Secondary Logo

Journal Logo

Systematic Review

Effect of Footwear on Joint Pain and Function in Older Adults With Lower Extremity Osteoarthritis

Wagner, Amy PT, DPT, GCS; Luna, Sarah PT, DPT, GCS

Author Information
Journal of Geriatric Physical Therapy: April/June 2018 - Volume 41 - Issue 2 - p 85-101
doi: 10.1519/JPT.0000000000000108
  • Free

Abstract

INTRODUCTION

Background and Purpose

Osteoarthritis (OA) is the most common form of arthritis in older adults, frequently occurs in weight-bearing joints, and interferes with functional activity such as walking and climbing stairs.1 According to the Centers for Disease Control and Prevention, 52.5 million adults in the United States reported being diagnosed with arthritis; this is estimated to increase to 67 million by the year 2030.2 In order to age optimally, these individuals need to maintain high functional ability despite the presence of OA.3 At last estimate, the total cost associated with arthritis and other rheumatic conditions was $128 billion per year, which includes medical expenditures and lost wages.4

In view of a growing population of older adults and increasing costs associated with arthritis, conservative and lower cost management options warrant further investigation. Less conservative management of arthritis, such as arthroscopic and total joint replacement surgeries, injections, and medication, has higher costs and more associated medical risks than conservative treatment such as exercise and weight loss.5–7 For example, risks associated with total joint replacement surgery include postoperative infection, deep vein thrombosis, myocardial infarction, cerebrovascular accident, nerve damage, and failure of the prosthesis.8 Pharmaceutical management risks include adverse drug reactions and drug-to-drug interactions related to polypharmacy.9

Footwear may be a cost-effective and safe intervention for managing arthritis symptoms. Clear recommendations exist in the literature about which types of shoes improve balance and gait. For example, flat, low-heeled shoes with a wide, firm sole are recommended to improve balance and gait quality.10–14 Type of footwear affects gait velocity, cadence, stride time, and stride length in older women when comparing slippers, high heels, standard walking shoe, and barefoot conditions.15 Menant et al12 demonstrated that wearing high-heeled shoes results in decreased gait speed and step length, as well as increased double stance time and postural sway.10 Wearing shoes with high heels also results in decreased forward reach and slower gait velocity.11 In addition, neutral sole hardness maximizes postural stability compared with very soft-soled shoes.10,12–14 These studies demonstrate the relationship between type of shoe and safety or quality of gait, but they do not address the relationship of footwear to pain with walking or function due to OA.

Biomechanical effects of footwear on knee joint load and knee adduction moment (KAM) in older adults have been studied extensively. Lateral wedge insoles (LWIs) unload the medial knee compartment by pronating the lower extremity and increasing valgus forces at the knee and subtalar joints.16–18 LWIs may shorten the knee-ground reaction force lever arm, which reduces the KAM.19,20 Although the literature demonstrates these biomechanical effects of footwear on forces in the lower extremity joints, a systematic review of the effects of different types of footwear on clinical improvement in pain with walking and functional tasks is lacking.

The purpose of this review of the literature was to systematically investigate which type of footwear, including shoe inserts, best reduces lower extremity (knee and foot) joint pain and improves gait, mobility, and quality of life in older adults with OA. The results of this analysis will guide health care providers' clinical decision making regarding recommendations of type of footwear for their patients with lower extremity OA.

METHODS

Search Strategy and Eligibility Criteria

The authors searched CINAHL, SPORTDiscus, PubMed, RECAL, and Web of Knowledge databases for publications from January 1990 to September 2014 using the terms “footwear,” “shoes,” “gait,” “pain,” and “older adult.” Inclusion criteria were (1) participants who were community-dwelling elders of 50 years and older; (2) individuals diagnosed with OA or joint pain in the lower extremity; (3) outcome measures of pain, comfort, function, or gait; (4) human subjects, both male and female; and (5) published in English. Exclusion criteria included studies (1) relating to patients with rheumatoid arthritis, amputation, diabetes, and multiple sclerosis; (2) using modified footwear or custom orthotics including ankle-foot orthoses; (3) exploring the impact of footwear on balance or falls only; (4) with outcomes exclusively measuring biomechanical variables; and (5) single-subject designs, qualitative or narrative reviews.

Article Selection

This systematic review of the literature used the Preferred Reporting Items for Systematic Reviews (PRISMA) guidelines, including the 27-item checklist and flow diagram.21 The PRISMA statement improves transparency in reporting of systematic reviews.21 Two authors independently reviewed the records identified through database searching and from searching other sources such as bibliographic references of included studies, conference proceedings, and dissertations. Duplicate records were removed, and then the authors independently reviewed each study by title and abstract and excluded those that did not meet inclusion and exclusion criteria. Both authors then independently reviewed the full texts of the remaining articles and removed articles that did not meet the study inclusion and exclusion criteria. After each of these steps in the methods, the authors discussed their results and came to consensus on which articles to include in the study.

Quality Assessment of Full Articles

Each article included in the study was first assessed for quality using the Physiotherapy Evidence Database (PEDro) Scale. The PEDro scale is a 11-item tool used to assess the quality of clinical trials.22 It has validity in assessing the methodology of clinical studies,22 and items on the PEDro scale have “fair” to “good” reliability.22,23 Both authors scored the included studies independently using the PEDro scale, and then discussed their results to reach consensus if needed. PEDro scores for each study are included in Table 1. Next, a data extraction form adapted from the Cochrane data extraction form24,25 (see the Appendix) was used for qualitative assessment of each study, by each author independently.

Table 1. - Summary of PEDro Scores for Included Studies
Criterion Study
Baker et al29 Bennell et al26 Dufour et al36 Fang et al27 Koca et al28 Lane et al34 Pham et al30 Toda et al33 Toda and Tsukimura31 Turpin et al35 Wallace32
Eligibility criteria Yes Yes N/A Yes Yes Yes Yes Yes Yes Yes Yes
Random allocation 1 1 N/A 0 1 0 1 0 1 0 0
Concealed allocation 0 1 N/A 0 0 0 0 0 1 0 0
Baseline comparability 1 1 N/A 0 1 0 1 1 1 0 1
Blinded participants 0 0 N/A 0 0 1 1 0 0 0 0
Blinded therapists 1 0 N/A 0 0 0 0 0 0 0 0
Blinded assessors 1 1 N/A 0 0 0 1 1 1 1 0
Adequate follow-up 1 1 N/A 1 1 1 0 1 1 1 1
Intention-to-treat analysis 0 1 N/A 1 1 1 1 1 1 0 0
Between-group comparisons 1 1 N/A 0 1 1 1 0 0 0 1
Point estimates and variability 1 1 N/A 1 1 1 1 1 1 1 1
Total PEDro score 7 8 N/A 3 6 5 7 5 7 3 4
Abbreviation: PEDro, Physiotherapy Evidence Database.

RESULTS

Literature Search Results

The initial systematic search yielded 417 titles. A search of other articles from the bibliographic references of included studies, conference proceedings, and dissertations revealed another 19 articles. After removal of duplicates, 376 records remained. These records were individually screened by title and abstracts with 315 removed as they did not fit the inclusion or exclusion criteria. The authors then individually assessed full text of the remaining 61 articles for eligibility, and then they came to a consensus, with 50 removed, resulting in 11 remaining articles for inclusion in the full qualitative synthesis (see the Figure).

F1
Figure.:
PRISMA flow diagram.

