Att. L. Response to Annie E. Casey Found. Letter

DEPARTMENT OF HEALTH & HUMAN SERVICES Public Health Service

Centers for Disease Control and Prevention

National Center for Health Statistics

3311 Toledo Road, Room 7204

Hyattsville, Maryland 20782

January 5, 2009

Laura A. Beavers, Senior Associate

National KIDS COUNT Coordinator

The Annie E. Casey Foundation

Baltimore, MD

Dear Laura:

This letter is in response to your public comment to the 60-day Federal register notice for Vital Statistics Report Forms, OMB No. 0920-0213. Thank you for your comments.

The proposal in this Federal Register notice is to continue to collect monthly record counts of vital events (births, deaths, infant deaths, marriages and divorces) from the registration areas. I believe your comments address more the viability of the National Vital Statistics System than the specifics of the 60-day notice. Because of the importance of your comments, however, the NCHS staff and I are taking this opportunity to provide our comments on your letter.

Budget Background

As with many Federal Programs, NCHS is experiencing severe budget shortfalls. For the past several years, NCHS budgets have not kept pace with rising data collection costs. We have been making program changes that have allowed us to keep most systems in the field; however, this approach is no longer viable from both a budget and data quality perspective. More significant cuts are needed. For FY2009 we have taken steps to reduce the current sample size of the National Health Interview Survey (NHIS) by 50 percent and without sufficient budget in FY2010 further reductions might be taken including suspending data collection. Although the sample for NHANES will not be cut in Calendar Year (CY) 2009, if NCHS cannot meet NHANES cost increases for FY2010, our plan is to either suspend data collection for CY2010 or reduce the sample size to half the design for CY2010. For our health care surveys, we will delay implementation of the National Nursing Home Survey for an indeterminate period and postpone fielding of the National Hospital Discharge Survey (NHDS) with the potential of no data collection in CY2009 and/or CY2010.

The budget situation for the Vital Statistics System continues to be problematic. Although NCHS has the authority to collect vital statistics from the States and US Territories, States are not required to provide these data to NCHS. Because of this, we enter into contracts with each of the States and US Territories to pay for their data. It currently costs NCHS approximately $19 million for 12 months of data and this does not include NCHS costs to process, code, conduct quality assurance and disseminate data. For several years, NCHS has not had adequate funds to pay for all 12 months of data, which has meant that we have had to spread our payments over two fiscal years and this in turn has affected the availability and quality of vital statistics. (The effect on quality is due to the increased time between the delivery of the data and the event, limiting investigation of the data quality.) In addition, many states have not been able to change to the 2003 revision of the birth and death certificates. There are many items on the new certificate that are not comparable with how the information was collected on the old certificate. The result is that, for many items, we cannot combine data across states. In addition, the number of states using the new certificate changes each year, making time trends very difficult if not impossible. Many of these items (such as prenatal care, smoking and education) are included in the enhanced item set. Essentially, NCHS maintains two systems each for both births and deaths, which results in added costs.

We recognize that limiting NCHS’s purchase to only core items could affect registration areas’ collection of items in the enhanced set. We believe this problem will be limited somewhat because of the recognition of the usefulness of these items to the states themselves.

The comparison to the marriage and divorce systems is not relevant here. There was never a complete marriage and divorce system in the US (only 42 states participated in the marriage system and 31 in the divorce system when it was discontinued in 1990). In 2003, NCHS estimated that reinstating a complete marriage and divorce system would cost $8.75 million per year for six years with all costs paid by the federal government as many states have no need for such a system at the state level.

Plan of Action

We are now in the process of negotiating new five-year contracts with the States, and based on our budget situation, we felt it prudent to have in those new contracts options other than delaying payments on the entire data set.  These new contracts will contain more deliverables allowing us to selectively pay for all or portions of the data. Fetal deaths and the linked birth and infant death files will be separate deliverables and most significantly, the birth and death files will each be split into two parts (core and enhanced) and each part will be a deliverable.

Our plan is to purchase a "core" set of items which will be consistent across states on the birth and death certificates from the States. This allows us to meet our primary commitment

to obtain twelve months of national data on the most central items on birth and death certificates and invest in the infrastructure required to improve quality and timeliness of vital statistics data. Any savings will be invested back into the system to assist all states in adopting the new certificates on reengineered systems. Should funding become available from other sources, we would purchase part or all of the remainder of the certificate data.

Summary

We share your concerns as well as those expressed by other organizations. However, if we continue with past practices, the system will become less and less timely and data quality will suffer even more. Our aim is to develop a high quality, timely, supportable system which will meet the data needs of as many users of the Vital Statistics System as possible.

Sincerely,

/Edward J. Sondik, Ph.D./

Edward J. Sondik, Ph.D.

Director

National Center for Health Statistics