Psychological factors associated with instrumental activities of daily living disability in older adults with moderate to severe knee osteoarthritis

Abstract

BACKGROUND:

The population of older adults with knee osteoarthritis (OA)-related disabilities is increasing globally. However, studies regarding instrumental activities of daily living (IADL) in older adults with knee OA are limited.

OBJECTIVE:

This study investigated the psychological factors associated with IADL disability in older adults with moderate to severe knee OA.

METHODS:

A cross-sectional study was conducted on 179 patients with knee OA aged $\geqslant$ 65 years. The six-item short form of the Pain Catastrophizing Scale (PCS-6), the four-item short form of the Pain Self-Efficacy Questionnaire (PSEQ-4), and the fifteen-item Geriatric Depression Scale (GDS-15) were used to assess psychological factors. The participants were divided into IADL disabled and non-disabled groups. Binary logistic regression analyses were performed with the IADL disability status as the dependent variable. The PCS-6, PSEQ-4, and GDS-15 tools were included as independent variables in the logistic regression model.

RESULTS:

Of the 179 participants, 88 (49.1%) showed disability in conducting IADL. PSEQ-4 (odds ratio $=$ 0.90, 95%; confidence interval $=$ 0.82–0.99, $p=$ 0.02) was a significant independent variable among all psychological factors.

CONCLUSION:

Even after controlling for cofounders, our study found that self-efficacy, assessed using the PSEQ-4, was related to IADL disability in older adults with moderate to severe knee OA.

Keywords

Gait speed activities of daily living knee osteoarthritis aged pain self-efficacy

1. Introduction

Knee osteoarthritis (OA) is a prevalent age-related chronic condition and a common cause of disabilities limiting daily activities in older adults [1]. Since the population of older adults with knee OA-related disabilities is increasing globally [2], healthcare professionals need effective countermeasures to address this concern.

Instrumental activities of daily living (IADL) are essential elements in the independent lives of older adults. It is defined as independent functioning in a given environment and includes activities such as meal preparation, shopping, commutation, financial management, and performance of other household chores [3]. Meanwhile, basic activities of daily living (BADL) include basic self-care activities such as bathing, dressing, eating, and other indoor activities [4]. IADL require more complex functioning than BADL. As IADL disability in older adults influences health-related quality of life [5], all-cause mortality [6], and decline of cognitive function [7], assessments and interventions for IADL disability are important for this population

Disabilities in patients with knee OA are affected by factors such as gait speed, knee muscle strength, and pain [8, 9]. Psychological factors, such as pain catastrophizing [10, 11, 12], self-efficacy [12, 13, 14] and depressive symptoms [14, 15, 16] have also been reported among disabled patients with knee OA. A meta-analysis has shown the importance of psychological intervention for disability in patients with knee OA [17].

However, most studies of psychological factors and disability among patients with knee OA [10, 11, 12, 13, 14, 15, 16] have treated disability as a concept that combines BADL and IADL. It is well known that disabilities in older adults progress hierarchically [18]; IADL disability occurs before BADL disability [19]. Especially among older adults with knee OA, countermeasures should be taken at an early stage to prevent an incident or deterioration of disabilities. In particular, older adults with moderate to severe knee OA are more likely to develop disabilities in their daily lives than healthy older adults [20]. Therefore, investigating specific factors related to IADL disability in older adults with moderate to severe knee OA is useful information to develop effective interventions. To our knowledge, there are limited studies of IADL disability among older adults with knee OA [16, 21]; only our previous study has specified that gait speed and pain intensity are related to IADL disability in older adults with knee OA [22]. Psychological interventions are recommended for such disabilities [17]; however, which specific psychological factors are associated with IADL disability remain unclear. Factors such as demographics [23], motor function [22, 24, 25, 26], and pain intensity [22, 27] have been shown to influence IADL disability in older adults. Additionally, most studies of psychological factors and disability among patients with knee OA [10, 11, 12, 13, 14, 15, 16] found associations between psychological factors and disability by including motor function and pain status as confounders. Some psychological factors, such as self-efficacy, are affected by physical function and pain intensity [28]; therefore, these factors must be considered as confounders when attempting to establish an association between psychological factors with IADL disability.

This study aimed to determine the psychological factors associated with IADL disability in older adults with moderate to severe knee OA. We hypothesized that psychological factors such as pain catastrophizing, self-efficacy, and depressive symptoms are independently associated with IADL disability, even when adjusted for demographics, motor function, and pain intensity as confounders.

2. Materials and methods

2.1 Study design

This study used a cross-sectional design, and its conducting and reporting were guided by the STROBE guidelines [29]. Sample recruitment was conducted at Shonan Kamakura General Hospital from May 17, 2019 to May 30, 2021. All participants provided informed consent before the study began. The study was conducted in accordance with the Declaration of Helsinki and was approved by the research ethics boards of the Tokushukai Group Ethics Committee (No. TGE01198-024).

