compared with the rest of Germany. Herein, we describe the association of the SCREEN project with mortality rates through the end of 2009 compared with the aforementioned areas.
MATERIALS AND METHODS
Preparation for the SCREEN project included multiple activities (see Table 1). One key aspect was that 1673 of 2614 general practitioners and 116 of 118 dermatologists in Schleswig-Holstein received an all-day training program in detecting skin cancer using a standardized whole-body examination. The training courses started in early 2003. The screening was organized as a 2-step intervention. The first whole-body examination was performed by general physicians (nondermatologists). If a suspicious lesion was detected, then the patient was referred to a dermatologist, who conducted a second whole-body examination and, if necessary, the patient underwent a biopsy. Alternatively screenees were free to consult a dermatologist for initial screening, but only 22.6% used this pathway. The population was motivated to attend the screening by physicians, health insurers, and different mass media campaigns. Each screened individual provided written informed consent. The details of these procedures have been reported elsewhere.11, 12
|Period||Activities||Who Was Involved?|
|1989-1998||Development of the 2-step procedure, testing the feasibility of the training, performance of the screening test, and documentation of the screening examinations||83 physicians; federal states of Hamburg and Lower Saxony|
|1998-1999||Preparation of the SCREEN project in Schleswig-Holstein|
|2000-2001||First pilot project; 8-h training courses; screening examinations||200 Physicians and 6000 screenees; federal state of Schleswig-Holstein|
|2001-2003||Continuous activities, including skin cancer awareness campaigns and preparation of the SCREEN project in collaboration with German Cancer Aid|
|2003||8-h training courses for physicians||>1700 Physicians; federal state of Schleswig-Holstein|
|July 2003 to June 2004||SCREEN project, population-based skin cancer screening||>360,000 Screenees; federal state of Schleswig-Holstein|
In this report, we describe melanoma mortality from 1998 to 2009 in Schleswig-Holstein and the 4 adjacent regions: Denmark to the north (population, 5.6 million) and the 3 adjacent federal states in Germany: Mecklenburg-Vorpommern to the east (population, 1.6 million), Hamburg in the south (population, 1.78 million), and Lower Saxony in the west (population, 7.9 million). In addition, we used the entire country of Germany (excluding Schleswig-Holstein) for comparison. Denmark has never had a systematic skin cancer screening program like the SCREEN project, whereas nationwide, systematic skin cancer screening in Germany started in the middle of 2008.
The official mortality statistics provided information on deaths from malignant melanoma, which were identified by the International Classification of Diseases (10th edition) code C43, stratified by age, sex. and year for the studied regions.13, 14 We calculated age-standardized rates per 100,000 individuals using the World standard population (World age standardized rates) and 95% confidence intervals.15 Ten-year mortality trends for the most recent period (2000-2009), expressed as annual percentage changes (APCs), were assessed by log-linear regression16; differences in 10-year trends for Schleswig-Holstein compared with the 5 other regions were assessed by comparing the slopes of the regression lines using t tests.17 Ethical approval was not required for this study, because we used administrative data that did not allow for the identification of individual patients.
In the 1-year period from July 1, 2003 to June 30, 2004, 360,288 residents of Schleswig-Holstein aged ≥20 years were screened, which comprised 19% of the population aged ≥20 years (94,982 men [10.4%] and 265,306 women [27.2%]). In total, 585 individuals were diagnosed with melanoma (213 men [36%] and 372 women [64%]), including 69% invasive melanomas and 31% in situ melanomas. During the same period, 1169 melanomas (again, including 69% invasive melanomas and 31% in situ melanomas) were recorded in the Schleswig-Holstein Cancer Registry. The proportion of all registered melanomas that were SCREEN-detected was 50% (men, 46%; women, 56%).11, 12
In the prescreening period (1998-1999), the melanoma mortality rate was 1.9 of 100,000 men and 1.4 of 100,000 women in Schleswig-Holstein (Fig. 1, Table 2). Relative to the 1998 to 1999 rates, melanoma mortality rates declined by 47% to 1.0 of 100,000 men and by 49% to 0.7 of 100,000 women during 2008 to 2009. Trend analyses in Schleswig-Holstein revealed a significant decline in melanoma mortality for the last 10 years (overall, −7.4% per year; men, −7.5% per year; women, −7.1% per year) (Table 3). The regions adjacent to Schleswig-Holstein (Denmark, Mecklenburg-Vorpommern, Hamburg, and Lower Saxony) and all of Germany excluding Schleswig-Holstein had no clear changes in melanoma mortality during the last 20 years or the last 10 years (Fig. 2, Table 3).
