KMRU_hhs_womac_network:
2023-09-11
Descriptive statistics
Descriptive stats at Baseline
## Number of subjects: 120## Number of subjects in control group: 60## Number of subjects in treatment group: 60##
## Age treatment## Mean: 67.9598448551221## Standard Deviation: 8.68625493698399## Range:## 48.5639972621492 - 87.3593429158111##
## Age control## Mean: 66.6892083048141## Standard Deviation: 11.2674775428205## Range:## 40.7008898015058 - 86.3846680355921##
## Height treatment## Mean: 169.947368421053## Standard Deviation: 8.07425594234889## Range:## 156 - 189##
## Height control## Mean: 171.077586206897## Standard Deviation: 7.82248318593223## Range:## 156 - 190##
## Weight treatment## Mean: 82.359649122807## Standard Deviation: 15.3796463321904## Range:## 59 - 129##
## Weight control## Mean: 82.801724137931## Standard Deviation: 13.0686244243079## Range:## 57 - 112##
## BMI treatment## Mean: 28.4109769648317## Standard Deviation: 4.16643954511763## Range:## 21.585557299843 - 37.448347107438##
## BMI control## Mean: 28.2710240901857## Standard Deviation: 3.98320553533097## Range:## 21.4535737137265 - 38.5674931129477##
## baseline womac treatment## Mean: 39.7692307692308## Standard Deviation: 14.843781803173## Range:## 10 - 74##
## baseline womac control## Mean: 40.6938775510204## Standard Deviation: 13.3607074438953## Range:## 16 - 73##
## baseline HHS treatment## Mean: 55.3428571428571## Standard Deviation: 19.2163434904473## Range:## 14.6 - 95.95##
## baseline HHS control## Mean: 68.2060344827586## Standard Deviation: 12.0385786215342## Range:## 39 - 92Descriptive stats at Follow-up
##
## follow up womac treatment## Mean: 9.22222222222222## Standard Deviation: 10.0065635025873## Range:## 0 - 44##
## follow up womac control## Mean: 34.8484848484849## Standard Deviation: 18.6750522290455## Range:## 4 - 62##
## follow up hhs treatment## Mean: 91.5166666666667## Standard Deviation: 7.17868530632683## Range:## 62.95 - 96##
## follow up hhs control## Mean: 76.6907894736842## Standard Deviation: 14.2466205346104## Range:## 53.25 - 96WP2 How the components of Harris hip score and WOMAC associate with physical capacity measures (baseline and follow-up)
Partial correlation network: Womac scores and physical capacity
Initially, partial correlation analysis was conducted using Pearson’s correlation coefficient. However, it was observed that the resulting partial correlation matrix was not positive definite. This indicated potential issues related to multicollinearity or other numerical stability concerns.
To address this, Spearman’s rank correlation coefficient was employed for the partial correlation analysis involving WOMAC scores. Spearman’s coefficient is based on ranks and is less sensitive to outliers and non-linear relationships, which can make it more suitable in cases where the assumptions of linearity and normality may not hold.
This adjustment allowed for a more robust assessment of the relationships between variables while controlling for the influence of covariates.
Baseline Womac scores and physical capacity
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/baseline_womac_physicalcapacity.pngMy observations from the plot:
The plot reveals that similar components of these scores tend to cluster together, such as stiffness, pain, and function.
Notably, some components exhibit very strong correlations, suggesting that they can be interchangeably used as proxies for one another. For example, tasks like taking socks on and off, engaging in heavy or light domestic activities, and the interplay of function and pain in various scenarios (e.g., standing, walking on flat surfaces, lying in bed) exhibit high intercorrelations.
The 40-meter walk test suggests an association with an individual’s capacity to engage in light domestic activities, highlighting its relevance as a measure of functional capability in this context.
The analysis reveals a limited overall association between objective physical measures and subjective self-reported experiences, suggesting that the two domains may capture distinct aspects of the individual’s health and well-being.