Interventions and Designs

Eleven studies from 2001 to 2014 were included in this final review. Eight of the final 11 articles investigated the effect of lateral shoe wedges on medial knee OA pain, function, and gait.26–33 One explored shoe sole hardness on comfort and pain,34 2 examined subtalar strapping methods on pain in addition to insoles,31,33 1 researched shock-absorbing insoles on walking pain with stair climbing in patients with knee OA,35 and another comparing “good” (athletic shoes or sneakers) to “average” (defined as leather soled or dress shoes) or “poor” (defined as heels, pumps, sandals, or slippers) footwear.36 Outcomes on pain and comfort were examined by 5 randomized control trials,26,28–31 5 quasiexperimental designs,27,32–35 and 1 observational cohort design.36 Outcomes on gait, function, and quality of life were examined by 5 randomized control trials,26,28–31 and 4 quasiexperimental studies.27,32,33,35 Summarized study characteristics including type of study design, duration of intervention, type of footwear, and materials, as well as statistical and clinical outcomes, can be found in Tables 2, 3, and 4.

Table 2. - Effect of Footwear on Pain and Comfort
Author PEDro Score Study Design/Duration Type of Footwear/Intervention Outcomes Significance Mean Difference MCID
Lateral wedge insoles
Bennell et al26 8 RCT
12 mo
Full-length bilateral LWIs (5°) of “high-density ethyl vinyl acetate” vs full-length flat insoles of “easily compressible low-density ethyl vinyl acetate but no wedging.” Worn daily, 6.8-9.1 h/d Within group:
LWIs
WOMAC Pain
WOMAC Stiffness
VAS Average
VAS Walking
Control insoles
WOMAC Pain
WOMAC Stiffness
VAS Average
VAS Walking
Between groups
WOMAC Pain
WOMAC Stiffness
VAS Average
VAS Walking
NR
NR
NR
NR
NR
NR
NR
NR
P > 0.05
P > 0.05
P > 0.05
P > 0.05
−3.5
−5.0
−0.9
−0.9
−6.00
−7.5
−1.3a
−1.4a
−2.0
−3.75
−0.3
−0.3
−9.7
−10.0
−1.1
−1.1
−9.7
−10.0
−1.1
−1.1



Baker et al29 7 RCT crossover
6 wk
LWIs (5°) vs neutral flat “(1/8-inch-thick) insole on side of affected knee” only. Both made of noncompressible NickelPlast material. Worn 7.2-7.7 h/d Between groups:
WOMAC Pain
P = 0.13 −0.03
Fang et al27 3 Quasiexperimental
4 wk
New Balance shoes (model 833) with full-length LWIs on symptomatic side(s), 4° vs neutral wedge on noninvolved side. Both constructed from “custom cork composite similar to Thermocork,” density 60 durometers (shock absorbing and resistant to deformation) Within group:
LWIs
WOMAC Pain
WOMAC Stiffness
WOMAC-Going up or down stairs item on pain scale
P = 0.03b
P = 0.05b
P = 0.0001b
−7.06
−8.15
−13.6
−9.7
−10.0
Koca et al28 6 RCT
Follow-up at 1 and 3 mo
LWIs (6 mm = 5°), quadriceps strengthening and NSAIDs vs no insole, quadriceps strengthening, and NSAIDs. Both insoles worn all day Within group (3 mo):
LWIs
VAS Resting
VAS Walking
VAS Standing
WOMAC Pain
WOMAC Stiffness
Ritchie Index
Control group
VAS Resting
VAS Walking
VAS Standing
WOMAC Pain
WOMAC Stiffness
Ritchie Index
Between groups
VAS Resting
VAS Walking
VAS Standing
WOMAC Pain
WOMAC Stiffness
Ritchie Index
P = 0.037b
P = 0.006b
P = 0.024b
P = 0.003b
P = 0.031b
P = 0.011b
P = 0.564
P = 0.257
P = 0.527
P = 0.521
P = 0.903
P = 1.0
P = 0.03b
P = 0.01b
P = 0.017b
P = 0.002b
P = 0.061
P = 0.039b
−0.79
−1.42a
−0.63
−15.25a
−12.5a
−0.42
−0.05
−0.17
0.12
−0.85
0
0
NR
NR
NR
NR
NR
NR
−1.1
−1.1
−1.1
−9.7
−10.0

−1.1
−1.1
−1.1
−9.7
−10.0






Pham et al30 7 RCT
2 y
Full-length bilateral LWIs vs neutrally wedged insoles. Both made of pure rubber Ledos material with cork powder mounted on leather strip. High capacity to absorb impact load Within group:
LWIs
VAS
WOMAC Pain
WOMAC Stiffness
Control insoles
VAS
WOMAC Pain
WOMAC Stiffness
Between groups
VAS
WOMAC Pain
WOMAC Stiffness
Number of days NSAIDs taken
P > 0.05
P > 0.05
P > 0.05
P > 0.05
P > 0.05
P > 0.05
P = 0.38
P = 0.37
P = 0.25
P = 0.003b
−0.58
−2.6
−1.2
−0.47
−3.9
0.4
1.1
1.3
0.8
97 d
−1.1
−9.7
−10.0
−1.1
−9.7
−10.0



Toda et al33 5 Quasiexperimental
8 wk
Bilateral half-length LWIs (6.35-mm wedge height) with ankle sprain supporter (Sofra-Wolfer) and subtalar strapping vs LWI only (6.35-mm wedge height). Both made from urethane rubber. Worn 3-6 h/d Within group:
Strapped LWI
VAS
Inserted LWI
VAS
P = 0.041b
P > 0.05
−0.88
0.15
−1.1
−1.1
Toda and Tsukimura31 7 RCT
12 wk
5 groups:
1. Neutral wedged insole (placebo) in walking shoes of flat full-length urethane sheet
2. LWI (6.35 mm = 5°) full-length sponge rubber insole with nylon seat in walking shoes
3. Sock-type ankle supporter with half-length sponge rubber heel wedge (6.35 mm = 5°) sewn in; sock extended to metatarsals
4. Urethane half-length lateral wedge (12 mm = 11.2°) with ankle and subtalar strapping in “figure 8” pattern. Strap was polyester/polyurethane, 230% stretch rate, with “everyday walking shoes”
5. Urethane half-length lateral wedge (12 mm = 11.2°) with ankle and subtalar strapping as in 4, without shoes
All groups wore footwear 5-10 h/d
Within group:
1. VAS
2. VAS
3. VAS
4. VAS
5. VAS
P = 0.13
P = 0.41
P = 0.31
P = 0.004b
P = 0.002b
0.26
0.23
−0.35
−0.63
−0.87
−1.1
−1.1
−1.1
−1.1
−1.1
Wallace32 4 Quasiexperimental
12 wk
Bilateral semirigid three-quarter length LWIs (7°) vs neutral insole. Made with “medium (62 durometers) density ethyl vinyl acetate shell covered with 2-mm blown polyvinyl chloride foam. Wedging material made of 80 durometer rating ethyl vinyl acetate foam” Within group
LWI
VAS Walking
VAS Stair Descent
WOMAC Pain
WOMAC Stiffness
Control insole
VAS Walking
VAS Stair Descent
WOMAC Pain
WOMAC Stiffness
P < 0.05b
P < 0.05
P < 0.05b
P > 0.05
P > 0.05
P > 0.05
P < 0.05b
P > 0.05
−1.59a
−1.95a
−15.2a
−1.5
0.14
−0.4
−7.5
−1.4
−1.1
−1.1
−9.7
−10.0
−1.1
−1.1
−9.7
−10.0
Other types of footwear
Dufour et al36 N/A Prospective cohort study
6-y period
Comparison of 3 conditions:
• “Good”: athletic shoe, casual sneaker
• “Average/fair”: work boots, leather shoe, hard or rubber-soled shoes, cowboy boots
• “Poor”: heels/pumps, sandals, slippers
Past shoe wear (“good” vs “average/fair”)
Causes hindfoot pain (yes/no)
P = 0.022b
OR = 0.33 (95% CI = 0.13-0.85)
NR
Lane et al34 5 Quasiexperimental 3 types of soles:
• Soft-soled (25 durometers)
• Medium-soled (40 durometers)
• Hard-soled (58 durometers)
All shoes were extra-depth shoes with 27-mm sole, 13 mm under forefoot, flat insole (no arch support)
Between conditions:
VAS (forefoot pain)
Soft vs medium
Soft vs hard
Medium vs hard
P = 0.223 −0.33
−0.95
−0.62


