8. Secondary Prevention – Prevention of Skin Cancer

Secondary prevention aims at the early detection and prevention of the progression of a disease. Secondary prevention can be described as an umbrella term that is centrally linked to the concept and possibilities of early detection. Early detection of diseases focuses on the timely identification of a disease, disorder, malformation, or other health impairment. In order to be suitable for early detection, it is therefore necessary for diseases to have a long symptom-free phase in which precursors or early stages of the disease are already detectable. By detecting disease at an early stage, secondary prevention aims to reduce or prevent mortality, morbidity, and thereby impaired quality of life.

Screening is a key component of early detection. The term "screening" comes from the English language and means "filter examination." Screening is designed to detect precursors, early stages, and risk factors of a disease. In screening, "undetected diseases or defects are detected using rapid and large-scale tests […]. Screening tests distinguish between apparently healthy people and those who may have the disease" [639] . Not every screening test takes the form of systematic screening. A screening is part of an overarching program that predefines the target population, examination method, and testing procedures in relation to the target disease. Such a screening programme is characterised by "addressing all persons of a defined target group and specifications for quality assurance from the first invitation to the evaluation of the endpoints" [640]. It complies with defined and verifiable quality standards in that the physicians carrying out the screening not only use a standardised examination method, but also document their examination results, thus making the screening evaluable.

SCS includes the recruitment of apparently healthy participants, collection of medical history and visual whole-body inspection (screening test) for the early detection of malignant skin tumours. As a rule, advice on risk factors and prevention of skin cancer should also be given in this context.

The term precaution, which is also frequently used in the context of secondary prevention, is a term used in the legal sense of the coverage of a risk by, e.g., an insurance. However, this term is misleading, as it could give the impression that regular participation prevents the occurrence of cancer. In the context of secondary prevention, the term early detection should therefore be used, as this is also implemented in the Cancer Early Detection Guideline of the G-BA.

The core concept of screening is the assumption that by diagnosing a disease (e.g. cancer) at an early stage, treatment is more likely to be successful and the risk of morbidity and mortality is reduced. This assumption implicitly assumes that if a disease is left untreated, the prognosis will worsen as it progresses. In the case of MM and SCC, reduced tumour thickness (i.e. a skin lesion at a less advanced stage) is the most important prognostic factor for improved survival (see, for example, for squamous cell carcinoma [641]). For patients with a BCC, detection of the disease at an early stage means improved quality of life due to fewer necessary therapeutic interventions.

For a cancer entity to be eligible for screening, a (long) "preclinical phase" in which early detection is possible is an essential prerequisite[642]. The disease starts at a certain point in time without being detectable yet. Only later can the disease be diagnosed, e.g. when a solid tumour has reached a minimal (i.e. visible) size. The phase before a disease would be diagnosed even without screening is known as the preclinical phase or "sojourn time" [643]. Accordingly, screening can only lead to an earlier diagnosis during this preclinical phase. The period of time by which the diagnosis is advanced is known as the "lead time." Neither the "lead time" nor the preclinical phase can be determined for single individuals. However, the distribution of these two time periods can be estimated for a population that has been screened. It is expected that in a screened group the average age at the time of diagnosis is lower (by the value of the respective "lead time") than in a comparison group without screening (Spix & Blettner, 2012).

Since the introduction of screening measures, the debate about cancer screening program endpoints has been in flux. Schwartz and others [644] defined increased quality of life and prolonged life span as measures of the effectiveness of the statutory cancer screening examination (CFE) program. Today, however, reduction in mortality is often considered the overriding goal. Therefore, the reduction of unfavourable disease stages, increase of life expectancy, reduction of morbidity, avoidance of unnecessary examinations (e.g. biopsies), reduction of costs of expensive treatments of advanced diseases, improvement of quality of life, as well as informing the population about the screening measures, are formulated as goals. These goals are also the parameters for the outcome-oriented evaluation of screening measures. 

The target group consists of persons who subjectively feel healthy and are also predominantly healthy. The target group does not include persons who come for screening because of complaints or with symptoms or who are in follow-up care. Therefore, few positive test results are to be expected in a screening.

According to Morrison [645], a screening divides the participants into "persons with a low probability of having the disease" and "persons with a high probability of having the disease," whereby the second group is to be submitted to follow-up diagnostics (suspected diagnostics and/or confirmatory diagnostics, see 8.5)

 to confirm the diagnosis. Accordingly, screening refers neither to diagnosis nor to treatment. If the examination is extended to the entire body for the diagnosis of a self-discovered skin lesion, this can also be referred to as "screening."

According to the screening criteria of Wilson and Jungner (1968), screening for cancer should cover the following points [642]:

1. The target disease should be a major health problem,
2. The natural history of the disease should be adequately understood,
3. There should be an identifiable early stage,
4. Treatment should be more effective in the early stage than in the late stage,
5. An effective test to detect early stages should be available,
6. The test should be acceptable,
7. Examination intervals should be known/established,
8. Adequate health care resources to cover the additional costs arising from screening,
9. Both physical and psychological risks should be less than the benefits,
10. The costs should be balanced with the benefits.

These were revised by Andermann et al. [646] based on more recent discussion. The following criteria are mentioned by them:

• The screening program should address an identified need/societal problem,
• The goals of the screening should be defined at the beginning,
• There should be a defined target population,
• There should be scientific evidence of the effectiveness of the screening program,
• The program should include as components statements about clinician training, tests used, clinical procedures, and program management,
• There should be quality assurance with a mechanism to minimize potential screening risk,
• The program should ensure informed choice, confidentiality, and respect for the autonomy of the individual,
• The program should ensure equity and provide access to screening for the entire target population,
• Program evaluation should be planned from the outset,
• The overall benefits of screening should outweigh the harms.

Skin cancer meets most, if not all, of these criteria, and is therefore suitable for screening. However, more evidence is needed to evaluate all of these 10 criteria, e.g. the harms of a SCS, cost-effectiveness, as well as the question of the appropriate screening interval.

In a screening programme, screening is usually carried out by specially trained clinicians. One way to screen is to screen an entire population (population-based or mass screening) without dividing the population into subgroups. Another possibility is a risk group screening only in certain population groups, e.g. in persons who have a higher probability of contracting skin cancer [642]. 

Given the dependence of the positive predictive value (PPV, see below) on the prevalence of a disease in a given population, and because prevalence is by definition higher in at-risk groups, a screening programme will usually be most productive and effective if it targets high-risk individuals (better PPV in at-risk groups compared with PPV in non-risk groups).

A screening programme may be systematic, by which is meant a highly organised programme with standardised and quality-assured screening. In contrast, a decentralized unsystematic screening is also referred to as "opportunistic" or "grey" screening [642]. 

Due to the lack of a standardized screening procedure, evaluations of opportunistic screening are difficult to conduct.

A systematic screening program should include the following components [647]; [643]:

• A target population,
• A recruitment strategy (strategies often include mass media campaigns with information specific to the target population and invitation or reminder letters personally addressed to those eligible to participate),
• A screening test,
• A standardized training program for performing physicians,
• A screening interval,
• Follow-up of patients,
• Evaluation.
• To generate evidence regarding the effectiveness of any screening program, the development of a comprehensive evaluation strategy is essential. This includes outcome evaluation in terms of mortality, incidence of stages, morbidity, and quality of life, as well as process evaluation, evaluation of training and evaluation of campaigns and the overall recruitment strategy.

The sole purpose of a screening test is to identify suspicious findings within a specific target group. Four groups are formed: the true positives (diseased and recognized as such), the false positives (not diseased but diagnosed as diseased), the false negatives (diseased but found to be healthy) and the true negatives (healthy and recognized as such).

The sensitivity of a diagnostic test procedure indicates the percentage of diseased patients in which the respective disease is actually detected by the application of the test.

The specificity of a diagnostic test procedure indicates the probability that actually healthy persons who do not suffer from the disease in question are also detected as healthy by the test.

The positive predictive value (PPV) or positive predictive value indicates how many people who have been diagnosed with a certain disease by a test procedure are actually ill. The PPV is influenced by the prevalence of the disease in the population. The higher the prevalence, the higher the PPV. [648]

The negative predictive value (NPV) or negative predictive value indicates how many people who have not been diagnosed with a particular disease by a testing procedure are actually healthy.

Cancer is a progressive disease and screening generally aims to detect cancer at an earlier stage than would be diagnosed in mainstream care - particularly before the tumour has reached an invasive stage.

A screening test, like almost any diagnostic test, is rarely 100% sensitive (i.e. all individuals who have the disease are detected as "true-positive") precisely because it is not a diagnostic test. Typically, no "gold standard" exists for comparing disease status. Most participants in the SCS are "true-negative' (negative test result and free of the disease) or "false-positive," i.e., a positive test result although the disease is not present. Only a fraction of screened individuals are "true-positive," i.e., they have a positive test and are affected by the disease.

The legal basis for cancer screening examinations is:

• Act to Improve the Rights of Patients (Patients' Rights Act), entered into force February 2013. The focus of the Act is on the standardisation of the treatment contract and the associated obligations of those providing treatment (e.g. the structuring of the obligations to provide information or clarification and regulations on the documentation of treatment), the strengthening of the rights of patients vis-à-vis the service providers and in the event of treatment errors in the context of social insurance law, as well as the strengthening of patient participation in self-administration.
• Act to Strengthen Health Promotion and Prevention (Prevention Act - PrävG), entered into force July 2015. Amendment of the Fifth Social Code (SGB V), §20. The focus of the Act is to improve the basis for cooperation between social insurance providers, Länder, and local authorities in the areas of prevention and health promotion. In addition, early detection examinations in all age groups are expanded and important measures for vaccination protection are regulated.
• Act on the Further Development of Early Cancer Detection and on Quality Assurance through Clinical Cancer Registries (Early Cancer Detection and Registries Act [KFRG]), which came into force in April 2013. The focus of the Act is the creation of the legal and financial framework for the establishment and operation of nationwide clinical cancer registries.
• Recommendations of the Council of the EU on Cancer Screening, December 2003. Promotes and supports cancer screening in all states of the European Community and makes recommendations on its implementation.
• Guideline of the Federal Joint Committee on the Early Detection of Cancer (Cancer Early Detection Guideline/KFE-RL). The guideline regulates the medical measures for the early detection of cancer, in particular the scope and timing of services, documentation, and evaluation.
• Guideline for Organised Cancer Screening Programmes (oKFE-RL). The guideline provides details on the implementation of organised cancer screening programmes. Essential structural elements are a regular invitation, combined with accompanying information for the insured persons about the respective examination, data protection, rights of objection, as well as the performance of the examination and the programme evaluation.

