Daily Sedentary Behavior Predicts Pain and Affect in Knee Arthritis

Abstract

Background

Sedentary behavior (SB), which has been linked with numerous adverse health outcomes, is prevalent among adults with osteoarthritis (OA). The associations between SB and daily physical and psychological health outcomes for OA patients, however, have received little attention.

Purpose

Using accelerometer and self-report data, the current study examined how the amount of time OA patients spent in SB was associated with their pain and affect in daily life, independent of physical activity.

Methods

Over 22 days, 143 older adults (mean age = 65 years) with knee OA wore an accelerometer to measure SB and physical activity, and also reported their pain and affect three times a day using a handheld computer. Multilevel analyses were conducted to examine the prospective within-person associations between SB and subsequent pain or affect within the same day and across days, independent of physical activity.

Results

The time spent in SB daily predicted less pain but worse affect at the end of that day, above and beyond the effects of physical activity, as well as demographics and individual differences in general health and depression. Moreover, cross-day lagged analyses indicated that time spent in SB on 1 day predicted higher negative affect the next morning. Finally, the average level of SB was also associated with worse average affect at the between-person level.

Conclusions

SB may be related to less pain in the short term but detract from patients’ emotional well-being. Future intervention should aim to reduce daily SB to improve OA patients’ emotional well-being.

Sedentary behavior, Physical activity, Affect, Pain, Osteoarthritis, Chronic condition

Introduction

Sedentary behavior (SB), defined as activities that do not increase energy expenditure substantially above the resting level (e.g., 1.0–1.5 metabolic equivalents) [1], is characterized by waking time spent sitting, reclining, or lying down. Globally, adults spend a considerable percentage of their daily time in SB such as workplace sitting, television watching, or computer and game-console use [2]. In fact, individuals can be highly active (e.g., meeting the 150 min/week moderate-to-vigorous physical activity recommendation) yet sedentary for most of their waking hours [3]. A growing body of literature has linked SB with numerous adverse mental and physical health outcomes and even all-cause mortality, independent of physical activity (see de Rezende et al. [4] and Teychenne et al. [5] for reviews). Adults with osteoarthritis (OA) are highly sedentary and spend even more time in SB than the general population [6]. A large-scale cohort study found that, on average, adults with knee OA spent 9.9 hr each day (66% of waking time) in SB [7]. The sedentary lifestyle of OA patients may have a profound impact on their health and well-being, given the important role of physical activity for reducing or managing OA symptoms [8]. However, little is known about how SB relates to the physical and psychological health outcomes, such as pain and affect, for OA patients in their daily lives.

SB and Pain

Pain is the primary symptom of OA and has significant influences on OA patients’ well-being and health-related quality of life [9]. Previous cross-sectional and longitudinal research has found that among healthy adults or adults with chronic pain, individuals with a higher level of SB report a higher level of pain than those who are less sedentary [10, 11]. However, recent research suggests that the levels of SB can vary from day to day within an individual, and the association between fluctuations of daily SB and pain may be negative. For example, one recent daily diary study found that more engagement in daily SB was associated with less subsequent bodily pain in the same day among older adults [12]. In addition, previous research suggests that daily ambulatory activity may increase knee joint loading in adults with OA and result in increased pain temporarily [13, 14]. Thus, it is likely that daily SB may relate to less pain subsequently due to the temporary relief of knee joint loading. However, it is unclear whether the pain-reduction effect of SB found by previous daily diary study is transient. The answer to this question has important health implications for adults with OA given that SB has been linked with decreased physical functions over time for OA patients [7], and exercise or physical activity has been recommended as one of the most effective treatment for OA symptoms [8].

SB and Affect

There is increasing evidence that the overall level of SB is associated with worse mental or psychological outcomes, such as lower positive affect and depression (see Teychenne et al. [5, 15] for reviews). In line with this, one recent daily dairy study of midlife women found a decrease in positive affect following more than usual time spent in SB [16]. A number of explanations have been proposed for the link between SB and psychological health. For example, time spent in SB such as screen-based entertainment has been shown to relate to the arousal of the central nervous system [17], a disrupted sleeping pattern [18], and greater stress-induced inflammation [19], all of which may in turn detract from psychological well-being. In addition, research supporting the social withdrawal hypothesis shows that the more time people spend in SB such as watching TV, the further they remove themselves from social interaction [15], which is linked to positive psychological outcomes [20].

Despite the growing number of studies examining the associations between SB and health outcomes, the majority of past research has only focused on between-person differences and addressed whether individuals with different levels of SB experience distinct levels of pain or affect. In contrast, few studies have taken a within-person approach to examine how SB relates to pain or affect within the same individual, in OA patients or healthy adults (exceptions include [12, 16]). In other words, how is an individual’s pain and affect different when this person engages in more versus less SB? Two recent studies found evidence suggesting that the within-person associations between daily SB and pain or affect may be different from the between-person associations [12, 16], thus highlighting the need to understand the effects of SB at both the within- and between-person levels.

The Current Study

The purpose of the current study was to investigate how the amount of time OA patients spend in SB every day was associated with their subsequent pain and affect, independent of the time they spend in moderate-to-vigorous physical activity (MVPA). Using data collected with handheld computers and accelerometers over a 22-day period among a sample of adults with OA, the current study disaggregated the within-person and between-person associations between SB and pain or affect. Based on previous empirical evidence [10–12], we predicted that on days when patients spent more time in SB than usual, they would report less pain at the end of the day (within-person effect), but in general, OA patients who spent more time in SB would report more pain compared with those who spent less time in SB (between-person effect), controlling for the time spent in MVPA. Regarding the psychological outcome, we predicted that on days when patients spent more time in SB than usual, they would report worse emotional well-being (lower positive affect and higher negative affect) at the end of the day (within-person effect), and patients who spent more time in SB on average would also report worse emotional well-being compared with those who spent less time in SB (between-person effect), independent of the level of MVPA. To test whether the effects of SB on pain or affect are transient, we also examined whether the within-person associations between SB and pain or affect could extend across days. Finally, we examined whether these cross-day within-person associations were bidirectional.

Method

Study Design and Participants

Data for the current study were from a larger study of couples in which one partner (i.e., the patient) was diagnosed with knee OA (for a detailed description, see Martire et al. [21]). This study included three in-person interviews conducted over a period of 18 months and a 22-day daily assessment immediately after the first interview. During the daily assessment period, all patients and spouses wore an accelerometer continuously for 22 days to assess their daily SB and physical activity. Patients and spouses also used handheld computers to complete a short survey three times per day (i.e., morning, afternoon, and end-of-day). The current report utilized OA patients’ data from the T1 interviews, accelerometers, and morning and end-of-day diary assessments.

Primary sources of recruitment were research registries for rheumatology clinics and for older adults interested in research, flyers distributed at the University of Pittsburgh, and word of mouth. To be eligible for this study, patients had to be diagnosed with knee OA by a physician, experience usual knee pain of moderate or greater intensity, be at least 50 years of age, and be married or in a long-term relationship and living with their spouse or partner. Exclusion criteria for couples included (a) patient had a comorbid diagnosis of fibromyalgia or rheumatoid arthritis; (b) patient planned to have hip or knee surgery in the next 6 months; (c) spouse had arthritis pain of moderate or greater intensity, or required assistance with personal care activities; (d) either patient or spouse used a wheelchair to get around; and (e) either patient or spouse was not cognitively functional as indicated by the accuracy of their answers to questions regarding the current date, day of the week, their age, and birth date.

A total of 606 couples were screened for eligibility. Of these, 221 couples declined to participate; the most frequent reasons were lack of interest (N = 87) and illness in the family (N = 55). An additional 233 couples were ineligible for the study; the most frequent reasons were lack of OA of the knee (N = 55) and OA pain that was mild (N = 47). A total of 152 couples were enrolled in the study and a total of 145 couples completed the diary assessment section. Due to missing accelerometer data, the sample size for the current analyses was 143 patients. Table 1 provides descriptive information for the final sample.

