- Split View
-
Views
-
Cite
Cite
Markus P. Anders, Sandra Nolte, Annika Waldmann, Marcus Capellaro, Beate Volkmer, Rüdiger Greinert, Eckhard W. Breitbart, The German SCREEN project – design and evaluation of the communication strategy, European Journal of Public Health, Volume 25, Issue 1, February 2015, Pages 150–155, https://doi.org/10.1093/eurpub/cku047
- Share Icon Share
Abstract
Background: Skin cancer is the most common cancer in light-skinned populations worldwide. Primary and secondary preventive activities such as skin cancer screening are intended to reduce skin cancer burden. In 2003, a population-based skin cancer screening project [SCREEN (Skin Cancer Research to Provide Evidence for Effectiveness of Screening in Northern Germany)] was conducted in Northern Germany with more than 360 000 people screened. SCREEN was supported by a communication intervention that was aimed at informing the population about skin cancer, its risk factors and the screening intervention as well as preparing the health professionals for the project. Within SCREEN both physicians and practice nurses were educated in counselling. The aim of the present article is to describe and evaluate the communication strategy accompanying SCREEN. Methods: Two computer-assisted telephone interview surveys were performed in April/May 2003 and May 2004. Participants had to be members of the statutory health insurance and be aged ≥20 years. They were asked about knowledge of skin cancer, perception of physicians’ performance and skin cancer screening in general. Data are mainly presented in a descriptive manner. For statistical analyses, Mann–Whitney U test and Pearson’s chi-square test were used. Results: Knowledge about sunburn in childhood and high ultraviolet exposure as skin cancer risk factors increased during SCREEN. Simultaneously, the awareness for early detection of skin cancer increased significantly from 41.3 to 74.0% (P < 0.001). A total of 21.5% of the interviewees participated in the skin cancer screening project, similar to the population-based participation rate reached. Conclusion: A comprehensive communication strategy accompanying a screening intervention improves the knowledge of potential screenees and may additionally increase the participation rate.
Introduction
Over the past decades, the burden of skin cancer comprising malignant melanoma (MM), basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) in light-skinned populations has increased worldwide.1–6 There is mounting evidence that skin cancer is mainly sun-induced,7,8 with ultraviolet radiation (UVR) being officially classified as a carcinogen by the International Agency for Research on Cancer.8 To reduce both the exposure to UVR and the burden of skin cancer, primary and secondary preventive activities are important to impact on people’s health behaviour. Primary preventive activities are predominantly aimed at increasing people’s awareness and knowledge of skin cancer and corresponding sun-related behaviour.9 Secondary preventive activities are aimed at the early detection of skin cancer.10
To promote primary and secondary preventive activities in Germany, the Subcommittee of Prevention (Unterausschuss Prävention), a body of the German Joint Federal Committee (JFC; Gemeinsamer Bundesausschuss), decided to improve the National Programme on Early Detection of Cancer in Germany in the late 1980s. In this political context, the Association of Dermatological Prevention (ADP; Arbeitsgemeinschaft Dermatologische Prävention e.V.) was founded. Private and governmental organizations such as the German Cancer Aid (Deutsche Krebshilfe e.V.) and the JFC contracted the ADP to develop and implement activities in primary and secondary prevention of skin cancer.
To date, the largest project of the ADP is SCREEN (Skin Cancer Research to Provide Evidence for Effectiveness of Screening in Northern Germany), a systematic population-based skin cancer screening project that was conducted in Schleswig-Holstein, the northernmost federal state of Germany. Methodology, procedure and key findings of the SCREEN project that ran from 1 July 2003 to 30 June 2004 are reported elsewhere.11–13 Results of the project had been crucial for the implementation of the nationwide Skin Cancer Screening Program that was introduced in Germany on 1 July 2008. This programme is targeted at every person from the age of 35 years who is a member of the German statutory health insurance, which applies to ∼90% of the target population. Eligibility qualifies for a free skin cancer screening once every 2 years. When SCREEN started in 2003, Schleswig-Holstein had one of the highest German-wide incidences of MM and non-melanocytic skin cancer (NMSC; Supplementary Table S1).14
Apart from the standardized training of physicians and practice nurses [In Germany, practice nurses (German: Medizinische Fachangestellte) are responsible for the administration in a medical practice. They also assist the physician in medical affairs, e.g. taking a sample of blood, but they are not allowed to decide the medical treatment of a patient) that aimed at preparing physicians and practice nurses for the project, SCREEN integrated a comprehensive communication intervention that was aimed at announcing SCREEN and increasing the population’s knowledge about skin cancer and its risk factors.
