Dr. Imran Haque on “Crowdsourcing” Medicine Feedback Through Monitoring Search Engine Queries

Imran Haque, an internist and general practitioner based out of North Carolina, is looking forward to a future of scaling up research like Elad Yom-Tov’s because of the information it can provide doctors.

Over the past decade, access to information is the easiest it has ever been and people have more knowledge at their fingertips than ever before. Every day, people execute billions of searches worldwide on search engines and being privy to the search trends can provide a tremendous amount of information. One member of Microsoft Research, Elad Yom-Tov, created an algorithm that scans search results from Bing to learn about the efficacy of certain drugs and whether or not they may be recalled in the near-future. Elad is an example of a pioneer who is leveraging search engine queries to draw conclusions for the future – and being able to apply this practice to queries on symptoms or diets could help give doctors and researchers more insight on the effectiveness of medicine.

Dr. Imran Haque, an internist and general practitioner based out of North Carolina, is looking forward to a future of scaling up research like Elad Yom-Tov’s because of the information it can provide doctors. Specifically, he looks forward to a future where the efficacy of medicines is much more transparent so he can provide patients with the best treatment available. Dr. Imran Haque will be using his extensive experience of over 15 years to help explain why a giving doctors better access to information can lead to much more precise and effective medicine.

The Many Medicines of Doctors

As a doctor, Dr. Imran Haque has hundreds of types of medication available for his patients, and there is little time to comb through his inventory to determine whether or not he has the best medicine for every illness he may encounter – and he is not alone. Doctors can have extremely long hours depending on their role and keeping up to date with new drugs, new generic versions of drugs, and old drugs being phased out becomes more and more difficult each year. Having transparency and immediate information on the efficacy of drugs is essential – especially if the drug may be recalled in the future. Elad Yom-Tov’s research is a first step towards aggregating information on drugs – his research can provide crucial feedback on ineffective drugs and how certain patients may react to them.

Research on Search Engine Query Trends

Elad Yom-Tov’s created an algorithm that can predict when drugs will be recalled based on user searches on Bing. He was able to train his algorithm via machine learning by feeding in millions of drug-related queries over the period of 240 days in order to find trends related to drug searches. Within the data set were about 300 drug names that were searched over 1,000 times each. Over time, it’s important to realize that the algorithm will become more accurate as it processes more search results – which means that Elad Yom-Tov’s algorithm will only become better as it processes more data.

Elad Yom-Tov also conducted a second study where he harnessed search engines in order to identify side effects in drugs that were previously unknown. In this study, he aggregated search data of large populations to extract information on drug performance and potentially adverse reactions. He analyzed the data to find correlations over a time period and found some previously unknown side effects. In particular, he found a number of adverse side effects from combinations of drugs and less serious long-term effects that were never captured in tests before the drug was released. He stated that monitoring search engines is a better way of capturing less acute later-onset reactions whereas tests today are best for determining acute early-onsite side effects. lad’s tests are essentially crowd-sourcing feedback on drugs in order to get information, and Dr. Imran Haque believes that making this feedback available to doctors when prescribing drugs can help them make better decisions for their patients.

More Information leads to more Effective Treatment

The research conducted by Elad provides essential information on the performance and safety of drugs – and hits towards a future of being able to monitor searches in real-time to provide continuous feedback on medicine as well. Dr. Imran Haque would like a future where the efficacy of drugs and side effects are monitored over time, and to be able to have the information broken down based on certain categories or criteria. Knowing that some patients may respond better to a secondary choice of medicine instead of what the doctor normally prescribes for an illness is crucial to a future with more effective and more efficient treatment. In addition, by monitoring user searches, it also becomes possible for pharmaceutical companies to get feedback on how their released drugs are functioning. More importantly, monitoring search terms may also give insight on potential drug combinations that work well together.

The Future of Monitoring Medicine

Dr. Imran Haque is very excited about monitoring search terms to pull additional information to determine whether a specific drug is effective or if there are previously unknown side effects. A lot of the difficulty today is that there are a huge number of drugs available but receiving feedback on efficacy of drugs for a large sample size is difficult, which means drug knowledge is relatively opaque today. Giving clarity on how certain people react to drugs and side effects on certain drug combinations gives doctors more power in providing effective and safe care to patients. In addition, monitoring search terms may give rise and insight to certain drug combinations that have unexpected, beneficial side effects that physicians could leverage as well. Most importantly, perhaps, is that monitoring search terms can also give physicians insight on people’s most common health concerns and give doctors the opportunity to address health concerns that may be regional.

 

 

Categories
HealthInternal Medicine

RELATED BY