Background
Low-dose computed tomography (LDCT) lung cancer screening has been increasingly implemented in high-income countries among high-risk populations. Cost-effectiveness analyses (CEAs) are commonly used to inform policy decisions. However, it remains unclear how published CEAs evaluate alternative screening start and stop ages and whether they identify optimal age-range strategies through mutually exclusive comparisons.
Objectives
To assess how informative cost-effectiveness analyses (CEAs) of low-dose computed tomography (LDCT) lung cancer screening are regarding optimal screening ages.
Methods
PubMed, Web of Science and Embase were searched for simulation-based CEAs of LDCT lung cancer screening. We determined whether each study simulated single or multiple birth cohorts; whether reported results were aggregated across cohorts or not when applying multiple cohorts; whether alternative screening age ranges were considered; and whether alternative age ranges were evaluated as separate subgroup analyses or compared as mutually exclusive competing strategies.
Results
This review retrieved 68 studies. Figure 1 illustrates age ranges distribution and how they were compared among 66 studies clearly report screening age ranges. Panel A shows the age ranges of the 31 studies that did not consider alternative age ranges.
Panels B and C report 35 studies assessing multiple age ranges. However, over half (19/35, Panel B) did not make mutually exclusive comparisons of strategies, which is necessary when seeking to determine optimally cost-effective policies.
Panel C illustrates 16 studies with mutually exclusive age range comparisons. Among them, 6 simulated a single birth cohort. For the remaining 10 studies that used multi-cohorts or were unclear about cohort inclusion, all reported outcomes aggregated across cohorts rather than reporting cohort-specific cost-effectiveness estimates.
Conclusions
Most CEAs reviewed were not adequate to clearly inform policy regarding the optimal age ranges in lung cancer screening. Some simply did not consider alternative screen age ranges, others did not make necessary comparisons. Other strategy comparisons were compromised by aggregation of outcomes across cohorts, obscuring potentially policy-relevant variation between younger and older cohorts. There is clear scope for greater clarity regarding optimal screening start and stop ages in lung cancer screening. Future studies should clearly specify birth cohort inclusion, report disaggregated outcomes when using multi-cohort models, and evaluate alternative age ranges as mutually exclusive competing strategies.
Low-dose computed tomography (LDCT) lung cancer screening has been increasingly implemented in high-income countries among high-risk populations. Cost-effectiveness analyses (CEAs) are commonly used to inform policy decisions. However, it remains unclear how published CEAs evaluate alternative screening start and stop ages and whether they identify optimal age-range strategies through mutually exclusive comparisons.
Objectives
To assess how informative cost-effectiveness analyses (CEAs) of low-dose computed tomography (LDCT) lung cancer screening are regarding optimal screening ages.
Methods
PubMed, Web of Science and Embase were searched for simulation-based CEAs of LDCT lung cancer screening. We determined whether each study simulated single or multiple birth cohorts; whether reported results were aggregated across cohorts or not when applying multiple cohorts; whether alternative screening age ranges were considered; and whether alternative age ranges were evaluated as separate subgroup analyses or compared as mutually exclusive competing strategies.
Results
This review retrieved 68 studies. Figure 1 illustrates age ranges distribution and how they were compared among 66 studies clearly report screening age ranges. Panel A shows the age ranges of the 31 studies that did not consider alternative age ranges.
Panels B and C report 35 studies assessing multiple age ranges. However, over half (19/35, Panel B) did not make mutually exclusive comparisons of strategies, which is necessary when seeking to determine optimally cost-effective policies.
Panel C illustrates 16 studies with mutually exclusive age range comparisons. Among them, 6 simulated a single birth cohort. For the remaining 10 studies that used multi-cohorts or were unclear about cohort inclusion, all reported outcomes aggregated across cohorts rather than reporting cohort-specific cost-effectiveness estimates.
Conclusions
Most CEAs reviewed were not adequate to clearly inform policy regarding the optimal age ranges in lung cancer screening. Some simply did not consider alternative screen age ranges, others did not make necessary comparisons. Other strategy comparisons were compromised by aggregation of outcomes across cohorts, obscuring potentially policy-relevant variation between younger and older cohorts. There is clear scope for greater clarity regarding optimal screening start and stop ages in lung cancer screening. Future studies should clearly specify birth cohort inclusion, report disaggregated outcomes when using multi-cohort models, and evaluate alternative age ranges as mutually exclusive competing strategies.
Screening age range comparisons across studies