Effectiveness of Different Types of Foot Orthoses for the Treatment of Plantar Fasciitis

Abstract

Clinicians often use foot orthoses to manage the symptoms of plantar fasciitis. Although there has been considerable research evaluating the effectiveness of orthoses for this condition, there is still a lack of scientific evidence that is of suitable quality to fully inform clinical practice. Randomized controlled trials are recognized as the “gold standard” when evaluating the effectiveness of treatments. We discuss why randomized controlled trials are so important, the features of a well-conducted randomized controlled trial, and some of the problems that arise when trial design is not sound. We then evaluate the available evidence for the use of foot orthoses, with particular focus on published randomized controlled trials. From the evidence to date, it seems that foot orthoses do have a role in the management of plantar fasciitis and that prefabricated orthoses are a worthwhile initial management strategy. At this time, however, it is not possible to recommend either prefabricated or customized orthoses as being better, and it cannot be inferred that customized orthoses are more effective over time and therefore have a cost advantage. Additional good-quality randomized controlled trials are needed to answer these questions.

Plantar fasciitis is one of the most common foot complaints, affecting approximately one in ten people at some time in their lives.[1] Plantar fasciitis is particularly prevalent in runners and people who are overweight[2–7]; however, it is also consistently reported in people with systemic inflammatory arthritis.[8–15] A recent study[16] found that there are approximately 1 million patient visits per year to office-based physicians and hospital outpatient departments each year in the United States for plantar fasciitis. This research did not, however, take into account podiatric physicians, so the true figure would undoubtedly be much higher. Although few other good-quality epidemiologic data are available, it has also been reported that:

• Fifteen percent (15%) of adult foot complaints requiring professional care in the United States are for plantar heel pain[17];
• Plantar fasciitis makes up one-quarter of all foot injuries[18] and up to 8% of overall injuries to runners and other athletes[19–21]; and
• One US orthotic laboratory found that one in every eight orthotic prescriptions were associated with heel spur–related symptoms.[22]

These figures highlight the prevalence of this condition.

Although there are many treatments used for plantar fasciitis, conservative management is usually very successful.[7,23] One conservative modality that is often used and advocated is foot orthoses. Consequently, there have been many articles published over the past 30 years discussing the effectiveness of foot orthoses for this condition. The vast majority suggest that orthoses are highly effective in reducing the symptoms.

Some of the earlier studies, although not specific to plantar fasciitis, retrospectively evaluated patients’ satisfaction or self-reported symptom improvement using a questionnaire.[24–27] The main findings from these studies were that self-reported symptom relief ranged from 70% to 94% and patient satisfaction from 83% to 91%. However, none of these studies controlled for the use of other treatments (eg, ice, physical therapy) or compared orthoses to no treatment. These are potential confounders because other treatments could have caused the result reported by respondents, or the condition may have improved over time even without treatment. Furthermore, patient satisfaction often indicates satisfaction with service quality rather than treatment outcome.[28,29] Finally, assessment of symptoms or satisfaction is not broad enough to expose the full impact of the condition on a person. For example, a respondent could have no symptoms because they have limited their activities—often ones they enjoy—rather than because of a direct pain-relieving effect of orthoses. As a result of these methodologic flaws, evidence from these studies is considered to be weak.

The majority of other literature on the effects of orthoses is also generally of poor quality, being either personal opinion[22,30–34] or case series.[7,35–44] Case series have concluded that foot orthoses are highly effective; however, none of these studies satisfy the requirements for good evidence.[45–48]

The purpose of this article, therefore, is to evaluate the effectiveness of different types of foot orthoses for the treatment of plantar fasciitis by applying currently accepted evidence-based principles[45–48] and by focusing on randomized clinical trials published to date. First, however, a brief discussion of some of the more important features of randomized trials is warranted.

Applying Evidence-Based Principles to Clinical Trials

In any clinical evaluation, it is essential that research methods ensure that the effect of a treatment on patients is directly attributable to that treatment and not to other causes. For example, a common problem in clinical trials is that researchers may unknowingly influence the outcome of the study. This influence may arise from how researchers deal with allocating patients to treatment groups (randomization), whether researchers or patients are aware of which treatment is being administered (blinding), or whether patient dropouts are ignored or dealt with inappropriately.

