SELECT admission_type
FROM admissions
WHERE admission_type IS NOT NULL Analysis Report One - What’s Your Data Strategy
Executive Summary
This weeks report takes a deeper dive into the roles of both offensive and defensive data strategies, specifically within healthcare organizations. By using information for “What’s Your Data Strategies?” (DalleMule and Davenport (2017)) and the MIMIC-III database the report will analyze why and how these healthcare organizations balance the protection of information that is sensitive, while using the data in a strategic way the will improve the overall performance and production of the operations being performed.
The difficulty yet necessary steps to take to ensure that defensive data is balanced, while also making sure that the offensive data is being used appropriately to improve outcomes and efficiency are crucial as we read in the article “The Rise of the Healthcare CIO” (Butcher (2021)) The ability to understand and utilize defensive data, which prioritizes minimizing risk and protection the sensitive information of patients, while also managing the offensive data, which is used to help these organizations look at things more in line with improving business is quintessential and should be a priority within all healthcare organizations.
Introduction
In “What’s Your Data Strategy” we are able to learn how the priority for healthcare organizations should be figuring out how to manage the data they have in a way to help improve and better business strategies, rather than that data being viewed as just a piece of the pie in reference to daily operations. It is very common to see these organizations not using the data to its fullest potential and often struggle to using and embracing the power it brings to them. The author of this article dives specifically into both defensive and offensive data in an effort to help us readers truly understand and embrace the necessity of understanding both. The defensive strategies are used to protect sensitive information. It is preventing things such as data breaches while also improving upon the security of the data and IT infrastructure (DalleMule and Davenport (2017)). Offensive strategies are catered toward improving the business aspect of things such as generating revenue and products while also being able to adapt with the market changes and understand overall customer analytics (DalleMule and Davenport (2017)). Being able to use both defensive and offensive data will help organizations allocate resources better while maintaining a healthy balance. Overall, the article explained to us how valuable data truly is to these companies. If the organizations took the time and effort to understand the data and use it to their advantage instead of pushing it away because they are confused by it, they would ultimately be put in better positions from it. Understanding the defensive and offensive data that is collected daily and in large amounts will help these organizations learn to utilize it in a way that will lead to an overarching improvement in business and operations.
The Healthcare Context
The healthcare industry is very reliant on data to not only improve in operation and performance, but also so the patient outcomes and experiences can improve. This is exactly why “What’s Your Data Strategy” is relevant to healthcare and the organizations that work in it. Offensive and defensive data can contribute a lot to increasing operational improvements, however it contains very sensitive information regarding patients thta must be kept protected at all times. The information collected through things like health records monitoring technology, and billing systems is so vast and useful, yet so sensitive and precious that it puts healthcare organizations in a bind on how to properly use it all.
The defensive side of data strategy is very important in healthcare specially. These organizations must comply with HIPAA and take extra steps to ensure their data is protected so the patients records can remain confidential. Poor data security and quality and data breaches can lead to a variety of punishments like financial penalties or legal trouble. With the increasing amount of data breaches in the healthcare industry specifically hacking and It incidents which was was reported as the most significant cause of attacks from 2010-2014 (Sun, Wu, and Ye (2026)), it is essential that defensive data is collected, understood, and utilized to the best of its ability to protect the organizations in the healthcare industry along with their patients.
While defensive data strategies are important, healthcare organizations also must use offensive data strategies. Hospitals often times use this data to improve the patient care and experience at said places. This data can also be used to identify gaps in services that will ultimately improve the care that is delivered before a patient is discharged. This is a prime xample as to how understanding offensive data can create value for both the healthcare organizations and the people utilizing them (Landro (2023)). These offensive strategies can lead to better staffing decisions, patient outcomes, hospital performance, and much more. This kind of data being not only protected, but used to create more value and better overall production is how healthcare organizations should be striving to use it.
The MIMIC-III database shows how balancing the offensive and defensive data is done and some effects of it. The information it has contributes in ways that support management decisions that also consider things such as background and administrative information that needs to be protected and manged carefully. This shows exactly why healthcare is a prime example of why balanced data strategies are essential to a successful organization.
Data Visualizations
Visualization One - Offensive
ggplot(data = myquery1, aes(x = admission_type)) +
geom_bar() +
theme_minimal() +
labs(
title = "Distribution of Admission Types",
subtitle = "Admission categories within MIMIC-III",
x = "Admission Types",
y = "Number of Admissions",
caption = "Source: MIMIC-III Clinical Database v1.4"
)
This visual is a graph depicting the different distributions of admission types in the MIMIC-III database. I believe this is offensive data as it is used in order to help the operational decision making and production of said healthcare organization. As we can see, the graph tells us that emergency admissions are the majority of patient admissions. Elective and urgent admissions are seen a lot less often. This information can help healthcare administrators in many ways. It can be used to improve staffing decisions like making sure more employees are allocated to the emergency admission part of the hosptial. Also it can help with resource allocation by making sure the emergency admissions have more resources allocated to them since they will be seeing many more patients and admissions in general compared to elective and urgent.
Visualization Two - Defensive
SELECT insurance, ethnicity, religion, marital_status
FROM admissions
WHERE insurance is NOT NULLggplot(data = myquery2, aes(x = insurance)) +
geom_bar() +
labs(
title = "Distribution of Insurance Categories",
subtitle = "Sensitive administrative data",
x = "Insurance Type",
y = "Number of Admissions",
caption = "Source: MIMIC-III Clinical Database v1.4"
)
This visual shows a graph hath tells you the different distribution levels of each insurance category within the admission data we were given. Due to the sensitive information that is contained in this both financially and personally I classified this as defensive data. As we can see fro mthe graph, Medicare is the biggest insurance category in this example. Following that would be private insurnace and Medicaid. This kind of information is crucial for healthcare organizations. It helps them in regards to properly reimbursing patients and reporting what treatment the patients had received. However, this information is very sensitive and must be kept private at all costs in order to maintain in compliance with HIPAA and other regulations in place. ## Recommendations for Industry After extensive reading of articles and databases it is clear that it is in the best interest of healthcare organizations to develop balanced data strategies. Supporting both offensive and defensive data will help strengthen data protection while making sure sensitive patient information is protected, while also being able to increase the overall production of operations for the organization and the patients.
Also healthcare organizations should used that increase in operational and offensive data to help make operational decisions. In the visual example we see how many more patients are coming in as emergency patients. If these organizations can use that information and allocate staffing, money, resources, etc accordingly it will not only produce a more efficient product, but a better experience and outcome for its patients.
Healthcare organizations in general just need to accept data as more than just every day number. If they are able to invest in ways that they can leverage the data that they receive it would open so many more possibilities. It would create better environments, work spaces, adn patient experiences. By aaccepting data and viewing it as an asset used to help improve the organization, all employees will be able to understand and take a balanced approach to maximizing it in order to help keep the industry growing while staying in compliance with regulations and all the sensitive informatiuon they must protect.
References
Butcher, Lola. 2021. “The Rise of the Healthcare CIO.” Physician Leadership Journal 8 (3): 20–22.
DalleMule, Leandro, and Thomas H Davenport. 2017. “What’s Your Data Strategy.” Harvard Business Review 95 (3): 112–21.
Landro, Laura. 2023. “How Hospitals Are Trying to Improve the Patient Experience.” The Wall Street Journal.
Sun, Mingyang, Yuxin Wu, and Rongtian Ye. 2026. “Research and Analysis of Healthcare Data Breach Risk Prediction in the US Based on Interpretable Machine Learning.” In ITM Web of Conferences, 84:01018. EDP Sciences.