Table 1

Demographic Characteristics of the Sample

Variable	M or %	SD
Gender (male)	41.96%
Age	65.39	9.53
Race (white)	86.71%
Employed (yes)	42.66%
Education (years)	16.03	2.02
Duration of OA (years)	16.17	12.23
General health	3.24	0.94
Depressive symptoms	6.56	5.29

N = 143. Range for self-reported general health is 1 (poor) to 5 (excellent); range for depressive symptoms scale is 0–30. M mean; OA osteoarthritis; SD standard deviation.

Measures

Daily SB and physical activity

Accelerometers were used to assess OA patients’ daytime SB and physical activities during the 22-day diary assessment period. Accelerometers are motion-sensitive monitors that count the number of movements taken per prespecified time interval. Participants wore the GT1M or GT3X model of the CSA/MTI triaxial ActiGraph, with placement on the hip secured with a belt and/or belt clip to best capture ambulatory activities. Participants were instructed to wear the monitor during the day and remove it at night; a reminder to put the monitor on in the morning was provided electronically via the handheld computers. Participants used a written log to record waking time, sleep time, time they put on the accelerometer, time they took off the accelerometer, and any waking periods during which they did not wear the accelerometer for longer than 30 min. Data were collected in 1-min epochs. Nonwear was defined by an interval of 90 consecutive minutes of zero counts/min, with allowance of up to 2 min of low (<100) activity counts [7, 22]. To be included in the analysis, at least four adherent days, with an adherent day indicating no less than 10 hr of accelerometer wear, were required. Data were also screened for anomalous values (activity counts >6,000 at any given minute), which affected <1% of the activity data.

The intensity thresholds used by the National Health and Nutrition Examination Survey (NHANES) were applied to identify SB and MVPA. SB was defined as <100 activity counts/min, and MVPA intensity was defined as ≥760 activity counts/min based on previous studies that incorporated more lifestyle physical activities [23] and studies on physical activity among OA patients [21, 24].

Daily positive and negative affect

During the 22-day diary assessment period, patients reported to what extent they felt positive or negative affect over the past 30 min in the morning and end-of-day assessments [25]. Eight items were rated on a 7-point scale (0 = not at all; 6 = extremely). Four items (e.g., frustrated, angry or hostile, unhappy, worried, or anxious) were averaged to create a mean score of negative affect (Cronbach’s α = .86 and .88 for morning and evening assessments), and four items (e.g., happy, joyful, pleased, enjoyment) were averaged to create a mean score of positive affect (Cronbach’s α = .98 and .97 for morning and end-of-day assessments).

Daily pain

Patients reported the pain or tenderness in 10 sets of joints over the past 30 min in the morning and end-of-day assessments [26]. Separate ratings were made for multiple joints or joint groups (i.e., knees, shoulders, hips, ankles, ball of foot, toe knuckles, elbows, wrists, hand knuckles, finger knuckles) on a scale from 0 to 3 (0 = no pain/tenderness; 3 = severe pain/tenderness). Items were averaged to create a mean score for patients’ pain (Cronbach’s αs = .90 for morning and end-of-day assessments).

Covariates

Covariates were selected based on previous research identifying factors that are related to daily affect and/or pain [27, 28]. Specifically, baseline physical health status was assessed by one item “In general, would you say your health is (1 = Poor, 5 = Excellent)?” Patients’ depressive symptoms were assessed by the 10-item The Center for Epidemiologic Studies Depression Scale [29, 30] at the baseline interview. Patients rated 10 statements of ways they might have felt or behaved in the past week on a 4-point scale (0 = rarely or none of the time, 3 = most of the time). Sum scores were created for depressive symptoms (Cronbach’s α = 0.82). In addition, patients’ gender, age, and OA duration measured at the baseline interview were included as covariates.

