The data contained on Our Healthy Massachusetts come from a variety of sources.

2010 Census – The U.S. Census Bureau conducts a survey of population and housing units every ten years. Read more about it online here.

The American Community Survey (ACS) – An annual sample survey conducted by the U.S. Census Bureau that collects a range of information about demographics and housing characteristics. Read more about it online here.

Behavioral Risk Factor Surveillance System (BRFSS) - The Behavioral Risk Factor Surveillance System (BRFSS) survey is an ongoing annual, random digit-dial telephone health survey of adults ages 18 years and older. The survey is conducted in all 50 states in collaboration with the Centers for Disease Control and Prevention (CDC) and state departments of public health and has been conducted in Massachusetts since 1986. BRFSS collects information about a variety of health concerns such as risk factors for chronic conditions and preventive behaviors. Respondents can choose to respond to the survey in one of three languages: English, Spanish, or Portuguese. The Health Survey Program within the Massachusetts Department of Public Health (MDPH) coordinates and analyzes BRFSS data. Data are analyzed in SAS software (v 9.1) using weighted variables to provide population-based estimates among Massachusetts adults. Data from 2007 through 2009 (except where noted) were analyzed in the aggregate to provide a larger sample of people with diabetes. Because BRFSS is a self-reported survey, it has several limitations which must be considered when interpreting the data:

1. People who do not have a telephone or who otherwise do not participate, due to such factors as unavailability, language barriers, or lack of interest, are not represented in the survey. Individuals less than 18 years of age and those who are institutionalized are excluded from the survey.
2. Self-reported responses, particularly those pertaining to time and frequency of behaviors from the BRFSS, are subject to recall bias.

Only in instances where statistical significance was clearly stated were the results statistically significant. The difference between two percentages is statistically significant (with 95% probability) if the 95% confidence intervals surrounding the two percentages do not overlap. More information on the BRFSS survey can be obtained online here.

Birth Data - Data on births are based on information from the Massachusetts Standard Certificate of Live Birth (1989 revision) filed with the Registry of Vital Records and Statistics. Medical information, such as birth weight and gestational age, is based on information supplied by hospitals and birthing facilities. Demographic and behavioral data, such as race and ethnicity and smoking during pregnancy, are supplied by the women who gave birth. For example, women chose their race from five categories: White, Black, Asian/Pacific Islander, American Indian, and Other. Mothers identified their ancestry by selecting one of the 38 ancestry/ethnicity groups. Vital statistics birth data may be presented in terms of either maternal residence or place of birth. Resident data include all events that occur to residents of the Commonwealth, regardless of where they happen. In Massachusetts, a resident is a person with a permanent address in one of the 351 cities or towns. Occurrence data include all events that occur within the state, whether to residents or nonresidents. All data in this publication are for Massachusetts residents unless otherwise stated. There is an agreement among the 50 states, District of Columbia, Puerto Rico, Virgin Islands, Guam, and Canadian provinces that allows for the exchange of statistical copies of birth and death records for events occurring in a state other than the state of residence.

Massachusetts Community Health Information Profile (MassCHIP) - MassCHIP is an on-line information service provided by the Massachusetts Department of Public Health and available to all Massachusetts residents. MassCHIP was developed by the Massachusetts Department of Public Health to assist communities and professionals in health planning and is located online here. The website provides access to community-level data, collected on a variety of Massachusetts populations, for use in assessing community needs, monitoring health status, and implementing program evaluation. Among the types of data available are vital statistics, communicable disease statistics, socio-demographics, and calculated statistics and measures of utilization specific to MDPH programs. MassCHIP was used for producing the information on diabetes mortality. Information from the year 2008 was used and was based on the ICD-9CM code 250 or its ICD-10 equivalent.

Massachusetts Hospital Discharge Data Set - Hospital discharge data from the Division of Health Care Finance and Policy was used to obtain information regarding diabetes-related hospitalizations in Massachusetts. All acute care inpatient facilities in Massachusetts are required to submit hospital discharge data to the Division of Health Care Finance and Policy. Data are categorized by fiscal year and include patient-level information on demographics, diagnosis, length of stay, procedures, and charges. Up to 15 diagnoses (one primary and fourteen “associated” or secondary diagnoses) and 15 surgical procedures are recorded and coded according to the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9CM).  The primary diagnosis is the reason for the hospitalization. Only data where the primary diagnosis was diabetes (i.e. starting with 250) were used in the analysis for this website to show diabetes hospitalizations. In this dataset, we omitted fields that did not have more than seven occurrences from analysis. Age-adjustment for this data was done using 2009 population estimates and the 2000 population from the US Census Bureau.

Mortality Data - Mortality information was obtained from death certificate data and analyzed in MassCHIP. Causes of death are classified and coded according to the ICD-9 and ICD-10. ICD-10 codes were applied to the data from 1999 to the most current year while comparability ratios were applied to the data from 1994 through 1998. In this dataset, we omitted fields that did not have more than seven occurrences from analysis. Age-adjustment for this data was done using 2009 population estimates and the 2000 population from the US Census Bureau.

Small Area Estimates - In Massachusetts, the Behavioral Risk Factor Surveillance System (BRFSS) has been the primary source of population-based information for many health indicators. However, the BRFSS does not provide health statistics at the community level, because it is designed primarily for statewide or metropolitan area health statistics. Small area estimation (SAE) is a statistical technique that can be used to make prevalence estimates for small sub-populations when sample sizes for the areas in question are small. The methodology calls for using survey data (e.g., BRFSS) and combining it with other administrative data (e.g., US Census data). The combination of these data sources is intended to increase the precision of estimates for geographic areas where the number of survey respondents is too small to use normal direct estimation methods. Two reporting rules outline specific measures that the SAE must meet in order to be publicly reported, specifically precision and sample size:

1. The precision (1/2 width of 95% CI) for the SAE should not be greater than 2.5%-4.0% (depending on the prevalence estimate).
2. The minimum number of BRFSS respondents for SAE to be accepted is 25 if time period is 1 to 3 years and 40 for 5 years; the maximum aggregating time period should be 5 years.

If the SAE prevalence for a given community meets both reporting rules, then there are no restrictions applied to dissemination of the SAE and 95% CI for that community. If the SAE prevalence for a given community meets one but not both reporting rules, then the SAE and 95% CI may be reported at the discretion of the program but the following language must accompany the report: "In order to provide data for more Massachusetts communities, we include town level estimates that may be based on relatively few respondents or have standard errors that are larger than average. The confidence interval for this community is wider than the normal limits set by MDPH.  Therefore, the estimate for this town should be interpreted with caution."

Community estimates that do not meet the reporting requirements are not displayed on this site.