Turpin et al35 3 Quasiexperimental
1 mo
Sulcus-length (metatarsal heads) shock-absorbing insoles made of “triple-density gel with a heel thickness of 8.35 mm and forefoot thickness of 4.31 mm” Within group:
LWIs
VAS Average
WOMAC Pain
P = 0.04b
P = 0.02b
−1.71a
−10.0a
−1.1
−9.7
Abbreviations: CI, confidence interval; LWIs, lateral wedge insoles; MCID, minimal clinically important difference; NR, not reported; NSAID, nonsteroidal anti-inflammatory drug; OR, odds ratio; PEDro, Physiotherapy Evidence Database; RCT, randomized controlled trial; RI, Ritchie Index; VAS, visual analog scale (0-10 points); WOMAC, Western Ontario and McMaster Universities Arthritis Index (reported on 0-100 VAS or % change).
aClinically significant difference (> MCID).
bStatistically significant difference (P> .05).

Table 3. - Effect of Footwear on Gait and Function
Author PEDro Score Study Design/Duration Type of Footwear/Intervention Outcomes Significance Mean Difference MCID
Lateral wedge insoles
Bennell et al26 8 RCT
12 mo
Full-length bilateral LWIs (5°) of “high-density ethyl vinyl acetate” vs full-length flat insoles of “easily compressible low-density ethyl vinyl acetate but no wedging.” Worn daily, 6.8-9.1 h/d Within group:
LWIs
WOMAC Function
PASE
Number daily steps
Control insoles
WOMAC Function
PASE
Number daily steps
Between groups:
WOMAC Function
PASE
Number daily steps
P > .05
P > .05
P > .05
P > .05
P > .05
P > .05
P > .05
P > .05
P > .05
−4.56
−16
14
−5.88
1
−673
−1.03
7.8
−797.6
−9.3


−9.3




Baker et al29 7 RCT crossover
6 wk
LWIs (5°) vs neutral flat “(1/8-inch-thick) insole on side of affected knee” only. Both made of noncompressible NickelPlast material. Worn 7.2-7.7 h/d Between groups:
WOMAC Function
50-foot walk-time (s)
Chair stand time (s)
P > .05
P > .05
P > .05
NR
NR
NR


Fang et al27 3 Quasiexperimental
4 wk
New Balance shoes (model 833) with full-length LWIs on symptomatic side(s), 4° vs neutral wedge on noninvolved side. Both constructed from “custom cork composite similar to Thermocork,” density 60 durometers (shock absorbing and resistant to deformation) Within group:
LWIs
WOMAC Function
P = .003a −7.16 −9.3
Koca et al28 6 RCT
Follow-up at 1 and 3 mo
LWIs (6 mm = 5°), quadriceps strengthening and NSAIDs vs no insole, quadriceps strengthening, and NSAIDs. Both insoles worn all day Within group (3 mo):
LWIs
WOMAC Function
WOMAC Total
Control insoles
WOMAC Function
WOMAC Total
Between groups
WOMAC Function
WOMAC Total
P = .001a
P = .001a
P = .146
P = .241
P = .004a
P = .012a
−9.58b
−11.01b
−3.43
−2.6
NR
NR
−9.3
−8.8
−9.3
−8.8

Pham et al30 7 RCT
2 y
Full-length bilateral LWIs vs neutrally wedged insoles. Both made of pure rubber Ledos material with cork powder mounted on leather strip. High capacity to absorb impact load Within group:
LWIs
WOMAC Function
Control insoles
WOMAC Function
Between groups
WOMAC Function
P > .05
P > .05
P = .25
1.2
0.4
0.8
−9.3
−9.3
Toda et al33 5 Quasiexperimental
8 wk
Bilateral half-length LWIs (6.35-mm wedge height) with ankle sprain supporter (Sofra-Wolfer) and subtalar strapping vs LWI only (6.35-mm wedge height). Both made from urethane rubber. Worn 3-6 h/d Within group:
Strapped LWI
Lequesne Index
Inserted LWI
Lequesne Index

P = .006a
P = .009a
−2.9
−1.3

Toda and Tsukimura31 7 RCT
12 wk
5 groups:
1. Neutral wedged insole (placebo) in walking shoes of flat full-length urethane sheet
2. LWI (6.35 mm = 5°) full-length sponge rubber insole with nylon seat in walking shoes
3. Sock-type ankle supporter with half-length sponge rubber heel wedge (6.35 mm = 5°) sewn in; sock extended to metatarsals
4. Urethane half-length lateral wedge (12 mm = 11.2°) with ankle and subtalar strapping in “figure 8” pattern. Strap was polyester/polyurethane, 230% stretch rate, with “everyday walking shoes”
5. Urethane half-length lateral wedge (12 mm = 11.2°) with ankle and subtalar strapping as in 4, without shoes
All groups wore footwear 5-10 h/d
1. Lequesne Index
2. Lequesne Index
3. Lequesne Index
4. Lequesne Index
5. Lequesne Index
P = .16
P = .20
P = .003a
P < .0001a
P < .0001a
−0.66
−0.44
−1.7
−3.0
−3.9




Wallace32 4 Quasiexperimental
12 wk
Bilateral semirigid three-quarter length LWIs (7°) vs neutral insole. Made with “medium (62 durometers) density ethyl vinyl acetate shell covered with 2-mm blown polyvinyl chloride foam. Wedging material made of 80 durometer rating ethyl vinyl acetate foam” Within group:
LWI
WOMAC Function
Stair descend time (s)
Control insole
WOMAC Function
Stair descend time (s)
P < .05a
P > .05
P < .05a
P > .05
−11.5b
−0.01
−9.71b
−0.01
−9.3

−9.3
Other types of footwear
Turpin et al35 3 Quasiexperimental
1 mo
Sulcus-length (metatarsal heads) shock-absorbing insoles made of “triple-density gel with a heel thickness of 8.35 mm and forefoot thickness of 4.31 mm” Within group:
LWIs
WOMAC Total
12 stair climbing time (sec)
P = .04a
P = .02a
−5.73
−0.42
−8.8
Abbreviations: LWIs, lateral wedge insoles; MCID, minimal clinically important difference; NSAID, nonsteroidal anti-inflammatory drug; NR, not reported; PASE, Physical Activity Scale for the Elderly (0-400 points); PEDro, Physiotherapy Evidence Database; RCT, randomized controlled trial; WOMAC, Western Ontario and McMaster Universities Arthritis Index (reported on 0-100 VAS or % change).
aStatistically significant difference (P> .05).
bClinically significant difference (< MCID).

