GithubHelp home page GithubHelp logo

gallvp / pain-stigma Goto Github PK

View Code? Open in Web Editor NEW
0.0 1.0 0.0 979 KB

Statistical Analysis: The Determinants and Effects of Chronic Pain Stigma

Home Page: http://dx.doi.org/10.1016/j.jpain.2022.05.006

License: MIT License

R 79.64% TeX 20.36%

pain-stigma's Introduction

Statistical Analysis Plan

A statistical analysis consisting of hierarchical linear regressions is conducted to evaluate the following hypotheses:

  1. People with chronic pain such as those who have a co-morbid mental health diagnosis, who use opiates, and who have more medical (“organic”) beliefs about pain, experience greater stigma.

  2. People with higher stigma have higher pain levels, greater disability, depression, and less social support compared to those reporting less stigma.

For each hypothesis, the analysis is conducted in two steps. In the first step, a linear regression model is constructed which includes all the available demographic and clinical variables. The health of this model is evaluated. If the model assumptions are violated, necessary adjustments are made. In the second step, the variables of interest are also added to the model. The health of the model is re-evaluated and hypothesis tests are conducted. For hypothesis 1, the step 1 independent variables are Age, Gender, Ethnicity, Relationship, Employment, Diagnosis, Pain location and Pain duration. For step 2, additional independent variables are: Organic beliefs, Psychological beliefs, Daily opioid intake and History of mental health. All these variables are discrete except Age, Pain duration, Daily opioid intake, Organic beliefs and Psychological beliefs. The dependent continuous variable is stigma. For hypothesis 2, four linear regression models are constructed corresponding to the four clinical outcomes: Pain intensity, Disability, Depression and Social support. All these variables are continuous. For each outcome, step I independent variables are: Age, Gender, Ethnicity, Relationship, Employment, Diagnosis, Pain location and Pain duration, Organic beliefs, Psychological beliefs, Daily opioid intake and History of mental health. For each outcome, the remaining outcomes are also included as independent variables so that the interdependence across the outcomes is accounted for. At step 2, Stigma is also added as an independent variable.

The two step hierarchical linear regression allows estimation of the amount of variance explained (Radj2; r-squared adjusted) by all the predictors and the additional variance explained (Δ**Radj2) by the variables of interest. To test whether a specific independent variable explains significant amount of dependent variable variance, a type II analysis of variance (ANOVA) F-test is used. This test considers the variance explained by a specific independent variable by comparing a linear regression model with the specific variable to the model which does not have the specific variable. Both models have all the other step 2 independent variables. Therefore, this test accounts for the variance explained by the correlated independent variables. Statistical significance level is set at 0.05. The health of the regression models is checked by evaluating the normality/homogeneity of variance assumptions for the residuals using histogram, QQ-plot and fitted versus residuals plot. Moreover, multicollinearity is checked with the variance inflation factor which is considered unacceptable if greater than 10 (McElreath, 2018). Quadratic-root transformation is applied to Daily opioid intake and Pain duration as the original variables are not normally distributed.

An exploratory path analysis is also conducted with the aim of collating the inferences drawn from the hypothesis tests into a single parsimonious model. This model is constructed using structural equation modeling. Variables which have significant associations in the above described linear regressions are included in this model. Moreover, any variables which resulted in better fit statistics were also included. Various plausible directions of relationships amongst the variables are also explored while monitoring the health of the model. The health of the model is evaluated against suggested thresholds for CFI (> 0.9), TLI (> 0.9) and RMSEA (< 0.08) statistics (Van de Schoot, Lugtig, & Hox, 2012).

All the analyses are conducted in R using packages: car, dplyr, lavaan, ggplot2, semPlot and report (Epskamp, 2019; Fox & Weisberg, 2019; Makowski, Ben-Shachar, Patil, & Lüdecke, 2021; R Core Team, 2021; Rosseel, 2012; Wickham, 2016; Wickham, François, Henry, & Müller, 2021).

Results of Primary Analysis

For Stigma, demographic and clinical variables explain 10% of the variance. Organic beliefs, Psychological beliefs, Use of strong opiates and History of Mental health together explain an additional 14% of the Stigma variance. When considered individually, Organic beliefs (F(1, 184) = 10.84, p = 0.001), Mental health (F(1, 184) = 14.5, p = 0.0002), Daily opioid intake (F(1, 184) = 6.46, p = 0.01) and Employment (F(5, 184) = 2.45, p = 0.04) explain a significant portion of the variance. The corresponding regression coefficients and their 95% confidence intervals suggest that an increase in Organic beliefs predicts an increase (0.018, 95% CI [0.007, 0.029]) in Stigma, not having a mental health diagnosis predicts lower (-0.270, 95% CI [-0.410, -0.130]) levels of Stigma , an increase in opioid intake predicts an increase (0.077, 95% CI [0.017, 0.137]) in Stigma, and being unemployed predicts higher (0.272, 95% CI [0.117, 0.427]) levels of Stigma.

For Pain intensity, demographic and clinical variables together with other outcomes (Disability, Depression and Social support) explain 41.5% of the variance. Addition of stigma as a predictor does not explain any additional variance in Pain intensity. Individually, Ethnicity (F(3, 180) = 3.09, p = 0.03) and Disability (F(1, 180) = 29.41, p = 1.869 x 10−7) explain significant variance. Model coefficients suggest higher levels of pain intensity for people of Pacific ethnicity (1.88, 95% CI [0.55, 3.21]) and an increase (0.37, 95% CI [0.24, 0.50]) in pain intensity with an increase in Disability.

For Disability, demographic and clinical variables together with other outcomes (Pain intensity, Depression and Social support) explain 59% of the variance. Stigma explains an additional 3.2% of the variance in Disability. Individually, Relationship (F(4, 180) = 4.8, p = 0.001), Employment (F(5, 180) = 2.4, p = 0.04), Pain location (F(8, 180) = 2.3, p = 0.03), Organic beliefs (F(1, 180) = 4.4, p = 0.04), Daily opioid intake (F(1, 180) = 4.33, p = 0.04), Pain intensity (F(1, 180) = 29.41, p = 1.869 x 10−7), Depression (F(1, 180) = 18.64, p = 2.604 x 10−5) and Stigma (F(1, 180) = 16.31, p = 7.94 x 10−5) explain significant variance in Disability. Model coefficients suggest lower (-0.516, 95% CI [-1.017, -0.015]) levels of Disability for people who are married or in a de facto relationship, higher levels of Disability (1.24, 95% CI [0.15, 2.33]) for people with “Other” employment status, lower levels of Disability for people with pain in Neck/back (-0.8, 95% CI [-1.53, -0.07]) or Abdomen/pelvis (-0.846, 95% CI [-1.63, -0.06]), an increase (0.04, 95% CI [0.002, 0.07]) in Disability with an increase in Organic beliefs, an increase (0.19, 95% CI [0.01, 0.37]) in Disability with an increase in Daily opioid intake, an increase (0.38, 95% CI [0.24, 0.52]) in Disability with an increase in Pain intensity, an increase (0.09, 95% CI [0.05, 0.12]) in Disability with increase in Depression, and an increase (1.05, 95% CI [0.54, 1.56]) in Disability with an increase in Stigma.

