SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

Background

It is known that the incidence of skin cancer is rising rapidly worldwide, but no reliable figures on multiple nonmelanoma skin cancer (NMSC) are available.

Aim

To determine the actual incidence of skin cancer in dermatology practice and to estimate how this relates to the first primary tumours (registered at the Eindhoven Cancer Registry).

Methods

We examined 1001 randomly selected patient records at Catharina Hospital Eindhoven for mention of skin cancer. For each patient, skin cancers were recorded in a database, starting from 1 January 2004 until 1 March 2010. The time interval between tumours and any history of skin cancer were also recorded.

Results

Of this group, 876 patients were treated for skin cancer during the study period. We recorded a total of 2106 tumours with a mean of 2.4 skin cancers per patient. Nearly half (46%) of patients developed multiple tumours, and the second tumour developed within a median period of 5 months. Over a quarter (28%) of patients were known to have had skin cancer before 2004, the start of the study period.

Conclusions

The number of NMSCs in practice differs substantially from the number of first primary histologically confirmed NMSCs, as usually reported by the Eindhoven Cancer Registry. To obtain the optimum benefit from registration of NMSC, it is recommended to register all NMSCs, because only this complete number will give an insight into the incidence of the rising skin-cancer numbers. Because subsequent tumours occur frequently, NMSC should be regarded as a chronic disease, and innovations in disease management are required for cost-effective control.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

The worldwide incidence of nonmelanoma skin cancer (NMSC) has risen dramatically over recent decades.[1, 2] The high incidence of NMSC is accompanied by a high incidence and prevalence of premalignancies, mainly actinic keratosis (AK). Registration of premalignancies does not take place, and thus the exact magnitude of NMSC incidence can only be estimated.[3] Precise figures for incidence and prevalence of skin cancer are lacking worldwide, owing to a lack of published data by most cancer registries on the incidence of squamous cell carcinoma (SCC) and basal cell carcinoma (BCC). Epidemiological reports of BCC focus only on the first primary histologically confirmed BCC, for which figures are available from the few registries that do monitor this disease.[2]

The Eindhoven Cancer Registry (ECR) is part of the Netherlands Cancer Registry, and has the most complete registry on NMSC of the Netherlands. The ECR registers all cancers for the south-east part of the Netherlands, and the Catharina Hospital in Eindhoven is one of the hospitals that belong to the ECR registration area. However, the actual incidence of NMSC in clinics (mainly dermatology clinics) differs from the number of first primary lesions reported by the ECR. This difference has implications for healthcare organization and costs.

In this study, we estimated the actual numbers of skin cancer in a dermatology department by screening the patient records, and we describe how and why these figures differ from first primary figures in the cancer registry.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

The study was performed using data from the Dermatology Department, Catharina Hospital (Eindhoven, the Netherlands), which is a regular Dutch dermatological practice and a tertiary referral centre for Mohs micrographic surgery. Data were collected in the period between July 2009 and March 2010. All records were screened, and variables were recorded by one investigator (SvdG).

For this study, we used the list of patients at our dermatology department with a diagnosis treatment code (DBC) of ‘closed malignant’ (meaning the treatment pathway for malignancy had come to an end) in 2007. DBCs have been used since 2005 in Dutch hospitals for registration and billing purposes. Skin cancers diagnosed from 1 January 2004 to 1 March 2010 were extracted from the hospital database, resulting in a list of 3563 DBCs in 1860 patients.

Of these 1860 records, 1005 were randomly selected by an independent nurse, and manually screened by one of the authors (SvdG). Of these 1005 records, 1001 (497 men, 504 women, mean ± SD age 64 ± 14 years; range 16–96) were unique patients, with the other 4 apparently being duplicate records (Fig. 1). Patient characteristics were recorded, including age, gender and zip code of residence.

image

Figure 1. Study design: selection of patients included in the study.

Download figure to PowerPoint

All available skin cancers (histologically proven and clinically diagnosed) since 2004 were recorded. The first tumour mentioned in the medical record after 1 January 2004 was named T1, which was not necessarily the first primary tumour. All subsequent tumours were recorded consecutively as T2, T3, etc. We registered all skin cancers that could be traced in our dermatological medical records, including tumours that were excised by the patient's general practitioner (GP) or by a surgeon in our hospital, after which the patient presented to us for follow-up. The characteristics of each tumour were also recorded, including clinical diagnosis, histological diagnosis, date of diagnosis, whether it was a recurrent tumour, treatment given, and date of treatment. We included BCC, SCC, Bowen disease, melanoma, lentigo maligna, and lentigo maligna melanoma. We also recorded any cases of lesions recorded as histologically benign when a malignancy was expected, and vice versa. In addition, we noted whether a patient had developed a skin malignancy in the years before 2004, or if they had received a diagnosis of actinic keratosis (AK), and reported the total number of visits to the hospital and the number of telephone calls to the hospital related to skin cancer during the follow-up period, which varied per patient.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

Ethics approval

Ethics approval was not required because the study involved only retrospective analysis of data.

