Tracking Healthy People 2010 Objectives: Methodological and Definitional Changes Kathleen M. Turczyn, MPH National Center for Health Statistics

Changes in tracking data can be due to: Coding classifications (ICD-9 to ICD-10) Revised OMB standards (1997) for race and ethnicity categories Age-adjustment groups Data source changes including question wording Error in original measure Preliminary analysis made final New science

Three examples of changes in methodology and definitions Objective 22-2 and LHI: Moderate physical activity among adults (NHIS) Objective 3-9: Use of sun- protective behaviors among adults (NHIS) Objective 5-12: Glycosylated hemoglobin measurement among adults with diabetes (BRFSS)

Objective 22-2 Moderate physical activity Increase the proportion of adults who engage regularly, preferably daily, in moderate physical activity for at least 30 minutes per day.

Objective 22-2 Measurement issues Change in NHIS questions asked between HP2000 and HP2010 HP2010 definition needed to be re-conceptualized (regular activity) Subsequent change in questions used for tracking HP2010 objective

HP2000 and HP2010 definitions are similar:  Frequency of moderate activity is 5 or more times a week  Duration is 30 or more minutes each time Healthy People definitions of moderate physical activity

HP2000 physical activity Periodic NHIS supplements ( ) List of leisure-time physical activities Continuum of activities No activity Moderate activity Vigorous activity

HP2010 physical activity Uses NHIS redesigned questionnaire core (1997+) Separate questions for moderate and vigorous activity Original baseline used 2 core questions:  Frequency of moderate activity  Duration of moderate activity

How often do you do light or moderate activities for at least 10 minutes that cause only light sweating or a slight to moderate increase in breathing or heart rate?  Never  Unable to do  ____ times per day/week/month/year  About how long do you do these light or moderate activities each time? _____ minutes/hours Original HP2010 questions used

Differences in estimates of moderate physical activity Definition HP2000 (from list) ---30% Original HP2010 (from mod q’s) 15%13%

Use both moderate and vigorous responses that meet criteria (closer to intent of objective – i.e., regular activity) Use 4 questions:  Frequency and duration of vigorous activity  Frequency and duration of moderate activity New recommendation for HP2010 definition of moderate physical activity

How often do you do vigorous activities for at least 10 minutes that cause heavy sweating or large increases in breathing or heart rate?  Never  Unable to do  ____ times per day/week/month/year  About how long do you do these vigorous activities each time? _____ minutes/hours Vigorous physical activity questions

Evolving estimates of HP2010 moderate physical activity Definition HP2000 (activities list)---30% Original HP2010 (2 moderate questions) 15%13% Revised HP2010 (4 vigorous + moderate questions) 32%30%

Light or moderate physical activity for at least 30 minutes 5 or more times a week, OR vigorous physical activity for at least 20 minutes 3 or more times a week Revised HP2010 operational definition of moderate physical activity

Final resolution for adult moderate physical activity objective No change in the wording of the objective New baseline: 32% in 1997 New target: 50%

Objective 3-9 Sun exposure and skin cancer Increase the proportion of persons who use at least one of the following protective measures that may reduce the risk of skin cancer:  avoid the sun between 10 am and 4 pm  wear sun-protective clothing when exposed to sunlight  use sunscreen with a sun protective factor (SPF) of 15 or higher  avoid artificial sources of ultraviolet light

Objective 3-9b Sun exposure and skin cancer among adults Increase the proportion of adults aged 18 years and older who follow protective measures that may reduce the risk of skin cancer.