Although screening interventions have the potential to reduce the mortality of a disease and save lives, death will not always be avoidable as an "outcome." Some individuals will die from cancer despite participation in screening because the tumor was not detected at screening, the diagnosed tumor does not respond to treatment, or because it is already at an advanced stage at the time of diagnosis.

In patients with a false-negative test result (they have a negative test result despite having the disease), the deceptive certainty can lead to a delayed diagnosis. In this case, the tumor remains undetected until it leads to symptoms or is discovered during the next screening examination.

Patients with false-positive test results (they have a positive test result although no disease is present) may undergo unnecessary multiple excisions. In addition, the time until the harmless / benign histopathological finding is experienced by many patients as a great psychological burden.

Interval carcinomas, ie tumors that are detected between two screening examinations [649][649] can also occur despite an effective screening program. These are not false-negative results. Interval carcinomas occur either because the screening interval is too long or because the patient is affected by a particularly fast-growing tumour. Missed tumours (false-negative test) are also counted as interval carcinomas. It is not possible to distinguish between the intermediate and the overlooked tumours at SCS.

Finally, during screening, one will also detect very slow growing tumours that, theoretically/statistically, are unlikely to have ever harmed the patient or become life-threatening to the patient at any time, the so-called overdiagnosis. In these cases, further diagnosis or treatment could harm rather than benefit these individuals.

The following must be considered when addressing the issue of overdiagnosis: the current scientific definition describes overdiagnosis as the detection of the early stages of a disease that would not in itself alter the user's lifespan or quality of life. The diagnosis and treatment therefore have a higher impact on the patient's quality of life and generate higher costs than the actual disease would ever have done without the early diagnosis.

Physicians are obligated from a medical ethics perspective to act according to the principle of harm avoidance and principle of care. From the perspective of the definition of overdiagnosis, surgical removal of a carcinoma would be equivalent to harm. Since the aim of screening is to detect tumours at an early and thus symptom-free stage, it is unclear to what extent the tumour would have affected the patient's health in the further course. Therefore, the dilemma exists that physicians act in the sense of care and patient well-being but also in the sense of a high-quality screening program if tumours are removed at this early stage in order to avert possible further harm to the patient, even if they can be described as overdiagnosis from an epidemiological point of view.

Numerous studies have addressed the potential benefits of SCS for individuals as well as for the community (eg, reduction in mortality, morbidity, and increase in quality of life). However, because screening aims to examine and test healthy individuals, potential harms and risks associated with these procedures need to be examined even more carefully.

Despite the rising incidence rates of MM, BCC and SCC worldwide, Germany is the only country with a nationwide population-based skin cancer screening program. Many countries with a higher burden of skin cancer remain reluctant to implement such a program. In order to introduce a screening program, the potential benefits must outweigh the potential risks and harms. Only then can the screening of apparently healthy populations be justified. Because studies are lacking on the burdens of a screening program on individuals and the health care system and because effectiveness has not yet been conclusively demonstrated, many organizations, including the United States Prevention Services Task Force (USPSTF, [653] ; [654]  and the Australian Cancer Council (published September 2019) with the New Zealand Guidelines Group [655][655] do not recommend routine SCS.

Many burdens are caused by screening tests because these tests also produce "false-positive" and "false-negative" results.

The following potential risks and harms are associated with false-positive test results:

• Many studies showed that suspicious skin lesions turn out to be benign lesions. The proportion of histopathologically confirmed benign lesions ranges from 70% to almost 90% [206] ; [651] ; [652] ; [650] . Only a small proportion of patients receive a "correct-positive" screening result. Sharing this knowledge by physicians with their patients could reduce the pressure and anxiety of patients with potentially "false-positive" test results.
• Superfluous further testing and/or investigation can lead to complications and harm the patient. In the case of skin cancer screening, unnecessary biopsies can cause complications in addition to unsightly and numerous scars. When these scars occur on visible parts of the body, such as the face, it can cause psychological stress to the patient. Different data are given on the number of excisions (NNE) needed to find a malignant skin tumour. For example, a skin cancer screening study showed an excision rate per newly diagnosed MM of 179:1, indicating low diagnostic specificity [650]. In contrast, in the German SCREEN project only 27 excisions had to be performed to detect MM, while the NNE for BCC was 8:1 and SCC 41:1 [206] . In the SCREEN project, dermatologists and other qualified physicians received additional training on early detection  [206] , which was not the case in the study by Schmitt et al [650] . This may have led to the lower NNE and highlights the importance of specific training and education for physicians participating in skin cancer screening.
• Expensive unnecessary excisions can be a burden on the health care system, as can overdiagnosis and unnecessary treatment.
• Quality of life may be affected by worry and stress during the waiting period for the final (negative) examination result. These negative psychological effects depend mainly on the amount of information available to the screening participant and also on the communication skills of the physician.
• Legal action may be taken by those who have been affected by complications during subsequent procedures. This may reduce public confidence in screening.

The following potential risks and harms are associated with false-negative test results:

• False-negative results can lead to a deceptive sense of security; for example, the patient may cancel medical appointments because the earlier screening suggested that everything was fine. In this case, the tumour remains undetected until it becomes apparent on its own or is discovered in the next round of screening. By then it may be too late for treatment or the tumour may be at a more advanced stage than it might have been if it had been diagnosed clinically without a screening offer. This can lead to increased morbidity, expensive therapies, and reduced quality of life due to the delayed diagnosis. In the extreme case of MM, which is associated with a potentially high likelihood of metastasis, a false-negative test result can potentially lead to death. Osborne et al (2003) reported on the accuracy to diagnose "false-negative" in different clinics. They found that the number of "false-negative" was lowest in specialized skin clinics ("pigmented lesion clinics"). They conclude that the experience of dermatologists working in such specialized clinics may be responsible for the improved diagnostic precision [656].
• Legal action may be taken by those affected by late-stage skin cancer despite having attended a SCS. This may also reduce public confidence in screening measures.

Most SCS participants have a "true-negative" test result and benefit from the SCS because physician confirmation that they are healthy is taken as positive. Patients with "true-positive" results may be pressured by the diagnosis because their disease phase is prolonged by the earlier diagnosis due to screening and must wait to see if they benefit from prompt treatment [651] ; [643] . Furthermore, delays in referral of suspicious lesions (through dermatologist consultation or more advanced procedures) may increase potential harms, e.g., increase in tumour thickness of MM and a decrease in survival rates of MM [657] .

Whole-body examination is performed without technical aids. It is a safe, inexpensive, and non-invasive screening test. In addition, it is not painful for patients or excessively time consuming for physicians. To date, there are no known disadvantages that directly result from a full body examination, other than the fact that the participant may be uncomfortable completely undressing for screening.

A skin tumour not detected during a screening may reach a symptomatic stage before the next screening examination is due. These tumours are called "interval carcinomas." Accordingly, "false-negative" test results may be used to determine the appropriate screening interval for SCS. This would reduce the potential negative consequences "false-negative" results, but increase the "false-positive." A screening interval that is too short, such as every three months for high-risk individuals, could cause long-term psychological distress and adversely affect a person's quality of life.

The possible harms listed are summarized and contrasted with possible benefits in the following table:

Need for Research

Most studies on negative consequences of SCS focus on unnecessary biopsies and the NNE. Further research on other factors that have an impact on potential harms is needed.

The following points should be considered:

• Investigate risk factors for "false-positive" and "false-negative" results in both high-risk groups and individuals without high-risk features,
• Overdiagnosis in SCS needs to be investigated,
• Interval cancer studies to determine optimal screening interval and reduce "false-negative" results,
• the NNE of trained vs. untrained physicians and the resulting consequences for patients,
• the communication skills of physicians and medical assistants to educate participants about potential benefits and risks of SCS,
• the negative psychological effects associated with the uncertainty of potential "false-negative" and "false-positive" results of a SCS,
• the negative impact of screening on clinicians (time, etc.).

Revision by G. Mehlhorn

As described in Chapter 4.3, the individual risk of developing skin cancer varies greatly. It depends on environmental, genetic, and acquired risk factors (e.g. immunosuppression in organ transplant patients). This chapter considers individuals at increased risk of skin cancer and reviews scientific evidence on screening these individuals.

High-Risk Individuals and Skin Cancer Screening

• Targeted SCS of high-risk individuals appears to be more efficient compared with population-based screening. However, the applicability at the population level has been unclear.

In the systematic review by Watts et al. [681], a total of 34 guidelines were included, with the aim of identifying strengths and weaknesses in the evidence for detecting, screening, and following up high-risk groups for cutaneous melanoma. Thirty-two percent of the studies made recommendations for screening high-risk groups, stating that long-term screening was necessary for patients at high risk, especially if there was a genetic predisposition and a family history of frequent melanoma occurrence. Screening should be done for patients with dysplastic nevi according to 32% of the guidelines and if there is a positive family history in 26% of the guidelines.

Guther et al. [652] attempted to develop a model to identify individuals at increased risk of skin cancer who would benefit from regular SCS. They used an open prospective point-prevalence study of consecutive patients who presented to dermatologists for a complete skin examination. Demographic characteristics and risk factors for skin cancer were documented, as was the histology of skin lesions. Results were analyzed univariately and multivariately, and a risk group model was developed to identify patients most likely to develop MM or NMSC [652]. The sensitivity of the risk model in contrast to the current mass screening in Germany can be described as follows: with a set sensitivity of >92% for melanoma and squamous cell carcinoma, one third of the study population is identified as a risk group and suggested for regular SCS. Overall specificity for the model (MM, SCC, BCC) is 67.24% versus 40.0%, sensitivity for melanoma is 92.3% versus 75.2%, sensitivity for squamous cell carcinoma is 92.4% versus 99.0%, and sensitivity for BCC is 61.8% versus 96.1%. Practical implementation at the population level has not been provided by the authors.

In the cohort study by Rat et al. (2015), recruitment of participants was done by 78 general practitioners in France. Participants are patients with increased risk of skin cancer, determined with a score collected independently of the reason for the visit to the doctor, without previous melanoma (n=3,923). In Rath et al. 2015, the SAM score was used (Self-Assessment of Melanoma risk score). In the score, questions about skin type, presence of freckles, nevian count >20 on both forearms, stay of one year in a country with increased sun exposure, melanoma history, and family history of melanoma are considered. A clinical SCS of the patients was done by the general practitioner, followed by a decision to refer to a dermatologist. At the dermatologist's office, a classification of the suspicious lesion was then performed with possible subsequent removal. This was followed by a telephone survey of patients over the course of treatment from 5 September 2012 to 14 October 2012. The follow-up period for each patient was one year. The cumulative melanoma incidence was 229.4/100,000 for the defined high-risk group. Melanoma occurred in nine of 3,923 patients, all with Breslow index <1. For patient compliance, there was a significant association with age (OR=1.02 per year, 95% CI 1.02-1.03, p<0.001), dermatology physician density high versus low (OR=2.28, 95% CI 1.78-2.92, p<0,001), the designation of a dermatologist by the general practitioner (OR=2.15, 95% CI 1.51-3.09, p<0.001), and the 4.2 times higher likelihood of seeing a dermatologist for 65-year-old patients compared to 20-year-old patients.

