Diagnosing Cancer Online

Pancreatic and colon cancer detection aided by online calculators

(RxWiki News) You're sitting with your family practitioner describing some pain and other symptoms you've been having. S/he goes to the computer and tells you it's time for tests. This online program - now available in Europe - could save your life.

An online calculator that cross-references patient symptoms and risk factors has been developed that can alert general practitioners to early signs of pancreatic and colon cancer.

"Tell your healthcare provider about ALL of your symptoms."

Researchers at The University of Nottingham and ClinRisk Ltd. have developed two new programs designed to red flag difficult-to-pinpoint symptoms related to often fatal pancreatic cancer, which has no reliable screening tools.  The objective is to diagnose this and colon cancer earlier to improve patient outcomes.

Professor Julia Hippisley-Cox in the University's Division of Primary Care led the research, which analyzed patient data from 564 general practitioner (GP) practices to develop and test the algorithm designed to predict which patients could have pancreatic cancer.

The data included a combination of symptoms such as loss of appetite, weight loss and abdominal pain, along with such risk factors as chronic pancreatitis, age, smoking and diabetes.

The program accurately predicted 62 percent of all pancreatic cancers diagnosed over the following two-year period and included the top 10 per cent of patients predicted to be most at risk.

Using the same GP practices, researchers also analyzed risk factors and symptoms for colon cancer, which has a poor survival rate in Europe. Symptoms of this cancer are often linked to other less serious illnesses. Symptoms include rectal bleeding, appetite and weight loss, diarrhea, constipation or abdominal pain.

But if doctors relied only on the most common signal of colon cancer - rectal bleeding - about 60 percent of these cancers would go undetected.

The algorithm was effective at predicting the patients who were likely to develop colon cancer over the next two years,

The study found that the algorithm was very successful in spotting which patients would be most likely to develop bowel cancer over the following two years, and 70 per cent of all bowel cancer patients who were diagnosed had been in the top 10 per cent of patients described as being at highest risk.

Both studies used data from patients between the ages of 30 and 84, all of whom had no symptoms of the diseases in the previous 12 months.

These algorithms could be loaded on to GP computer systems to help reduce diagnosis time. The simple calculators could also become web-based programs for consumers, alerting them to seek medical attention for worrisome symptoms.

QResearch programs have already been shown to be effective in identifying patients at high risk of lung, stomach and esophageal cancers, heart disease, type 2 diabetes, bone fractures, kidney disease and blood clots

This research is published in the January, 2012 edition of the British Journal of General Practice.

Review Date: 
December 15, 2011