Health Data Analytics PDF

# SECTION 6 ## Health Data Analytics Christy L. Beaudin, Pradeep S.B. Podila, and Deborah J. Bulger ## SECTION CONTENTS ### Introduction * 310 ### Health and Data Informatics * 310 ### Data Analytics * 311 ### Data for Improvement * 318 * Step 1. Planning and Organizing for Data Collection,...

# SECTION 6 ## Health Data Analytics Christy L. Beaudin, Pradeep S.B. Podila, and Deborah J. Bulger ## SECTION CONTENTS ### Introduction * 310 ### Health and Data Informatics * 310 ### Data Analytics * 311 ### Data for Improvement * 318 * Step 1. Planning and Organizing for Data Collection, Interpretation, and Use * 319 * Step 2. Verifying and Correcting Data * 319 * Step 3. Identifying and Presenting Findings * 319 * Step 4. Ongoing Study and Recommendations for Change * 321 * Step 5. Taking Action * 321 * Step 6. Monitoring Performance * 322 * Step 7. Communicating Results * 322 ### Information Systems * 322 * Administrative Support Information System * 324 * Management Information System * 324 * Clinical Information System * 324 * Decision Support System * 324 * Distributed Health Data Network * 325 ### Design and Management * 327 * Data Fundamentals * 327 * Healthcare Data * 327 * Common Types of Healthcare Data * 327 * Patient-Generated Health Data * 328 * Quality Measurement * 331 * Data Specifications * 336 * Code Specifications * 336 * Data Protocol * 338 * Measure Documentation * 339 * Measure Calculations * 339 * Measurement Concepts * 342 * Reliability * 342 * Validity * 343 * Data Types * 343 * Nominal * 344 * Ordinal * 344 * Interval * 345 * Ratio * 345 * Sampling Design * 345 * Probability Sampling * 346 * Nonprobability Sampling * 346 * Sample Size * 346 ### Statistical Techniques * 347 * Comparison Groups * 347 * Statistical Techniques * 347 * Statistical Power * 347 * Measures of Central Tendency * 347 * Mean * 347 * Median * 348 * Mode * 348 * Measures of Variability * 349 * Range * 349 * Interquartile Range * 349 * Standard Deviation * 349 * Statistical Tests * 350 * Parametric Tests * 350 * Nonparametric Tests * 352 * Chi-Square Tests and Categorical (Attributes) Data * 352 * Multiple Regression Analysis * 352 * Tests of Significance * 352 * Data-Informed Decision Making * 353 * Displaying Data * 353 * Frequency Distribution * 354 * Histogram * 354 * Pareto Diagram/Pareto Chart * 355 * Pie Chart * 355 * Run Chart * 356 * Scatter Diagram/Scatter Plot * 356 * Stratification Chart * 356 * Statistical Process Control * 356 * Reporting for Improvement * 358 ### Information Systems * 322 ### Distributed Health Data Network * 325 ### Data Governance Protocols and Procedures such as * Business Associate Agreements, Data Use Agreements and Data Sharing agreements should be in place before sharing any sensitive Institutional information to external entities * DHDNs help improve the sample size by fostering the data sharing in a federal model, by giving due importance to privacy and security of sensitive, personal health information * DHDNs exist based on the presumption that organizations participating in multi-institutional activities standardize their in-house or organizational-level data to a single data schema known as a common data model. ### Quality, Safety, and Performance Improvement and * Research exist on a continuum of 'rigor' of soft science and 'hard science'. Hard science is the most sophisticated method of acquiring knowledge and involves inductive and deductive reasoning. Hard science might be considered superior based on tradition, authority, and experience * Organizations are going to make gains and sustain their operations in the future, using voluminous data and processing it with advanced methodologies to identify actionable intelligence. * Healthcare data sets serve as a key source of information for understanding the disease burden and health disparities that exist within patient populations, and they also inform strategies for addressing health inequities. * Drawing knowledge from mathematics and statistics, healthcare data were often classified into two major categories: quantitative data (e.g., age, admission dates and discharge dates, lab values) and qualitative data (text captured within medical records such as a progress note). ### Patient-Generated Health Data * Digital technologies now allow patients to track health data outside of the clinical setting. ### Data Specifications * Data collected are aggregated to preserve the confidentiality of information or protect vulnerable patient populations. On the other hand, disaggregated data are divided and broken down into smaller information units. * Defining the population, consider clinical and demographic factors (social determinants of health). * Defining the stratification variables is a precursor to structuring the