Standardized TB Screening for People Living with HIV
Author Information
Author(s): Getahun Haileyesus, Kittikraisak Wanitchaya, Heilig Charles M., Corbett Elizabeth L., Ayles Helen, Cain Kevin P., Grant Alison D., Churchyard Gavin J., Kimerling Michael, Shah Sarita, Lawn Stephen D., Wood Robin, Maartens Gary, Granich Reuben, Date Anand A., Varma Jay K.
Primary Institution: World Health Organization
Hypothesis
Can a simple, standardized tuberculosis (TB) screening rule effectively identify people living with HIV who need further investigation for TB disease in resource-constrained settings?
Conclusion
A simplified screening rule using any one of the symptoms of cough, fever, night sweats, or weight loss can effectively identify people living with HIV who need further diagnostic assessment for TB.
Supporting Evidence
- The overall prevalence of TB disease was found to be 5.8% among the participants.
- The best performing screening rule was the presence of any one of the symptoms: cough, fever, night sweats, or weight loss.
- The sensitivity of the proposed screening rule was 78.9% with a specificity of 49.6%.
- Sensitivity increased to 90.1% among participants selected from clinical settings.
Takeaway
This study found that asking about symptoms like cough and fever can help doctors figure out who might have tuberculosis (TB) when treating people with HIV, especially in places where resources are limited.
Methodology
The study conducted a meta-analysis of 12 observational studies, evaluating the screening performance of various combinations of symptoms in diagnosing TB among people living with HIV.
Potential Biases
Potential biases may arise from the variability in how symptoms were reported and the differences in study settings.
Limitations
The study's findings may not be generalizable to all populations, as it primarily included participants from sub-Saharan Africa and Southeast Asia.
Participant Demographics
The median age of participants was 34 years, with a majority from sub-Saharan Africa.
Statistical Information
P-Value
0.001
Confidence Interval
95% CI 58.3%–90.9%
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website