Table 2 Overview of characteristics and results of studies reporting on interventions aiming to enhance physical activity behavior after THA/TKA

Beck, Beyer [40]

Germany

THA

RCT

N = 160

IG: N = 80, 52.5% female, median age 59.0 yrs. (51.1; 69.7), median BMI 26.4 kg/m² (23.8; 28.6).

CG: N = 80, 63.8% female, median age 61.9 yrs. (52.5; 70.0) median BMI 25.9 kg/m² (23.7; 30.4).

General medical eligibility for hip rehab sports therapy, stable implant, age 18 yrs. or older.

Measurements at baseline, 6 and 12 months after surgery

IG: rehabilitation sports therapy program (endurance, strength, coordination, flexibility).

CG: no rehab sports therapy.

General

Isokinetic dynamometry, postural stability, lactate threshold, WOMAC, HHS, pain (VAS), UCLA scale, EuroQol, EQ-5D.

Strength capacity not significantly better in IG. At one year IG subjects had less pain (WOMAC pain score (p = 0.023), size of effect small (r = 0.27). Health-related quality of life higher in intervention group at six months, size of effect small (p = 0.036, r = 0.25). The other parameters showed no significant changes differences. Median UCLA score was 7 in both groups at both six and twelve months.

No benefit of sports rehabilitation on functional outcomes compared to controls. Positive trends seen in some parameters. The unexpectedly high dropout rate had been underestimated in the planning of the trial.

Heiberg and Figved [41]

Norway

THA

RCT

N = 60, mean age 70 yrs. (range 50–87)

IG: N = 30, 70% female, education > 12 yrs. 57%

CG: N = 30, 43% female, education > 12 yrs. 57%

Primary THA for OA and residence within an approximate 30-km radius

from the hospital.

October 2008 to March 2010. Measurements preop and 3 & 5 months, 1 & 5 yrs. post-THA.

IG: a supervised walking skills-training program 3–5 months post-THA. 12 sessions, 70 min per session, 2x week.

CG: not allowed to attend supervised physiotherapy during the same period, but encouraged to continue training on their own and to keep generally active.

Walking

6MWT, SCT, active hip ROM flexion/extension, 30-CST, HOOS, Self-efficacy (self-constructed), UCLA activity scale

IG and CG were equal on outcome measures of physical functioning, pain, and self-efficacy at 5 years (p > 0.05). In the total group, recovery course was unchanged from 1 to 5 years (p > 0.05), except for 9% improvement in ROM (p < 0.001) and increase in time on SCT of 18% (p = 0.004). Preop HOOS pain (p = 0.022) and HOOS sport (p = 0.019) predicted UCLA activity scale 5 years post-THA.

5 yrs. post-THA, the CG had caught up with the IG on physical functioning, participants led an active lifestyle. Those with worse preop scores on pain and physical functioning in sport were at risk of being less physically active in the long-term post-THA.

TKA

RCT (no blinded allocation)

N = 48 60% female, mean age 67 yrs.

IG: N = 25

CG: N = 23

Persons post-TKA (55–75 yrs) 1–5 yrs. postop, committed to hiking 2–3 times/week over a 3-month period.

July–December 2015 Measurements prior to intervention period (pre-test), immediately after the 3-month intervention period (post-test) and 2 months after (retention-test).

IG: 3-month guided hiking program (2–3 times/week)

CG: activities of daily living.

Hiking

SCT, KOOS, SF-36, extensor and flexor torque.

After hiking program, IG achieved faster overall walking times on the SCT. Time decreased from 4.3 ± 0.6 s (pre-test) to 3.6 ± 0.4 s (posttest) for the stair ascent (p = 0.060) and from 3.6 ± 0.6 s (pre-test) to 3.2 ± 0.5 s (post-test) for the stair descent (p = 0.036). IG showed significant improvement on KOOS subscales (symptoms/sport, recreation/QOL) from pre-test to retention-test (p < 0.01). No significant changes observed in IG. No effect on SF-36.

Results indicate moderate improvement in functional abilities and QoL of persons post-TKA who participated in a 3-month guided hiking program compared with CG subjects. Hiking did not have any acute detrimental effects on persons post-TKA during this study period.

TKA

RCT

N = 97 58% female, mean age 58 yrs. (SD 4.8).

Persons < 65 yrs. suffering from

debilitating knee OA and awaiting TKA, participating in a paid or voluntary job or working as an informal caregiver, and able to define and perform personal rehabilitation goals.

October 2015 to

November 2017. Measurements preop and 6 months postop.

IG: Intervention using GAS. 3 personal activity goals: 1 ADL activity, 1 work activity, 1 leisure-time activity.

CG: regular outpatient

physical therapy

General

Accelerometer, Activ8. 5–7 consecutive days (24/7 in the month prior to TKA and 6 months post-TKA).