reporting and assures the underlying data model can support the analysis. ### Code Specifications * Most measures require the use of standard code sets to define the patient's condition and status as a means of creating comparable, reliable, consumable measures. ### International Classification of Diseases (ICD) * On January 16, 2009, the U.S. Department of Health & Human Services (HHS) released the final rule mandating that everyone covered by HIPAA implement International Classification of Diseases ICD-10 for medical coding. ### Systemized Nomenclature of Medicine-Clinical Terms (SNOMED CT) * The Systemized Nomenclature of Medicine-Clinical Terms (SNOMED CT) provides healthcare professionals the ability to use different terms that mean the same thing in their clinical documentation systems. ### Current Procedural Terminology (CPT) * Current Procedural Terminology (CPT) codes are a uniform language for coding medical services and procedures to streamline reporting and increase accuracy and efficiency. ### Logical Observation Identifiers Names and Codes (LOINC) * Logical Observation Identifiers Names and Codes (LOINC) is a clinical terminology for the standardization of laboratory test names in the transactions between healthcare facilities, laboratories, laboratory testing devices, and public health authorities. ### Data Protocol * The measure developer must explicitly identify types of data and how to aggregate or link these data so that calculation of the measure is reliable and valid. ### Measure Documentation * Measure documentation occurs throughout the measure development process. ### Measure Calculations * Measure calculation is specific to each value-based reporting initiative, the type of measure, and the source of data. ### Benchmarking * Benchmarking compares an organization against other organizations that are well known for best practices within the same domain. * Balanced scorecards and dashboards are key reporting tools, that help organizations track and sustain the improvement of outcomes. ## Dashboards * In the early 1980s, executive information systems, also known as executive support systems, supported senior executives with information necessary for decision making by providing easy access to both internal and external information relevant to meeting strategic goals of the organization. * Organizations often develop dashboards to represent key management and performance indicators. ## Key Performance Indicators * What gets measured gets done. Key performance indicators (KPIs) are the critical or key indicators to measure performance or progress toward an intended result. They are used to set targets (i.e., desired level of performance) and track progress against the target. ## Reporting for Improvement * The management of data, analysis of information, and management through knowledge are necessary to change and improve healthcare organizations. ## Key Points * Data governance protocols and procedures such as business associate agreements, data use agreements, and data sharing agreements should be in place before sharing any sensitive institutional information to external entities. * DHDNs help improve the sample size by fostering the data sharing in a federal model by giving due importance to privacy and security of sensitive, personal health information. * DHDNs exist based on the presumption that organizations participating in multi-institutional activities standardize their in-house or organizational-level data to a single data schema known as a common data model. * Instruments such as questionnaires, surveys, and rating scales are the devices that healthcare quality professiona.ls and researchers use to obtain and record data received from the subjects. * Reliability is the extent to which an instrument measuring procedure yields the same results under similar circumstances by the same or different individuals on repeated trials. * Validity is the degree to which an instrument measures what it intends to measure and is usually more difficult to establish than reliability. * Parametric tests are used with data measured on a continuous scale (i.e., interval or ratio data, which also are known as variables data). * Nonparametric tests are used with categorical (attributes) data and with ordinal data, especially if the ordinal categories have a small range of possible values or a nonnormal distribution. * CI provides a range of possible values around a sample estimate (a mean, proportion, or ratio) that is calculated from data. * Data in general are abundant and wisdom is rare. * Good presentation of data creates interest and enhances understanding. Data are reported and analyzed on a regular basis. * The three most common measures of central tendency are mean, median, and mode. * Healthcare quality professionals