2.2 Participants

Patients with knee OA scheduled for primary unilateral total knee arthroplasty (TKA) or unicompartmental knee arthroplasty (UKA) were eligible for this study. The inclusion criteria were those aged $\geqslant$ 65 years and with a diagnosis of knee OA based on the clinical guidelines: knee pain for $>$ 3 months and Kellgren-Lawrence score (KL score) of $\geqslant$ 2 [30]. Conversely, the exclusion criteria were those diagnosed with rheumatoid arthritis, post-traumatic knee OA (e.g., knee OA after ligament injury), dementia (mini-mental state examination score $<$ 23), BADL disability (Barthel index $<$ 100), a history of lower extremity surgery, serious pathologies (e.g., cancer during treatment), neurological findings (e.g., motor paralysis) that could influence test performance, insufficient vision to answer questionnaires, and non-native speakers of Japanese. Patients who intended to get TKA or UKA on their other knee within six months after this surgery were also excluded to eliminate the effects of pain in the contralateral knee. A total of 357 patients were identified at the start of the study.

2.3 Procedures

All participants were given a two- to three-month waiting period after deciding to undergo TKA or UKA. On the day participants decided to undergo TKA or UKA, well-trained physical therapists taught them self-exercises to do during waiting periods. Self-exercises were designed to improve knee muscle tightness and strength based on individual knee conditions (e.g., muscle strength and range of motion [ROM] exercises). All physical function measurements, such as gait speed, were performed by well-trained physical therapists, and participants completed the self-administered questionnaires one month before the surgery.

2.4 Outcome measures

2.4.1 IADL status

IADL status was assessed on six activity items (preparing food, shopping, housekeeping, doing laundry, using transportation, and handling finances), referencing the IADL scale proposed by Lawton and Brody [3]. Taking medication and using a telephone were excluded, as it was expected that the condition of knee OA would not affect these items. Participants were asked to rate their abilities to perform these IADL activities by “able,” “need help,” or “unable.” Based on previous cross-sectional reports [24, 27], we defined those participants who opted for “need help” or “unable” to perform for at least one item as “disabled”; otherwise, they were defined as “non-disabled.”

2.4.2 Psychological factors

Pain catastrophizing, self-efficacy, and depressive symptoms were assessed as psychological factors based on previous reports that discussed the relationship between disability and psychological factors in patients with knee OA, as they show strong relation to disability even when adjusted for other confounders such as demographic factors and pain intensity [10, 11, 12, 13, 14, 15, 16].

Pain catastrophizing was assessed using the Japanese version of the six-item short form of the Pain Catastrophizing Scale (PCS-6) [31]. The Pain Catastrophizing Scale consists of subscales related to magnification, rumination, and helplessness [32]. PCS-6 has the same properties as the original version, correlates to pain intensity assessed by numerical rating scales ( $r=$ 0.30, $p<$ 0.001) [31], and has good internal consistency (Cronbach’s alpha $=$ 0.90) [33]. Participants were asked to rate the degree to which they experienced each of the six thoughts and feelings when experiencing pain on a scale from 0 (“not at all”) to 4 (“all the time”). The total PCS-6 score ranges from 0 to 24, with higher scores indicating higher levels of pain catastrophizing.

Self-efficacy was assessed using the Japanese version of four-item short form of the Pain Self-Efficacy Questionnaire (PSEQ-4) [34]. The Pain Self-Efficacy Questionnaire is a 10-item self-report questionnaire used to assess self-efficacy in individuals with chronic pain [35]. The PSEQ-4 has the same properties as the original version [34], correlates pain intensity assessed by numerical rating scales ( $r=-$ 0.35, $p<$ 0.001), and has good internal consistency (Cronbach’s alpha $=$ 0.90) [33]. The PSEQ-4 consists of four questions, and participants were asked how confident they were to perform the given activities despite pain, on a scale of 0 (“not at all confident”) to 6 (“completely confident”). The PSEQ-4 scores range from 0 to 24 with higher scores indicating more confidence in performing the given activities despite pain.

Depressive symptoms were assessed using the Japanese version of the fifteen-item Geriatric Depression Scale (GDS-15) [36]. The GDS-15 can screen depression (area under the curve of the receiver characteristic operating curve $=$ 0.96) and has good internal consistency (Cronbach’s alpha $=$ 0.83) [36]. The participants answered with a “yes” or “no” response. The total score was calculated from 0 to 15, with higher scores indicating more depressive symptoms.

2.5 Confounders

We assessed gait speed and pain intensity as confounders for outcome measures, referring to previous studies that report factors related to IADL disability in older adults [22, 24, 25, 26]. Knee muscle strength and joint ROM were also assessed as confounders, since they are considered specific factors related to disability in patients with knee OA [37]. Based on previous studies regarding IADL disability and gait speed [22, 24, 25, 26], usual gait speed (UGS) was measured using a 5-meter gait test [38]. If participants use a cane on a daily basis, we permitted using it during measurement.

