Screening finds skin cancer, but does it save lives?

Hi all

This was reported by Reuters this week from Archives of Dermatology:


Screening finds skin cancer, but does it save lives?

Thursday, May 24, 2012

By Andrew M. Seaman

NEW YORK (Reuters Health) – Doctors find a high number of malignant tumors when a state-wide skin cancer screening program is introduced, says a new study.

Based on results from a program in Germany, researchers say 116 people need to be screened for skin cancer and five people need to have a biopsy to find one malignant tumor.

They, however, cannot say whether the screenings actually saved lives.

Still, the numbers reported in the new study are “quite good,” said Dr. Alexander Katalinic, one of the study’s coauthors, in an email to Reuters Health.

In the United States, the last time the government-backed U.S. Preventive Services Task Force (USPSTF) looked at the screenings in 2009, the group said there was not enough evidence to recommend full-body exams to check for signs of skin cancer in adults. The USPSTF, however, did not recommend against it either.

Dr. Virginia Moyer, the chair of the USPSTF, said the group reviews its guidelines every few years, and as for now its 2009 recommendation stands.


For the study, the researchers examined data from the Skin Cancer Research to Provide Evidence for Effectiveness of Screening in Northern Germany (SCREEN) program, which was conducted in the German state of Schleswig-Holstein between 2003 and 2004. (Germany has had a national skin cancer screening program in place since July 2008.)

More than 360,000 people older than 20 years old and living in the state chose to be screened by doctors who went through a special training to identify suspicious skin lesions or moles.

Some people decided to see a dermatologist while others went to a general practitioner who referred them to a dermatologist if they suspected skin cancer.

Overall, about 16,000 people had a biopsy — about one for every 23 people who were screened. Doctors identified about 3,100 malignant tumors from those biopsies.

The cost for each screening is about $27. A biopsy can run over $100, and the removal of a malignant skin lesion can cost about $800.

The researchers cannot say how many screenings led to an unnecessary biopsy or treatment, because of the program’s design. But, Katalinic said “of course there are false positives.”

Overall, there were 3,103 malignant skin tumors, and 585 of those were malignant melanomas, the most deadly type.

More than 50 people between the ages of 20 and 49 years old had to have a biopsy to identify one melanoma. That’s more than double the 20 biopsies needed to find one in people over 65 years old.

Katalinic said, as an epidemiologist, he thinks the number needed to screen or biopsied should be improved, especially among younger people. There were also 1961 basal cell carcinomas, 392 squamous cell carcinomas and 165 were other types of malignant tumors.


The United States’ National Cancer Institute says about two million people in the U.S. are treated for basal cell or squamous cell cancers every year. As for melanomas, the American Cancer Society says more than 76,000 people will be diagnosed with one in 2012 and about 9,000 will die from it.

“The main question is do we find the right skin cancers to prevent deaths and morbidity (or costs),” said Katalinic.

Another study Katalinic coauthored reported that there was a significant drop in deaths from skin cancer following the program.

Deaths from skin cancer fell by about 50 percent — to one melanoma death or fewer per every 100,000 people each year. The number of deaths from skin cancer in neighboring regions that didn’t do screening remained the same over time (see Reuters Health story of May 7, 2012.)

Even that drop, however, could be due to an increased awareness of skin cancer and not the screening itself. The study also didn’t look at whether there was a decrease in overall deaths.

“This is terrific information and certainly will be useful,” said Moyer of the new study. But, she added, it does not answer the question of whether the screening led to better health outcomes. Also, she said the study could have benefited from a comparison group, who did not get screened.

Moyer said the approach of the SCREEN program may also be a way for the U.S. to solve the problem of there not being enough dermatologists to screen everyone for skin cancer.

Basically, specially trained doctors who are not dermatologists act as gatekeepers and refer patients who have suspicious lesions or moles to dermatologist.

“In terms of feasibility, if it were the case that screening were shown to result in better health outcomes, this would be one possible way to manage that,” said Moyer.

SOURCE: Archives of Dermatology, online May 21, 2012.

Reuters Health

  1. #1 by Cliff Rosendahl on May 29, 2012 - 9:45 am

    Interesting. One in 615 people screened had a melanoma. In Australia one in 800 people (all ages and includes infants) get a melanoma each year. In my practice there is one new melanoma for each 30 unique patients seen. I think the message is that you cannot extrapolate these figures to Australia.

  2. #2 by Bob Shirlaw on May 29, 2012 - 10:57 am

    i think the authors of the study need to join in the SCARD audit and see how well Australian doctors can diagnose and treat skin cancer and how cheaply compared to these figures. I suggest they upskill so they don’t have to biopsy so many lesions to find a cancer. If you are not getting at least 75% correct for non melanoma skin cancer and 1 in 7 or less being melanoma then you could benefit from a crash course in dermoscopy.

    • #3 by Dr Ian Katz on May 29, 2012 - 11:02 am

      Bob – where do you get that figure of 1 in 7 to diagnose melanoma as a target? Surely that is so dependent on the population studied to be meaningless?

      • #4 by Bob Shirlaw on May 29, 2012 - 1:06 pm

        Hi Ian. My audit figures using Cliffs SCARD audit gives me 1 melanoma for every 4.2 lesions tested for melanoma and percentage of lesions tested for NMSC that were NMSC of 86.17% but i gave the rough figures pooled results to be more conservative which is 1 melanoma for 7.56 lesions tested and 71.97% for % of lesions tested for NMSC that were NMSC. Cliff knows how these figures are calculated and i find it helpful to compare to the audit to make sure i am doing a good job. The days of cutting out 100 moles to find a melanoma are long gone. Most of the melanomas are still in situ so we are not boosting our figures by just excising obvious ones. Bob

      • #5 by Dr Ian Katz on May 30, 2012 - 5:24 am

        Hi Bob
        I respectively think you are mistaken. The 100 to 1 ratio may be quite OK in a population with a very low prevalence of melanoma. The doctor who cuts out 1 in 100 in a GP practice in the middle of Sydney or Germany for that matter may be just a good a skin cancer dermoscopist as the doctor sitting in in a dedicated skin cancer clinic in Queensland. They may have exactly the same sensitivity and specificity for detecting melanomas and be equally as good. One cannot use the b9 to malignant ratio to make statements how good a doctor is or not without taking into account the prevalence of melanoma in the population they are each seeing.

  3. #6 by Dr Jeff Keir on May 29, 2012 - 11:20 am

    Same old banging on about this – since when is life saved the only measure of outcome? What about reduced morbidity from earlier less disfiguring treatment of even non lethal malignacies? No one seems to want to measure that!? As pointed out, the cost benefit ratios would depend on disease prevalence as well… so we need to use Aussie figures.

  4. #7 by amlilley on June 10, 2012 - 11:54 am

    Screening program for skin cancer – bring it on! Depends on what your goal is. If it’s to save lives from melanoma then the cohort age will be wide and the number of normal results will be high. If you believe the patients we normally see who self select to be checked are a representative cohort of the broader population in your area, then NNT of 8 to 1 would be a reasonable aim in my area (Newcastle, NSW). We do not have as many melanomas as QLD where NNT approaches 4 to 1 but more than VIC where NNT is probably closer to 15 to 1. Same dermatoscopy skill, different disease incidence.

    The Germans are at an advantage when it comes to healthcare as they appear to have more money available (just ask Greece!) to spend on their population.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: