Predictive Analysis in the health care industry

This article discusses the applications and benefits of predictive analysis in the healthcare industry. It explores how healthcare organizations can use predictive analytics to improve patient outcomes, reduce costs, and enhance operational efficiency. The article also highlights the importance of data infrastructure and provides insights on how to start using predictive analytics in the healthcare industry.

Running Head: PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
1
1
Predictive Analysis in the health care industry
Name
Professor
Course
Institution
Date

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
2
Predictive Analysis in the health care industry
The predictive analysis encompasses statistical techniques from machine learning,
predictive modeling, and data mining. It analyzes historical and current facts to predict future
unknown events. Healthcare organizations are developing sophisticated capabilities of big data
analytics [1]. They are moving from basic descriptive analytics to predictive insights realm.
Predictive analytics does not only present past information but it also estimates the future
outcome. Predictive analysis alerts clinicians, administrative staff and financial experts about the
future potential event. It allows them to make informed choices before uncertainties. It becomes
more useful when the knowledge is transferred into action. Predictions for making predictions
only are a waste of money and time. Willingness to intervene harnesses the power real-time and
historical data. To add value and efficacy in healthcare, there should be the integration of the
intervention and predictor.
How to start using predictive analytics in the health care industry
It is important first to establish fundamental analytic and data infrastructure. Afterward,
move the organization up the levels of the healthcare analytics adoption model [2]. It starts with
an enterprise data warehouse in combination with discovery and foundational analytic
applications.
Enterprise Data Warehouse
Healthcare organizations require enterprise data warehouse platform to manage patient
populations. It is the central platform for building a scalable analytics approach to make sense of
data and integrate it systematically. Catalysts in healthcare deploy a late-binding data warehouse
to automate aggregation, integration, and extraction of clinical, patient experience, financial and

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
3
administrative data [3]. They apply advanced analytics to measure and organize patient
satisfaction, safety, and outcomes, cost, and clinical processes.
Predictive modeling three basic steps
Source:https://downloads.healthcatalyst.com/wp-content/uploads/2013/10/predictive-
modeling.png
To be effective in predictive analysis, lean practitioners should understand the actual
workflow, type of data, the targeted audience and actions to be prompted after the prediction.
First, health care analysts should define the problem, gather initial data and evaluate the available
algorithm approaches [4]. Second, they select the best models and use a separate data set to
validate the process. The third step is to apply the model in a real-world setting.
Predictive analytics includes evidence, actions, and recommendations for the predicted
outcome. They should link to measurable events such as patient outcomes, clinical protocols, and
clinical priorities. There are many options to stratify patient risk and develop predictive

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
4
algorithms. Healthcare providers should partner with people with commercial tools and leading
academic knowledge to develop more appropriate prediction models.
When adopting predictive analytics healthcare organizations should not confuse more
insight with more data. They should not overestimate their ability in data interpretation. They
should not confuse value with insight [5]. They should not underestimate the implementation
challenges. Subsequent intervention and clinical should be clinician and content-driven. The
overall goal of predictive analysis is to improve patient outcomes by use of their historical data.
Predictive analysis use cases in healthcare
Predictive analysis in healthcare improves chronic disease management, supply chain
efficiencies, patient care, and hospital administration. Across the care continuum, predictive
analysis supports health management, better outcomes, and financial success. Healthcare
organizations use predictive analysis in the following ways:
Chronic diseases risk scoring
Healthcare organizations use predictive analysis to identify people at risk of having
chronic conditions early to help them avoid those health problems. They use lab tests, claim,
biometric data, health social determinants, and patient-generated health data to create risk scores.
The risk scores give healthcare providers insights to enhance wellness activities and services.
Management, stratification, and identification of patients at high risks improve cost and quality
outcomes.
Avoidance of hospital readmissions in 30 days

