Read A Data Table
These How To pages provide information on how to read and interpret the
data tables from the reports produced by the following CDC surveillance
- The Pediatric Nutrition Surveillance System (PedNSS)
- The Pregnancy Nutrition Surveillance System (PNSS)
What can I learn from the tables?
The following paragraphs describe the types and number of tables
included in the reports, the level of analysis, and how the records are
What types of data tables are included in the reports?
The PedNSS and PNSS reports contain three types of tables focusing on
- demographic characteristics.
- prevalence of health indicators. These tables
include geographic comparisons and analysis of health indicators by
demographic variable, birth outcome, and risk factor analysis.
- trends in population.
Note that these three categories are also represented in the Table of
Contents on the left side of your screen.
What health indicators are reported by demographic variables, geographic
comparisons, and trends?
The PedNSS tables provide race/ethnicity-specific or age-specific
prevalence, geographic comparisons and trends for all of the health
- low birthweight
- high birthweight
- short stature
- children ≥ 2 years of age overweight (85th to <95th and ≥95th)
For detailed information, see
The PNSS tables provide race/ethnicity-specific or age-specific
prevalence, geographic comparisons and trends for only selected health
indicators that are included in groups and listed in the table titles. The
PNSS health indicator groups and the selected health indictors for each
group are the following:
- Maternal health indicators
- Prepregnancy BMI
- Weight Gain
- Diabetes During Pregnancy
- Hypertension During Pregnancy
- Maternal behaviors indicators
- Medical Care
- WIC Enrollment
- Multivitamin Consumption
- Smoking Indicators
- Smoking Status
- Household Smoking
- Infant health indicators
- Full Term LBW
- Breastfeeding Status
For detailed information, see PNSS health indicators.
How many tables are produced?
The CDC produces annual nutrition surveillance reports for each
contributor as follows:
- The PedNSS report contains 22 data tables, starting with the 2010 PedNSS reports. Table 22 is the annual PedNSS Comparison of Record Volume and Data Quality table and is generated from the same annual PedNSS unique child file as the other 21 tables. It is sent to contributors separately from the other 21 tables. Note that although the report lists 22 data tables, only 20 of those tables have data available. Table 14 and Table 15 are not available.
- The PNSS report contains 23 data tables.
What is the analysis level of the data tables?
The PedNSS and PNSS summary tables of demographic, health indicators,
health indicators by demographic variables, and trends are produced
annually at the national and contributor levels.
The PedNSS and PNSS geographic tables are produced annually at all levels
(nation, contributor, clinic, or school, county, local agency, metro area,
and region/school district).
The following abbreviations are used to indicate the level of analysis:
- A = Clinic
- B = County
- C = Contributor (state, District of Columbia, U.S. territory, or Indian Tribal Organization)
- D = Nation
- E = Region
- F = Local Agency
- G = Metro Area
What records are included in the tables?
The records included in each table are described below.
PedNSS is a child-based system with only one record per child
included in the table analysis.
- Data are collected based at each child
visits and a child may have more than one visit (or record)
during a reporting period of one year.
- For children with multiple records during the reporting
period, one unique child record is created using valid
demographic, anemia, growth, birthweight, and breastfeeding data
from multiple visit records. For more information, see
How is the unique child
PNSS is a pregnancy-based system with only one record for each
pregnancy included in the table analysis.
- Prenatal data are collected at the
first prenatal visit to the clinic and postpartum data are
collected at the postpartum visit after delivery. PNSS
contributors combine information from these different time
periods into a single record.
- Records are selected based on the infant's date of birth or
expected date of birth for the reporting period of January 1
through December 31.
In the PNSS data tables, records are coded as follows:
- A complete record contain both prenatal and
- A prenatal record contains only data that were
collected when the woman enrolled in the public health program
- A postpartum record contains data that were collected
when the woman entered the program after delivery.
Prenatal and postpartum records represent records that could
not be linked as a complete record because data from one time
period was missing. These records are included in the table
How do I know if the data is of good quality?
Surveillance data are useful for program management and policy
development only if the data are of good quality. The CDC evaluates the
completeness and quality of the data submitted from contributors and
provides them with a data quality report.
- The data quality report, Summary of Record Volume and Data Quality,
is generated each time a PedNSS or PNSS transaction file is sent to CDC
- The report contents include a record volume update, a summary of
identified data quality problems, and a description of data quality
problems related to missing data, completion code and record linkage
errors, mis-codes, biologically implausible values (BIVS), cross-check
errors, unusual data distributions, and low or high standard deviations.
- The section How To Review Data Quality provides guidance on how to evaluate the quality of your data so erroneous conclusions are not drawn from poor quality data.
- The annual PedNSS Comparison of Record Volume and Data Quality table should be used to assess data quality before analyzing the prevalence data in tables of the PedNSS and PNSS annual reports. For example, the PedNSS Comparison of Record Volume and Data Quality report shows if any health indicators have more than 40% missing, miscoded or biologically implausible data. If a health indicator is this extensively flawed, it is excluded from the data tables in annual PedNSS reports and applied to trend tables. The annual PNSS Comparison of Record Volume and Data Quality table is forthcoming.
Are publication standards applied to annual PedNSS reports?
Only production-quality monthly and quarterly files from PedNSS contributors are added to the CDC PedNSS database and referenced for annual PedNSS reports. Such files have adequate data quality and minimum or vital field requirements for PedNSS participation. (PDF-220k) CDC also excludes missing and miscoded data, as well as biologically implausible values (BIVs) from annual report analysis.
Additional standards for producing annual PedNSS reports were implemented starting with the 2010 PedNSS reports. Data are now excluded from analysis if >5% of records have biologically implausible values, or if > 40% of records cumulatively have missing, miscoded or BIV data. Annual PedNSS report prevalence rates are also suppressed if they are unbelievably low or high. Questionable data and large shifts in prevalence rates in current and prior year annual reports are also flagged (but not excluded from analysis).
For detailed information, see Annual PedNSS Report Standards.
How do I learn about the individual tables?
The following pages contain descriptions of the tables, explanations of
their use, information about the table data items, sample tables, and
links to actual tables.
- Each sample data table will familiarize you with the table's format and
the information it contains.
- Most of the sample tables are taken from the PedNSS reports, since the designs of the PedNSS and PNSS tables are
similar. If you understand how to read a PedNSS table, you should be able
to read a similar PNSS table without difficulty.
For more information about the general structure of tables, see
Basic information about tables.
The section How To Interpret Data will provide you with more
information on how to interpret the data in the tables and apply the basic
principles of epidemiology. The section
How To Review Data Quality will
provide guidance on how to evaluate the quality of your data so erroneous
conclusions are not drawn from poor quality data.
back to top
Page last reviewed: March 22, 2011
Page last updated: March 22, 2011
Content Source: Division of Nutrition, Physical Activity and Obesity,
National Center for Chronic Disease
Prevention and Health Promotion