Please remember, just because something is predicted doesn’t mean it is going to happen!
Still, this seems to be a good “tool”.
Analyzing medical records from thousands of patients, statisticians have devised a statistical model for predicting what other medical problems a patient might encounter.
Like how Netflix recommends movies and TV shows or how Amazon.com suggests products to buy, the algorithm makes predictions based on what a patient has already experienced as well as the experiences of other patients showing a similar medical history.
“This provides physicians with insights on what might be coming next for a patient, based on experiences of other patients. It also gives a predication that is interpretable by patients,” said Tyler McCormick, an assistant professor of statistics and sociology at the University of Washington.
The algorithm will be published in an upcoming issue of the journal Annals of Applied Statistics. McCormick’s co-authors are Cynthia Rudin, Massachusetts Institute of Technology, and David Madigan, Columbia University.
McCormick said that this is one of the first times that this type of predictive algorithm has been used in a medical setting. What differentiates his model from others, he said, is that it shares information across patients who have similar health problems. This allows for better predictions when details of a patient’s medical history are sparse.
- New statistical model lets patient’s past forecast future ailments (esciencenews.com)
- Algorithm Predicts Future Medical Conditions – Works Somewhat Like Netflix Based on Patient and Other Patient Experiences – Good for Clinical Arena and Words of Caution With Potential “For Profit” Utilization (ducknetweb.blogspot.com)
- Beware of False Positives (infocus.emc.com)
- False Positive Science: Why We Can’t Predict the Future (freakonomics.com)
Since I’ve heard through the rumor mill that search engines and blog readers like bullet points, I’ve decided to toss Google a bone. This post is a series of statistics from The Real State of America Atlas: Mapping the Myths and Truths of the United States. These numbers may surprise you.
Poverty and housing
- In 2000, 12 percent of Native American houses on reservations lacked complete plumbing. This situation is almost nonexistent in the rest of the United States.
- In 2009, 32 percent of Native Americans were living below the federally set poverty line. The matching statistic for whites was 9 percent.
- Subprime mortgage lending has led to many people losing their homes. 61 percent of African-American women who borrowed mortgages in 2005 received subprime ones. The matching statistic for white women was 22 percent.
Journalism and diversity
- In 2008, 88 percent of United States radio reporters and 76 percent of TV journalists were white. (In 2009, 75 percent of United States residents identified as white.)
- 53 percent of foreign-born residents of the United States are from Latin America.
- The national average number of foreign-born workers in the labor force is 16 percent.
- Meanwhile, 64 percent of United States newspapers reduced their coverage of international news between 2007 and 2009. It’s unlikely immigrants made those newsroom decisions.
Environmental emotions and actions
- 61 percent of Americans surveyed said they were sympathetic to the environmental movement in 2010.
- As of November 2010, there were 1,280 Superfund sites in the United States in line for cleanup, according to the Environmental Protection Agency. Does sympathy equal action? Not necessarily.
- In 2005, women became the majority of motor vehicle owners in the United States. However, only 26 percent of auto industry employees are women. Women are also more likely than men to believe global warming is a serious concern. Guess who’s designing our cars?
- 30 Statistics That Show That The Middle Class Is Dying Right In Front Of Our Eyes As We Enter 2012 (zerohedge.com)
- From the U.S. Census Bureau- by the Numbers: Interesting Aspects of U.S Population (worldtradedaily.com)
Health Data Tools and Statistics on PHPartners.org | Health Information Literacy – for health and well being
The following was posted on the BHIC Blog; Sep 11, 2011 10:26 PM by Cheryl Rowan
Do some statistically laden advertisements and Web sites seem misleading? Is there a disconnect between the displayed data in some ads with your gut feelings? But you just cannot put your finger on why you feel distrustful?
Just plain sloppily represented infographics could be creating some of the confusion. Infographic combines an interesting graphical element with hard data. They are commonly seen in the media, including USA Today.
Here are some excerpts from An Epidemic of Bad Infograhics: Depression
In an effort to keep trying to get people’s attention in an increasingly attention-deficit world, we get a lot of inquiries for links to websites promoting education programs and other affiliate websites. The latest effort is focused around “infographics,” those graphics made popular by the USA Todaynewspaper that combines an interesting graphical element with hard data. A well done infographic ostensibly makes data more engaging. A fantastic infographic puts data into proper perspective and gives it valuable context.
What these marketing firms send me, however, are not fantastic or even well-done. So in the interests of demonstrating that any infographic can be worse than no infographic, I’m going to critique one of the latest ones to have come across my desk. It’s about depression, one of the most common and serious mental disorders….
What about your level of depression? Well, according to the infographic — but not the research or mental health professionals — you can have different “depression levels” ranging from “Normal” (what’s a “Normal” depression?) to “Situational” or even “Major.”
Of course, the Diagnostic and Statistical Manual of Mental Disorders (the DSM-IV) doesn’t divide major depression in this manner. Instead, it specifies that major depression can be Mild, Moderate, Severe without Psychotic Features, Severe with Psychotic Features, In Partial Remission, In Full Remission, or Chronic.
