Osteoarthritis and functional disability: results of a cross sectional study among primary care patients in Germany

Disability has a substantial impact on the quality of life of patients and affects daily living [1]. It is associated with extensive direct and indirect costs and represents a considerable burden for the health care system and the society in general. In the US, direct medical costs for persons with disability were estimated to be $260 billion during 1996 [2]. The leading reason for disability in the general population is osteoarthritis (OA) [3, 4]. Data from the Centers for Disease Control (CDC) show that approximately 1 in 3 adults (37.6%) with arthritis reported limitation in their usual activities [5]. In the general population data from the Nottingham study, 14% of subjects in the age group of 40–79 years had knee pain with disability on most days of the previous month [6]. Due to the aging of most societies in the western world, the high prevalence of OA is expected to increase further in the upcoming years [7]. Interestingly, the degree of disability varies widely in subjects with the same degree of pathomorphological damage to the joint [8–10]. Prior studies focused on the question to what extent arthritis contributes to the risk of developing difficulties in activities of daily living (ADL) and revealed multiple factors: Song and colleagues for instance found that arthritis is associated with an odds ratio of 1.5 (95% CI 1.2–1.8) to develop ADL disability [1]. Obviously, many factors influence functional disability of patients with OA. Comorbidities as for example depression, which is highly prevalent among OA patients [11], can be assumed to contribute substantially [12–14]. Even though the causality of the association is yet unclear, the relation seems to be bidirectional: depression leads to decreased physical activity and functional limitation is associated with increased prevalence of depression. Since it is obvious that many factors contribute to functional disability, studies assessing functional disability need to control a variety of factors. Therefore, in this study, in a large sample of OA patients in primary care, factors associated with functional disability shall be assessed which have not yet been assessed altogether, i.e. depression and other comorbidities, obesity, duration of OA and amount of physical activity.

The data analyzed in this study were part of the baseline assessment of the PraxArt project. The project is financed by the German Ministry for Education and Research over a period of six years starting from 2003 and comprises data from a representative sample of 75 general practitioners' (GPs) practices in the area of Baden-Wuerttemberg and Bavaria. The project aims at learning about needs of patients with OA and at developing tailored multifaceted interventions to increase quality of life [15]. The Ethical Board of the University of Heidelberg approved the study. The study protocol has been published elsewhere [15].

Overall, the GPs addressed 1311 patients, 1250 (95.3%) agreed to receive the package of questionnaires. The main reason given for not participating was time effort. Finally, 1021 of the distributed 1250 questionnaires (81.7%) were returned and analyzed. As far as data were available in patient files, the non-respondents were compared to the respondents. In terms of sociodemographic variables as age and gender, as well as in available disease characteristics (as duration of disease, number of comorbidities, health service utilization, prescription of pain relievers), both groups did not differ significantly. Over 90% of the enrolled patients were Caucasians. Missing data occurred mainly within the same package of questionnaires, in total in 271 of the 1021 questionnaires. In 123 cases the data could be completed by means of the patients' medical files. Most missing data occurred in the IPAQ (148, 14.5% of questionnaires), whereas the PHQ-9 was completed by 344 men (98.9%) and 668 women (99.1%). For 448 women (66.5%) and 287 men (82.7%) a radiological scoring according to Kellgren was available.

Depression, pain intensity, educational level, BMI and social contacts are the most important factors associated with disability of the lower limb in patients suffering from osteoarthritis of the knee or hip.

The impact of depression on physical activity is well known. A study among patients with rheumatoid arthritis even revealed a linear relationship between depression severity and functional disability [42]. Even though the causality of the relationship remains unclear, it seems to be bidirectional [17, 43]. Besides its relation to disability, depression also has a substantial effect on the perception of pain [44].

Since significant more men than women in our study sample were married or lived in a partnership, which could be correlated to the prevalence of depression, we assessed a possible correlation between the PHQ-9 score and the marital status. Spearman's rho for the correlation between "being married/living in a partnership" and the PHQ-9 score was -0.046, but this correlation achieved no significance (p = 0.191). Consequently, the differences in the marital status between gender in our study sample can be excluded as confounder for our results.