A follow-up cohort study of the study just described was done by Rat et al. (2015) in the same year. An invitation was sent by email to re-present to the GP one year after inclusion in the cohort described above. Patients who developed melanoma (n=9) or died (n=11) within one year of starting the original participation were excluded. The melanoma incidence of participants was now 160/100,000. Eighty-three participants had a lesion removed, six patients were diagnosed with melanoma, five with squamous cell carcinoma, and 15 with BCC. The incidence of melanoma was 183.7/100,000 in men and 98.7/100,000 in women. Men over the age of 50 had an increased proportion of lesions removed (21% versus 11.6%, p=0.029) and a higher proportion of malignant lesions identified after removal (66.7% versus 21.5%, p<0.001). Patients over 60 years of age had a higher proportion of identified malignant lesions after removal (66.7% versus 19.4%, p<0.001). Patient compliance is described below. After written re-presentation, 54% of participants consulted the general practitioner, 7% directly consulted the dermatologist, and 31% of participants had no SCS. The proportion of patients who were referred to and saw a dermatologist increased (68.8% versus 59.1%, p<0.001). The proportion of patients who consulted a dermatologist overall decreased (15.8% versus 23.9%, p<0.001). This can be explained by the fact that the proportion of patients who were referred to a dermatologist decreased (12.2% versus 38.8%, p<0.001).

Patients at Risk and Suspected/Confirmed Diagnosis

In the cohort study by Moloney et al. (2014), a total of 311 participants recruited through the Sydney Melanoma Diagnostic Centre and Melanoma Institute Australia outpatient clinic from 2006 to 2009 were allocated to one or more of the following groups. Group 1 (n=219) includes all patients with personal history of at least one invasive melanoma and syndrome of dysplastic nevi (at least 100 nevi, of which at least six have atypical changes and at least one nevi is larger than 8mm). Group 2 (n=52) was assigned to patients with personal history of at least one invasive melanoma and family history of malignant melanoma in at least three first- and second-degree relatives. Group 3 (n=146) included patients with personal history of at least two primary invasive melanomas. For patients with only two melanomas, one of these must have occurred in the ten years prior to the patient's enrollment. The final group, group 4 (n=17), includes all patients with confirmed CDKN2A (OMIM 600160) or CDK4 (OMIM 123829) gene mutations. At baseline, all nevi were recorded, whole-body skin screening was performed, and suspicious nevi were examined. Dermatologic examination of patients was performed every six months with a median follow-up duration of 3.5 years (2.4-4.2 years). At baseline, 14 melanomas were diagnosed, with a total of 75 during the course of the study. The risk of developing a new melanoma was 12.7% after two years and 18.2% after four years. A comparison of the incidence of new primary melanomas in the last three study years to the first two study years showed the following: Incidence density ratio = 0.43 (95 CI 0.25-0.74, p=0.002). The effect of diagnostic tools in identifying new melanomas revealed the following: 16% of melanomas were detected by dermatoscopic examination of dermatologists without change in whole-body review photography (TBP)/sequential digital dermoscopy SDDI. Eight percent of melanomas were self-discovered by the patient without TBP and 91% of postbaseline detected melanomas had a Breslow tumour thickness of <1mm. Sequential digital dermoscopy (SDD/SDDI) detected 39% of postbaseline melanomas (16% by short-term SDDI, 23% by long-term SDDI). TBP detected 38% (n=23) of postbaseline melanomas (20 by TBP, 3 by TBP alone). None of the diagnosed melanomas were detected solely by changes in TBP by the patient.

In the systematic review by Watts [681], 70% of the guidelines recommend specific training for users of dermoscopy, whole-body photography (for patients with high numbers of nevi, for early detection of lesions), and sequential digital dermoscopy (to improve diagnostic accuracy). Photographic documentation of changes is recommended by ten guidelines. Prophylactic removal of nevi is not recommended by any guideline. The evidence base for targeted screening with dermoscopy and SDDI to improve diagnostic accuracy is considered very good. The evidence base for whole-body photography is lower.

Examination Intervals of Patients at Risk

A guideline from Australia and New Zealand (Clinical Practice Guidelines for the Management of Melanoma in Australia and New Zealand) states that regular surveillance reduces the average thickness of MM. For high-risk individuals, this guideline recommends a combination of self-examination and screening and advises a screening interval of six months. However, no studies are available that systematically compare alternative methods; these recommendations are based on expert opinion only. The individual frequency of skin examinations in persons at risk should be made dependent on the individual risk factors, i.e. the frequency should be determined in such a way that a reduction of mortality and morbidity as well as a stage shift is achieved.

In the above systematic review by Watts [681], 35% of the included guidelines recommend screening based on prior risk assessment with intervals from six months to annually at regular intervals or for life. Recommendations for screening intervals and –duration for high-risk groups are based on EC.

Overall, there is insufficient evidence regarding optimal screening intervals for individuals at increased risk of skin cancer.

High-Risk Individuals and Skin Self-Examination 

There are different results regarding the benefit of skin self-examination. Because the effectiveness of skin self-examination has already been discussed in sufficient detail in Chapter 8.1.6.1, it is only briefly touched on here in the context of recommending SSE for people at increased risk of skin cancer.

Oliveria et al. [682] showed that regular self-examination has a benefit in detecting thin lesions. Identifying factors associated with performing skin self-examination (gender, age, education, marital status, "skin awareness," previous benign biopsy, presence of atypical moles) will allow health care providers to focus on those individuals who do not perform skin self-examination despite an increased risk of melanoma. They emphasize the importance of identifying factors that increase the likelihood of an individual self-examining their skin, as early detection and excision of lesions has the potential to reduce morbidity and mortality from MM [682].

The review by Watts [681] showed that as part of patient education, skin self-screening is included in 76% of guidelines and specifically named for managing high-risk groups in 38% of guidelines. The intervals for self-screening varied from monthly, tri-monthly, six-monthly, to no indication. These recommendations are based on EC.

Need for Research

Studies should be conducted comparing the effectiveness of mass screening with risk group screening for individuals at increased risk of skin cancer and evaluating it in terms of mortality, morbidity, and stage shift. In this context, economic aspects should also be considered.

Since 1 July 2008, the SCS has been part of the statutory early cancer detection (regulated by the Cancer Early Detection Guideline of the Joint Federal Committee). Statutorily insured persons over the age of 35 are entitled to participate every two years. A new screening examination for skin cancer is only possible after the end of the calendar year following the previous examination. The SCS is designed as a two-stage mass screening. Either the participant is examined by a general practitioner and, if skin cancer is suspected, is referred to a dermatologist who carries out the standardised visual full-body examination again, or the dermatologist carries out the SCS directly.

The following figure illustrates the screening procedure:

J. Hübner

This chapter reviews and evaluates studies that address the population-based German SCS and the preceding feasibility study "SCREEN" (Skin Cancer Research to Provide Evidence for Effectiveness of Screening in Northern Germany). International studies that deal with screening measures in general can be found in Chapter 8.1.6.2. The reason for this separation is that Germany is the only country with a population-based screening programme and comparability with other, smaller, and more timely screening interventions is limited. In addition, despite some differences in detail, the nationwide screening and the SCREEN project share conceptual similarities, so that the latter provides indirect evidence for the current nationwide SCS of primary interest.

In the one-year pilot project, 360,288 legally insured women and men (about 19% of the eligibile population) were screened by visual whole-body examination. The project was preceded by several years of preparatory activities, including, in particular, publicity measures and training for physicians willing to participate. In contrast to the nation-wide screening, gynaecologists, urologists, and surgeons were also entitled to participate besides dermatologists and general practitioners and internists. Another difference to the current screening is that the intervention was aimed at insured persons already from the age of 20 years. Furthermore, primary screening non-dermatologists were obliged to refer to a dermatological specialist for re-examination not only in case of suspicious cancer findings but also in case of the presence of one of the defined risk factors.

The aim of the screening examination for skin cancer is the early detection of MM, BCC, and SCC. The relevant wording in § 28 sentence 1 KFE-RL deviates in this respect from the normative objective of breast cancer early detection, which exclusively mentions the reduction of breast cancer mortality (§ 9 para.1 sentence 1 KFE-RL). For the evaluation, this means that a one-sided focus on mortality does not do justice to the intention of the norm setter. On the other hand, early detection is not an end in itself. On the contrary: by bringing forward the diagnosis, healthy people are turned into sick people, which initially represents a harm. The intention is rather that the early diagnosis opens up more effective treatment options, which should not only reduce mortality but also reduce the overall burden associated with skin cancer (cf. § 2 sentence 2 oKFE-RL). This may also include the avoidance of invasive, potentially disfiguring therapies, as well as the detection of precancerous lesions (e.g. actinic keratoses), which may lead to the prevention of cancer.

Follow-Up Evaluations of the SCREEN Projectt

In the temporal context of the SCREEN project and the preparatory activities, epidemiological trend changes were observed in Schleswig-Holstein, which are to be expected in the case of effective population-based early detection.

Initially, there was a pronounced increase in skin cancer incidence. Compared to 1999, the age-standardized incidence of invasive melanoma (ICD-10: C43) was about 20% higher in 2003/2004 [198], only to fall back below the pre-screening level thereafter [687]; [207]. In the Saarland as a comparative region without systematic skin cancer early detection there were hardly any changes in the C43 incidence in the same period [206]; [207]. Particularly strong increases in incidence were recorded in the SCREEN project for in situ melanoma (+46% (women and men overall, comparison July 2003-June 2004 with July 2001-June 2003) [206] and +133% (women) and +108% (men) (comparison July 2003-June 2004 with 1998-2000) [207]). For invasive non-melanocytic skin cancer (C44), the rate of new cases in July 2003-June 2004 was 47% (women) and 34% (men) higher than in 1998-2000 [686]. This was followed by a delayed slight decline, but not back to pre-screening levels. This could be an expression of a long-term rising screening-independent trend of increase in non-melanocytic skin cancer, as described for Saarland for example [686].

