The following is a summary of “Influence of initial misdiagnosis on mortality in patients with bacteraemia: propensity score matching and propensity score weighting analyses,” published in the April 2024 issue of Infectious Diseases by Eikenboom et al.

Diagnostic errors are common in infectious diseases due to their complex nature, and they can be harmful to patients who are infected. 

Researchers conducted a retrospective study using propensity scores to evaluate the impact of initial misdiagnosis of the infection source on 14-day mortality in patients with bacteremia. 

They utilized data from the Leiden University Medical Centre (LUMC) patient cohort with mono-bacterial bloodstream infections (BSI) (2013 to 2015). Propensity score matching and inverse probability of treatment weighting (IPTW) corrected for confounding. The average effect of initial misdiagnosis on the misdiagnosed (AEMM) was estimated. Addressing methodological issues, sensitivity analyses adjusted caliper size in propensity score matching and used various truncated weights in IPTW.

The results showed that out of 887 patients, 9.9% (n=80) had high propensity scores (>0.95), which ranged between 0.015 and 0.999. In matched analysis, 20.5% (n=35) misdiagnosed died within 14 days, compared to 5.8% (n=10) correctly diagnosed, making a difference of 14.6% (7.6%; 21.6%). Propensity score reweighting estimated that 19.8% of misdiagnosed patients would die within 14 days versus 6.5% if correctly diagnosed (difference 13.3% (95% CI 6.9%; 19.6%)). Adjustment for all variables showing disbalance showed a difference of 13.7% (7.4%; 19.9%). Sensitivity analyses produced similar results, but untruncated weighted analyses were unstable.

Investigators concluded that 14-day mortality in patients who were initially misdiagnosed showed a significant increase. Also, some patients had nearly one propensity score, indicating high chances of initial misdiagnosis. 

Source: bmcinfectdis.biomedcentral.com/articles/10.1186/s12879-024-09299-9