**3. Results**

A total of 21 women with FM from a local association participated in this study. The main characteristics of the participants are shown in Table 1.

**Table 1.** Descriptive characteristics of the participants.


Abbreviations: N, sample; SD, standard deviation; BMI, body mass index, FIQR, Fibromyalgia Impact Questionnaire Revised.

Table 2 shows the relative reliability (ICC) and absolute reliability (SEM and MDC) of the performance obtained in the 3MBWT, under the single and dual-task conditions in both test and retest with the different devices. Following the recommendations by Koo et al. [41], the 95% confidence intervals of the ICC were used to interpret the reliability values. Regarding the single condition, "poor" to "good" and "moderate" to "excellent" reliability values were obtained for the stopwatch and Chronopic, respectively. On the other hand, in the dual-task condition, the reliability values for the stopwatch were "moderate" to "excellent" and for the Chronopic were "good" to "excellent".



Abbreviations: 3MBWT, 3 m backward test; s, seconds; SD, standard deviation; ICC, intraclass correlation coefficient; CI, confidence interval; SEM, standard error of measurement; MDC, minimal detectable change; DT, dual-task.

Reliability values obtained by comparing the different devices in both the single and dual-task conditions in the test and retest are shown in Table 3. Taking into account the 95% confidence intervals of the ICC, a reliability value of "good" to "excellent" was obtained for the test, and an "excellent" value was obtained for the retest in the single and dual-task condition.

**Table 3.** Reliability of the 3MBWT under single and dual-task conditions using stopwatch and Chronopic in the test and retest.


Abbreviations: 3MBWT, 3 m backward test; s, seconds; SD, standard deviation; ICC, intraclass correlation coefficient; CI, confidence interval; DT, dual-task.

Figure 1 shows the Bland–Altman plots of the times obtained by stopwatch and Chronopic in test and retest in the single and dual-task conditions, and the times obtained between the two devices in both the single and dual-task conditions in the test and retest, respectively.

Table 4 shows the concurrent validity analysis results of 3MBWT and TUG test. All correlation values obtained were classified as strong [43] in the test and retest and the different devices.


**Table 4.** Concurrent validity between 3MBWT and TUG under single and dual-task conditions, in test and retests, with stopwatch and Chronopic.

Abbreviations: 3MBWT, 3-m backward test; TUG, timed up and go. \*\*\* *p*-value < 0.001.

Finally, the relationship between the performance of the 3MBWT in test and retest conditions under single and dual-task conditions measured with different devices and the impact of the disease were strong, excepting the 3MBWT in test condition under single condition measured with a stopwatch (r: 0.488), which was moderate. In analyzing the dimensions that make up the FIQR questionnaire, strong correlations were obtained between the "symptoms" dimension and all the conditions and devices. Similarly, strong correlations were also obtained between the dimension "overall impact" except for the single and dual-task conditions in the test period, measured with the stopwatch and the Chronopic, respectively, where moderate correlations were reported. As for the dimension "function," only moderate correlations were found in the single condition in the retest and test periods, measured with a stopwatch and Chronopic, respectively. Results are shown in Table 5.

**Figure 1.** (**a**) differences between test and retest vs. the mean of the two measurements under the single condition using a stopwatch; (**b**) differences between test and retest vs. the mean of the two measurements

under the single condition using a Chronopic; (**c**) differences between test and retest vs. the mean of the two measurements under the dual-task condition using a stopwatch; and (**d**) differences between test and retest vs. the mean of the two measurements under the dual-task condition using a Chronopic; (**e**) differences between stopwatch and Chronopic vs. the mean of the two measurements under the single condition in test; (**f**) differences between stopwatch and Chronopic vs. the mean of the two measurements under the single condition in retest; (**g**) differences between stopwatch and Chronopic vs. the mean of the two measurements under the dual-task condition in test; and (**h**) differences between stopwatch and Chronopic vs. the mean of the two measurements under the dual-task condition in retest.

**Table 5.** Correlation between 3MBWT and FIQR under single and dual-task conditions, in test and retest, with stopwatch and Chronopic.


Abbreviations: 3MBWT, 3-m backward test; FIQR, fibromyalgia impact questionnaire revised. \* *p*-value < 0.05; \*\* *p*-value < 0.01; \*\*\* *p*-value < 0.001.

Figure 1 Bland–Altman plots of the times obtained by stopwatch and Chronopic in test and retest under the single and dual-task conditions and Bland–Altman plots of the times obtained between the two devices under the single and dual-task conditions in the test and retest.

