Sodium Consumption Data Quality Challenge
by Guest Blogger, 7/9/2003
On May 14, 2003 the U.S Chamber of Commerce (USCC) and the Salt Institute filed a data quality petition with the National Heart, Lung and Blood Institute (NHLBI) within the National Institutes of Health (NIH). The petition challenges the objectivity, transparency and reproducibility of NHLBI information that claims or suggests that reduced sodium consumption will result in lower blood pressure in all individuals. The challenge further requests that NHLBI make additional information from a key study publicly available, claiming that the publicly released studies do not support such conclusions for “all” individuals. This request is not an accurate request for data correction and is another example of how the data quality guidelines create undue burden on agencies. Request for Correction The core of the data quality challenge revolves around a study entitled “Dietary Approaches to Stop Hypertension – Sodium Trial (DASH-Sodium)” that concluded a reduction in sodium intake below the current recommended level substantially lowered blood pressure, and these results should apply to most people in the United States. USCC and the Salt Institute assert that sufficient detailed data supporting the DASH-Sodium study was not released, making it impossible to evaluate the study’s conclusions or reproduce them. Interestingly, the petition does not question the data quality of the DASH-Sodium study itself, nor does it present any indications that the petitioners possess any evidence to refute the conclusions the study makes. Instead, the petitioners name six documents that they believe are in violation of the Data Quality Act and data quality guidelines produced by the Office of Management and Budget (OMB), Health and Human Services (HHS) and the NIH. The contested documents include two press releases, two fact sheets, a Journal of the American Medical Association article, and NHLBI’s “Seventh Report of the Joint national Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure.” The petition claims that each of these documents contains assertions that all Americans can reduce their blood pressure by limiting their sodium intake. The key claim within the petition is that the DASH-Sodium study, and by association the various documents specifically challenged, does not meet the data quality standard of objectivity or its underlying requirements of transparency and reproducibility. USCC and the Salt Institute claim that without the underlying data of the DASH-Sodium trial the public is unable to assess whether there is some reason to question the objectivity of the sources. The petition goes on to claim that the information listed in the petition represents “influential scientific” information and is therefore obligated to meet higher standards of transparency and reproducibility. However, the petition does not present too strong a case that the information is “influential,” instead implying that the information has the potential to be influential. Additionally, the petition utilizes a very narrow concept of reproducibility when it claims that for the DASH-Sodium conclusions to be reproducible all of the underlying subgroup data must be disclosed. This outlook ignores the possibility to the DASH-Sodium results could be considered reproducible if all of the study’s assumptions and methodology are presented clear enough that a third party could reproduce the results with an acceptable degree of imprecision. There are several problems with this data quality challenge. First, the petition does not appear to be a legitimate request for correction of data under the DQA. The USCC and the Salt Institute do not provide any evidence that information in the documents specified requires any correction, nor does the petition offer any proposed correction even though the data quality guidelines require such recommendations. The petition is not an effort to correct data but is actually a request for information using the DQA as an excuse. Second, the petition seems to be schizophrenic in its complaints. While all of the criticism and complaints center on the DASH-Sodium study, the petition does not list that study as the challenge data and instead lists various minor publications without even attempting to detail their data quality failings. OMB Watch supports access to government information and believes that the subgroup data of the DASH-Sodium study should be publicly available unless NHLBI can provide some legitimate reason for withholding the data. However, this challenge seems to be more a fishing expedition then a correction of data. The data quality guidelines have raised serious concerns about inappropriately diverting agency resources aware from protecting the public and industry attacking data to undercut policies that protect the public simply because they adversely affect profits. The USCC and Salt Institute should make their request for the DASH-Sodium data through more appropriate channels and stop forcing agencies to waste data quality resources. Agency Response In a response letter dated August 19, 2003, NIH denied the request for correction deeming the data quality process the incorrect method for the petitioners to acquire the information. Since the Chamber of Commerce and the Salt Institute are not seeking correction of specific information but looking to gain access to data, NIH feels the proper administrative mechanism is the Freedom of Information Act (FOIA). The grantee of the DASH-Sodium study has apparently previously released the information and is preparing a data set for public distribution, which would provide a much higher degree of accessibility for those interested in the data. The letter does go on to explain that if the challengers were requesting correction of the documents cited in the request for correction, they would satisfy NIH’s data quality guidelines and no correction would be warranted. The agency states that the challenged documents were rigorously and independently peer reviewed in addition to meeting NHLBI’s publications procedures. Furthermore, NHLBI used many sources, including the DASH-Sodium study, to make its recommendations on sodium intake. NIH also rejects the information as “influential” On September 3, 2003, NHLBI advised the Chamber of Commerce and the Salt Institute that is was treating the data quality petition as a FOIA request and then denied the request. The agency claimed that the data had been turned over to government-funded third-party investigators. The organizations could only obtain data that were produced under a separate grant. Salt Institute Appeal According to a September 22, 2003 appeal by the Chamber of Commerce and the Salt Institute, NHI’s reasons for denial are inadequate and it must release the requested information. The petitioners argue that the information NHLBI released based on the Sodium-DASH study violate the DQA because:
- The information is incomplete, and neither reproducible nor objective because the requested data is not available;
- Peer review is not adequate to demonstrate data quality reliability under the guidelines;
- NHLBI cannot rely on the totality of scientific evidence instead of only the DASH study because all evidence must adhere to the data quality guidelines; and
- The information is influential and must adhere to higher standards of quality.
- Assume jurisdiction of this case;
- Declare the NHLBI has violated the Data Quality Act;
- Order the agency to comply with the Data Quality Act, and to produce the mean blood pressures, standard deviations, and sample sizes of the requested subgroups on each of the three levels of sodium intake for both the control and the DASH Diet in a useable format;
- Order the agency to comply with the Data Quality Act in any dissemination of information pertaining to dietary salt;
- Order the agency to develop and make available procedures under the Shelby Amendment through which the public may be able to obtain all agency-funded data;
- Award Salt Institute its attorney fees and other expenses; and
- Grant such other and further relief as the Court deems just.