The results of this study indicate that there may be some association between average numbers of steps per day and psychological symptoms, with increased levels of fear of movement and pain catastrophizing associated with a reduction in the number of steps. Higher self-efficacy may be associated with increased activity; however our results suggest depressive symptoms have a weak association with mobility-related activity.
Our results are consistent with previous research demonstrating greater fear of movement [11, 29–32] and greater pain catastrophizing [14, 29] are associated with less physical activity among several populations including patients with knee OA. Our quantitative findings are also supported by the fear-avoidance model [33], whereby pain catastrophizing and pain-related fear of movement are theorized to lead to avoidance of physical activity [33, 34]. Further, a qualitative study has demonstrated that fear of pain is a psychological barrier to engaging in physical activity in people with knee OA [16]. Our findings suggest a small relationship (16% of the variability) which is not unsurprising given the great number of variables that can be associated with an individual’s physical activity behaviour [35, 36].
In our study, fear of movement and pain catastrophizing were associated with activity after controlling for pain levels. This suggests that one’s personal attitude toward pain, rather than pain severity itself, is related to physical activity. However, other findings in the OA literature related to pain are conflicting. Some studies have demonstrated that higher pain is associated with lower physical activity in patients with knee OA [37, 38] while others have not [39]. Lazaridou et al. [14] similarly reported that the association between physical activity levels and pain intensity was moderated by pain catastrophizing. Interestingly, physical activity levels (measured via accelerometry) have not been found to increase following total joint replacement despite substantial improvements in pain [40].
Lower self-efficacy for managing OA pain and other symptoms may be related to lower physical activity levels as the standardized coefficient suggested a similar-sized relationship (0.16) as fear of movement and pain catastrophizing, but the confidence interval around the standardized coefficient included zero indicating a lower level of confidence in the existence of a relationship (Table 2). Overall self-efficacy has previously been shown to be is a significant determinant of health behavior [41]. Another study found that self-efficacy could be directly targeted by treatment to improve physical function for individuals with early knee OA [42]. Therefore, increasing self-efficacy for OA symptoms may also be a clinically important intervention goal for people with knee OA and warrants testing in a randomised controlled trial.
There is some conflict in the literature about the association between depressive symptoms and physical activity in the general population [43–45] and in people with knee OA [12, 13]. This may relate to differences in cohort characteristics and in study methodology including the measurement of depressive symptoms and defining and assessing physical activity. Our results showed no statistically significant association between depression levels and average daily steps. However, people who scored more than 21 on the depression subscale of the DASS [20] were excluded from the study itself, which constrained the spread of the data and may have influenced the results. Nonetheless, our finding of no statistically significant relationship concurs with that of two systematic reviews in people with knee OA [12, 13].
Although exploratory, our results suggest that fear of movement, pain catastrophizing and possibly self-efficacy for symptom management, are associated with physical activity behaviour, and therefore we speculate that efforts to increase walking behavior in this patient population might be enhanced by strategies aimed at reducing fear of movement and pain catastrophizing, and improving self-efficacy for symptom management. There is evidence that specific pain neurophysiology education can reduce pain catastrophizing and increase knowledge about pain in people with chronic pain [46]. One of the primary aims of such education is to reconceptualise thinking about pain, away from the belief that “hurt” always equates to “physical harm”. Whether pain neurophysiology education subsequently increases physical activity levels has not yet been studied. Other psychological interventions may also be beneficial. Pain coping skills training, a form of cognitive behavioral therapy, has been investigated in people with knee OA. Bennell, et al. [47] found that pain coping skills training reduced pain catastrophizing but did not increase general physical activity levels unless it was combined with a structured exercise program. Further clinical trials are needed to investigate whether improvements in these psychological parameters mediate improvements in physical activity levels following targeted interventions.
Our study has several limitations. First, as this is a cross-sectional study, we cannot establish the temporal relationship between the psychological characteristics and physical activity levels. It is possible that being more physically active might lead to lower fear of movement and pain catastrophizing. The causal relationship needs to be investigated in randomized controlled trials. Second, we enrolled people classified as sedentary or insufficiently physically active based on the Active Australia Survey [19], a self-report instrument. This could have potentially constrained our ability to detect relationships in those with higher physical activity levels. However, based on our objective measure of physical activity, the average number of daily steps of the participants was indeed relatively high, around 8000. This would be deemed “somewhat active” [48]. A possible reason why participants had higher steps/day than expected is that wearing an accelerometer to measure daily steps might motivate participants to be more physically active. This might confound true relationships between psychological status and physical activity levels. Third, participants were recruited for a clinical trial investigating an intervention that included physical activity. As such, results may not necessarily generalize to all people with knee OA as psychological characteristics may differ in those who volunteer for research of this nature.