Patients and tumours

Maximum follow-up was 71 months (median 26, range 0–71). Of the 1001 patients, 125 were not diagnosed with a (new) skin malignancy after 2004. Of these 125 patients without skin cancer, 81 had only been seen for follow-up after a skin malignancy before 2004. There were 44 patients who did not have a malignancy, but did seem to have AK or a benign tumour. AK was present in 448 of the 1001 patients (45%).

For the 876 patients with cancer, 2106 tumours were recorded (Table 1), of which 271 appeared to be recurrent or residual tumours, thus a mean number of 2.4 tumours occurred per patient. Two or more tumours occurred in 406 patients (46%) (Table 2). The second tumour (T2) developed within a median follow-up of 5 months (Table 3). In 137 patients, T2 was diagnosed at the same time as T1, and in 39 patients T1, T2 and T3 were diagnosed at the same time. Multiple tumours were mainly BCCs (Table 2). Of the 876 patients, 280 (32%) developed a second tumour within the first year after T1. Development of subsequent tumours occurred mostly in the first years of follow-up, with a decrease in later years (Fig. 2).

Table 1. Number of tumours in 1001 patients during the study period
Tumour type n
Basal cell carcinoma1738
Squamous cell carcinoma147
Bowen disease110
Melanoma77
Lentigo maligna7
Lentigo maligna melanoma2
Other25
Total2106
Table 2. Occurrence of multiple tumours during the study period
PatientsSkin cancer, anyBCC
  1. BCC, basal cell carcinoma.

Total, n876724
With 2 tumours177140
With ≥ 3 tumours229183
Table 3. Time interval between the occurrences of multiple tumours in the study period
 T2T3T4T5
  1. a

    Median (range). T1–T5, tumour number.

Patients, n406225143105
Intervala from T1, months5 (0–57)11 (0–62)13 (0–58)19 (0–63)
image

Figure 2. Percentage of the 876 patients developing a second and third tumour in the following years after development of T1.

Download figure to PowerPoint

Skin cancer before 2004

Based on the medical records, 284 patients (28%) had a history of skin cancer before 2004. Nearly 50% (n = 140) of the 284 patients developed ≥ 2 or more tumours after 1 January 2004 (Table 4). Of the 463 patients in the database who were diagnosed with one tumour, 69 (15%) were known to have skin cancer before 2004.

Table 4. New tumours after 1 January 2004, in 284 patients known to have had skin cancer prior to 2004
New tumours, nPatients
n %
07526
16925
≥ 214049

For 195 of the 284 patients with a history of cancer (69%), we could trace the date of the first tumour before 2004, which was known at our hospital. The median period between the first tumour that was diagnosed before 2004 and the first tumour diagnosed after 2004 (T1) was 6 years (range 0–33). Diagnosis of skin cancer was made at various institutions, including our department, other departments in our hospital, other hospitals or general practices.

Treatments

Patients underwent a total of 2061 treatments during the study period (Table 5). The remaining 45 tumours were still waiting for treatment or the patients had refused treatment. A total of 1577 telephone calls and 10 620 appointments were recorded for 2106 skin tumours, giving a mean of 0.75 telephone calls and 5 appointments per tumour. Appointments for AK were excluded this analysis. Per patient, with a mean of 2.4 tumours, 1.8 telephone calls and 12 appointments were needed over a mean period of 29 months.

Table 5. Treatments performed in the study period for the 2106 skin cancers in 876 patients
Treatments n
  1. *Aldara and †Efudix (both from Meda Pharmaceuticals, Bishops Stortford, Cambridgeshire, UK).

Surgical excision999
Mohs' micrographic surgery375
Photodynamic therapy606
Imiquimod cream*15
5-fluorouracil 5% cream2
Cryotherapy35
Slow Mohs surgery22
Radiotherapy7
Total2061

High-risk patients

The presence of patients with high-risk skin cancer in the database was random; such patients were neither specifically selected nor excluded. There was one patient with naevoid BCC syndrome, who developed 65 BCCs, and one patient with familial atypical multiple mole/melanoma syndrome who developed 2 melanomas, while 8 patients who had undergone organ transplant had 58 tumours between them (Table 6). Three patients had been treated with local radiotherapy for other diseases; two of these developed one BCC in the treated area, while the third developed five BCCs in the treated area.