Data Analysis

Multilevel modeling was used to examine the associations between the daily time patients spent in SB and subsequent pain and affect at the end of the day [31]. The data were structured hierarchically, with daily assessments (level 1) nested within persons (level 2). Thus, daily variables (e.g., SB, MVPA, pain, and affect) could vary over days within a person as well as across persons. Analyses were conducted using SAS PROC MIXED with restricted maximum likelihood (REML) to handle missing data [32, 33] and robust standard errors to adjust against the skewness of the outcomes [34]. The within-person interdependence of the repeated outcome assessments was accounted for by using autoregressive, AR [1], covariance structure for residual errors.

Analyses were conducted in a series of steps. In the first step, an empty model with a random intercept but no predictor was tested for each key variable to estimate the intraclass correlation coefficient (ICC), which indicates the variance at between- versus within-person levels. In the second step, the prospective associations between patients’ SB throughout the day and pain or affect at the end of the day, independent of the effects of MVPA, were examined. In this model (Model 1), predictors included Level 1 person-mean centered SB and MVPA scores which captures each person’s daily deviation from his or her own mean score of SB or MVPA (within-person effect), Level 2 mean scores of SB and MVPA which captures the individual differences in the average score of SB and MVPA (between-person effect), as well as the daily accelerometer wear time. In each model, a random intercept was included to allow the mean score of outcome to vary across individuals, and a random slope of the within-person effect of SB was also included if it was statistically significant to allow the association between daily SB and each outcome to vary between individuals. In the third step of analysis, covariates were added to the model (Model 2). Specifically, affect (positive and negative) and pain reported in the morning were person-mean centered and added into the model to control for the “carry-over” effects of the same health variables measured earlier and thus provide a strong test of the prospective association between SB and subsequent health outcomes. Baseline assessments of general physical health, depressive symptoms, OA duration, as well as OA patients’ gender and age were grand-mean centered and added as Level 2 covariates to provide a robust test of the unique predictive effects of SB on health outcomes above and beyond these factors. Following the recommendation of previous research [35], Cohen’s f² was computed as the estimate of local effect size for the association between daily SB and outcomes.

In addition to within-day analysis, we examined the cross-day lagged effect of SB by predicting Day t health outcomes (morning assessments) from Day t − 1 SB (end-of-day assessment) controlling for Day t − 1 MVPA, affect and pain (end-of-day assessment). Finally, the potential bidirectional lagged effect was tested by predicting Day t SB (end-of-day assessment) from Day t − 1 pain and affect (i.e., lag-1, end-of-day assessment) controlling for Day t pain and affect (end-of-day assessment). We also tested multivariate models predicting three health outcomes (i.e., positive and negative affect, pain) simultaneously from SB and MVPA in Mplus V8.0; the primary results were unchanged and thus findings are presented for each outcome separately. (Multivariate analysis results are included in the Supplementary Material.)

Results

Compliance Rates and Preliminary Analyses

Out of 3,146 potential daily accelerometer assessments (143 participants × 22 days) from valid participants (i.e., having at least 4 days of 10 or more hours of wearing time), data for 2,648 accelerometer assessments were available (84%). On average, patients wore the accelerometer for 18.52 (SD = 3.19) days in the study and for 14.20 hr each day on average (SD = 1.11). In addition, out of a potential 3,146 morning or end-of-day assessments, 2,944 morning assessments (94%) and 2,952 end-of-day assessments (94%) were completed. Morning assessments that were completed more than 120 min after waking and end-of-day assessments that were completed more than 120 min before bedtime were excluded from analysis. Using this criterion, 2,623 morning and 2,648 end-of-day assessments were included in analysis.

On average, patients spent 9.61 hr (65% of their waking time) in SB and 1.16 hr in MVPA each day. As shown in Table 2, SB was moderately correlated with MVPA at the within-person level (r = −.33). The ICCs indicated that 50% of the total variance in SB and 61% of the total variance in MVPA time occurred between persons, suggesting considerable variability in daily SB and MVPA at both between-person and within-person levels. In addition, SB was significantly correlated with less pain at the within-person level and lower positive affect at the between-person level. Time spent in MVPA was significantly correlated with more pain at the within-person level.

