DATA ANALYTICS IN HEALTHCARE

 

A basic overview of healthcare data analysis and steps in the data analysis process to advance the discovery process as a natural consequence of patient care and ensure the quality of innovation, safety and value in healthcare. Consider various information systems, so far, at the hospital will likely have an electronic health record system, as well as specialized departmental systems for billing nutritional services in pharmacy diagnostics in laboratories, etc. Each of these systems is designed and provided for clinical, ie patient care, and therefore collect specific data about the patient.

However, none of these systems have a complete set of data for an individual patient or for a group of patients, e.g. B. for all patients admitted with a specific diagnosis that can be used for analysis and reporting. Individual patients and across groups of patients require the aggregation of data from many systems. In order to obtain a deep insight into what is happening with individual patients and between patient groups, data from many systems must be aggregated and statistical analyzes of this aggregated data must be carried out and in contrast to the various clinical systems, clinical data warehouse collects patient data in a single coordinated location and this location is used for reporting and for analysis. It does this through a process called Extraction Transformation Load (ETL), which pulls data from various clinical data systems, synchronizes data formats in a process called transformation, cleans up the data, and then imports the data into the data warehouse database.

The transformation process is especially important as data can be stored in various ways across systems, for example a laboratory system may use the letters M for a man's gender or F for a woman's gender or U for an unknown gender, the Radiology Information System could use a number; however, they must match the designations used in the clinical data store and that conversion process to match what is called transformation. Another important step is to ensure that all patient records from various systems are linked to each other, this usually requires a master patient index sometimes called a master person index to link the various identifiers of a patient across the systems. Understand the need for a centralized coordinated location for patient data that can be used for analysis and reporting, define the term analysis, and explore the different types.

What is analytical is not the same as statistics. Analysis can be used in different ways and with different meanings. BI intelligence is an application-oriented initiative. The term analysis refers to the discovery of significant patterns in the data and is one of the following steps in the data lifecycle of raw data collection Preparing the pattern information analysis to synthesize knowledge and action for value creation As shown in this diagram, the analysis encompasses the whole process of data collection, extraction transformation , Analysis interpretation and reporting, including statistical data analysis as one of the most recent steps, the most recent showing that analytics is used to relate to the methods, their implementation in tools and the results of using the tools as interpreted by the skilled person .

The analysis process is the synthesis of knowledge from IBM information in 2013 categorized analyzes into three types. The descriptive uses business intelligence and data mining to ask what happened; the predictor uses statistical models and forecasts to ask what might happen; the prescriptive use of optimization and simulation to ask what we should do; the fourth type of diagnostic analysis is added. define as a form of advanced analysis that examines data or content to answer the question of why it happened. Diagnostic tests are more valuable to the institution, but also more difficult to perform; Even more difficult and valuable are our predictive analytics.

Common statistics such as the number of laboratory tests, the average age of the patients, or the average length of hospital stay are used for patients with a specific diagnosis. The descriptive analysis is often presented as a pie chart, bar chart, or column chart. Tables are written narratives and diagnostic analytical analysis as a form of advanced analytics that examines data or content to answer the question of why this happened.

 Tools used for diagnostic analysis include drill-down techniques, data discovery, and correlations. Let's start with an example before turning to the formal definitions: "Lithium Michael Wu claims that the purpose of predictive analysis is not to tell you what will happen in the future, it cannot." In fact, no predictive analysis can only lead to a prediction of what this might happen in the future, since all predictive analysis is probabilistic in nature. This brings us to the highest level of analysis, prescriptive analysis. Data analysis inv performs a sequence of steps: 1. identify the problem, 2.identify what data is needed and where that data is located, 3. develop an analysis plan and recovery plan, 4.extract the data, 5 mark clean and prepare the data for analysis, 6. analyze and interpret the data, 7. Visualize data, 8. disseminate new knowledge and 9. implement knowledge in the organization.

Next, the data needed for the analysis should be identified, where the data elements are in which system or systems and which tables are in the database. Who is the contact person for each system who is responsible for retrieving the data? If there is a clinical if not, the required data items may be stored in different systems requiring multiple extraction steps.

A plan for retrieving data from the different systems, as well as the plan for verifying that all required data has been retrieved, must be developed. There Must Be Some To determine how many records are expected and then actually retrieved, this may involve cross-checking with other systems. This step requires the participation of the people who normally pull data from the systems involved. An analysis plan should be drawn up. Should a Statistician Be Consulted Questions to be addressed here include population size, sample size, and condition. The next step is extracting the data from the system or the systems involved, after the data has been retrieved, the data should be checked to verify that it is complete.

 Once the data set is complete, if all the data sets are retrieved Descriptive statistics such as the need to run the accounts, changes in the extraction plan and further extraction of the source systems may be necessary at this point. Once a full set of records has been pulled from the source system, errors in the records need to be identified and corrected, and any data. If there are errors such as transposed letters and names, as well as incorrect values, decisions about how to deal with the blank fields need to be made.

The following data also needs to be synchronized or transformed, for example, the patient's gender in one system in the hospital can be stored as M. while another system can use 1 to 91 sets of values must be changed for all records to use. same values after all necessary transformation steps have been completed The data, then imported into the target system where the actual data analysis and reporting will be performed, this can be as complex a system as the clinical data warehouse or as simple as a desktop computer.

 The data is now on the system where the analysis will be run and should be a complete set of data. You need to verify that everything is ready for analysis. Did you get what you needed? Please check and verify this with the analysis plan that was developed and that it has everything to address the problem you identified. You are now ready to do the actual analysis to run the analysis plan that was developed earlier. the statistical analysis and with the help of the statistician to confirm the interpretations and conclusions of the analysis, you now need to be able to communicate the results of the analysis and how the results have solved the problem since step one this communication must be very clear and fast understandable to the takers decision-making in the institution, so selecting an appropriate representation of the findings is essential to choose a visualization that is appropriate for the type of data, for example, categorical data can be represented with tables of column or bar graphs and tables dynamic Quantitative data can be displayed with histograms and a wide variety of other types of graphs, such as scatter plots and star plots.

Some common tools. Tableau and Microsoft Excel charting function once the interpretation analysis and visualizations are complete, a report needs to be developed, it could be a formal document written in email or presentation, regardless of the delivery method, The report should clearly state the original problem the process that was used to address the problem and then the results of the analysis along with the supporting visualization, this represents new knowledge and should be distributed to the stakeholders that were identified in step one. Finally, the new knowledge must be implemented to address the original problem, this will require stakeholder participation.