The Japanese version of the pain subscale of the Knee injury and Osteoarthritis Outcome Score (KOOS-pain) was used for the representative index of pain intensity [39]. The KOOS-pain consists of nine questions. Participants were asked about their condition one week before the evaluation date. Standardized answer choices were provided, and each question was assigned a score ranging from 0 to 4. A normalized total score of 0 to 100 was calculated, and higher scores indicated that patients reported less pain. A score of $<$ 100 points is considered to indicate knee pain. The KOOS-pain correlates with the body pain subscale of the short form-36 health survey ( $r=$ 0.67, $p<$ 0.01) and has good internal consistency (Cronbach’s alpha $=$ 0.90) [39].

Isometric knee extension strength (IKES) was measured using a handheld dynamometer ( $\mu$ -tus F-1, Anima Co., Japan) [40]. Knee extension and flexion ROM were measured using a goniometer [41]. The knee scheduled for surgery was defined as the affected side, and the opposite side was defined as the unaffected side. IKES and knee ROM measurements were conducted on both knees. In addition to these factors, age, sex, body mass index, Charson Commodity Index (CCI) [42] and KL score of both knees were obtained from clinical records and used as confounders. The original method proposed by Charlson et al. [42] was used to calculate CCI. KL scores were assessed and validated by at least two of four orthopedic surgeons specializing in knee joint surgery.

2.6 Statistical analysis

The participants were divided into two groups according to their IADL status: those who answered “need help” or “unable” to perform on at least one item were assigned to the IADL disabled group, and those who answered “able” in all items were assigned to the IADL non-disabled group. All outcome measures and confounders were compared between the two groups. The chi-square test was used for categorical variables, and Student’s $t$ -test and Mann-Whitney $U$ test were used for normally and non-normally distributed variables, respectively. In the IADL disabled group, the proportion of each item of IADL disability was shown.

Which psychological factor can be associated with IADL disability was determined using a binary logistic regression model with IADL status as the dependent variable (disabled $=$ 1 or not $=$ 0). As previous reports have shown that age, sex [23], gait speed [22, 24, 25, 26], and pain intensity [22, 27] are factors related to IADL disability in older adults, these variables were included as confounders. In addition, other confounders that had $p$ -values $<$ 0.05 in the two groups of comparison were added to the logistic regression model. As the first step, PCS-6, PSEQ-4, and GDS-15 were included as the independent variable in each model (Models 1–3) separately. Furthermore, all psychological factors (e.g., PCS-6, PSEQ-4, GDS-15) were included as the independent variable simultaneously (Model 4) because we consider them to confound with each other. All analyses were performed using R version 4.0.3 (Foundation for Statistical Computing, Austria) for all tests; a $p$ -value $<$ 0.05 was considered statistically significant.

To determine the sample size for the logistic regression analysis, the number of cases (N) $=$ 10 $k/p$ are needed, where $k$ is the number of independent variables as covariates and $p$ is a defined ratio of responders to non-responders at the follow-up points [43]. To avoid overfitting and comply with the recommendations of Vittinghoff and McCulloch [44], we assumed that we would arrive at nine factors out of all the variables in the measurements and demographics. In a study involving older adults with joint pain ( $n=$ 407), 60.9% of participants had IADL disability [27]. Assuming that the ratio p of disabled to non-disabled is 1:1.5, the minimum sample size in this study was calculated as 10 $\times$ 9/0.6 $=$ 150.

3. Results

Of 357 patients, 166 were excluded due to a history of TKA or UKA on the other knee ( $n=$ 106), history of lower extremity surgery except knee joint ( $n=$ 32), rheumatoid arthritis ( $n=$ 13), Barthel index $<$ 100 ( $n=$ 9), mini-mental state examination score $<$ 23 ( $n=$ 3), or post-traumatic knee OA ( $n=$ 3). A total of 191 patients were invited to participate in this study, of which 12 declined; finally, 179 patients were included.

Figure 1.

Participant flowchart. BADL $=$ basic activities of daily living; OA $=$ osteoarthritis; TKA $=$ total knee arthroplasty; UKA $=$ unicompartmental knee arthroplasty.

All participants completed all assessments, and all confounders were collected from their medical records. Participants’ characteristics are presented in Table 1. Of all participants, 88 (49.1%) had IADL disabilities. Sixty-one (69.3%) showed disability in shopping and 47 (53.4%) showed disability in using transportation (Table 2).

Table 1

Baseline characteristics of the study population

Variable	ALL ( $n=$ 179)
Women; number (%)	141 (78.8)
Age; mean $\pm$ SD	76.5 $\pm$ 5.4
Body mass Index (kg/m ${}^{2}$ ); median IQR)	23.1 (21.1–26.2)
Family status (living alone); number (%)	36 (20.1)
CCI; median (IQR)	0 (0–1.0)
KL score on affected side
II; number (%)	19 (10.6)
III; number (%)	101 (56.4)
IV; number (%)	59 (33.0)
KL score on unaffected side
II; number (%)	62 (34.6)
III; number (%)	82 (45.8)
IV; number (%)	35 (19.6)

CCI $=$ Charlson comorbidity index; IQR $=$ interquartile range; KL score $=$ Kellgren-Lawrence score.