Table 2

Descriptive Information and Intercorrelations of Key Daily Study Variables

Study variables	M	SD	ICC	1.	2.	3.	4.	5.
1. Daily sedentary time (hr)	9.61	1.49	0.50	—	−.33***	−.03	.03	−.09***
2. Daily MVPA time (hr)	1.16	0.72	0.61	−.63***	—	−.02	.03	.12***
3. Positive affect (Eod)	2.89	1.30	0.65	−.23**	.13	—	−.43***	.003
4. Negative affect (Eod)	0.49	0.65	0.49	.11	−.04	−.33***	—	.08***
5. Pain (Eod)	0.61	0.49	0.86	−.05	−.11	−.05	.14	—

N = 143 subjects. M and SD were calculated based on scores averaged at the individual level. Possible range for daily positive and negative affect scale is 0–6; for daily pain scale is 0–3. Between-person correlations are below the diagonal and within-person correlations are above. Eod end-of-day assessment; ICC intraclass correlation; M mean; MVPA moderate to vigorous physical activity; SD standard deviation.

**p < .01, ***p < .001.

Primary Analyses

We first tested the prospective association between patients’ SB throughout the day and pain at the end of the day independent of MVPA (Table 3). Consistent with our hypothesis, daily SB time predicted less pain at the end of the day, independent of time spent in MVPA, suggesting that on days when patients spent more time than usual in SB, they reported less pain at the end of the day (Model 1). This effect remained significant with covariates in the model (Model 2), indicating that on days when patients spent one more hour in SB than usual, they reported 0.02 points less than usual pain (4-point scale) at the end of the day (f² = 0.004). No significant between-person effect of SB was found for pain, suggesting that patients with higher SB in general did not significantly differ from those with lower SB in terms of overall level of pain. In addition, among covariates, MVPA had a positive within-person effect on pain, suggesting that on days when patients spent more time than usual in MVPA, they reported more pain at the end of the day. Patients’ baseline physical health was negatively associated with their daily pain.

Table 3

Daily Sedentary Time Predicting Pain

Fixed effects	Pain (end-of-day)
	Model 1		Model 2
	Est.	95% CI	Est.	95% CI
Intercept	1.04**	[0.37, 1.71]	1.44**	[0.58, 2.30]
Sedentary time (WP)	−0.01**	[−0.02, −0.004]	−0.02**	[−0.03, −0.01]
Sedentary time (BP)	−0.001	[−0.001, 0.000]	−0.000	[−0.001, 0.001]
MVPA time (WP)	0.02*	[0.003, 0.05]	0.02*	[0.004, 0.04]
MVPA time (BP)	−0.002*	[−0.004, −0.000]	0.000	[−0.001, 0.002]
Wearing time	0.01	[−0.002, 0.02]	0.01**	[0.003, 0.02]
Morning pain			0.17***	[0.10, 0.24]
Morning positive affect			−0.001	[−0.01, 0.01]
Morning negative affect			−0.01	[−0.02, 0.01]
Age			0.005	[−0.01, 0.01]
Gendera			0.04	[−0.09, 0.18]
Duration of OA			0.01	[−0.002, 0.02]
General health			−0.15***	[−0.24, −0.05]
Depressive symptoms			0.01	[−0.003, 0.02]

BP between-person; MVPA moderate to vigorous physical activity; OA osteoarthritis; WP within-person.

^a1 = male, 2 = female.

*p < .05, **p < .01, ***p < .001.

We then tested the prospective association between patients’ SB throughout the day and emotional well-being at the end of the day independent of MVPA (Table 4). As hypothesized, on days when patients spent one more hour than usual in SB, they reported 0.1 point lower positive affect (f² = 0.01, 7-point scale) and 0.04 point higher negative affect (f² = 0.002, 7-point scale) at the end of the day, independent of time spent in MVPA (Model 1). These within-person effects of SB remained with covariates in the model, supporting the unique contribution of patients’ daily SB in predicting subsequent positive and negative affect beyond the effects of morning pain and affect, as well as patients’ demographics, general health status, and depressive symptoms (Model 2). Also, in line with our hypothesis, SB time had significant between-person effects on positive and negative affect controlling for MVPA and other covariates, suggesting that people who spent more time in SB in general also reported lower positive affect and higher negative affect than their counterparts who spent less time in SB.

