In this study, the prevalence of both multimorbidity and depression was high and social support was associated with multimorbidity and depression. Hypertension, cardiovascular disease, CVD, and cataract had strong associations with depression when they co-occurred with other conditions. Compared with having a single disease, there were more diseases associated with depression in a multimorbid state. In addition, three patterns of multimorbidity were identified: a musculoskeletal pattern, cardiometabolic pattern, and degenerative disease pattern. Among these, the cardiometabolic and degenerative disease patterns were associated with a higher risk of the presence of depressive symptoms.
The prevalence of multimorbidity in this study was 64.7%, which is in line with the findings of Hu that older people in Chinese communities have rates of multimorbidity between 6.4% and 76.5%(8); however, our results were higher than the 41.56% found in communities of eastern China(32). Depression is known to be the most common psychological disorder among older adults. In the United States, approximately 15–20% of older adults experience depression each year(33). Studies that have investigated the prevalence of depression among older people in mainland China have reported depression rates ranging from 15.8–30.8%(34–36), probably owing to the use of different methodologies. The prevalence of depression was 64.6%. This may be because our participants were covered by long-term care insurance and had poor physical conditions, with most over age 80 years. Their physical functions were greatly reduced, but multimorbidity may share common pathologic pathways with mental disorders(37, 38). The impact of social support on multimorbidity and depression has been reported. Social support is negatively related to depression(39, 40). In our research as well, social support scores were lower for people with multimorbidity and depression. Impaired social support and feelings of loneliness are considered to be risk factors for depression in older people(41, 42). When we performed regression analysis, social support was included as an adjustment variable, taking full account of the impact of social support on multimorbidity and depression.
Previous studies concerning the associations between chronic disease and depression have mostly focused on a single condition or the number of comorbid conditions. A large volume of literature has demonstrated that depression is associated with several somatic conditions such as hypertension, heart diseases, and stroke(43–45), which was also observed in our study. Indeed, the simultaneous presentation of diseases in one individual may have a multiplicative, rather than additive, effect on the individual’s health(46). Thus, compared with those who have no chronic diseases or without multimorbidity, people with multimorbidity are indeed more prone to depressive symptoms.
The multimorbidity patterns in this study were similar to those observed previously, and common antecedents and disease pathways may explain these patterns. A systematic review found three major multimorbidity patterns among older adults living in Western countries, which are cardiovascular–metabolic, mental health, and musculoskeletal disorders(28). In China, a cohort study also identified four multimorbidity patterns: a cardiometabolic pattern, respiratory pattern, arthritic–digestive–visual pattern, and hepatic–renal–skeletal pattern(31). In our research, the musculoskeletal model includes chronic low back pain, rheumatoid arthritis, and intervertebral disc disease. Previous studies have also found the existence of this model, and all of them included at least one musculoskeletal disease(28). Musculoskeletal patterns or arthritis included in these patterns are often associated with the occurrence of depressive symptoms(47–49). In this study, there may be no correlation between the two owing to the low prevalence of related diseases; however, this warrants further investigation. The cardiometabolic pattern is the most widely described pattern in previous studies. The diseases included in this model differ in each study, but the core diseases are the same, such as heart disease, hypertension, diabetes, and others(28, 50). Moreover, the relationship between this model and depression has also been confirmed in many studies(13). Among the three multimorbidity patterns found, the cardiometabolic pattern had the strongest association with depression, which is consistent with previous studies. Degenerative diseases include cataract, severe vision loss, and osteoporosis. In previous studies, multimorbidity patterns including joint and eye diseases have been found(9, 31, 51, 52). Such multimorbidity patterns are not scarce in such studies; previous evidence supports an association between conditions in this pattern through factors such as inflammation, side effects of medications, and so on(53, 54). The strong correlation between bone diseases and eye diseases may permit the discovery of potential links. The differences in multimorbidity patterns across studies might be partly attributable to remarkable heterogeneity in the number, type, and assessment approach of chronic conditions as well as in characteristics of the study samples. However, common components have been identified.
In estimation of the three multimorbidity patterns and depression, two were found to be associated with depression, namely, the cardiometabolic and degenerative disease patterns. In addition, strong correlations were found between cardiovascular diseases and some degenerative diseases and depression in previous studies(55, 56). Further, the cardiovascular–degenerative disease pattern has been identified, and this pattern has the strongest association with poor HRQL and is the only pattern associated with poor HRQL in the mental health dimension(9). Some confirmed multimorbidity patterns and their overlap may indicate common underlying pathological mechanisms(52). In future research, these connections should be considered.
This is the first study of the association of multimorbidity patterns and depression among people covered by China's long-term care insurance. Our findings may be helpful for the improvement of China's long-term care insurance system and the optimization of service content. In addition, exploratory factor analysis was used to explore the multi-disease model and the oblimin rotation method was used; factors were allowed to be associated with each other, which is helpful in researching comorbidity between diseases(29, 57). Identifying specific disease patterns related to depression can help to improve understanding of the impact of multimorbidity on depression, not only focusing on the physical health of older people but also on mental health, which is equally important.
Our research also has some limitations. First, the cross-sectional nature of the study design does not allow causal inference of the observed associations. Thus, caution is needed when interpreting the findings. Second, some chronic diseases were ascertained based only on self-reported information, such that differential recall bias cannot be ruled out, with more severe diseases being more likely to be reported. Third, we did not consider the impact of disease severity; future research should incorporate more objective diagnoses and take disease severity into consideration.