Related Experiment Videos
Latent Transition Analysis of Symptom Burden Before and After Pulmonary Rehabilitation in Postoperative Lung Cancer
Ying Wu1, Qun Yang1, Jinguo Wang1
1Department of Cardiothoracic Surgery, Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine, Nanjing, Jiangsu, People's Republic of China.
Aim:
To identify latent classes of symptom burden in lung cancer patients before and after pulmonary rehabilitation (PR) and to examine factors associated with transitions between classes.
Design:
A longitudinal study.
Methods:
The M. D. Anderson Symptom Inventory (Chinese version) was completed by 273 postoperative lung cancer patients before (T1) and 12 weeks after PR (T2). Latent profile analysis identified symptom burden classes, and latent transition analysis estimated the probability of moving between classes. Measurement invariance was tested, and the two-class solution was retained on the basis of fit indices, entropy and interpretability. Univariate-significant variables entered a logistic regression model.
Results:
Two latent classes were identified at both T1 and T2 (high and low symptom burden). The two-class solution was retained because it showed the lowest AIC, BIC and aBIC, adequate entropy (0.844 at T1; 0.902 at T2) and interpretable profiles, whereas the three-class model did not improve fit and yielded small classes. Among 262 patients (95.97% valid), the low burden class was stable (0.863), and 61.07% of patients in the high burden class at T1 moved to the low burden class at T2; these probabilities denote the proportion of patients moving between classes. Age <45 years (OR = 2.345), no financial burden (OR = 2.841), satisfactory sleep (OR = 3.130) and family living (OR = 2.949) were associated with transition from the high to the low burden class, whereas age ≥60 years (OR = 0.724), unmarried (OR = 0.418) and divorced (OR = 0.405) status with lower odds of transition.
Conclusion:
Symptom burden in postoperative lung cancer patients is heterogeneous and follows class-specific trajectories. Younger age, better sleep and stronger family and financial resources were associated with a favourable transition. As this was a single-arm observational study, the findings indicate associations rather than causal effects of PR.