Table 2

Content of IADL disability in IADL disabled group ( $n=$ 88)

Item	Number (%)
Preparing food	33 (37.5)
Shopping	61 (69.3)
Housekeeping	40 (45.5)
Doing laundry	6 (6.8)
Using transportation	47 (53.4)
Handling finances	15 (17.0)

IADL $=$ Instrumental activity of daily living.

The IADL non-disabled group was significantly younger ( $p=$ 0.001) and had a greater number of men ( $p=$ 0.009) than the IADL disabled group. The IADL non-disabled group showed significantly higher values of UGS and IKES on both sides, KOOS-pain and PSEQ-4. Conversely, the IADL non-disabled group showed significantly lower values of PCS-6 and GDS-15 (Table 3).

Table 3

Comparison of measurements and confounders between IADL non-disabled and disabled groups

Variable		IADL non-disabled ( $n=$ 91)	IADL disabled ( $n=$ 88)	$p$ -value	Effect size
Women; number (%)		64 (70.6)	77 (87.5)	0.009	0.20
Age; mean $\pm$ SD		75.2 $\pm$ 5.2	77.8 $\pm$ 5.2	0.001	0.24
Body mass Index (kg/m ${}^{2}$ ); median (IQR)		22.9 (21.4–26.1)	23.1 (21.4–26.1)	0.74	0.02
Family status (living alone); number (%)		22 (24.2)	14 (15.9)	0.23	0.09
CCI; median (IQR)		0 (0–1.0)	0 (0–1.0)	0.63	0.04
KL score on affected side
II; number (%)		9 (9.9)	10 (11.4)	0.44	0.10
III; number (%)		48 (52.7)	53 (60.2)
IV; number (%)		34 (37.4)	25 (28.4)
KL score on unaffected side
II; number (%)		33 (36.3)	29 (33.0)	0.50	0.09
III; number (%)		38 (41.8)	44 (50.0)
IV; number (%)		20 (22.0)	15 (17.0)
Psychological factors
PCS-6; median (IQR)		14.0 (7.5–17.0)	16.0 (12.0–19.0)	0.003	0.23
GDS-15; median (IQR)		3.0 (1.0–6.0)	4.0 (2.0-7.3)	$<$ 0.001	0.25
PSEQ-4; median (IQR)		15.0 (12.0–18.5)	13.0 (11.0–15.0)	$<$ 0.001	0.29
UGS (m/sec); mean $\pm$ SD		1.0 $\pm$ 0.3	0.8 $\pm$ 0.3	$<$ 0.001	0.37
KOOS-pain; mean $\pm$ SD		48.6 $\pm$ 17.8	42.5 $\pm$ 15.1	0.01	0.18
IKES (Nm/kg); median (IQR)	Affected side	1.2 (1.0–1.6)	1.0 (0.7–1.3)	$<$ 0.001	0.27
	Unaffected side	1.3 (1.0–1.6)	0.9 (0.8–1.3)	$<$ 0.001	0.29
Knee flexion ROM ( ${}^{\circ}$ ); median (IQR)	Affected side	130 (115–135)	130 (115–140)	0.86	0.03
	Unaffected side	125 (120–138)	130 (115–140)	0.63	0.03
Knee extension ROM ( ${}^{\circ}$ ); median (IQR)	Affected side	$-$ 5 ( $-$ 10 to 0)	$-$ 5 ( $-$ 10 to 0)	0.51	0.05
	Unaffected side	$-$ 5 ( $-$ 5 to 0)	$-$ 5 ( $-$ 10 to 0)	0.25	0.07

Effect size was estimated using $r$ for continuous variables ( $r=\sqrt{{t}^{2}/{(t}^{2}+df)}$ , for normally distributed variables; $r=z/\sqrt{N}$ , for non-normally distributed variables) and Cramér ’s $V$ for categorical variables [53]. CCI $=$ Charlson comorbidity index; GDS-15 $=$ the fifteen-item Geriatric Depression Scale; IADL $=$ instrumental activities of daily living; IKES $=$ isometric knee extension strength; IQR $=$ interquartile range; KL score $=$ Kellgren-Lawrence score; KOOS-pain $=$ pain subscale of the Knee injury and Osteoarthritis Outcome Score; PCS-6 $=$ six-item short form of the Pain Catastrophizing Scale; PSEQ-4 $=$ four-item short form of the Pain Self-Efficacy Questionnaire; ROM $=$ range of motion; UGS $=$ usual gait speed.

In the binary logistic regression analysis with IADL disability status as a dependent variable, IKES on the unaffected side was adopted as a confounder. IKES on both sides showed significant differences between the two groups, but the effect size on the unaffected side was greater than that on the affected side. Therefore, only IKES on the unaffected side was included to avoid multicollinearity. In Models 1–3, PCS-6, PSEQ-4, and GDS-15 were significantly independent variables. In Model 4, PSEQ-4 (odds ratio [OR] $=$ 0.90, 95% confidence interval [CI] $=$ 0.82–0.99, $p=$ 0.02), UGS (OR $=$ 0.13, 95% CI $=$ 0.02–0.72, $p=$ 0.02), and sex (OR $=$ 0.38, 95% CI $=$ 0.15–0.96, $p=$ 0.04) were significantly independent variables. PCS-6 (OR $=$ 1.06, 95% CI $=$ 0.94–1.19, $p=$ 0.13) and GDS-15 (OR $=$ 1.06, 95% CI $=$ 0.94–1.19, $p=$ 0.45) were not significantly independent variables (Table 4).