Our study again emphasizes the need to stop the vicious circle of depression, pain and functional disability in order to increase physical activity and muscle strength. Physical activity and muscle strength are known to have a positive influence on the course of OA [45]. Interestingly, even if the regression coefficient beta was similar, depression was a stronger predictor for disability than pain. One reason might be that pain in OA is mostly not constant but shows peaks e.g. after intense activity.

The finding that depressed mood and/or depression increases pain perception in patients suffering from musculoskeletal disorders has been known for a long time [46, 47]. Since several studies have pointed out the importance of the social network for coping with chronic diseases and especially with pain, our results are plausible [42, 48, 49]. Even simple interventions, providing some kind of social support, as for instance monthly telephone calls, have shown to improve patients' QoL [50]. And interventions for osteoarthritis patients, involving for instance the spouse, have shown to achieve better results [51].

Increased body weight is one of the major risk factors for developing OA, but also for aggravation of joint damage and related symptoms [52]. Therefore it seems quite plausible that the BMI is associated with functional impairment [52, 53]. Radiological severity of OA has shown to be correlated with the BMI, especially in women [54]. Nevertheless, the fact that the Kellgren score did not remain in the final model by slightly surpassing the required significance displays that the impact of the BMI on disability in our study sample was more significant. The fact that the educational level remained in the regression model as a predictor is in accordance with prior findings emphasizing the importance of a higher education level for coping strategies and managing daily live in chronic diseases [55]. Interestingly, in OA a low educational level has been found to be associated with both more severe radiographic findings as well as more pain in the knee [56]. However, it has to be considered that these effects could be due to barriers in accessing the health care system [44].

An interesting finding is the fact that the social network contributes to physical activity. This result is supported by previous findings showing that interventions offered in a group setting or involving the spouse result in increased success of these programs [51]. A possible correlation between the marital status and social contacts would represent a possible confounder for our findings, consequently we assessed this correlation but no significance could be revealed.

Prevalence of OA increases with age [57]. Moreover, since OA is a degenerative disease, severity of pain and functional disability increase. This was confirmed by our findings and is in line with prior research where age was also revealed as predictor for increased disability in OA [1].

Some weaknesses of the study have to be noted. Due to the cross-sectional design, the revealed factors could not be confirmed as predictors over time. The fact that a control group with people without OA was not available may have led to an underestimation of the impact of OA. Furthermore, we did not assess factors as for instance self-efficacy, which showed to contribute to the course of the disease and functional ability [46]. Because of the large number of different questionnaires already used in this study, we abandoned including further instruments to assess self-efficacy. The study population had a remarkably high educational level, a finding which is mainly due to the setting in private solo practices. This may limit the comparison with data collected in hospital settings or Health Maintenance Organizations (HMO). Even though many findings of this study confirm prior findings, it is superior to prior studies in some terms: No other study before assessed more factors including depression simultaneously. It may therefore give a more realistic reflection of the situation in primary care.

In conclusion, our results confirm in particular prior findings as those by Sharma et al. [33] who found baseline laxity, knee pain intensity, self-efficacy and social support as predictors for decreased functional ability. Contrary to their findings, increased physical activity was not associated with a decreased risk of disability in our study sample. An explanation might be that different assessment instruments for physical activity were used. A new aspect revealed in this study was the substantial impact of depression on disability among OA patients. None of the prior studies among OA patients did control for depression. Furthermore, in an average sample of primary care patients our data showed the small predictive value of x-rays for individuals' complaints and for disability and confirmed once again the weak association between structural changes and functional disability.

As in rheumatoid arthritis [58], our results emphasize that psychological as well as physical factors need to be addressed similarly in aiming at improving functional ability. More research with multifaceted and tailored interventions is needed to determine how these factors can be targeted appropriately.