For Depression, demographic and clinical variables together with other outcomes (Pain intensity, Disability and Social support) explain 48.7% of the variance. Stigma explains an additional 4.4% of the variance. Individually, Mental health (F(1, 180) = 6.2, p = 0.01), Disability (F(1, 180) = 18.64, p = 2.604 x 10−5) and Stigma (F(1, 180) = 17.74, p = 3.999 x 10−5) explain significant variance in Depression. Model coefficients suggest lower levels of Depression (-1.98, 95% CI [-3.55, -0.41]) for people who do not have a mental health diagnosis, an increase in Depression (1.1, 95% CI [0.6, 1.61]) with an increase in Disability and an increase in Depression (3.92, 95% CI [2.08, 5.75]) with an increase in Stigma.

For Social support, demographic and clinical variables together with other outcomes (Pain intensity, Disability and Depression) explain 11% of the variance. Stigma explains an additional 6.1% of the variance. Individually, only Stigma (F(1, 180) = 14.29, p = 0.0002) explain significant variance in Social support. Model coefficients suggest a decrease in social support (-1.77, 95% CI [-2.7, -0.85]) with an increase in Stigma.

Results of Exploratory Path Analysis

The exploratory path analysis suggests higher levels of Stigma associated with higher opioid intake, having more Organic pain beliefs, having a mental health history and being unemployed. Higher Stigma itself predicts higher Disability and Depression but lower Social support. Pain intensity is not predicted by any variable in the data. Rather it is a strong predictor of Disability. Higher Pain intensity predicts higher Disability. Higher Disability itself predicts higher Depression.

Being married or having a de facto relationship predicts lower levels of Disability, having more organic beliefs predicts higher Disability and having neck/back/abdomen/pelvis pain predicts lower Disability compared to upper limb/chest/hips/lower limb/widespread pain. Having a mental health history predicts higher Depression. Some of the relationships which stand out in the hierarchical linear regression analysis are removed from the model as their presence results in poor model fit statistics. Changing the presented direction of relationship amongst the variables also results in poor fit statistics. The fit statistics for the presented model are CFI = 0.95, TLI = 0.93 and RMSEA = 0.067 90% CI [0.036, 0.097].

These results are illustrated in the following figure where the direction and width of the arrows represent the direction and strength of relationship between the variables. Red arrows represent a decrease in the consequent variable with an increase in the antecedent variable. Green arrows represent an increase in the consequent variable with an increase in the antecedent variable.

Hypothesis I

Step I: Stigma and demographic/clinical variables

Model Setup

Model.stageI.stepI <- lm(Stigma.total ~ Age + Gender + Ethnicity +
                           Relationship + Employment +
                           Diagnosis + Pain.loc + Pain.duration,
                         data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.573878 3.100753 1.604331 0.3885188 0.3225023 0.4588955 Low (< 5)
Gender 1.414058 1.671667 1.189142 0.7071844 0.5982051 0.7966530 Low (< 5)
Ethnicity 1.608568 1.907132 1.268293 0.6216710 0.5243477 0.7100909 Low (< 5)
Relationship 3.773982 4.592752 1.942674 0.2649721 0.2177344 0.3182884 Low (< 5)
Employment 3.249806 3.940799 1.802722 0.3077107 0.2537556 0.3674885 Low (< 5)
Diagnosis 11.405325 14.091773 3.377177 0.0876783 0.0709634 0.1078733 High (> 10)
Pain.loc 7.984997 9.833816 2.825774 0.1252349 0.1016899 0.1533010 Moderate
Pain.duration 1.337317 1.581134 1.156424 0.7477660 0.6324574 0.8362644 Low (< 5)

High and moderate variance inflation factor is indicated for Diagnosis and Pain location, respectively. Plotting these factors against each other reveals the underlying clustering of participants across them. If pain locations(s) is known, one can predict the Diagnosis thus eliminating the need for inclusion of Diagnosis in the model.

Reduced Model Setup

Model.stageI.stepI <- lm(Stigma.total ~ Age + Gender + Ethnicity +
                           Relationship + Employment +
                           Pain.loc + Pain.duration,
                         data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.308445 2.808475 1.519357 0.4331920 0.3560652 0.5136977 Low (<5)
Gender 1.244953 1.499299 1.115775 0.8032429 0.6669783 0.8927194 Low (<5)
Ethnicity 1.353792 1.622538 1.163526 0.7386660 0.6163184 0.8325966 Low (<5)
Relationship 2.730012 3.339572 1.652275 0.3662987 0.2994396 0.4387371 Low (<5)
Employment 2.396827 2.919732 1.548169 0.4172183 0.3424972 0.4959449 Low (<5)
Pain.loc 2.123456 2.575839 1.457208 0.4709304 0.3882230 0.5552654 Low (<5)
Pain.duration 1.188891 1.443709 1.090363 0.8411202 0.6926605 0.9255725 Low (<5)

Fit Statistics

Parameter Fit
R2 0.210
R2 (adj.) 0.109
Sigma 0.457

Step II: Explaining stigma with pain beliefs, …

Model Setup

Model.stageI.stepII <- lm(Stigma.total ~ Age + Gender + Ethnicity +
                            Relationship + Employment +
                            Pain.loc + Pain.duration +
                            Beliefs.organic + Beliefs.psych +
                            Daily.opioid.intake + Mental.health,
                          data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.411769 2.925443 1.552987 0.4146333 0.3418285 0.4913680 Low (< 5)
Gender 1.371882 1.636403 1.171274 0.7289256 0.6110966 0.8214838 Low (< 5)
Ethnicity 1.461648 1.744080 1.208986 0.6841591 0.5733682 0.7773513 Low (< 5)
Relationship 3.084261 3.769151 1.756206 0.3242268 0.2653117 0.3892924 Low (< 5)
Employment 3.011249 3.677485 1.735295 0.3320882 0.2719250 0.3982808 Low (< 5)
Pain.loc 2.516393 3.056587 1.586314 0.3973941 0.3271623 0.4721235 Low (< 5)
Pain.duration 1.206908 1.452681 1.098594 0.8285635 0.6883826 0.9135990 Low (< 5)
Beliefs.organic 1.328054 1.584990 1.152412 0.7529816 0.6309189 0.8446176 Low (< 5)
Beliefs.psych 1.158984 1.411020 1.076561 0.8628244 0.7087074 0.9420667 Low (< 5)
Daily.opioid.intake 1.268841 1.517734 1.126428 0.7881211 0.6588769 0.8775015 Low (< 5)
Mental.health 1.225183 1.471013 1.106880 0.8162044 0.6798037 0.9028071 Low (< 5)