The effect of communicative activities on knowledge is well examined in cancer prevention; however, results are controversial. A public campaign in Nashville and Atlanta (USA) showed little or no effects on knowledge. The project was aimed at delivering health information about cancer prevention to African–American communities by broadcasts, print media, lectures and workshops. The main effect observed was an increase in cancer screening experiences for breast and prostate cancer.15 Skinner et al. (2000)16 also found no changes in knowledge in their study that was carried out in St. Louis (USA). The corresponding intervention was aimed at informing women about breast cancer. In contrast, a multimedia programme improved UVR knowledge of participating men significantly, whereas knowledge of women did not change. The electronic device used (touch screen) was implemented in a kiosk located in a Swedish suburban middle-class community.17 In another investigation, Glazebrook et al. (2006) recognized an increase in both knowledge and protective skin behaviour. The study evaluated the application of an interactive multimedia intervention (the Skinsafe programme) to patients with an increased risk of developing skin cancer. The study took place in five family practices in the UK where patients received education in a single self-directed sitting of 10–15 min.18
Health professional training in primary and secondary prevention of skin cancer was also investigated in several studies. A meta-analysis reviewed 20 publications of 13 different educational programmes. All programmes contained education in diagnosis of MM. Epidemiology of skin cancer was part of the education in 10, management and counselling in 8 and dermoscopy in 2 of the programmes. Eighteen of the 20 studies found a significant improvement in at least one of the following outcomes: knowledge, competence, confidence, diagnostic performance of participating health professionals or system outcomes such as behaviours in practice and effects on patients.19 For example, Mikkilineni et al. (2001) assessed the impact of an education intervention targeted at physicians’ counselling in prevention. After the intervention, more patients were counselled in sun protection and skin self-examination (SSE). Furthermore, the participating physicians informed patients more frequently about their individual skin cancer-related risk factors.20
The aim of the present article is to provide a description of the communication intervention that was implemented as part of SCREEN. In addition, the intervention is evaluated at a population level. The evaluation is based on two surveys conducted before and towards the end of the 1-year project, with particular focus on knowledge about skin cancer, skin cancer risk factors and skin cancer screening in potential screenees on one hand and perceived quality of counselling of physicians and practice nurses on the other hand.
The communication intervention
The communication intervention included a social marketing campaign and training courses for physicians as well as practice nurses (see figure 1). The marketing part was aimed at informing the entire population of Schleswig-Holstein (2.8 million) about the screening project. Especially those people (1.7 million) who were eligible to participate in the screening (statutory health insured, aged ≥20 years). Moreover, the educational part of the intervention was targeted at preparing participating health professionals for skin cancer screening activities. The objectives of the intervention were to increase the informed participation in skin examinations and the awareness of skin cancer in the population. The provision of comprehensive information and the counselling in medical practices should also support the self-responsibility of potential sreenees through the increase of their knowledge about skin cancer and its risk factors.
In preparation for the communication intervention, a survey was conducted to collect data on knowledge about skin cancer, skin cancer risk factors and primary and secondary skin cancer prevention of the target population. The Health Belief Model21 and the Stages of Change Model22 (see figure 2) determined the development of the social marketing campaign: advertisement in magazines and newspapers, posters, post cards, flyers and radio broadcasts were aimed at making people think about their skin cancer ‘susceptibility’ and informing them about the ‘severity’ of skin cancer (‘stage of precontemplation and contemplation’). Lectures and newsletters delivered at companies, professional associations (e.g. firemen, farmers, pharmacists) and sports clubs were aimed at reaching eligible screenees in their everyday life. Display materials in medical offices were targeted at patients waiting for their doctor’s appointment. Furthermore, a telephone hotline and a web page were established. Both services provided comprehensive information about benefits and harms of skin cancer screening. Additionally, people were encouraged to talk about the screening with their physicians and/or practice nurses. All materials and services were established to support an informed decision-making, especially at the ‘stage of preparation’.
