The average minutes of MVPA were measured through a connected activity tracker (Actigraph device) worn by the participants to evaluate the effects of Voltaren Gel on physical activity. A mixed model with repeated measures (MMRM) was used to analyse the change from Baseline in MVPA average minutes. Change from Baseline was calculated by subtracting the Baseline average MVPA value from the average MVPA value at Week 1. Baseline value was calculated as the sum of minutes of MVPA over the Baseline period divided by duration of Baseline period, where Baseline period was the number of non-missing days between Day -6 and Day 0. A positive change from Baseline indicated improvement. Modified Intent-To-Treat (mITT) population included all participants who met the inclusion/exclusion criteria, who used study product at least once and had data from at least one post Baseline quality of life (QoL) questionnaire to support at least one of the secondary endpoint assessments.

mITT Population. This study was designed as a single-arm real-world evidence (RWE) study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the three formulations within the population; therefore, all participants were analyzed as a single group. The reported results apply to the entire study population ("all groups") because there were no separate Arms/Groups in the study design.

The average minutes of MVPA were measured through a connected activity tracker (Actigraph device) worn by the participants to evaluate the effects of Voltaren Gel on physical activity. A MMRM was used to analyse the change from Baseline in MVPA average minutes. Change from Baseline was calculated by subtracting the Baseline average MVPA value from the average MVPA value at Week 2. Baseline value was calculated as sum of minutes of MVPA over the Baseline period divided by duration of Baseline period, where Baseline period was the number of non-missing days between Day -6 and Day 0. A positive change from Baseline indicated improvement.

mITT population. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the three formulations within the population; therefore, all participants were analyzed as a single group. The reported results apply to the entire study population ("all groups") because there were no separate Arms/Groups in the study design.

The average minutes of MVPA were measured through a connected activity tracker (Actigraph device) worn by the participants to evaluate the effects of Voltaren Gel on physical activity. A MMRM was used to analyse the change from Baseline in MVPA average minutes. Change from Baseline was calculated by subtracting the Baseline average MVPA value from the average MVPA value at Week 3. Baseline value was calculated as sum of minutes of MVPA over the Baseline period divided by duration of Baseline period, where Baseline period was the number of non-missing days between Day -6 and Day 0. A positive change from Baseline indicated improvement.

mITT population. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the three formulations within the population; therefore, all participants were analyzed as a single group. The reported results apply to the entire study population ("all groups") because there were no separate Arms/Groups in the study design.

Daily average number of steps taken were measured through a connected activity tracker (Actigraph device) worn by the participants to evaluate the effects of Voltaren Gel on functional mobility. MMRM was used to analyse the change from Baseline in daily average number of steps taken. Change from Baseline was calculated by subtracting the average number of steps taken at Baseline from the average number of steps taken at each indicated post-Baseline timepoint. Baseline value was calculated as the sum of steps taken over the Baseline period divided by duration of Baseline period, where Baseline period was the number of non-missing days between Day -6 and Day 0. A positive change from Baseline indicated improvement.

mITT population. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the three formulations within the population; therefore, all participants were analyzed as a single group. The reported results apply to the entire study population ("all groups") because there were no separate Arms/Groups in the study design.

Sedentary/non-sedentary time was measured using a connected activity tracker (Actigraph device) worn by the participants to evaluate the effects of Voltaren Gel on functional mobility. Ratio of sedentary/non-sedentary time was calculated as = Total number of minutes defined as sedentary behaviour divided by Total number of non-sedentary behaviour minutes where number of non-sedentary behaviour minutes in the day was derived as the sum of light activity and MVPA. Change from Baseline was calculated by subtracting the Baseline ratio of sedentary/non-sedentary time from the ratio at each indicated post-Baseline timepoint. Baseline was defined as the number of non-missing days between Day -6 and Day 0. A negative change from Baseline indicated improvement.

mITT population. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the three formulations within the population; therefore, all participants were analyzed as a single group. The reported results apply to the entire study population ("all groups") as there were no separate Arms in the study design.

Gait was assessed through speed and step irregularity measured via cadence using a connected activity tracker (Actigraph device) worn by the participants. Change from Baseline was calculated as by subtracting the average cadence at Baseline from the cadence at each indicated post-Baseline timepoint. Baseline was defined as the number of non-missing days between Day -6 and Day 0. A positive change from Baseline indicated improvement.

mITT population. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the three formulations within the population; therefore, all participants were analyzed as a single group. The reported results apply to the entire study population ("all groups") because there were no separate Arms/Groups in the study design.