An increase in the incidence of the target disease in the initial phase of a screening programme is a typical consequence of the measure if it cannot be explained by secular trends. It indicates that screening has detected diseases that would otherwise not have been detected or would have been detected at a later time. It also includes diseases that would have been detected in an earlier screening round under the hypothetical assumption of a longer existing programme (“prevalent" cases). The initial prevalence peak, as well as a persistent incidence surplus accompanying the programme, is not as such a suitable indicator of the intended success of the programme. A remaining incidence surplus indicates diseases that would never have appeared during the lifetime of the affected person without early detection. Such overdiagnoses represent a damage that is in principle unavoidable, although in the case of the one-year SCREEN project it is difficult to quantify.

The crucial indicators for assessing a screening program are incidence and mortality; in the absence of other influencing factors, the age-standardized rate of advanced tumours and the mortality rate should decline.

Indeed, a significant decrease in melanoma incidence rates was observed in the follow-up of the SCREEN project at stage T2 and T3, and also T4 in women (Eisemann et al. 2014b). When these stages are combined, there is a 43% (women) and 40% (men) decrease in the standardized rate when comparing 2006/07 to 1999/2000. Missing stage data have been replaced by multiple imputation in this analysis. A particular contributor to the population-based decline, as suggested by another study, was the reduced burden of disease in participants of the SCREEN project [690]. In this study, the incidence of interval carcinomas was investigated. This refers to carcinomas that occur after unremarkable participation in screening outside the programme within the regular screening interval (here assumed to be 24 months). Invasive interval melanomas were recorded in 150 of 350,307 inconspicuously screened individuals, which is less overall than would have been expected based on incidence rates in the pre-SCREEN era (1999-2002) (age-adjusted relative incidence: 0.71 (95% CI: 0.60-0.84). Among these, in turn, the proportion of T2-T4 melanomas was significantly smaller (16.7%) than in the reference period (28.8%). In addition, it should be taken into account that, on the one hand, the proportion of invasive melanomas with unknown stage (Tx) was higher before the start of the project (42.4% vs. 36.0%), and that, on the other hand, 21.5% of SCREEN participants were recommended for a follow-up examination, which increases the number of interval melanomas compared to a screening-free situation. Both circumstances suggest that the study on interval melanomas underestimates the impact of screening on the incidence of advanced melanomas in a screening cohort. However, it cannot be ruled out in both studies that a change in T classification from 2004 onwards artificially reduced the number of unfavourable stages [687].

In temporal connection with the SCREEN project, there was a significant decrease in melanoma mortality in Schleswig-Holstein [691]; [206]; [198]; [692]. Compared to 1998/99, the age-standardised mortality rate in 2008/09 was 51% lower for women and 47% lower for men, while in neighbouring regions (Denmark, Mecklenburg-Western Pomerania, Hamburg, Lower Saxony, and Germany without Schleswig-Holstein) the rates were stable or increased [199]. The plausibility of the finding has been questioned with different considerations. On the one hand, the significant mortality decline with a SCREEN participation rate of only 19% is unexpectedly high. On the other hand, the decline seems to have started before 2003, although a reliable determination of the time of the trend change is problematic due to statistical noise in the data. Preparatory activities may have contributed to the unexpectedly favourable development, which may have increased awareness of skin cancer early detection among physicians and the population and promoted corresponding examinations already before the actual start of the project. These are not taken into account in the participation rate of 19%. The magnitude of the decrease may also be influenced by the fact that persons with an above-average risk of dying from skin cancer were overrepresented among the project participants [198]; [692]. It is unclear whether and to what extent incorrect coding of cause of death contributed to the observed transient decline in melanoma mortality. During the same period in which the largest decrease in melanoma mortality was recorded (2008-2010), a significant increase in mortality due to malignant neoplasms of vaguely defined, secondary, and unspecified sites (C76-C80) was recorded [692].

Evaluation Results for Nationwide Skin Cancer Screening

Even after the launch of the nationwide SCS in July 2008, there was a significant increase in skin cancer incidence with a participation rate of eligible persons of approximately 30 percent. For invasive melanoma, an increase in age-standardized incidence of 24% (2006-2010) [691] and 28% (2003/04 - 2010/11) [198] was reported. An analysis of cancer registry data restricted to North Rhine-Westphalia showed an annual percentage change (APC) of 3.8% (95% CI: 0.4%-7.3% (females) or 0.1%-7.7% (men)) and a similar increase for non-melanocytic skin cancer (APC women: 5.2% (95% CI: 2.1%-8.3%,); APC men: 3.6% (95% CI: 0.6%-6.7%) [688]. The increase was more robust for in situ tumours of the skin (in situ melanoma: APC women: 11.2% (95% CI: 5.2%-17.6%,); APC men: 12.0% (95% CI: 5.8%-18.5%); in situ SCC: APC women: 14.3% (95% CI: 8.0%-20.9%,); APC men: 15.1% (95% CI: 8.4%-22.3%) (Stang et al. 2018). In parallel, North Rhine-Westphalia showed an increase in melanoma-associated work disability cases as well as work disability cases and hospitalizations for non-melanocytic skin cancer. A positive association between the nationwide SCS and the frequency of relevant hospital discharge diagnoses (as a surrogate for incidence; C43 and C44 combined) was also found in a study with panel data from 22 European countries over the period 2000-2013 [685]. Only one of the studies identified and appropriately designed found no incidence increase for skin cancer. This was an analysis of health insurance data from Saxony comparing 6-month incidences from the first and second half of 2008 [689]. The negative result is not very meaningful due to the very short observation period, the low participation rates at the beginning of the program, and the relatively small study population.

The same study examined the respective disease severity of cases included in the health insurance data over the period 2008 to 2012. Severe courses (defined as cases with LK and/or distant metastases and/or interferon treatment) were more common in nonscreening participants than in screening participants, although the differences were not significant in each case. Population-based changes in stage-specific incidence of melanoma after the introduction of nationwide screening were examined in the aforementioned study from North Rhine-Westphalia [688].

While the incidence of T1 melanoma increased by 14.6% (95% CI 8.4%-21.2%) annually in women and by 14.3% (95% CI 7.0%-22.2%) in men from 2007 to 2014, the annual increase in advanced melanoma (T2-T4) ranged from 2.9% (95% CI 0.5%-5.4%) (T3 women) to 6.6% (95% CI 2.3%-11.1%) (T4 men). However, it is reasonable to expect that the reported small increases in advanced tumours may actually underlie weaker increases or even declines in new disease rates, as the substantial number of unknown stages (49% of invasive melanomas in women and 50% in men at the start of the observation period) fell to 20% each in parallel. Of note, in the non-modeled annual rates, despite further increasing completeness of staging information at the end of the observation period (2014 vs. 2013), a decline is evident in all advanced stages in both men and women.

Robust evidence for the incidence of overdiagnosis and interval carcinomas is not yet available for the nationwide SCS. None of the identified studies found a decrease in melanoma mortality as a result of the program. The observation periods extend to 2012 [685], 2013 [691]; [198]; [692] and 2015 [688], respectively. The discrepancy with the described mortality trends as a result of the SCREEN project has led to controversial interpretations, which, in addition to the aforementioned possibility of miscoded causes of death and changes in the population data base (Census 2011), are also based on differences in detail in design, implementation quality, acceptance, and awareness effects of the two interventions (participation of specialists from different disciplines, training measures, motivation of the examiners, age of those eligible, participation rates, self-selection of at-risk individuals, referral of at-risk individuals to specialists, and public relations) [692]). As a result, the SCREEN project [198], in part, and the nationwide screening [691] are rated as questionably more intensive. If one focuses on the initial increase in incidence, neither the one nor the other interpretation can be proven with certainty. The latter interpretation would make the absence of an observed decrease in mortality all the more disappointing. However, it is possible that the time series evaluated so far are too short to detect effects of nationwide screening on mortality [692]. It is generally assumed that mortality-reducing effects of cancer screening do not become visible before five years have elapsed [691]. In the case of the SCREEN project, the second mortality reduction, which was more significant in terms of magnitude, also began with approximately this latency [198]. For the nationwide SCS, this would mean that changes in the mortality trend would only become visible from around 2013.

Need for Research

The currently available evidence on the effectiveness of the nationwide SCS is insufficient. Promising results from the SCREEN project have not yet been replicated for the current programme. Harms that counterbalance the presumed benefits have not been sufficiently investigated. The evaluation reports commissioned in accordance with § 35 KFE-RL [694] focus on indicators of structural and process quality and also show weaknesses in this respect, which are largely due to inadequacies in the data basis.

As essential inconsistencies of the last report [694] are to be emphasized [252]:

• For the physician participation rates among dermatologists values of >100% have been mentioned for some KV-districts.
• Due to the lack of an insured person pseudonym, it was not possible to clearly link GP and dermatological screening documentation on an individual level. Therefore, the reported participation rates double-count all insured persons who were screened in two stages, which leads to an overestimation of the utilization.
• Biopsies were documented in only 71.5% of all cases of suspected cancer documented by a specialist. On the other hand, nearly 60,000 cases were documented in which biopsies were ordered without suspicion of cancer. The authors assume artifacts, which in individual cases were also provoked by incomplete answer options.

Although the documentation requirements that have been in effect since 01/01/2019 suggest certain improvements with regard to evaluability, the most pressing question of whether the SCS provides a net benefit to participants in terms of mortality, morbidity, and quality of life will still not be answered validly by the documentation data.

Randomized controlled trials, which would be most suitable for this purpose, are not meaningful because of the contamination to be expected in the control arm (in Germany) and would require – at least for the outcomes mortality and occurrence of advanced tumours – numbers of participants and observation times that are not realistically achievable.

The most obvious option would be to conduct a cohort study, which, because of the necessary number of person-years to be observed, would most likely be realized as a retrospective cohort study with health insurance data. The main difficulty here is the a priori definition and collection of possible confounders, i.e. factors that are associated with both screening participation and outcome.

Therefore, a case-control study with the possibility of primary data collection is advocated here. In such a study, cases, e.g. individuals with metastatic skin cancer, would be compared in terms of exposure screening participation with control subjects from the general population. Death due to skin cancer could also justify case status, with relevant confounders and exposure collected during lifetime.

High priority continues to be given to conducting ecological analyses, i.e. studies that examine effects of SCS on a group basis and detached from individual uptake of the screening offer. Current figures of the cause of death statistics show that the mortality decline, which was not yet discernible in the cited studies, started in 2013. Since then, age-standardized skin cancer mortality has been declining significantly and steadily. A latency of five to seven years corresponds to what is generally to be expected in the case of cancer screening [691]. However, it can be assumed that improved treatment options for melanoma (immune checkpoint inhibitors and BRAF inhibitors, approved from 2011) may at least have contributed to this development [208].

