Plantar heel pain / plantar fasciopathy · Follow-up analysis · 2025
How did heel pain change over a year for people in a shockwave trial?
In this follow-up of 200 adults from a heel-pain trial, pain during activity fell from about 6 to about 3 out of 10 over one year. Everyone had insoles and advice; some also had shockwave, sham shockwave or exercise, and this analysis looked at all the groups together.
Improved over time- adults with heel pain
- 200
- treatment groups, looked at together
- 4
- months of follow-up
- 12
Good to know: This analysis combined all four treatment groups, so it shows how people did overall.
Røe C, Heide M, Søberg HL, et al. Journal of Foot and Ankle Research, 2025 · Read the paper ↗
The treatment
Shockwave + insoles and advice
Radial
- Numbing
- Not mentioned
- Device
- Not named in the paper
- Everyone also got
- Advice and custom insoles
Compared with: Insoles and advice alone, or with sham shockwave or heavy, slow exercise
Settings are what each study used, not a prescription.
Who took part
200 adults aged 18 to 70 at a Norwegian specialist clinic; 76 had pain in both heels.
Side effects
This analysis did not report side effects.
All the numbers and the full appraisal
The results, in numbers.
200 adults aged 18–70 in a Norwegian specialist outpatient clinic; 160–182 attended each 3-, 6- and 12-month follow-up. 76 (38%) had bilateral heel pain.
The secondary trajectory analysis pools participants because parent-trial outcome trajectories did not differ by arm; per-arm counts are not given in this paper.
The secondary trajectory analysis pools participants because parent-trial outcome trajectories did not differ by arm; per-arm counts are not given in this paper.
The secondary trajectory analysis pools participants because parent-trial outcome trajectories did not differ by arm; per-arm counts are not given in this paper.
The secondary trajectory analysis pools participants because parent-trial outcome trajectories did not differ by arm; per-arm counts are not given in this paper.
Pain during activity (PainA): linear time trajectory3 reported outcomes
Pain during activity (PainA): linear time trajectory
Numeric Rating Scale (0–10) NRS points per model-coded time unit; time-unit coding not specified
Within-cohort linear time coefficient, pooled across treatment arms; not an ESWT treatment contrast
200 participants in this estimate
Linear mixed-effects model with reported covariates
The model reports a significant average trajectory over time. The coefficient is per model-coded time unit; the paper does not define that unit in the table.
Analysis details and source
Cohort-level longitudinal association; randomized intervention arms were pooled after no between-arm trajectory differences in the primary analysis. This coefficient is not an isolated treatment effect.
Pain during rest (PainR): linear time trajectory
Numeric Rating Scale (0–10) NRS points per model-coded time unit; time-unit coding not specified
Within-cohort linear time coefficient, pooled across treatment arms; not an ESWT treatment contrast
200 participants in this estimate
Linear mixed-effects model with reported covariates
The model reports a significant average trajectory over time. The coefficient is per model-coded time unit; the paper does not define that unit in the table.
Analysis details and source
Cohort-level longitudinal association; randomized intervention arms were pooled after no between-arm trajectory differences in the primary analysis. This coefficient is not an isolated treatment effect.
Pain-at-rest trajectory difference associated with bilateral versus unilateral pain
Numeric Rating Scale (0–10) NRS points per model time unit
Difference in modeled time trajectory: bilateral versus unilateral pain
200 participants in this estimate
Linear mixed-effects model
The interaction indicates a different modeled trajectory for bilateral versus unilateral pain; it is not a randomized treatment effect.
Analysis details and source
Coefficient per model-coded time unit; the time-unit coding is not specified in the table. The authors report that participants with bilateral pain improved in pain at rest and foot function up to 6 months but showed a lack of improvement from 6 to 12 months. Table 4's title calls PainR "pain during activity"; the Methods, Table 3, the Figure 3 caption and the Results text use PainR for pain at rest, as here.