|Population||Schleswig-Holstein: SCREEN Project Area||North: Denmark||East: Mecklenburg- Vorpommern||South: Hamburg||West: Lower Saxony||Germany Excluding Schleswig- Holstein|
|2009 Population||1.39 Million||2.74 Million||820,000||870,000||3.90 Million||38.8 Million|
|No. of MM deaths/y|
|MM mortality: WASR (95% CI)|
|A: 1998-1999||1.9 (1.5-2.4)||2.9 (2.4-3.3)||1.1 (0.7-1.5)||1.3 (0.9-1.8)||1.7 (1.5-1.9)||1.6 (1.6-1.7)|
|2003-2004||1.6 (1.2-1.9)||2.7 (2.4-3.1)||1.6 (1.1-2.2)||2.0 (1.5-2.5)||1.8 (1.5-2.0)||1.8 (1.7-1.9)|
|B: 2008-2009||1.0 (0.7-1.3)||3.2 (2.8-3.5)||1.6 (1.2-2.1)||1.4 (1.0-1.8)||2.0 (1.8-2.3)||1.8 (1.7-1.9)|
|2009 Population||1.44 Million||2.79 Million||840,000||910,000||4.04 Million||40.3 Million|
|No. of MM deaths/y|
|MM mortality: WASR (95% CI)|
|A: 1998-1999||1.4 (1.1-1.8)||1.8 (1.5-2.1)||0.8 (0.5-1.2)||1.2 (0.8-1.6)||1.3 (1.1-1.4)||1.1 (1.0-1.1)|
|2003-2004||1.3 (0.9-1.6)||1.8 (1.5-2.1)||1.0 (0.6-1.3)||1.1 (0.7-1.4)||1.2 (1.0-1.4)||1.1 (1.0-1.1)|
|B: 2008-2009||0.7 (0.5-1.0)||1.9 (1.6-2.1)||1.1 (0.7-1.5)||1.1 (0.7-1.5)||1.1 (0.9-1.3)||1.2 (1.1-1.2)|
|Total 2009 population||2.83 Million||5.53 Million||1.66 Million||1.78 Million||7.94 Million||79.1 Million|
|No. of MM deaths/y|
|MM mortality: WASR (95% CI)|
|A: 1998-1999||1.7 (1.4-2.0)||2.3 (2.0-2.5)||1.0 (0.8-1.3)||1.2 (0.9-1.5)||1.4 (1.3-1.6)||1.3 (1.3-1.4)|
|2003-2004||1.4 (1.2-1.6)||2.3 (2.0-2.5)||1.3 (1.0-1.6)||1.5 (1.2-1.8)||1.5 (1.3-1.6)||1.4 (1.3-1.4)|
|B: 2008-2009||0.9 (0.7-1.1)||2.5 (2.2-2.7)||1.3 (1.0-1.6)||1.2 (1.0-1.5)||1.5 (1.4-1.7)||1.4 (1.4-1.5)|
|APC (95% CI)|
|Population||Schleswig-Holstein: SCREEN Area||North: Denmark||East: Mecklenburg- Vorpommern||South: Hamburg||West: Lower Saxony||Germany Excluding Schleswig-Holstein|
|Men||−7.5 (−14.0, −0.5)a||2.0 (0.0, 4.0)||−0.4 (−3.6, 2.9)||−3.4 (−8.0, 1.5)||−1.2 (−4.1, 1.7)||0.2 (−0.8, 1.3)|
|Women||−7.1 (−10.5, −2.9)a||−2.1 (−4.4, 0.2)||−1.2 (−8.1, 6.2)||1.1 (−4.4, 6.9)||−0.4 (−2.8, 2.2)||0.7 (−0.4, 1.8)|
|Total||−7.4 (−11.9, −2.7)a||0.3 (−1.2, 1.8)||−0.3 (−3.8, 3.4)||−1.4 (−5.1, 2.5)||−0.8 (−2.4, 0.7)||0.5 (−0.4, 1.3)|
|P vs Schleswig-Holstein||—||.002||.012||.033||.008||.001|
We tested the overall mortality trend of Schleswig-Holstein against the trends for the adjacent regions to assess statistical differences in the annual percentage changes. The declining melanoma mortality trend in Schleswig-Holstein was significantly different from that in each of the other regions (P < .05). The decline in melanoma mortality was consistent in Schleswig-Holstein for both sexes and was much greater for both men and women than that in any of the 5 comparison regions, and the decline also was observed in young, middle-aged, and older individuals (Fig. 3).