Table 4. - Effects of Footwear on Quality of Life
Author PEDro Score Study Design/Duration Type of Footwear/Intervention Outcome Significance Mean Difference MCID
Bennell et al26 8 RCT
12 mo
Full-length bilateral LWIs (5°) of “high-density ethyl vinyl acetate” vs full-length flat insoles of “easily compressible low-density ethyl vinyl acetate but no wedging.” Worn 6.8-9.1 h/d AQoL: Within group
LWIs
Control
Between group
P > .05
P > .05
P > .05
−0.02
−0.01
−0.01


Abbreviations: AQoL, Assessment of Quality of Life Instrument (0-1.0 scale, where 1.0 is full health and 0 is death-equivalent health state); LWIs, lateral wedge insoles; MCID, minimal clinically important difference; RCT, randomized controlled trial.

Methodological Quality

PEDro scores varied between 3 and 8, with Bennell et al26 receiving the highest score of an 8. Fang et al27 and Turpin et al35 scored the lowest with a 3; a PEDro score could not be calculated for Dufour et al36 due to use of an observational design. PEDro scores for each study are reported in Table 1.

Study Characteristics

Participants

All articles included older adults with medial knee OA, or foot OA. Mean age of all participants was 65.1 years (range, 55.1-73.2 years). Inclusion criteria for knee OA studies included the American College of Rheumatology definition of OA, and graded stage of disease by the Kellgren/Lawrence (KL) system, which classifies knee OA via radiographs in 5 grades from 0 (no features of OA) to 4 (severe findings).37 Nine of the 11 studies included participants with medial knee OA with at least KL grade 2,26–33,35 with Baker et al29 reporting 93% of participants having KL grade 3 or 4. Two studies only had participants with grade 2 or 3.26,28 Average body mass index (BMI) was 28.4 (range, 25-33). Average femorotibial angle using radiography ranged from 178.8° to 181.6° where reported,26,31,33,35 for studies with medial knee OA diagnosis. These were all indicative of a varus abnormality in these studies. The frequency of women in participant groups for experimental and quasiexperimental studies was 73.8% on weighted average (range, 36%-100%).26–35

Protocols

Eight of the articles26–33 examined LWIs, but varied in angle (range, 4°-11.2°; mode = 5°) or height of the insole (6-6.35 mm), material structure of the insole, and full versus half-length. Of the 9 studies examining inserts, 3 reported use of full-length inserts,26,27,30 1 used three-quarter length,32 1 used sulcus length,35 1 used half-length heel wedges,33 and 1 used both full-length and half-length within the same study.31 Two insert studies did not report the length of the insole.28,29 Seven of the 8 studies used the participants' everyday walking shoes,26,28–33 with 1 requiring participants to wear New Balance® walking shoes for the duration of the investigation.27 All LWI studies used neutral or flat insoles in their control groups,26,27,29–33 except Koca et al28 who used no insert for the control group. Concomitant interventions included the following: 2 of the LWI studies added subtalar strapping to the LWIs,31,33 whereas Koca et al28 added exercise and nonsteroidal anti-inflammatory drugs (NSAIDs) to both treatment and control groups (exercise protocol not specified but stated quadriceps strengthening performed).

Materials of footwear in all studies varied between 25 and 80 durometers of hardness, where a lower number represents the softest material. Footwear and insert materials included high-density ethyl vinyl acetate,26,32 noncompressible NickelPlastTM,29 shock-absorbing cork similar to Thermocork®27 and rubber with cork,30 urethane rubber,31,33 or shock-absorbing triple-density gel.35 Study durations in randomized controlled trials and quasiexperimental designs varied from 4 weeks to 2 years. Wearing schedules of shoes varied from 3 to 10 hours per day. See Tables 2, 3, and 4 for details on intervention protocols and footwear materials.

Reported Outcomes

Pain

Visual analog scale (VAS) and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and use of pain medications. The 11 included articles showed heterogeneity of design and outcomes. However, all measured pain with footwear and mobility as an outcome. The VAS and WOMAC measured pain with OA. The VAS is reliable,38 valid,39 and responsive to change (minimal clinically important difference [MCID] = 11 points on a 100-point scale)40 in the population with arthritis. The WOMAC is a reliable, valid, responsive, and efficient self-report measure used to measure pain, stiffness, and physical function in individuals with OA of the hip or knee.41

Baker et al29 found no difference in WOMAC pain subscale (P = 0.13) with LWIs. Koca et al28 determined that use of LWIs resulted in the decreased VAS in standing (P = 0.046), and WOMAC pain score (P = .003) at 1-month follow-up compared with the control group. The Ritchie Index, a numerical measure of joint tenderness with passive range of motion,42 (P = .039), the VAS at rest (P = .03), the VAS with walking (P = .01), the VAS with standing (P = .017), and the WOMAC pain score (P = .002)—all decreased at the 3-month follow-up compared with the control group.28 Fang et al27 reported improved WOMAC scores for pain (P = .03), and pain with going up or down stairs (P < .001), but not with walking on a flat surface (P = .096), at night while in bed (P = .084), sitting or lying (P = .198), or standing upright (P = .87) with use of LWIs and New Balance® shoes.

Bennell et al26 compared the effect of LWIs to no intervention on pain over 12 months. Overall average pain or pain with walking (VAS) was not significantly different between groups.26 Pham et al30 investigated use of LWIs for a period of 2 years, resulting in a significant decrease in number of days NSAIDs were taken (P = .003). However, there was no significant change in WOMAC pain scores (P = .37) or number of days analgesics were taken (P = .40). Toda and Tsukimura31 evaluated the effects of LWIs with subtalar strapping. The LWIs with subtalar strapping combined with flat shoes (P = .004) or no shoes (P = .002), as opposed to heeled shoes, improved pain within subjects over the course of 12 weeks as compared with baseline.31 According to Toda et al,33 LWIs with subtalar strapping significantly improved the VAS pain score (P = .04) compared with baseline assessments. A neutral wedge without strapping with shoes (P = .13), LWIs with shoes (P = .41), and LWIs without shoes (P = .31)—all resulted in no significant changes in pain on the VAS from baseline.31 Finally, Wallace32 reported decreased pain using the VAS with stair descent in the treatment group, using LWIs in older adults with medial knee OA (P < .05), and WOMAC pain (P < .05) after 12 weeks of use.

Three studies did not investigate the effects of LWIs.34–36 Turpin et al35 reported that joint pain with walking due to medial knee OA decreased on the VAS with use of shock-absorbing insoles (P = .04); WOMAC pain scores decreased as well (P = .02). A long-term observational design followed up a cohort of older individuals and examined the relationship between types of footwear and likelihood of foot pain.36 Women are less likely to have hindfoot pain in the future with “good” (defined as athletic shoes or sneakers) versus “average” (defined as leather soled or dress) shoes (P = .022), when the data were adjusted for age and weight.36 Finally, Lane et al34 reported no difference between soft, medium, and hard-soled shoes for comfort with gait on a VAS (P = .223).

