Objective <p>Pulmonary rehabilitation (PR) improves conditions for lung cancer (LC) surgery patients, but PR behavior is suboptimal. This study investigated the effect of message framing on PR behavior in this population, as this area is underexplored.</p> Methods <p>A three-arm randomized controlled trial enrolled 102 patients randomly assigned to a gain-framed message (GFM) group (information emphasizing positive outcomes of PR adherence), a loss-framed message (LFM) group (information highlighting negative consequences of non-adherence to PR), or a no-framed message (NFM) group (PR with no framed messages). Data were collected at admission, the day before surgery, the day of discharge, and one month after discharge. Primary outcomes were postoperative exercise adherence and exercise self-efficacy. Secondary outcomes included exercise capacity, dyspnea, anxiety, and depression.</p> Results <p>The baseline characteristics of the three groups were comparable (P &gt; 0.05). Repeated measures analysis of variance revealed a significant time effect (P &lt; 0.001) and a significant group effect (P &lt; 0.001) for exercise adherence. For exercise self-efficacy, exercise capacity, dyspnea, and anxiety, significant time effects, group effects, and interaction effects were observed (all P &lt; 0.05). However, the interaction effect for depression was not statistically significant (P = 0.109). Both the GFM and LFM groups demonstrated significantly better outcomes across all measures compared to the NFM group (P &lt; 0.05). Furthermore, the GFM group showed significantly greater improvement than the LFM group in all outcomes (P &lt; 0.05).</p> Conclusion <p>Message frames appear to be a promising communication strategy, with GFM in particular showing potential. When combined with PR, they may have the potential to improve PR behaviors and outcomes among patients undergoing LC surgery.</p> <p>Trial registration number (ChiCTR2300070445) and date of registration (2023–04-12).</p>