Gait was assessed through speed and step irregularity measured via gait speed using a connected activity tracker (Actigraph device) worn by the participants. Change from Baseline was calculated by subtracting the average gait speed at Baseline from the Average gait speed at each indicated post-Baseline timepoint. Baseline was defined as the number of non-missing days between Day -6 and Day 0. A positive change from Baseline indicated improvement.

mITT population. Overall number analyzed is the number of participants with data available for analyses of this outcome measure. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population; So, all participants were analyzed as a single group. The reported results apply to the entire study population (all groups) as there were no separate Arms in the study design.

Indices of morning stiffness were assessed through levels of mobility 30 minutes post-wake. Functional mobility was measured through a connected activity tracker (Actigraph device) worn by the participants. Daily morning stiffness 30 minutes post-wake was defined as the total vector magnitude counts from Actigraph device 30 minutes post-wake. Change from Baseline was calculated as = (Average morning stiffness 30 minutes post-wake at each indicated post-Baseline timepoint minus Average Baseline morning stiffness 30 minutes post-wake). Baseline was defined as the number of non-missing days between Day -6 and Day 0. A negative change from Baseline indicated improvement.

mITT population. Overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

Indices of morning stiffness were assessed through levels of mobility 60 minutes post-wake. Functional mobility was measured through a connected activity tracker (Actigraph device) worn by the participants. Daily morning stiffness 60 minutes post-wake was defined as the total vector magnitude counts from Actigraph device 60 minutes post-wake. Change from Baseline was calculated as = (Average morning stiffness 60 minutes post-wake at each indicated post-Baseline timepoint minus Average Baseline morning stiffness 30 minutes post-wake). Baseline was defined as the number of non-missing days between Day -6 and Day 0. A negative change from Baseline indicated improvement.

mITT Population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

Participants completed the WOMAC questionnaire daily using eDiary. The WOMAC questionnaire consisted of 24 questions categorised in 3 subscales - Pain, Stiffness, Physical function. The physical function subscale consisted of 17 questions related to level of difficulty in performing functions, on average, during the last 48 hours scored on a 5-point Likert scale with scores ranging from 0 to 4, where score 0=none, 1=slight, 2=moderate, 3=severe, 4=extreme. Total WOMAC physical function subscale score was normalized on a 0 to 100 scale calculated as = (sum of raw score items)*1.47. Higher scores indicated more difficulty in performing physical functions. Change from Baseline was calculated by subtracting the normalized WOMAC subscale score at Baseline from the normalized WOMAC subscale score at each indicated post-Baseline timepoint. Baseline was defined as Day 0. A negative change from Baseline indicated improvement.

mITT Population. Overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

The average minutes of MVPA on each day were measured through a connected activity tracker (Actigraph device) worn by the participants to assess the functional mobility. Baseline was calculated as the average of MVPA values collected during the Baseline period, where Baseline period was the number of non-missing days between Day -6 and Day 0.

mITT population. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population; So, all participants were analyzed as a single group. The reported results apply to the entire study population (all groups) as there were no separate Arms in the study design.

Physical function subscale score of the WOMAC questionnaire was used to study participant's perceived ability to exercise more regularly. It consisted of 17 questions related to the level of difficulty in performing functions, on average, during the last 48 hours scored on a 5-point Likert scale with scores ranging from 0 to 4, daily using eDiary where score 0=none, 1=slight, 2=moderate, 3=severe, 4=extreme. Total WOMAC physical function subscale score was normalized on a 0 to 100 scale calculated as = (sum of raw score items)*1.47. Higher scores indicated more difficulty in performing physical functions.

mITT population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

Participants rated the pain intensity experienced by them daily within the eDiary, using NRS. NRS is an 11-point scale with scores ranging from 0 to 10 where score 0 = no pain, 1 = slight pain, 2-3 = mild pain, 4-6 = moderate pain, 7 = severe pain, 8 to 9 = extreme pain, 10=pain as bad as it could be. Higher scores indicated greater pain intensity. Change from Baseline was calculated by subtracting the Baseline NRS score from the NRS score at each indicated post-Baseline timepoint. Baseline was defined as Day 0. A negative change from Baseline indicated improvement.

