BCC indicates basal cell carcinoma; NMSC, nonmelanoma skin cancer; SCC, squamous cell carcinoma. Unspecified NMSCs were not coded as either BCC or SCC.
eAppendix. Online Addendum
Rogers HW, Weinstock MA, Feldman SR, Coldiron BM. Incidence Estimate of Nonmelanoma Skin Cancer (Keratinocyte Carcinomas) in the US Population, 2012. JAMA Dermatol. 2015;151(10):1081-1086. doi:10.1001/jamadermatol.2015.1187
Understanding skin cancer incidence is critical for planning prevention and treatment strategies and allocating medical resources. However, owing to lack of national reporting and previously nonspecific diagnosis classification, accurate measurement of the US incidence of nonmelanoma skin cancer (NMSC) has been difficult.
To estimate the incidence of NMSC (keratinocyte carcinomas) in the US population in 2012 and the incidence of basal cell carcinoma (BCC) and squamous cell carcinoma (SCC) in the 2012 Medicare fee-for-service population.
Design, Setting, and Participants
This study analyzes US government administrative data including the Centers for Medicare & Medicaid Services Physicians Claims databases to calculate totals of skin cancer procedures performed for Medicare beneficiaries from 2006 through 2012 and related parameters. The population-based National Ambulatory Medical Care Survey database was used to estimate NMSC-related office visits for 2012. We combined these analyses to estimate totals of new skin cancer diagnoses and affected individuals in the overall US population.
Main Outcomes and Measures
Incidence of NMSC in the US population in 2012 and BCC and SCC in the 2012 Medicare fee-for-service population.
The total number of procedures for skin cancer in the Medicare fee-for-service population increased by 13% from 2 048 517 in 2006 to 2 321 058 in 2012. The age-adjusted skin cancer procedure rate per 100 000 beneficiaries increased from 6075 in 2006 to 7320 in 2012. The number of procedures in Medicare beneficiaries specific for NMSC increased by 14% from 1 918 340 in 2006 to 2 191 100 in 2012. The number of persons with at least 1 procedure for NMSC increased by 14% (from 1 177 618 to 1 336 800) from 2006 through 2012. In the 2012 Medicare fee-for-service population, the age-adjusted procedure rate for BCC and SCC were 3280 and 3278 per 100 000 beneficiaries, respectively. The ratio of BCC to SCC treated in Medicare beneficiaries was 1.0. We estimate the total number of NMSCs in the US population in 2012 at 5 434 193 and the total number of persons in the United States treated for NMSC at 3 315 554.
Conclusions and Relevance
This study is a thorough nationwide estimate of the incidence of NMSC and provides evidence of continued increases in numbers of skin cancer diagnoses and affected patients in the United States. This study also demonstrates equal incidence rates for BCC and SCC in the Medicare population.
Nonmelanoma skin cancer (NMSC) is the most common malignancy in the United States, with substantial associated morbidity and cost, as well as relatively small but significant mortality.1 Surveys using US national claims and survey databases have demonstrated significant increases in numbers of NMSCs and affected individuals (summarized in Table 1).2- 7 The most recent peer-reviewed national incidence estimate from 2006 estimated 3 507 069 NMSC cases in that year,2 greatly exceeding the number of all other cases of human malignancies combined.8 That study also estimated the total number of persons in the United States treated for NMSC in 2006 at 2 152 500. This estimate was a substantial increase from the previous US national estimate of 900 000 to 1.2 million NMSC cases from 1994.3 Recently, a published report by Guy et al4 of skin cancer prevalence using data from the Medical Expenditure Panel Survey estimated an increase from 3.1 to 4.3 million US adults in the treated for NMSC annually from 2002 to 2011. Stern’s study5 from 2010 was based on mathematical incidence modeling of NMSC and estimated that approximately 13 million white non-Hispanics living in the United States at the beginning of 2007 had had at least 1 NMSC. Moreover, about 1 in 5 white non-Hispanic 70-year-olds had had at least 1 NMSC, and most of those affected had multiple NMSCs.5 Another recent US report evaluated basal cell carcinoma (BCC) incidence trends from a US female cohort, the Nurses’ Health Study (1986-2006), and a US male cohort, the Health Professionals’ Follow-up Study (1988-2006).6 Age-adjusted BCC incidence rates increased from 519 cases to 1019 cases per 100 000 person years for women and increased from 606 cases to 1488 cases per 100 000 person-years for men during the follow-up period. Finally, evaluation of the National Ambulatory Medical Care Survey database revealed a 70% increase in NMSC-related office visits (from 910 to 1660 per 100 000 person years) from 1995 to 2007.7 Despite these increases in incidence estimates, NMSC is not typically reported to cancer registries, and interval changes in incidence are not systematically measured.