To avoid these problems, medical science has moved toward a structured approach in which the best evidence available is used to evaluate clinical questions.[48] Thus the randomized controlled trial (RCT) is widely recognized as the gold standard for evaluating treatments.[49–53] The RCT offers the highest level of evidence when clinicians want to assess the effectiveness of treatments for a given condition.

Table 1. Benefits and Limitations of Randomized Controlled Trials. 

lists some of the benefits and limitations of RCTs.

Although RCTs are considered the gold standard in clinical research, they are not appropriate in certain circumstances. Also, if conducted poorly, RCTs provide results that could be distorted or could require cautious interpretation. To appreciate the power of RCTs, particularly as they relate to research on orthoses and plantar fasciitis, it is important to explore these issues further.

As discussed previously, RCTs are the best method to date to protect clinical trials from bias.[54] Nevertheless, they are still open to abuse if certain methodologic issues are not addressed appropriately. Some of the more important issues include the following:

• The primary outcome measures should be fully validated and have been found to be reliable.[55]
• Researchers should conduct a prospective sample size calculation to decrease the chances of a type II error occurring. (A type II error occurs when the null hypothesis is accepted incorrectly, with the conclusion that no statistically significant difference exists between treatment groups when in fact one does exist. [55] This difference depends largely on sample size. A small sample size is less likely to detect a statistically significant difference. A large sample size is more likely to detect a statistically significant difference; however, this difference may not be clinically relevant. A prospective sample size calculation should be performed to ensure that a statistically significant result is detected when a clinically meaningful difference is present.)
• Appropriate outcome measures should be used,[56] including those that measure patient-based health status, as they offer a broader investigation of the effect of an intervention compared with surrogate outcome measures.[57]
• The randomization order is best determined by a computer, and the allocation sequence must be concealed from clinicians until the last possible moment (ie, clinicians and participants should be blinded to treatment allocation).[58,59] This prevents clinicians from gaining previous knowledge of the allocation sequence and consequently being in a position to bias recruitment. Although blinding prevents researchers and participants from introducing bias, it is difficult in a trial with foot orthoses to blind researchers once participants have received their orthoses.[60]
• The dropout rate must be reported and should be kept to a minimum; anything greater than 15% should be viewed with caution.[61] Excessive dropout may lead to distortion of results, particularly if more dropouts occur in one group.
• An intention-to-treat analysis should be conducted as the primary analysis. With this type of analysis, outcome measures are obtained regardless of compliance with the trial protocol, and data from all participants are analyzed according to allocation even if the participants had adverse events or unexpected outcomes.[62–65] Intention-to-treat analysis maintains the balance of confounders (reducing variability between groups)[64] and provides a more pragmatic estimate of the benefit of a treatment.[66]
• If a pragmatic trial is planned, then interventions, clinicians, and study protocols should represent common practice as much as possible to ensure external validity and generalizability.[67]

By addressing each of these issues, researchers conducting RCTs minimize bias and make the results as valid as possible. Clearly, each point must be considered before the trial begins. With these points in mind, a detailed evaluation of published RCTs evaluating plantar fasciitis and its treatment with foot orthoses is now presented.

RCTs Evaluating Foot Orthoses and Plantar Fasciitis

There have been six published RCTs to date evaluating foot orthoses for the treatment of plantar fasciitis.[68–73] A detailed evaluation of these trials is presented below. First, though, it is worth highlighting that the first of these trials[68] was published in 1997, so use of this methodology to evaluate orthoses for this condition is still relatively new. In this context, it is important to note that all suffer from certain methodologic flaws. Therefore, their findings need to be discussed alongside their strengths and weaknesses.

Two of the six trials have evaluated the effect of magnetic insoles on plantar heel pain, and these will be discussed first. Caselli et al[68] evaluated the effect of relatively soft insoles with and without magnetic foil in 34 participants with heel pain. Participants were assessed using the Foot Function Index before and after 4 weeks of treatment. No significant difference was found between the two groups—the overall mean score (0, best; 100, worst) decreased from 34 to 31 for the magnetic foil insole and from 29 to 28 for the standard insole. Therefore, it was concluded that the magnetic foil offered no advantage over the plain insole. This study did not follow key recommendations for conducting rigorous RCTs and was generally of poor quality.

