Mini-Lecture
I. Modules for Information Systems
Keep in mind that information systems do not start with re-engineering, but the information flows are inherent to the organization. The best approach to reveal the information flow in an organization is to start modeling the organization. A model is a symbolic or abstract representation of reality and results in visualizing structures. Three types of models for information systems exist:
Data models: are the blueprints for information systems. Data models depict data and the relationships, enforcing organizational/unit procedures and policies or interrelationships between data.
Activity models: record the activities of interest; involve the functional decomposition of organizational/unit processes from the highest level to the lowest (elementary processes), as well as the specification of process dependency events, to refine the decomposition of the processes.
Interaction models: define how organizational/unit activities/events affect things of interest to the organization/unit (data).
The concepts of data models and Data Flow Diagrams (DFDs), activity models, interaction models, and decomposition diagrams are the pre-cursors to database design.
Examples of models can be found at: http://www.umsl.edu/~sauter/analysis/dfd/dfd.htm (Links to an external site.)
The construction of models and more specific of DFDs are explained at: http://en.wikipedia.org/wiki/Unified_Modeling_Language (Links to an external site.)
II. Understanding Databases
A. Goal of the Database
The goal of a database is to generate a set of schemas that allow us to:
Store information without unnecessary redundancy.
Retrieve information easily (and accurately)
A database is a structured electronic collection of related data elements similar to paper databases such as an address book, a telephone book, etc.
B. Anatomy of a Database
Understanding databases starts with understanding their composition. Following is an example of a table of a relational database.
Data Fields
last name
first name
middle initial
DOB
Nr, street
City
zip
Country
record1
Adams
Beth
F
12/31/1967
2 East St.
London
34670
UK
record2
Stevens
Jeff
S
2/9/1989
6 West Ct.
Orlando
52215
US
record3
Brown
George
F
7/7/1977
67 Ridge Ln.
Vancouver
6745
CA
Table is the collection of all structured records. This structure makes it possible to manipulate or select records or fields based on specific values in the fields.
Fields are the columns in a table. They represent the characteristics, i.e. attributes of interest. A field is identified by its fieldname. Field types are e.g. text/alphanumeric, numeric, memo, date, look-up, and auto-number.
Records or entities are the rows of the table. They represent the values of the data pertaining to an entity.
Queries can be performed from linked tables in a database. Example: select record for city=”Vancouver”. The query will return the values of record3.
More information on databases is provided by Windows Enterprise Database Services (http://www.liberty.edu/media/1414/%5B6330%5DERDDataModeling.pdf (Links to an external site.) ). Be familiar with the terms of the databases.
C. Types of Databases and Database Management Systems
Data from an electronic database are managed, stored, and retrieved by a software packet called a database management system (DBMS). Please watch a YouTube about Database Management and SQL:
What is Database & SQL? (Links to an external site.)
What is Database & SQL?
1) Hierarchical databases
This type of database is conceptualized as an upside-down tree in which the data elements are organized hierarchically. Relationships in such a system are thought of in terms of children and parents such that a child may only have one parent but a parent can have multiple children. Parents and children are tied together by links called “pointers” assuring that data can be accessed systematically. Hierarchical databases were developed for the older mainframes. Although redundancy generally is avoided for one-to-many relationships (e.g. a parent has one or many children), redundancy in hierarchical databases will occur for many-to-many relationships as well. This is because a child may only have one parent. Please see more information from the following link: https://en.wikipedia.org/wiki/Hierarchical_database_model (Links to an external site.)
2) Relational Databases
Relational databases were developed in the 1980s from the work done by E.F. Codd at IBM. The model was built on principles of the Boolean Algebra. At the core of the relational model is the concept of a table in which all data are stored. Each table is made up of records (horizontal rows also known as tuples) and fields (vertical columns also known as attributes). Tables in a relational database are linked through their relationships. Data in a relational database are accessed on the values of the data (i.e. the records). In addition, relational systems can use operations recursively in order to provide information about the database which facilitates their management.
Relational databases are very popular because of their conceptual simplicity and query languages. For more complex and very large databases this simplicity poses a problem in implementing database operations and specifying the semantics of the data. An Object Oriented Database has been designed to provide a solution to the issue. Related records are grouped in an object, and associated with the object are methods for entering and maintaining fields in the object (e.g., a vehicle is an object, a car, a bus, a truck are classes in that object. They all inherit attributes from the object vehicle such as registration required, number of wheels, cylinder capacity, etc. but have also specific attributes that separate them from each other. Please see more information from the following link: https://en.wikipedia.org/wiki/Relational_model (Links to an external site.)
Also, for training on databases and lots of other software, go to https://www.lynda.com/ (Links to an external site.), (Links to an external site.) you may need to set up an account. please see the direction, https://servicecenter.twu.edu/TDClient/KB/ArticleDet?ID=11562 (Links to an external site.) and search âdatabases.â This training is free to all TWU students.
III. Health Care Databases
A. Clinical Database
A clinical database contains well-defined, discrete, and continuous data elements concerning patients and is routinely coupled with health care outcome descriiptors. Compared to administrative databases, clinical databases contain accurate diagnostic information, clinical risk factors, health status, and process of care. Administrative data may answer certain questions such as differences in health service utilization across health markets; while clinical data supply detailed longitudinal information about comorbidity, severity of diseases, and complications which allows for accurate severity adjustment of care. The advantages of clinical data over administrative data for evaluating utilization, quality, and costs are shown in the following table.