To be eligible to carry out the screening, all physicians, regardless of their specialization, had to attend a mandatory 8-h training course to get certified. Almost all (98%) dermatologists and about two-thirds (64%) of non-dermatologists (general practitioners, gynaecologists, internists, surgeons, urologists) in the intervention area participated in SCREEN.11 The emphasis of the first part of the training was placed on communication aspects (counselling). The second part of the training was focused on carrying out the screening test, i.e. a visual standardized whole body examination. Furthermore, the intervention included education of physicians’ practice nurses who were trained to be able to inform their patients about the screening.
The training encouraged physicians and practice nurses to offer the screening directly to their patients (‘cue to action’). If a patient participated, the physician performed the screening test and counselled the screenee about UVR protective behaviour including SSE and individual skin cancer risk (‘stage of action’). To support preventive behaviour, the physician handed out additional information about primary and secondary prevention. The provision of a screening booklet, the so-called ‘prevention pass’, was aimed at reminding screenees of their next appointment (‘stage of maintenance’).
Methods
The baseline survey (t0) was carried out in April/May 2003. The same and further questions were asked in a follow-up survey in May 2004 (t1). Both surveys used computer-assisted telephone interviews. Participants had to be aged ≥20 years, be members of the statutory health insurance and be German speaking. Interviews were carried out by interviewers of TNS Health Care, Munich. A multistage sampling process was used to randomly select telephone numbers of study participants by applying a random digit dialling algorithm. The corresponding household was contacted by phone. The person with the next birthday was chosen to participate if there was more than one person from the defined target population.
SCREEN was approved by the JFC that represents all stakeholders of the German statutory health insurance including patient’s representatives. Assessment of the project by the JFC guaranteed adherence of ethical standards during the project.
Variables
Age was grouped into six categories (20–29 years, 30–39 years, 40–49 years, 50–59 years, 60–69 years and ≥70 years). Using the Winkler Index, socio-economic status (SES) was calculated. It is based on four determinants: participant’s education (originally two variables: highest school and occupational education), net household income and current occupational status. Participant’s school and occupational education determined the value of the first component of the index. Then, each component was recoded with a value between 1 and 7 points and added up, leading to a range between 3 and 21 points. SES was then interpreted as follows: between 3 and 8 points, SES was considered low; from 9 to 14, medium; and from 15 to 21, high. If one of the three determinants was missing, it was replaced with the mean of the other two valid determinants.23,24
Interviewers read seven possible risk factors successively to the participants, while interviewees had to decide whether the respective factor increases or decreases the risk of skin cancer or whether the factor has no influence. ‘High UV exposure’, ‘sunburns in childhood’, ‘sunbed use’, ‘naevi’, ‘freckles’ and ‘fair skin’ increase the risk of skin cancer, whereas ‘dark skin’ has no influence on the risk of skin cancer. Finally, a new variable that indicated whether a participant has answered correctly was calculated.
Interviewers also asked whether the interviewees knew of the different terms of skin cancer: ‘black skin cancer’, ‘MM’, ‘BCC’, ‘basaliom’ and a range of German-specific terms for ‘SCC’. In the analysis, all terms that describe SCC were combined into one variable.
In both surveys, interviewers asked whether participants were aware of activities in early detection of skin cancer. In the follow-up survey in May 2004, participants indicated additionally whether they had heard about the introduced ‘skin cancer screening’ and whether they had participated.
