Health Indicator Tracking Dashboard Overview

Description of Dashboard Indicators

Indicators

Indicators are specific, measurable targets created to track health change over time.  The indicators in this dashboard belong to one or more sets of indicators based on their origin.  That is, some of the indictors are state versions of national Healthy People 2020 goals, some are used for a Health Improvement Plan, and some were designed by the agency to track program progress or simply selected by programs for display because of interest.

Finding an Indicator

A specific indicator can be found by choosing an indicator set and a topic which likely contains the desired indicator, or by searching for a word likely to be found in the title of the indicator.

Limitations of Indicators

Each indicator presents a single point of data for each year that it is available.   Consequently, an indicator is a very narrow view of the problem.  For instance, an indicator designed to measure current smoking rates among adults tells us nothing about the broader issues related to smoking such as sub-populations with higher or lower smoking rates, youth smoking, the average amount people smoke or attempts by smokers to quit.

Some indicators may be very similar to each other, representing only a difference in the target date or representing different years of data.  For some sources of data, changes in data collection methodology required the creation of two indicators with different years of data because older and newer data cannot be directly compared.

Limitations of Data

All datasets have limitations.   For example, data points may be missing from the dataset; data may have been incorrectly recorded, persons responding about their own health may misrepresent the truth (e.g., calculations from self-reported height and weight tend to under estimate obesity because women tend to under-report their weight and men tend to over-report their height).  If the data is likely to seriously misrepresent population health, it is not used. 

Slight differences in results from various reporting sources may also occur.  For instance, results derived directly from state datasets may be slightly different than that derived from a federal reporting site (e.g., HTTP://wonder.cdc.gov).  The reasons for this are complex and varied, but the difference is almost always quite small, and hence of little consequence.

Setting Targets

Each indicator has a target date and a target value.  Generally the state target value will represent a 10% improvement compared to the value for 2010, or the last available data prior to 2010.   However, sometimes a 10% improvement is unreasonable.  For instance, the rate of diabetes in North Dakota is rising most likely due to rising obesity.  Although obesity incidence in North Dakota appears to be leveling off, diabetes is going to continue to rise.   Even if obesity were to fall markedly over the next few years, diabetes is going to rise.  Consequently, limiting the rate of rise of diabetes represents a more realistic target than reducing the rate.   For some indicators, a federal target may also exist that may or may not be similar to the state target (e.g., the rate of teen pregnancy in North Dakota is much lower than the rate for the nation; therefore, the targets are very different).

The Display

The display provides a graphical presentation of the data and a brief written description of the findings.  Additional details, including a data table, can be found by clicking on the View Details button.  Technical details (e.g., lists of ICD codes used) can be found under the Technical Notes link and abbreviations can found under the Abbreviations link.

Currency of Data

The data will be updated periodically to add new years of data as they become available.  For some datasets, the data will be two to three years old before it is ready for addition to the dashboard.  Some data is not available every year; it might be collected every other year or even every fifth year, or not even on a regular schedule.  Missing years of data will be extrapolated by the graph, but the data table will show the data as missing. 

Data from Multiple Years

In order to obtain stable risk estimates when numbers are small, multiple years of data may be combined.  This may be done with non-overlapping time periods or using running averages (overlapping time periods).  Data points that are created using multiple years of data will only display the first year of the span of years on the x axis of the graph, but the table will give the specific years covered by that data point. 

Information about the Datasets

For most of the datasets found in the dashboard, additional information about the dataset can be found in the metadatebase located at http://www.ndhealth.gov/healthdata/healthdatametadata/.