The database which collects information about cancer in the US has been updated this month. The HARD DATA is in for 2013. Yes, I know that was 6 years ago, but the data always lags reality by about four years. It is a difficult collection of complex data, from several different sources, to collect in any more rapid means. The important thing about cancer data, is not a single year number, but TREND LINES in the data. The upward/increasing trend lines in oral and oropharyngeal cancers remain unchanged from my early years at OCF, and the projections are that this will continue to increase in incidence for most of our lifetimes. I will post soon with a follow up that speaks to this reality which has been written about in peer reviewed scientific papers with projections as far out as 2060. I have always believed that some of this is alterable, if not in our generation, in our children’s. But much has to change for that to occur. You should be aware that this is a raw batch of data and interpreting it and parsing it to get any given number can be done in many different manners. The database of raw unsorted data is called the SEER database, kept by the National Cancer Institute (NCI). SEER stands for surveillance, epidemiology, and end results, or essentially what happened, what caused it, and what was the ultimate outcome. More on this further down the post.
The Oral Cancer Foundation makes an annual prediction for the incidence of oral and oropharyngeal cancers each year in the US. This year the number has increased again, and OCF estimates that 54,000 Americans will be newly diagnosed with oral or oropharyngeal cancer in 2019. About 9,750 individuals will die from this cancer in 2019, also an increase over last year. The foundation has been issuing estimates since 2002, and with a retrospective eye, we have been very accurate in our estimates. When I started OCF around 2000, the annual incidence rate had held steady for many decades at about 30,000 people. Today we are almost at 54,000. I find these numbers sobering and sad. Much has changed to cause this, and much needs to be done to reduce these numbers. Many mechanisms to improve this are available to us today, but they are poorly implemented. In this post I would like you to understand these numbers and how they are collected, so that you have confidence in their validity. In the next post tomorrow I will speak to the issues of why we are doing so poorly at bringing this number at least to a plateau, and the obstacles to eventually seeing it reduced.
First, here is an overview of cancer in the US. If you think that we track all cancers in the US, all individuals that get cancer, and ultimately what happens to them, you would be wrong. Understanding the enormity of the idea will make it clear why it is done the way that it is. With a population of about 323 million individuals at the end of 2016, it is estimated that somewhere near 1,685,000 Americans were newly diagnosed with a cancer. That number does NOT include most skin cancers (squamous cell and basal cell carcinomas, nor very early finds of carcinoma in situ). About 595,700 deaths were caused by cancers in 2016 in America, or about 1,650 people a day. At the end of the year (based on end of 2013 data), we had access to numbers on survivorship which can be parsed out by cancer types and stages from the dataset, again yielding a best estimate. The number of Americans that have survived a cancer or were currently in treatment was estimated to be about 14.5 million.
A system that would tell us about every case, covering this large a data set of people does not exist, and we have no accurate mechanism to employ such a data collection. What we do is collect data from representative areas of the country, and extrapolate that into a national number. The SEER network of institutions that report covers about 28% of the US population. They are located in many different geographic locations from rural to urban. But keep in mind that when you see different organizations report slightly different numbers, they are all estimates, as exact numbers are not knowable, and some organizations may sort the SEER data differently. When I calculate OCF’s number, I use ALL incidences of the disease. The database can be parsed in an almost infinite number of ways, by years, by ethnicity, by gender, by age, by stage of disease, and much more. We do not split hairs unless we are trying to look at a particular age group or sub population. The number OCF is issuing today is drawn from “all comers” with no distinctions for race, age, gender, etc. Brian