Statistical analysis
The sample was weighted by gender and age, with the weighting factor being based on information about all members of statutory health insurance from October 2002, as provided by the German Federal Ministry of Health. Data were mainly analysed in a descriptive manner. To test for potential differences between baseline and follow-up data, the Mann–Whitney U test was applied. Bivariate analyses of respective baseline and follow-up data were performed using Pearson’s chi-square statistic. Statistical analyses were carried out using IBM SPSS Statistics 20.0® (SPSS Inc., Chicago, IL, USA). The level of significance was set at 0.05.
Results
To achieve a sample size of 600 interviews in each survey, interviewers had to call 1055 at t0 (completion rate: 56.9%) and 1056 persons at t1 (completion rate: 56.8%). As shown in table 1, substantially more women participated in the surveys compared with men. Weighted mean age was 49.7 years at baseline and 49.8 years in the follow-up survey. The majority of respondents were married (t0: 62.1%; t1: 62.4%). In both surveys almost half of the participants were of medium SES, whereas at baseline, the proportion of those with low SES was significantly larger than in the follow-up survey (t0: 40.7%; t1:30.9%).
. | . | t0 . | t1 . |
---|---|---|---|
n = 600 . | n = 600 . | ||
Mean (SD; 95% confidence interval) . | Mean (SD; 95% confidence interval) . | ||
. | |||
Age, years . | . | 49.7 (17.06; 48.34–51.07) . | 49.8 (17.04; 48.46–51.19) . |
. | |||
. | Subgroups . | % (n) . | % (n) . |
Age groups, years | 20–29 | 13.2 (79) | 13.2 (79) |
30–39 | 20.4 (122) | 20.4 (122) | |
40–49 | 18.0 (108) | 18.0 (108) | |
50–59 | 15.0 (90) | 15.0 (90) | |
60–69 | 17.1 (103) | 17.1 (103) | |
70– | 16.3 (98) | 16.3 (98) | |
Sex | Female | 55.0 (330) | 55.0 (330) |
Male | 45.0 (270) | 45.0 (270) | |
Marital statusa | Married, living together | 60.9 (365) | 60.4 (363) |
Married, separated | 1.2 (7) | 2.0 (12) | |
Single | 21.0 (126) | 22.8 (137) | |
Divorced | 7.6 (46) | 6.1 (36) | |
Widowed | 9.3 (56) | 8.8 (53) | |
SESa | Low | 40.7 (243) | 30.9 (185) |
Medium | 47.5 (284) | 52.2 (311) | |
High | 11.8 (70) | 16.9 (101) |
. | . | t0 . | t1 . |
---|---|---|---|
n = 600 . | n = 600 . | ||
Mean (SD; 95% confidence interval) . | Mean (SD; 95% confidence interval) . | ||
. | |||
Age, years . | . | 49.7 (17.06; 48.34–51.07) . | 49.8 (17.04; 48.46–51.19) . |
. | |||
. | Subgroups . | % (n) . | % (n) . |
Age groups, years | 20–29 | 13.2 (79) | 13.2 (79) |
30–39 | 20.4 (122) | 20.4 (122) | |
40–49 | 18.0 (108) | 18.0 (108) | |
50–59 | 15.0 (90) | 15.0 (90) | |
60–69 | 17.1 (103) | 17.1 (103) | |
70– | 16.3 (98) | 16.3 (98) | |
Sex | Female | 55.0 (330) | 55.0 (330) |
Male | 45.0 (270) | 45.0 (270) | |
Marital statusa | Married, living together | 60.9 (365) | 60.4 (363) |
Married, separated | 1.2 (7) | 2.0 (12) | |
Single | 21.0 (126) | 22.8 (137) | |
Divorced | 7.6 (46) | 6.1 (36) | |
Widowed | 9.3 (56) | 8.8 (53) | |
SESa | Low | 40.7 (243) | 30.9 (185) |
Medium | 47.5 (284) | 52.2 (311) | |
High | 11.8 (70) | 16.9 (101) |
a: Differences between t0 and t1 significant at the P < 0.05 level (Mann–Whitney U test).
. | . | t0 . | t1 . |
---|---|---|---|
n = 600 . | n = 600 . | ||
Mean (SD; 95% confidence interval) . | Mean (SD; 95% confidence interval) . | ||
. | |||
Age, years . | . | 49.7 (17.06; 48.34–51.07) . | 49.8 (17.04; 48.46–51.19) . |
. | |||
. | Subgroups . | % (n) . | % (n) . |
Age groups, years | 20–29 | 13.2 (79) | 13.2 (79) |
30–39 | 20.4 (122) | 20.4 (122) | |
40–49 | 18.0 (108) | 18.0 (108) | |
50–59 | 15.0 (90) | 15.0 (90) | |
60–69 | 17.1 (103) | 17.1 (103) | |
70– | 16.3 (98) | 16.3 (98) | |
Sex | Female | 55.0 (330) | 55.0 (330) |
Male | 45.0 (270) | 45.0 (270) | |
Marital statusa | Married, living together | 60.9 (365) | 60.4 (363) |
Married, separated | 1.2 (7) | 2.0 (12) | |
Single | 21.0 (126) | 22.8 (137) | |
Divorced | 7.6 (46) | 6.1 (36) | |
Widowed | 9.3 (56) | 8.8 (53) | |
SESa | Low | 40.7 (243) | 30.9 (185) |
Medium | 47.5 (284) | 52.2 (311) | |
High | 11.8 (70) | 16.9 (101) |
. | . | t0 . | t1 . |
---|---|---|---|
n = 600 . | n = 600 . | ||
Mean (SD; 95% confidence interval) . | Mean (SD; 95% confidence interval) . | ||
. | |||
Age, years . | . | 49.7 (17.06; 48.34–51.07) . | 49.8 (17.04; 48.46–51.19) . |
. | |||
. | Subgroups . | % (n) . | % (n) . |
Age groups, years | 20–29 | 13.2 (79) | 13.2 (79) |
30–39 | 20.4 (122) | 20.4 (122) | |
40–49 | 18.0 (108) | 18.0 (108) | |
50–59 | 15.0 (90) | 15.0 (90) | |
60–69 | 17.1 (103) | 17.1 (103) | |
70– | 16.3 (98) | 16.3 (98) | |
Sex | Female | 55.0 (330) | 55.0 (330) |
Male | 45.0 (270) | 45.0 (270) | |
Marital statusa | Married, living together | 60.9 (365) | 60.4 (363) |
Married, separated | 1.2 (7) | 2.0 (12) | |
Single | 21.0 (126) | 22.8 (137) | |
Divorced | 7.6 (46) | 6.1 (36) | |
Widowed | 9.3 (56) | 8.8 (53) | |
SESa | Low | 40.7 (243) | 30.9 (185) |
Medium | 47.5 (284) | 52.2 (311) | |
High | 11.8 (70) | 16.9 (101) |
a: Differences between t0 and t1 significant at the P < 0.05 level (Mann–Whitney U test).
Knowledge about skin cancer risk factors increased during the intervention (table 2). In the follow-up survey, more participants knew that ‘sunburns in childhood’ and ‘high UV exposures’ increase the risk of skin cancer. In contrast, changes in knowledge about ‘sunbed use’ and ‘naevi’ as risk factors of skin cancer were not significant. However, those risk factors were already widely known in the population, with 76.7% (‘sunbed use’) and 74.3% (‘naevi’) having knowledge of respective risk factors at baseline. In contrast, the risk of ‘freckles’, ‘fair skin’ and ‘dark skin’ was generally not judged correctly. Knowledge about some risk factors was significantly different between women and men. Women more frequently assessed ‘high UV exposure’ (96.7 vs. 93.0%; P = 0.038) and ‘sunbed use’ (83.0 vs. 69.1%; P < 0.001) as risk factors of skin cancer at t0, and they also knew more often about ‘naevi’ as a risk factor in both surveys (t0: 84.5 vs. 68.9%; P = 0.003/t1: 79.0 vs. 68.5%; P < 0.001).