The most compelling reason to conduct a skin cancer screening is to save lives that otherwise would be lost to melanoma. We used the unique experience of the Schleswig-Holstein population-based skin cancer screening (the SCREEN project) to evaluate the potential of this public health activity to prevent death from melanoma. We observed a decline in mortality in Schleswig-Holstein associated with the SCREEN project but no parallel significant decline in melanoma mortality in the adjacent areas to the north, south, east, and west of Schleswig-Holstein or in all of Germany. The observations reported here provide important evidence that screening may lead to the prevention of a substantial proportion of melanoma deaths in Germany and elsewhere.
Mortality rates for women in Schleswig-Holstein were quite constant at rates of 1.4 per 100,000 from 1990 to 2003 and began to drop during and immediately after the implementation of the state wide program. In 2009, mortality rates were almost 50% lower than they were during the 1990 to 2003 period. Mortality rates for men in Schleswig-Holstein ranged between 1.8 and 2.1 per 100,000 from 1990 to 2006 and also dropped by almost 50% to 1.0 per 100,000 during the most recent period.
Our data do not represent absolute proof that skin cancer screening caused the decline in melanoma deaths, and we must be concerned about possible bias or confounding. We considered the possibility that there may be a problem in counting melanoma deaths in Schleswig-Holstein and the adjacent areas, but we did not observe any such problem, and the quality of these data have historically been regarded as high. To further address the possibility of an isolated change in the coding practice of death certificates in Schleswig-Holstein, we contacted the state-wide statistics office to confirm that neither changes in coding practice nor changes in other procedures had occurred during the study years. Furthermore, the statistics office in Schleswig-Holstein also processed data from Hamburg (which adjoins the federal state to the south), in which a similar decline in melanoma mortality was not observed. We also considered whether initiatives to prevent melanoma, such as strong primary prevention programs, which should lead to a decline in incidence, were responsible for the sharp decline in mortality. We observed neither differences in primary prevention in the surrounding areas nor a decline in melanoma incidence in Schleswig-Holstein. Ideally, we would have preferred to confirm these findings with matched mortality data using additional measures, such as thickness-specific incidence rates; however, this information was not available at the time this report was prepared.
We considered possible improvements in therapy as an explanation for the observed mortality reduction, but melanoma therapy did not change substantially in the study years, nor did the therapeutic approach used in Schleswig-Holstein differ substantially from the approaches used in the other regions that were evaluated. Furthermore, the prognosis for patients with late-stage melanoma remained almost unchanged, as reflected by the very stable rates of melanoma mortality in Germany over the last 20 years.4, 13
We searched for large disparities in the mortality decline by age and sex but observed consistent results. We identified no factors, apart from the screening itself and its accompanying educational activities (Table 1), that could explain the difference between mortality trends in Schleswig-Holstein and in adjacent areas.
A limitation of our study is that it was not possible to quantify all of the potential factors that led to the timing (early onset) and the magnitude of the mortality decline in Schleswig-Holstein. Factors that may have contributed to the mortality decline include 1) preliminary screening activities, physician training, and public awareness campaigns in Schleswig-Holstein before the formal start of the state-wide screening in mid-2003 and 2) a high participation rate of at-risk groups in the SCREEN project, as detailed below.
1) The lag time expected between screening activity and the subsequent mortality decline is unclear and should be investigated with appropriate modeling. Data presented here suggest that the lag time may be as short as a few years. Australian data indicating that individuals with a clinical skin examination within 3 years before diagnosis had a decreased risk of thick melanoma provide further evidence that early onset mortality reduction is conceivable.18 Before the launch of the SCREEN project throughout Schleswig-Holstein in July 2003, 2 separate but interconnected activities took place: preliminary skin cancer screening activities and efforts to increase public awareness of skin cancer (Table 1). After the development of skin cancer screening from 1989 on, the first pilot projects in Schleswig-Holstein were implemented in the year 2000 in which approximately 200 nondermatologists and dermatologists participated in an 8-hour training course and 6000 individuals were screened (Table 1). In early 2003, the 8-hour training program greatly expanded to involve more than 1700 physicians in advance of the population-based SCREEN project. Advertisements to recruit participants for the SCREEN project began in early 2003. Therefore, the time span from 2000 to the official start of skin cancer screening in July 2003 can be regarded as a period of awareness building regarding early detection for the public and professionals as well as training of professionals in skin cancer detection. Both factors may have contributed to an early reduction in mortality. In addition, patients who contacted their physicians for other health concerns or because of skin cancer symptoms (which were exclusion criteria for skin cancer screening) may have benefited from improved skills of the physicians because of raised awareness and their skin cancer screening training. This also may have contributed to the mortality reduction.
2) The magnitude of the observed decline in mortality may have been caused in part by the large proportion of participants who belonged to groups at high risk for melanoma. Ten percent of men and 27% of women (aged ≥20 years) were screened in the 1-year period between July 1, 2003 and June 30, 2004. However, 45% of men and 51% of women who were screened had at least 1 defined risk factor for melanoma, indicating that a particularly high-risk group was included in the SCREEN project.11 This is supported by the finding that half of the melanomas registered in the Schleswig-Holstein cancer registry for the period from July 2003 to June 2004 were detected in individuals who had full-body skin examinations as part of the SCREEN project (men, 46%; women, 56%).
It is likely that targeted publicity and screening of high-risk groups, such as middle-aged and older men, within a population-based skin cancer screening campaign will lead to increased participation of these individuals. The worse prognosis of melanoma for middle-aged and older men is well known.8 Increasing the rate of men in this age group who attend screenings would raise the possibility that the impact of screening on mortality would be even greater than that observed in the current analysis. Because of the observational nature of our study, we cannot differentiate clearly between the individual effects of the screening examinations versus heightened awareness as reasons for the observed mortality decline. In principle, a randomized controlled trial is required to provide stronger evidence that screening leads to a reduction in mortality. To our knowledge, there have been no studies to date on mortality after randomized controlled trials for skin cancer screening. Given the expense of conducting such a trial and the number of years required to follow screened and nonscreened individuals, it is highly unlikely that new randomized controlled trials for skin cancer will be funded. We have learned from the prostate-specific antigen screening trials for prostate cancer19, 20 that, even if such a trial to measure the efficacy of skin cancer screening were to be conducted, a meaningful answer would not be guaranteed.
The substantial benefit of skin cancer screening has been demonstrated repeatedly in carefully conducted observational studies. The results include findings that monthly skin self-examination was associated with a substantially lower risk of advanced or fatal melanoma,21 that clinical skin examination was associated with a decreased risk of thick melanoma,18 and that a “natural experiment” of intensive clinical screening and promotion of skin self-examination at a large work place was associated with a major decline in thick melanomas.8
In combination with the data reported herein, we conclude that, although we do not have complete proof from a randomized controlled trial, the evidence is now strong enough to support the efficacy of screening programs for melanoma (with their associated professional training, publicity, and education) as a means to reduce melanoma mortality. In the public health arena, absolute proof is not necessarily required when lives are at stake.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
- 1World Health Organization, International Agency for Research on Cancer. GLOBOCAN 2008. Cancer Incidence and Mortality Worldwide in 2008. Available at: http://globocan.iarc.fr/. Accessed April, 1 2011.
- 2American Cancer Society. Cancer Facts & Figures 2009. Available at: http://www.cancer.org/acs/groups/content/@nho/documents/document/ 500809webpdf.pdf. Accessed January 20, 2011.
- 3Association of the Nordic Cancer Registries. Cancer Stat Fact Sheets. Nordic Countries—Melanoma of Skin. Available at: http://www-dep.iarc.fr/NORDCAN/english/StatsFact.asp?cancer=290&country=0. Accessed March 27, 2012.
- 4Robert Koch-Institute and Association of the Epidemiological Cancer Registries in Germany, eds. Cancer in Germany 2005/2006. Incidence and Trends. 7th ed. Saarbrucken, Germany: Robert Koch-Institute and the Association of the Epidemiological Cancer Registries in Germany; 2010.
- 13Federal Statistics Office. Official mortality statistics on deaths of malignant melanoma. Available at: http://www.gbe-bund.de/oowa921-install/servlet/oowa/aw92/dboowasys921.xwdevkit/xwd_init?gbe.isgbetol/xs_start_neu/&p_aid=3&p_aid=53989859&nummer=327&p_sprache=D&p_indsp=-&p_aid=67908399. Accessed April 1, 2011.
- 14Sundhedsstyrelsen. Dodsarsager, lands od regionstal (ny gruppering). (National Board of Health, Denmark. Causes of death, country and region specific (new groups). Web-based Interactive Database. Available at: http://www.sst.dk/Indberetning%20og%20statistik/Sundhedsdata/Doedsaarsager/DSN_0.aspx. Accessed March 27, 2012.
- 15Chapter 2. Rates and rate standardization. In: Breslow NE, Day NE, eds. Statistical Methods in Cancer Res. Volume II: The Design and Analysis of Cohort Studies. Lyon, France: International Agency for Research on Cancer, 1987: 48-79., .
- 17Regression With Social Data: Modeling Continuous and Limited Response Variables (Wiley Series in Probability and Statistics). 1st ed. New York: Wiley-Interscience; 2004..