Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to site content Skip directly to page options
CDC Home

Spirometry Longitudinal Data Analysis (SPIROLA)

NIOSH Scientific Information Quality - Peer Review Agenda

Document

Spirometry Longitudinal Data Analysis (SPIROLA)

Related Materials

Peer Reviewers’ Comments [PDF - 1 MB]

NIOSH Response to Peer Reviewers’ Comments


NIOSH Response to SPIROLA Reviewer 1
  • We have sorted out minor bugs with data format that have been identified by reviewer 1. They were mainly related to his data input. The new version of SPIROLA uses free format and can detect the compatibility of user’s data to reduce error in data input.
  • SPIROLA provides information on the change in the rate of FEV1 decline with increasing duration of follow-up. From 4 years of follow-up, SPIROLA tracks the running change in the rate of decline based on all the data points, including the most recent one; this enables the physician to discern the change due to the last data point. From 8 years of follow-up, when the slope has relatively high precision, SPIROLA provides the running change in the slope for the 8 previous years and enables the physician to discern changes in the rate of decline due to the last 8 years. Based on the comment by reviewer 1, we have made changes to the software so that it provides this information only when there are at least three data points available for the calculation of the changes in the rate of decline.
  • SPIROLA monitors the overall precision of longitudinal spirometry data over time in a monitoring program using the pair-wise estimate of within-person variation. This measure is based on spirometry measurements repeated within 18 months of follow-up on a group of workers. Furthermore, this measure is used to estimate the limit of longitudinal decline appropriate for a specific program. To provide a measure of repeatability as suggested by reviewer 1, we would need to have datasets that include all spirometry measurements done within a single test in the database or that store results for the repeatability statistics for quality control as the OMI system now does. Most databases do not have this information. Nevertheless, based on the suggestion by reviewer 1, we are conducting a pilot study in the Phoenix Fire Department monitoring program to evaluate how well the American Thoracic Society repeatability criterion corresponds with the pair-wise within-person variance values. This study would allow us to establish how useful the repeatability criteria is and how does it relate to the measure of within-person variation that we currently use to set the limit of longitudinal decline.
  • To improve the reader’s understanding of the underlying theory used in deriving longitudinal limits of decline, we have created a HELP menu. This enables the reader to obtain pertinent information on the theory of the limit of longitudinal decline method and other methodological issues, and provides references.
  • The “Risk List” function enables the health care professional t identify and tag individuals who are at risk of developing respiratory disease. Because the time required for processing of the “Risk List” depends on the dataset size, based on the comment by reviewer1, we have added a running statistic that shows how many records have been processed so the user knows how long he still needs to wait.
  • SPIROLA now shows the number of repeated observations that were available for the calculation of pair-wise within-person variation shown by a chart that monitors data variability for the group. We display a message informing the user if the number of repeated observation is too small to obtain a reliable estimate of the within-person variation.
  • If the program within-person variation is in excess of 180 mL, which should be achievable in workplace monitoring programs, based upon previously published data, SPIROLA prints a message warning the user that the overall program data precision requires improvement.
  • With further software development, we plan to conduct research in the variability of FVC and add the same functions for FVC as we have for FEV1. Currently only the trend in FVC and percent predicted values are displayed.
  • Inclusion of other tests into SPIROLA could be done. We could, for specific users expand SPIROLA functions to include other tests. However, most users of SPIROLA have mainly spirometry data.
  • Based on the previous initiative by reviewer 1, we have created an option to include a set of reference equations that caters for specific Australian needs and includes an option to expand the reference equation by height squared and height cubed terms.
    We were able to address all the recommendations by reviewer 1.
NIOSH Response to SPIROLA Reviewer 2
  • We now provide theoretical justifications for the limit of longitudinal decline in the User Manual and in the Help menu. We will also make available through internet the relevant publications on which the theory is based. This will help the user to improve his understanding of the method. We also plan to write a manuscript on the application of SPIROLA when there is more widespread experience with the software.
  • We have created a Help menu that allows the user to go to specific sections on the manual using a content list. This Menu now provides all definitions that the reader needs. In future, we plan to create glossary wit definition terms for a faster reference.
  • Currently SPIROLA provides by default three reference equations based on NHANES data: for Caucasians, African-American, and Mexican-American. In addition, user can specify only one own equation. We will work with reviewer 2 to allow him to use SPIROLA so that he can specify several own reference equations and this way he will develop SPIROLA further based on the needs and experience of various health care professionals.
  • A program blog will be initiated to create a forum for sharing experience and improving the quality and utility of SPIROLA, if we establish that this is needed.

Required Elements for Initial Public Posting

Title: Spirometry Longitudinal Data Analysis (SPIROLA)

Subject: Computer software on the NIOSH Web site. SPIROLA is an integrated visual and quantitative tool.

Purpose: To aid qualified healthcare professionals who conduct workplace spirometry monitoring in maintaining acceptable sprirometry data precision and in identifying individuals who may be experiencing excessive lung function decline.

Type of Review: Individual

Timing of Review: Late 2006 to early 2007

Number of Reviewers: 2

Primary Disciplines or Expertise Needed for Review: Pulmonary medicine, workplace spirometry monitoring

Reviewers Selected by: NIOSH

Public Nominations Requested for Reviewers: NIOSH

Opportunities for the Public to Comment: Yes

Peer Reviewers Provided with Public Comments Before Their Review: No

Peer Reviewers:

Alan Crockett
Academic and Professional Credentials: PSM, MPH, PhD, FANZSRS
Organizational Affiliation: Associate Professor, Director, Primary Care Respiratory Unit School of Population Health and Clinical Practice, University of Adelaide, Australia
Areas of Expertise, Discipline, or Relevant Experience: Pulmonary medicine, workplace spirometry monitoring
Recommended by: NIOSH

Robert Dowdeswell
Academic and Professional Credentials: MD
Organizational Affiliation: Group Occupational Health Physician, Anglo Platinum, Rustenberg, South Africa
Areas of Expertise, Discipline, or Relevant Experience: Pulmonary medicine, workplace spirometry monitoring
Recommended by: NIOSH

Charge to Peer Reviewers:

SPIROLA reviewers were asked to provide technical review comments.

 
Contact Us:
  • Page last reviewed: December 9, 2007
  • Page last updated: December 9, 2009
USA.gov: The U.S. Government's Official Web PortalDepartment of Health and Human Services
Centers for Disease Control and Prevention   1600 Clifton Rd. Atlanta, GA 30329-4027, USA
800-CDC-INFO (800-232-4636) TTY: (888) 232-6348 - Contact CDC-INFO