Understanding skin cancer incidence is critical for planning prevention and treatment strategies and allocating medical resources. There are very significant costs associated with the treatment of skin cancer. Skin cancer is the fifth most costly malignancy to treat in the United States.9,10 A recent report estimates the average annual cost of treating NMSC in the United States at $4.8 billion from 2007 to 2011, which increased 74% since the 2002-2006 estimate.4 With evidence of continuing increases in the incidence and public health burden of NMSC, the purpose of this study is to provide the most up-to-date estimate of incidence of NMSC in the US population and BCC and squamous cell carcinoma (SCC) in the Medicare fee-for-service population.
Our analyses were based primarily on 2 distinct Medicare databases and on national survey data. The Physician/Supplier Procedure Summary Master File (Total Claims Data Set)11 was analyzed for the years 2006 through 2012. For our primary approach to the estimation of incident NMSC, the Total Claims Data Set was used to provide total numbers of approved fee-for-service Medicare claims categorized by Current Procedural Terminology (CPT) procedure code number. However, the Total Claims Data Set does not contain information relating to patient age or diagnosis associated with each procedure code. The Medicare Limited Data Set Standard Analytic File 5% Sample Physician Supplier Data (5% Sample Data Set) is the nationally sampled Medicare database that contains information on claims filed for approved procedures with their associated International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes, patient age stratification, and counts of unique persons receiving the services. Hence, the 5% Sample Data Set allowed estimation of the proportion of procedures for skin cancer that were for NMSC, the proportion of procedures that were conducted on enrollees 65 years or older, and the mean number of procedures per affected beneficiary.
The National Ambulatory Medical Care Survey (NAMCS) is a cross-sectional survey system of ambulatory-based physicians wherein participating physicians complete a questionnaire for patient visits during a random 1-week period of the year. These visit observations are then used to provide a national estimate of physician visits and limited characteristics of these visits for that year. The NAMCS allowed estimation of the proportion of visits for NMSC for US patients 65 years or older. The NAMCS data from the years 2011 and 2012 were not available; NAMCS estimates for NMSC physician visits were evaluated from the years 2008 through 2010 to ensure there was no substantial change over time. The results were averaged and applied to create a 2012 estimate. The NAMCS database could not be used to differentiate visits for BCC and SCC because the more specific diagnosis codes were not available in the examined years.
This study does not report data from studies involving human participants; therefore, formal review and approval, or formal review and waiver, by an appropriate institutional review board or ethics committee is not applicable.
The number of skin cancers in the fee-for-service Medicare population was estimated in this study as the total of approved skin cancer treatment procedures (malignant destructions, malignant excisions, and Mohs micrographic surgeries) for that year from the Total Claims Data Set. Thus, the crude number of skin cancers for a given year was estimated by adding the numbers of approved claims for skin cancer procedure code series (11600-11606, 11620-11626, and 11640-11646 for malignant excisions; 17260-17266, 17270-17276, and 17280-17286 for malignant destructions; 17304 for Mohs surgeries in 2006 and 17311 and 17313 for Mohs surgeries in subsequent years). The total specific to NMSC was determined by multiplying the estimated crude number of skin cancers by the proportion of skin cancer procedure code claims associated with the ICD-9-CM diagnoses for invasive NMSC (173.0-173.9 for the time from January 2006 through October 2011 and 173.00-173.90, 173.01-173.91, 173.02-173.92, and 173.09-173.99 for November 2011 through 2012) and in situ carcinoma (232.0-232.9) from the 5% Sample Data Set.