. | Correct answera . | Percentage of correct answers . | Comparison of t0 vs. t1b . | |
---|---|---|---|---|
t0 . | t1 . | P-value . | ||
n = 600 . | n = 600 . | |||
% (n) . | % (n) . | |||
High UV exposure | Higher risk | 94.9 (570) | 97.9 (587) | 0.018 |
Sunburns in childhood | Higher risk | 80.2 (481) | 84.6 (507) | 0.018 |
Sunbed use | Higher risk | 76.7 (459) | 79.4 (477) | 0.234 |
Naevi | Higher risk | 74.3 (445) | 77.6 (465) | 0.175 |
Freckles | Higher risk | 44.0 (264) | 45.5 (273) | 0.538 |
Fair skin | Higher risk | 66.5 (399) | 66.6 (399) | 0.563 |
Dark skin | No influence | 28.7 (172) | 27.8 (167) | 0.332 |
. | Correct answera . | Percentage of correct answers . | Comparison of t0 vs. t1b . | |
---|---|---|---|---|
t0 . | t1 . | P-value . | ||
n = 600 . | n = 600 . | |||
% (n) . | % (n) . | |||
High UV exposure | Higher risk | 94.9 (570) | 97.9 (587) | 0.018 |
Sunburns in childhood | Higher risk | 80.2 (481) | 84.6 (507) | 0.018 |
Sunbed use | Higher risk | 76.7 (459) | 79.4 (477) | 0.234 |
Naevi | Higher risk | 74.3 (445) | 77.6 (465) | 0.175 |
Freckles | Higher risk | 44.0 (264) | 45.5 (273) | 0.538 |
Fair skin | Higher risk | 66.5 (399) | 66.6 (399) | 0.563 |
Dark skin | No influence | 28.7 (172) | 27.8 (167) | 0.332 |
a: Example: ‘If I had had sunburns in childhood, the risk of skin cancer would have been (a) increasing, (b) decreasing, (c) remaining constant, (d) I do not know’. [(a) is the only correct answer].
b: Mann–Whitney U test.
. | Correct answera . | Percentage of correct answers . | Comparison of t0 vs. t1b . | |
---|---|---|---|---|
t0 . | t1 . | P-value . | ||
n = 600 . | n = 600 . | |||
% (n) . | % (n) . | |||
High UV exposure | Higher risk | 94.9 (570) | 97.9 (587) | 0.018 |
Sunburns in childhood | Higher risk | 80.2 (481) | 84.6 (507) | 0.018 |
Sunbed use | Higher risk | 76.7 (459) | 79.4 (477) | 0.234 |
Naevi | Higher risk | 74.3 (445) | 77.6 (465) | 0.175 |
Freckles | Higher risk | 44.0 (264) | 45.5 (273) | 0.538 |
Fair skin | Higher risk | 66.5 (399) | 66.6 (399) | 0.563 |
Dark skin | No influence | 28.7 (172) | 27.8 (167) | 0.332 |
. | Correct answera . | Percentage of correct answers . | Comparison of t0 vs. t1b . | |
---|---|---|---|---|
t0 . | t1 . | P-value . | ||
n = 600 . | n = 600 . | |||
% (n) . | % (n) . | |||
High UV exposure | Higher risk | 94.9 (570) | 97.9 (587) | 0.018 |
Sunburns in childhood | Higher risk | 80.2 (481) | 84.6 (507) | 0.018 |
Sunbed use | Higher risk | 76.7 (459) | 79.4 (477) | 0.234 |
Naevi | Higher risk | 74.3 (445) | 77.6 (465) | 0.175 |
Freckles | Higher risk | 44.0 (264) | 45.5 (273) | 0.538 |
Fair skin | Higher risk | 66.5 (399) | 66.6 (399) | 0.563 |
Dark skin | No influence | 28.7 (172) | 27.8 (167) | 0.332 |
a: Example: ‘If I had had sunburns in childhood, the risk of skin cancer would have been (a) increasing, (b) decreasing, (c) remaining constant, (d) I do not know’. [(a) is the only correct answer].
b: Mann–Whitney U test.