Objective 3-9b Measurement issues 1998 baseline questions differ from 2000 and 2003 questions To use or not to use new 2000 data on respondents who offer that they “do not go out in the sun” in the analysis To use or not to use the SPF question added to the 2000 and 2003 questionnaires

If you were to go outside on a very sunny day for MORE than one hour, are you very likely, somewhat likely, or unlikely to…  wear protective clothing such as wide brimmed hats or long sleeved shirts?  avoid the sun by staying in the shade?  use sunscreen or sun block lotion? Questions used for 1998 baseline

When you go outside on a very sunny day for more than one hour, how often do you…  stay in the shade?  wear a hat that shades your ears, face, and neck?  wear a long sleeved shirt?  wear sunscreen? [Response choices read after each question: always; most of the time; sometimes; rarely; never. “Don’t go out in the sun” was also coded as a response, but not asked.] Questions used for 2000 and 2003 updates

Why changes in question wording? Questions and response choices should relate to the frequency of real behavior and not to a hypothetical likelihood of behavior A 5-point scale is generally considered to be better than a 3-point scale

People who don’t go out in the sun (DGOITS) The response “I don’t go out in the sun” to the sun protective behavior questions was recorded and coded but not specifically asked for in 2000 How to use this information?  Are DGOITS at risk but engaging in a protective behavior? (radical sun protectionists)  Or are they not at risk because they spend their lives indoors? (not applicable subgroup) Resolution: Exclude DGOITS from analysis (don’t know reason for behavior)

Use SPF question? HP2010 objective wording includes the words “use sunscreen with a sun protective factor (SPF) of 15 or higher” 1998 baseline did not include SPF numbers in the analysis (question was not asked) In 2000, if respondents said they used sunscreen “all” or “most of the time,” they got a follow-up question:  What is the SPF number of the sunscreen you use MOST often? Resolution: Use the SPF question

Year DGOITS in numer- ator? DGOITS in deno- minator? SPF used ? Estimate % 2000Yes No62% 2000NoYesNo60% 2000NoYes 55% 2000No Yes Evolving estimates of people reporting sun protective behaviors 58%

Final resolution for objective 3-9b: Sun protection behaviors Change the baseline data from 1998 to 2000 New baseline is 58% Target remains the same at 75%

Objective 5-12 Glycosylated hemoglobin measurement among adults with diabetes Increase the proportion of adults with diabetes who have a glycosylated hemoglobin measurement at least once a year.

Objective 5-12 Measurement issues 1998 baseline; BRFSS questions changed in 2000 Substantial, real increase in people with diabetes getting A1C tests To include or exclude “Never heard of A1C test” in the denominator in 2000

Numerator:  Yes to “ever heard of glycosylated hemoglobin or hemoglobin A1C”  Had seen a health professional about diabetes one or more times in the last year  Had been checked by health professional for glycosylated hemoglobin A1C one or more times in the last year Denominator:  Ever been told by a doctor that you have diabetes (for females, exclude if told only when pregnant) 1998 baseline questions

Numerator:  Had seen a health professional about diabetes one or more times in the last year  “A test for hemoglobin A1C measures the average level of blood sugar over the past three months. About how many times in the past 12 months has a doctor, nurse, or other health professional checked you for hemoglobin A1C?”  Answer categories include “Never heard of A1C.” Denominator: Ever been told by a doctor that you have diabetes (for females, exclude if told only when pregnant) 2000 questions

Why changes in question wording? 1998 questions were confusing to respondents -- too many terms Public health educators wanted people with diabetes to know specifically about the A1C lab test Funding issue: The change shortens the questionnaire from 3 to 2 questions

A short evolution of the estimates Year Standard of care for testing Number of annual Hemoglobin A1C tests Incl “Never heard of A1C” in denominator Estimate 1998 Type I only1+24% 2000All Types1+88% 2000All Types1+85% 2000All Types 2+ (new guidelines)63% *Respondents who had “Never heard of A1C” were not asked about tests. *

Final resolution for objective 5-12: Glycosylated hemoglobin measurement Change the wording of the objective from “one or more” to “two or more…” Possible other wording changes Change the baseline year from 1998 to 2000 New baseline is 63% New target has yet to be determined (original target is 50%)

Websites NCHS Healthy People Homepage: DHHS Healthy People Homepage: DATA2010: wonder.cdc.gov/data2010