Therefore, it remains important to monitor the incidence of advanced, particularly metastatic tumours. Declining incidences in advanced stages that cannot be explained by a decline in “natural" incidence reliably indicate successful early detection. Methodological difficulties with respect to lack of completeness in stage data can be addressed by appropriate imputation procedures. When focusing on mortality, possibilities to control the different influences of early detection and therapy should be considered. This can be done, for example, by analysing regional trends in relation to screening participation rates or by making comparisons with other, non-screening countries. Elaborated epidemiological methods (e.g. age period cohort models) can contribute to a deeper understanding of the data. In all group-based analyses, the possibility of ecological fallacy (cross-level bias) should be considered.

Last but not least, the nationwide SCS should be subjected to an in-depth health economic analysis, including harms (including overdiagnosis) and possible modifications, e.g. in the direction of risk-adapted screening.

E. Grossmann, I.-M. Hübner

• A high general relevance assessment of free statutory skin cancer screening (gSCS) can be seen among adults with statutory health insurance. However, at most half of the eligible persons know about the possibility to make use of it.

The cross-sectional study by Augustin et al. [718], which is representative for Germany, investigates the public perception of the gSCS. The gSCS was introduced in July 2008 as a standard benefit of the statutory health insurance in Germany from the age of 35 and serves the early detection of skin cancer. The study is a standardized telephone survey of 1,014 legally insured adults with German citizenship. Overall, 44% of respondents were aware of the possibility of free gSCS, with a higher level of awareness in eastern Germany, among women and at an older age. Fifty-six percent of respondents consider the possibility of free SCS as personally very important, 32% find it somewhat important. Hope for early detection and good treatment options, physician recommendation, interest in information about skin cancer, and own risk are named as the most important drivers of utilization. For more than half of the participants there is no reason not to go to the gSCS.

Eissing et al. [719]

conducted a representative cross-sectional survey on the public's knowledge and perception of gSCS. For this purpose, 1,004 adult persons with German citizenship were interviewed in standardized telephone interviews. The screening was rated positively overall. Persons who participated rate the advantages of the gSCS higher than the disadvantages. In general, 50% of respondents know about their eligibility for the gSCS, with a higher proportion of women than men. This represents a stagnation compared to previous years.

An American population representative cross-sectional study (Rutten et al., 2009) relates the data collected regarding perceptions of screening to the evidence, and the evidence base here is considered unclear. Nevertheless, 88% of respondents believe that regular checks increase the chance of diagnosing skin cancer at an early stage. This assessment may be due to the fact that despite the perceived lack of clarity of evidence, many organisations recommend SCS.

Sullivan et al. [720], referring to the same population-representative cross-sectional study as Rutten et al. [424], compared knowledge of screening opportunities for different cancers. For skin cancer, the possibility of screening is less known than for colorectal and lung cancer. However, for all three cancers, there is an assumption that screening leads to successful early detection (87-90% of respondents), which is why the value of screening is appreciated.

• There is a clear gender difference in willingness to participate in SCS. Men are less likely to take up screening and also show a lower willingness to participate.

Davis et al. [721] investigated attitudes towards skin cancer screening and its uptake among US adults. For this purpose, 1,148 persons were interviewed in standardized telephone interviews. In general, the SCS is perceived as effective by both men and women, but strong gender differences are evident in willingness to participate and participation overall. Men show a lower willingness to participate in the SCS than women. Willingness is higher when specific information on screening is provided. The gender difference is also evident in the uptake of screening: 41% of men have never used screening, compared to only 5% of women.

• Knowledge of gSCS as well as concern about skin cancer lead to an overall increase in uptake.

Eissing et al. [719] conducted a cross-sectional survey representative for Germany on the public's knowledge and perception of gSCS. It was found that half of the respondents knew about the gSCS. Thirty-nine percent had participated in screening at least once. Uptake is higher in West Germany and among women than among men and in East Germany. In general, uptake is positively associated with knowledge of screening and concern about skin cancer.

A working group of the National Cancer Plan (NCP) dealt with the uptake of cancer screening measures (Goal 1 from Field of Action 1). In the context of this development, a change of direction was made, which places the informed decision for or against a cancer screening measure above the highest possible participation rate [724]. In the following, it was necessary to address the concept of “Informed Decision.” In this context, the members of this working group agreed on the following definition, following Rimer et al. [725] [726] :

An “Informed Decision" occurs when an individual:

• understands the condition being addressed and grasps what the medical service entails, including benefits, risks, limitations, alternatives, and uncertainties,
• has considered his or her preferences, and
• makes the decision in accordance with these,
• believes he or she has participated in the decision to the extent desired, and
• has made the decision voluntarily and with the highest degree of personal autonomy.

Participatory decision-making is present when a decision is made jointly by the health care provider/health professional and the patient and implies the health professional's constructive support of the process at all levels. Overall, a three-stage process is assumed: 

1) Conversation about need for decision and communality (Choice/Team Talk):

• Communicate that decision is coming up
• Formulate equal rights of the partners

2) Conversation about possibilities (Option Talk):

• Discuss advantages and disadvantages of the option
• Possibly use of decision aid (Decision Aid)

3) Decision Talk (Decision Talk): 

• Determine preferences of the user
• Negotiate decision
• Reach a joint decision
• Reach agreements on the implementation of the decision

In the context of screening measures, the informed decision for or against participation in screening is particularly relevant. If the screening result is found to be abnormal and further diagnostic steps become necessary, the process of participatory decision-making also kicks in.

Together with representatives of the German Cancer Society (Deutsche Krebsgesellschaft e. V.), German Cancer Aid (Deutsche Krebshilfe e. V.) and the Association of German Tumour Centres (Arbeitsgemeinschaft Deutscher Tumourzentren e. V.), the Federal Government launched the National Cancer Plan in 2008 to coordinate the activities of all those involved in the fight against cancer more effectively and to optimise the care situation for cancer patients in Germany. The goals of the NCP are primarily the further development of early cancer detection and care structures as well as more quality assurance in oncology, ensuring efficient drugs for treatment, and patient orientation. This also includes improving the communication skills of physicians as well as the information, counselling and support services [726].

In this context, Objective 1 (improvement of information and participation in early cancer detection) from Field of Action 1 (further development of early cancer detection) deals with the development of criteria that are necessary with regard to the formal and content-related design of information offers in order to enable citizens to make an “informed decision.” Since cancer screening examinations are aimed at people without symptoms and involve risks as well as benefits, an “informed and participatory decision" for or against participation is particularly important. Therefore, objective, understandable, and comprehensive information about potential advantages and disadvantages must be available to citizens. In order to do justice to this fact, the following checklist for recommended contents of information on screening measures was formulated by the members of Objectives Paper 1. It represents a consensual basis for the production of health information, which will be further tested within research projects [724].

Checklist: recommended contents of information on screening measures (modified) [724]:

1. Introduction,
2. Target groups,
3. Aims of the information,
4. Explanation of the disease for which the measure is used:
1. description of the disease and its course (without early detection measures),
2. health significance/impairment,
3. epidemiology (incidence of disease, mortality; it may be helpful to present these risks in comparison with other diseases; presentation of risks in natural numbers and, if possible, graphically),
4. treatment options,
5. prevention,
5. Description of the screening measure:
1. objective of the measure (reduction of incidence/morbidity/mortality),
2. explanation of the method/description of the procedure of the examination,
3. description of further clarification examinations after findings,
4. accuracy of the method (frequency of false-positive and false-negative findings; positive predictive value of a finding),
5. description of the benefit and quantification (comparative with and without screening),
6. level of evidence (or the degree of certainty of scientific proof that the intervention actually achieves its goals),
7. description of risks and harms,
8. direct risks associated with screening (e.g. radiation, complications),
9. indirect risks resulting from a finding,
1.  … due to false-positive findings,
2.  … by false-negative findings,
3.  … due to advance diagnosis,
4.  … due to overdiagnosis/overtreatment,
6. Access to screening,
7. Information on costs incurred or cost coverage,
8. Information on the quality of the screening measure,
9. Description of quality assurance measures (e.g. certification of service providers, training programme, double assessment) and verifiability (quality indicators that the participant can check, e.g. advice on possible findings, need to undress in case of SCS),
10. Further information:
1. Reference to supplementary information that is missing due to lack of space,
2. Reference to the fact that other people who knew this information came to different decisions,
3. Reference to the fact that there are no pressures on content or deadlines,
4. Reference to patient guidelines or specific further information,
5. Decision aids (where validated aids are available for individual decision making),
6. Reference to data protection or data use or consent to data sharing,
7. Self-examination,
8. Reference to the absence of symptoms (i.e. symptoms should and will be clarified irrespective of eligibility for the screening examination),
9. Personal responsibility (each person is responsible for taking care of themselves and making decisions for or against preventive measures. The knowledge of one's own risk about effectiveness, benefits, risks, and limits of methods and consequences – even in case of non-use – are the basis for taking personal responsibility),
10. Risk groups,
11. Imprint/source information/status of information,
12. Financing of the information medium, source of information, etc.,
13. Information on conflicts of interest,
14. Expiry date of the information.

Need for Research

The available studies on the determinants of secondary skin cancer prevention provide little information on the effect of communication measures on informed and participatory decision-making. The focus of the analyses is on participation in screening and knowledge about the risks of the sun or skin cancer and early detection of melanoma, but not on knowledge about the risks and opportunities of using the SCS.

Research is therefore still needed regarding the identification and description of predictors, moderators and mediators acting on the “Informed decision" regarding the secondary prevention of skin cancer, even though studies already provide evidence for this. In this context, the areas of information, context and medium need to be considered. Furthermore, as described, the checklist must be (empirically) tested and further developed, e.g. by means of criteria catalogues or other measuring instruments.

For more general information on the presentation of risks and opportunities of screening, we refer to the guideline evidence-based health information [727]. Although the guideline on evidence-based health information contains only one study on the presentation of risks in information on mammography screening [728], a Cochrane Review compiles the effects of different formats of risk presentations (such as natural frequencies, percentages, absolute risk reduction, relative risk reduction, number needed to treat/screen/harm) on different health topics [729]; [728]. However, these findings may be transferable to the SCS or indicate a need for further research on this topic.

This also includes the identification of relevant multipliers (e.g. family members, life partners, doctors, pharmacists, teachers, employers, peers), who can be essential for reaching the actual target group(s).

In the first step, it is necessary to identify the different target groups (segmentation) (e.g. within the framework of the formative evaluation, see Chapter 5.4.6) in order to subsequently describe them and, in a second step, to select, develop and compile suitable strategies and measures (targeting). The function of segmentation is to delineate more homogeneous subgroups from the heterogeneous overall group in order to be able to address and serve them more effectively and in a more targeted manner [731]; [732]. This is possible in particular through digital forms of communication in which the content can be specifically tailored to the user groups.

