By Chiraag Kala


Introduction/Background (why we should care):

Diabetes is one of the biggest epidemics we're currently facing with nearly 26 million diabetics in the country and where the trends show that "at least one in five girls and one in four boys born in the year 2000 will be diagnosed with diabetes in their lifetimes" (Mehmet). In fact, "Type 2 diabetics are now up to four times as likely to die from a heart disease, and it is estimated that women diagnosed by the time they're 40 will lose around 14 years, on average, from their lives, while men will lose almost 12" (Mehmet).


In the US, one of our major health problems we face is that we spend more on treatments than on disease prevention, unlike other leading developed countries. In the case of diabetes, many young individuals who currently have prediabetes are not diagnosed and have no idea about it. According to recent stats approximately 35% of the the adults (20 or older) who have prediabetes have not been diagnosed. The issue here is that there is no way for doctors to monitor the general health of their patients, and even though there are health tracking devices - there are currently no technologies that the hospitals/clinics are using to use that data from these devices and predictive algorithms to give doctors a better idea of how "healthy" patients really are and when they can be at the risk of diseases such as prediabetes.

Additional stats about the current state of Diabetes in the US.


To be able to use data from a health tracking device to monitor the general health of a patient over-time and prevent diabetes and diagnose a case of pre-diabetes early on so that the person can take steps to prevent diabetes. The solution is to give dietitians/doctors access to data about patients not just during the time of a check-up but data over-time, which the dietitians/doctors can then use with predictive technologies to better understand the patient's history and give better recommendations.

The data used for this can include fitness data (calories burned everyday, activity level, heart rate, number of steps, etc) to lifestyle data (sleep + diet).

external image 11092-3764-DSC00401-l.jpg


With the use of predictive algorithms and regular health data from patients' trackers, the doctors will be able to better predict when patients are at a risk of diseases such as diabetes or detect if they are currently prediabetic.

Furthermore, studies have indicated that that tracking -- and the awareness that comes with it -- works. One study found that people who wore pedometers naturally increased their activity by 27%. (And they lowered their blood pressure and weight as well). Other studies have found that people with diabetes who used apps to record food, exercise, and other behavior had better long-term blood sugar control.

external image BN-HN556_istock_G_20150322144822.jpg external image Fotolia_69622291_S-725x543.jpg


To divide a similar group of people with very similar lifestyles and the same health tracking devices into two groups - one treatment and one control group. In order to maintain the credibility of the RCT, it must be crucial to make sure that the two groups are randomly assigned and are similar in most ways other than the fact that one will be involved the treatment and the other group won't be.

The treatment group's health tracking data will be obtained every year for 5 years and be given to the doctor with predictive analytics about health trends and general lifestyle choices they make, while the control group's data will not be given to the doctor.

The experiment results will then look if the treatment group could detect diseases such as prediabetes more easily than the control group and whether the health tracker data helped the doctors recommend lifestyle changes to individuals or not.


The following things could be measured for both the control and treatment groups:
  • number of those who detected prediabetes / total number of individuals who were at the risk of prediabetes or had prediabetes.
  • total number of preventions from diabetes
  • ratings from patients about how helpful they feel their doctors are (same doctor between control and treatment group).

This could give an indication of how useful/effective collecting and analyzing the data from health tracking devices for doctors is in monitoring patients at the risk of diseases such as prediabetes.

external image Infographics-DPP-rev2013-1in3-rgb300ppi.jpg

End Result:

The end result of this study would be to give doctors/dietitians more information about their patients which is recorded everyday (a snapshot of the lifestyle choices patients make everyday) for doctors to be able to treat their patients more effectively. This would also allow faster/easier detection of diseases such as prediabetes/diabetes and allow patients to be more aware of what's going on.

external image Healthy_Life_badge.jpg