A recent, more rigorous trial by Winemiller et al[73] also evaluated the effect of cushioning insoles with and without static bipolar magnets in 101 participants with plantar heel pain. No significant differences were found between the two groups after 4 and 8 weeks of treatment. The overall mean visual analog pain scale scores decreased from 6.9 to 3.9 for the nonmagnetic insole and from 6.7 to 3.9 for the magnetic insole at 8 weeks. As in the trial by Caselli et al,[68] it was concluded that magnets do not provide additional benefits compared with nonmagnetic insoles for the treatment of plantar heel pain. Therefore, there is no evidence to support the use of magnets for plantar fasciitis.

The other four trials focused on comparing customized foot orthoses with prefabricated or over-the-counter devices. Both types of devices are commonly used in practice by podiatrists in Australia and New Zealand[74] as well as in other countries. The comparison of customized and prefabricated foot orthoses has arisen primarily because of the difference in cost between them. A substantial amount of money is spent each year on foot orthoses—one United Kingdom study in 1994[75] reported that the annual budget for orthoses in the National Health Service was UK£12 million. The issue of cost-effectiveness of foot orthoses is further highlighted by a number of surveys and audits conducted in the United Kingdom in the past 10 years.[39,75–77] Clearly, cost is an important issue, but equally important is whether customized or prefabricated devices are more effective. The following randomized trials have attempted to evaluate the effectiveness of these devices.

Lynch et al[69] investigated the effect of foot orthoses in a three-arm trial involving 85 participants. The three interventions were 1) a mechanical treatment consisting of low-Dye taping and functional foot orthoses, 2) an anti-inflammatory treatment consisting of a corticosteroid injection followed by nonsteroidal anti-inflammatory drugs, and 3) an accommodative viscoelastic heel cup. After 3 months, the mechanical treatment was found to provide better relief of symptoms and fewer dropouts than the other groups. Visual analog pain scores improved 44 mm for the mechanical group, 34 mm for the anti-inflammatory group, and 22 mm for the accommodative group. Nevertheless, this trial evaluated quite different treatment regimens, and the functional foot orthoses group had the added short-term treatment of taping.

Concentrating more on the effect of orthoses alone, Turlik et al[72] evaluated functional foot orthoses against another type of generic heel pad in 55 participants. They found after approximately 3 months that functional foot orthoses were associated with better outcomes on all measures. However, the researchers used outcome measures that did not have demonstrated validity or reliability. Further, certain methodologic issues (eg, allocation concealment) that could have led to bias were not addressed or reported on in the above studies, making the findings questionable.

In a larger, multicenter trial, Pfeffer et al[71] evaluated five treatments—a silicone insert, a rubber insert, a felt insert, a customized orthosis, and stretching only (all orthosis groups also received stretching)—in 200 participants. After 8 weeks, they found that the prefabricated (over-the-counter) groups had significantly better outcomes than the customized group and the stretching-only group. Pain scores (0, best; 100, worst) from the Foot Function Index improved 22.9 points for all the prefabricated groups combined versus 16.9 points for the customized group and 17.2 points for the stretching-only group. The study design for this trial was generally in accordance with good research practice, although the researchers’ experience with prescription foot orthoses seemed to be minimal; thus their experience with casting and prescription of customized orthoses could be questioned. For example, the authors state that “each centre had reviewed a Prolab instructional video on casting technique” prior to the commencement of the trial, indicating that they may have been unfamiliar with the casting and prescription techniques normally used for customized foot orthoses. In contrast, it is likely that the other RCTs used experienced clinicians. Nonetheless, this study provides a useful test of the effects of prefabricated orthoses and demonstrates a slightly better outcome of stretching and prefabricated orthoses compared with stretching alone.

Finally, a recent study by Martin et al[70] compared customized orthoses with over-the-counter arch supports and tension night splints in 193 participants. At 12 weeks, there was no significant difference in pain reduction between the groups. Visual analog pain scores for pain during the day improved 34 mm for the custom-made group, 32 mm for the over-the-counter group, and 28 mm for the tension night splint group. Pain scores rating the pain upon the first step of the day improved 53 mm for both the customized and the over-the-counter groups and 61 mm for the tension night splint group. Interestingly, the customized group had fewer dropouts than the other groups: 16% for customized versus 27% for over-the-counter versus 29% for the tension night splints. Although this is the largest study to date (initial recruitment), indicating good statistical power and therefore less likelihood of occurrence of a type II error, there were some methodologic issues, including a very high dropout rate (24%), that may have biased the study’s results.