Table 4.

Daily Sedentary Time Predicting Positive and Negative Affect

Fixed effects	Positive affect (end-of-day)				Negative affect (end-of-day)
	Model 1		Model 2		Model 1		Model 2
	Est.	95% CI	Est.	95% CI	Est.	95% CI	Est.	95% CI
Intercept	6.13***	[4.12, 8.12]	5.31***	[2.92, 7.70]	−0.33	[−1.31, 0.65]	−1.79**	[−3.06, −0.53]
Sedentary time (WP)	−0.10***	[−0.14, −0.05]	−0.10***	[−0.14, −0.05]	0.04*	[0.0001, 0.07]	0.04*	[0.002, 0.10]
Sedentary time (BP)	−0.01***	[−0.01, −0.002]	−0.01***	[−0.01, −0.003]	0.001	[−0.000, 0.003]	0.003***	[0.001, 0.005]
MVPA time (WP)	−0.14**	[−0.24, −0.04]	−0.12**	[−0.22, −0.03]	0.07	[−0.002, 0.14]	0.04	[−0.02, 0.10]
MVPA time (BP)	−0.003	[−0.01, 0.004]	−0.005	[−0.01, 0.002]	0.001	[−0.002, 0.004]	0.003	[−0.00, 0.01]]
Wearing time	0.07***	[0.03, 0.11]	0.08***	[0.04, 0.12]	−0.02	[−0.05, 0.01]	−0.02	[−0.05, 0.02]
Morning pain			−0.04	[−0.34, 0.26]			0.12	[−0.07, 0.30]
Morning positive affect			0.17***	[0.10, 0.24]			0.01	[−0.02, 0.05]
Morning negative affect			−0.02	[−0.13, 0.19]			0.15***	[0.06, 0.24]
Age			0.004	[−0.02, 0.03]			−0.01*	[−0.02, −0.001]
Gendera			0.27	[−0.16, 0.70]			0.27*	[0.05, 0.50]
Duration of OA			−0.01	[−0.03, 0.01]			0.004	[−0.004, 0.01]
General health			0.14	[−0.10, 0.37]			−0.01	[−0.10, 0.08]
Depressive symptoms			−0.06**	[−0.10, −0.02]			0.06***	[0.03, 0.08]

BP between-person; MVPA moderate to vigorous physical activity; OA osteoarthritis; WP within-person.

^a1 = male, 2 = female.

*p < .05, **p < .01, ***p < .001.

Among covariates, baseline depressive symptoms were associated with lower positive affect and higher negative affect whereas younger age and being female were associated with higher negative affect. Furthermore, time spent in MVPA was associated with lower positive affect at the end of the day. To explain this finding, we conducted mediation analysis to test whether the increased pain due to MVPA can explain the within-person association between MVPA and lower positive affect at the end of the day. Results failed to support this speculation because pain was not significantly associated with lower positive affect within individuals (b = 0.001, p = .990).

We also explored the possible mechanisms underlying the within-person associations between daily SB and subsequent affect. Specifically, we tested the mediation effects of some daily social support/interaction variables: spouse’s activity-related support (autonomy support, persuasion, and pressure/control) and marital interactions (tension and enjoyment, see Martire et al. [21] for details on these scales) reported by the patients at the end of each day. The results demonstrated that, among all tested mediators, only the lower levels of autonomy support from the spouse partially explained the within-person effects of SB on the lower levels of positive affect (indirect effect = −0.01, 95% CI [−0.014, −0.001], direct effect of SB = −0.09, 95% CI [−0.14, −0.04]), as well as the higher levels of negative affect at the end of the day (indirect effect = 0.003, 95% CI [0.0001, 0.006], direct effect of SB = 0.03, 95% CI [−0.005, 0.07]).