Our definition for NMSC is based on the ICD-9-CM code series 173 (invasive nonmelanoma cutaneous malignancy) and 232 (in situ nonmelanoma cutaneous malignancy). Prior to November 2011, the 173 code series had 1 decimal place to define the body site of the malignancy. Limitations and assumptions of using this code set have been previously discussed.2 In November 2011, the 173 code series was modified to 2 decimal places. The first decimal place conveys the same body site information as previously, and the second decimal place defines the subsets of NMSC: 173.X1 indicates BCC; 173.X2 indicates SCC; 173.X0 indicates unspecified cutaneous malignancy; and 173.X9 indicates other specified cutaneous malignancy. In evaluating the numbers of each NMSC subtype in this study, we defined SCC as the total of invasive and in situ SCCs. Unspecified and other specified malignancies are all classified as “unspecified.” The number of procedures per affected individual and the number of unique persons who underwent at least 1 procedure were also derived from the 5% Sample Data Set. Because of the inclusion of unspecified cutaneous malignancies in our totals, we have chosen the broader term NMSC for the body of our report, instead of keratinocyte carcinoma, which refers to BCCs and SCCs only.
In our current study, the percentage of NMSCs treated in the Medicare population attributed to those beneficiaries 65 years or older was established directly from the 5% Sample Data Set. In our group’s previous study,2 the percentage of NMSCs treated in the Medicare fee-for-service population attributed to beneficiaries 65 years or older was derived by multiplying the rate of NMSC procedures per person by the number of beneficiaries in the 2 age categories (<65 and ≥65 years).2 Our current calculation results in a more accurate estimate by using data directly from the Medicare 5% Sample Data Set.
The proportion of the entire US population (age ≥65 years) covered under Medicare was derived from the Centers for Medicare & Medicaid Services 2013 Trustee’s report12 and US census data,13 allowing estimation of the number of NMSCs in the entire population 65 years or older. The proportion of total office visits for NMSC ICD-9-CM codes (173.x and 232.x) for patients 65 years or older in 2008 through 2010 was obtained from the NAMCS. The number of NMSCs in the US population 65 years or older was divided by the proportion of office visits for NMSC in that same patient group to estimate the total number of skin procedures for NMSC in the United States. The total number of persons in the United States diagnosed with NMSC in a given year was calculated from the skin cancer procedure totals and the number of NMSCs per affected Medicare patient. More detailed representation of the calculations described herein is provided in the eAppendix in the Supplement.
The Total Claims Data Set does not contain age-stratified data. Therefore, although the crude skin cancer procedure rates were derived using this database, age-specific rates could not be determined. The 5% Sample Data Set allowed calculation of diagnosis-specific, age-specific, and age-adjusted procedure rates. The age-adjusted NMSC Medicare Part B fee-for-service procedure rate was calculated, standardized to the year 2006, as in our group’s previous study.2 Age-adjusted rates were determined by a direct standardization method. Age intervals were used (<65, 65-69, 70-74, 75-79, 80-84, and ≥85 years).
The total number of fee-for-service Medicare skin cancer procedures increased by 13% over the 7 years of the study (2006-2012) from 2 048 517 to 2 321 058 (Table 2). The age-adjusted rate of skin cancer procedures increased from 6075 per 100 000 beneficiaries in 2006 to 7320 in 2012 (Table 2). Even though the population is aging, age adjustment in Medicare did not have a large effect on the rate of skin cancer procedures because of the trend of increased numbers of enrollees who are younger than 65 years. Using the diagnosis-specific 5% Sample Data Set, we found that the overall number of NMSC-specific procedures increased by 14% from 2006 to 2012 from 1 918 340 to 2 191 100 (Table 3). The number of persons undergoing at least 1 procedure for NMSC increased by 14% from 2006 to 2012 (1 177 618 to 1 336 800), while the average number of procedures per beneficiary remained stable during the study period (Table 3).
Our calculations of the incidence of NMSC and the number of affected individuals in the US population in 2006 and 2012 are detailed in the eAppendix in the Supplement. In Table 4, the key figures of the calculations are listed, including the total procedures for NMSC in fee-for-service Medicare patients 65 years or older, the percentage of the US population 65 years or older enrolled in fee-for-service Medicare, the number of procedures for NMSC in the United States in persons 65 years or older, and the NAMCS estimates of the proportion office visits for NMSC diagnoses in the US population ascribed to patients 65 years or older.