can use various charts and diagrams when considering the data collected to make sense of them. * The management of da.ta, analysis of information, and management through knowledge are necessary to change and improve healthcare organizations. * The goal remains that governing body reports contain only the critical information needed for effective decision making. * Organizations rely on a bottom-up strategy for performance management where each operating unit (e.g., cardiology, radiology, and dialysis) or key pillar (e.g., finance, quality, and growth) sets its own targets toward meeting the overall organizational strategic goals. * Balanced scorecards focus on multiple dimensions that underlie care delivery * In the early 1980s, executive information systems, also known as executive support systems, supported senior executives with information necessary for decision making by providing easy access to both internal and external information relevant to meeting strategic goals of the organization. * Healthcare quality professionals are uniquely positioned to advance the use of data in their organizations and the comrnunity through various approaches and strategic thinking, including the following. * Key performance indicators (KPIs) are the critical or key indicators to measure performance or progress toward an intended result. * The performance improvement team can determine whether it wants to be average or raise the bar to a much higher level of performance. ## Online Resources * **Agency for Healthcare Research and Quality** * www.ahrq.gov * **Guidelines and Measures** * https://www.ahrq.gov/gam/index.html * **National Quality Measures Clearinghouse** * https://www.ahrq.gov/gam/index.html * **Quality and Patient Safety Resources** * https://www.ahrq.gov/patient-safety/resources/index.html * **State Snapshots** * https://www.ahrq.gov/data/state-snapshots.html * **Surveys on Patient Safety Culture** * https://www.ahrq.gov/sops/index.html * **American Health Information Management Association** * http://www.ahima.org/ * **American Society for Quality** * www.asq.org * **Australian Commission on Safety and Quality in Health Care** * www.safetyandquality.gov.au * **Centers for Disease Control and Prevention** * https://www.cdc.gov/ * **Behavioral Risk Factor Surveillance System: Annual Survey Data** * https://www.cdc.gov/brfss/annual_data/annual_data.htm * **National Center for Health Statistics** * https://www.cdc.gov/nchs/ * **National Vital Statistics System** * https://www.cdc.gov/nchs/nvss/ * **Centers for Medicare & Medicaid Services-Program Statistics** * https://www.cms.gov/research-statistics-data-and-systems/statistics-trends-and-reports/cmspro-gramstatistics * **Centre for Effective Practice** * https://cep.health/ * **Dartmouth Atlas Benchmarking Tool** * https://www.ruralcenter.org/resource-library/dartmouth-atlas-benchmarking-tool * **DARTNet Institute-Informing Practice: Improving Care** * http://www.dartnet.info/ * **eCQI Resource Center** * http://ecqi.healthit.gov/ * **Health Deata.gov** * https://www.healthdata.gov/ * **Health Information Management and Systems Society** * http://www.himss.org/ * **Health Quality Ontario** * https://www.hqontario.ca/ * **Health Resources & Services Administration Data Warehouse** * https://data.hrsa.gov/ * **Institute for Healthcare Improvement** * www.ihi.org * **International Society for Quality in Health Care** * https://isqua.org/ * The Joint Commission: https://www.jointcommission.org/ * The Leapfrog Group: https://www.leapfroggroup.org/ * National Academies of Sciences, Engineering, and Medicine: https://www.nationalacademies.org/ * Sharing Health Data: https://nam.edu/sharing-health-data-the-why-the-will-the-way-forward/ * National Association for Healthcare Quality: https://nahq.org/ * National Committee for Quality Assurance: https://www.ncqa.org/ * National Health Service-UK: https://www.england.nhs.uk/ * National Institutes of Health: https://www.nih.gov/ * National Quality Forum: https://www.qualityforum.org/Home.aspx * RAND Corporation: https://www.rand.org/ * Substance Abuse and Mental Health Services Administration: https://www.samhsa.gov/ * Drug Abuse Warning Network: https://www.samhsa.gov/data/data-we-collect/dawn-drug-abuse-warning-network?msclkid=e8b76dd5cf4311ecb656fdb0a8af21ff * National Survey on Drug Use and Health: https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health?msclkid=04aafca4cf441lecal8c11b9c2d245b9 * National Mental Health Services Survey: https://wwwdasis.samhsa.gov/dasis2/nmhss.htm * National Survey of Substance Abuse Treatment Services : https://wwwdasis.samhsa.gov/dasis2/nssats.htm * Treatment Episode Data Set: https://wwwdasis.samhsa.gov/dasis2/teds.htm * Visualizing Health: http://www.vizhealth.org/ ==End of OCR for page 64==