Foot Function Index Revised Short Form: linear time trajectory1 reported outcome
Foot Function Index Revised Short Form: linear time trajectory
FFI-RS (0–100) FFI-RS points per model-coded time unit; time-unit coding not specified
Within-cohort linear time coefficient, pooled across treatment arms; not an ESWT treatment contrast
200 participants in this estimate
Linear mixed-effects model with reported covariates
The model reports a significant average trajectory over time. The coefficient is per model-coded time unit; the paper does not define that unit in the table.
Analysis details and source
Cohort-level longitudinal association; randomized intervention arms were pooled after no between-arm trajectory differences in the primary analysis. This coefficient is not an isolated treatment effect.
RAND-12 physical component score: linear time trajectory1 reported outcome
RAND-12 physical component score: linear time trajectory
RAND-12 PCS (0–100) score points per model-coded time unit; time-unit coding not specified
Within-cohort linear time coefficient, pooled across treatment arms; not an ESWT treatment contrast
200 participants in this estimate
Linear mixed-effects model with reported covariates
The model reports a significant average trajectory over time. The coefficient is per model-coded time unit; the paper does not define that unit in the table.
Analysis details and source
Cohort-level longitudinal association; randomized intervention arms were pooled after no between-arm trajectory differences in the primary analysis. This coefficient is not an isolated treatment effect.
RAND-12 mental component score: linear time trajectory1 reported outcome
RAND-12 mental component score: linear time trajectory
RAND-12 MCS (0–100) score points per model-coded time unit; time-unit coding not specified
Within-cohort linear time coefficient, pooled across treatment arms; not an ESWT treatment contrast
200 participants in this estimate
Linear mixed-effects model with reported covariates
The model reports a significant average trajectory over time. The coefficient is per model-coded time unit; the paper does not define that unit in the table.
Analysis details and source
Cohort-level longitudinal association; randomized intervention arms were pooled after no between-arm trajectory differences in the primary analysis. This coefficient is not an isolated treatment effect.
FFI-RS trajectory difference associated with bilateral versus unilateral pain1 reported outcome
FFI-RS trajectory difference associated with bilateral versus unilateral pain
Foot Function Index Revised Short Form (0–100) FFI-RS points per model time unit
Difference in modeled time trajectory: bilateral versus unilateral pain
200 participants in this estimate
Linear mixed-effects model
The interaction indicates a different modeled trajectory for bilateral versus unilateral pain; it is not a randomized treatment effect.
Analysis details and source
Coefficient per model-coded time unit; the time-unit coding is not specified in the table. The authors report that participants with bilateral pain improved in pain at rest and foot function up to 6 months but showed a lack of improvement from 6 to 12 months. The Results text says the final FFI-RS model included "the quadratic time interaction of bilateral pain", but Table 4 prints only a "Time × bilateral pain" row; the paper does not say whether this coefficient is the linear or the quadratic term.