For the total group, a significant increase in PA of 9 min (±37) per day (p = 0.01) was observed and a significant decrease in sedentary time of 20 min (±79) per day (p = 0.02). No difference in standing time (p = 0.11). No difference CG and IG regarding changes in PA.

A small but significant increase in overall PA post-TKA, but no difference between GAS-based rehabilitation and standard rehabilitation was found.

Losina, Collins [44]

USA

TKA

A factorial RCT

N = 202 57% female, mean age 65 yrs. (SD 8), 68% Bachelor degree, mean BMI 31 (SD 6)

IG THC: N = 49

IG FI: N = 50

IG THC + FI: N = 52

CG: N = 51

Prior to TKR, participants walked a mean of 5032 steps/day (SD 2771). With the exception of step count, all characteristics are balanced across the arms.

Participants excluded if < 40 yrs., did not speak English, resided in nursing home, scheduled to undergo contralateral TKR or other surgery requiring hospitalization within 6 months, previously diagnosed with inflammatory arthritis or osteonecrosis affecting the knee, had a comorbidity that might prevent safe performance of moderate ambulatory PA, required a wheelchair or walker to ambulate preoperatively or did not have regular Internet access.

November 2013 through January 2016. Measurements preop and 6 months postop

4 groups: Attention control (CG), telephonic health coaching (THC), financial incentives (FI), THC + FI.

Walking

Accelerometer (Fitbit Zip), demographics, social and employment history, resource utilization, Knee injury Outcomes and Osteoarthritis Score (KOOS), EuroQol-5D (EQ-5D-3L), a general health Visual Analogue Scale (VAS), Risk Taking Index, Work Productivity and Activity Impairment questionnaire, Yale Physical Activity Survey, self-reported knee range of motion, components of the SF-36, MHI-5, Vitality Score.

Average daily step count at 6 months ranged from 5619 (SD 381) in THC arm to 7152 (SD 407) in THC + FI arm. Daily step count 6 months post-TKR increased by 680 (95% CI: − 94–1454) in control arm, 274 (95% CI: − 473–1021) in THC arm, 826 (95% CI: 89–1563) in FI arm, and 1808 (95% CI: 1010–2606) in THC + FI arm. PA increased by 14 (SD 10), 14 (SD 10), 16 (SD 10), and 39 (SD 11) minutes in the control, THC, FI, and THC + FI arms, respectively.

A dual THC + FI intervention led to substantial improvements in step count and PA post-TKR.

Paxton, Forster [45]

USA

TKA

RCT

N = 45

IG: N = 22, female 50%, age 64 yrs. (SD 6), BMI 26.4 (SD 8.6)

CG: N = 23, female 57%, age 63 yrs. (SD 7), BMI 29.9 (SD 10.7).

Participants 50–75 yrs. who underwent unilateral TKA

Initial assessments after completion of outpatient rehabilitation (6–8 weeks postop). Final assessments 12 weeks after beginning of intervention

IG: 12-wk program real time PA and face-to-face feedback

CG: no PA feedback (current standard of care post-TKA)

General

Feasibility: retention, adherence, dose goal attainment, and responsiveness with pre- and post-intervention testing.

PA: accelerometer (GT3X Actigraph Activity Monitor)

Functional performance: TUG, 6-MWT, 4-MWT.

IG: 100% retention, 92% adherence (frequency of feedback use), and 65% dose goal attainment (frequency of meeting goals). IG average daily step count increased from 5754 (2714) (preop) to 6917 (3445) steps/day (postop).

The PA feedback intervention is a feasible intervention to use as an adjunct to conventional rehabilitation for persons with TKA and seems to be effective.

Piva, Almeida [46]

USA

TKA

RCT

N = 44

IG: N = 22, female 82%, age 68.1 yrs. (SD 7.5), BMI 31.2 (SD 3.6)

CG: n = 22, female 59%, age 68.3 yrs. (SD 5.5), BMI 29.3 (SD 4.1)

Participants > = 50 yrs., unilateral TKA 3–6 months before, no regular participation in exercise program

October 2011 to August 2013

6 months FU

IG: CBI program with exercise and education component. The education component of CBI to promote PA and healthy eating included two 30-min educational lectures during intervention week 1; mini-sessions of PA promotion were delivered in the subsequent weeks.

CG: SCE.

3-month program followed by 3 months home exercise program (same for both groups)

Exercises

Feasibility of interventions assessed by adherence to supervised exercises, attrition and knee pain (WOMAC pain).

Outcome measures: physical function (WOMAC PF, SF-36 PF, battery of performance-based tests) and PA using 7 days accelerometry.