Model Summary

Parameter Coefficient CI (low) CI (high) Std. Coefficient Fit
(Intercept) 2.191 1.578 2.805 0.444
Age -0.003 -0.009 0.004 -0.081
Gender [Female] -0.146 -0.380 0.087 -0.303
Gender [Other] 0.098 -0.558 0.753 0.202
Ethnicity [Māori] 0.028 -0.192 0.248 0.058
Ethnicity [Pacific] 0.432 -0.018 0.882 0.893
Ethnicity [Asian] 0.411 -0.030 0.852 0.849
Relationship [In a relation] -0.177 -0.388 0.034 -0.366
Relationship [Married/De facto] -0.077 -0.248 0.095 -0.159
Relationship [Divorced/sep.] -0.146 -0.397 0.105 -0.302
Relationship [Widowed] 0.020 -0.450 0.489 0.041
Employment [Unemployed/benefit] 0.272 0.117 0.427 0.563
Employment [Work at home] 0.074 -0.213 0.360 0.153
Employment [Retired] 0.087 -0.203 0.377 0.180
Employment [Student] 0.072 -0.200 0.345 0.150
Employment [Other] 0.071 -0.301 0.444 0.148
Pain loc [Neck/back] -0.066 -0.316 0.185 -0.136
Pain loc [Upper limb] -0.127 -0.396 0.143 -0.262
Pain loc [Chest] -0.187 -0.625 0.251 -0.387
Pain loc [Abdomen/pelvis] 0.011 -0.257 0.280 0.024
Pain loc [Hips/buttocks] 0.048 -0.238 0.334 0.099
Pain loc [Lower limb] -0.069 -0.344 0.206 -0.142
Pain loc [Other/widepread] -0.056 -0.378 0.265 -0.117
Pain loc [Missing] 0.271 -0.285 0.827 0.560
Pain duration -0.024 -0.113 0.064 -0.036
Beliefs organic 0.018 0.007 0.029 0.227
Beliefs psych 0.014 -0.002 0.030 0.113
Daily opioid intake 0.077 0.017 0.137 0.171
Mental health [No diagnosis] -0.270 -0.410 -0.130 -0.559
R2 0.342
R2 (adj.) 0.242
Sigma 0.421

Hypothesis Tests

Anova Table (Type II tests)

Response: Stigma.total
                    Sum Sq  Df F value    Pr(>F)    
Age                  0.134   1  0.7575 0.3852411    
Gender               0.363   2  1.0235 0.3613587    
Ethnicity            1.217   3  2.2876 0.0800890 .  
Relationship         0.632   4  0.8909 0.4705062    
Employment           2.168   5  2.4450 0.0357358 *  
Pain.loc             0.828   8  0.5837 0.7905089    
Pain.duration        0.052   1  0.2927 0.5891808    
Beliefs.organic      1.923   1 10.8427 0.0011894 ** 
Beliefs.psych        0.550   1  3.1031 0.0798065 .  
Daily.opioid.intake  1.146   1  6.4600 0.0118577 *  
Mental.health        2.571   1 14.4948 0.0001913 ***
Residuals           32.634 184                      
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Hypothesis II

Pain Intensity

Step I: Pain intensity and demographic/clinical variables

Model Setup

Model.stageII.pain.stepI <- lm(Pain.intensity ~ Age + Gender + Ethnicity +
                                 Relationship + Employment +
                                 Pain.loc + Pain.duration +
                                 Beliefs.organic + Beliefs.psych +
                                 Daily.opioid.intake + Mental.health +
                                 Disability + Depression.phq9 + Support.oss3,
                               data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.489637 3.012918 1.577859 0.4016649 0.3319042 0.4756492 Low (< 5)
Gender 1.401272 1.665734 1.183753 0.7136376 0.6003361 0.8052394 Low (< 5)
Ethnicity 1.623599 1.936011 1.274205 0.6159157 0.5165258 0.7064840 Low (< 5)
Relationship 3.372716 4.117185 1.836496 0.2964970 0.2428844 0.3563733 Low (< 5)
Employment 3.492912 4.267603 1.868933 0.2862941 0.2343236 0.3446034 Low (< 5)
Pain.loc 3.019683 3.675505 1.737723 0.3311606 0.2720714 0.3960997 Low (< 5)
Pain.duration 1.235755 1.476384 1.111645 0.8092218 0.6773306 0.8955184 Low (< 5)
Beliefs.organic 1.464366 1.741567 1.210110 0.6828891 0.5741957 0.7747225 Low (< 5)
Beliefs.psych 1.191811 1.432446 1.091701 0.8390593 0.6981064 0.9215932 Low (< 5)
Daily.opioid.intake 1.342393 1.596159 1.158617 0.7449384 0.6265042 0.8356682 Low (< 5)
Mental.health 1.436347 1.707766 1.198477 0.6962104 0.5855603 0.7880133 Low (< 5)
Disability 2.453480 2.967758 1.566359 0.4075844 0.3369546 0.4822510 Low (< 5)
Depression.phq9 2.264017 2.731263 1.504665 0.4416929 0.3661310 0.5200548 Low (< 5)
Support.oss3 1.316873 1.566545 1.147551 0.7593749 0.6383474 0.8494523 Low (< 5)

