Radiographic knee osteoarthritis severity has no impact on fall risk: the locomotive syndrome and health outcomes in the aizu cohort study (LOHAS): a cross-sectional study

Falls are a leading cause of injury-related morbidity and mortality among older adults, with more than one in four older adults falling each year [1]. Costs of care after falls in the elderly are burdensome, reaching over $31 billion dollars in the United States in 2015 [2]. Falls may also lead to a fear of falling, decreased physical function, limited physical activity, loss of independence, and a higher mortality risk [3, 4].

Knee osteoarthritis (KOA) is a major public health problem associated with aging in musculoskeletal disorders [5]. The prevalence of radiographic KOA in adults aged ≥ 60 years in the U.S. was estimated to be 37% [6]. Common symptoms of KOA include knee pain, stiffness, swelling, decreased mobility, muscle weakness, and difficulty walking and climbing stairs [7]. Pain, dysfunction, and disability are caused by progressive loss of articular cartilage [8] and synovitis [9] of the knee joint, and the progression of KOA tends to lead to increased physical limitations, pain, and functionality restrictions [10].

An association between KOA and the risk of falls has been reported in cross-sectional studies [11‐14], prospective cohorts [15‐18], and longitudinal studies [19‐23] (Table 1), but the results are inconsistent and controversial [17]. Although the diagnosis of KOA is often based on the combination of plain radiographs and subjective symptoms, several studies have described the involvement of psychological factors such as depression in the severity of symptoms and physical function in patients with KOA [24]. The need to consider psychological factors in assessing the association between KOA and falls has been addressed [13], but we found no published investigation of the impact of radiographic KOA severity on fall risk that also takes into account psychological factors. Even with the same radiographic KOA severity, pain and physical function vary from case to case. Prior studies have shown that the radiographic KOA severity itself is associated with falls [14, 15, 20]. By considering a variety of confounding factors, including psychological factors that have not been examined previously, this would be a new finding to determine whether the radiographic KOA severity itself is a risk for falls. Furthermore, if the radiographic KOA severity itself is not a risk for falls, this finding sends an important clinical message that falls can be prevented by improving factors other than the radiographic KOA severity, which are difficult to improve with current medicine. We conducted the present study to (i) investigate factors that have an impact on the risk of falls, and (ii) clarify the impact of the radiographic KOA severity on the risk of falls.

Our findings revealed that knee pain, decreased mobility, and depressive mood were significantly associated with a fall history. On the other hand, the radiographic KOA severity was not associated with a fall history.

Osteoarthritis has been considered an established risk factor for falls [35]. Regarding KOA, several studies [11‐23] have obtained inconsistent results, based on different diagnoses and classifications of KOA. We analyzed the influence of radiographic KOA itself on the risk of falls, including as covariates factors that have been reported as a correlation or risk factor for falls (i.e., gender, pain, limited physical ability, and imbalance)^11–23 and a psychological factor (depression) [13]. Falls exacerbate knee pain [20]. In this study, to minimize the impact of falls on regular knee pain, we assessed knee pain after the falls. Our present investigation is the first to analyze the impact of radiographic KOA on falls, taking into account not only the symptoms of KOA but also psychological factors and physical function. Our results corroborate and extend the prior studies' findings.

According to the present results, having a fall history was significantly associated with knee pain. Knee pain and other symptoms have been reported to be risk factors for falls among individuals with KOA [11, 12, 18, 20‐22], and knee pain was significantly associated with falls in the present study as well. Pain related to OA leads to decreased use of the affected limbs and joints, muscle weakness, and poor functional performance, resulting in an increased risk of falls [16]. However, in KOA, the degree of knee pain is not necessarily related to radiographic severity [36]. In addition, we detected no significant association between radiographic KOA severity and fall history in a multivariate logistic analysis. In clinical practice, the radiographic severity tends to be more focused on the diagnosis of KOA, but we must not forget to properly evaluate knee pain when assessing its relationship to falls. Regardless of the severity of the imaging findings, the treatment of knee pain may improve an individual's risk of falling.