Pain during activity4 reported outcomes
Pain during activity — cohort mean at baseline
PainA NRS points
Cohort-level descriptive mean 6.31 (SD 1.91); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain during activity — cohort mean at 3 months
PainA NRS points
Cohort-level descriptive mean 4.32 (SD 2.55); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain during activity — cohort mean at 6 months
PainA NRS points
Cohort-level descriptive mean 3.56 (SD 2.63); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain during activity — cohort mean at 12 months
PainA NRS points
Cohort-level descriptive mean 2.81 (SD 2.59); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain at rest4 reported outcomes
Pain at rest — cohort mean at baseline
PainR NRS points
Cohort-level descriptive mean 3.74 (SD 2.41); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain at rest — cohort mean at 3 months
PainR NRS points
Cohort-level descriptive mean 2.79 (SD 2.38); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain at rest — cohort mean at 6 months
PainR NRS points
Cohort-level descriptive mean 2.30 (SD 2.34); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Pain at rest — cohort mean at 12 months
PainR NRS points
Cohort-level descriptive mean 1.86 (SD 2.40); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Foot Function Index Revised Short Form4 reported outcomes
Foot Function Index Revised Short Form — cohort mean at baseline
FFI-RS FFI-RS points
Cohort-level descriptive mean 45.36 (SD 18.06); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Foot Function Index Revised Short Form — cohort mean at 3 months
FFI-RS FFI-RS points
Cohort-level descriptive mean 32.38 (SD 22.27); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Foot Function Index Revised Short Form — cohort mean at 6 months
FFI-RS FFI-RS points
Cohort-level descriptive mean 26.64 (SD 22.31); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
Foot Function Index Revised Short Form — cohort mean at 12 months
FFI-RS FFI-RS points
Cohort-level descriptive mean 21.84 (SD 21.75); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 physical component summary4 reported outcomes
RAND-12 physical component summary — cohort mean at baseline
PCS12 RAND-12 points
Cohort-level descriptive mean 39.30 (SD 9.45); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 physical component summary — cohort mean at 3 months
PCS12 RAND-12 points
Cohort-level descriptive mean 42.85 (SD 11.00); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 physical component summary — cohort mean at 6 months
PCS12 RAND-12 points
Cohort-level descriptive mean 45.88 (SD 11.27); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 physical component summary — cohort mean at 12 months
PCS12 RAND-12 points
Cohort-level descriptive mean 47.20 (SD 10.88); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 mental component summary4 reported outcomes
RAND-12 mental component summary — cohort mean at baseline
MCS12 RAND-12 points
Cohort-level descriptive mean 42.29 (SD 10.71); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 mental component summary — cohort mean at 3 months
MCS12 RAND-12 points
Cohort-level descriptive mean 43.89 (SD 12.13); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 mental component summary — cohort mean at 6 months
MCS12 RAND-12 points
Cohort-level descriptive mean 46.12 (SD 12.25); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
RAND-12 mental component summary — cohort mean at 12 months
MCS12 RAND-12 points
Cohort-level descriptive mean 47.09 (SD 12.09); this is not assigned to an intervention arm and is not an ESWT treatment effect.
Analysis details and source
Table 2 observed cohort mean (SD), distinct from the mixed-model coefficient in Tables 3–4. No arm-level estimate or outcome-specific n is reported in this row.
How treatment was delivered
Standardized oral and written advice plus customized semi-rigid foot orthosis for all participants.
Orthosis/advice alone, or with radial ESWT, sham radial ESWT, or heavy-slow exercises. This article refers to the parent protocol for detailed settings and does not report device parameters or session schedule.
Adverse events are not reported in this secondary trajectory article.
The authors note that participants in an RCT may not represent the broader plantar-fasciopathy population.
Twenty percent did not attend the 3-month visit; under 10% were lost at one year. Model-based missing-data handling may be biased if missingness was not random.
The bilateral-pain trajectory findings require replication.
Treatment arms are combined for this secondary analysis; it does not isolate ESWT efficacy.
Detailed device, dose and session parameters are in the cited parent protocol, which was not part of this article source and was not used to populate this package.
Findings checked against the full source by a separate automated reviewer.
Limitations noted by the reviewers
- The authors note that participants in an RCT may not represent the broader plantar-fasciopathy population.
- Twenty percent did not attend the 3-month visit; under 10% were lost at one year. Model-based missing-data handling may be biased if missingness was not random.
- The bilateral-pain trajectory findings require replication.
- Treatment arms are combined for this secondary analysis; it does not isolate ESWT efficacy.
- Detailed device, dose and session parameters are in the cited parent protocol, which was not part of this article source and was not used to populate this package.
Source-linked critical reading
Secondary trajectory analysis of a randomized trial
Technical source record
Full text read: all 17 sections, Tables 1–4 and footnotes, Figures 1–3 captions, and disclosure/data statements. Raw XML has four table-wraps, matching parsed tables. Parent-trial protocol is cited but not included in this article source; it was not used to fill device settings or treatment details.