In addition, cross-day lagged analyses were conducted in which patients’ SB today predicted pain and affect the next morning controlling for pain and affect today. One significant effect was found: patients’ SB today predicted higher negative affect the next morning (b = 0.02, p = .046, f² = 0.001). Patient’s SB today did not significantly predict their positive affect or pain the next morning. We also explored the bidirectional associations between SB and pain or affect by testing whether tonight’s pain or affect predicted the next day’s SB and these associations were not statistically significant.

Finally, we explored whether the associations between daily SB and subsequent pain or affect (within a day or cross-day) were moderated by patients’ general level of SB, MVPA, as well as baseline health status or depressive symptoms. We found no significant moderation effects.

Discussion

Older adults are the most sedentary group [4] and a sedentary lifestyle has important health implications, especially for the older adults with OA. The purpose of the present study was to examine how SB was associated with pain and affect in daily life for older adults with knee OA. Our findings revealed that time spent in SB throughout the day was associated with less pain but worse affect (lower positive affect and higher negative affect) at the end of the day, independent of the effects of time spent in MVPA that day, as well as individual differences in age, gender, OA duration, general physical health, and baseline depressive symptoms. In addition, the connection between SB and increased negative affect extended to the next day. Consistent with previous studies [5], we also found evidence that individuals who spent more time in SB in general reported lower positive affect and higher negative affect in general than those who spent less time in SB.

Our finding that daily SB was associated with less pain at the end of the day supported our hypothesis and was consistent with a recent study among older adults [12]. We also found that daily MVPA was associated with more pain at the end of day, in line with previous research [13, 21]. However, these findings should not be viewed as evidence contradicting previous research suggesting that SB is a health risk [4] and that physical activity is an effective treatment for OA [8]. Our cross-day lagged analyses revealed that the reduction in pain after SB was transient and did not extend to the next morning. In addition, we also found that individuals with higher overall levels of SB did not report significantly less (or more) pain than their counterparts. In fact, the short-term decreases in pain after SB may explain why OA patients tend to be highly sedentary [7] and have difficulty engaging in physical activity or adhering to exercise interventions [36], and underscores the challenge faced by OA patients to break up prolonged SB and engage in more physical activity in daily life.

Our findings regarding negative affective responses to daily SB provide strong evidence to support the unique deleterious effect of SB on psychological health outcomes, independent of physical activity. Most previous studies used a cross-sectional design and thus could not draw any conclusions regarding the temporal ordering of SB and health outcomes. By showing the predictive effects of SB on subsequent affect (e.g., at the end of the day or the next morning) controlling for affect measured in the morning, our results demonstrate the temporal order between SB and affect in an OA sample. The mediation analyses further suggest that prolonged SB was associated with lower levels of spouses’ autonomy support, which partially explained the worse emotional well-being following SB. Furthermore, given the known link between daily emotional experiences and depression [37], our findings shed light on the daily processes through which SB could be linked with long-term psychological well-being.

Our finding that SB was associated with both positive affect and negative affect (within-person) is not entirely consistent with one recent daily diary study which found significant within-person associations between SB and subsequent positive affect but not negative affect among midlife women [16]. These divergent findings may be due to differences in populations that were sampled and their physical health status. Specifically, negative associations between SB and health may be more pronounced for the older adults with OA in our study due to their worse physical health in general [12].

Finally, we also found that more than usual time spent in MVPA was associated with less positive affect at the end of the day, and this association could not be explained by increased pain after MVPA. Similar to this finding, a previous study among older adults with OA found that sedentary older adults with knee OA did not experience the same psychological benefits following acute exercise as younger, healthy individuals [38]. One possible explanation is that adults with OA experience elevated fatigue or physical exhaustion after physical activity [13], which may reduce their positive feelings. Future research is needed to better understand the mechanisms through which SB and physical activity relate to OA patients’ emotional well-being on a daily basis.

Together, our findings extend the literature in several ways. First, among one of the first studies focused on the within-person associations between SB and daily outcomes, our study provides useful information regarding the daily dynamics between SB and physical and psychological health outcomes. Second, by simultaneously examining the effects of SB and MVPA, our results provided strong evidence to support the unique detrimental effects of daily SB on emotional well-being, independent of the effects of physical activity. Third, our finding regarding the transient “pain-reduction” effect of SB highlights the potential challenge of reducing prolonged SB for adults with chronic conditions, such as OA, suggesting an interesting direction for future research.