mITT Population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

Participants completed the WOMAC questionnaire daily. The WOMAC questionnaire consisted of 24 questions each scored on a 5-point scale: 0=none, 1=slight, 2=moderate, 3=severe, 4=extreme. The WOMAC questionnaire was categorized in 3 subscales - Pain (5 questions, score ranges 0-20), Stiffness (2 questions, score ranges 0-8), Physical function (17 questions, score ranges 0-68). Each subscale score was normalized on a 0 to 100 scale calculated as: Pain subscale score= (sum of raw score items in dimension)*5; Stiffness subscale score= (sum of raw score items in dimension)*12.5; Physical function subscale score= (sum of raw score items in dimension)*1.47. WOMAC Total score = sum of the 3 normalized WOMAC subscale scores. Thus, WOMAC total score ranges from 0 to 300 with 0 being the best and 300 being the worst. Change from Baseline was calculated by subtracting the Baseline score from score at each indicated post-Baseline timepoint. A negative change from Baseline indicated improvement.

mITT Population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

The pain subscale of the WOMAC questionnaire consisted of 5 questions related to severity of pain experienced by the participant when walking, stair climbing, at night, at rest, weightbearing, on average, during the last 48 hours. Participants scored the questions on a 5-point Likert scale ranging from 0 to 4 daily using eDiary where score 0=none, 1=slight, 2=moderate, 3=severe, 4=extreme. The total pain subscale score was normalized on a 0 to 100 scale calculated as = (sum of raw score items in dimension)*5. Higher scores indicated more pain. Change from Baseline was calculated by subtracting the Baseline normalized WOMAC pain subscale score from the normalized score at each indicated post-Baseline timepoint. Baseline was defined as Day 0. A negative change from Baseline indicated improvement.

mITT Population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

The stiffness subscale of the WOMAC questionnaire consisted of 2 questions related to severity of stiffness experienced by the participant at morning and during the day, on average, during the last 48 hours. Participants scored the questions on a 5-point Likert scale with scores ranging from 0 to 4 daily using eDiary, where score 0=none, 1=slight, 2=moderate, 3=severe, 4=extreme. The total stiffness subscale score for was normalized on a 0 to 100 scale calculated as = (sum of raw score items in dimension)*12.5. Higher scores indicated more stiffness. Change from Baseline was calculated by subtracting the Baseline normalized WOMAC stiffness subscale score from the normalized score at each indicated post-Baseline timepoint. Baseline was defined as Day 0. A negative change from Baseline indicated improvement.

mITT Population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

KSS measured the subjective level of sleepiness/alertness during the day. Participants rated how sleepy or alert they were feeling using KSS with scores ranging from 1 to 9 where 1 = extremely alert, 2 = very alert, 3 = alert, 4 = fairly alert, 5 = neither alert nor sleepy, 6 = some signs of sleepiness, 7 = sleepy, but no efforts to keep alert, 8 = sleepy, some efforts to keep alert, 9 = very sleepy, great effort to keep alert, fighting sleep. Higher scores indicated greater sleepiness. Change from Baseline was calculated by subtracting the Baseline score from score at each indicated post-Baseline timepoint. Baseline was defined as Day 0. A negative change from Baseline indicated improvement.

mITT Population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").

Health-related QoL of participants was assessed using the EQ-5D-5L questionnaire. It consisted of 5 dimensions (mobility, self-care, usual activities, pain/discomfort, anxiety/depression) with 5 response levels for each dimension where Level 1 = no problems, Level 2 = slight problems, Level 3 = moderate problems, Level 4 = severe problems, Level 5 = unable to perform activities/extreme problems. Participants were asked to indicate their health by ticking the box next to the most appropriate response level for each dimension. The response levels were used to derive a health index score based on the Crosswalk Value Set for the United States ranging from less than 0 (worst imaginable health) to 1 (best imaginable health) where higher scores indicated better health. Change from Baseline was calculated by subtracting the Baseline score from score at each indicated timepoint. Baseline was defined as Day 0. A positive change from Baseline indicated improvement.

mITT population. Here, overall number analyzed is the number of participants with data available for analyses of this outcome measure. Number analyzed is the number of participants with non-missing data at the indicated timepoint. This study was designed as a single-arm RWE study to evaluate the effectiveness of Voltaren gel. There was no intention to compare the effectiveness of the 3 formulations within the population. The reported results apply to the entire study population ("all groups").