Table 6. Skin cancers in organ-transplant patients in the study period
 Tumours, n
Basal cell carcinoma20
Squamous cell carcinoma21
Bowen disease10
Other7
Total58

Tertiary referrals/Eindhoven Cancer Registry registration area

Of the 1001 patients, 834 (83%) had been referred by their GP, and were living in the ECR region, while 55 patients (5%) were not inhabitants of the ECR registration area. There were 157 tertiary referrals, of which 45 were not inhabitants of the ECR registration area.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

Epidemiological studies on NMSC have focused on first primary tumours. The incidence figures of the Eindhoven ECR, for example, have been used in multiple studies.[2, 4, 5] A recent study based on the ECR showed that the incidence of BCC is still rising, and that the BCC incidence reported by de Vries et al. was higher than expected.[2, 4] Calculations using only first primary BCC figures would give a total of 724 primary BCCs for 724 of our patients, but in fact we found 1738 BCCs. Hence, our study indicates that the actual incidence of NMSC, and especially BCC, is far higher than expected, and that a large difference exists with the official first primary tumour figures. Our figures are still an underestimate of the actual incidence of NMSC, as skin cancers treated elsewhere, cancers not known at our hospital, and AKs were left out of this calculation.

Our results are comparable with the literature, in that we found that most patients develop a second tumour within 5 years of their first BCC.[6] Based on the medical records, 28% of patients were already known to have had some type of skin cancer, and on average, they had developed that skin cancer during a median of 6 years prior to 2004. This suggests that for patients for whom only one tumour was recorded, this tumour might actually have been their second or even third (or more). These findings indicate that NMSC should be regarded as a chronic disease, i.e. one with long duration and slow progression.

Detailed data on the numbers of NMSCs and their treatment in everyday clinical practice are important for surveillance, prediction of prognosis, improvements in quality of care and treatment, and research purposes.[7-9] In addition, it is necessary to have accurate figures on the incidence of NMSC for the calculation of healthcare costs. Because of the large gap between the number of first primary NMSCs and the actual incidence of tumours found in our study, we consider that the numbers for first primary tumours are unsuitable for such calculations.

In Denmark, an attempt is being made to set up a clinical database for NMSC.[9] Its completeness of registration was compared with that of the Danish Pathology Registry (DPR), and the medical records of a number of patients were screened; the overall completeness of registration was 62% compared with the registration of the DPR, and the completeness was 76% based on the medical records. Those results illustrate the difficulty of obtaining complete registration of NMSC in dermatology practices. Incomplete clinical registration is probably a result of the time pressure on physicians and the lack of manpower to register all information in the database.

In our outpatient clinic, the medical records comprise a full electronic patient file. The large amount of information that is reported in these patient files could be an excellent resource for monitoring the number of skin cancers and their treatment, complications and prognosis. Feedback features could be used to evaluate complications and recurrence rates, and to determine the adjustments needed to improve outcome. So-called ‘process-mining techniques’ have recently been applied for this purpose in the disease management of strokes.[10]

A limitation of our study is the fact that it was a single-centre study, and figures from other hospitals in the Netherlands were not included. As our institution is a tertiary referral centre, we do diagnose and treat more skin cancers than do other institutions, but the tertiary referrals are a small part of the dermato-oncology care that we handle.

The large number of skin cancers that we recorded all needed treatment, and patients were followed up according to the guidelines existing at that time. Treatment of AK in 44% of patients was not included in the number of hospital visits. These figures highlight the need for innovations in the overall management of skin cancer, which should be considered a chronic disease needing ongoing monitoring and treatment.[7]

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

We found a high incidence of NMSC in our dermatology practice, and a significant difference from the official skin-cancer figures, which are based only on first primary lesions. This difference has considerable implications for healthcare systems and healthcare economics related to skin cancer. Registration of NMSC will remain a challenge, owing to the very high number of skin cancers and the workload required for appropriate registration. Methods to reduce the workload involved in recording data are needed, and process-mining techniques may be a promising method for effective use of the data available in electronic patient files. Having accurate data on the numbers of NMSCs in dermatology practices is needed to improve management of this chronic disease.