WOMAC Stiffness, Function, and Total Scores

Seven studies examined WOMAC subscales of stiffness, physical function, and/or total scores.26–30,32,35 Koca et al28 reported improved WOMAC total scores (P = .011) at 1-month follow-up, as well as improved WOMAC physical function scores (P = .004) and WOMAC total scores (P = .012) at 3-month follow-up, but not WOMAC stiffness at 1-month (P = .408) or 3-month follow-up (P = .061) compared with the control group. Fang et al27 found improved WOMAC stiffness (P = .05) and function scores (P = .003) with use of LWIs. Turpin et al35 showed improved WOMAC total scores with shock-absorbing insoles (P = .04). When comparing LWIs over a 12-month period, Bennell et al26 noted no difference in WOMAC stiffness or function. Baker et al29 reported no change in WOMAC disability scores in LWI versus control groups. Pham et al30 demonstrated no change in LWI versus control groups on WOMAC stiffness (P = .25) or function (P = .25), but Wallace32 did find improved WOMAC physical function scores with LWI versus control (P < .05) after 12 weeks of use.

Lequesne Index and Ritchie Index

The Lequesne Index (LI) is a measure of discomfort with OA, and includes domains of pain, walking distance, limitations in functional activity, and activities of daily living.43 The Ritchie Index (RI) measures pain with passive range of motion and is a reliable and sensitive tool for measuring pain in individuals with OA.44 Two studies measured outcomes using the LI31,33 and 1 used the RI.28 Koca28 measured improvement in the RI with LWIs, exercise, and NSAIDs versus no insole, exercise, and NSAIDs (exercise protocol was not stated). The RI score significantly improved in the treatment group (P = .039) compared with the control group, and within the LWI group as well (P = .011). Toda et al33 noted both subtalar strapping (P = .006) and LWI groups (P = .009) improved on the LI pre- to postintervention after 1 month of use. Toda and Tsukimura31 reported significant within-group changes on the LI for LWIs without shoes (P = .003), subtalar strapping when used with flat shoes (P < .0001), or strapped insole with no shoes (P < .0001) in older adults with medial knee OA compared with heeled shoes. However, in that same study, the neutral wedge with shoes had no effect (P = .16) nor did the LWI with shoes show a change in the LI (P = .20).31

Functional Mobility and Quality of Life

Four studies explored the effects of footwear on ability to perform functional tasks, including stair-climbing speed, walk time, chair stand time, distance walked, and quality-of-life participation scales.26,29,32,35 Turpin et al35 showed improvement in time to ascend 12 stairs with use of shock-absorbing insoles (P = .02). Bennell et al26 did not find significant differences in the Physical Activity Scale for the Elderly scores, or number of daily steps taken with use of LWIs. The Physical Activity Scale for the Elderly is a self-assessment measuring activity in the domains of leisure, household, and occupational activity in adults older than 65 years.45,46 Bennell et al26 also noted no difference between groups on the Assessment of Quality of Life Instrument (AQoL). Baker et al29 reported no difference in 50-ft walk time or chair stand time with use of LWIs in adults with medial knee OA compared with neutral insoles worn 7 to 8 hours per day.

Adverse Events

Five studies reported adverse effects associated with trials. Although rare, adverse events associated with use of LWIs included sweating, foot sole pain, back and foot pain, blisters on the tops of toes, popliteal pain, and “foot pain due to the foot being cramped inside their shoes.”26,29,31,33,35

DISCUSSION

OA of the lower extremity is highly prevalent in older adults and can lead to a decline in functional mobility and loss of independence.1 This systematic review sought to compare various footwear interventions that decreased lower extremity joint pain and improved functional mobility, gait, or quality of life in older adults with lower extremity OA. The heterogeneity, both in terms of design (participant characteristics, protocols, footwear materials) and outcomes measured, made comparison between studies difficult. The results of this review may assist clinicians in making evidence-based recommendations regarding footwear for individuals with lower extremity OA.

Effects on Pain and Comfort

Six27,28,31–33,35 of the 11 studies found at least 1 positive statistically significant outcome on pain, using the VAS or WOMAC. However, only 3 of those had both statistically and clinically meaningful improvement, when comparing mean differences to the MCID for the VAS (−1.1 on a 0- to 10-point scale)40 and WOMAC.28,32,35 The MCID for the WOMAC is −9.7 points or −18% change in pain, −10.0 points or −22% change in stiffness, all points on a 0- to 100-point VAS for the WOMAC (see Tables 2 and 3).47–49 Therefore, caution should be used when interpreting these results in terms of clinically meaningful improvement.

Fang et al27 performed an uncontrolled pilot study of 4 weeks' duration that did show significant improvement in knee pain using the VAS, and the WOMAC with the use of the same LWIs. However, there was no blinding of subjects or raters, or a control group that increased potential bias. Koca et al28 also reported improvements in pain (VAS) with walking, standing, and at rest, as well as improved RI at 3 months, and WOMAC pain scores when comparing the addition of LWIs used with quadriceps-strengthening exercise and analgesics, when compared with exercise and analgesic use only. However, Koca et al28 did not state the specific quadriceps-strengthening exercise protocol, which may have been useful, as quadriceps strength is known to provide shock absorption to the knee and mitigate pain due to OA. A smaller sample size (n = 37) and no masking of participants, therapists, or assessors may have contributed to potential bias. Bennell et al26 did allow participants to take analgesics or exercise. In the clinic, patients may be taking medication and exercising, giving this study increased external generalizability. Pham et al30 did find a significant decrease in days NSAIDs were taken with LWI use, which is clinically important due to the risk of gastrointestinal distress in older adults with chronic NSAID use.9

Five studies that included shock-absorbing qualities as a part of the footwear (such as use of cork, rubber, or gel) demonstrated significant reductions in pain on at least 1 outcome measure.27,30,31,33,35 Despite lower quality (PEDro score of 3) and a small sample size of 16 participants in the study by Turpin et al,35 this was the only study that measured the effect of strictly shock-absorbing footwear on pain for individuals with knee OA. Even though that study failed to show a biomechanical effect through a decrease in the KAM for all phases of gait, important reductions in pain were found. This is clinically important in that shock-absorbing material in both insoles and footwear appears to be of benefit and warrants both clinical recommendation and further investigation for effects on pain in older adults with OA.

Subtalar strapping is thought to cause both valgus angulation of the talus upon x-ray, and correction of the femorotibial angle, and it appears more powerful than LWIs alone.33 Subtalar strapping (in addition to LWIs) by Toda et al33 demonstrated significant differences in the strapping group with LWIs compared with a traditional insole. This type of bracing may improve pain associated with OA more effectively than a lateral wedge alone, although it did have an increased rate of adverse effects such as popliteal pain compared with other LWIs.31 Furthermore, Toda and Tsukimura31 reported that subtalar strapping with or without shoes gave the greatest pain relief in the knee by the VAS and LI. This method warrants further investigation on clinical outcomes from this method, which includes a more proximal structure.

The study on effects of shoe hardness revealed that there was no difference in comfort with harder versus softer soled shoes, including pain with walking.34 However, there was no blinding of assessors or randomization of subjects in this study. In addition, the subjects walked on an 8-m walkway, which may not have been enough distance to detect foot pain or comfort with walking when worn for longer periods or outdoor terrain.34 More high-quality research on the effects of shoe hardness on community-distance walking and lower extremity joint comfort is needed.