Follow-up Womac scores and physical capacity
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso## Warning in EBICglassoCore(S = S, n = n, gamma = gamma, penalize.diagonal =
## penalize.diagonal, : A dense regularized network was selected (lambda < 0.1 *
## lambda.max). Recent work indicates a possible drop in specificity. Interpret
## the presence of the smallest edges with care. Setting threshold = TRUE will
## enforce higher specificity, at the cost of sensitivity.
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/followup_womac_physicalcapacity.pngMy observations from the plot
Similar observations as the previous plot
Partial correlation network: Harris hip score components and physical capacity
Partial correlation analysis was conducted to assess the relationship between variables while controlling for the influence of one or more covariates. Pearson’s correlation coefficients were utilized in this analysis. This method allows for the isolation of the unique association between the variables of interest, while accounting for the effects of additional covariates.
Baseline Harris hip score components and physical capacity
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/baseline_HHS_physicalcapacity.pngMy observations
Similar to the WOMAC scores, we observed that analogous domains within the Harris Hip Score (HHS) tend to aggregate together
The 40 meters walk test demonstrated a correlation with the subject’s ability to ascend stairs according to the HHS assessment.
The range of internal rotation and flexion of the leg exhibited an association with the subject’s capability to independently put on or take off socks.
A broader range of adduction was found to be linked with higher limp score (i.e. lower limp) on HHS scale
Follow-up Harris hip score components and physical capacity
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/followup_HHS_physicalcapacity.pngAt the follow-up assessment, the 40-meter walk test demonstrated was associated not only with the stair-climbing component but also with the support and limp components of the Harris Hip Score (HHS).”
other observations similar as the last plot.
NOTE: The variable transport had only one value (i.e. every subject was able to get in a car). Despite that it shows correlation with other components, it should not because it has no variation. We need to consult a statistician about this. One solution is to remove it since it provides no additional information.
Additional observations
- In all the tests, the 40m walk test was most consistently associated with components in both WOMAC and HHS scales
WP3 What HHS and WOMAC components are affected by treatment
Partial correlation network: Womac scores in treatment vs control groups
Initially, partial correlation analysis was conducted using Pearson’s correlation coefficient. However, it was observed that the resulting partial correlation matrix was not positive definite. This indicated potential issues related to multicollinearity or other numerical stability concerns.
To address this, for partial correaltion analyses involving WOMAC scores, Spearman’s rank correlation coefficient was employed. Spearman’s coefficient is based on ranks and is less sensitive to outliers and non-linear relationships, which can make it more suitable in cases where the assumptions of linearity and normality may not hold.
This adjustment allowed for a more robust assessment of the relationships between variables while controlling for the influence of covariates.
Baseline Womac score in treatment vs control groups
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/baseline_womac_tvc.pngObservations
- At baseline, there was no differnce in WOMAC scores between treament and control groups
Baseline Womac score in treatment vs control groups
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso## Warning in EBICglassoCore(S = S, n = n, gamma = gamma, penalize.diagonal =
## penalize.diagonal, : A dense regularized network was selected (lambda < 0.1 *
## lambda.max). Recent work indicates a possible drop in specificity. Interpret
## the presence of the smallest edges with care. Setting threshold = TRUE will
## enforce higher specificity, at the cost of sensitivity.
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/followup_womac_tvc.pngAt follow-up, one group (treatment?) had better scores on standing pain and morning stiffness components of the WOMAC score. There was no difference in terms of physical capacity measures between treatment and control groups
Partial correlation network: Effect of treatment on Harris hip score components
Partial correlation analysis was conducted to assess the relationship between variables while controlling for the influence of one or more covariates. Pearson’s correlation coefficients were utilized in this analysis. This method allows for the isolation of the unique association between the variables of interest, while accounting for the effects of additional covariates.
Baseline HHS and treatment vs control groups
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/baseline_hhs_tvc.pngAt baseline, one group (control most likely) had lower score on stairs and walking compenents of HHS.
Baseline HHS and treatment vs control groups
## Estimating Network. Using package::function:
## - qgraph::EBICglasso for EBIC model selection
## - using glasso::glasso
## Output stored in C:/Users/saran/OneDrive/Work/Research/KMRU/Pihla/followup_hhs_tvc.pngObservations
At follow up, the opposite group (treatment most likely) had higher score on pain and internal rotation components of HHS. There was no difference in terms of physical capacity measures between treatment and control groups
NOTE: The variable transport had only one value (i.e. every subject was able to get in a car). Despite that it shows correlation with other components, it should not because it has no variation. We need to consult a statistician about this. One solution is to remove it since it provides no additional information.