Related Experiment Videos
Are standardized mortality ratios valid for public health data analysis?
1Expert Health Data Programming Inc., Austin, Texas 78756, USA. dgoldman@ehdp.com
Statistics in Medicine
|May 3, 2000
Summary
Standardized mortality ratios (SMRs) are as valid as standardized rate ratios (SRRs) for public health analysis. Simulations show both methods yield similar conclusions for comparing mortality across geographic areas.
Area of Science:
- Epidemiology
- Public Health
- Biostatistics
Background:
- Standardized mortality ratios (SMRs) face criticism regarding their validity in public health data analysis.
- Recommendations suggest using standardized rate ratios (SRRs) as a more valid alternative.
- Existing literature and textbooks show disagreement on SMR validity and lack supporting data.
Purpose of the Study:
- To rigorously assess the validity of SMRs in public health data analysis.
- To compare the performance of SMRs and SRRs in measuring and ranking mortality.
- To provide empirical evidence for the utility of SMRs versus SRRs.
Main Methods:
- Conducted simulations using diverse study population age distributions and disease rates.
- Compared SMRs and SRRs as absolute measures of increased mortality.
- Evaluated the impact of changing age-adjustment standards on both SMRs and SRRs.
- Assessed the consistency of population rankings by SMRs and SRRs.
Main Results:
- SMRs exhibited minor changes (6-8%) with shifts in age distribution (young to old).
- SRRs showed comparable changes (4-5%) when the age-adjustment standard changed (1940 to 1990 U.S. Census).
- County mortality rankings by SRR were only slightly more consistent than those by SMR.
Conclusions:
- SMRs demonstrate similar usefulness to SRRs in public health data analysis.
- Both SMRs and SRRs lead to comparable conclusions when analyzing mortality data.
- SMRs remain a viable tool for comparing mortality across different geographic areas.