In this context, it is beneficial if the segmentation is oriented towards health psychology and behavioural science constructs, such as lifestyle, health-related attitudes and motives, risk perception and behaviour, and self-efficacy experience. These factors are in turn linked to information and communication needs, preferences and barriers, to the type and intensity of health-related information search and media use, and to topic perception and processing. With regard to health communication, this means that information and communication objectives, communication channels and message strategies should be developed on this basis in a target group-specific way [733]; [504]; [730]. 

In addition, it is necessary to record the health literacy and media literacy, the risk status of the persons (e.g. skin type or typical sun exposure duration) as well as life-related settings (e.g. outdoor workplace) of the individual target groups in order to be able to locate and delimit the individual segments and to adapt the contents of the communication measures. Since such comprehensive information is only available through appropriate risk screenings or when using corresponding digital applications, segmentation is in many cases carried out on the basis of more readily available determinants of the aforementioned characteristics.

Here, sociodemographic, socioeconomic, sociological, and psychographic criteria as well as health status, health awareness, and risk profile play an important role. Frequently, the combination of several criteria (hybrid segmentation) is also possible and necessary. In addition, the segmentation should be process-oriented, since a decision is often part of a behavioural change process consisting of several stages (cf. transtheoretical model), the target group differentiation should also take into account that the target persons are to be located at different stages of the health behaviour change and thus have different information needs. In addition, target group segmentation should constantly adapt to changes in target group characteristics (dynamic segmentation) [732].

Need for Research

However, there is still a need for research to examine how the effectiveness of target group segmented measures differs from those that address the entire population. This applies in particular to opportunities for digital communication measures and instructions for self-examination, e.g. within apps. In addition, empirical evidence is needed to show which criteria applied as part of a segmentation or personalisation strategy have a higher efficiency and effectiveness than others. In addition, strategies and measures for the dissemination of messages and information must be evaluated in terms of their strategy by comparing target groups. In doing so, it is necessary to examine measures in comparison to each other and their suitability for specific target groups.

The systematic review by Austoker et al. [469] deals with interventions that promote cancer awareness and early medical consultation in suspected cases. Five studies were included in the analysis that examined interventions targeting individuals and distributing (personalized) information by mail or web-based means. In addition, a further ten studies were analysed, some of which described public education campaigns, but also setting-based (i.e. related to the living environment, e.g. the workplace) interventions. Most studies focused on a specific cancer type. A total of four studies focused exclusively on malignant melanoma. The means of intervention ranged from information leaflets, telephone education, and computer-based learning programmes to mass media education campaigns, educational seminars, lectures, and information stands. As a result, it was found that cancer awareness, attention to possible cancer symptoms, active seeking of help for suspicious symptoms, or knowledge about melanoma risk reduction could be increased by the respective interventions. It can be inferred from the study that interventions tailored to the individual (tailoring) have the highest effectiveness. Embedding or tailoring an information intervention in or to the social setting/immediate lifeworld is one way of target group segmentation [469].

The question at hand is only indirectly examined in the review, so the level of evidence can only be related to the statement of the recommendation in a very limited way. Due to this, the wording “can" has been chosen in statement 8.32.

Need for Research

As the studies show, there are hardly any studies to date that evaluate setting-based interventions (e.g. at the workplace or in the doctor's practice) against interventions that work without a setting. It would be important to provide evidence for the benefits of setting-based interventions and to generate detailed research results in order to be able to adapt interventions to the respective setting. This is because it is precisely through such interventions that hard-to-reach target groups can be reached.

In the context of information on screening measures, there is evidence that family involvement, such as partner education, can improve self-confidence in self-examination [734] and approaching high-risk patients about family members affected by skin cancer (e.g. siblings) can increase willingness to and participation in SCS [735]; [736]. Robinson et al. [734] show in their study that partner education can improve self-confidence in self-examination in melanoma patients.

Educational videos with information about risk groups or internet training can be offered via GPs and pharmacists. However, there is insufficient evidence on the effect of such educational videos on specific skin cancer risk groups or internet education among GPs [704]; [706] and pharmacists [737].

The knowledge about melanoma and the performance of the self-examination of the skin can be increased by the multiple as well as multimedia approach of adults. These effects were significantly different from the results of the respective control group [450]; [460].

Multiple communication achieves better effects in promoting skin self-examination than handing out a standard brochure once [450]. Repeated multimedia health education with animations, photos, and brief information in clinical settings leads to better knowledge of melanoma and increase in control of moles in at-risk individuals [459].

Multimedia communication (e.g., videos) appears to be superior in effectiveness to purely text-based communication [461]. Evidence for the superiority of multimedia forms of delivery (video) over conventional routes via brochures can be found in Idriss et al. and Janda et al. [461]; [463]. Group-based interventions can lead to better risk awareness. Austoker et al. [469] describe an increase in physician consultations within three months of recognition of melanoma symptoms from 16% to 67% [469].

Finally, women and men seem to be equally motivated to visit screening facilities regardless of education level [457]. This is attributed to well-designed promotional materials [738]. Overall, it should be noted that the effects in some studies cannot be clearly related to the attributes multiple or multimedia, as both were applied together and therefore a differentiated view is not possible.

However, on the basis of the available studies, the positive effect of such prevention programmes can only be assumed for complex training programmes that integrate various textual, visual, and audiovisual elements. In the study by Glazebrook et al. [459], tthe programmes were not tested against the effect of other programme profiles (other delivery channels or other forms or combinations of information presentation and processing), so that on this basis – despite the partly high level of evidence of the studies – it is not possible to make any statements about which measures or which components of a training programme have an effect and which do not contribute to an improvement in knowledge, attitude, and behavioural parameters. In studies in which different forms and ways of presenting information are compared, e.g. [461]; [463], other biasing factors may have been at work. Moreover, these studies also only provide indications of the effect of a complex and multi-part bundle of measures, so that no statement can be made about the potential preventive influence of individual components.

Against this background, the evidence of the use of multimedia interactive training materials for optimizing the cognitive prerequisites necessary for an “Informed Decision" to participate in screening must be regarded as limited.

Need for Research

Accordingly, research is needed to comparatively test the short-, medium-, and long-term delivery performance of different training programs. Conclusions can only be drawn about the mediation potential of individual forms of presentation or programme profiles if the presentation and mediation parameters vary systematically, other parameters (e.g. target group, communication content) are kept constant and confounding variables are controlled or eliminated. Above all, it must be ensured that the information content of the different mediation channels used is comparable. In addition, comparative studies should examine the transferability of findings on the effect of different training programmes in countries and regions with an above-average risk potential from solar radiation (e.g. Australia [678]; [463]) and on specific target groups (e.g. older men [460]; [463]).

Furthermore, in the context of content and design planning and implementation of prevention and intervention programmes, there is a need for research in formative, in-process and summative evaluation (see also the following Chapter 5.4.6). For the strategic planning and conception of campaigns, it is particularly important to determine which communication channels can be used to reach which risk groups and how these should be designed in terms of media in order to also have an attention-grabbing effect in the natural settings of the target groups. In combination with formal design elements, various message strategies (“message frames") should also be tested for their effectiveness in raising awareness of SCS.

A systematic literature review [740] of 

25 English-language studies on self-examination and melanoma detection concluded that visual representations improve knowledge, attitudes, and self-efficacy related to skin self-examination, increase frequency and accuracy of examination, and improve melanoma detection in patients. Text descriptions alone prove ineffective. Effective images here include labels of abnormal body sites in the form of diagrams or body sketches, clinical example images of benign and malignant skin lesions, dermoscopy images, and photographs of the patient's own skin and moles as a basis for comparison. The authors concluded that patients should have access to images and a large number of case histories at all times to enhance their visual memory and pattern recognition of melanomas.

However, the potential impact of such information materials as a single intervention should not be overestimated. Even if learning effects on the knowledge level can be demonstrated with appropriately well-designed materials, this can only be transferred to a limited extent to the ability to distinguish benign and malignant lesions in reality [460]). Here, a media communication measure alone does not seem to have a sufficient effect, especially in risk groups with below-average health literacy, so that a combination of such measures with interpersonal counselling and support offers is probably required. King [739]  also points to the possibility of unintended effects through visual representations of case studies. He was able to show that visual representations increased perceived informedness about the detection of atypical melanoma, but that willingness to be reviewed by a physician varied.

Research Needed

Accordingly, there is a need for research to find out which depth of information and which type of information presentation is appropriate for which target group, i.e. can be easily understood and transferred into action-relevant knowledge, and to what extent a combination of media and interpersonal training measures is appropriate in each case. In this context, taking into account the target- and risk-group-specific resources and barriers to information processing, above all the process of directing attention and learning should be analysed in a differentiated manner.

Health information tailored to personal characteristics, behaviour patterns, needs, and beliefs is more likely to be perceived as personally relevant and therefore has a stronger motivating character than information that contains general information and advice. This so-called tailoring should take the form of, for example, personalized feedback on risk status, behavioural recommendations tailored to this, and reminders. This is particularly possible in digital communication formats (e.g. decision-making aids or risk calculators), which can interactively adapt to personal information.

For prevention and intervention programmes that address single individuals via personal contact with a health professional or also in the form of direct media, there is stronger evidence of their effect on the perception of a cancer risk than is the case for interventions that start at the collective level, i.e. do not specifically address single individuals

[469]. 

The systematic review of studies also provides evidence that individualised targeting or information tailored to individual risk status (“tailoring") is more effective than general information.

Evidence for the effectiveness of “tailoring" is also provided by Glanz et al. (2010). For adults at moderate to high risk of skin cancer, the authors were able to demonstrate a positive impact of personalised feedback in the context of a three-times mailed information package at two-week intervals compared to a non-personalised intervention in the form of general educational material on skin cancer prevention and self-examination as well as a brochure on sun protection measures and behavioural tips. Feedback was personalized based on individual risk status and personal risk factors, as well as on practiced sun protection and self-examination behaviours, behaviour change readiness, and perceived barriers to behaviour change. The constructs “risk perception,” “cost and benefit trade-offs of behaviour change,” “action-relevant knowledge and skills,” and “social norms" were included as mediating variables [450].

A systematic review of personalisation of risk information for informed decision-making about screening participation [741] includes three studies on skin cancer screening [735]; [459]; [736]. In two studies, personalized risk communication in the form of a personalized list of risk factors [736] and in the form of a numerical risk score [459] is better suited than generalized risk communication to increase screening participation in high-risk groups. However, in a study [735] this relationship was not found, only a benefit of the personalized list of risk factors on intention to participate in screening. The effectiveness of personalized risk information is also indicated by the successful use of a tool in which individuals were asked to self-assess their risk as part of a self-examination [495]. However, this tool was used in combination with medical advice, so that no conclusions can be drawn about its effectiveness independently of this.