We estimate the total number of NMSCs treated in the United States in 2012 at 5 434 193. From the 5% Sample Data Set, we found the number of skin cancers treated per affected patient in 2012 to be 1.64. If this number is extended to the US population, the number of persons in the United States who were treated for NMSC in 2012 can be estimated at 3 315 554. Using improved methodology as described herein, the proportion of the procedures for NMSC in 2006 performed on Medicare fee-for-service patients 65 years or older is 96%. This figure was previously estimated at 84%. Using this new percentage and completing the calculations on the previously published data from 2006, we estimate the number of incident NMSC cases in the US population at 4 013 890 for that year. The number of persons who were treated for at least 1 skin cancer in the United States in 2006 can be estimated at 2 463 567.
Using the diagnosis-specific 5% Sample Data Set, we calculated the number of NMSC treatment procedures specific for BCC, SCC, and unspecified carcinoma. In the Medicare fee-for-service population in 2012, there were 1 029 660 BCCs, 1 027 700 SCCs, and 133 740 unspecified cancers (Table 3). The numbers of persons with at least 1 of the 3 NMSC subtypes treated in 2012 were 726 840 for BCC, 708 540 for SCC, and 98 380 for unspecified cancers. Evaluation of age-specific procedure rates for NMSCs revealed increasing rates of both BCC and SCC as beneficiaries age, with a levelling off for SCC and a decrease for BCC in the oldest individuals (Figure). The age-adjusted procedure rates per 100 000 beneficiaries specific to NMSC, BCC, and SCC in 2012 were 6983, 3280, and 3278, respectively. The ratio of treated BCC to SCC was 1.0. Across all Medicare fee-for-service CPT codes, the ratio of BCC- to SCC-specific diagnosis codes for 2012 is 1.0. If our estimate incidence excludes in situ SCC, the ratio of BCC to invasive SCC in the Medicare fee-for-service population is 1.23.
Nonmelanoma skin cancer is the most common malignancy in the United States, but it is generally not reported to registries and is treated mainly in the office-based setting. Therefore, the exact incidence of NMSC and its subtypes is not accurately known. The present study provides an updated US incidence estimate for NMSC in 2012 as well as a more accurate 2006 estimate. The data presented in this study indicate that the incidence rates of skin cancer continue to rise dramatically, with a 100% increase from1992 to 2012 in the Medicare fee-for-service population and a 35% increase in NMSC in the US population over the 6-year period 2006 through 2012. This study also indicates that an approximately equal number of BCCs and SCCs were treated in the Medicare fee-for-service population in 2012.
Despite the huge number of NMSCs in the US population, estimation of the proportion of BCCs and SCCs in the NMSC cases has been difficult. We were surprised to find the ratio of BCC to SCC of 1.0 in the Medicare population. Historically, the ratio of BCC to SCC in general has been estimated at 4 to 1.14 However, recent reports indicate a shift in the numbers of SCC in relation to BCC. Data from Australia have shown a significant change in relative proportions of NMSC with time as the BCC to SCC ratio has decreased from 4 to 1 in 1985 to 2.5 to 1 in 1995.15 A survey of US Mohs fellowship directors indicates an increasing ratio of SCC to BCC treated by Mohs surgery from 1985 through 2006.16 In South Florida, a study of NMSC treated in both Medicare and privately insured patients in 1996 indicated that SCCs were 2.5 times more prevalent than BCCs.17 There is also evidence that although BCCs are much more common than SCCs in the younger population, there is a disproportionately rising incidence of SCC in older age groups.18,19 Thus, an increasing incidence of SCC relative to BCC in our aging Medicare population with heavy, chronic UV exposure may be an underappreciated emerging trend.
In this study, we have updated our group’s previously published 2006 US estimate of NMSC2 upward from 3 507 069 to 4 013 890, and the number of affected individuals is also revised upward from 2 152 500 to 2 463 567. This update is based on 1 improvement in the calculation methodology for the proportion of the procedures for NMSC in Medicare B fee-for-service procedures were performed in beneficiaries 65 years or older. The previous technique multiplied the rate of NMSC procedures per affected person by the number of persons enrolled in Medicare B fee for service. However, for the present study, we used the 5% Sample Data Set to provide a more direct measure of this statistic.