Our analyses revealed that decreased mobility was significantly associated with fall history. Decreased mobility has been reported as a risk factor for falls [37], which is consistent with our present observations. Our present univariate analysis detected a significant association between balance disability and fall history, but the multivariate analysis did not. In a multivariate analysis in which both OLST and TUGT variables were entered separately, we also found that decreased mobility was significantly associated with falls, but balance disability was not significantly associated with falls. This suggests that the association between balance disability and fall history was confounded by other factors. The OLST assesses static balance and is considered a strong predictor of falls [33], but the parameters of the OLST may vary between different testing sessions for the same individuals [38]. In this study, statistically, there was a significant difference in OLST between the positive fall experience group and no fall group, with lower mean values in the positive fall experience group, but the SD was larger in the positive fall experience group than in the no fall group. In other words, we showed that OLST in the positive fall experience group varied widely among individuals. Therefore, OLST may not have been extracted as a factor significantly related to falls in logistics analysis. Further, factors such as fatigue can influence the balance ability [38]. Although static balance is a clinically important factor in the risk of falling, this limited reproducibility of the OLST may have led to the lack of a significant association in the present study.

Our analyses also demonstrated a significant association between depression and fall history. Individuals with more than moderate depressive symptoms are more likely to have a fear of falling than those without depression [39], and such symptoms were reported to be associated with an increased risk of falls in the future [40]. We speculate that a fear of falling may increase in individuals with KOA as their depression worsens, eventually contributing to falls to the same degree as decreased gait and balance functions. In addition, depression is considered a potential mediator of knee pain and falls [41]. Our results indicated a significantly weak positive correlation and correlation with little clinical meaning between them (correlation coefficient; 0.2, (p)-value of < 0.0001). It should be noted that knee pain was significantly associated with falls in this study as well, and that depression itself is not only associated with falls, but may also influence falls through the mediation of knee pain. Prior research indicated that psychological factors (e.g., depression) and cognitive factors should be considered when assessing the association between KOA and falls [13]. Our present findings suggest that assessing and treating psychological factors might be important in reducing the risk of falls as well as improving pain.

We also analyzed excluding no OA pattern to account for the effects of depression and decreased mobility not associated with KOA. Our results showed that knee pain, depression, and decreased mobility affected the risk of falling, regardless of the presence or absence of a radiographic KOA.

Our results also demonstrated no significant direct relationship between radiographic KOA severity and fall history. However, it has previously been reported that the risk of falling increases with more severe KOA [20], which contradicts our results. Although KOA is usually diagnosed based on radiographic severity, symptoms of KOA do not always match imaging findings [36, 42]. Similarly, the risk of falls in KOA should not be based on imaging findings alone, but additionally include the patient's clinical symptoms, knee pain, physical function, and psychological factors.

Despite being considered risks for falls in previous studies, gender [12, 18, 19], crepitus [22], and sleeping pill use [26] were not significantly associated with fall history in the present study. it has been indicated that differences in fall rates by gender may reflect differences in lifestyle [43]. Because the subjects in this study were residents of a limited rural region, their lifestyles may not have varied significantly between gender. Crepitus is reported to be poorly understood, especially among the elderly [44]. In this study, self-reports of knee crepitus may have been lower than actual. In addition, since this study was conducted at the same time as the annual health checkups, subjects may have been concerned about their health. Thus, it is possible that there were more healthy subjects and that sleeping pill use was not significantly associated with fall history. Therefore, gender, crepitus, and sleeping pill use may not have been significantly associated with fall history.

Although clinical symptoms and image findings of KOA patients are often the clinical focus, we must not forget to evaluate psychological factors (such as depressive symptoms) and physical function. Regardless of the radiographic KOA severity, the treatment of knee pain and treatment for depression and reduced mobility may reduce the risk of falls. Our present findings provide key points for clinical practice because they show the importance of therapeutic interventions for knee pain, depression, and mobility in preventing falls in the elderly. Further research and intervention studies are required to test our present findings.

Several study limitations must be addressed. First, because this study was a cross-sectional analysis, a causal relationship cannot be determined. Second, we were unable to include polypharmacy, malnutrition, smoking, alcohol consumption, stroke, and Parkinson's disease, which are commonly considered risk factors for falls²⁶, in our analysis. Third, we did not investigate the subjects' detailed history of treatment for KOA, which might have affected their symptoms. Fourth, in this study, because plain radiographs of the knee, responses to the questionnaire, and physical function assessment were optional, subjects enrolled in the study differed significantly from those excluded in demographic and clinical variables. Many symptomatic or health-conscious subjects may have participated, possibly biasing the study. The generalizability of the results must be carefully considered. Finally, a subject's fall history may be more excessive or may be insufficient compared to the actual fall history, due to recall bias.

In conclusion, the results of this study demonstrated that knee pain, decreased mobility, and depression were significantly associated with a history of falling and hence, a risk of future falls. However, the radiographic KOA severity was not significantly associated with fall history. Regardless of the patients' radiographic KOA severity, their risk of falling may be reduced by treating their knee pain, improving their mobility function, and addressing psychological factors.