ASPE comments NCI response

Date: March 31, 2014

TO: Office of Management and Budget (OMB)

Through: Keith Tucker, Reports Clearance Officer, HHS

Susan Queen, Director of Data Policy, HHS/ASPE

Seleda Perryman, Report Clearance Officer, NIH

Vivian Horovitch-Kelley, PRA OMB Clearance Liaison, NCI

FROM: Anne M. Hartman, Health Statistician, ARP/RFMMB

Division of Cancer Control and Population Sciences

National Cancer Institute

SUBJECT: Next Series of Tobacco Use Supplements to the Current Population Survey (TUS-CPS) (NCI), (OMB No. 0925-0368)

This memo contains NCI responses to ASPE’s review of the proposed collection of information “Next Series of Tobacco Use Supplements to the Current Population Survey (TUS-CPS) (NCI), (OMB No. 0925-0368).” ASPE commented on both Supporting Statement Part A, and the corresponding survey comparison Table. Below are ASPE’s comments, and NCI’s responses.

ASPE: Supporting Statement Part A: There are several references to the support for CDC’s National Adult Tobacco Survey (NATS) provided by FDA’s Center for Tobacco Products (CTP), stating that CTP has decided not to continue funding for the NATS after the 2013-2014 cycle. The supporting statement claims that, because of this, the survey is ending. Although the support from CTP will not continue, CDC has not confirmed that the NATS will not be fielded in future years, therefore, the supporting statement should be revised to omit this reference. (See attached Part A for changes on pages 5, 10, 12, and 14)

NCI RESPONSE: Thank you for your suggestion. The Supporting Statement A has been updated to omit the cited references. Additionally, in consultation with CDC, we (NCI, FDA) have further clarified the status of the NATS survey in Section A.4, where we provide a description of the NATS:

“….. FDA funded the 2012-13 and 2013-14 NATS as a partnership with CDC-OSH to collect time-critical data to monitor the impact of new regulatory actions. However, as part of an ongoing assessment of its tobacco-related surveillance and monitoring activities, FDA has decided not to continue funding the NATS beyond the 2013-14 wave.  At the present time, it is not known if CDC-OSH will continue to conduct the survey periodically to support their tobacco control program evaluation.”

ASPE: The A4 “Comparison of Key Features of Adult National Tobacco Surveys” table is particularly valuable for the review from OMB and complements the narrative from the supporting statement section on duplication; however, there are several issues that should be addressed or corrected. Incorrect Links: several of the links are not functional addresses. These are either old sites or incorrect ones.

• NHIS should be: http://www.cdc.gov/nchs/nhis/htm - this link appears to have an additional space in the address which prevents it from functioning.

• NATS should be: http://www.cdc.gov/tobacco/data_statistics/surveys/nats/index.htm

• NSDUH should add the SAMHSA link to the RTI link: http://www.samhsa.gov/data/NSDUH.aspx

• NHANES should be: http://www.cdc.gov/nchs/nhanes.htm The current link appears to be directed towards the Youth Tobacco Survey but has extra spaces embedded into the address and doesn’t work.

• PATH should be: http://www.fda.gov/TobaccoProducts/PublicHealthScienceResearch/ucm337005.htm The current link is directed toward the main page for FDA’s Tobacco Products and not to the survey.

NCI RESPONSE: Thank you for the corrections. We have incorporated the correct links into the A4 “Comparison of Key Features of Adult National Tobacco Surveys” table.

ASPE: The Comparison document did not include the Health Information National Trends Survey (HINTS) on the list in Tab A Overview of Surveys, even though the original spreadsheet that was sent with the supporting statement and other material contained a sheet with the HINTS question by question comparison to the other major HHS surveys. We recommend that HINTS be included in Tab A, Overview of Surveys.

NCI RESPONSE: We apologize for the confusion. Our table was adapted from the table submitted for the HINTS OMB package. Leaving in the sheet with the HINTS question by question comparison to the other major HHS surveys was an oversight on our part. In reference to including HINTS in the overview of surveys table, we agree with ASPE and now have added a row for HINTS in the table.

ASPE: Information for Column Headings: Please provide information on the following column headings, and describe the criteria that were used to categorize the surveys under each heading: Precision for non-oversampled subgroups; Policy and social norms. While the supporting statement has a good discussion of the information contained in Table 4, some additional detail would facilitate using the table for the survey comparisons.

For example, how was it determined that the NHIS, NHANES, and NSDUH surveys do not have precision for subgroups that weren’t oversampled under the heading “Precision for non-oversampled subgroups?” Oversampling is done in NHANES to increase the reliability and precision of estimates for the population subgroups that are of particular public health interest, but NHANES can be used to produce estimates of subgroups that weren’t oversampled. Researchers need to use the NHANES analytic guidelines to ensure that they take into account the proper use of the weights as well as the survey design, etc. Please explain why BRFSS is categorized as having precision for non-oversampled subgroups but NHIS, NHANES, and NSDUH do not have this precision according to the table.