Finally, the effect of previous type of shoes worn on hindfoot pain with gait in older adults revealed that athletic shoes and sneakers worn while younger predicted less pain compared with previous wearing of heels, pumps, sandals, and slippers.36 This observational study had no blinding of participants and was the only design that was not an experimental or quasiexperimental design making a true cause-and-effect relationship difficult. However, it does suggest that the use of athletic shoes early in life in older adults can lead to a lower likelihood of developing hindfoot pain from OA.36 Other literature suggests that even modest (1.5-inch) heels can create a significant torque on the knee and may be relevant to the progression of medial knee OA.50 This is clinically relevant in suggesting minimal or no use of high heels will aid in decreasing or preventing joint pain associated with heel use.

Effects on Gait, Function, and Quality of Life

Inconclusive findings were present in studies that examined effects of footwear on gait, function, and quality of life. Two studies found mean differences from baseline that were greater than the MCID for the WOMAC function scale with use of LWIs.28,32 The MCID for the WOMAC is −9.3 points or −17% change in function, and −8.8 points or −18% for total score change.47–49 Turpin et al35 found statistically but not clinically meaningful improvements in the WOMAC total as well as stair climbing time with shock-absorbing insoles, when compared with the MCID. That study was of 1-month duration, which may not have been enough time to see clinically meaningful improvements. Pham et al30 did use a high load-bearing capacity cork material LWI, but found no clinically or statistically significant improvements in gait or function over 2 years. This study did have a significant dropout rate over 2 years (32%), which may have underpowered results. An intention-to-treat analysis was performed, possibly giving a more conservative estimate.30

Quality of life is an important clinical domain to measure in older adults with OA. Only Bennell et al26 used a quality-of-life instrument to measure changes using LWIs, and found no changes over 12 months within or between groups. However, the AQoL instrument measures multiple constructs such as mental health, relationship, and coping abilities in addition to limitation in functional activities due to pain.51 When the AQoL was evaluated for validity in populations with hip and knee disease, some issues were reported with “limited options for reporting pain,” as well as ceiling effects were found.52 Future research on footwear in older adults with OA would benefit from using both self-report and performance-based measures from the activity and participation domains of the International Classification of Function, Disability and Health, such as 36-Item Short Form Health Survey (SF-36) quality-of-life scale. The SF-36 has been shown to be a “psychometrically sound tool” in measuring changes in quality of life for adults with OA.53

Several factors may contribute to the inconclusive findings with the LWIs. There was a wide heterogeneity of participants, including factors such as BMI, wearing time each day, materials, degree of wedging, KL grade of disease severity, concomitant therapies, and sex ratio of participants. In addition, only 1 study required participants using LWIs to use a specific pair of shoes, such as the New Balance® shoes,27 whereas all other studies stated that participants wore their “everyday walking shoes.”26,28–33 Variability in choice of walking shoes in the research with LWIs may have been a confounding variable that had an effect on outcomes. Furthermore, the American College of Rheumatology recommends against discontinuing concomitant treatments while performing research on treatments for OA.54 Use of concomitant treatments such as medications during the study period may affect the outcome of such investigations.

The majority of participants in the studies reviewed were women, making the results harder to generalize to men, as anatomical structure of the knee differs between sexes, and may have affected outcomes. Variability in BMI may have affected outcomes as well. Average BMI was 25 in both of the studies by Toda et al,31,33 which is the high end of “normal” weight. BMI was highest (mean = 33) in the study by Baker et al.29 A subgroup analysis showed that those with a BMI of 30 or more had a lower improvement in pain compared with those with a BMI of less than 30.29 This could make generalization of findings difficult for individuals with higher BMIs, and results should be interpreted with caution.

All participants, where reported, had a varus deformity in the studies investigating effects of footwear on medial knee OA.26,31,33,35 This makes results more valid for individuals with varus deformities only, but not valgus. In addition, there may have also been individual differences in other unreported variables, such as “quadriceps strength or ligamentous laxity,”27 or amount of daily activity, which may have contributed to outcomes. Baker et al29 reported participants as having more severe grades of knee OA, with 93% K/L grade 3 or 4. A subgroup analysis showed that those with a K/L grade less than 4 had greater improvements in pain compared with those with a grade of 4.29 Other literature also suggests individuals with knee OA may be more responsive to LWIs in earlier stages (K/L grades 1 and 2).17 Finally, pain is a subjective construct, and highly variable over time.26 Bennell et al26 stated that taking more measurement time points and estimating average pain may be a more valid measure than just measuring pain at baseline and 12-month points, due to the fluctuating nature of pain.

Participants' adherence to wearing time of insoles varied and may have affected outcomes. Optimal wearing time for wedge insoles has been previously shown to be 5 to 10 hours.55 Bennell et al26 demonstrated a difference in wearing time of LWIs between the LWI (6.8 hours) and control groups (9.1 hours). This difference could have contributed to different outcomes in pain between groups. Clinically, adherence is highly variable due to individual tolerance and needs to be considered when prescribing an orthotic device such as an LWI or footwear. LWIs varied in wedging from 4°,27 5°,26,28,29,31,33 7°,32 and 11.2°31 making comparison difficult across groups. However, previous research has shown that greater wedging (10°) is more beneficial than lower degrees of wedging (5°) for reducing pain and KAM, although may be less tolerated.56

Variable length of the LWIs may also account for inconclusive findings. Previous research has shown that full-length insoles reduce the KAM more effectively than half-length, due to increased leverage through the metatarsal heads.57 Use of sulcus-length shock-absorbing insoles demonstrated effects on pain and function,35 but not biomechanical factors such as KAM for all phases of gait,35 although a different mechanism other than length may have contributed to the outcomes due to use of triple-density gel. Future research may be useful in combining full-length insoles with gel material. Three-quarter-length insoles were used by Wallace,32 with significant outcomes in pain and function. It is unclear whether sulcus or three-quarter-length insoles are as beneficial as full-length insoles in reducing pain; future research would benefit from investigation of sulcus-length versus full-length insoles on biomechanical and functional outcomes. Toda et al33 found that only the half-length wedge with subtalar strapping showed improvements in the femorotibial angle and talar tilt angle versus the half-length wedge alone. This confirms that half-length wedges may be inferior in correcting joint alignment in medial knee OA. Toda and Tsukimura31 also found that half-length wedges with ankle support had benefit for pain but did not compared with unstrapped half-length wedges. As discussed previously, subtalar strapping may have been the corrective intervention, rather than the wedge alone. Length of the LWI could not be determined in a few studies by review of the methodology.28,29 As a result, clear comparison among these studies is difficult. Future research would benefit from use of full-length insoles due to their superiority in reducing KAM57 and potentially pain.

Previous research shows improvement in radiographic or biomechanical findings such as KAM with the use of LWIs and other footwear by moving the center of pressure of the ground force laterally,58 or other biomechanical mechanisms; however, this may not always translate to improvement in clinical findings such as pain, stair climbing ability, and distance walked. Consideration of individual differences is critical when prescribing footwear or orthotics for conservative pain management. Chapman et al59 reported that individuals with increased ankle/subtalar joint complex inversion were more likely to be a nonresponder to LWIs, and had an increase in KAM rather than a decrease. Further high-quality research into other predictive factors or a clinical prediction rule for responsiveness to footwear and LWIs would be beneficial.