Table 4

Logistic regression model with the dependent variable as IADL disabled or not

		Model 1	Model 2	Model 3	Model 4
Variables	Crude OR	Adjusted OR (95% CI)	Adjusted OR (95% CI)	Adjusted OR (95% CI)	Adjusted OR (95% CI)
Age	1.10	1.07 (0.99–1.14)	1.07 (0.99–1.14)	1.07 (0.99–1.14)	1.07 (1.00–1.15)
Sex	0.33	0.39 ${}^{*}$ (0.16–0.97)	0.38 ${}^{*}$ (0.15–0.94)	0.40 ${}^{*}$ (0.16–0.98)	0.38 ${}^{*}$ (0.15–0.96)
UGS	0.05	0.13 ${}^{{\dagger}}$ (0.02–0.65)	0.10 ${}^{{\dagger}}$ (0.02–0.52)	0.14 ${}^{*}$ (0.03–0.73)	0.13 ${}^{*}$ (0.02–0.72)
IKES unaffected side	0.35	0.61 (0.28–1.34)	0.80 (0.36–1.78)	0.65 (0.29–1.42)	0.73 (0.32–1.64)
KOOS-pain	0.98	1.01 (0.98–1.03)	1.01 (0.98–1.03)	0.10 (0.98–1.02)	1.01 (0.99–1.04)
PCS-6	1.07	1.07 ${}^{*}$ (1.00–1.14)			1.06 (0.94–1.19)
PSEQ-4	0.86		0.88 ${}^{{\dagger}}$ (0.81–0.96)		0.90 ${}^{*}$ (0.82–0.99)
GDS-15	1.17			1.13 ${}^{{\dagger}}$ (1.02–1.26)	1.06 (0.94–1.19)

${}^{*}$ : $p<$ 0.05, ${}^{{\dagger}}$ : $p<$ 0.01, Model 1: Hosmer-Lemeshow test $=$ 0.35, area under the curve (95% CI) $=$ 0.77 (0.70–0.84). Model 2: Hosmer-Lemeshow test $=$ 0.90, area under the curve (95% CI) $=$ 0.78 (0.71–0.85). Model 3: Hosmer-Lemeshow test $=$ 0.27, area under the curve (95% CI) $=$ 0.77 (0.70–0.84). Model 4: Hosmer-Lemeshow test $=$ 0.47, area under the curve (95% CI) $=$ 0.80 (0.73–0.86). CI $=$ confidence interval; GDS-15 $=$ the fifteen-item Geriatric Depression Scale; IADL $=$ instrumental activity of daily living; IKES $=$ isometric knee extension strength; OR $=$ odds ratio; KOOS-pain $=$ pain subscale of the Knee injury and Osteoarthritis Outcome Score; PCS-6 $=$ six-item short form of the Pain Catastrophizing Scale; PSEQ-4 $=$ four-item short form of the Pain Self-Efficacy Questionnaire; UGS $=$ usual gait speed.

4. Discussion

The present study aimed to investigate the psychological factors associated with IADL disability in older adults with moderate to severe knee OA. The values of the PCS-6, PSEQ-4, and GDS-15 tools significantly differed between the IADL non-disabled and IADL disabled groups. However, in the logistic regression analysis, which included all of the selected psychological factors and confounders, self-efficacy, assessed using PSEQ-4, was identified as a significant psychological factor associated with IADL disability.

In the logistic regression model, sex and UGS were also significant independent variables. Alexandre et al. [23] have shown that sex is a factor associated with IADL disability in older adults. Our results support this finding. A reduction in gait speed is also strongly associated with IADL disability in older adults [24, 25, 26]. Our current findings are consistent with those of our previous study, which showed gait speed to be a discriminatory factor for IADL disability in older adults with knee OA [22]. In our study population, the main items pertaining to IADL disability were outdoor activities such as shopping (68.6%) and using transportation (50.6%), which require a degree of gait speed. Furthermore, because the effect size of UGS was the highest among all the measurements when compared between the two groups, it could be a crucial factor associated with IADL disability in older adults with moderate to severe knee OA.

