Elsevier

The Foot

Volume 18, Issue 1, March 2008, Pages 15-19
The Foot

Minimal important difference: Values for the Foot Health Status Questionnaire, Foot Function Index and Visual Analogue Scale

https://doi.org/10.1016/j.foot.2007.06.006Get rights and content

Abstract

Background

There is a common misperception that a statistically significant result in clinical research equates to a meaningful or worthwhile outcome for patients. However, not all statistically significant findings are considered important by patients.

Objective

This research project aimed to calculate the minimal important difference for commonly used outcome measures in research related to the feet.

Methods

Data from 175 participants from two trials that evaluated conservative interventions for plantar fasciitis were used to determine minimal important differences for the following outcome measures: the Foot Health Status Questionnaire, the original Foot Function Index and a Visual Analogue Scale used to measure pain. The mean change in the outcome measure was compared to a simple global question of improvement. The amount of change in the outcome measure that related to “a little change” in the global improvement question represented the minimal important difference.

Results

For the Foot Health Status Questionnaire the following minimal important differences were found: 14 for pain, 7 for function and 9 for general foot health. Similar results were found for the Foot Function Index: 12 for pain, 7 for disability and 7 for Total Foot Function Index. The minimal important difference for the Visual Analogue Scale was 9 mm.

Conclusion

Our findings can be used to help interpret results from clinical trials that have used these outcome measures to evaluate the effectiveness of interventions, particular for the treatment of plantar fasciitis. Researchers can also utilise these minimal important differences to assist in prospective sample size calculations for clinical trials, although caution is required for trials on disorders other than plantar fasciitis.

Introduction

It is common for clinicians and researchers to view statistical significance as being equal to clinical significance [1]. Any treatment that is demonstrated to be ‘significantly’ different (i.e. better) from a statistical standpoint to another treatment is commonly then suggested to be more effective and accordingly, incorporated into clinical practice.

Statistical significance only refers to the probability that the result of an analysis between two treatments is truly different. Thus, a significant finding provides us only with a degree of confidence that the result is not a chance finding [2]. However, the mere existence of a statistically significant treatment effect does not always mean the effect is large, or even clinically worthwhile. That is, the effect may not be important to patients [3]. There is a very real chance, therefore, that a statistically significant result might be so small as to be meaningless to patients because they are unable to detect it. The problem of equating statistical significance with clinical importance is that it ignores the possibility that a statistically significant result may not always be clinically relevant, or equally, a clinically important finding may not always be statistically significant [1], [4], [5].

The most effective method of examining a study's results for clinical importance is by comparing the results against an already established minimal important difference (MID) for the outcome measure being implemented [1]. A minimal important difference has been defined “as the smallest difference in score in the domain of interest which patients perceive as beneficial and which would mandate, in the absence of troublesome side effects and excessive cost, a change in patients’ management” [6, p. 408]. This value was formerly known as a minimal clinically important difference (MCID), however deletion of the word ‘clinical’ was made to remove the focus on the clinical context rather than the patient's experience [7].

Once established, the minimal important difference for an outcome measure is not only useful for interpreting trial results, it is also important for sample size calculations [1], [8]. An appropriate sample size ensures sufficient power to detect, using statistical tests, a clinically worthwhile result (i.e. a minimal important difference) if one indeed exists. This point is fundamental to good clinical trial design—without an appropriate, prospective sample size calculation the chance of statistically invalid findings will be high due to a lack of power.

This study, therefore, aimed to investigate the minimal important difference of three commonly used outcome measures used in foot-related research: the Foot Health Status Questionnaire, the Foot Function Index and the Visual Analogue Scale.

Section snippets

Methods

Data for this study were taken from two trials evaluating conservative treatments for plantar fasciitis. Both trials used similar methodology – the same general protocol and outcome measures – although one trial evaluated a short-term treatment (outcomes measured over 2 weeks) and the other a more long-term intervention (outcomes measured over 12 months). The long-term trial ran from 1999 to 2001 and the short-term trial in 2002. Findings relating to the effectiveness of the treatments

Results

A total of 175 participants were recruited into the two trials. The mean change scores from all time points for both trials were combined. To appreciate the number of individual participant responses that were analysed, the total pooled number of data entries for each domain of the outcome measure are presented in Table 1.

The mean ‘no change’ and ‘a little change’ scores are presented with the minimal important differences for all outcomes in Table 2, Table 3, Table 4. The ‘no change’ score

Discussion

When evaluating the effectiveness of interventions in trials relating to the foot, pain is a common outcome measure. While pain is an important outcome, it does not take into account the effect a disorder has on the patient's life in general. Recently, outcome measures have begun to focus on health status and health-related quality of life, where a broader interpretation of the impact of a condition on a person's life is taken into account [21]. The Foot Health Status Questionnaire and Foot

Conclusion

This research project calculated minimal important differences for the Foot Health Status Questionnaire, the Foot Function Index and the Visual Analogue Scale. Our findings will assist with interpreting results from clinical trials that have used these outcome measures to evaluate the effectiveness of interventions, particularly for the treatment of plantar fasciitis. Further, with the limitations outlined in mind, researchers can cautiously use these minimal important differences to assist in

Conflict of interest

The authors have no financial or personal relationships with other people or organizations that could inappropriately influence (bias) this work.

Acknowledgement

We thank Anne-Maree Keenan and Anthony Redmond for their role in this project.

References (28)

  • L.G. Portney et al.

    Foundations of clinical research: applications to practice

    (2000)
  • L.E. Kazis et al.

    Effect sizes for interpreting changes in health status

    Med Care

    (1989)
  • G.A. Wells et al.

    Minimum important difference between patients with rheumatoid arthritis: the patient's perspective

    J Rheumatol

    (1993)
  • H.J. Schunemann et al.

    Commentary—Goodbye M(C)ID! Hello MID, Where do you come from

    Health Serv Res

    (2005)
  • Cited by (134)

    View all citing articles on Scopus
    View full text