Fit Statistics

Parameter Fit
33
34 R2 0.501
35 R2 (adj.) 0.415
36 Sigma 1.202

Step II: Explaining pain intensity with stigma

Model Setup

Model.stageII.pain.stepII <- lm(Pain.intensity ~ Age + Gender + Ethnicity +
                                  Relationship + Employment +
                                  Pain.loc + Pain.duration +
                                  Beliefs.organic + Beliefs.psych +
                                  Daily.opioid.intake + Mental.health +
                                  Disability + Depression.phq9 + Support.oss3 +
                                  Stigma.total,
                                data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.498308 3.020370 1.580604 0.4002710 0.3310852 0.4736765 Low (< 5)
Gender 1.415544 1.680913 1.189766 0.7064422 0.5949148 0.7977052 Low (< 5)
Ethnicity 1.652702 1.969686 1.285575 0.6050696 0.5076952 0.6947635 Low (< 5)
Relationship 3.539066 4.320385 1.881241 0.2825604 0.2314608 0.3399512 Low (< 5)
Employment 3.599777 4.396275 1.897308 0.2777950 0.2274653 0.3344399 Low (< 5)
Pain.loc 3.128161 3.806856 1.768661 0.3196767 0.2626840 0.3826179 Low (< 5)
Pain.duration 1.238331 1.477252 1.112803 0.8075388 0.6769327 0.8936408 Low (< 5)
Beliefs.organic 1.478936 1.757267 1.216115 0.6761617 0.5690655 0.7675157 Low (< 5)
Beliefs.psych 1.205612 1.443725 1.098004 0.8294542 0.6926528 0.9130119 Low (< 5)
Daily.opioid.intake 1.360458 1.615502 1.166387 0.7350465 0.6190027 0.8256983 Low (< 5)
Mental.health 1.440556 1.710929 1.200232 0.6941762 0.5844778 0.7855404 Low (< 5)
Disability 2.678109 3.244759 1.636493 0.3733978 0.3081893 0.4435572 Low (< 5)
Depression.phq9 2.481287 2.999137 1.575210 0.4030166 0.3334293 0.4767400 Low (< 5)
Support.oss3 1.420714 1.687103 1.191937 0.7038715 0.5927319 0.7951633 Low (< 5)
Stigma.total 2.416042 2.917757 1.554362 0.4139000 0.3427291 0.4888579 Low (< 5)

Model Summary

Parameter Coefficient CI (low) CI (high) Std. Coefficient Fit
(Intercept) 1.979 -0.308 4.267 -0.148
Age 0.015 -0.003 0.034 0.136
Gender [Female] -0.009 -0.681 0.663 -0.006
Gender [Other] -0.256 -2.139 1.626 -0.163
Ethnicity [Māori] -0.188 -0.825 0.450 -0.119
Ethnicity [Pacific] 1.879 0.549 3.209 1.195
Ethnicity [Asian] 0.662 -0.624 1.949 0.421
Relationship [In a relation] 0.069 -0.545 0.683 0.044
Relationship [Married/De facto] 0.305 -0.191 0.801 0.194
Relationship [Divorced/sep.] -0.446 -1.170 0.278 -0.283
Relationship [Widowed] 0.784 -0.572 2.139 0.498
Employment [Unemployed/benefit] 0.261 -0.205 0.727 0.166
Employment [Work at home] -0.535 -1.359 0.289 -0.340
Employment [Retired] 0.347 -0.492 1.186 0.221
Employment [Student] 0.539 -0.249 1.327 0.343
Employment [Other] -0.220 -1.303 0.863 -0.140
Pain loc [Neck/back] 0.077 -0.653 0.806 0.049
Pain loc [Upper limb] -0.046 -0.825 0.734 -0.029
Pain loc [Chest] 0.029 -1.239 1.297 0.019
Pain loc [Abdomen/pelvis] 0.134 -0.648 0.915 0.085
Pain loc [Hips/buttocks] -0.366 -1.186 0.454 -0.233
Pain loc [Lower limb] -0.071 -0.873 0.730 -0.045
Pain loc [Other/widepread] -0.097 -1.023 0.829 -0.062
Pain loc [Missing] 2.300 0.682 3.918 1.462
Pain duration 0.073 -0.184 0.330 0.033
Beliefs organic 0.024 -0.009 0.057 0.092
Beliefs psych -0.044 -0.090 0.002 -0.108
Daily opioid intake -0.012 -0.189 0.164 -0.008
Mental health [No diagnosis] -0.155 -0.589 0.280 -0.098
Disability 0.369 0.235 0.503 0.467
Depression phq9 0.029 -0.011 0.069 0.118
Support oss3 -0.014 -0.094 0.067 -0.021
Stigma total -0.192 -0.716 0.332 -0.059
R2 0.502
R2 (adj.) 0.414
Sigma 1.204

Hypothesis Tests

Anova Table (Type II tests)

Response: Pain.intensity
                     Sum Sq  Df F value    Pr(>F)    
Age                   3.896   1  2.6872   0.10290    
Gender                0.109   2  0.0377   0.96305    
Ethnicity            13.451   3  3.0927   0.02835 *  
Relationship          9.542   4  1.6455   0.16477    
Employment            8.066   5  1.1127   0.35523    
Pain.loc             18.405   8  1.5869   0.13143    
Pain.duration         0.458   1  0.3156   0.57494    
Beliefs.organic       3.027   1  2.0879   0.15021    
Beliefs.psych         5.112   1  3.5260   0.06203 .  
Daily.opioid.intake   0.027   1  0.0187   0.89125    
Mental.health         0.716   1  0.4940   0.48307    
Disability           42.633   1 29.4077 1.869e-07 ***
Depression.phq9       2.955   1  2.0381   0.15514    
Support.oss3          0.162   1  0.1117   0.73860    
Stigma.total          0.757   1  0.5220   0.47094    
Residuals           260.950 180                      
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Disability

Step I: Disability and demographic/clinical variables

Model Setup

Model.stageII.dis.stepI <- lm(Disability ~ Age + Gender + Ethnicity +
                                Relationship + Employment +
                                Pain.loc + Pain.duration +
                                Beliefs.organic + Beliefs.psych +
                                Daily.opioid.intake + Mental.health +
                                Pain.intensity + Depression.phq9 + Support.oss3,
                              data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.508362 3.036307 1.583781 0.3986665 0.3293475 0.4723006 Low (< 5)
Gender 1.393612 1.656605 1.180513 0.7175599 0.6036443 0.8090898 Low (< 5)
Ethnicity 1.677143 2.001959 1.295046 0.5962520 0.4995107 0.6860470 Low (< 5)
Relationship 3.203525 3.905485 1.789839 0.3121562 0.2560501 0.3743696 Low (< 5)
Employment 3.420644 4.177162 1.849498 0.2923426 0.2393970 0.3515851 Low (< 5)
Pain.loc 2.967350 3.610050 1.722600 0.3370010 0.2770045 0.4027534 Low (< 5)
Pain.duration 1.237891 1.478641 1.112605 0.8078257 0.6762968 0.8942665 Low (< 5)
Beliefs.organic 1.424363 1.693367 1.193467 0.7020680 0.5905396 0.7938243 Low (< 5)
Beliefs.psych 1.215159 1.455123 1.102342 0.8229378 0.6872274 0.9076749 Low (< 5)
Daily.opioid.intake 1.291236 1.537268 1.136326 0.7744520 0.6505046 0.8636553 Low (< 5)
Mental.health 1.418583 1.686435 1.191043 0.7049288 0.5929669 0.7966548 Low (< 5)
Pain.intensity 1.720562 2.055566 1.311702 0.5812054 0.4864839 0.6702962 Low (< 5)
Depression.phq9 1.862290 2.231127 1.364657 0.5369734 0.4482039 0.6234596 Low (< 5)
Support.oss3 1.295597 1.542210 1.138243 0.7718449 0.6484203 0.8612144 Low (< 5)