Our findings also have high public health significance and implications for clinical practice. Particularly, the current recommendations for OA patients’ symptoms management has only focused on engaging in physical activity, especially MVPA, but our findings, along with previous research [3], suggest that adults can meet the physical activity recommendations yet be sedentary. Therefore, future interventions should aim to reduce or break down the prolonged sitting or excessive time spent in SB, in addition to increasing physical activity. Our findings that MVPA related to more pain and less positive affect at the end of the day also highlight the challenge of encouraging older adults with OA to reduce SB and engage in MVPA. There is growing evidence supporting the health benefits of light intensity activity, such as leisure walking, shopping, or light housework [39, 40]. As suggested by previous researchers [12, 39], changing sedentary lifestyle by increasing light intensity activity may not only confer important health benefits, but it may also be a more feasible intervention target for older adults, especially those with chronic conditions.

In addition, our finding regarding the immediate negative emotional responses to longer-than-usual SB on the same day points to the need of real-time intervention. Advances in wearable technology provide opportunities to monitor and provide feedback to reduce SB in daily life. For example, prompts triggered by extended periods of inactivity have proven to be effective at reducing sedentary time [41]. Finally, it would be important for future interventions to incorporate education about the potential health risks of SB and remind OA patients that it may be appealing for them to sit longer or move less, due to the temporary reduction of pain, but reducing SB and engaging in physical activity are important for psychological well-being and health.

Limitations and Future Directions

There are several limitations of this study that present promising avenues for future research. First, measuring SB objectively using accelerometer is a strength but we were not able to differentiate the impact of different types of SB. For example, previous studies found that TV watching but not computer use or reading time was associated with cardio-metabolic risk factors [42] and depression [15]. Future research that incorporates objective assessment and self-reported information on different SB types is needed to better understand the health consequences of distinct types of SB and develop more tailored interventions. Second, it is worth noting that the participants in our study reported on average a low level of daily pain and thus our results may not generalize to OA patients who experience higher levels of daily pain. Patients in severe pain may have a different pattern of daily activity (e.g., spend even more time in SB and less time in physical activity), and/or experience different physical and psychological responses. Third, the underlying daily mechanisms through which SB relates to daily pain and affect are still not thoroughly understood. Particularly, identifying the factors that contribute to the temporary “pain-reduction” effect of SB would help patients to overcome the psychological barriers of reducing SB and increasing physical activity (e.g., fear of pain). Even though we examined the mediation effects of several social interaction factors, future research would benefit from exploring other daily or momentary biological and social processes, which could help to further elucidate the pathways linking SB with daily affect. Forth, the within-person effects of daily SB on pain and affect were small (Cohen’s f²≤ 0.01) and accounted for 1% of the variance in daily pain or affect. Although small in magnitude, the effect sizes found in our study are comparable to those reported in previous EMA or dairy studies on physical activity, pain, and affect (e.g., [43, 44]). Given the prevalence of sedentary lifestyles among older adults, especially those with OA, it is plausible that these small effects of SB could accumulate and have significant influences on older adults’ physical and mental health over time. Therefore, future research would benefit from examining how short-term associations between SB and daily outcomes predict long-term health outcomes over months or even years. Finally, our study utilized the morning and end-of-day assessments of pain and affect and thus was not able to capture change in these variables over shorter periods. Future studies with more momentary assessments per day would be able to examine the diurnal variations in pain and affect, as well as the duration of the effects of SB and physical activity on these outcomes.

Authors’ Statement of Conflict of Interest and Adherence to Ethical Standard Authors Ruixue Zhaoyang and Lynn M. Martire declare that they have no conflict of interest.

Ethical Approval All procedures, including the informed consent process, were conducted in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000.

Informed Consent Informed consent was obtained from all individual participants included in the study.

AcknowledgementPreparation of this article was supported by National Institutes of Health Grants (R01 AG026010 and K02 AG039412 to L. Martire).