Positive Predictive Value: A Clinician's Guide to Avoid Misinterpreting the Results of Screening Tests

Area of Science:

• Psychiatry
• Epidemiology
• Biostatistics

Background:

• Self-administered screening scales are frequently used in research to identify potential cases of various disorders.
• However, the statistical validity of conclusions drawn from these studies is often questionable due to a lack of consideration for key statistical principles.

Purpose of the Study:

• To review the principles and statistics of diagnostic screening.
• To differentiate screening from case-finding.
• To highlight studies that have overlooked the positive predictive value (PPV) of screening scales.

Main Methods:

• Review of diagnostic screening principles and statistics.
• Analysis of studies that employed screening scales as case-finding instruments.
• Examination of the impact of low positive predictive value on research conclusions.

Main Results:

• Many studies use screening scales as proxies for diagnosis without assessing their PPV.
• In low-prevalence disorders, screening scales often have insufficient PPV to reliably identify true cases.
• Examples include studies on borderline personality disorder and bipolar disorder where PPV was critically low, leading to potentially erroneous conclusions.

Conclusions:

• Conclusions from studies using screening scales without considering PPV are often not justified.
• Researchers must address the PPV of screening measures, particularly in low-prevalence populations.
• Accurate case-finding requires a nuanced understanding of screening statistics beyond simple scale administration.