Fit Statistics

Parameter Fit
33
34 R2 0.650
35 R2 (adj.) 0.590
36 Sigma 1.274

Step II: Explaining disability with stigma

Model Setup

Model.stageII.dis.stepII <- lm(Disability ~ Age + Gender + Ethnicity +
                                 Relationship + Employment +
                                 Pain.loc + Pain.duration +
                                 Beliefs.organic + Beliefs.psych +
                                 Daily.opioid.intake + Mental.health +
                                 Pain.intensity + Depression.phq9 + Support.oss3 +
                                 Stigma.total,
                               data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.509798 3.034706 1.584234 0.3984384 0.3295212 0.4716303 Low (< 5)
Gender 1.401393 1.664009 1.183804 0.7135757 0.6009583 0.8047370 Low (< 5)
Ethnicity 1.709072 2.039150 1.307315 0.5851127 0.4904005 0.6739264 Low (< 5)
Relationship 3.313059 4.037905 1.820181 0.3018359 0.2476532 0.3621664 Low (< 5)
Employment 3.481691 4.248669 1.865929 0.2872167 0.2353679 0.3453290 Low (< 5)
Pain.loc 3.055871 3.716538 1.748105 0.3272389 0.2690676 0.3912547 Low (< 5)
Pain.duration 1.239589 1.478588 1.113368 0.8067190 0.6763209 0.8929033 Low (< 5)
Beliefs.organic 1.460479 1.734944 1.208503 0.6847069 0.5763873 0.7760886 Low (< 5)
Beliefs.psych 1.223432 1.461669 1.106089 0.8173727 0.6841493 0.9024183 Low (< 5)
Daily.opioid.intake 1.328614 1.578290 1.152655 0.7526640 0.6335971 0.8426478 Low (< 5)
Mental.health 1.430101 1.698362 1.195868 0.6992512 0.5888025 0.7905846 Low (< 5)
Pain.intensity 1.727013 2.061294 1.314159 0.5790346 0.4851322 0.6675471 Low (< 5)
Depression.phq9 2.273916 2.740575 1.507951 0.4397701 0.3648870 0.5174991 Low (< 5)
Support.oss3 1.419391 1.685519 1.191382 0.7045273 0.5932890 0.7958121 Low (< 5)
Stigma.total 2.221692 2.675510 1.490534 0.4501074 0.3737605 0.5288788 Low (< 5)

Model Summary

Parameter Coefficient CI (low) CI (high) Std. Coefficient Fit
(Intercept) -1.078 -3.416 1.260 0.260
Age 0.013 -0.006 0.032 0.091
Gender [Female] 0.317 -0.365 0.999 0.159
Gender [Other] -0.674 -2.585 1.237 -0.339
Ethnicity [Māori] 0.477 -0.167 1.122 0.240
Ethnicity [Pacific] 0.728 -0.648 2.105 0.366
Ethnicity [Asian] 0.009 -1.303 1.320 0.004
Relationship [In a relation] 0.195 -0.429 0.819 0.098
Relationship [Married/De facto] -0.516 -1.017 -0.015 -0.259
Relationship [Divorced/sep.] 0.657 -0.075 1.390 0.330
Relationship [Widowed] -1.201 -2.572 0.170 -0.603
Employment [Unemployed/benefit] -0.321 -0.794 0.151 -0.161
Employment [Work at home] 0.699 -0.136 1.534 0.351
Employment [Retired] 0.291 -0.562 1.144 0.146
Employment [Student] 0.118 -0.687 0.923 0.059
Employment [Other] 1.239 0.153 2.325 0.622
Pain loc [Neck/back] -0.800 -1.533 -0.068 -0.402
Pain loc [Upper limb] -0.648 -1.435 0.139 -0.326
Pain loc [Chest] 0.161 -1.128 1.449 0.081
Pain loc [Abdomen/pelvis] -0.846 -1.631 -0.062 -0.425
Pain loc [Hips/buttocks] -0.175 -1.010 0.660 -0.088
Pain loc [Lower limb] -0.792 -1.598 0.014 -0.398
Pain loc [Other/widepread] 0.226 -0.715 1.167 0.114
Pain loc [Missing] -0.982 -2.656 0.692 -0.493
Pain duration 0.048 -0.213 0.309 0.017
Beliefs organic 0.035 0.002 0.069 0.107
Beliefs psych -0.022 -0.069 0.025 -0.043
Daily opioid intake 0.187 0.010 0.365 0.101
Mental health [No diagnosis] -0.300 -0.740 0.140 -0.151
Pain intensity 0.381 0.242 0.519 0.301
Depression phq9 0.085 0.046 0.124 0.275
Support oss3 -0.022 -0.104 0.060 -0.027
Stigma total 1.046 0.535 1.557 0.254
R2 0.679
R2 (adj.) 0.622
Sigma 1.224

Hypothesis Tests

Anova Table (Type II tests)

Response: Disability
                     Sum Sq  Df F value    Pr(>F)    
Age                   2.771   1  1.8508   0.17539    
Gender                2.836   2  0.9468   0.38989    
Ethnicity             4.545   3  1.0117   0.38883    
Relationship         28.916   4  4.8275   0.00101 ** 
Employment           17.755   5  2.3715   0.04105 *  
Pain.loc             25.836   8  2.1567   0.03285 *  
Pain.duration         0.199   1  0.1326   0.71618    
Beliefs.organic       6.572   1  4.3891   0.03757 *  
Beliefs.psych         1.277   1  0.8528   0.35699    
Daily.opioid.intake   6.489   1  4.3334   0.03879 *  
Mental.health         2.716   1  1.8135   0.17978    
Pain.intensity       44.036   1 29.4077 1.869e-07 ***
Depression.phq9      27.911   1 18.6392 2.604e-05 ***
Support.oss3          0.419   1  0.2795   0.59767    
Stigma.total         24.429   1 16.3138 7.937e-05 ***
Residuals           269.536 180                      
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Depression