Clinical Data
Administrative Data
Covers the entire enrolled population
Covers only users of health care
More accurate for severity adjustment
Limited for severity adjustment
More accurate clinical diagnoses
Diagnoses subject to coding errors
Allow assessment of the process of care
content of care processes limited
Clinical information systems are used for studying:
Efficiency of care
Monitoring practice changes
Evaluating compliance with screening recommendations
Outcome assessment
Epidemiological surveillance
Development of prompts and reminders
AHRQ compared the advantages and disadvantages of different types of data for health care quality measures. Please follow the link for the information: https://www.ahrq.gov/talkingquality/measures/understand/index.html (Links to an external site.)
B. Public Databases
Public health databases have been built and released to the public as there has been an increasing need for timely information about the health and health economics of the population, communities, states, and countries. The data are used to identify, assess, and monitor health problems and priories and evaluate appropriateness and cost-effectiveness of health care. The public data is also used to formulate public policies.
Public Health Informatics is the application of informatics in areas of public health, including surveillance, prevention, preparedness, and health promotion (American Medical Informatics Association, 2021). Public health informatics and the related population informatics, work on information and technology issues from the perspective of groups of individuals. Dr. John Snow has been designated the father of public health informatics. Florence Nightingale should also be recognized as an early public health informaticist. Her recommendations about medical reform and the need for improved sanitary conditions were based on the data about morbidity and mortality that she complies from her experiences in Crimea and England. Her efforts led to a total reorganization of how and what healthcare statistics should be collected (Dossey, 2000; McGonigle & Mastrian, 2012). For more information on public health informatics, please review the following link: https://www.phii.org/defining-public-health-informatics (Links to an external site.)
A lot of statistical information with the aggregated data and user-friendly query system for accessing individual data and aggregated statistics are available at public health databases. You might be able to the following data manipulation:
Sorting
Querying
Simple query
Boolean query
Calculated query
Generating reports, lists
Generating forms (mostly for online data entry)
The following link gives you examples of public databases: A collaboration of U.S. government agencies, public health organizations, and health sciences libraries, https://phpartners.org/health_stats.html (Links to an external site.)
IV. References
American Medical Informatics Association. (2021). Public Health Informatics. https://www.amia.org/applications-informatics/public-health-informatics (Links to an external site.)
Dossey, B. M. (2000). Florence Nightingale: Mystic, visionary, healer. Springhouse, PA: Springhouse Corporation.
McGonigle, D., & Mastrian, K. G. (2012) Nursing Informatics and the Foundation of Knowledge. (2nd ed.). (pp.530). Burlington, MA: Jones & Bartlett Learning.
Assignment
The goal of this assignment is to locate and use public databases that can be useful for decision-making for practice, management, and research. Each student is supposed to individually complete and submit the assignment. Note: Some databases are easier to use than others.
By 7/10 Sunday midnight (11:59 pm), answer the following questions using publicly available databases:
What are the rates of adults ages 18-64 at high risk (e.g., COPD) who received an influenza vaccination in the last flu season by the insurance status (insured vs. uninsured) in 2019 in Texas comparing its national rate? Provide a comparison table (Texas vs. National by insurance status) and a brief interpretation. (Use the Data Query system of the National Healthcare Quality and Disparities Reports: https://nhqrnet.ahrq.gov/inhqrdr/data/query (Links to an external site.)) (up to 2 points)
A) Which county in North Texas has the smallest rate of non-physician primary care providers in 2021? B) In that county, what is the most problematic ‘health/cancer’ issue compared with the values of overall Texas and the United States 2015-2019? C) Do you think it is possible to make a policy proposal that XXX County (you found) needs to increase the number of non-physician primary care providers to solve the most problematic health/cancer issue found in that county? D) What data should be additionally collected to strengthen your proposed workforce policy to increase non-physician primary care providers? (Go to http://www.healthyntexas.org/index.php?module=indicators&controller=index&action=index (Links to an external site.) You would need to start searching by Indicator. Then, for further information on the topics in the selected county, explore Search by Location and select ‘Health/Cancer’ topic) (up to 2 points)
What are the top 10 leading causes of deaths in the United States in the year 2010 and the year 2020? Obtain and submit the charts. Describe any difference in the causes of deaths between the two years. (Use the CDCâs WISQARS⢠(Web-based Injury Statistics Query and Reporting System: https://webappa.cdc.gov/sasweb/ncipc/leadcause.html (Links to an external site.)) (up to 2 points)
Explore the Texas Health Data query system (http://healthdata.dshs.texas.gov/Home (Links to an external site.)). Go to âHospitalsâ tab and select âPatient Safety Indicators (PSI)â in the drop-down list of âQuality Indicatorsâ. What are the top 5 hospitals reporting high death (observed) rate among surgical inpatient with serious treatable conditions in 2019? (up to 2 points) (If you want to know the conditions, please refer to the following link: https://www.dshs.texas.gov/thcic/publications/hospitals/Patient-Safety-Indicators/Patient-Safety-Indicators-With-Diagnosis-Present-on-Admission-(POA)-2016/#NewBookmark (Links to an external site.)
What do THSA and HIETexas stand for in Texas, respectively? What are their purposes? (up to 2 points)
Extra points (0.9 point) Briefly state your opinions on which public database querying interface is better, any recommendations to any of these public databases, and lessons or challenges of this public database exercise.
Please submit your answers in MS Word along with attachments of graphs or charts. Label your assignment files with “PublicDB_your username” as the file name (0.1 point).