Predicting adverse outcome from lower respiratory tract infection in primary care: The 3C cohort study of LRTI in primary care

Talk Code: 
1B.1
Presenter: 
Michael Moore
Co-authors: 
Stuart B, Smith S, Thompson MJ, Knox K, van den Bruel A, Lown, M, Little P, Mant D
Author institutions: 
University of Southampton, University of Oxford, University of Washington

Problem

Lower respiratory tract infection (LRTI) is one of the commonest acute infections presenting in primary care. Antibiotics are frequently prescribed although there is little evidence to support symptomatic benefit. One of the drivers of continued prescribing is concern over adverse outcome and hospital admission.

Approach

Adult patients presenting in UK general practice with lower respiratory tract infections had symptoms signs and treatment recorded. Participants were followed-up for 30 days to determine clinical outcome including admissions and deaths. All admissions were reviewed and a decision regarding whether potentially attributable to the LRTI illness. The predictive value of patient characteristics, presenting symptoms, and clinical findings for admission or death was assessed but non-attributable admissions eg elective surgery were not included.

Findings

A cohort of 28867 adult patients with acute cough was recruited with informed consent by 522 practices between October 2009 and April 2013. There were a total of 258 hospitalisations recorded. There were 30 deaths, of which 15 were to patients who were not hospitalised. 171 of the 258 hospital admissions were for conditions possibly related to the LRTI and 87 were not. Of the 171, 63 were coded LRTI, 56 coded COPD, and 2 were coded as adverse reactions to antibiotics; the other 50 were either other respiratory conditions (e.g. measles, pulmonary fibrosis, asthma) or issues that might have arisen because of LRTI (e.g. septicaemia, hypotension). Preliminary analysis suggests there are 10 variables that predict hospitalisation or death with a RR of 1.5 or higher: age 60+, comorbidity, shortness of breath, chest pain, crackles, higher severity score, high pulse, high temperature, low oxygen saturation and low blood pressure. These 10 items can be combined into a total score which ranges from 0 (none of these) to 10 (all of these). The AUC of this score is 0.73 (Bootstrapped 95% CI 0.70, 0.76).

Consequences

Hospitalisation and death is uncommon following LRTI presentation in primary care. The prediction model shares many features of that predicting pneumonic infiltrates. The implications of the model and its clinical utility for predicting adverse outcomes will be discussed.

Submitted by: 
Michael Moore
Funding acknowledgement: 
NIHR Programme Grant