Experimental Design and Study Limitations

This systematic review examined 11 articles from 2001 to 2014, in English language, that used randomized control trials, quasiexperimental design, or observational cohort studies. Within randomized controlled trials, only 1 had a “true” control group, using no insole as a comparison.28 The neutral insole control groups may have shown an improvement in pain due to the material in the neutral, nonwedge insole by virtue of shock absorption alone.26,27,29–33 Future research in this area would benefit from a true control group to make an accurate comparison of difference in outcomes between groups. Several studies also had small sample size, which may have underpowered the results. Although most of the studies looked at interventions of footwear use and measured pain with a VAS, they used different outcomes, making comparison between studies difficult. Three studies32–34 out of the 11 had PEDro scores of 4 to 5 indicating “fair” quality, as well as 1 study being an observational design.36 One rated a 6,28 whereas 4 rated between 729–31 and 826 on the PEDro scale, which is considered “good.”60 Because of the variable range of study quality, results may need to be interpreted with caution. Finally, the reviewers may not have fully covered the full expanse of literature due to limitation of studies to English language only, incomplete retrieval of literature, or there may have been lack of publication of negative findings, limiting full coverage of all results.

Conclusions and Suggestions for Future Research

This systematic review of the literature shows that footwear may have an impact on lower extremity joint pain due to OA in older adults. Specifically, the use of shock-absorbing insoles, athletic footwear, avoidance of heels, and subtalar strapping appear effective in reducing lower extremity joint pain from OA, and reducing stair climbing time, whereas LWIs have inconclusive evidence of efficacy in improving functional mobility.27,28,31–33,35,36 Because of the potential benefits of being a low-cost and low-risk intervention, further research on the effects of footwear on mobility and gait in older adults with lower extremity OA is needed. Suggestions for further research include the relationship between types of footwear, including LWIs, subtalar strapping, and shock-absorbing insoles and function. In addition, duration of use and clinical outcomes such as gait distance, gait speed, chair stand time, steps taken per day, and self-report functional outcome measures should be investigated in future work.

ACKNOWLEDGMENTS

The authors thank Anthony Micchelli, MAT, MLIS for his assistance with searching the literature and managing references.