The chief strengths of our current NMSC estimate are that it is up to date and based on the largest, most representative databases available. Revised ICD-9-CM coding in 2012 provides for the most specific definition of treated cutaneous malignancy to date, allowing for estimation of BCC and SCC rates in the Medicare fee-for-service population. Another strength of our NMSC subtype analysis is that the ratio of BCC to SCC of 1.0 holds across all CPT codes reported in Medicare fee for service, across pathology code 88305 (which is reported by pathologists in the diagnosis of skin biopsies and skin excisions), and across all skin cancer treatment codes.
The recent ICD-9-CM update for NMSC coding provides valuable diagnostic information on BCC and SCC, but there is some lack of specificity of the fifth digit of the code series for NMSC. The 173.X0 and 173.X9 codes designate “unspecified” and “other specified” NMSCs, respectively. These codes apply to 6% of treated NMSCs, but the composition of the skin cancers represented by these codes is unclear. Given the large numbers of cancers involved, we suspect that they are mainly BCCs and SCCs coded to a lower level of specificity. However, they also include other uncommon cutaneous malignancies. Diagnostic drift between actinic keratosis and SCC in situ may also contribute to increases in SCC incidence.20 Once data from the 2012 NAMCS becomes available, an additional independent national estimate of SCC and BCC numbers may be possible. Large private insurer databases may also be useful to assess the generalizability of our conclusions.
The assumptions and relative limitations of our estimation model for US incidence of NMSC have been discussed previously and should hold true in this study as well.2 Of note, in this and our group’s previous NMSC incidence estimate,2 we have assumed that the rate of NMSCs per affected individual is the same in the US population as a whole as for the Medicare fee-for-service population.
To our knowledge, no US study has evaluated the average number of skin cancers treated or diagnosed in a year on individuals affected with skin cancer. However, a study from Australia shows that individuals across all age groups referred for NMSC treatment had an average of 1.87 NMSC lesions that needed treatment.21 Although our data indicate a small decrease in the rate of NMSCs per affected individual in the youngest Medicare age group, we chose to use the conservative estimate from the entire Medicare population (1.64 NMSCs per affected individual), which may result in an underestimation of affected persons when applied to the non–Medicare fee-for-service population. In addition, diagnosis and age estimates from the 5% Sample Data Set were extrapolated to the Total Claims Data Set in both the present study and our group’s earlier incidence study.2 However, given that the 5% Sample Data Set was designed as a nationally sampled and statistically representative data set for the Medicare fee-for-service population, we are confident that little error is added with these calculations. Another limitation is that it is unclear how the use of nonsurgical treatments of NMSC, including radiation therapy and topical treatments such as imiquimod and fluorouracil, has changed in the years between the NMSC estimates.
The public health ramifications of increased US skin cancer incidence levels has been acknowledged by the federal government with the release of the US Surgeon General’s Call to Action to prevent Skin Cancer.22 This call to action proposes a series of goals and strategies to increase education about the risk of UV radiation exposure; increase opportunities for UV radiation protection; and strengthen research, surveillance, and monitoring related to skin cancer prevention. Our current report provides the strongest NMSC estimate published to date. We hope that it provides further evidence and motivation to support skin cancer prevention and treatment efforts and that it underscores the need for a national system of sentinel registries to better track the incidence of the most common, and largely preventable, cancer in the United States.
Corresponding Author: Howard W. Rogers, MD, PhD, Advanced Dermatology, 111 Salem Turnpike, Ste 7, Norwich, CT 06360 (firstname.lastname@example.org).
Accepted for Publication: March 31, 2015.
Published Online: April 30, 2015. doi:10.1001/jamadermatol.2015.1187.
Author Contributions: Dr Rogers had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Rogers, Weinstock, Coldiron.
Acquisition, analysis, or interpretation of data: Rogers, Weinstock, Feldman.
Drafting of the manuscript: Rogers.
Critical revision of the manuscript for important intellectual content: Rogers, Weinstock, Feldman, Coldiron.
Statistical analysis: Rogers, Weinstock.
Study supervision: Weinstock, Coldiron.
Conflict of Interest Disclosures: Dr Weinstock reports financial relationships in the past year with AbbVie, Castle, Celgene, Melafind, Merck, Palatin Technologies, and Personal Care Products Council; he reports financial relationships 2 or more years ago with Astrozeneca, Navigant, and Idera. No other conflicts are reported.
Previous Presentation: This article was presented at the 47th Annual Meeting of the American College of Mohs Surgery; April 30, 2015; San Antonio, Texas.