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
5
Predictive analysis deploys strategies in care coordination and improves care transitions.
It also warns providers on risk factors in patients and indicates the likelihood of readmission in
30 days. Medicare’s Hospital Readmissions Reduction Program (HRRP) subjects hospital
significant penalties to prevent patients’ unplanned returns in inpatient settings [6]. The
predictive analysis identifies traits in patients that produce a high likelihood of readmission. It
gives providers an indication to focus on follow-up and design discharge protocols to prevent
readmissions.
Prevent patient self-harm and suicide
The predictive analysis identifies individuals likely to harm them. They are given mental
healthcare to avoid suicide and other serious events. They use predictors such as substance abuse
diagnosis, depression questionnaires, previous suicide attempts and the use of psychiatric
medications [7]. A combination of electronic health record data and other tools identifies people
at risk of a suicide attempt.
Ensures data security
Predictive analysis plays a significant role in cyber security as the sophisticated attacks
increase. Analytics tools monitor patterns in data utilization, access, and sharing. They give
providers warnings when an intruder penetrates the network [8]. Predictive tools calculate real-
time scores for requests and transactions and respond depending on the event. The strategy
prevents software from affecting healthcare organizations data security.
Get ahead of deterioration of patients
In hospitals, there are potential threats facing patients’ wellbeing. They can develop
sepsis, sudden downturn or hard-to-treat infection due to their clinical conditions [9]. Healthcare

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
6
providers use predictive analysis to identify changes in patients’ vitals and react before
symptoms of upcoming deterioration manifest.
Predicting the utilization of patients of patients
The predictive analysis enables providers to know when their clinics will get busy. It
alerts care sites operating without permanent schedules such as agent care centers and emergency
departments. They then vary their number of staff members to account for patient flow
fluctuations. They ensure enough beds for patients whom they expect to admit [10]. They also
ensure increased patient flow does not increase their waiting times to raise their satisfaction.
Altering scheduling procedures also reduces nurses’ burdens.
Develop new therapies and precision medicine
Predictive analysis allows providers to drug discovery techniques and supplements
tradition medicine trials. They use simulation and modeling to predict clinical outcomes, support
effectiveness evidence, predict product safety, optimize dosing and evaluate potential risks. They
identify the subgroups of patients who need adjustments in their dosages [11]. Analysts do trials
using models. The systems can predict a patient’s response to treatments. It matches genetic
information with patient cohorts. It allows doctors to choose successful therapy hence improve
outcomes.
Stages of analytics maturity

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
7
Source:https://media.licdn.com/dms/image/C4E12AQF35rcPbqvO3A/article-
inline_image-shrink_1500_2232/0?
e=1564617600&v=beta&t=7LYHA_KWjXaCVkLvC_BxRjQJL41OIYWsEvViBDrnXGE
Predictive analytics makes use of historical data to predict the future. An example is the
use of records in the hospital with other weather forecasts or social media forecasts to improve
staffing levels.
Prescriptive analytics is an extension of predictive analytics. It uses the predictions to
include recommended actions. They employ data techniques such as machine learning and
simulation and cognitive systems. In the previous example, the prescriptive analysis will
recommend workflow adjustments and prioritization to reduce waiting time and treat all patients.
Cognitive analytics use systems such as deep learning, logic systems and machine
learning which lie under artificial intelligence technologies. They provide suggestions and

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
8
insights to human decisions or uses human-like analysis to automate decisions. They create a full
view of the state of health of the patient by considering scans and records.
The predictive analysis predicts the unknown
Predictive analysis is determined by the problem and the question one asks. Historical
questions are answered with high certainty. However, asking questions for future events the
certainty in the prediction decreases. For example, asking “Will I get diabetes” is higher to
predict than asking “what I ate today”.
https://rockhealth.com/wp-content/uploads/2015/06/Predictive-Analytics_Website_PNGs.003-
1200x721.png
Predictive analysis underlies traditional healthcare and medicine

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
9
https://rockhealth.com/wp-content/uploads/2015/06/Predictive-Analytics_Website_PNGs.004-
1200x721.png
Most of the traditional medicines use predictive analysis as a patient is not tested in labs.
Algorithm production and big data reignite the excitement and interest in predictive analysis.
New technologies collect continuous patient-reported data.
Current applications of predictive analytics in the healthcare industry
The rapid growth of predictive analysis is due to increased healthcare data. Apixio is
software that uses HCC profiler software to sort through patient charts found in electronic health
records. It predicts the risk of insurance plants such as the affordable care act. The software
analyzes data from medical billing and patient records to provide risk scores for the healthcare
centers plans [12]. Software is Lumiata which helps insurance providers and hospitals to improve
operational efficiency and reduce risk. Ayasdi software offers solutions to clinical variation
using predictive analysis. It uses learning algorithms to find patterns in EMR data. It clusters