Although ‘black skin cancer’ was widely known in both surveys [t0: 80.8% (n = 485); t1: 83.9% (n = 503); P = 0.318], knowledge of the term ‘MM’ increased significantly from 49.5% (n = 297) at baseline to 60.7% (n = 359) at the follow-up survey (P < 0.001). Further, the proportion of interviewees who were familiar with at least one of four items referring to SCC also increased significantly between the two surveys [t0: 32.7% (n = 194); t1: 39.2% (n = 228); P = 0.013]. Finally, the awareness of ‘basalioma’ increased from 12.8 (n = 76) to 16.3% (n = 97), P = 0.025, whereas the awareness of ‘BCC’ did not change [t0: 15.9% (n = 95); t1: 21.0% (n = 125); P = 0.099]. In both surveys, women knew more often all terms for skin cancer compared with men (P ≤ 0.002).
In 2003, 41.3% (n = 232) of respondents were aware that physicians offered an examination for the early detection of skin cancer. This proportion increased significantly to 74.0% (n = 360; P < 0.001) in 2004. Results of the follow-up survey also suggest that 87.6% (n = 524) of the interviewees knew the term ‘skin cancer screening’. The awareness was higher among women (89.7%; n = 295) than among men (84.8%; n = 229; P = 0.074). One hundred and twenty-nine interviewees (21.5%) participated in the screening. Women (26.7%; n = 88) participated more frequently (P = 0.001) than men (15.2%; n = 41). Although patients who were addressed directly in medical offices participated frequently in skin cancer screening (62.9%; n = 83), only 11.9% (n = 46) of patients who were not addressed directly were screened (P < 0.001).
Four in five screenees (n = 102) indicated that their physician informed them about SSE and 51.6% (n = 66) remembered being counselled about UVR. Again, four in five (n=102) screenees felt well informed about skin cancer and skin cancer screening and 94.6% (n=122) remembered the examination and the counselling positively. Eighty-six percent (n=111) stated that physicians had enough time for them, whereas 6.2% (n=8) felt uncomfortable during the examination. Finally, 44.2% (n=57) of the screenees received further information from their physician about other early detection activities.
Discussion
It was found that during the comprehensive communication intervention accompanying the SCREEN project, knowledge about some skin cancer risk factors as well as knowledge about skin cancer in general increased in the target population. Additionally, findings suggest that physicians counselled screenees comprehensively. Glazebrook et al. (2006) also reported an increase in knowledge as a result of their interactive multimedia intervention promoting primary prevention of skin cancer.18 In contrast, Skinner et al. (2000)16 did not recognize significant differences in knowledge in their study. A further analysis showed only little or no effects on cancer knowledge.15 In the presented analysis, the awareness of especially ‘sunburns in childhood’ and ‘high UV exposure’ as risk factors increased. ‘Freckles’ and ‘light skin’ as risk factors were relatively unknown, whereas ‘sunbed use’ and ‘naevi’ were widely known as risk factors but did not increase significantly during the project. Miles et al. (2005)25 reported that 92.4% knew ‘fair skin’ as a risk factor of skin cancer, and awareness of known factors like ‘sunburns in childhood’ (81.5%) and ‘using sunbeds’ (82.4%) was similar to our results. Furthermore, the terms MM, SCC and basalioma were known more frequently after SCREEN. Four of five potential screenees already knew the term black skin cancer both before and towards the end of the intervention. The literature on the awareness of skin cancer in the general population is scarce. A survey conducted in 1989 showed that 92.6% of general practice patients in South Australia had heard of MM.26 Additionally, similar results to our analysis were reported in an international market research survey where 30% of all and 20% of German interviewees confirmed that they were aware of BCC.27 Moreover, women’s knowledge of skin cancer and its risk factors was much higher than that of men, especially at the beginning of the intervention, an observation in line with other studies reporting higher knowledge rates of women in skin cancer risk factors.25 However, over the course of the SCREEN project, we observed that gender differences declined.
