9. Health Economic Evaluation – Prevention of Skin Cancer

Inga-Marie Hübner and Jessica Achter

• Model calculations based on international data and a retrospective cost-of-illness study indicate an economic benefit/cost-effectiveness of reducing tanning bed use

The study by Guy et al. [846]investigated the impact of a reduction in tanning bed use in the USA on melanoma prevention and treatment costs. The estimates were made using a Markov model. A cohort of 61.2 million individuals aged 14 years or younger was studied at annual intervals over their lifetime. Five different scenarios were examined: restricting tanning bed use among minors and reducing use by 20%, 50%, 80% and 100%. Compared to the no age restriction case, banning tanning bed use for minors was estimated to prevent 61,839 melanoma cases (4.9% reduction) and 6,738 melanoma deaths (4.7% reduction).This would result in a gain of 142,659 life years and elimination of treatment costs over the lifetime of the 61.2 million adolescents aged 14 years or younger of approximately $ 342.9 million. The estimated health and economic benefits increase with greater reductions in the prevalence of tanning bed use, according to the model. For a 20% reduction in prevalence compared to 100%, the estimated number of melanoma cases prevented increases from 40,410 to 202,662, as does the estimated number of melanoma deaths prevented from 4,286 to 23,266. In addition, the number of life years gained increases from 91,229 to 458,592 and the therapy costs saved increases from $ 219 million to $ 1.1 billion over the lifetime of the cohort.

Hirst et al. [847] provide a prediction of preventable skin cancer cases and associated cost savings to the government associated with regulation of the tanning industry in Australia. For this purpose, a Markov model was set up in order to map the future costs and health effects of sunbed use at the current standard compared with the enforcement of regulations. Regulations are understood here as a ban on sunbed use for minors and for people with very light skin. The Markov model used comprises a hypothetical cohort of 100,000 individuals aged 15 years who go through various cycles until they reach the age of 85 years.Health effects were measured using the number of new melanoma and squamous cell carcinoma cases and life years gained. With stricter regulations, it is estimated that between 18 and 31 melanomas, 200 to 251 squamous cell carcinomas, and associated costs of $ 256,054 per 100,000 people could be prevented. This would result in the prevention of melanoma deaths with a gain of 31 life years.

Waters and Adamson [848] studied the health and economic consequences of using tanning devices. To do this, they estimated the number of health problems associated with tanning device exposure in the United States and calculated the cost of medical care in the form of therapy. The major unit of analysis for the study represented the number of individuals living in the US who sought therapy for BCC, SCC, or melanoma. To estimate the percentage of these cases attributable to tanning device exposure, a population-attributed risk (PAR) was calculated for each disease. YPLL (years of potential life lost) was also used to calculate annual medical costs on a per case basis and indirect productivity costs. The analysis shows that in 2015, there were 8,947 incidences of melanoma in the US, including 5,176 invasive and 3,771 in situ, as well as more than 86, 600 cases of SCC and 168,000 cases of BCC attributable to tanning device exposure. The cost of indirect medical care for these cases is $ 343.1 million annually and will result in an economic loss of $ 127.3 billion over the lifetime of the affected individuals. In summary, the use of tanning devices represents a significant contribution to disease and premature mortality in the United States. Furthermore, use represents a significant economic burden in terms of medical care costs and lost productivity. the use of tanning devices represents a significant contribution to disease and premature mortality in the United States. Furthermore, use represents a significant economic burden in terms of medical care costs and lost productivity. the use of tanning devices represents a significant contribution to disease and premature mortality in the United States. Furthermore, use represents a significant economic burden in terms of medical care costs and lost productivity.