Catastrophizing about pain has been linked to increased physical disability in patients with knee OA [45]. Somers et al. reported that pain catastrophizing explained a significant proportion of the variance in physical disability measures [11]. Depressive symptoms are also significant psychological issues in public health [46]. According to a recent systematic review and meta-analysis, 19.9% of people with OA had depressive symptoms [47]. A cohort study of functional decline in daily life in patients with arthritis over two years found that depressive symptoms were a significant risk factor for functional decline, even after controlling for physical function and demographic factors [16]. For patients with chronic pain, the fear-avoidance model, which explains the relationship between psychological factors and disability progression in daily life, has been proposed. In the model, pain catastrophizing and depressive symptoms have been shown to be some of the leading causes of disability [48], and should be considered when accounting for relationships between psychological factors and disability. Moreover, self-efficacy has also been reported a mediator in the relationship between pain intensity and disability [49]. Our results implied that pain catastrophizing and depressive symptoms could be factors associated with IADL disability. However, based on the results of our multivariate analysis, self-efficacy as assessed by the PSEQ-4 was the most important factor associated with IADL disability in older adults with moderate to severe knee OA.

With respect to the relationship between pain self-efficacy and disability in patients with knee OA, Sinikallio et al. [12] suggested that self-efficacy assessed using the Pain Self-Efficacy Questionnaire was the only psychological factor associated with disability in multiple regression analysis that included demographics as confounders. Some cross-sectional studies have shown that self-efficacy, which partially includes pain elements, is a factor related to disability in patients with knee OA [14, 15]. A cohort study of older adults with knee pain indicated that self-efficacy is a predictor of self-reported disability [13]. A new finding of our study is that self-efficacy assessed by PSEQ-4 was also a factor related to disability even if the disability in daily life was limited to IADL. Regarding IADL disability and self-efficacy, using a logistic regression analysis that included demographics and pain intensity as confounders, a previous cross-sectional study of older adults with chronic joint pain showed that self-efficacy is associated with IADL disability [27], which is consistent with our results. Self-efficacy describes the confidence that a person has in their ability to achieve a desired outcome [50]. In patients with knee OA, those with high self-efficacy for controlling arthritis pain have higher pain thresholds than those with low self-efficacy [51]. Therefore, older adults with knee OA who have high self-efficacy might be considered less likely to voluntarily restrict activities included in IADL even if they have knee pain. Consequently, we considered self-efficacy, assessed using the PSEQ-4, associated with IADL disability even after adjusting for other confounders.

This study has several limitations. First, since this was a cross-sectional study, causal relationships between psychological factors and IADL disability incidence could not be determined. Second, this study included patients with knee OA scheduled for TKA or UKA, who were in a different treatment situation compared with other older adults with knee OA. Therefore, it might be difficult to generalize the results to older adults with early-stage knee OA or in a low pain status. In particular, awaiting surgery could affect pain intensity and psychological factors related to pain. Nonetheless, the average KOOS-pain score in this study population was not inferior to those in other studies [52]. The median PCS-6 and PSEQ-4 scores were also not inferior to those in previous studies in patients with chronic pain [33, 34]. Therefore, effects of awaiting TKA or UKA to pain intensity and psychological status were small. Third, because IADL disability was assessed using categorical variables as the dependent variable in our logistic regression analyses, we could not identify the effect of each factor on IADL disability. However, when comparing the two groups, the effect sizes of the variables that were significantly related to IADL disability in logistic regression models were approximately the same.

As we excluded patients with BADL disability, as measured by the Barthel index, from our study population, no participants required assistance with BADL. Additionally, CCI revealed that only a few participants had moderate to severe comorbidity affecting BADL. However, many patients may struggle with long-distance walking or stair climbing due to pain or poor knee function. Hence, it was not surprising that 49.1% of all participants had IADL disability, which was defined as requiring assistance to perform IADL activities. Previous cohort studies have shown that even one IADL disability affected all-cause mortality [6] and the rate of cognitive function decline [7]. We believe that those with IADL disability in our study population are more likely to experience health problems in the near future and should be closely monitored. The purpose of this study was not to predict IADL disability using assessments; rather, it was conducted with the conviction that greater knowledge about psychological factors in older adults with knee OA will aid in understanding the risk of IADL disability or deterioration.

5. Conclusion

Regarding psychological factors, self-efficacy, assessed using the PSEQ-4, was associated with IADL disability even after adjusting for confounders. Sex and gait speed were also associated with IADL disability in older adults with moderate to severe knee OA. Our study demonstrated the importance of self-efficacy, assessed using the PSEQ-4, as the factor most associated with the presence of IADL disability in older adults with moderate to severe knee OA. Future studies are necessary to elucidate the mechanism through which self-efficacy is related to IADL disability in this population. Additionally, the sample size should be increased, and multivariate analysis should be performed using other factors that can affect IADL disability, such as living environment and educational level. Furthermore, longitudinal studies are needed to identify predictors of the development of IADL disability in older adults with moderate to severe knee OA.

Ethical approval

The study was approved by the research ethics boards of the Tokushukai Group Ethics Committee (No. TGE01198-024).

Funding

The authors report no funding.

Informed consent

All participants provided informed consent before the study began.

Author contributions

KN: Conceptualization, Methodology, Investigation, Formal analysis, Writing – original draft, Project administration. TI: Methodology, Writing – original draft. NN: Investigation, Writing – Review and Editing. TS: Methodology, Writing – Review and Editing. TJ: Methodology, Writing – Review and Editing, Supervision.