NCI RESPONSE: We see how some of the table headings do not clearly demonstrate our intended purpose and thus for simplicity and clarity, we deleted the “Precision for non-oversampled subgroups,” which was to represent the ability to make even smaller subgroup comparisons based on larger sample size for TUS than other surveys, because we realized this should be apparent from the sample size already in the table and is better described already in the text of SSA, Section A.

ASPE: Policy and Social Norms: ASPE suggests that clarification is needed to the column with the heading “Policy and Social Norms,” as many of the surveys that are listed with “no” in this column have used “policy relevance” as a stated purpose for obtaining OMB approval. NSDUH is listed under this heading as not able to address policy issues and social norms. However, one of the purposes in the supporting statement for NSDUH is “to meet the needs of Federal policy makers and the substance abuse research, prevention, and treatment communities.” This is true for the NATS as well, and CDC justified their tobacco-related questions to OMB as “critical to FDA in monitoring the effectiveness and public perceptions of its regulatory activities.” (As stated earlier, although the CPS TUS supporting statement indicates that FDA is pulling support from NATS, CDC has not confirmed that the survey will be terminated and discontinued, therefore, the supporting statement should be revised to omit references to the discontinuation of NATS.) The NHIS has a stated purpose “for epidemiologic and policy analysis of time issues…” The information provided in this column contradicts the information that has been given to OMB to justify approval of these other information collection activities – one suggestion is to make this heading more specific, if possible. For example, NSDUH asks respondents about their perception of potential health risks related to smoking cigarettes, but does not have any questions in this area about other tobacco products, and there are no questions in the survey that ask about switching products. CPS TUS has questions on key policy areas that many of the other major surveys (except perhaps PATH and NATS) do not include are: flavored products, price, cessation, quit lines, etc. ASPE suggests that some clarity be provided in the headings to highlight key policy areas that are unique to CPS TUS.

NCI RESPONSE: As ASPE suggested that some clarity be provided in the headings to highlight key policy areas that are unique to TUS-CPS, we thought it would be most efficient to create sub-headings under the heading “Policy and social norms,” and further change the heading to read, “Policy and social norms/attitudes.” We created two sub-headings that more clearly demonstrate type of policies/social norms that are more unique to TUS-CPS (“Work and home smoke-free policies and attitudes,” and “Cigarette prices”). We also annotated some of the responses for clarity.

ASPE: Detailed Non-Cigarette Tobacco Product Use: The spreadsheet indicates that NSDUH does not contain detailed questions on non-cigarette tobacco use; however, NSDUH actually does have detailed questions on some of these products. There is a set of questions on snuff, including on age first used, past 30 day use, frequency of use, brand names, brand used most often, etc. The same series of questions is also asked separately for chewing tobacco and cigars.

ASPE: The spreadsheet also indicates that NHANES does not have detailed questions on non-cigarette tobacco use; however, NHANES has a set of questions on non-cigarette tobacco use, with separate questions on e-cigarettes, chewing tobacco, snuff, snus, and dissolvables. The focus in NHANES is on most recent use of these products as the questions are asked in the Mobile Examination Center (MEC) and will be analyzed in connection with the physical measures collected in the MEC. While not as detailed as the CPS TUS, it is important to acknowledge that these surveys do include such questions, since OMB has reviewed and approved their use and the table would suggest that such questions are only on the CPS TUS, NATS, and PATH.

NCI RESPONSE: Again, we took ASPE’s advice to be more specific in our headings in our attempt to highlight unique aspects of TUS-CPS. The table now has 4 sub-headings under the heading “Detailed non-cigarette tobacco product use.” We have added columns that specify those unique details, “Flavored Tobacco Product,” “Nicotine Dependence,” “Quit attempts,” and “Switching to quit tobacco product.”

ASPE: State Level Estimates: NSDUH is described as not able to produce state level estimates however, NSDUH is specifically designed to produce state estimates as well as national estimates, and can also be used for estimates on a limited number of areas within states. The sample is sufficient to support small area estimates in each state and the District of Columbia while maintaining efficiency for national estimates. Information to make the state estimates is not available on the public use file and has restricted access, but SAMHSA publishes a report annually with prevalence estimates and 95% confidence intervals by state. See: http://www.samhsa.gov/data/NSDUH/2k12State/NSDUHsae2012/Index.aspx

NCI RESPONSE: As we agree with ASPE that our “No” for NSDUH under the heading “State level estimates,” does not clearly describe our intent to distinguish flexibility/access to TUS state and sub-state identifiers compared with NSDUH state and sub-state identifiers, we have clarified that column for NSDUH response by saying the following: “Yes, mainly small area estimates reported for 3 items, limited state ID access in data files.”

2 3/31/2014