2012 National Conference on Health Statistics
Monday: Learning Institute
The Learning Institute is offered in four sections that include both hands-on and educational sessions.
- Hands-on sessions are held in an interactive, classroom style setting. Participants use computers and work directly with data from specific NCHS surveys and other data sources.
- Lecture sessions are conducted in a theatre-style setting, with one or more individuals presenting information on data topics derived from NCHS surveys and other data sources.
Section 1: 9:00-10:30 a.m.
A1. Understanding and Analyzing Ambulatory Health Care Data From the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS): Overview and Basic Analyses (Hands-on Session) [LI-05]
This session will provide a general introduction to two record-based surveys of ambulatory medical care: NAMCS and NHAMCS. The session will include an overview of both surveys, as well as basic data analyses using SAS and Stata. Participants will learn how to create analytic datasets using downloadable public-use microdata files. Programming examples will include generating weighted frequencies with and without standard errors, subsetting data to examine specific diagnoses, producing visit rates by selected characteristics, using write-in procedure data, and calculating mean time spent with physician.
A2. Advanced Session: The National Health Interview Survey (NHIS) (Hands-on Session) [LI-12]
Using data from NHIS, participants will learn how to merge multiple years of data files, access and use the NHIS imputed income files, and use SAS-callable SUDAAN in logistic regression and other multivariate analyses. NHIS staff members will be available to assist and answer questions. Participants should have some experience using SAS and some experience with NHIS datasets. Assistance with Stata syntax and programs will also be provided.
A3. Tools, Techniques, and Framework for Analysis of Vital Statistics Data on External Causes of Injury Death (Lecture Session) [LI-13]
The International Collaborative Effort on Injury Statistics has produced two matrices that serve as a framework for the presentation and analysis of data collected on the external causes of injury and on injury diagnoses. These are, respectively, the External Cause Matrix and the Injury Mortality Diagnosis (IMD) Matrix adapted for ICD–10. The presenters will discuss the origins of these instruments, as well as their rationale and organization. Examples based on NCHS reports and microdata will highlight the usefulness of these tools in analyzing mortality data.
- Tools, techniques and framework for the analysis of vital statistics data on external causes of injury death [PDF - 3 MB]
- The ICD injury matrices [PDF - 1.3 MB]
A4. Web Tutorial for the National Health and Nutrition Examination Survey (NHANES) and Centers for Medicare & Medicaid Services (CMS) Linked Records (Hands-on Session) [LI-17]
NHANES data linked with Medicare and Medicaid records from CMS were recently released for use in the NCHS Research Data Center (RDC). This hands-on session will demonstrate how to use a newly developed special-topic Web tutorial for NHANES and CMS linked records. This supplemental self-learning tool complements the existing NHANES tutorials and is designed for users of NHANES and CMS linked data. The session will provide instruction using actual examples of statistical programs and outputs. Main topics include how to (a) access the NHANES and CMS linked data through the RDC, (b) work with the feasibility files (publicly available limited variable files that can be downloaded directly from the NCHS website) to determine the maximum available sample size overall and for subgroups for each linked file, and (c) how to produce summary statistics using the feasibility files with SAS survey procedures and SUDAAN. Important analytic issues regarding linked data will also be discussed, including how to assess additional “nonresponse” due to records that could not be linked, the use of survey weights for linked files, and how to define an analytic cohort.
A5. Finding Key Resources From NCHS (Hands-on Session) [LI-21]
This workshop will include several learning modules that should equip students and researchers to locate statistical information for health research. Participants will be asked to solve a series of scenarios using the NCHS and CDC websites as the main data sources.
Section 2: 11:00a.m.—12:30 p.m.
B1. An Overview of the National Health and Nutrition Examination Survey (Lecture Session) [LI-02]
This session will provide an overview of the NHANES, discuss NHANES history and selected survey design issues and describe data collection in NHANES. It will also provide examples of how the data are used to inform science and policy and future directions for the survey.
B2. The National Survey of Residential Care Facilities (NSRCF) (Hands-on Session) [LI-04]
This workshop will present an overview of the first-ever NSRCF, conducted in 2010. The session will provide computer exercises that demonstrate how to access and analyze the NSRCF public-use files, including the facility file and the resident file. The facility file includes data on facility characteristics, structure and physical environment, staffing, operating procedures, services offered, fees charges, and the distribution of residents at the facility level. The resident file includes data on a selected set of facility-level variables from the facility file, sociodemographic characteristics, health and physical functioning, cognitive functioning, service use, involvement in activities, and charges. The workshop is intended for experienced analysts skilled in SAS or Stata.
B3. An Overview of the National Health Interview Survey (NHIS) (Lecture Session) [LI-09]
Users new to NHIS data, or those who need a refresher, are encouraged to attend this workshop, which will include a brief introduction to the history, development, purpose, data access, and current activities of NHIS. Question content, sample design, weighting, and variance estimation will also be presented, along with two examples of research using NHIS.