Footnotes

Acknowledgments

The authors thank D. Kurihara, K. Imahira, and K. Suda for their cooperation with data collection.

Conflict of interest

The authors declare that they have no conflict of interest.

References

Bijlsma

Berenbaum

Lafeber

. Osteoarthritis: An update with relevance for clinical practice. Lancet. 2011; 377(9783): 2115-26.

Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. Lancet. 2016; 388(10053): 1545-1602.

Lawton

Brody

. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969; 9(3): 179-186.

Katz

Ford

Moskowitz

, et al. Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. JAMA. 1963; 185: 914-919.

Williams

Egede

. The association between multimorbidity and quality of life, health status and functional disability. Am J Med Sci. 2016; 352(1): 45-52.

Hennessy

Kurichi

Pan

, et al. Disability stage is an independent risk factor for mortality in Medicare beneficiaries aged 65 years and older. PM&R. 2015; 7(12): 1215-1225.

Rajan

Hebert

Scherr

, et al. Disability in basic and instrumental activities of daily living is associated with faster rate of decline in cognitive function of older adults. J Gerontol A Biol Sci Med Sci. 2013; 68(5): 624-630.

Pereira

Peleteiro

Araújo

, et al. The effect of osteoarthritis definition on prevalence and incidence estimates: A systematic review. Osteoarthritis Cartilage. 2011; 19(11): 1270-1285.

van Dijk

Veenhof

Lankhorst

, et al. Limitations in activities in patients with osteoarthritis of the hip or knee: The relationship with body functions, comorbidity and cognitive functioning. Disabil Rehabil. 2009; 31(20): 1685-1691.

10.

Odole

Ekediegwu

Ekechukwu

END

, et al. Correlates and predictors of pain intensity and physical function among individuals with chronic knee osteoarthritis in Nigeria. Musculoskelet Sci Pract. 2019; 39: 150-156.

11.

Somers

Keefe

Pells

, et al. Pain catastrophizing and pain-related fear in osteoarthritis patients: relationships to pain and disability. J Pain Symptom Manage. 2009; 37(5): 863-872.

12.

Sinikallio

Helminen

Valjakka

, et al. Multiple psychological factors are associated with poorer functioning in a sample of community-dwelling knee osteoarthritis patients. J Clin Rheumatol. 2014; 20(5): 261-267.

13.

Rejeski

Miller

Foy

, et al. Self-efficacy and the progression of functional limitations and self-reported disability in older adults with knee pain. J Gerontol B Psychol Sci Soc Sci. 2001; 56(5): S261-S265.

14.

Maly

Costigan

Olney

. Determinants of self-report outcome measures in people with knee osteoarthritis. Arch Phys Med Rehabil. 2006; 87(1): 96-104.

15.

Zambon

Siviero

Denkinger

, et al. Role of osteoarthritis, comorbidity, and pain in determining functional limitations in older populations: European project on osteoarthritis. Arthritis Care Res. 2016; 68(6): 801-810.

16.

Dunlop

Semanik

Song

, et al. Risk factors for functional decline in older adults with arthritis. Arthritis Rheum. 2005; 52(4): 1274-1282.

17.

Dixon

Keefe

Scipio

, et al. Psychological interventions for arthritis pain management in adults: A meta-analysis. Health Psychol. 2007; 26(3): 241-250.

18.

Barberger-Gateau

Rainville

Letenneur

, et al. A hierarchical model of domains of disablement in the elderly: A longitudinal approach. Disabil Rehabil. 2000; 22(7): 308-317.

19.

Edjolo

Proust-Lima

Delva

, et al. Natural history of dependency in the elderly: A 24-year population-based study using a longitudinal item response theory model. Am J Epidemiol. 2016; 183(4): 277-285.

20.

Logerstedt

Zeni

Snyder-Mackler

. Sex differences in patients with different stages of knee osteoarthritis. Arch Phys Med Rehabil. 2014; 95(12): 2376-81.

21.

George

Ruiz

Sloan

. The effects of total knee arthroplasty on physical functioning in the older population. Arthritis Rheum. 2008; 58(10): 3166-3171.

22.

Nanjo

Ikeda

Nagashio

, et al. Gait speed and pain status as discriminatory factors for instrumental activities of daily living disability in older adults with knee osteoarthritis. Geriatr Gerontol Int. 2021; 21(8): 683-688.

23.

Alexandre

Corona

Nunes

, et al. Disability in instrumental activities of daily living among older adults: Gender differences. Rev Saude Publica. 2014; 48(3): 379-389.

24.

Lee

Hsu

Tsai

, et al. Criterion-referenced values of grip strength and usual gait speed using instrumental activities of daily living disability as the criterion. J Geriatr Phys Ther. 2018; 41(1): 14-19.

25.

Zhang

Guo

, et al. Role of physical performance measures for identifying functional disability among Chinese older adults: Data from the China Health and Retirement Longitudinal Study. PLoS ONE. 2019; 14(4): e0215693.

26.