In a retrospective cost-of-illness study, Pil et al. [849][849] used a Markov model to analyze the current and future economic burden of skin cancer in Belgium and the cost-effectiveness of primary prevention of skin cancer. The health-related burden of skin cancer was estimated based on the registered prevalence of skin cancer lesions in therapy or follow-up. To estimate the total economic burden of skin cancer on the population, retrospective data in the form of questionnaires completed by patients between March and June 2015 were used. A total of 287 patient-completed questionnaires were included. The median age of the participants was between 61 and 70 years. Based on the questionnaires, costs per skin cancer type were calculated for six months each, separately for the diagnosis and treatment, intensive follow-up, and long-term follow-up phases. In order to calculate the future health and economic costs of skin cancer, a Markov model with a time horizon of 20 years was set up. Costs were expressed separately as costs to the health care payer, costs to the patient, and costs due to productivity losses. The total economic cost of skin cancer in 2014 in Belgium was estimated at 106 million euros. The total cumulative costs over a 20-year period were estimated at 3.2 billion euros and over 50 years at 8 billion euros. The Markov model simulation over 50 years showed that of the 8 billion euros, 238 million euros (2.9%) could be saved by a complete ban on sunbeds. Furthermore, the budget impact analysis showed that each euro invested in an awareness campaign would save the health care payer 3.6 euros in the long run. After a period of 50 years, a ban on sunbeds would lead to a reduction in prevalence of 8.6% (absolute number: 9,491 in men and 11,335 in women). Banning the use of sunbeds would further lead to a reduction in the prevalence of squamous cell carcinoma by 22.7% (absolute number: 35,934 in men and 52,565 in women). By reducing the prevalence of squamous cell carcinomas and ma-lign melanomas, fewer tumors would subsequently develop to later stages, and therefore a reduction in skin cancer mortality can be expected. In the model, over a 50-year period, 3,927 deaths (1,602 men and 2,329 women) were predicted to be prevented by banning the use of public tanning beds. 934 in men and 52,565 in women). By reducing the prevalence of squamous cell carcinomas and ma-lign melanomas, fewer tumors would subsequently develop to later stages, and therefore a reduction in skin cancer mortality can be expected. In the model, over a 50-year period, 3,927 deaths (1,602 men and 2,329 women) were predicted to be prevented by banning the use of public tanning beds. 934 in men and 52,565 in women). By reducing the prevalence of squamous cell carcinomas and ma-lign melanomas, fewer tumors would subsequently develop to later stages, and therefore a reduction in skin cancer mortality can be expected. In the model, over a 50-year period, 3,927 deaths (1,602 men and 2,329 women) were predicted to be prevented by banning the use of public tanning beds.

Inga-Marie Hübner and Jessica Achter

• Economic evaluations of international population-based primary skin cancer prevention interventions demonstrate economic benefit/cost effectiveness.

Kyle et al. [850] provide an economic evaluation of the United States Environmental Protection Agency's so-called SunWise program. The SunWise program is a school-based teaching program for children with the goal of teaching children how to protect themselves from overexposure to the sun. The students are each given a type of test before as well as a while after the teaching sessions, regarding the students' knowledge, attitude, use as well as intended use. The purpose of the study was to assess the health benefits of the programme by means of an economic analysis in order to determine the net benefits and cost-effectiveness in a next step and to compare these data with a control group. Health outcomes were measured as number of skin cancer cases and prevention of premature mortality and QALYs saved over the 17-year course of the program. Prevented costs are measured as direct medical costs and prevented costs of lost productivity resulting from the SunWise program. Net benefits are the difference between prevented costs and program costs. The economic analysis indicates that the SunWise program should prevent more than 50 premature deaths, approximately 11,000 skin cancer cases, and 960 QALYs (undiscounted) among participants if current U.S. government financial support continues. For every dollar invested in the SunWise program, between $1.95 and $4.02, concerning medical care costs and lost productivity, are saved, according to the analysis.

Shih et al. [851] provide with their study an economic evaluation of skin cancer prevention measures in Australia. The cost-effectiveness of the SunSmart skin cancer prevention program since its inception was assessed along with its potential cost-effectiveness as an improved and ongoing national program. The primary endpoint was the reduction in melanoma incidence attributable to the SunSmart program. A utility analysis was used in which costs are expressed as dollars and outcomes are expressed as DALYs (disability-adjusted life-years). DALYs are used to ensure comparability with outcomes from a larger Australian research concerning 100 different prevention programs. This study uses data from three states, Victoria, New South Wales, and Queensland, to assess program effectiveness. These three states are representative of three latitude zones with different UV exposures. Melanoma incidence rates from the states are used to model key health outcomes. Because of the SunSmart program, more than 103,000 skin cancer cases were prevented in Victoria between 1988 and 2003, including 9,000 melanomas and 94,000 white skin cancer cases. Furthermore, more than 1,000 deaths were prevented, which equated to 28,000 DALYs and 22,000 life years. Investment in the SunSmart program in the past yielded 3.6 Australian dollars per dollar invested with an assumed reduction rate of 10% per year for white skin cancer. A retrofitted national program is estimated to prevent 120,000 DALYs over the next 20 years with associated reductions in health service resource use.