A comparison of the four studies that focused on prefabricated versus customized orthoses is presented in

Table 2. Summary of Randomized Controlled Trials Comparing Prefabricated and Customized Foot Orthoses for the Management of Plantar Fasciitis. 

Abbreviations: NSAIDs, nonsteroidal anti-inflammatory drugs; VAPS, visual analog pain scale.

and

Table 3. Methodologic Comparison of Randomized Trials Using Prefabricated and Customized Foot Orthoses for the Management of Plantar Fasciitis. 

Abbreviations: VAPS, visual analog pain scale; FFI, Foot Function Index. Note: Prospective sample size calculations, whether appropriate randomization and allocation concealment took place, and whether patients were blinded to treatment were not stated in any of the studies.

. Table 2 summarizes general information about these trials (eg, aims, participants, interventions, and findings), whereas Table 3 reports on whether or not they followed important methodologic recommendations (eg, prospective sample size calculation, allocation concealment, intention-to-treat analysis).

Clearly, there are many methodologic issues, such as allocation concealment, high dropout rates, and blinding, that have not been addressed in these trials. With these limitations in mind, the evidence suggests that foot orthoses produce beneficial reductions in pain and disability in people with plantar fasciitis. It is important to note, however, that true comparisons with control groups are largely lacking. Although difficult to achieve, comparison with a control group is important because conditions such as plantar fasciitis have a natural tendency to improve on their own. Without a true control group, a study may simply measure the natural improvement of the condition rather than a true therapeutic effect of the intervention.

Aside from these issues remains an important question: Are prefabricated or customized orthoses more effective? All of the trials appraised above have attempted to answer this question in one way or another. Some trials have demonstrated that customized orthoses provide better outcomes,[69,72] while others either show no difference[70] or show that prefabricated orthoses perform better.[71] Interestingly, the two largest trials[70,71]—and arguably the best designed from a methodologic standpoint—demonstrated either no difference or that the prefabricated orthoses tested performed better than the customized orthoses tested.

Therefore, while there are discrepancies relating to the effect of orthoses in these trials, the evidence so far suggests that certain prefabricated orthoses performed as well as, and in some circumstances better than, customized orthoses. This is an important finding because one of the most crucial issues relating to foot orthoses is their cost. Clearly, if two devices have similar therapeutic effects, but one costs less, then the less expensive device is more cost-effective. The cost of prefabricated foot orthoses is often much less than that of customized or prescription orthoses. Therefore, if a prefabricated orthosis is as effective as a customized orthosis that costs more, the prefabricated device provides better value for money. However, the cost of fitting and modifying prefabricated orthoses (ie, professional expertise) must be considered when assessing their overall cost.

If patients obtain a clinically worthwhile outcome (ie, a health status they are satisfied with) from prefabricated devices, then prescribing fully customized, prescription orthoses with every possible modification may be unnecessary. If customized orthoses are no more effective over a given period, then the only justification for them would be evidence demonstrating that prefabricated devices had a reduced effect over time and needed to be replaced more often. As yet, though, the issue of cost-effectiveness has not been properly evaluated.

Conclusion

Foot orthoses are a common form of treatment for plantar fasciitis. Although many studies recommend foot orthoses for this condition, there is limited good-quality research evaluating their effectiveness. Six RCTs have evaluated different types of foot orthoses for the treatment of plantar fasciitis, four of which have compared prefabricated and customized devices. Each trial has methodologic strengths and weaknesses, so caution is needed when interpreting their results. Nevertheless, the evidence suggests that foot orthoses produce reductions in pain and disability associated with plantar fasciitis, although caution is needed because true comparisons with control groups are largely lacking. At this stage, it is not possible to conclude that either prefabricated or customized orthoses are better, and it cannot be suggested that customized orthoses are better over time and therefore have a cost advantage. Additional good-quality RCTs are needed to provide this evidence.