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
10
patient procedures with similar issues to generate clinic pathways. They improve patient
outcomes and reduce costs and time.
Advantages of predictive analysis in the healthcare industry
It advances the patient’s care. Health records collect medical data such as medical
conditions, clinical data, lab tests, and diagnosis. It helps practitioners provide quality care. It
improves operational efficiency. Predictive analysis analyzes staff efficiency and examines
historical patient admission. It enables healthcare providers to cut down costs and provide
enhanced care. It reduces readmissions and medication errors as well as increased administrative
and financial performance. It helps providers to find a cure for diseases. It uncovers hidden
patterns, unknown correlations, and insights. The predictive analysis makes it possible to address
health issues before they extend. They assist in fraud detection. It uses advanced algorithms to
sift through reports to find mistakes fast.
Disadvantages
Predictive analysis compromises privacy in confidential medical records. They access
private records and take away individual privacy to benefit other people. Predictive analysis
tends to replace doctors. It lacks personal touch as currently, patients are using technology to
find out there health issues instead of visiting the licensed doctors.

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
11
References
[1] I. Duncan, “Mining health claims data for assessing patient risk,” in Data Mining:
Foundations and Intelligent Paradigms, ser. Intelligent Systems Reference Library, D. E. Holmes
and L. C. Jain, Eds. Springer Berlin Heidelberg, 2012, vol. 25, pp. 29–62
[2] J. Donze, D. Aujesky, D. Williams, and J. L. Schnipper, “Potentially Avoidable 30-Day
Hospital Readmissions in Medical Patients: Derivation and Validation of a Prediction Model,”
JAMA Internal Medicine, vol. 173, pp. 632–638, 2013.
[3] O. Hasan, D. O. Meltzer, S. A. Shaykevich, C. M. Bell, P. J. Kaboli, A. D. Auerbach, T. B.
Wetterneck, V. M. Arora, J. Zhang, and J. L. Schnipper, “Hospital readmission in general
medicine patients: A prediction model,” Journal of General Internal Medicine, vol. 25, pp. 211–
219, 2010.
[4] E. Coiera, Y. Wang, F. Magrabi, O. P. Concha, B. Gallego, and W. Runciman, “Predicting
the cumulative risk of death during hospitalization by modeling weekend, weekday and diurnal
mortality risks,” BMC Health Services Research, vol. 14, 2014
[5] R. B. Cumming, D. Knutson, B. A. Cameron, and B. Derrick, “A comparative analysis of
claims-based methods of health risk assessment for commercial populations,” Final Report to the
Society of Actuaries, 2012.
[6] Y. Zhao, A. S. Ash, R. P. Ellis, J. Z. Ayanian, G. C. Pope, B. Bowen, and L. Weyuker,
“Predicting pharmacy costs and other medical costs using diagnoses and drug claims,” Medical
Care, vol. 43, pp. 34–43, 2015.

PREDICTIVE ANALYAIS IN THE HEALTH CARE INDUSTRY
12
[7] D. Bertsimas, M. V. Bjarnadottir, M. A. Kane, J. C. Kryder, R. Pandey, S. Vempala,
and G. Wang, “Algorithmic prediction of health-care costs,” Operations Research, vol. 56,
pp. 1382–1392, 2018.
[8] K. Pietz, C. M. Ashton, M. McDonell, and N. P. Wray, “Predicting healthcare costs in a
population of veterans affairs beneficiaries using diagnosis-based risk adjustment and self-
reported health status,” Medical Care, vol. 42, pp. 1027–1035, 2014.
[9] C. A. Powers, C. M. Meyer, M. C. Roebuck, and B. Vaziri, “Predictive modeling of
total healthcare costs using pharmacy claims data - A comparison of alternative
econometric cost modeling techniques,” Medical Care, vol. 43, no. 11, pp. 1065–1072, 2015.
[10] Ali Serhan Koyuncugil and Nermin Ozgulbas “Financial early warning system model and
data mining application for risk detection”, Expert Systems with Applications, Vol. 39, pp.
6238–6253, 2012.
[11] H. G. Dove, I. Duncan, and A. Robb, “A prediction model for targeting low cost,
high-risk members of managed care organizations,” American Journal of Managed Care,
vol. 9, no. 5, pp. 381–389, 2013.
[12] B. Fireman, J. Bartlett, and J. Selby, “Can disease management reduce health care
costs by improving quality?” Health Affairs, vol. 23, no. 6, pp. 63–75, 2014.