A further study by Shih et al. [852], building on the 2009 study, looked at the economic evaluation of future skin cancer prevention in Australia. The aim of the study was to update and expand the economic references for skin cancer prevention. Economic evaluations were conducted in 2015 using a range of methods applied, including a cost-effectiveness analysis and a cost-benefit analysis, as well as different study perspectives and a counterfactual analysis related to cancer incidence between the years 1982 and 2011. Modeled outcomes include "cases prevented," "deaths prevented," and "health-adjusted life-years (HALYs)." The analysis finds that implementing interventions in the form of a coordinated set of skin cancer prevention interventions over a 20-year period (2011 to 2030) would incur an additional program cost of $ 63 million Australian dollars to the government, but would also deliver a range of health, financial and economic gains for Australia. The projected health gains include 140,000 cases of skin cancer prevented at an additional per capita investment in future skin cancer prevention of 0.16 Australian dollars, the prevention of 6,200 premature deaths, and 111,000 life years gained and 92,600 health-corrected life years. The financial gains include health sector cost savings of over 200 million Australian dollars and productivity gains in the economy of 2, 269 million in the human capital approach or 221 million Australian dollars in the friction cost approach. Depending on the study perspective and methodology, the improved program (see Shih et al. 2009 study) turns out to be either dominant or highly cost-effective. Dominant in this sense refers to both the achievement of health gains and cost savings. The program is considered cost-effective if health gains are achieved at a moderate net cost. The return on investment (ROI) according to the analysis is 3.2 Australian dollars per dollar invested by the government in the program (present value, 3% discount rate), with a net societal benefit of 1.43 trillion Australian dollars. Dominant in this sense refers to both the achievement of health gains and cost savings. The program is considered cost-effective if health gains are achieved at a moderate net cost. The return on investment (ROI) according to the analysis is 3.2 Australian dollars per dollar invested by the government in the program (present value, 3% discount rate), with a net societal benefit of 1.43 trillion Australian dollars. Dominant in this sense refers to both the achievement of health gains and cost savings. The program is considered cost-effective if health gains are achieved at a moderate net cost. The return on investment (ROI) according to the analysis is 3.2 Australian dollars per dollar invested by the government in the program (present value, 3% discount rate), with a net societal benefit of 1.43 trillion Australian dollars.

In the retrospective cost-of-illness study by Pil et al. (2016) mentioned earlier, additional analyzes were conducted related to an awareness campaign. The Markov model simulation over 50 years showed that of the 8 billion Euros, 228 million (2.8%) could be saved through an awareness campaign. Furthermore, the budget impact analysis showed that each euro invested in an awareness campaign would save the health care payer 3.6 euros in the long run. After a period of 50 years, according to the analysis, the awareness campaign would lead to a reduction in the prevalence of diagnosed malignant melanoma stage I of 11.3% (absolute number: 10,954 in men and 15,053 in women). The model predicted over a 50-year period that 3,991 deaths (1,593 men and 2,398 women) would be prevented using an annual awareness campaign.

The systematic review by Gordon and Rowell [853] includes a total of 16 studies reporting national estimates of the costs of skin cancer and 11 studies reporting the cost-effectiveness of skin cancer prevention or screening. Studies were included through August 2013 via the Medline, Cochrane Library, and National Health Service Economic Evaluation Database programs. The focus here is on skin cancer as defined by MM, BCC and SCC. Depending on the size of the population considered, the annual direct health system costs of skin cancer are highest in Australia, New Zealand, Sweden and Denmark. Skin cancer prevention initiatives were found to be highly cost-effective and could also save costs. Programmes to screen for melanoma in high-risk individuals, such as older men or individuals with a family history of melanoma, may also be cost-effective. However, further analysis is needed to confirm these observations. Furthermore, a significant cost burden of skin cancer exists in many of the countries analyzed. Health care expenditures for skin cancer will increase in the future as a result of an increase in the incidence of skin cancer. Public investment in skin cancer prevention and screening programmes shows great potential for health and economic benefits.

