Assignment Help
cov_psy1
W4A2_.docxAPPENDIX
Reporting and Analysis Tools
Reporting and descriptive analytic tools are widely available from vendors of all size and scale. The table presented here is a list of application software types designed to extract or retrieve and report data. The tools at least read data that has been stored in a data warehouse or datamart. The chart indicates possible data analytics generally available as part of these tools (check marked as indicated). Some tools can link directly to a process application.
Reporting and Analysis Tools
Important here is the use of appropriate tools for a task. Often tools will extract or report, however additional different tools have to be engaged to analyze the data beyond a simple summary level. These are often in an application suite where differing tools in the suite can be linked to work sequentially. However, can one of these tools enable one to obtain the data desired and turn it into information through appropriate analysis for the monitoring and planning situation posed? Will the tool enable one to apply the types of analytic techniques to be discussed in module five?
Not all of these tools will. Notice that only one application type is able to engage predictive and prescriptive analytics.
If the data is in a warehouse, these vendors supply extraction and reporting tools; however they may well suffer the limitations of process applications discussed earlier.
Reporting and Analysis Tools
Important as well is the fact that these more general component tool packages require a level of distinct competence in constructing reports and analysis out of application components. It may be beyond the scope of an organization’s ability to staff, for cost or other reasons. With consultants or special jobs then being required, project volume and depth is limited. The organization may have data it cannot use!
Reporting and Analysis Tools
The expertise question presented here specifically affects:
I. Moving beyond scheduled executable jobs (especially to
integrate components for more automated processing), as
these are not typically designed around ease of automation.
II. Handling the typically required manual maintenance and
intervention.
III. Ability to assemble predictive and prescriptive analytics from
components contained in some applications.
Reporting and Analysis Tools
Rather than use more general component tools, specific applications are available to accept, report on, and analyze data related to specific aspects of the business. For example, there are numerous budget planning and monitoring applications.
On the operations clinical side there are a number of applications aimed at tracking and aiding in directing the providing of services to inpatient healthcare users. Most of the applications available currently only perform descriptive analytics. Predictive and prescriptive analytic tools of this nature are just beginning to emerge.
Data Systems Architecture
Data architecture is the high level design and functioning of the IT system. As with data warehousing, architecture is in the purview of IT. It is the set of system governing rules and definitions for how data is used, managed, stored, etc. in specific applications and among and across all applications. Developing common data operations among applications for control and predictability is one primary function of the design. This point to point coordination of data flow is a key aspect of architecture effecting the BI/Analytics Consultant’s access to and use of data and should be understood at, at least, a high level. This understanding of how data flows is most important when it comes to situations that are Operational in nature due to the need for real-time immediate access to data and response back to operating systems.
Data Systems Architecture
An item of importance for the BI/Analytics Consultant is that as applications change and evolve conflicts with systems architecture may arise and knowingly (or unfortunately unknowingly) compromises or deviations from the architecture occur. An understanding of how things are supposed to look and work help when they might not match.
For more details on this topic see Data Systems Architecture and Enterprise Information System Data Architecture Guide, page 1 at http://www.sei.cmu.edu/reports/01tr018.pdf. The latter provides an important contextual example of the topic
Data Warehousing
While the deep details of data warehousing are generally the work of the IT department, it is imperative that the BI/analytics consultant understand the construction and operation of these system entities, which are the foundation for the "one truth." (See data warehouse, paragraphs 1 - 4, for a basic explanation.) See also the Enterprise Information System Data Architecture Guide, page 5 at http://www.sei.cmu.edu/reports/01tr018.pdf, for another viewpoint and explanation of data warehousing in a specific context
Data Warehousing
Presented here is a simple general diagram suitable for understanding data warehousing in the healthcare context and the work of a BI/analytics consultant.
Data Warehousing
Note that there is no representation of an integrated data source, as this is not yet widely the case in healthcare. Individual applications dominate. Every organization differs in the construction of these systems. One would do well to know more about the specifics of their organization.
Data Warehousing
In particular, it is in the ETL decisions made during set up about extraction, cleaning, translation, transforming, and mapping—from the process applications on the far left of the illustration through the Integration Layer—how everything is combined and mapped that effects one’s ability to access data desired. Based on the ETL work limitations will exist in the warehouse data.
Data Warehousing
The critical linchpin in healthcare is construction and maintenance of the Master Patient Record. Also in retrieving data, the data one wants in the format desired or required might not have made it through the ETL process.
Last and very important is the fact that data in a warehouse is retrospective/historical only. This can only be used for monitoring and reporting, not operational situations. This will be discussed more in discussing operational situations.
In sum: one must know of the limitations existing in the warehouse; knowing in advance would be better.
A Pivotal Role
Key tasks of the BI/analytics consultant are the achievement of "one truth," a focus on items and areas of significance, and the presentation of information in ways that promote understanding, all in service of enabling the organization to make better decisions.
The role will be a pivotal one in that the BI/analytics consultant along with the clinical informaticist(s) sits at the point of power for engaging the analytics needed just to be in business in the future. This module has developed a framework for thinking through the data needed to address a situation, finding it, retrieving it, and displaying the analytic output that results from it, so the organization can make better decisions.
Key Points to Remember
As you move on, please keep in mind the following points.
Business Intelligence (BI) analytics - The goal is to obtain the information that addresses the questions or issues at hand. We can refer to this as "one truth," and this one truth needs to be clearly conveyed at the point in time when it matters.
Historically, healthcare provider organizations' use of databases and reporting has been narrowly directed and concentrated, with concomitant issues, which are as follows:
The business intelligence/analytics work traditionally performed, and much of the work done today, is a decision-support function involving the framing and formatting of reports and relatively basic descriptive analytics, if any analytics at all.
This work in health care often is still performed within the silos of individual process applications.
Most of this work has been financially focused in nature.
This work is typically based on performing recurring, routine reporting and analysis.
Regardless of the approach, the output is typically of a simplistic, descriptive, analytic nature providing quantities, comparisons, percentiles, and maybe trend lines on graphs.
Most often, application report writers are insufficient to reveal the information needed for powerful decision-making.
The five key "Power Decision" attributes are as follows:
Replicable - done in the same way, over and over again
Adaptable to differing circumstances