Compared to the SCE group, the CBI group had less pain (p = 0.035) and better physical function based on the SF-36 (p = 0.017) and the single-leg stance test (p = 0.037). The other outcome measures did not demonstrate statistically significant differences between the two groups. Results from the responder analysis demonstrated that the CBI group had a 36% higher rate of responders in physical function than the SCE group. Also, the CBI group had 23% more responders in the combined domains of physical function and PA.

The CBI was found to be safe and well-tolerated, showing better outcome than the standard of care exercise program.

Pozzi, Madara [47]

USA

THA

Case-series (n = 2)

N = 2 62 yrs., one female, one male

Historical cohort as comparison (N = 32)

Persons 40–70 yrs., 3–9 months after unilateral THA

Measurements at baseline, end of intervention, 12 months post-THA

Exercise and education intervention, 18 supervised sessions over 6 weeks

Exercises

Feasibility and preliminary efficacy.

HOS, hip abductor muscle strength, maximal voluntary isometric strength for quadriceps muscle, functional performance (TUG, SCT, 6-MWT, FSS), IPAQ, PSFS

Outcomes reported at individual level. Improved leg strength, weekly PA, and ability to perform demanding recreational and sports participation, without producing adverse effects. Feedback on the additional value of the health coach differed, leading to the conclusion that not all patients may benefit from this type of behavioral intervention.

This intervention could potentially increase activity levels and restore recreational participation in patients post-THA. Identifying those who may benefit from this intervention may help optimize outcomes without overusing resources.

Smith, Zucker-Levin [48]

USA

TKA

RCT

N = 60 Female 65%, BMI 36.4 (SD 4.7) 10–18 months post-TKA

IG: N = 30

CG: N = 30

Both groups: N = 24 ompleted final testing

Obese persons 1 year after unilateral TKA

Measurements at baseline, 8 weeks, end of intervention (16 weeks)

Both groups: 16-week tailored resistance and aerobic training designed to be completed at home with no supervision and minimal equipment based on ACSM guidelines for exercise prescription

IG: exercise program and fitness tracker

CG: exercise program only

Exercises

6-MWT, WOMAC, SF-36, ROM, knee extension strength.

Improvement on all outcome measures. The anecdotal reports from patients who received the fitness tracker technology indicated that many participants were engaged by the device and found it motivational (but no improvement in compliance with prescribed exercises).

The 16-week home-based exercise program is feasible and effective in improving strength and walk performance.

Trudelle-Jackson, Hines [49]

USA

TKA

RCT

N = 13 Female 85%, age 63.5 yrs. (SD 7), BMI 34.8 (SD 7.6)

IG: N = 7

CG: N = 6

Persons at least 6 months after primary unilateral or bilateral TKA, > 40 yrs

Pre- and post-test

IG: High-Velocity Training Exercises Plus Step-Monitoring, 8 weeks

CG: Step monitoring only

Exercises

Muscle strength, muscle power, functional performance (6-MWT, SCPT), habitual walking behavior: number of steps/day along with minutes/week of moderate and/or vigorous PA (pedometer)

PA behavior: differences between pre-intervention and post-intervention values of PA behavior were not significant for minutes of MVPA (p = 0.09, r = −0.39) or for average daily steps (p = 0.09, r = 0.39) for the high-velocity training intervention group. The CG had significant improvement in number of daily steps (p = 0.01, r = 0.64), but not in minutes of MVPA (p = 0.38, r = 0.11).

No significant differences between IG and CG on amount of change in any of the outcomes. Based on these results, we could argue that providing a step-monitoring device like the simple pedometer used in this study or one of the many commercially available wearable technology may be more cost-effective than prescribing and monitoring a high-velocity training program.

Van der Walt, Salmon [50]

Australia

TKA

THA

RCT

N = 163

IG (FB): N = 81

CG (NFB): N = 82

Adults undergoing primary elective THA or TKA, 1 day postop

May–December 2016. Accelerometer measurements on days 1–14 postop, (PROMs) preop and 6 months postop.

FB group: feedback by means of accelerometer on daily step goal.

NFB group: no feedback for 2 weeks postop and no daily step goal.

Walking

Garmin Vivofit 2 accelerometer, KOOS or HOOS,EuroQol-5D, satisfaction component of KSS, satisfaction with outcome of surgery, one-item question if they would have the same surgery again under the same circumstances

FB subjects had a significantly higher (p < 0.03) mean daily step count by 43% in week 1, 33% in week 2, 21% in week 6, and 17% at 6 months, compared with NFB. FB subjects were 1.7 times more likely to achieve a mean 7000 steps/day than NFB subjects at 6 weeks postop (p = .02). No significant difference in PROMs at 6 months. 90% of FB and 83% of NFB participants reported satisfaction with surgery results (p = 0.08). 6 months postop, 70% of subjects had a greater mean daily step count compared with their preop level.

The CBI program improves physical function and PA in patients several months post-TKA.