Glazebrook et al. (2006), as part of their interactive PC training for at-risk individuals, worked with individualised feedback on risk status, which, as a fear appeal, aimed to increase perceived threat and, with regard to the practice of protective behaviour, simultaneously provided information to lower barriers and increase perceived benefits. It contributed to an increase in knowledge, particularly among individuals with higher risk status [459]. A

gain, however, the programme was not tested against non-personalised training, so the evidence on the evidence of individualised information and feedback elements in this regard remains limited.

Beyond the need for an individualised approach, the studies provide clear evidence that theoretical underpinning of programme designs is important and useful. The elements of the health belief model provide a coherent theoretical framework for personalization that underlies many interventions [735]; [459], from which the information to be conveyed individually about the SCS can be derived and which provides explanatory approaches for the target group-specific effectiveness of programs for the prevention of skin cancer as well as starting points for the design of the messages and levels of individualized feedback and evaluation [449]. Glanz et al. (2010) also included the above constructs as mediating variables derived from the health belief model and social cognitive theory to measure the effectiveness of personalized feedback [450]. In addition, the Transtheoretical Model and the Theory of Planned Behaviour are also used as a theoretical basis [742]; [735]; [736].

Need for Research

Accordingly, research is needed to empirically substantiate the short-, medium-, and long-term effectiveness of individualized training programs and feedback elements in intervention measures compared to programs that do not include elements of tailoring and feedback opportunities. In doing so, programs should draw on established theoretical approaches to modeling programs and explaining health behaviour change. Accordingly, the design of such programs should be based on a theoretical foundation and systematically tested against those interventions that do not include tailoring. In this context, the technical possibilities of digital and mobile communication measures in particular should be given greater consideration.

In order to develop and plan communicative interventions in the context of secondary skin cancer prevention in a targeted manner, data collection is necessary even before the actual implementation of the intervention (formative evaluation). This has two aims: collection of information for evidence- and theory-based conceptualization and implementation of the intervention (preproduction research), and a preliminary testing of the finalized intervention and its instruments and materials (product testing). Measurements and monitoring of the entire process are also beneficial in order to be able to take external and internal disturbance variables into account over time (process evaluation). Process evaluation includes not only the survey of content-related aspects but also the inclusion of variables that describe the quality of the intervention organisation (controlling) (e.g. organisational processes). Summative evaluation makes it possible to examine the defined intervention goals of a communicative intervention and to record the effects, effectiveness, and efficiency of the measure. The entire period during and after the intervention must be taken into account. Summative evaluation provides information necessary to identify and, if necessary, quantify possible changes brought about by the intervention. For this purpose, it is at least necessary to collect the relevant variables in each case before (can already be done within the formative evaluation (preproduction research)) and after the intervention. For an example, see the study by Anders et al. [742] on the SCREEN skin cancer screening project. Furthermore, it is important within the evaluation not only to examine variables that are directly related to the communication, but also to include the relevant health indicators and their change over time [504]; [505]; [506]; [507]; [743]; [509].

The evaluation parameters used within an evaluation should be derived from a theoretically proven model. According to the Transtheoretical Model, different stages of information processing are passed through before an intervention becomes behaviourally relevant. Continuum models, such as the Health Belief Model and the Theory of Planned Behaviour, also model the process of health behaviour change initiated by a prevention or intervention measure in a differentiated manner. The stage of behavioural change at which the target person or test person is in each case, or which constellations of individual predispositions are present in the members of a target group, also influences their receptiveness to different information and communication offers that are part of an intervention, as well as their evaluation and the resulting mediation potentials. Which end variables are measured and evaluated at the attitudinal and behavioural levels should therefore be derived from the theoretical model on the basis of which the intervention was designed [510]; [511]; [512]; [507].

Research Needs

Research is needed in testing evaluation strategies for reliability and in developing a set of criteria for testing the quality of evaluation measures. In addition, the explanatory power and predictive power of different theoretical models for different objectives and measures should be identified and the model parameters specified for the secondary prevention of skin cancer.

The evaluation should take place at several measurement points and measure short- and long-term effects. Validated and standardised scales should be used to measure the different endpoints. If these are not available, evaluation findings should be empirically validated by comparing the findings generated by different survey and analysis procedures.

Austoker et al. (2009) conclude in their systematic review of prevention measures to increase cancer awareness, which also includes studies on skin cancer prevention, that a higher methodological quality and comparability of study designs is required: “future research evaluating individual-level interventions to promote cancer awareness should attempt to use study designs that generate high-quality evidence, measure outcomes over a longer term (months/years) and attempt to measure behavioural and stage outcomes, as well as knowledge and attitudes. We also highlight the need for standardised and validated measures of cancer awareness […]” (p.38 in [469]).

This results in the consequences formulated in the recommendation for the parameters to be evaluated and the way they are measured. This also concerns the choice of survey instruments used for the measurement of attitude- and behaviour-related outcome variables.

In order not to underestimate possible effects of an intervention by the fact that the chosen evaluation method may not capture certain effects due to the nature of the survey, different methods should be used to measure the dependent variables that are complementary to each other and, when combined, allow for a more comprehensive picture [450].

Research Needs

There is a need for research into methods to optimise study designs with regard to the evaluation of prevention or intervention measures and the measurement procedures used in this context. The aim is to develop a catalogue of criteria for the evaluation of prevention and intervention measures in order to generate more empirically validated and comparable evaluation findings, e.g. by developing standardised and validated scales.

A prerequisite for the unfolding of an attitudinal and behavioural preventive effect of the prevention or intervention measure is how frequently and intensively the individual communication offers and messages are perceived, whether they generate attention, how they are evaluated at the level of content and design, and whether they are understood, retained, and subjectively perceived as useful. In order to measure the immediate communication success that precedes a longer-term effect at the attitudinal and behavioural level, evaluation parameters are therefore also required that directly address the perception of the campaign message or training measure and measure the dispositions of the test persons in a differentiated manner at the respective level of behavioural change. Effective interventions must therefore also have a positive influence on the outcome variables upstream of the behavioural change if the behavioural change is to be attributed to the intervention. Also, only recipient feedback on the actual information or training material provides concrete indications of how information and training offers as a whole, or how individual elements as well as the content and design of the information in multimedia interventions, are accepted by different target groups and what optimisation potentials result from this.

Glanz et al. (2010a) evaluated the assessment of personalized and non-personalized stimulus material by the subjects, although they did not statistically include it as a mediator variable in the impact model. However, it can be shown that all personalized information is rated significantly better than the non-personalized across all items. Also in Glazebrook et al. (2006) the positively rated way of preparation and presentation of the information as well as the perceived user-friendliness of a deployed interactive PC training in the risk group might have contributed to the learning success [459].

A relevant concept in this context is also the decision balance derived from the decision-making model of Janis and Mann, which expresses the weighing of positive and negative consequences of an action or behaviour of the target person. It plays an essential role in the context of (health-related) behavioural change according to the Transtheoretical Model. Empirical social research has developed two main ways to capture the concept of decision balance. One way is to contrast the advantages and disadvantages of an action, e.g. a behaviour, i.e. to subtract the disadvantages from the advantages. Another way to operationalize decision balance is to contrast or subtract the advantages of a particular behaviour from the advantages of the opposite behaviour or of not adopting the recommended behaviour [744]; [745].

Accordingly, decision balance should be included as a mediator or surrogate parameter of sun-protective behaviour when evaluating skin cancer prevention interventions.

Need for Research

There is a need for research to systematically evaluate the significance of the parameters directly related to communication (e.g. range and attention-grabbing of the means of communication, comprehensibility, and evaluation of the information offered or measure) for the effect of the prevention or intervention measure at the attitude and behavioural level. In this context, it is also important to empirically model the different variants of the decision balance and to examine them with regard to their mediatizing effect on sun-protective behaviour in order to draw conclusions about suitable forms of address in prevention campaigns.

Studies in which media messages are used and associated with attitudinal and behavioural outcomes would have to fulfil the necessary prerequisites for drawing conclusions about the effect on the campaign and should only be interpreted as evidence of changes at the attitudinal and behavioural level if it is empirically confirmed that this change results from the contact of the target groups with the campaign content (reach) and the processing of these messages. Previous studies have not yet provided sufficient evidence for this. For example, Del Mar et al. do not provide sufficient evidence that the increased number of excisions by doctors during two TV campaigns can be clearly attributed to them in terms of causality, so that the assumptions on the effect relationship remain of a speculative nature despite a statistical correlation between the campaign period and the number of excisions [513]. Also in Oivanen et al., visits for skin examination cannot be causally attributed to contact with campaign messages [514].

In the evaluation of such measures, detailed information on the disseminated messages and advertising materials as well as a measurement of the contact probability with the campaign and its reach up to the perception and evaluation of the same in the target population should therefore be measured or ensured before evidence of the effectiveness of a campaign can be assumed.

Based on the definition, three dimensions can be derived that an “Informed Decision" includes: understanding, preferences, and decision. Mullen et al. also describe a fourth: participation. The individual dimensions can be operationalized in different ways. Understanding, for example, can be mapped by recording knowledge and risk perception. Personal preferences can be assessed with parameters such as perceived benefits or barriers, values, and attitudes. Aspects of participation can be mapped with the help of concepts such as self-efficiency or shared decision-making. Finally, the decision dimension is reflected in the intention to participate or in participation itself [746]. An assignment of individual parameters to a specific dimension is not always possible in a clear, comprehensive and/or exclusive manner, i.e. in some cases a parameter also captures aspects of several dimensions or only fragments of a single dimension.

Marteau et al. (2001) developed a multidimensional model for the measurement of an “Informed Decision,” which comprises knowledge, attitude and behaviour (participation), i.e. proportions of the aforementioned dimensions of understanding, preferences, and decision are taken into account. All three parameters are considered dichotomized in the model: knowledge (high, low), attitude (positive, negative), and behaviour (participation, non-participation). Finally, the resulting combinations of the three variables are used to derive whether a decision is informed or uniformed. It is important for an “Informed Decision" that there is congruence between attitude and behaviour, with a high level of knowledge at the same time (Table "Informed Decision Algorithm").

On the contrary, incongruencies between attitude and behaviour are a sign of a uniformed decision (Table "Informed Decision Algorithm"). The model gets its theoretical basis from the Theory of Planned Behaviour [510].

