Table 1.
Estimated probabilities, sensitivity values and sources
| Probabilities | Used in model | Sensitivity valuesa | Sources/assumptions |
|---|---|---|---|
| Melanoma | |||
| Proportion of cases of suspected melanoma are seen in public or private setting | 20 %/80 % | 15 %/85 %, 25 %/75 % | Expert opinion, persons with white skin are more susceptible to skin cancers and seen in private settings |
| Proportion of cases seen by GP who suspects a melanoma refers to a dermatologist or surgeon | 80 % | 70 %, 90 % | Assumption based on convenience and high likelihood in public hospital to refer to specialist |
| Proportion of suspected melanoma that were malignantb | 80 % | 75 %, 85 % | Expert opinionc, Fong 2014 [12] |
| Melanoma is surgically excised | 100 % | 87.9 % | Expert opinion (all melanomas including advanced) Vallejo-Torres 2014 [10] |
| Melanoma is greater than 1 mm thick | 30 % | 20 %, 40 % | Expert opinionc |
| Melanoma greater than 1 mm thick has metastasized | 20 % | 15 %, 25 % | Expert opinionc |
| Melanoma with no metastases is treated with interferon 2b alpha | 3 % | 2 %, 4 % | Expert opinionc, Fong 2014 [12] |
| Melanoma greater than 1 mm thick has metastases in lymph nodes | 30 % | 25 %, 35 % | Expert opinionc, published literature - ranges from 4 to 44 % |
| Melanoma with lymph node metastases is treated with radical LND | 100 % | – | Expert opinionc: All those with SLNB get RLND. |
| Melanoma is treated by radiotherapy | 5 % | 4 %, 6 % | Expert opinionc, most with metastases will only get palliative care, Fong 2014 [12] |
| Melanoma is treated with chemotherapy | 10 % | 5 %, 7 % | Expert opinionc, Fong 2014 [12] |
| Non-melanoma (NM)SCC or BCC | |||
| NM is treated by a GP in the public setting | 100 % | – | Expert opinionc, all seen first by a GP, (same for Aust. and England) |
| NM case is referred to a dermatologist | 60 % | 50 %, 70 % | Expert opinionc 60 % for dermatologist or surgeon |
| Suspected NM is confirmed to be malignantc | 85 % | 80, 90 % | Expert opinionc |
| SCC is >2 cm diameter | 10 % | 8 %, 12 % | Expert opinionc |
| Large SCC is positive and surgeon treats by radical LND | 20 % | 15 %, 20 % | Expert opinionc, Fong 2014 [12] |
| NM is treated by: | |||
| Surgical excision | 80 % | 86.0 % | Expert opinionc, Vallejo-Torres 2014 [10], Fong 2014 [12] |
| Cryotherapy | 10 % | 3.1 % | As above |
| Curette and diathermy/electrodesiccation | 5 % | 7.5 % | As above |
| Topical cream | 3 % | 0.5 % | As above |
| Photodynamic therapy | 1 % | 0.8 % | As above |
| Radiotherapy | 1 % | 1.7 % | As above |
LND lymph node dissection, GP general practitioner, SCC squamous cell carcinoma, BCC basal cell carcinoma
aThe sensitivity values are the high and low estimates used in the sensitivity analysis. These are based on sources in the literature or judged as plausible ranges around the best estimate used in the model base case. For probabilistic sensitivity analyses, beta distributions were assigned to probabilities to account for uncertainty
bExpert opinion is from practicing dermatologists and practicing doctors
cLesions suspected of being Malignant are often investigated and later diagnosed as benign