Crude rates: A crude rate is the number of events (such as deaths) in a specified time period divided by the number of people at risk of these events in that period (typically, a state or county population unless the event only affects a subset of the population). This figure is generally multiplied by a constant such as 1,000 or 100,000 (the “multiplier”) to get a number that is easy to read and compare and is reported as “per 1,000” or “per 100,000.” Rates calculated in this manner are called crude rates.

Age-adjusted rates: Crude rates adjust for differences in population size but not differences in population characteristics. To see if a rate is high due age differences in the population, we need to use age-adjusted rates. These rates are computed by taking crude rates for each age group and applying them to a standard population. Since 1999, the standard has been the 2000 U.S. population.

The major use of age-adjusted rates is to allow comparisons between different areas and/or different time periods, especially for metrics highly reliant on age. Users should be aware that an age-adjusted death rate has no absolute meaning; it is an artificial number based on a hypothetical population and is only useful for comparing with other rates calculated in the same manner.

The number of newly diagnosed cases of a disease or condition during a specific time period.

The number of cases of a disease or condition, including both newly diagnosed and pre-existing, for people alive on a certain date.

An estimated range of values that are likely to include the true unknown population parameter.  The confidence interval is a measure of the variability in the data.  If the confidence interval is 8.1% - 15.8%, then we are 95% confident that the actual percentage is between these numbers.

A mathematical measure of the difference between groups. The difference is said to be statistically significant if the difference between the groups is greater than what is expected to happen by chance alone 95% of the time.

When the confidence intervals for two groups do not overlap (include any values in common), then the two groups are statistically significantly different.  When the confidence intervals overlap just slightly, a statistical test can be performed to check for a statistically significant difference.

The value exactly in the middle of all the values in the sample or in the data.  Half of the values are above and half are below.

The amount of change in a statistic over a given time interval. A positive percent change corresponds to an increasing trend, while a negative percent change corresponds to a decreasing trend. 

​The change in rate, or number of events, over time, which is usually expressed as an increasing or decreasing trend over a period of years. Although there may be some change over time, the trend is not identified unless it is statistically significant (i.e. greater than what is expected to happen by chance alone 95% of the time).  A change from one year to the next is not a trend. There must be at least 3 years (or periods) of data to show a trend.  

When two things have a mutual relationship or connection between them, they are said to be correlated. A strong correlation, association, or connection between two things does not mean that one of things caused the other.

Example: There is a correlation between eating ice cream and getting sunburned, but neither one influences, causes, or affects the other. They are both associated with a separate factor, sunny weather. The sunny weather is the confounder (see definition).

The relationship between two events or two variables (cause and effect), where a change in the cause results in a change in the effect.

A variable that is associated with two other variables, and because of its association, the confounder may mask a true association between the variables or suggest an association where there is none.

School district boundaries outline the 5 subcounty geographic areas in Kitsap:

• Bainbridge Island School District
  

• Bremerton School District
  

• Central Kitsap School District
  

• North Kitsap School District
  

• South Kitsap School District

ZIP code boundaries outline the same 5 subcounty geographic areas in Kitsap, but their outlines are slightly different.

• Bainbridge Island - 98061, 98110

• Bremerton - 98310, 98312, 98314, 98337

• Central Kitsap - 98311, 98315, 98380, 98383, 98393

• North Kitsap - 98340, 98342, 98345, 98346, 98364, 98370, 98392

• South Kitsap - 98322, 98353, 98359, 98366, 98367, 98378, 98384, 98386

Census Tract boundaries outline the same 5 subcounty geographic areas in Kitsap, but their outlines are slightly different than ZIP codes and School Districts. Census Tracts are used at a subcounty level to organize Census data. 

The Behavioral Risk Factor Surveillance System (BRFSS) is the largest, continuously conducted, telephone health survey in the world. It enables the Centers for Disease Control and Prevention (CDC), state and local health departments, and other health agencies to monitor the health and health behaviors of adults to guide policy and programs.

Office of the Superintendent of Public Instruction (OSPI) provides aggregate school enrollment numbers, dropout and graduation rates for Washington schools, as well as aggregate demographic and testing data. 

The United States Census Bureau completes a census of all residents every 10 years in the United States. The last census was completed in 2020. The census has a limited number of questions about the demographics of our population. 

In addition, the Census Bureau conducts the American Community Survey (ACS) every year.  The ACS is a mandatory, ongoing statistical survey that samples a small percentage of the population every year to gather information about population characteristics, housing, and economics among other topics.

Due to the impact of the COVID-19 pandemic, the Census Bureau did not release 2020 ACS 1-year data products, but 5-year data products are available for all 5-year periods. For more information about the American Community Survey, see census.gov/programs-surveys/acs

Washington State Department of Health (DOH) makes data and statistical reports on health, healthcare and the environment available to the public.  Topics range from cancer and communicable diseases to drinking water and births and deaths, among others. 

The Washington Tracking Network (WTN), a DOH data system, is an easily accessible source for environmental and public health data, including tables, charts and maps. 

Washington State Department of Social and Health Services (DSHS) Facilities, Finance and Analytics Administration (FFA) provides leadership in financial, operational and risk management services. 