Need for Research

There is a need for research concerning the development of a survey instrument that covers all four dimensions (understanding, preferences, participation, decision) of the “Informed Decision" and thus enables a more precise measurement of it. In addition, predictors, moderators and mediators are to be identified and described that act on the parameter “Informed Decision" as an overall concept. In this context, the areas of information, context and medium need to be considered. Similarly, not only decision-making processes in the context of secondary prevention (of skin cancer) but also primary prevention are to be evaluated with regard to informed decision-making. In addition, it must be examined whether persons who have made informed decisions actually achieve different short-term and long-term outcomes with regard to primary and secondary prevention behaviours and their consequences in comparison with uniformed decision-makers.

Doctor-patient communication is a psychosocial aspect of health care. It is important for the accurate communication and understanding of health-related information, the perception and adequate management of emotional distress, the establishment of a respectful and trusting relationship between doctor and patient, and shared decision-making and patient participation in treatment [747].

Patient-centeredness means taking into account patients' values, needs, attitudes, perspectives, preferences, and experiences [747] and patients' abilities and prior knowledge in doctor-patient conversations and the chosen form of interaction.

In general, patient-centered communication is associated with a variety of positive short- and long-term effects for patients [748]; [749]. Using a Randomized Controlled Trial (N = 80 women, aged 18-30 years), Bientzle, Fissler, Cress, and Kimmerle [748] make a comparison of the effects of a physician-centered versus a patient-centered communication style of physicians. The results show that physicians with a patient-centered communication style are perceived as more empathic, socially, and professionally competent. Attitude changes are more likely and influence on decisions is stronger. Only for knowledge transfer do no significant differences between the two communication styles emerge. Findings by Fox, Heritage, Stockdale, Asch, Duan, and Reise [750] also indicate that a patient-centered relationship in combination with a recommendation by the physician to participate in screenings increases willingness to participate. In addition, based on a systematic review of 40 studies, Riedl and Schuessler [751] show that a high degree of patient-centeredness is associated with perceived better emotional health and the use of fewer diagnostic tests.

Relevant components of patient-centered communication that were reviewed include patient activation and active engagement [752]; [753]. Specifically, communication readiness should be encouraged. It describes to what extent the patient feels able and willing to actively engage in the conversation with the physician and to voice his or her concerns [754]. 

In addition to the patient's experience and skills, willingness is influenced by the physician's communication style. Physicians should communicate to patients that they can express their needs, attitudes, and opinions. This is both a positive predictor of satisfaction and adherence to treatment recommendations.

Similarly, this includes the physician using clear language that is understandable to laypersons and either explaining or avoiding technical terms. According to a survey of cardiology patients (N = 119) by Thomas, Hariharan, Rana, Swain, & Andrew [755], language that is difficult to understand and uses many technical terms leads to lower patient satisfaction and poorer recall of the physician's information and recommendations. In addition, it is significant to offer the patient the opportunity to ask comprehension questions at any time and to receive answers such as explanations of the relevant information [754].

A high degree of patient-centeredness is achieved when communication is culturally sensitive. This describes that communication should be adapted for different audiences and address individual needs and characteristics of the situation such as person (Betsch et al., 2015; Epstein & Street, 2007).

For the the degree of patient-centeredness, nonverbal communication (including eye contact, gestures, facial expressions, listening) is also taken into account. Especially listening represents a relevant behaviour of the physician, which according to a meta-analysis can also increase patient satisfaction (Henry, Fuhrel-Frobis, Rogers, & Eggly, 2011). Another influencing factor of satisfaction is also an affect-oriented communication of the physician who shows empathy (Schrooten, & de Jong, 2017; Verheul, Sanders, & Bensing, 2010).

An effective communication between doctor and patient is supported by a discussion agenda and control. According to Jünger [756] , two phases must be distinguished. At the beginning of a conversation, it is important to clarify the patient's concerns. To this end, physicians should give their patients time to present their concerns, encourage their willingness to communicate and contribute to activating the patient (e.g. in the form of open questions), not interrupt the patient's explanations, instead allow pauses and listen to the patient [756]. Jünger also describes this as the patient-centered phase.

The second phase (physician agenda) should deal with what information, what examinations and therapies are necessary. This is about precise information for decision making. In the course of this, it may seem necessary to ask specific follow-up questions in order to understand the problem and gather information about it (especially in the form of closed questions). In addition, the information is evaluated and information about examinations, therapies is given. Thus, the second phase can also be described as a doctor-centered phase [756].

Finally, it should be noted that basic communicative knowledge and techniques for doctor-centred conversation, as described for example by Schweickhardt and Fritzsche or Jünger (2018), are helpful in the context of the doctor-patient conversation to create successful communication [757]. Jünger describes, for example, the “WRMS" technique (wait, repeat, mirror, summarize) as a central technique of patient-centered communication.

The doctor-patient interview is important to provide the potential participant with information about primary and secondary measures of skin cancer prevention. This can reduce knowledge deficits and uncertainties on the part of the potential participant regarding behaviour and measures (e.g. full body examination). In addition, the potential participant should be given the opportunity to weigh up the advantages and disadvantages in connection with his or her own preferences, attitudes, and abilities and to make an “informed decision" for or against a measure/behaviour[758]. The information content that is necessary in this regard has been developed by the members of Goals Paper 1 as part of the National Cancer Plan. These are summarised in a checklist. It serves as a basis for informing potential participants in screening examinations. It is planned to supplement this checklist in a further step with a criteria catalogue. This criteria catalogue will be used to review and evaluate information concepts (e.g. brochures, leaflets, verbal communication processes) [758].

Within the doctor-patient discussion, successful risk communication is also important, which shows the potential participant his individual risk and, if necessary, his risk behaviour and allows him to assess it. The recording of individual risk factors is to be determined by self and family history as well as the clinically recordable picture. In this context, it should be noted that a “positive" family history is sometimes not or insufficiently communicated within one's own family and to health professionals (health care professionals). This may be due to the fact that health professionals, for their part, do not communicate the significance of this risk factor in a clear and comprehensible manner. These findings are presented in a qualitative study using the example of malignant melanoma, in which 22 people from seven families were interviewed. In each family there were persons suffering from melanoma (n=11). They were interviewed about (risk) communication within the family and towards health professionals [759]; [760].

Only one study deals with the recipients' memory of possible risks after an explanatory talk using the example of a dermatological surgical method (Mohs micrographic surgery). The study shows that the general recollection of risks 20 minutes and one week after the explanation does not differ significantly (arithmetic mean of recalled risks: 2.65 (20 minutes) vs. 2.44 (one week – difference: 0.21). In conclusion, it can be assumed that within a period of one week the knowledge base relevant for a decision remains relatively stable [761]. However, these results can only be transferred to screening to a limited extent, since in the study the subjects are already patients, i.e. there is already a disease. This increases the need for decision-making and prioritizes the importance of information differently compared to healthy subjects. 
Need for Research

• Clarification of the importance of the time factor in the information and decision-making process with regard to weighing the available facts and memory.
• Studies that assess knowledge and other decision factors multiple times over a longer period of time after being informed.
• Identification and description of predictors, moderators, and mediators acting on the “Informed Decision" for or against participation in SCS. In this context, the areas of communication, information, context and medium should be considered. Questions to be addressed include what competencies and content are necessary to enable health professionals to enable the potential participant to make an “Informed Decision" and how to communicate the content described:
• Studies to identify the physician communication strategies and –pathways (including comparison of different communication styles and modes of delivery) that best succeed in enabling different target and patient groups to make an informed decision for or against participation in SCS. 
• Intervention studies that examine the differential impact of an information intervention developed according to the criteria of an “Informed Decision" on different target/patient groups.
• Intervention studies examining the impact of different information channels and mediators in the delivery of an information intervention developed according to the criteria of an “Informed Decision.”
• Instrument development study that captures possible dimensions of an “Informed Decision" and describes their validity in terms of these, in order to review or develop existing or new instruments to quantify an “Informed Decision.”
• Intervention studies, which examine further training courses, which are to convey the criteria of the “Informed Decision" regarding the informing of potential participants, to health professionals in a controlled manner.
• (Empirical) verification of the checklist, e.g. by developing criteria catalogues within validation studies.
• Risk communication within (skin cancer) screening measures. 
• Intervention studies in which different types of risk communication and their effects on the decision-making process or on informed decision-making are examined in a controlled manner.

The breaking of bad news, such as a diagnosis of skin cancer, causes anxiety in many patients. For this reason, sharing the diagnosis should be done in a calm environment, in an understandable manner and within a reasonable time frame. The conversation should be tailored to the patient and his or her intellectual abilities and preferences. Consideration is given to the realization that usually only a small amount of information can be absorbed at one time. Only as little information is given as is compatible with the patient's information needs. Sensitive care is taken to provide education only to the extent that patients signal that they can absorb and process it [763]. In addition, emotional support for the patient is useful. In advance, the patient should be offered the opportunity to include a trusted person in the discussion; this is particularly desired by married persons. The presence of other health professionals is largely perceived as uncomfortable [764].

Depending on the preference or wish of the patient and/or the relative, it may also be useful to address questions about further diagnostic and therapeutic steps, the effects of the disease and treatment on everyday life, and possibly also questions about the prognosis of the further course of the disease. For the discussion about the prognosis, it should always be taken into account that in the general population cancer is often associated with death and dying and that it is important to address this association and the fears associated with it. A sound source of information should be chosen for prognosis. In addition, patients prefer a brief information sheet with answers to the most important and frequently asked questions, as well as references to other support services.

The above recommendations for the educational interview are based on a standard publication on the delivery of serious diagnoses [765].

The general goals of the educational interview are to convey information to the patient in a way that is understandable, to provide psychosocial support for the patient, and to design strategies for action together with the patient. The conversation can be divided into six steps according to the SPIKES model of Baile et al. (2000):

At the beginning, an appropriate atmosphere is to be created and important persons are to be involved (e.g. partners). This is followed by an elicitation of the patient's attitude and clarification of the level of information about the previous diagnosis. After obtaining permission to communicate findings, this is done in adequate language (no technical terms) and by not communicating too much information at once. The understanding of the communication of findings and information is checked regularly during the interview. After communicating the findings, it is useful to address the patient's feelings, identify reactions, and acknowledge them appreciatively. At the end, further planning is discussed [765]. Patients are encouraged to ask further questions themselves [763]. In particular, at the end of the interview, the patient is asked if any questions remain unanswered. The patient is also offered the possibility of seeking psychosocial support from cancer counselling centres or self-help groups [763].

Need for Research

There is a need for research on how long patients wait on average to be informed of a confirmed diagnosis. This can be determined by the retrospective collection and analysis of data from patient files, compiled by personnel within the medical practice, with due regard for data protection. The quality of the patient interview should also be recorded, but this may prove difficult due to the sensitive nature of the situation. Qualitative and quantitative interviews with affected patients may play a role.