- Studying Disability in Households Using Multiple Years of the National Health Interview Survey [PDF - 126 KB]
- From the Lab to the Field: Preliminary Results and Next Steps in Field Testing the National Health Interview Survey Audio Computer-Assisted Self-Interviewing Module [PDF - 496 KB]
- A Brief History of the National Health Interview Survey and Plans for Its Future [PDF - 96 KB]
- Aspects of the National Health Interview Survey (NHIS) [PDF - 64 KB]
B4. Using the National Survey of Family Growth (NSFG) (Hands-on Session) [LI-15]
This workshop will introduce the National Survey of Family Growth (NSFG). The session will include the topics covered in the new 2006–2010 NSFG; how to obtain the public-use, ACASI (audio computer-assisted self-interview), , and contextual data files; how to find and choose the variables you need; and how to run example analyses of the male and female respondent files, and the female pregnancy file, including how to estimate variances for complex samples. Time will be reserved for answering participants’ questions.
B5. Health Indicators Warehouse (Hands-on Session) [LI-18]
The Health Indicators Warehouse (HIW) was launched in February 2011 and contains more than 1,200 pretabulated health-related indicators at the national, state, county, and hospital referral region levels. Indicators are derived from more than 200 different data sources, including vital statistics, survey data, surveillance data, and Medicare and Medicaid administrative data. The HIW includes indicators created for Healthy People 2020, the County Health Rankings, and 151 new utilization and quality indicators created by the Centers for Medicare & Medicaid Services. The purpose and uses of the HIW will be described, and an overview of HIW functionalities will be demonstrated. Participants will learn through hands-on examples how to search for indicators; display the data in table, chart, and map formats; and download data by exporting and by using API/Web services. Time will be reserved for participants to work on their own in the HIW with help from NCHS staff.
Section 3: 2:00—3:30 p.m.
C1. Understanding and Analyzing Ambulatory Health Care Data From the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS): Advanced Analysis (Hands-on Session) [LI-08]
This session will provide an introduction to advanced programming topics and techniques for experienced users of NAMCS and NHAMCS. Participants will use SAS and Stata programming examples to conduct the following analyses: manipulate the physician weight to obtain physician-level estimates, modify the patient visit weight to approximate patient-level estimates, use linear regression to model time spent with physician, perform logistic regression to model visits for essential hypertension, and combine multiple years and ambulatory care settings to examine utilization by selected patient characteristics.
C2. The National Survey of Children With Special Health Care Needs (NS-CSHCN) and the Survey of Pathways to Diagnosis and Services (Hands-on Session) [LI-10]
This workshop is designed to assist users in managing SLAITS data to create analytical data sets using NS-CSHCN and the follow-up Survey of Pathways to Diagnosis and Services. Through syntax demonstrations using SAS and SUDAAN, special emphasis will be placed on teaching users how to merge files and use the proper weights.
C3. Beginner’s Session: The National Health Interview Survey (NHIS) (Hands-on Session) [LI-11]
Participants will learn how to access NHIS data, create analytic files, and run basic statistical programs. NHIS staff members will be available to assist and answer questions. Participants should have some experience using SAS, but experience with NHIS datasets is not required. Assistance with Stata syntax and programs will also be provided.
C4. Understanding Mortality Statistics: The Importance of Cause-of-Death Certification and Coding (Lecture Session) [LI-14]
Understanding how cause-of-death data are collected and coded is important for the interpretation of mortality statistics. This session is divided into two parts. The first part will be a primer on how cause of death is to be reported by physicians, medical examiners, and coroners. This will be followed by a discussion of common problems with cause-of-death certification practices and how these problems affect the data. The second part will be a description of basic coding rules and procedures and how NCHS implements these according to the International Classification of Diseases (ICD). Also discussed will be changes in the ICD and how these should be handled in the analysis of cause-of-death data.
- Understanding Mortality Statistics: The Importance of Cause-of-Death Certification and Coding [PDF - 337 KB]
- Coding and classification of causes of death in accordance with the Tenth Revision of the International Classification of Diseases [PDF - 1 MB]
C5. Health Data Interactive (Hands-on Session) [LI-19]
This session will present information on the purpose of Health Data Interactive (HDI) and give some examples of how to use the data. Information on the types of data sources used to create HDI tables, some background information on how data are tabulated to be included in the HDI tables, and the strengths and limitations of the system will also be provided. Uses of HDI will be illustrated, drawing from recent NCHS Data Briefs and QuickStats that used the pretabulated HDI data. Participants will then be presented with example questions that can be answered using information contained in HDI, and then shown how to find and display the results in tables, charts, and maps as well as download the data for further manipulation. Participants will be given time to work on their own as well as with the group to utilize this data tool.