Inga-Marie Huebner and Jessica Achter

Gordon et al. [854]

 analyzed in their study the cost-effectiveness of an intervention with the aim to encourage men over 50 years of age to self-examine their skin in order to detect skin cancer at an early stage. The cost-effectiveness analysis was conducted using data from a skin awareness study and data obtained from a review of literature. A lifetime Markov model was constructed to combine these data. In the skin awareness study, a total of 929 participants were randomly assigned to either the intervention or control group. The intervention group received a DVD on skin self-examination and the importance of seeking medical attention for suspicious lesions. In addition, participants received reminders to watch the DVD, a graphic representation of the body to note possible locations of skin lesions, and a brochure on the differentiation between benign and malignant skin lesions. Participants in the control group received only the brochure. All participants completed an assessment at baseline and at six and twelve months. The primary outcomes represented self-examinations of the skin, clinical examinations of the skin by a physician, self-efficacy, and perceived social support. For a cohort of older men over their remaining lifetime, the average cost per person of a screening intervention for skin cancer was $ 5,298 Australian compared with $ 4,684 Australian for usual care, according to the analysis. The corresponding average QALY was 7.58 for the intervention group and 7.77 for the control group. Consequently, as the intervention involved higher costs and lower health gains in terms of QALYs compared with usual care, usual care dominates the intervention in this case. When considering the gains in terms of survival, the model predicted a cost of 1,059 Australian dollars per life year saved. The probability that the intervention was cost-effective at up to 50,000 Australian dollars gained per QALY was 43.9%. In conclusion, although the intervention improved participant behavior regarding skin self-examination, which spurred men to seek medical attention for abnormal lesions, the overall costs and effects of a higher detection rate of SCC, BCC, and benign lesions outweighed the positive health gains from detecting more thin melanomas. 000 Australian dollars gained per QALY was 43.9%. In conclusion, although the intervention improved participant behavior regarding skin self-examination, which spurred men to seek medical attention for abnormal lesions, the overall costs and effects of a higher detection rate of SCC, BCC, and benign lesions outweighed the positive health gains from detecting more thin melanomas. 000 Australian dollars gained per QALY was 43.9%. In conclusion, although the intervention improved participant behavior regarding skin self-examination, which spurred men to seek medical attention for abnormal lesions, the overall costs and effects of a higher detection rate of SCC, BCC, and benign lesions outweighed the positive health gains from detecting more thin melanomas.

Hirst et al. [855] examined the lifetime cost-effectiveness of skin cancer prevention by promoting daily sunscreen use. A Markov model was used to integrate data from a community-based randomized controlled trial conducted in Australia, as well as other epidemiological and published sources. In the controlled trial, a total of 1,621 residents from Nambour in Queensland were randomly assigned to either the sunscreen intervention group or a control group. The intervention group was encouraged to apply sunscreen with SPF 15+ to their face, neck, arms, and hands every morning. Furthermore, they were given one or more bottles of sunscreen as needed. Participants in the control group, on the other hand, were instructed to use the sunscreen at their discretion. The primary outcome of the study represented the incremental cost per QALY. The discounted incremental benefit per QALY gained of the sunscreen intervention was 40,890 Australian dollars. Over the projected lifetime of the intervention cohort, this would prevent a total of 33 melanomas, 168 squamous cell carcinomas, and four melanoma deaths at an incurred cost of approximately 808,000 Australian dollars. The probability that the sunscreen intervention was cost-effective was 64% at a willingness-to-pay threshold of 50,000 Australian dollars per QALY gained.

Research Needed

The studies presented here demonstrate the high (economic) benefits of primary prevention interventions. As these are exclusively international studies, a clear need for research can be formulated for cost-benefit analyses of primary prevention measures specific to Germany. However, the challenge here is that the interventions currently implemented mostly aim to change endpoints that can be measured in the short term (such as a change in behaviour or an increase in knowledge, cf. Chapter 5) and therefore the possibilities for using relevant effect measures of economic evaluations (see above) are limited. The discussion on the integration of health economic issues in practice projects should therefore be advanced in the future.