Awareness of interventions aimed at the early detection of skin cancer increased during the project, and the term skin cancer screening was introduced successfully. About one in five persons of the target population (21.5%) participated in the screening in the first 11 months, whereas the official participation rate was 19% for the 1-year project.11 In the nationwide skin cancer screening, introduced in July 2008, 24.4% of eligible persons participated over a period of 2 years. Furthermore, more women than men participated in our study, a phenomenon reported by others in this field.28
The performance of screening examination including counselling reached a high standard during the project as most screenees felt well informed and stated that the physician spent enough time for the examination and the counselling. In contrast, only a small proportion indicated feeling embarrassed during the examination. Furthermore, over 80% of screenees remembered instructions for SSE. Martin et al. (2007)29 concluded that the understanding of SSE increases regular and correct performance of self-examination behaviour. Additionally, more than half of the participants remembered receiving consultation about correct handling of UVR. This complies with the postulation formulated by the Council of Europe in 1994. It claims that health professionals who promote activities of secondary prevention, such as early detection interventions, should simultaneously inform potential screenees about primary prevention measures.30 Apart from information about UVR, screenees were informed about other secondary preventive activities. In accordance with our results, Mikkilineni et al. (2001)20 stated that the education of physicians increases counselling in preventive activities.
Patients who were addressed directly by office staff were more likely to participate than patients who were not addressed directly. In this context, practice nurses play an important role, as they are the first point of contact for patients. In an additional survey conducted during SCREEN (data not published), 64% of practice nurses reported that they offered screening to almost all eligible patients; in contrast, only 37% of the physicians did.
The present study has some limitations. Both the baseline and the follow-up survey were based on a cross-sectional design. This limits conclusions about individual change. Furthermore, the complexity of the intervention hampers a thorough evaluation. Although our findings suggest an influence of the intervention on the population’s knowledge, there is no empirical evidence of causal association. Because of the lack of a control cohort, we cannot rule out that factors beyond the intervention influenced the results. A different distribution of socio-economic factors in the two surveys could also bias our results. Sensitivity analyses (Supplementary Tables S2 and Supplementary Data) show that in some cases persons with a high SES benefited more from the intervention, but nevertheless increases in knowledge can be observed in all socio-economic groups. Analyses also show that screenees benefited more than non-participants. Moreover, aspects such as changes in UVR-related behaviour or people’s exposure to UVR were not included in the evaluation. Finally, no comprehensive cost-effectiveness analysis was performed within SCREEN. The costs of the information materials amounted to ∼1.1 million €, i.e. ∼3 € per recruited screenee (data not published). This has not yet been valued against potential savings through the project such as those possible through stage shifts towards earlier less-advanced stages of skin cancer achieved through the earlier detection of malignant skin tumours.
Conclusion
A comprehensive communication strategy as implemented during SCREEN improves outcomes in knowledge. The lack of knowledge among men compared with women was reduced during the project. Apart from social marketing activities, both physicians and practice nurses play a crucial role in the recruitment of screenees. Further, such communication strategy accompanying a screening intervention may also increase the participation rate. In future research, however, elements used in the present intervention should be tested separately and in different combinations to detect the most effective information and recruitment strategies. Finally, training of included health professionals guarantees a high standard of the counselling in screening.
Funding
The sponsors of SCREEN (German Cancer Aid (Deutsche Krebshilfe e.V.), National Association of Statutory Health Insurance Physicians (Kassenärztliche Bundesvereinigung), German Federal Association of Health Insurance Funds (Spitzenverbände der Krankenkassen)) had no role in the design and conduct of the study, in the collection, analysis, and interpretation of data, or in the preparation, review, or approval of the presented publication.
Conflicts of interest: None declared.
Skin cancer is the most common cancer in developed countries. Activities of primary and secondary prevention are predestined to reduce the burden of skin cancer by improving sun-related behaviour and offering skin cancer screening.
The presented analysis evaluates a population-based communication intervention accompanied with a skin cancer screening project (SCREEN) conducted in Northern Germany. Two surveys during SCREEN investigate changes in knowledge and quality of physicians’ counselling.
The findings suggest that a comprehensive communication intervention is able to increase knowledge of potential screenees significantly. Moreover, high quality and satisfaction with screening counselling can be observed after physicians and practice nurses were trained.
Comments