Many participants improved in pain, foot function and quality of life during a year of care, but this study cannot tell whether shockwaves caused that improvement. In people with pain in both feet, rest pain and foot function improved up to month 6 but did not improve further between months 6 and 12.
How do pain, foot function and health-related quality of life change over one year in patients with plantar fasciopathy, and do recovery trajectories differ by bilateral pain or other patient characteristics?
Secondary longitudinal analysis of a sham-controlled, observer-blinded and partly patient-blinded four-arm randomized trial. The parent trial is registered as NCT03472989. This article models recovery across the combined cohort after finding no trajectory differences between intervention arms in the primary analysis.
200 adults aged 18–70 years were enrolled from a specialized Oslo University Hospital outpatient clinic. Eligibility required heel pain for over three months, localized plantar fascia insertion tenderness and activity pain NRS ≥3. 160–182 participants attended each follow-up (3, 6 and 12 months); 76 (38%) had bilateral pain.
All participants received standardized advice and customized semi-rigid foot orthoses. They were randomized to orthoses/advice alone or with radial ESWT, sham radial ESWT, or heavy-slow exercises. This trajectory paper refers readers to the parent protocol and primary analysis for detailed intervention delivery; device settings and session parameters are not reported here.
Four randomized groups are named, but this secondary analysis combines participants because the primary mixed models found no differences among intervention-arm trajectories. The article does not present a new pairwise treatment comparison.
Pain during activity and rest used an 11-point NRS (0–10). Foot function used FFI-RS (0–100; lower is better). Health-related quality of life used RAND-12 physical and mental component scores (0–100; higher is better). Assessments were baseline and 3, 6 and 12 months. Linear mixed models used maximum-likelihood handling of missing values.
Across the cohort, mean activity pain fell from 6.31 (SD 1.91) at baseline to 2.81 (2.59) at 12 months; rest pain 3.74 (2.41) to 1.86 (2.40); FFI-RS 45.36 (18.06) to 21.84 (21.75); PCS12 39.30 (9.45) to 47.20 (10.88); MCS12 42.29 (10.71) to 47.09 (12.09). All changes over time were reported p<0.001. The source reports within-cohort effect sizes of 1.56, 0.78, 1.18, 0.78 and 0.41, respectively; these are not between-group treatment effects. Bilateral pain was associated with a different trajectory in pain at rest and foot function; the bilateral group showed a lack of improvement in these outcomes from 6 to 12 months.
The article does not report adverse-event outcomes for this secondary trajectory analysis. This is not evidence of no harms in the parent trial.
Authors note that a selected RCT sample may not represent usual clinical populations, 20% did not attend the three-month follow-up, follow-up loss at one year was under 10%, and model-based missing-data handling may be biased if missingness was not random. Trajectory findings for bilateral pain require replication. Treatment trajectories were pooled after no differences between parent trial arms were found; this paper is not an ESWT-versus-control efficacy analysis.
The results describe recovery in a selected outpatient cohort receiving advice and orthoses, with randomized add-on treatment varying. They do not isolate ESWT effects. Bilateral symptoms were associated with slightly less improvement in rest pain and function and a lack of improvement from 6 to 12 months, which may inform follow-up planning but is not a validated individual prognosis.
Authors call for replication of bilateral-pain trajectories, investigation of bilateral pain causes and evaluation of treatment strategies for this subgroup.
Read the original article.
One-Year Trajectory of Pain, Function, and Health-Related Quality of Life in Patients With Plantar Fasciopathy.
Røe C, Heide M, Søberg HL, Brunborg C, Hoksrud AF, Myhre K, Brox JI, Mørk M. · Journal of foot and ankle research · 2025
Open the publication record or available full-text files.
Checked against the full paper · Clinical review: Dr. Marcus Yu Bin Pai, · How we review →