Data Mining: how it can be utilized in Ethiopia’s health sector

By Senayt Nur

Data mining: the future of the health care system

Collecting data that can be analyzed using patterns and trends dates back to the Bayes’ theorem and ‘regression analysis’ that were used in the 1700s and 1800s respectively. For decades, supercomputers have been used to collect data, analyze market research reports, determine patterns of customer preferences, product usage, and the general demand and supply patterns in order to increase revenue and decrease unnecessary costs.

In plain terms, “data mining is a process of analyzing data from different perspectives and summarizing it into useful information”. Although the technology have been around for a while, advancements in computer’s processing power, disk storage, and statistical software have dramatically increased the accuracy of ‘data analysis’ while driving down it’s cost.

Before digging any deeper into data mining, let me first explain the difference between data and information. Data is collection of any facts, numbers, or text that can be processed by a computer.

Today, organizations are accumulating vast and growing amounts of data in different formats and different databases. This includes: operational or transactional data such as, sales, cost, inventory, payroll, and accounting; Non Operational data, such as industry sales, forecast data, and macro-economic data and Metadata- data about the data itself, such as logical database design or data dictionary definitions”. (Palace)

On the other hand, information is the patterns, associations, or relationships among all this data can provide information.

How could data mining be used in health care system?

Data mining works in different ways. One of the most common ways is ‘mining by association’, which works by finding patterns where one event is linked to another. Another well known method is ‘sequential patterns’ that works by figuring out where one event can initiate another, and last but not least is ‘forecasting’ and it’s focused on establishing patterns that can help lead to predictions regarding the future.

Now let’s see the prospects of Data Mining in Ethiopia’s health sector. The idea behind using data mining in the health care system is to provide patients with accurate and reliable diagnosis, treatment, and general help which at the end of the day will lead to better health care for those who are in real need.

Data mining can introduce quite a few interesting prospects for the future; new treatments of diseases can be discovered. Since Data mining finds patterns in sets of data and then associations that could be interpreted into a meaningful outcome, by feeding the raw data [of a certain disease pathology] to computers, diagnosis would be more accurate and reliable.

For example, when a patient comes with a complaint, the algorithm skims through a huge load of raw data (data about patient, disease diagnosis, electronic patient records, medical devices that have been recorded through time, hospitals’ resources) in the database and comes up with the best possible diagnosis, the most effective treatment regimen with the least amount of expenditure.

It can also be used to measure ‘treatment effectiveness’.

This application of data mining involves comparing and contrasting symptoms, causes and courses of treatments to find the most effective course of action for a certain illness or condition. For example, patient groups who are treated with different drug regimens can be compared to determine which treatment plans work best and save the most money. Furthermore, the continued use of this application could help standardize a method of treatment for specific diseases, thus making the diagnosis and treatment process quicker and easier”. (Vataloro)

The other important aspect of data mining in healthcare is ‘detecting fraud and abuse’. “This involves establishing normal patterns, then identifying unusual patterns of medical claims by clinics, physicians, labs, or others. This application can also be used to identify inappropriate referrals or prescriptions and insurance fraud and fraudulent medical claims” (Vataloro). This would help save a lot of money that is being put into corrupt health workers’ pockets and help patients save unnecessary expenditures.

When we come to Ethiopia, most of the hospitals and health offices use manual data collection and data storage. By converting this into a digital health informatics system, the health care system could benefit a lot. For starters, when a patient comes into the hospital instead of searching for the patient info among thousands of other files in an old dusty storage room– which is a massive waste of time not only for the patient, but as well the certain health care center’s too– every history that the patient has ever had can be found in a matter of seconds. This search is going to include the previous admissions in any health center in the country or even across the globe. This will also make sure important past medical histories of patients are not lost.

Having accurate data on previous admissions, treatments, and surgical procedures that the patient had is going to help the current treating physician on the course of management he/she is going to follow. For example, let’s assume a patient who have been treated for pneumonia a week back with the first line drug and who isn’t responding to the treatment is now back with similar symptoms. Having this patient’s treatment history will save the attending physician from repeating the same antibiotic and will give him an idea to give a stronger drug this time.

This doesn’t stop at the entry; by the time the patient enters the primary physician’s office, the doctor has every necessary information on his/her hands including essential referral papers to have a basic background of the patient. Instead of subjecting the patient to unnecessary cost expenditure and exposure to radiation taking one test after another trying to figure out the problem, by the time the physician gets a complaint from the patient and has possible differential diagnoses, the physician can reach to a few possible alternatives to the problem just by utilizing the benefits of data mining.

This will not only save the patient a lot of money and unnecessary exposures, this will also help the physicians reach to a diagnosis faster. This means, physicians will be able to save a lot of lives with better accuracy. We cannot take this lightly in a country with scarce medical resources. If we can spare even a few supplies and equipments by utilizing them strategically, those who are in real need will get the necessary tests which will end up saving more lives at the end.

Despite how beneficial data mining is to various organizations, privacy concerns can result in lots of problems. Individuals sharing their data must be aware of how the data is being used (for example, the purpose of its collection, how it will be used and why). This actually becomes a real concern in the health care system where patient’s private information might be accessed by the wrong people.

Hence, privacy may be a setback for the future of data mining. But at the end of the day, the goal of health care is not to protect privacy; it’s main objective is to save lives. It is like what Thomas Graf, chief medical officer at Geisinger Health System, told The Washington Post, “It’s not an irrational fear. At the same time, people die driving every year and we still choose to drive cars, or most of us do. It’s a risk every person has to decide where they fall on the line.”



Herland, Matthew, et al. “A Review of Data Mining Using Big Data in Health Informatics.” Journal Of Big Data 1.1 (2014): 2. Web. 13 December.2016.
Desikan, Prasanna, and Kuo-Wei Hsu. “Data Mining For Healthcare Management.” 2011 Siam International Conference on Data Mining. Ed. Jaideep Srivastava. Arizona, USA: n.p., 28 Apr. 2011. 3–35. Print.
Koh, Hian Chye, and Gerald Tan. “Data Mining Applications in Healthcare.” Journal of healthcare information management 19.2 (2005): 64–72. Web. 28 Oct 2016.
McFarland, Matt. “The Incredible Potential and Dangers of Data Mining Health Records.” Washington Post. washingtonpost, 1 Oct. 2014. Web. 15 Dec.2016.
Palace, Bill. Data mining: What is data mining? 1998. Web. 18 Nov.2016.
Sahle, Geletaw. “Ethiopic Maternal Care Data Mining: Discovering the Factors That Affect Postnatal Care Visit in Ethiopia.” 4. (2016): n.pag. Web. 28 Oct.2016.
USF Health. “Improve Healthcare Efficiency and Quality with Data Mining.” University of South Florida. n.d. Web. 13 Dec. 2016.
Vataloro, Ron. “Improve healthcare efficiency and quality with data mining.” Big Data Made Simple – One source. Many perspectives., 29 July 2016. Web. 13 Feb. 2017.


Social tagging: > >

Leave a Reply

%d bloggers like this: