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

Statistics for clinicians. 3. Nominal data (II)

A S Nanivadekar1, A R Kannappan

  • 1Pfizer Limited, Bombay.

The Journal of the Association of Physicians of India
|February 1, 1991
PubMed
Summary
This summary is machine-generated.

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

Effect of physiological doses of oral vitamin B12 on plasma homocysteine: a randomized, placebo-controlled, double-blind trial in India.

European journal of clinical nutrition·2010
Same author

Ineffectiveness of iron polymaltose in treatment of iron deficiency anemia.

The Journal of the Association of Physicians of India·2004
Same author

Prazosin GITS vs sustained release nifedipine in patients with hypertension and abnormal lipid profile: a randomized, controlled, multicenter study. Madras Hypertension Study Group.

The Journal of the Association of Physicians of India·2001
Same author

Prazosin GITS vs atenolol in patients with hypertension and normal lipid profile: a randomized, controlled multicenter study. Hyderabad Hypertension Study Group.

The Journal of the Association of Physicians of India·2001
Same author

A randomized, controlled, multicenter study to compare prazosin GITS with enalapril in hypertensive patients with diabetes mellitus. Bombay Hypertension Study Group.

The Journal of the Association of Physicians of India·2001
Same author

Effect of prazosin GITS, atenolol, nifedipine SR, and enalapril on ADP-induced platelet aggregation.

The Journal of the Association of Physicians of India·2001

Standard statistical tests may miss trends in event occurrence. Specialized methods like the z-test for linear trend and life table analysis (logrank/Mantel-Haenszel) are crucial for accurate significance and timing of events. Predictive value of diagnostic tests also depends on prevalence.

Area of Science:

  • Biostatistics
  • Epidemiology
  • Clinical Research Methodology

Background:

  • Standard chi-square tests may fail to detect linear trends in event occurrence across groups.
  • Individual studies might show non-significant trends, necessitating pooled analysis.
  • Assessing the timing of events (onset or delay) requires methods beyond simple frequency counts.

Purpose of the Study:

  • To highlight appropriate statistical methods for analyzing event occurrence trends.
  • To introduce advanced techniques for analyzing pooled study data and event timing.
  • To emphasize the importance of predictive value in diagnostic test evaluation.

Main Methods:

  • Z-test for linear trend to analyze significance of trends in event percentages.
  • Pooled analysis using z-test for linear trend across multiple studies.

Related Experiment Videos

  • Life table method (logrank or Mantel-Haenszel) for analyzing event occurrence over time.
  • Consideration of predictive value alongside sensitivity and specificity for diagnostic tests.

    • Main Results:

    • The z-test for linear trend can reveal significance missed by standard chi-square tests.
    • Pooled analysis enhances the power to detect trends when individual studies are not significant.
    • Life table methods provide insights into event timing (time-to-event analysis).
    • Predictive value is a critical, prevalence-dependent metric for diagnostic tests.

    Conclusions:

    • Employing the z-test for linear trend is essential when analyzing event occurrence with linear trends.
    • The life table method is vital for studies requiring analysis of event timing.
    • Pooled analysis of multiple studies can strengthen trend detection.
    • Clinicians must consider predictive value, influenced by prevalence, for effective diagnostic test interpretation.