Introduction

Beginning in the 2007-2008 school year, the annual collection of diabetes prevalence data was begun by the Massachusetts Department of Public Health (MDPH), Bureau of Environmental Health (BEH). The source of the data is school health records for all public and private schools in Massachusetts serving grades Kindergarten through 8 th grades (approximate ages 5-14). School nurses or administrative staff are requested to provide the number of children with diabetes in their school, enumerating whether a child has Type 1, Typ2, or an unknown type of diabetes. No personal identifiers such as name or street address of a child are provided to the MDPH.

2007-2008 Prevalence

Prevalence data for the first year of data collection, the 2007-2008 school year, are presented only by county location of the school. Most importantly, these data represent the prevalence of a county which served as the location of the school. It therefore is possible that students with diabetes actually reside in a different county from which the student's school is located. The numbers for each county were calculated by summing together the number of children with diabetes attending a public or private school whose address location was within the county of interest. The denominator was calculated by summing together the enrollment number of all schools in that county. Tables are provided for Type 1, Type 2, and All Types combined. The All Types table includes children with an unknown type of diabetes (less than <50 children statewide). Prevalence is presented per 100,000 population and confidence intervals are shown to indicate the precision of the prevalence estimate.

As stated above, when examining the 2007-2008 prevalence of diabetes, caution should be taken in their interpretation because calculations were based on the physical location of the school and not the actual residence of the child. This can result in either an overestimate or an underestimate of the true prevalence for residents of that county. For example, some children who are enrolled in a school in Cambridge may reside in Boston, so the prevalence estimate for Suffolk County would exclude those children. In addition, enrollment information is estimated by school nurses and may not necessarily represent the most precise accounting of enrollment. It should also be noted that the prevalence estimates have not been age-adjusted.

Approximately 98% of Massachusetts schools (about 2,100 schools) submitted diabetes data in the 2007-2008 school year. The prevalence of diabetes (all types combined) was estimated to be 264 per 100,000 persons (95% Confidence Interval 252 - 275). Prevalence was highest for Barnstable County (321 per 100,000) and lowest for Dukes County (191 per 100,000). U.S. statistics on the prevalence of diabetes (all types) for the roughly equivalent age group of the student population included in a national survey demonstrated 183 cases per 100,000 (NHANES 2005-2006). While the US prevalence is lower than the MDPH statewide estimate, the methods used by the CDC and MDPH for their respective prevalence estimates are significantly different. This could account for some of the difference observed (e.g., CDC estimates are based upon a sample of individuals who participated in their National Health and Nutrition Examination Survey, while MDPH data are based upon school health information for all children who attend grades K-8 in public and private schools in Massachusetts). In order to determine if the prevalence of diabetes was higher or lower than expected in a specific county, it would be more appropriate to compare that figure with the prevalence in the state of Massachusetts as a whole rather than with CDC figures. When the county prevalence estimates are compared with the statewide prevalence estimate, none of the counties appear statistically significantly higher or lower. This suggests a relatively geographically stable prevalence across the state.

Documents

2008-2009 School Year

Data collection for year 2 (2008-2009 school year) of the diabetes surveillance was enhanced in order to be able to determine prevalence estimates by community of residence of the children in addition to estimates for the county locations of the schools.

Participation of public and private schools serving grades K through 8 was again excellent with more than 99% of schools reporting. Estimates of prevalence by county of school location were very similar to the previous school year for all counties. Importantly, there were again no statistically significant differences between any counties and the statewide prevalence for 2008-2009.

When prevalence is examined by community of residence of the children, 9 of the 351 communities had prevalence estimates that were statistically significantly higher than that observed for the state as a whole. Seventy-four (74) communities had prevalence estimates that were statistically significantly lower than the statewide prevalence.

Documents

Future Surveillance

Diabetes prevalence data for 2009-2010 are currently being collected. These data will also be available by community of residence of the children. Analyses of these data will include combining multiple years of data that will enable more stable estimates of prevalence and examination of geographic patterns.


This information is provided by the Environmental Health within the Department of Public Health.