REFERENCES

1. Felson DT, Lawrence RC, Dieppe PA, et al. Osteoarthritis: new insights. Part 1: the disease and its risk factors. Ann Intern Med. 2000;133(8):635–646.
2. Barbour KE, Helmick CG, Theis KA, et al. Prevalence of doctor-diagnosed arthritis and arthritis-attributable activity limitation-united states, 2010-2012. Morb Mortal Wkly Rep. 2013;62(44):869–873.
3. Brummel-Smith K. Optimal aging, part I: demographics and definitions. Ann Longterm Care. 2007;15(11):26.
4. Centers for Disease Control and Prevention. Arthritis: Cost statistics. http://www.cdc.gov/arthritis/data_statistics/cost.htm. Updated August 2011. Accessed July 29, 2015.
5. Hafez AR, Al-Johani AH, Zakaria AR, et al. Treatment of knee osteoarthritis in relation to hamstring and quadriceps strength. J Phys Ther Sci. 2013;25(11):1401–1405.
6. Aaboe J, Bliddal H, Messier SP, Alkjaer T, Henriksen M. Effects of an intensive weight loss program on knee joint loading in obese adults with knee osteoarthritis. Osteoarthritis Cartilage. 2011;19(7):822–828.
7. Waller B, Ogonowska-Slodownik A, Vitor M, et al. Effect of therapeutic aquatic exercise on symptoms and function associated with lower limb osteoarthritis: systematic review with meta-analysis. Phys Ther. 2014;94(10):1383–1395.
8. Talmo CT, Aghazadeh M, Bono JV. Perioperative complications following total joint replacement. Clin Geriatr Med. 2012;28(3):471–487.
9. Guccione AA, Wong RA, Avers D. Geriatric Physical Therapy. 3rd ed. St. Louis, MO: Mosby; 2012:50–52.
10. Menant JC, Steele JR, Menz HB, Munro BJ, Lord SR. Effects of footwear features on balance and stepping in older people. Gerontology. 2008;54(1):18–23.
11. Arnadottir SA, Mercer VS. Effects of footwear on measurements of balance and gait in women between the ages of 65 and 93 years. Phys Ther. 2000;80(1):17–27.
12. Menant JC, Steele JR, Menz HB, Munro BJ, Lord SR. Effects of walking surfaces and footwear on temporo-spatial gait parameters in young and older people. Gait Posture. 2009;29(3):392–397.
13. Menant JC, Perry SD, Steele JR, Menz HB, Munro BJ, Lord SR. Effects of shoe characteristics on dynamic stability when walking on even and uneven surfaces in young and older people. Arch Phys Med Rehabil. 2008;89(10):1970–1976.
14. Koepsell TD, Wolf ME, Buchner DM, et al. Footwear style and risk of falls in older adults. J Am Geriatr Soc. 2004;52(9):1495–1501.
15. Roman de Mettelinge T, Calders P, Danneels E, Geeroms S, Du Four C, Cambier D. Does footwear matter when performing spatiotemporal gait analysis among older women? J Geriatr Phys Ther. 2015;38(4):155–161.
16. Shakoor N, Lidtke RH, Sengupta M, Fogg LF, Block JA. Effects of specialized footwear on joint loads in osteoarthritis of the knee. Arthritis Care Res. 2008;59(9):1214–1220.
17. Shimada S, Kobayashi S, Wada M, et al. Effects of disease severity on response to lateral wedged shoe insole for medial compartment knee osteoarthritis. Arch Phys Med Rehabil. 2006;87(11):1436–1441.
18. Kakihana W, Akai M, Nakazawa K, Takashima T, Naito K, Torii S. Effects of laterally wedged insoles on knee and subtalar joint moments. Arch Phys Med Rehabil. 2005;86(7):1465–1471.
19. Hinman RS, Bowles KA, Metcalf BB, Wrigley TV, Bennell KL. Lateral wedge insoles for medial knee osteoarthritis: effects on lower limb frontal plane biomechanics. Clin Biomech. 2012;27(1):27–33.
20. Kean CO, Bennell KL, Wrigley TV, Hinman RS. Modified walking shoes for knee osteoarthritis: mechanisms for reductions in the knee adduction moment. J Biomech. 2013;46(12):2060–2066.
21. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264–269.
22. de Morton NA. The PEDro scale is a valid measure of the methodological quality of clinical trials: a demographic study. Aust J Physiother. 2009;55(2):129–133.
23. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther. 2003;83(8):713–721.
24. Portney LG, Watkins MP. Foundations of Clinical Research: Applications to Practice. 3rd ed. Upper Saddle River, NJ: Prentice Hall; 2008:366–367.
25. Higgins J, Deeks JJ. Selecting studies and collecting data. In: Higgins JPT, Green S eds. Cochrane Handbook for Systematic Reviews of Interventions: Cochrane Book Series. The Cochrane Collaboration; 2008:151–185.
26. Bennell KL, Bowles KA, Payne C, et al. Lateral wedge insoles for medial knee osteoarthritis: 12-month randomised controlled trial. BMJ. 2011;342:d2912.
27. Fang MA, Taylor CE, Nouvong A, Masih S, Kao KC, Perell KL. Effects of footwear on medial compartment knee osteoarthritis. J Rehabil Res Dev. 2006;43(4):427–434.
28. Koca B, Oz B, Olmez N, Memis A. Effect of lateral-wedge shoe insoles on pain and function in patients with knee osteoarthritis. Türk Fiz Tıp Rehab Derg. 2009;55(4).
29. Baker K, Goggins J, Xie H, et al. A randomized crossover trial of a wedged insole for treatment of knee osteoarthritis. Arthritis Rheum. 2007;56(4):1198–1203.
30. Pham T, Maillefert J, Hudry C, et al. Laterally elevated wedged insoles in the treatment of medial knee osteoarthritis: a two-year prospective randomized controlled study. Osteoarthritis Cartilage. 2004;12(1):46–55.
31. Toda Y, Tsukimura N. Influence of concomitant heeled footwear when wearing a lateral wedged insole for medial compartment osteoarthritis of the knee. Osteoarthritis Cartilage. 2008;16(2):244–253.
32. Wallace DA. Efficacy of Lateral Heel Wedge Orthotics for the Treatment of Patients with Knee Osteoarthritis [dissertation]. Corvallis, OR: Oregon State University; 2006.
33. Toda Y, Segal N, Kato A, Yamamoto S, Irie M. Effect of a novel insole on the subtalar joint of patients with medial compartment osteoarthritis of the knee. J Rheumatol. 2001;28(12):2705–2710.
34. Lane TJ, Landorf KB, Bonanno DR, Raspovic A, Menz HB. Effects of shoe sole hardness on plantar pressure and comfort in older people with forefoot pain. Gait Posture. 2014;39(1):247–251.
35. Turpin KM, De Vincenzo A, Apps AM, et al. Biomechanical and clinical outcomes with shock-absorbing insoles in patients with knee osteoarthritis: immediate effects and changes after 1 month of wear. Arch Phys Med Rehabil. 2012;93(3):503–508.
36. Dufour AB, Broe KE, Nguyen US, et al. Foot pain: is current or past shoewear a factor? Arthritis Rheum. 2009;61(10):1352–1358.
37. Brandt KD, Fife RS, Braunstein EM, Katz B. Radiographic grading of the severity of knee osteoarthritis: relation of the Kellgren and Lawrence grade to a grade based on joint space narrowing, and correlation with arthroscopic evidence of articular cartilage degeneration. Arthritis Rheum. 1991;34(11):1381–1386.
38. Ferraz MB, Quaresma MR, Aquino LR, Atra E, Tugwell P, Goldsmith CH. Reliability of pain scales in the assessment of literate and illiterate patients with rheumatoid arthritis. J Rheumatol. 1990;17(8):1022–1024.
39. Downie WW, Leatham PA, Rhind VM, Wright V, Branco JA, Anderson JA. Studies with pain rating scales. Ann Rheum Dis. 1978;37(4):378–381.
40. Wolfe F, Michaud K. Assessment of pain in rheumatoid arthritis: minimal clinically significant difference, predictors, and the effect of anti-tumor necrosis factor therapy. J Rheumatol. 2007;34(8):1674–1683.
41. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation study of WOMAC: a health status instrument for measuring clinically important patient-relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol. 1988;15:1833–1840.
42. Ritchie DM, Boyle JA, McInnes JM, et al. Clinical studies with an articular index for the assessment of joint tenderness in patients with rheumatoid arthritis. Q J Med. 1968;37(147):393–406.
43. Lequesne M. Indices of severity and disease activity for osteoarthritis. Semin Arthritis Rheum. 1991;20(6):48–54.
44. Doyle DV, Dieppe PA, Scott J, Huskisson EC. An articular index for the assessment of osteoarthritis. Ann Rheum Dis. 1981;40(1):75–78.
45. Washburn RA, Smith KW, Jette AM, Janney CA. The physical activity scale for the elderly (PASE): development and evaluation. J Clin Epidemiol. 1993;46(2):153–162.
46. Washburn RA, McAuley E, Katula J, Mihalko SL, Boileau RA. The physical activity scale for the elderly (PASE): evidence for validity. J Clin Epidemiol. 1999;52(7):643–651.
47. Angst F, Aeschlimann A, Michel BA, Stucki G. Minimal clinically important rehabilitation effects in patients with osteoarthritis of the lower extremities. J Rheumatol. 2002;29(1):131–138.
48. Ehrich EW, Davies GM, Watson DJ, Bolognese JA, Seidenberg BC, Bellamy N. Minimal perceptible clinical improvement with the western Ontario and McMaster universities osteoarthritis index questionnaire and global assessments in patients with osteoarthritis. J Rheumatol. 2000;27(11):2635–2641.
49. Williams VJ, Piva SR, Irrgang JJ, Crossley C, Fitzgerald GK. Comparison of reliability and responsiveness of patient-reported clinical outcome measures in knee osteoarthritis rehabilitation. J Orthop Sports Phys Ther. 2012;42(8):716–723.
50. Kerrigan DC, Johansson JL, Bryant MG, Boxer JA, Della Croce U, Riley PO. Moderate-heeled shoes and knee joint torques relevant to the development and progression of knee osteoarthritis. Arch Phys Med Rehabil. 2005;86(5):871–875.
51. Hawthorne G, Richardson J, Osborne R. The assessment of quality of life (AQoL) instrument: a psychometric measure of health-related quality of life. Qual Life Res. 1999;8(3):209–224.
52. Ackerman IN, Busija L, Tacey MA, et al. Performance of the assessment of quality of life measure in people with hip and knee joint disease and implications for research and clinical use. Arthritis Care Res. 2014;66(3):481–488.
53. Kosinski M, Keller SD, Hatoum HT, Kong SX, Ware JE Jr. The SF-36 health survey as a generic outcome measure in clinical trials of patients with osteoarthritis and rheumatoid arthritis: tests of data quality, scaling assumptions and score reliability. Med Care. 1999;37(5):MS10–MS22.
54. Altman R, Brandt K, Hochberg M, et al. Design and conduct of clinical trials in patients with osteoarthritis: recommendations from a task force of the osteoarthritis research society: results from a workshop. Osteoarthritis Cartilage. 1996;4(4):217–243.
55. Toda Y, Tsukimura N, Segal N. An optimal duration of daily wear for an insole with subtalar strapping in patients with varus deformity osteoarthritis of the knee. Osteoarthritis Cartilage. 2005;13(4):353–360.
56. Kerrigan DC, Lelas JL, Goggins J, Merriman GJ, Kaplan RJ, Felson DT. Effectiveness of a lateral-wedge insole on knee varus torque in patients with knee osteoarthritis. Arch Phys Med Rehabil. 2002;83(7):889–893.
57. Hinman RS, Bowles KA, Payne C, Bennell KL. Effect of length on laterally-wedged insoles in knee osteoarthritis. Arthritis Care Res. 2008;59(1):144–147.
58. Sasaki T, Yasuda K. Clinical evaluation of the treatment of osteoarthritic knees using a newly designed wedged insole. Clin Orthop. 1987;221:181–187.
59. Chapman GJ, Parkes MJ, Forsythe L, Felson D, Jones R. Ankle motion influences the external knee adduction moment and may predict who will respond to lateral wedge insoles?: an ancillary analysis from the SILK trial. Osteoarthritis Cartilage. 2015;23(8):1316–1322.
60. Bhogal SK, Teasell RW, Foley NC, Speechley MR. The PEDro scale provides a more comprehensive measure of methodological quality than the Jadad scale in stroke rehabilitation literature. J Clin Epidemiol. 2005;58(7):668–673.

Appendix. Data Extraction Form

table5

Keywords:

footwear; function; osteoarthritis; pain

© 2018 Academy of Geriatric Physical Therapy, APTA.