Content uploaded by Paul Campbell
Author content
All content in this area was uploaded by Paul Campbell on Sep 07, 2015
Content may be subject to copyright.
Structural Equation Modelling of the Moderation Effect of Health Locus of Control on the
Pain-Depression Pathway
Analysis of the critical ratio values (Table 2), and pathway coefficients (Figures 1 and 2)
show:
A significant moderation effect of HLOCi on the bothersomeness to depression
pathway, with an almost doubling of strength of association for those with low internality
compared to those with high levels of internality
A non significant trend (p = 0.07) on the pathway between pain interference and
bothersomeness, suggestive that pain interference plays a lessor role in judgements of
bothersomeness for those with low HLOCi
Funding: This research was supported by the Wellcome Trust [083572]
Kate Hope, Kate M Dunn, Paul Campbell
Aim
i. to construct a viable pain/disability to depression pathway model
ii. to test the moderation effect of HLOCi on the pain/disability-depression pathway
Introduction
Low back pain (LBP) is common, and is a major health concern
Psychological consequences of LBP, such as depression, are significant barriers
to recovery
Recent research has shown that the mechanisms of how psychological factors
exert their influence on LBP is unclear, with evidence of considerable variation in
what psychological factors are important, and evidence of conceptual overlap1
This may be suggestive of an underlying latent factor, or mechanisms that
underpin and determine psychological expression
One potential mechanism is an individuals’ Health Locus of Control internality
(HLOCi) Individuals with lower levels of HLOCi believe their health is
beyond the control of their own actions, and see themselves as
passive agents in the management of their health2
Method
Setting/procedure - Cross sectional study of participants (n = 637) who had taken
part in two longitudinal cohort studies3,4. In both studies patients who had consulted
their General Practitioner (GP) about back pain were invited to take part.
Participants received questionnaires at regular intervals over a period of 12 months
and were then followed up years later. This current analysis used data gathered at
their long term follow up of 7 years3 and 5 years4
Measures
Outcome: Depression (Hospital Anxiety and Depression Scale, HADS)
Predictors:
oPain intensity (0 to 10 scale of average, lowest and current pain combined)
oDisability (Roland Morris Disability Questionnaire, RMDQ)
oPain interference (single item, 0 to 10 scale)
oBothersomeness (single Item, 5 point Likert scale)
Moderator: HLOC Internality Scale (5 questions, 6 point Likert scale, upper and
lower quartile groups compared)
Analysis
Structural Model: Pain intensity and disability were placed as exogenous
predictors, pain interference as a endogenous mediator and bothersomeness as
an endogenous predictor with adjustment for age and gender (see Figures 1and
2)
Premise for the model is that depression will manifest when pain and disability
are perceived as interfering and bothersome
Model fit was assessed using Comparative Fit Index (CFI), Goodness of Fit
Index (GFI) and Root Mean Square Error Approximation (RMSEA)
Moderation analysis was carried out using multi group analysis on AMOS version
21
Pairwise Comparison Tests were carried out on pathway coefficients between
HLOCi Models (i.e. low and high internality) to determine critical ratio differences.
Pathway coefficients that differ between models (≥ 1.96 critical ratio difference)
are significant (p ≤ 0.05)
Contact
p.campbell@keele.ac.uk
Mean (SD) Median IQR
Percentage
Age
58.8 (8.8) 57 14
Gender (Female)
61.9%
HADS Depression Scale
4.9 (4.0) 4 5
Pain intensity
2.8 (2.7) 2 4
RMDQ
5.7 (6.1) 3 8
Pain interference
3.1 (2.9) 2 5
Bothersomeness
2.4 (1.2) 2 2
SD
– Standard Deviation, IQR – Inter Quartile Range
Table 1. Participant characteristics
Table 2. Critical ratio (CR) parameter test
Results
Model fit indices indicated an acceptable fit for both independent high internality and low
internality models, and for the multi-group model (CFI > 0.95, GFI > 0.94, RMSEA <
0.09)
Figure 1. High internality pathway model (standardized
beta coefficient values shown)
Conclusion
HLOCi significantly moderates the pain/disability to depression pathway in those
who report back pain
People who report having a low level of control over their own general health report
greater levels of depressive symptoms in relation to their pain/disability
This finding may signify a potential factor that may predict depression among people
with pain, and could potentially be a target for intervention, for example increasing
perceptions of control leading to improved self-management
Further prospective work is now required to track the influence of HLOC beliefs on the
development of depression in those with back pain
References
1) Campbell, P., Bishop, A., Dunn, K. M., Main, C. J., Thomas, E., & Foster, N. E. (2013). Conceptual overlap of psychological
constructs in low back pain. Pain, 154(9), 1783-1791.
2) Wallston, B. S., & Wallston, K. A. (1978). Locus of control and health: a review of the literature. Health Education &
Behavior, 6(1), 107-117.
3) Dunn, K. M., Jordan, K., & Croft, P. R. (2006). Characterizing the course of low back pain: a latent class analysis. American
Journal of Epidemiology, 163(8), 754-761.
4) Foster, N. E., Bishop, A., Thomas, E., Main, C., Horne, R., Weinman, J., & Hay, E. (2008). Illness perceptions of low back
pain patients in primary care: what are they, do they change and are they associated with outcome?. Pain,136(1), 177-187.
Research Institute for Primary Care & Health Sciences, Keele University, Keele,
Staffordshire, ST5 5BG, UK
Figure 2. Low internality pathway model (standardized
beta coefficient values shown)
Pathway CR Value p value (2 tailed)
Pain intensity to pain interference
1.66 0.10
Pain interference to
Bothersomeness 1.82 0.07
Pain Intensity to
Bothersomeness 1.30 0.20
Disability to
Bothersomeness 0.12 0.90
Bothersomness
to Depression 2.84 0.005
