ObjectiveTo evaluate the comparative effectiveness of current treatment options for plantar heel pain (PHP).DesignSystematic review and network meta-analysis (NMA).Data sourcesMedline, EMBASE, CINAHL, AMED, PEDro, Cochrane Database, Web of Science and WHO Clinical Trials Platform were searched from their inception until January 2018.Study selectionRandomised controlled trials (RCTs) of adults with PHP investigating common treatments (ie, corticosteroid injection, non-steroidal anti-inflammatory drugs, therapeutic exercise, orthoses and/or extracorporeal shockwave therapy (ESWT)) compared with each other or a no treatment, placebo/sham control.Data extraction and analysisData were extracted and checked for accuracy and completeness by pairs of reviewers. Primary outcomes were pain and function. Comparative treatment effects were analysed by random effects NMA in the short term, medium term and long term. Relative ranking of treatments was assessed by surface under the cumulative ranking probabilities (0–100 scale).ResultsThirty-one RCTs (total n=2450 patients) were included. There was no evidence of inconsistency detected between direct and indirect treatment comparisons in the networks, but sparse data led to frequently wide CIs. Available evidence does not suggest that any of the commonly used treatments for the management of PHP are better than any other, although corticosteroid injections, alone or in combination with exercise, and ESWT were ranked most likely to be effective for the management of short-term, medium-term and long-term pain or function; placebo/sham/control appeared least likely to be effective; and exercise appeared to only be beneficial for long-term pain or function.ConclusionsCurrent evidence is equivocal regarding which treatment is the most effective for the management of PHP. Given limited understanding of long-term effects, there is need for large, methodologically robust multicentre RCTs investigating and directly comparing commonly used treatments for the management of PHP.PROSPERO registration numberCRD42016046963.

Background Although exercise is a core treatment for people with knee osteoarthritis (OA), it is currently unknown whether those with additional comorbidities respond differently to exercise than those without. We explored whether comorbidities predict pain and function following an exercise intervention in people with knee OA, and whether they moderate response to: exercise versus no exercise; and enhanced exercise versus usual exercise‐based care. Methods We undertook analyses of existing data from three randomized controlled trials (RCTs): TOPIK (n = 217), APEX (n = 352) and Benefits of Effective Exercise for knee Pain (BEEP) (n = 514). All three RCTs included: adults with knee pain attributable to OA; physiotherapy‐led exercise; data on six comorbidities (overweight/obesity, pain elsewhere, anxiety/depression, cardiac problems, diabetes mellitus and respiratory conditions); the outcomes of interest (six‐month Western Ontario and McMaster Universities Arthritis Index knee pain and function). Adjusted mixed models were fitted where data was available; otherwise linear regression models were used. Results Obesity compared with underweight/normal body mass index was significantly associated with knee pain following exercise, as was the presence compared with absence of anxiety/depression. The presence of cardiac problems was significantly associated with the effect of enhanced versus usual exercise‐based care for knee function, indicating that enhanced exercise may be less effective for improving knee function in people with cardiac problems. Associations for all other potential prognostic factors and moderators were weak and not statistically significant. Conclusions Obesity and anxiety/depression predicted pain and function outcomes in people offered an exercise intervention, but only the presence of cardiac problems might moderate the effect of exercise for knee OA. Further confirmatory investigations are required.

One‐stage individual participant data meta‐analysis models should account for within‐trial clustering, but it is currently debated how to do this. For continuous outcomes modeled using a linear regression framework, two competing approaches are a stratified intercept or a random intercept. The stratified approach involves estimating a separate intercept term for each trial, whereas the random intercept approach assumes that trial intercepts are drawn from a normal distribution. Here, through an extensive simulation study for continuous outcomes, we evaluate the impact of using the stratified and random intercept approaches on statistical properties of the summary treatment effect estimate. Further aims are to compare (i) competing estimation options for the one‐stage models, including maximum likelihood and restricted maximum likelihood, and (ii) competing options for deriving confidence intervals (CI) for the summary treatment effect, including the standard normal‐based 95% CI, and more conservative approaches of Kenward‐Roger and Satterthwaite, which inflate CIs to account for uncertainty in variance estimates. The findings reveal that, for an individual participant data meta‐analysis of randomized trials with a 1:1 treatment:control allocation ratio and heterogeneity in the treatment effect, (i) bias and coverage of the summary treatment effect estimate are very similar when using stratified or random intercept models with restricted maximum likelihood, and thus either approach could be taken in practice, (ii) CIs are generally best derived using either a Kenward‐Roger or Satterthwaite correction, although occasionally overly conservative, and (iii) if maximum likelihood is required, a random intercept performs better than a stratified intercept model. An illustrative example is provided.

A one‐stage individual participant data (IPD) meta‐analysis synthesizes IPD from multiple studies using a general or generalized linear mixed model. This produces summary results (eg, about treatment effect) in a single step, whilst accounting for clustering of participants within studies (via a stratified study intercept, or random study intercepts) and between‐study heterogeneity (via random treatment effects). We use simulation to evaluate the performance of restricted maximum likelihood (REML) and maximum likelihood (ML) estimation of one‐stage IPD meta‐analysis models for synthesizing randomized trials with continuous or binary outcomes. Three key findings are identified. First, for ML or REML estimation of stratified intercept or random intercepts models, a t‐distribution based approach generally improves coverage of confidence intervals for the summary treatment effect, compared with a z‐based approach. Second, when using ML estimation of a one‐stage model with a stratified intercept, the treatment variable should be coded using “study‐specific centering” (ie, 1/0 minus the study‐specific proportion of participants in the treatment group), as this reduces the bias in the between‐study variance estimate (compared with 1/0 and other coding options). Third, REML estimation reduces downward bias in between‐study variance estimates compared with ML estimation, and does not depend on the treatment variable coding; for binary outcomes, this requires REML estimation of the pseudo‐likelihood, although this may not be stable in some situations (eg, when data are sparse). Two applied examples are used to illustrate the findings.