21 December 2022 in EBioMedicine
Background Stratification of patients with post-acute sequelae of SARS-CoV-2 infection (PASC, or long COVID) would allow precision clinical management strategies. However, long COVID is incompletely understood and characterised by a wide range of manifestations that are difficult to analyse computationally. Additionally, the generalisability of machine learning classification of COVID-19 clinical outcomes has rarely been tested. Methods We present a method for computationally modelling PASC phenotype data based on electronic healthcare records (EHRs) and for assessing pairwise phenotypic similarity between patients using semantic similarity. Our approach defines a nonlinear similarity function that maps from a feature space of phenotypic abnormalities to a matrix of pairwise patient similarity that can be clustered using unsupervised machine learning. Findings We found six clusters of PASC patients, each with distinct profiles of phenotypic abnormalities, including clusters with distinct pulmonary, neuropsychiatric, and cardiovascular abnormalities, and a cluster associated with broad, severe manifestations and increased mortality. There was significant association of cluster membership with a range of pre-existing conditions and measures of severity during acute COVID-19. We assigned new patients from other healthcare centres to clusters by maximum semantic similarity to the original patients, and showed that the clusters were generalisable across different hospital systems. The increased mortality rate originally identified in one cluster was consistently observed in patients assigned to that cluster in other hospital systems. Interpretation Semantic phenotypic clustering provides a foundation for assigning patients to stratified subgroups for natural history or therapy studies on PASC. Funding NIH (TR002306/OT2HL161847-01/OD011883/HG010860), U.S.D.O.E. (DE-AC02-05CH11231), Donald A. Roux Family Fund at Jackson Laboratory, Marsico Family at CU Anschutz.
1 January 2021 in eClinicalMedicine
While household contact investigation is widely recommended as a means to reduce the burden of tuberculosis (TB) among children, only 27% of eligible pediatric household contacts globally received preventive treatment in 2018. We assessed the cost-effectiveness of household contact investigation for TB treatment and short-course preventive therapy provision for children under 15 years old across 12 high TB burden countries. We used decision analysis to compare the costs and estimated effectiveness of three intervention scenarios: (a) status quo (existing levels of coverage with isoniazid preventive therapy), (b) contact investigation with treatment of active TB but no additional preventive therapy, and (c) contact investigation with TB treatment and provision of short-course preventive therapy. Using country-specific demographic, epidemiological and cost data from the literature, we estimated annual costs (in 2018 USD) and the number of TB cases and deaths averted across 12 countries. Incremental cost effectiveness ratios were assessed as cost per death and per disability-adjusted life year [DALY] averted. Our model estimates that contact investigation with treatment of active TB and provision of preventive therapy could be highly cost-effective compared to the status quo (ranging from $100 per DALY averted in Malawi to $1,600 in Brazil; weighted average $383 per DALY averted [uncertainty range: $248 – $1,130]) and preferred to contact investigation without preventive therapy (weighted average $751 per DALY averted [uncertainty range: $250 – $1,306]). Key drivers of cost-effectiveness were TB prevalence, sensitivity of TB diagnosis, case fatality for untreated TB, and cost of household screening. Based on this modeling analysis of available published data, household contact investigation with provision of short-course preventive therapy for TB has a value-for-money profile that compares favorably with other interventions. Unitaid (2017–20-IMPAACT4TB).