Section 4: 4:00 – 5:30 p.m.
D1. The International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM): Understanding the Differences in ICD-9-CM and ICD-10-CM and Its Potential Impact on Data Analysis (Lecture Session) [LI-01]
On October 1, 2013, the ICD-9-CM code sets used by health care agencies to report medical diagnoses and inpatient procedures will be replaced by the ICD-10-CM and ICD-10-PCS code sets. In preparation for this transition, NCHS’ Classification and Public Health Data Standards Section will conduct a workshop on the changes from the ICD-9 to ICD-10 coding systems and how these changes will impact public health data analyses. The session will focus on major differences between the two code sets and will offer suggestions for assessing trends and data analysis, using examples from areas such as injury, environmental health, and chronic disease.
- Understanding the Impact of the Differences in ICD-9-CM and ICD-10-CM and Its Potential Impact on Data Analysis [PDF - 76 KB]
D2. The National Hospital Discharge Survey (NHDS) (Hands-on Session) [LI-03]
This workshop will provide an opportunity for users of NHDS to learn to perform trend analyses with NHDS data. Each registered participant will receive the NHDS 1979–2010 Multi-Year Data File on CD–ROM and will gain proficiency in loading the data file, reading it into SAS, creating various types of analytic variables from medical diagnoses and procedures coded on the record, and running elementary SAS procedures such as PROC FREQ, PROC MEANS, and PROC TABULATE. Although formal instruction will be provided, the session will emphasize hands-on exercises. NCHS analysts who have worked with the survey will be available to provide technical support and to answer questions about programming and statistical techniques appropriate for the data.
D3. What We Eat in America: National Health and Nutrition Examination Survey (NHANES) Dietary Data—What to Know and How to Use It (Hands-on Session) [LI-06]
This hands-on session will help participants gain an understanding of NHANES’ complex dietary data and documentation—commonly called What We Eat in America (WWEIA)—and the technical support files of the USDA’s Food and Nutrient Database for Dietary Studies. Particular emphasis will be placed on the specifics of collection and dissemination of dietary data—both data for analysis and summarized dietary data tables and briefs. Participants will learn what information on food, nutrient intakes, and sociodemographic characteristics is included in the data files that are important in dietary intake analysis. Dietary components include food codes and descriptions, WWEIA Food Categories, nutrient values, combination codes, eating occasions, food source, day of the week, and time of day. The basic steps in using NHANES’ WWEIA data will be discussed, with special attention to identifying appropriate types of analyses.
- WWEIA, NHANES Dietary Data: Data Preparation Steps for Dietary Analysis [PDF - 249 KB]
- What We Eat in America, NHANES Dietary Data: What To Know and How to Use It [PDF - 2.7 MB]
D4. Using Drug Data From the National Ambulatory Medical Care Survey (NAMCS) and National Hospital Ambulatory Medical Care Survey (NHAMCS) (Hands-on Session) [LI-07]
This workshop is for anyone interested in using NAMCS and NHAMCS drug data. It presupposes basic familiarity with the surveys. Since the 2006 survey year, NAMCS and NHAMCS have utilized Multum Lexicon therapeutic class categories and have produced a separate data file for drug ingredients. The characteristics and structure of the drug variables will be discussed, and SAS and Stata programs will be used to explore techniques for analyzing current and trended medication data.
D5. VitalStats—Accessing Natality Data Online (Hands-on Session) [LI-16]
This session will introduce VitalStats—the online data access tool for birth, infant mortality, and fetal death. The session will include training in how to use VitalStats, how to retrieve prebuilt tables, how to find and choose the variables you need to create your own tables, and how to extract created tables to Excel. Time will be reserved for answering participants’ questions.
D6. Analysis of Data From the Early Release Program of the National Health Interview Survey (Lecture Session) [LI-20]
The Early Release (ER) Program of the National Health Interview Survey is a mechanism for releasing selected estimates and making selected microdata accessible on an expedited schedule, just 6 months after data collection. This program enables health practitioners, researchers, and policymakers to have the latest information on health measures that reflect issues of public health concern, and allows for monitoring of the health status of the U.S. population. Along with an overview of the evolution of the ER Program, this session will provide information on the various ER products, including reports on key health indicators, health insurance, wireless substitution, and health care access and utilization. Presenters will also discuss applications for the preliminary quarterly microdata files that are produced as part of the ER Program.
- Wireless Substitution: Semiannual Estimates from the NHIS ER Program [PDF - 550 KB]
- Analytic Uses of National Health Interview Survey Early Release Products: Health Insurance Reports, Preliminary Microdata Files, Special Topics [PDF - 794 KB]
- National Health Interview Survey Early Release Program: Overview and Key Health Indicators Report [PDF - 91 KB]