BACKGROUND: Lung cancer is the most common cause of cancer death in the UK, and incidence is strongly associated with increasing age. Screening can improve survival by detecting cancer earlier. Targeted Lung Health Check (TLHC) programme is a lung cancer screening pilot offered to smokers or ex-smokers aged 55-74 years in certain areas of England. However, uptake remains as low as 20% in London. We did a service evaluation to assess the effect of adding messaging informed by behaviour science to invitation letters on TLHC uptake. METHODS: The intervention involved adding to the standard invitation letter a box with a two-sentence behavioural message targeted at mitigating fatalistic beliefs and low risk perception, as well as promoting clinician endorsement and the value of early detection. Allocation to receive either standard or intervention letters was done at West London GP practice level, with approximate pairing between practices. From Sept 22, 2022, to Nov 28, 2022, 9464 invitations were sent to eligible individuals (4842 standard and 4622 intervention) covering 20 GP practices. The primary outcome (uptake) was a booked appointment date within 4 months of the invitation letter being sent. TIDieR checklist was used for methodology and reporting guidelines. We used χ2 analysis to test for significant differences in uptake, and mixed-effects logistic regression to control for demographic covariates. FINDINGS: Demographic characteristics were similar in the control and intervention groups, with mean ages of 63·3 and 63·0 years, average index of multiple deprivation (IMD) deciles of 6·33 and 5·53, and percentage of female participants 42·1% (n=2038) and 43·7% (n=2022), respectively. Uptake of TLHC was significantly higher in the intervention group (25·8%, n=1192) than in the control group (20·4%, n=987; χ2(1)=38·762, p<0·0001). A logistic regression model estimated likelihood of uptake was 37·5% (95% CI 12·1-63·2) higher in the intervention group (p=0·004). Patients aged 60-64, 65-69 and 70-75 years were 23·5% (10·7-36·3, p=0·0003), 28·3% (14·5-42·0, p<0·0001), and 32·6% (18·3-47·0, p<0·0001) more likely to attend than those aged 55-59 years. The likelihood of attendance decreased on the basis of deprivation (IMD decile; b=0·060, 95% CI 0·035-0·087, p<0·0001), female gender (b=0·156, 0·057-0·254, p=0·002), and for ex-smokers compared with smokers (b=0·580, 0·467-0·693, p<0·0001). INTERPRETATION: This simple, no-cost addition of behavioural messaging to invitation letters can significantly improve screening uptake and is recommended for wider rollout. These findings are consistent with other studies. However, a limitation is that the allocation to study arm was at GP practice level and the study was not randomised. FUNDING: RM Partners, the West London Cancer Alliance.