Step I: Depression and demographic/clinical variables

Model Setup

Model.stageII.dep.stepI <- lm(Depression.phq9 ~ Age + Gender + Ethnicity +
                                Relationship + Employment +
                                Pain.loc + Pain.duration +
                                Beliefs.organic + Beliefs.psych +
                                Daily.opioid.intake + Mental.health +
                                Pain.intensity + Disability + Support.oss3,
                              data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.464729 2.981808 1.569946 0.4057241 0.3353670 0.4801776 Low (< 5)
Gender 1.375806 1.635465 1.172948 0.7268465 0.6114471 0.8181649 Low (< 5)
Ethnicity 1.686654 2.013693 1.298712 0.5928898 0.4966001 0.6825357 Low (< 5)
Relationship 3.372557 4.116987 1.836452 0.2965109 0.2428961 0.3563894 Low (< 5)
Employment 3.391392 4.140556 1.841573 0.2948642 0.2415135 0.3544921 Low (< 5)
Pain.loc 3.114230 3.793773 1.764718 0.3211066 0.2635898 0.3846190 Low (< 5)
Pain.duration 1.221142 1.461196 1.105053 0.8189057 0.6843708 0.9041292 Low (< 5)
Beliefs.organic 1.479814 1.760271 1.216476 0.6757608 0.5680943 0.7675690 Low (< 5)
Beliefs.psych 1.183404 1.424780 1.087844 0.8450201 0.7018627 0.9266237 Low (< 5)
Daily.opioid.intake 1.342727 1.596548 1.158761 0.7447532 0.6263512 0.8354903 Low (< 5)
Mental.health 1.376704 1.636527 1.173330 0.7263726 0.6110499 0.8177032 Low (< 5)
Pain.intensity 1.985594 2.384353 1.409111 0.5036276 0.4194010 0.5876487 Low (< 5)
Disability 2.329004 2.812354 1.526107 0.4293681 0.3555740 0.5064414 Low (< 5)
Support.oss3 1.298275 1.545252 1.139419 0.7702527 0.6471435 0.8597205 Low (< 5)

Fit Statistics

Parameter Fit
33
34 R2 0.562
35 R2 (adj.) 0.487
36 Sigma 4.610

Step II: Explaining depression with stigma

Model Setup

Model.stageII.dep.stepII <- lm(Depression.phq9 ~ Age + Gender + Ethnicity +
                                 Relationship + Employment +
                                 Pain.loc + Pain.duration +
                                 Beliefs.organic + Beliefs.psych +
                                 Daily.opioid.intake + Mental.health +
                                 Pain.intensity + Disability + Support.oss3 +
                                 Stigma.total,
                               data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.488632 3.008300 1.577540 0.4018271 0.3324137 0.4754131 Low (< 5)
Gender 1.401656 1.664322 1.183915 0.7134418 0.6008451 0.8046054 Low (< 5)
Ethnicity 1.723900 2.057451 1.312974 0.5800801 0.4860384 0.6686455 Low (< 5)
Relationship 3.569352 4.358243 1.889273 0.2801629 0.2294503 0.3371794 Low (< 5)
Employment 3.575388 4.365788 1.890870 0.2796899 0.2290537 0.3366323 Low (< 5)
Pain.loc 3.249759 3.958800 1.802709 0.3077151 0.2526018 0.3689177 Low (< 5)
Pain.duration 1.228376 1.466789 1.108321 0.8140828 0.6817615 0.8994965 Low (< 5)
Beliefs.organic 1.494741 1.776431 1.222596 0.6690121 0.5629264 0.7603124 Low (< 5)
Beliefs.psych 1.208752 1.446823 1.099433 0.8272993 0.6911695 0.9111388 Low (< 5)
Daily.opioid.intake 1.358541 1.613247 1.165565 0.7360836 0.6198677 0.8267026 Low (< 5)
Mental.health 1.396390 1.658048 1.181690 0.7161322 0.6031190 0.8072492 Low (< 5)
Pain.intensity 1.986670 2.383080 1.409493 0.5033548 0.4196251 0.5868967 Low (< 5)
Disability 2.823293 3.426031 1.680266 0.3541963 0.2918830 0.4218870 Low (< 5)
Support.oss3 1.420454 1.686792 1.191828 0.7040003 0.5928413 0.7952907 Low (< 5)
Stigma.total 2.205680 2.655565 1.485153 0.4533750 0.3765677 0.5324682 Low (< 5)

Model Summary

Parameter Coefficient CI (low) CI (high) Std. Coefficient Fit
(Intercept) 5.730 -2.675 14.135 -0.254
Age -0.063 -0.130 0.004 -0.137
Gender [Female] -0.988 -3.447 1.472 -0.153
Gender [Other] 2.839 -4.046 9.724 0.441
Ethnicity [Māori] 1.165 -1.166 3.497 0.181
Ethnicity [Pacific] 0.358 -4.617 5.334 0.056
Ethnicity [Asian] 1.753 -2.967 6.473 0.272
Relationship [In a relation] 1.230 -1.013 3.474 0.191
Relationship [Married/De facto] 0.852 -0.969 2.673 0.132
Relationship [Divorced/sep.] -0.957 -3.617 1.703 -0.149
Relationship [Widowed] 3.747 -1.205 8.699 0.582
Employment [Unemployed/benefit] 1.511 -0.186 3.208 0.235
Employment [Work at home] -0.968 -3.998 2.061 -0.150
Employment [Retired] 1.314 -1.758 4.386 0.204
Employment [Student] -1.768 -4.659 1.122 -0.275
Employment [Other] -0.981 -4.949 2.986 -0.152
Pain loc [Neck/back] 2.266 -0.387 4.919 0.352
Pain loc [Upper limb] 1.044 -1.808 3.896 0.162
Pain loc [Chest] 1.926 -2.711 6.563 0.299
Pain loc [Abdomen/pelvis] 2.663 -0.173 5.499 0.414
Pain loc [Hips/buttocks] 1.708 -1.293 4.709 0.265
Pain loc [Lower limb] 2.550 -0.361 5.462 0.396
Pain loc [Other/widepread] 2.006 -1.376 5.387 0.312
Pain loc [Missing] 1.594 -4.457 7.645 0.248
Pain duration -0.633 -1.569 0.304 -0.070
Beliefs organic -0.025 -0.146 0.097 -0.023
Beliefs psych 0.149 -0.019 0.317 0.090
Daily opioid intake -0.171 -0.818 0.476 -0.029
Mental health [No diagnosis] -1.977 -3.544 -0.411 -0.307
Pain intensity 0.388 -0.148 0.923 0.095
Disability 1.104 0.600 1.609 0.341
Support oss3 -0.057 -0.351 0.238 -0.021
Stigma total 3.917 2.082 5.752 0.294
R2 0.601
R2 (adj.) 0.531
Sigma 4.411

Hypothesis Tests

Anova Table (Type II tests)

Response: Depression.phq9
                    Sum Sq  Df F value    Pr(>F)    
Age                   66.1   1  3.3975   0.06694 .  
Gender                36.2   2  0.9298   0.39653    
Ethnicity             28.4   3  0.4869   0.69177    
Relationship          97.3   4  1.2497   0.29162    
Employment           132.9   5  1.3659   0.23922    
Pain.loc             106.7   8  0.6856   0.70386    
Pain.duration         34.6   1  1.7768   0.18422    
Beliefs.organic        3.2   1  0.1625   0.68732    
Beliefs.psych         59.3   1  3.0492   0.08248 .  
Daily.opioid.intake    5.3   1  0.2727   0.60214    
Mental.health        120.7   1  6.2028   0.01366 *  
Pain.intensity        39.7   1  2.0381   0.15514    
Disability           362.7   1 18.6392 2.604e-05 ***
Support.oss3           2.8   1  0.1447   0.70414    
Stigma.total         345.1   1 17.7389 3.999e-05 ***
Residuals           3502.2 180                      
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Support