What's already known about this topic
  • The worldwide incidence of skin cancer is rising rapidly.
  • However, precise numbers for multiple NMSCs are not available because of a lack of registration or guidelines on registration of NMSC.
  • In addition, treatment based on clinical diagnosis of NMSC is performed without histological confirmation.
What this study adds
  • We found a far higher incidence of NMSC than that indicated by the official figures, which use data for first primary lesions only.
  • Multiple tumours, particularly BCCs, are common, as is development of subsequent tumours.
  • Consequently, NMSC should be regarded as a chronic disease, and innovations in disease management will be required for cost-effective control.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions
  • 1
    Rigel D, Friedman R, Kopf A. Lifetime risk for development of skin cancer in the U.S. population: current estimate is now 1.5. J Am Acad Dermatol 1996; 35: 101213.
  • 2
    de Vries E, Poll-Franse L, Louwman W et al. Predictions of skin cancer incidence in The Netherlands up to 2015. Br J Dermatol 2005; 152: 4818.
  • 3
    Bernard P, Dupuy A, Sasco A et al. Basal cell carcinomas and actinic keratoses seen in dermatological practice in France: a cross-sectional survey. Dermatology 2008; 216: 1949.
  • 4
    Flohil S, de Vries E, Neumann H et al. Incidence, prevalence and future trends of primary basal cell carcinoma in the Netherlands. Acta Derm Venereol 2011; 91: 2430.
  • 5
    de Vries E, van der Rhee H, Coebergh J. Trends, causes, treatment and consequences of skin cancer epidemic in the Netherlands and Europe (in Dutch). Ned Tijdschr Geneeskd 2006; 150: 110815.
  • 6
    Kiiski V, de Vries E, Flohil S et al. Risk factors for single and multiple basal cell carcinomas. Arch Dermatol 2010; 146: 84855.
  • 7
    van der Geer S, Reijers H, van Tuijl H et al. Need for a new skin cancer management strategy. Arch Dermatol 2010; 146: 3326.
  • 8
    Trakatelli M, Ulrich C, del Marmol V et al. Epidemiology of nonmelanoma skin cancer (NMSC) in Europe: accurate and comparable data are needed for effective public health monitoring and interventions. Br J Dermatol 2007; 156 (Suppl. 3): 17.
  • 9
    Lamberg A, Cronin-Fenton D, Olesen A. Registration in the Danish Regional Nonmelanoma Skin cancer Dermatology Database: completeness of registration and accuracy of key variables. Clin Epidemiol 2010; 2: 12336.
  • 10
    Mans R, Schonenberg H, Leonardi G et al. Process mining techniques: an application to stroke care. In: Proceedings of the 21st International Congress of the European Federation for Medical Informatics (MIE 2008). Studies in Health Technology and Informatics, Vol. 136 (Andersen SK, ed). The Netherlands: IOS Press, 2008; 5738.

CPD questions

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

Learning objective

To provide readers with an up-to-date understanding of the effect of skin cancer on health care.

Question 1
  • Skin cancer is a chronic disease. Why?
    • a)
      Because patients develop multiple lesions.
    • b)
      Because it is a disease with long duration.
    • c)
      Because it is a disease with slow progression.
    • d)
      Because it is a disease with long duration and slow progression.
    • e)
      Because it is often a disfiguring disease.
Question 2
  • Figures for nonmelanoma skin cancer (NMSC) from registries such as the Eindhoven Cancer Registry literature include which of the following?
    • a)
      All NMSC.
    • b)
      All histologically proven NMSC.
    • c)
      Histologically proven NMSC plus actinic keratoses.
    • d)
      Histologically proven primary NMSC plus actinic keratoses.
    • e)
      Histologically proven primary NMSC.
Question 3
  • The risk for recurrent of subsequent tumours after a primary tumour is highest in which year after diagnosis?
    • a)
      The first.
    • b)
      The second.
    • c)
      The third.
    • d)
      The fourth.
    • e)
      The fifth.
Question 4
  • Where can reliable figures for nonmelanoma skin cancer be obtained?
    • a)
      Hospitals.
    • b)
      Registries such as the Eindhoven Cancer Registry.
    • c)
      Insurance companies.
    • d)
      Pathology laboratories.
    • e)
      No reliable figures are available.
Question 5
  • Computer systems could be used to monitor the number of skin cancers that are diagnosed and treated at hospital. This can be best performed with the help of which of the following features?
    • a)
      Process mining.
    • b)
      Workflow management system.
    • c)
      Data mining.
    • d)
      Store and retrieve features.
    • e)
      Feedback features.

Instructions for answering questions

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  9. CPD questions
  10. Instructions for answering questions

This learning activity is freely available online at http://www.wileyhealthlearning.com/ced.

Users are encouraged to

  • Read the article in print or online, paying particular attention to the learning points and any author conflict of interest disclosures
  • Reflect on the article
  • Register or login online at http://www.wileyhealthlearning.com/ced. com and answer the CPD questions
  • Complete the required evaluation component of the activity

Once the test is passed, you will receive a certificate and the learning activity can be added to your RCP CPD diary as a self-certified entry.