Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Some conceptual problems in multivariable analyses of perinatal mortality.

J L Kiely1

  • 1Gertrude H. Sergievsky Center, Faculty of Medicine, Columbia University, New York, New York.

Paediatric and Perinatal Epidemiology
|July 1, 1991
PubMed
Summary

Researchers frequently misuse multivariable analysis in maternal and child health studies. This review identifies common statistical errors and suggests improved methods for analyzing perinatal and infant mortality to ensure accurate causal inference.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Late-onset central hypoventilation syndrome: a family genetic study.

The European respiratory journal·2007
Same author

Intranasal corticosteroid therapy for obstructive sleep apnoea in patients with co-existing rhinitis.

Thorax·2003
Same author

Epidemiological trends in multiple births in the United States, 1971-1998.

Twin research : the official journal of the International Society for Twin Studies·2001
Same author

Maternal race/ethnicity and twinning rates in the United States, 1989-1991.

The Journal of reproductive medicine·2001
Same author

A comparison of recent trends in infant mortality among twins and singletons.

Paediatric and perinatal epidemiology·2001
Same author

Controlled oxygen therapy and carbon dioxide retention during exacerbations of chronic obstructive pulmonary disease.

Lancet (London, England)·2001

Area of Science:

  • Biostatistics
  • Epidemiology
  • Maternal and Child Health Research

Background:

  • Multivariable analysis is commonly used to study perinatal and infant mortality.
  • A review of 35 years of published research revealed recurring methodological issues.

Purpose of the Study:

  • To identify and address common problems in multivariable analyses of perinatal and infant mortality.
  • To propose solutions for more accurate causal inference in maternal and child health research.

Main Methods:

  • Systematic review of multivariable analyses in maternal and child health literature over 35 years.
  • Identification of four recurring problems in statistical analysis and variable handling.

Main Results:

Related Experiment Videos

  • Problem 1: Intervening variables incorrectly controlled as confounders.
  • Problem 2: Inadequate categorization of continuous variables like birthweight.
  • Problem 3: Neglect of statistical interactions between variables (e.g., social factors and birthweight).
  • Problem 4: Simultaneous entry of highly intercorrelated variables representing similar constructs.

    • Conclusions:

    • Current multivariable analyses often contain significant methodological flaws.
    • Improved analytical approaches incorporating subject-matter expertise are needed for valid confounding and causal inference.
    • Addressing these issues can lead to more accurate understanding of perinatal and infant mortality determinants.