Step I: Support and demographic/clinical variables

Model Setup

Model.stageII.sup.stepI <- lm(Support.oss3 ~ Age + Gender + Ethnicity +
                                Relationship + Employment +
                                Pain.loc + Pain.duration +
                                Beliefs.organic + Beliefs.psych +
                                Daily.opioid.intake + Mental.health +
                                Pain.intensity + Disability + Depression.phq9,
                              data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.523941 3.055768 1.588692 0.3962057 0.3272500 0.4695501 Low (< 5)
Gender 1.399734 1.663900 1.183104 0.7144215 0.6009978 0.8060097 Low (< 5)
Ethnicity 1.677963 2.002971 1.295362 0.5959605 0.4992583 0.6857427 Low (< 5)
Relationship 3.461084 4.227770 1.860399 0.2889268 0.2365313 0.3476438 Low (< 5)
Employment 3.482213 4.254213 1.866069 0.2871737 0.2350611 0.3456195 Low (< 5)
Pain.loc 3.037080 3.697266 1.742722 0.3292636 0.2704701 0.3939361 Low (< 5)
Pain.duration 1.221985 1.462060 1.105434 0.8183406 0.6839667 0.9036303 Low (< 5)
Beliefs.organic 1.479691 1.760122 1.216425 0.6758170 0.5681424 0.7676256 Low (< 5)
Beliefs.psych 1.215728 1.455697 1.102601 0.8225522 0.6869563 0.9073369 Low (< 5)
Daily.opioid.intake 1.335938 1.588631 1.155828 0.7485376 0.6294729 0.8391207 Low (< 5)
Mental.health 1.439863 1.711997 1.199943 0.6945107 0.5841133 0.7863232 Low (< 5)
Pain.intensity 2.003124 2.406163 1.415318 0.4992201 0.4155995 0.5828844 Low (< 5)
Disability 2.810258 3.413607 1.676383 0.3558392 0.2929452 0.4241362 Low (< 5)
Depression.phq9 2.251749 2.715959 1.500583 0.4440992 0.3681940 0.5227065 Low (< 5)

Fit Statistics

Parameter Fit
33
34 R2 0.241
35 R2 (adj.) 0.111
36 Sigma 2.282

Step II: Explaining support with stigma

Model Setup

Model.stageII.sup.stepII <- lm(Support.oss3 ~ Age + Gender + Ethnicity +
                                 Relationship + Employment +
                                 Pain.loc + Pain.duration +
                                 Beliefs.organic + Beliefs.psych +
                                 Daily.opioid.intake + Mental.health +
                                 Pain.intensity + Disability + Depression.phq9 +
                                 Stigma.total,
                               data = Datasource)

Diagnostics

Residuals plot.

Multicollinearity statistics.

Term VIF VIF_CI_high SE_factor Tolerance Tolerance_CI_low Tolerance_CI_high Label
Age 2.533618 3.064426 1.591734 0.3946925 0.3263254 0.4674443 Low (< 5)
Gender 1.412414 1.677169 1.188450 0.7080077 0.5962429 0.7992511 Low (< 5)
Ethnicity 1.697443 2.024804 1.302860 0.5891215 0.4938750 0.6781255 Low (< 5)
Relationship 3.608561 4.407256 1.899621 0.2771188 0.2268985 0.3336572 Low (< 5)
Employment 3.627052 4.430371 1.904482 0.2757060 0.2257147 0.3320215 Low (< 5)
Pain.loc 3.116994 3.792903 1.765501 0.3208219 0.2636503 0.3839272 Low (< 5)
Pain.duration 1.222125 1.460326 1.105498 0.8182466 0.6847788 0.9031919 Low (< 5)
Beliefs.organic 1.494031 1.775569 1.222306 0.6693301 0.5631997 0.7606334 Low (< 5)
Beliefs.psych 1.229147 1.467591 1.108669 0.8135725 0.6813886 0.8990419 Low (< 5)
Daily.opioid.intake 1.347210 1.599955 1.160694 0.7422749 0.6250175 0.8326815 Low (< 5)
Mental.health 1.444140 1.715242 1.201724 0.6924538 0.5830082 0.7838250 Low (< 5)
Pain.intensity 2.007919 2.409486 1.417010 0.4980281 0.4150264 0.5811387 Low (< 5)
Disability 3.110817 3.785186 1.763751 0.3214589 0.2641878 0.3846552 Low (< 5)
Depression.phq9 2.507367 3.031673 1.583467 0.3988248 0.3298509 0.4720618 Low (< 5)
Stigma.total 2.244812 2.704312 1.498270 0.4454716 0.3697799 0.5237802 Low (< 5)

Model Summary

Parameter Coefficient CI (low) CI (high) Std. Coefficient Fit
(Intercept) 14.120 10.449 17.790 0.022
Age 0.006 -0.027 0.040 0.037
Gender [Female] -0.028 -1.258 1.202 -0.012
Gender [Other] 1.118 -2.322 4.558 0.462
Ethnicity [Māori] -0.299 -1.465 0.868 -0.124
Ethnicity [Pacific] 1.438 -1.037 3.913 0.594
Ethnicity [Asian] 1.964 -0.378 4.306 0.812
Relationship [In a relation] -0.457 -1.579 0.664 -0.189
Relationship [Married/De facto] 0.439 -0.470 1.348 0.182
Relationship [Divorced/sep.] 0.140 -1.189 1.470 0.058
Relationship [Widowed] 0.385 -2.102 2.872 0.159
Employment [Unemployed/benefit] -0.682 -1.531 0.167 -0.282
Employment [Work at home] -0.394 -1.907 1.119 -0.163
Employment [Retired] -0.589 -2.124 0.945 -0.244
Employment [Student] -0.979 -2.420 0.463 -0.404
Employment [Other] -0.878 -2.856 1.099 -0.363
Pain loc [Neck/back] 0.137 -1.198 1.472 0.057
Pain loc [Upper limb] 0.281 -1.145 1.706 0.116
Pain loc [Chest] -1.688 -3.994 0.618 -0.698
Pain loc [Abdomen/pelvis] -0.471 -1.898 0.957 -0.195
Pain loc [Hips/buttocks] 0.067 -1.436 1.571 0.028
Pain loc [Lower limb] 0.637 -0.825 2.100 0.263
Pain loc [Other/widepread] 0.412 -1.282 2.105 0.170
Pain loc [Missing] 3.347 0.364 6.330 1.383
Pain duration -0.389 -0.855 0.078 -0.114
Beliefs organic 0.015 -0.045 0.076 0.038
Beliefs psych -0.005 -0.089 0.080 -0.008
Daily opioid intake 0.218 -0.104 0.540 0.097
Mental health [No diagnosis] -0.087 -0.882 0.709 -0.036
Pain intensity -0.046 -0.314 0.223 -0.030
Disability -0.071 -0.335 0.194 -0.058
Depression phq9 -0.014 -0.088 0.059 -0.038
Stigma total -1.771 -2.695 -0.847 -0.354
R2 0.297
R2 (adj.) 0.172
Sigma 2.202

