Yangjun Liu

Yangjun Liu
Karolinska Institutet | KI · Department of Medical Epidemiology and Biostatistics

Doctor of Medicine

About

10
Publications
623
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
38
Citations

Publications

Publications (10)
Article
Full-text available
Purpose To examine the trajectory of psychological distress from 1 to 2 years after esophageal cancer surgery, and whether dispositional optimism could predict the risk of postoperative psychological distress. Methods This Swedish nationwide longitudinal study included 192 patients who had survived for 1 year after esophageal cancer surgery. We me...
Article
Full-text available
PURPOSE We aimed to evaluate the efficacy and feasibility of patient-reported outcome (PRO)-based symptom management in the early period after lung cancer surgery. METHODS Before surgery, patients with clinically diagnosed lung cancer were randomly assigned 1:1 to receive postoperative PRO-based symptom management or usual care. All patients repor...
Article
Full-text available
Background The latest European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire-Lung Cancer 29 (QLQ-LC29) has been translated and validated in several languages but not yet in simplified Chinese. This study aimed to translate this questionnaire into simplified Chinese and adapt it for use in Chinese patients w...
Article
Background: Approximately 30% of patients suffer from severe reflux after surgery for esophageal cancer, which may serve as a continuous reminder of the cancer and catalyze fear of recurrence. Objective: The aim of this study was to investigate the association between severe reflux and symptoms of anxiety and depression after esophageal cancer s...
Article
Full-text available
Purpose To assess whether higher dispositional optimism could predict better health-related quality of life (HRQL) after esophageal cancer surgery. Methods This Swedish nationwide longitudinal study included 192 patients who underwent esophagectomy for cancer. The exposure was dispositional optimism measured by the Life Orientation Test-Revised (L...
Preprint
Full-text available
Background The latest European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire-Lung Cancer 29 (QLQ-LC29) has been translated and validated in several languages but not yet in simplified Chinese. This study aimed to translate this questionnaire into simplified Chinese and adapt it for use in Chinese patients w...
Article
Full-text available
Background: Patients are at higher risk of suffering from psychological distress and reduced health-related quality of life (HRQoL) after oesophageal cancer surgery. This Swedish nationwide population-based longitudinal study aimed to evaluate the association between psychological distress and HRQoL up to 2 years after oesophageal cancer surgery....
Preprint
Life orientation test-revised (LOT-R) is a common scale used to assess dispositional optimism. There is no consensus regarding the dimensionality of this scale. To date, no psychometric study regarding this scale has been conducted among surgically treated esophageal cancer patients. In this Swedish nationwide population-based study, we conducted a...
Article
The sulfonylurea receptor 1 (SUR1)-regulated NCca-ATP channels were progressively upregulated and demonstrated unchecked opening in central nervous system (CNS) injury, which induced cerebral damage. Glibenclamide (GLI) can block NCca-ATP channels and consequently exert protective effects. Recent studies have found that GLI has antioxidative effect...

Questions

Questions (3)
Question
Hi,
I am kind of a beginner of statistics and Stata. And I did a linear mixed model which contains an interaction term (exposure*time or exposure#time, both exposure and time are catergorical variables), and I want to get the estimated marginal means. However, I am confused by two technical questionas.
Firstly, I found that in stata, when I use the margins command, I could add option "over". Since I have an interaction term in my model and I want to assess the mariginal means for the the diffferent combined levels. I am wondering what is the difference of interpretation between the results from command "marginal exposure#time" and "marginal, over (exposure time)".
Secondly, it is said that in R, SAS, SPSS, and JMP, the marginal means procedure by default assumes a balanced population, but in Stata, it assumes an unbalanced population by default. Now I am wondering in which situation I should assume a balanced population and which situation an unbalanced population should be assumed?
Thanks a lot in advance!
Question
Hi,
I got a problem regarding method factor while doing the CFA.
I found that if I added correlated errors among positively worded items, the model could be assessed smoothly by the software. However, if I used method factor instead to explain the variance, after iterated 16000 times, the software suggested "convergence not achieved".
I am wondering the mechanism behind this. I am not so familiar with CFA and method factor. To me, adding correlated error among positively worded items and adding method factor to explaining variance among positively worded items should output similar results (e.g. goodness of fit), because they are kind of different ways to explain similar things. Currently, I am puzzling over the discordance of the outputs. And I am really appreciate if someone could give some hints.
Thank you very much!
Question
I got two questions after I did the confirmatory factor analysis, and I hope that I could get some help from this forum.
The first question is: my CFA result showed negative loading on one reversed indicator (this variable has already been reversed because of negative wording). I am wondering what should I do next? Since I didn't do EFA beforehand, should I back to the beginning and do EFA firstly (maybe the EFA results will suggest me to remove this item from the scale or do not reverse it)? Or shouldn't I reverse this item based on my CFA results (but it is a negatively worded item)? Or should I just present the negative loading and try to explain why this happened based on my data?
The second question is: If all negatively or positively worded items are correlated, should I add a new latent factor to replace the correlation errors in order to explain the variance? I saw some papers did like this. In order to make this question clearer, I attached two example models (v1 v2 v3 are positively items, and v4 v5 v6 are negatively worded items), my question is what is the difference between these two models, and how to decide which one should be selected?

Network

Cited By