I assume “Situational” refers to a completely different mental disorder — Adjustment Disorder with Depressed Mood. The person designing this graphic was obviously not too familiar with the actual information he was asked to illustrate……
Related Health Information Evaluation Resources
- Evaluating Health Information on the Internet summarizes tips and pointers
- Consumer’s Guide to Taking Charge of Health Information has 10 tips, including how to interpret risk statements, a “reliability chart”, and where to get more information. Bonus feature: snappy cartoons!
- Evaluating Health Information on the Internet by the US National Cancer Institute summarizes 10 key points. Links to related federal agencies
- Evaluating Health Information: MedlinePlus provides trusted links to overviews, specific conditions (as cancer and complementary medicine), organizations, and more
- Quackwatch is physician published guide to health fraud including 23 hot topics (as immunization,autism, homeopathy, chiropractic). While the site is a bit controversial, it does have useful information and links.
- National Council Against Health Fraud is a private nonprofit, voluntary health agency that focuses upon health misinformation, fraud, and quackery as public health problems
- patientInform is a collaborative effort of health care related professionals who interpret research articles, in the form of summaries or news items.
- What to look for when reading medical research outlines the different types of scientific studies and which ones are the best
- Participating organizations provides links to news items from over 25 publishers and organizations. “The publishers allow readers following links from patientINFORM material on the health organizations’ sites to access the full text of these articles without a subscription, and they provide patients and caregivers with free or reduced-fee access to other articles in participating journals.”
Related Statistics Resources
- Guide to Biostatistics (MedPage Today) is a bit technical, but a good introduction to biostatistical terms used in medical research
- Misleading Statistical Information in Ads: A Drug Ad Analyzed and Related Evaluation Resources (jflahiff.wordpress.com)
- Misleading information on health social sites (and tips on how to evaluate health/medical information) (jflahiff.wordpress.com)
- Women And Prescription Drugs: One In Four Takes Mental Health Meds (worldwright.wordpress.com)
- 1 in 5 of U.S. adults on behavioral meds (worldwright.wordpress.com)
- Ethical Implications of the Use of Data and Statistics (lynnmunoz.wordpress.com)
- Information Designers (fusionfinds.wordpress.com)
- 1 in 5 Adults on Behavioral Meds (abcnews.go.com)
State fact sheets provide information on population, income, education, employment, federal funds, organic agriculture, farm characteristics, farm financial indicators, top commodities, and exports, for each State in the United States. Links to county-level data are included when available.
Data last updated on April 28, 2011.
Don’t know a randomized control study from a descriptive study? Keep forgetting the difference between sensitivity and specificity in diagnostic testing?
MedPage today has a great 4 page guide with definitions and diagrams in these areas
- Study designs (as research classifications and terminology)
- Descriptive statistics (as measures and terms)
- Clinical Research & Clinical Trials(National Institute of Child Health and Human Development)
- Increasing Knowledge — How to Read a Research Paper(Lewy Body Dementia Association)
- JAMA Patient Page: Basic Science Research(American Medical Association) - PDF
Trust for America’s Health is a “non-profit, non-partisan organization dedicated to saving lives by protecting the health of every community and working to make disease prevention a national priority”.”By focusing on PREVENTION, PROTECTION, and COMMUNITIES, TFAH is leading the fight to make disease prevention a national priority, from Capitol Hill to Main Street. We know what works. Now we need to build the resolve to get it done.”
Trust for America’s Health includes the following
- How Healthy Is Your State? (home page link) where you can select a state from the drop-down menu. Statistics include adult health indicators (as cancer and asthma), child and adolescent health facts (as pre-term labor), obesity and diabetes rates, and program spending
- Current Report topics include the flu, liver diseases, preventative health care, how public dollars are spent on health care, and keeping America’s food supply safe
- Advice is given on Advocacy issues, including writing representatives and joining grassroot organizations
- The Resource Library provides links to organization and other Web sites in topics as Health Reform Legislation, Health Disparities, Environmental Health, and Food Safety
[As of July 10, 2009]
The National Prescription Audit, derived from IMS (a health care information company)Health’s Xponent, captures roughly 70% market share of all prescriptions, then uses projection methodology from a stratified and geographically balanced sample to represent 100% market share coverage of US prescription activity at retail, mail service, long-term care, and managed care outlets. Highlights include:
- During each of the last several years, fewer prescriptions have been dispensed in the United States for menopause hormone therapy (HT). In 2008, 42.262 million prescriptions were dispensed, down 6.2% from 45.054 million during 2007. This compares to 57.861 million HT prescriptions dispensed in 2004.
- Oral Premarin (conjugated estrogens) remained the most dispensed estrogen therapy during 2008, responsible for 23.3% of all dispensed scripts, down 13% versus 2007.
- Oral medroxyprogesterone acetate (MPA) remained the most dispensed progestogen.
- Estrogen-progestogen combinations were responsible for 14.8% of dispensed prescriptions during 2008, about even with 2007 levels. The most dispensed combination remained Prempro (low-dose).
- Estrogen-androgen combinations were responsible for 4.3% of dispensed prescriptions during 2008.