This administration produces a comprehensive time-series collection of county and school district-level data related to substance use and abuse, and the risk factors that predict substance use among youth, called the Risk and Protection Profiles for Substance Abuse Prevention for Washington State and its Communities. 

In addition, they provide aggregate client counts, use rates and cost by age groups for many DSHS services, as well as research articles on mental health and substance use in Washington. 

Washington State Healthy Youth Survey (HYS) is a collaborative effort of the Office of the Superintendent of Public Instruction, the Department of Health, the Department of Social and Health Services Division of Behavioral Health and Recovery, and the Liquor Control Board.

HYS provides information about the self-reported health and health behaviors of youth in grades 6, 8, 10 and 12 in Washington to guide policy and programs that serve youth. 

Washington State Office of Financial Management (OFM) provides population estimates by age, sex, race and Hispanic origin, as well as estimates of population density and change. 

The Washington State Statistical Analysis Center provides access to crime and justice statistics and conducts studies and analysis to inform Washington policy and budget development.  They maintain a clearinghouse for crime and justice statistics, with data on crime, arrests, court filings and sentences, jail and prison populations and juvenile justice. 

The County Health Rankings measure the health of nearly all counties in the nation. They are produced every year by the Robert Wood Johnson Foundation and the University of Wisconsin Population Health Institute.

Kitsap Public Health District’s Assessment & Epidemiology program uses RStudio for data analysis.

RStudio is a free, open-source Integrated Development Environment (IDE) developed by Posit that serves as a user-friendly interface for the R programming language.

The R programming language is used by Kitsap Public Health District (KPHD) to create standardized, reproducible summaries of data and perform statistical analysis.

Kitsap Public Health District’s Assessment & Epidemiology program uses Joinpoint for trend analysis.

Joinpoint Trend Analysis Software is statistical software for the analysis of trends using joinpoint models where several different lines are connected at the inflection points or “joinpoints” (for example, zero joinpoints is a straight line). 

The software takes trend data (data over time) and fits the simplest joinpoint model that the data allow, starting with the minimum number of joinpoints, and tests whether more joinpoints are statistically significant and must be added to the model, up to the maximum number of joinpoints the data allow. This allows the user to test whether an apparent change in trend is statistically significant.  

The models can analyze crude and age-adjusted rates and proportions, use a Poisson model of variation, and tests of significance use Monte Carlo permutation methods. KPHD typically uses a log-transformed model which calculates the annual percentage change in the outcome.  

When an annual percentage change is statistically significantly different from zero for a set of data points over time, we say that the trend is increasing or decreasing. When the annual percentage change is not statistically significant, we say the trend is not changing over time.

Some analyses may be conducted using Joinpoint regression methods in R and some are done in the Joinpoint Trend Analysis Software.

The number or count of events best estimates:

• The size of a problem in any area (e.g., how many people die of cancer) OR

• Population changes due to birth, death, and other factors.

Numbers best answer questions like “How many people are impacted?” and “How many more did we see this year compared to 5 years ago?”

But, using just numbers, we cannot readily compare two areas or two time periods where the population has changed substantially over time. Such comparisons should take the size of the population into account to avoid wrong conclusions.

To eliminate the effect of different-sized populations, we compare rates (a type of ratio). These rates could take the form of rate per 100,000, rate per 1,000, a percentage, or a proportion. 

Ratios are a direct expression of one measure relative to another, such as the ratio of pink dots to green dots. A real world example is the ratio of males to females in a population.

Proportions are a type of ratio that relate a part to a whole. The numerator is included in the denominator, and it can be expressed as a decimal, fraction, or percentage.

Rates are a type of ratio that incorporate the dimension of time. It is a measure of the frequency of an event in a defined population over a specified period of time. A percentage is actually the same thing as a rate per 100.

After evaluating data sources for accuracy, KPHD epidemiologists may still suppress (not show) some of our data and information intentionally.

Although it may vary under certain situations, like in the case of an outbreak, there are typically two reasons you will see missing data in our work:

• Confidentiality: We do not show data based on fewer than 10 residents because of a concern about confidentiality. Small numbers under 10 are a concern because of their potential to reveal personal information about an individual or lead to inaccurate conclusions that may cause harm to residents in Kitsap. We intend to provide general information about our community as a whole rather than information about individuals. 

• Data Reliability: Even when confidentiality is not a concern, we may not show data if we are concerned about the reliability of the data. Unreliable data may fluctuate widely from year to year, making it difficult to draw meaningful conclusions. When we do show unreliable data, we will provide you with additional information to help you interpret the data and a note telling you the data does not meet our reliability standards. One way we provide you with more information is by providing confidence intervals. When a percentage is highly unreliable, the confidence interval will be extremely wide, indicating that the data could vary across this range. 

NOTE: Kitsap Public Health District sometimes releases data that would otherwise be suppressed when data are needed to protect the community from disease outbreaks or other urgent health risks. For example, we might release data about three cases of illness in a specific area of the county so that other people in that area can take steps to protect their health. In these situations, the Health District weighs the public benefits of releasing information against the potential for identifying individuals and releases the least amount of information needed to protect public health.