Hypothesis Tests

Anova Table (Type II tests)

Response: Support.oss3
                    Sum Sq  Df F value    Pr(>F)    
Age                   0.68   1  0.1411 0.7075976    
Gender                2.30   2  0.2372 0.7890858    
Ethnicity            20.80   3  1.4297 0.2356452    
Relationship         14.50   4  0.7472 0.5610590    
Employment           20.22   5  0.8337 0.5273360    
Pain.loc             64.92   8  1.6732 0.1076448    
Pain.duration        13.13   1  2.7066 0.1016800    
Beliefs.organic       1.20   1  0.2482 0.6189793    
Beliefs.psych         0.06   1  0.0120 0.9130059    
Daily.opioid.intake   8.68   1  1.7890 0.1827308    
Mental.health         0.22   1  0.0461 0.8302206    
Pain.intensity        0.54   1  0.1117 0.7386034    
Disability            1.36   1  0.2795 0.5976744    
Depression.phq9       0.70   1  0.1447 0.7041442    
Stigma.total         69.31   1 14.2919 0.0002128 ***
Residuals           872.96 180                      
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Exploratory Path Analysis

Model Setup

Model <- '
Stigma            ~ Unemployment + Daily.opioid.intake + Mental.health + Beliefs.organic
Social.support    ~ Stigma
Disability        ~ Relationship + Beliefs.organic + Stigma + Pain.location + Pain.intensity
Depression        ~ Mental.health + Stigma + Disability
Depression        ~~ 0*Social.support
'

Model Diagram

Model Summary

lavaan 0.6-12 ended normally after 1 iterations

  Estimator                                         ML
  Optimization method                           NLMINB
  Number of model parameters                        17

  Number of observations                           215

Model Test User Model:
                                                      
  Test statistic                                41.438
  Degrees of freedom                                21
  P-value (Chi-square)                           0.005

Model Test Baseline Model:

  Test statistic                               482.493
  Degrees of freedom                                34
  P-value                                        0.000

User Model versus Baseline Model:

  Comparative Fit Index (CFI)                    0.954
  Tucker-Lewis Index (TLI)                       0.926

Loglikelihood and Information Criteria:

  Loglikelihood user model (H0)              -1583.040
  Loglikelihood unrestricted model (H1)      -1562.321
                                                      
  Akaike (AIC)                                3200.079
  Bayesian (BIC)                              3257.380
  Sample-size adjusted Bayesian (BIC)         3203.510

Root Mean Square Error of Approximation:

  RMSEA                                          0.067
  90 Percent confidence interval - lower         0.036
  90 Percent confidence interval - upper         0.097
  P-value RMSEA <= 0.05                          0.161

Standardized Root Mean Square Residual:

  SRMR                                           0.051

Parameter Estimates:

  Standard errors                             Standard
  Information                                 Expected
  Information saturated (h1) model          Structured

Regressions:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
  Stigma ~                                                              
    Unemployment      0.205    0.065    3.160    0.002    0.205    0.195
    Daily.opid.ntk    0.057    0.027    2.086    0.037    0.057    0.126
    Mental.health     0.284    0.065    4.378    0.000    0.284    0.264
    Beliefs.organc    0.022    0.005    4.585    0.000    0.022    0.274
  Social.support ~                                                      
    Stigma           -2.028    0.310   -6.540    0.000   -2.028   -0.407
  Disability ~                                                          
    Relationship     -0.650    0.184   -3.533    0.000   -0.650   -0.168
    Beliefs.organc    0.037    0.017    2.185    0.029    0.037    0.116
    Stigma            1.567    0.201    7.789    0.000    1.567    0.393
    Pain.location    -0.328    0.186   -1.758    0.079   -0.328   -0.084
    Pain.intensity    0.527    0.062    8.477    0.000    0.527    0.429
  Depression ~                                                          
    Mental.health     2.649    0.705    3.758    0.000    2.649    0.188
    Stigma            5.226    0.746    7.005    0.000    5.226    0.399
    Disability        1.076    0.183    5.877    0.000    1.076    0.328

Covariances:
                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 .Social.support ~~                                                      
   .Depression         0.000                               0.000    0.000

Variances:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
   .Stigma            0.175    0.017   10.368    0.000    0.175    0.744
   .Social.support    4.853    0.468   10.368    0.000    4.853    0.834
   .Disability        1.782    0.172   10.368    0.000    1.782    0.477
   .Depression       19.350    1.866   10.368    0.000   19.350    0.482

R-Square:
                   Estimate
    Stigma            0.256
    Social.support    0.166
    Disability        0.523
    Depression        0.518

References

Epskamp, S. (2019). semPlot: Path diagrams and visual analysis of various SEM packages’ output. Retrieved from https://CRAN.R-project.org/package=semPlot

Fox, J., & Weisberg, S. (2019). An R companion to applied regression (Third). Retrieved from https://socialsciences.mcmaster.ca/jfox/Books/Companion/

Makowski, D., Ben-Shachar, M. S., Patil, I., & Lüdecke, D. (2021). Automated results reporting as a practical tool to improve reproducibility and methodological best practices adoption. CRAN. Retrieved from https://github.com/easystats/report

McElreath, R. (2018). Statistical rethinking: A bayesian course with examples in r and stan. Chapman; Hall/CRC.

R Core Team. (2021). R: A language and environment for statistical computing. Retrieved from https://www.R-project.org/

Rosseel, Y. (2012). lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48(2), 1–36. Retrieved from https://www.jstatsoft.org/v48/i02/

Van de Schoot, R., Lugtig, P., & Hox, J. (2012). A checklist for testing measurement invariance. European Journal of Developmental Psychology, 9(4), 486–492.

Wickham, H. (2016). ggplot2: Elegant graphics for data analysis. Retrieved from https://ggplot2.tidyverse.org

Wickham, H., François, R., Henry, L., & Müller, K. (2021). Dplyr: A grammar of data manipulation. Retrieved from https://CRAN.R-project.org/package=dplyr

pain-stigma's People

Contributors

gallvp avatar

Watchers

avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.