Association of depressive symptoms and risk of knee pain: the moderating effect of sex

BACKGROUND: Depression has been shown in some studies to be associated with knee pain. Females were widely recognized as more vulnerable to depression and knee pain than males. However, the role of sex in this correlation was under-researched. This study aimed to investigate the association between depressive symptoms and subsequent knee pain, as well as whether and how sex would moderate this association based on a four-wave (Wave 1 in 2010–2011, Wave 2 in 2013, Wave 3 in 2015, and Wave 4 in 2018) longitudinal study among middle-aged and elderly Chinese adults. METHODS: Seventeen thousand seven hundred eight participants were recruited and followed in the China Health and Retirement Longitudinal Study (CHARLS). Ten thousand four hundred fifty-one entered the final analysis based on the inclusion and exclusion criteria. Knee pain was assessed by self-report. Depressive symptoms were evaluated using the validated 10-item Center for Epidemiological Studies-Depression Scale (CESD-10). Cox proportional hazards models were used to calculate hazard ratios with 95% confidence intervals (CIs) after controlling potential confounders to examine the association between depressive symptoms and subsequent incident and persistent knee pain. Non-linear association of depressive symptoms score (CESD-10) and risk of knee pain was also investigated via applying 3-knotted restricted cubic spline regression. An interaction term of depressive symptoms status and sex was added to investigate the moderating effect of sex on the relationship between depressive symptoms status and the risk of knee pain. RESULTS: The median follow-up time was seven years for all the outcomes. Participants with depressive symptoms were 1.45 times (95% CI: 1.34–1.56) and 2.16 times (95% CI: 1.85–2.52) more likely to develop the incident and persistent knee pain after multivariable were adjusted, compared with those without depressive symptoms. There was a non-linear association between CESD-10 score and risk of knee pain. Compared with females, males had an enhanced correlation between depressive symptoms status and knee pain (multivariable-adjusted HR: 1.22, 95% CI: 1.05–1.42 and HR: 1.57, 95% CI: 1.14–2.17 for the incident and persistent knee pain, respectively). CONCLUSION: Depressive symptoms are independently associated with an excess risk of knee pain, with a stronger correlation for males than females among middle-aged and elderly Chinese adults. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12891-021-04511-2.