Donoghue

Savva

Cronin

, et al. Using timed up and go and usual gait speed to predict incident disability in daily activities among community-dwelling adults aged 65 and older. Arch Phys Med Rehabil. 2014; 95(10): 1954-1961.

27.

Hermsen

Leone

Smalbrugge

, et al. Frequency, severity and determinants of functional limitations in older adults with joint pain and comorbidity: Results of a cross-sectional study. Arch Gerontol Geriatr. 2014; 59(1): 98-106.

28.

Maly

Costigan

Olney

. Determinants of self-efficacy for physical tasks in people with knee osteoarthritis. Arthritis Rheum. 2006; 55(1): 94-101.

29.

von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies. J Clin Epidemiol. 2008; 61(4): 344-349.

30.

Altman

Asch

Bloch

, et al. Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Arthritis Rheum. 1986; 29(8): 1039-1049.

31.

Nishigami

Mibu

Tanaka

, et al. Psychometric properties of the Japanese version of short forms of the Pain Catastrophizing Scale in participants with musculoskeletal pain: A cross-sectional study. J Orthop Sci. 2017; 22(2): 351-356.

32.

Sullivan

MJL

Bishop

Pivik

. The pain catastrophizing scale: Development and validation. Psychol Assess. 1995; 7(4): 524-532.

33.

McWilliams

Kowal

Wilson

. Development and evaluation of short forms of the Pain Catastrophizing Scale and the Pain Self-efficacy Questionnaire. Eur J Pain. 2015; 19(9): 1342-1349.

34.

Adachi

Enomoto

Yamada

, et al. Evaluating the psychometric properties of two-item and four-item short forms of the Japanese Pain Self-Efficacy Questionnaire: A cross-sectional study. J Anesth. 2019; 33(1): 58-66.

35.

Nicholas

. The pain self-efficacy questionnaire: Taking pain into account. Eur J Pain. 2007; 11(2): 153-163.

36.

Sugishita

Hemmi

, et al. A Validity and reliability study of the Japanese Version of the Geriatric Depression Scale 15 (GDS-15-J). Clin Gerontol. 2017; 40(4): 233-240.

37.

Kauppila

Kyllonen

Mikkonen

, et al. Disability in end-stage knee osteoarthritis. Disabil Rehabil. 2009; 31(5): 370-380.

38.

Shinkai

Watanabe

Kumagai

, et al. Walking speed as a good predictor for the onset of functional dependence in a Japanese rural community population. Age Ageing. 2000; 29(5): 441-446.

39.

Nakamura

Takeuchi

Sawaguchi

, et al. Cross-cultural adaptation and validation of the Japanese Knee Injury and Osteoarthritis Outcome Score (KOOS). J Orthop Sci. 2011; 16(5): 516-523.

40.

Katoh

Yamasaki

. Comparison of reliability of isometric leg muscle strength measurements made using a hand-held dynamometer with and without a restraining belt. J Phys Ther Sci. 2009; 21(1): 37-42.

41.

Norkin

White

. Measurement of joint motion: a guide to goniometry. 2nd ed. Philadelphia: F.A. Davis; 1995.

42.

Charlson

Pompei

Ales

, et al. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis. 1987; 40(5): 373-383.

43.

Peduzzi

Concato

Kemper

, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996; 49(12): 1373-1379.

44.

Vittinghoff

McCulloch

. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol. 2007; 165(6): 710-718.

45.

Keefe

Lefebvre

Egert

, et al. The relationship of gender to pain, pain behavior, and disability in osteoarthritis patients: The role of catastrophizing. Pain. 2000; 87(3): 325-334.

46.

Moussavi

Chatterji

Verdes

, et al. Depression, chronic diseases, and decrements in health: Results from the World Health Surveys. Lancet. 2007; 370(9590): 851-858.

47.

Stubbs

Aluko

Myint

, et al. Prevalence of depressive symptoms and anxiety in osteoarthritis: A systematic review and meta-analysis. Age Ageing. 2016; 45(2): 228-235.

48.

Vlaeyen

JWS

Linton

. Fear-avoidance and its consequences in chronic musculoskeletal pain: A state of the art. Pain. 2000; 85(3): 317-332.

49.

Arnstein

. The mediation of disability by self efficacy in different samples of chronic pain patients. Disabil Rehabil. 2000; 22(17): 794-801.

50.

Bandura

. Self-efficacy: Toward a unifying theory of behavioral change. Psychol Rev. 1977; 84(2): 191-215.

51.

Keefe

Lefebvre

Maixner

, et al. Self-efficacy for arthritis pain: Relationship to perception of thermal laboratory pain stimuli. Arthritis Care Res. 1997; 10(3): 177-184.

52.

Collins

Prinsen

Christensen

, et al. Knee Injury and Osteoarthritis Outcome Score (KOOS): Systematic review and meta-analysis of measurement properties. Osteoarthritis Cartilage. 2016; 24(8): 1317-1329.

53.

Field

. Discovering Statistics Using IBM SPSS Statistics. 4th ed. London: Sage Publications Ltd; 2012.