#### **4. Discussion**

This study aimed to investigate the test–retest reliability and concurrent validity of the 3MBWT in women with FM under single and dual-task conditions. We also aimed to investigate the agreement between a manual stopwatch and a Chronopic automatic stopwatch. In order to provide clinical and objective directions, the SEM, MDC, and Bland–Altman plots were reported. Generally, good reliability values were obtained for the 3MBWT test when measured with a stopwatch and Chronopic in both single and dual-task conditions. Similarly, the 3MBWT and the TUG achieved good concurrent validity. Lastly, performance between 3MBWT and the impact of the disease were analyzed, and consistent relationships were found between both.

In the present study, walking backward has been analyzed because it is a complex task that requires high neuromuscular and proprioceptive control [22]. Walking backward usually occurs in activities of daily living, such as opening a door, avoiding an obstacle, or backing up to a chair [23]. Therefore, it is a sensitive measure for assessing mobility and balance deficits [24,25]. In people with FM, performing a functional assessment of mobility and balance is essential to manage the disease. However, one of the main tests used to evaluate these characteristics in this population is the TUG [9,11,17]. However, this test does not include backward gait. In this sense, we consider it relevant to analyze the psychometric properties of the 3MBWT in people with FM since previous studies have focused on other populations obtaining good reliability and validity values [26–29]. Therefore, the 3MBWT is a tool that can be used elsewhere in the clinical and research context. Moreover, due to the fact that more than one task is performed at the same time in activities of daily living, and the impairment in dual-task ability detected on people with FM [32,44,45], we decided to incorporate the dual-task paradigm in this cross-sectional study.

Results showed that the 3MBWT could be considered reliable under single and dualtask conditions when measured with both a stopwatch and a Chronopic. However, it is necessary to consider the ICC fluctuation range [41]. Nevertheless, the data presented in the Bland–Altman plots (Figure 1) showed that most of the 3MBWT values were close to the mean of the test–retest differences in both single and dual-task conditions, having a bias close to zero and a reduced variability seeing the limits of agreement. Therefore, there is a high level of agreement between the test–retest measures evaluated in the 3MBWT since the values obtained provide consistent results. The reported ICC in the single condition with a stopwatch (0.71, 95% CI 0.283–0.882) is lower than those reported in previous studies that investigated test–retest reliability in stroke [26] (0.985, 95% CI 0.973–0.992), community-dwelling older adults [27] (0.940, 95% CI 0.90–0.96), and primary total knee arthroplasty [46] (0.942).

Our findings might be due to the symptomatology of FM [47], characterized by widespread pain and fatigue, which can fluctuate in intensity and severity over time [48,49]. For this reason, a person's symptoms and level of functioning can vary daily and affect the physical fitness performance. In this regard, previous studies highlighted what can affect performance in physical fitness tests [50,51]. The same rationale can also affect cognitive function [52], including attention, memory, and information processing. These factors may have contributed to the variability of results between the test and retest. Nevertheless, the ICC values obtained in the present study are acceptable, so the test analyzed seems to be stable enough to be used in the characterization of people with FM.

The optimal time interval between tests may vary depending on the construct being measured, the stability of the construct over time, and the target population [53]. For this study, a seven-day period was selected, the same as previous studies conducted in people with FM [11,18,38,39], to minimize the impact of potential confounding variables, such as recovery or learning effects [54]. However, other studies have used shorter test–retest times to assess test–retest reliability on the 3MBWT [26–29]. In addition, the tests (TUG and 3MBWT), as well as the conditions (single and dual-task), were randomized to ensure that the order of administration may not bias the results. In this way, obtaining more accurate and reliable results is possible by eliminating biases and reducing the influence of external factors that may affect the results.

As expected, the ICC values were slightly higher when using Chronopic versus stopwatch, since the use of a manual stopwatch adds human variability to the measurement by the evaluator [9,55]. Therefore, using an automatic timer can be a cost-effective alternative to assess performance in the 3MBWT in both single and dual conditions. However, although the ICC values obtained by the stopwatch are slightly lower than those obtained by the Chronopic, our data suggest that the use of a manual stopwatch could be also very useful for this test, since it yields relatively good reliability values in both single (0.71, 95% CI 0.283–0.882) and dual-task conditions (0.89, 95% CI 0.718–0.954) (see Table 2). Similarly, the lower ICC obtained in the single and dual-task conditions in the test performance was probably due to human error experienced when using a manual stopwatch. This may be suggested since the scores between the stopwatch and the Chronopic differed slightly, unlike the values obtained in the new test. Nevertheless, the ICC values obtained are classified as good to excellent (see Table 3).

Our study also provided the SEM and MDC values of the 3MBWT under all conditions and devices. These values are important for interpreting the results of the 3MBWT, because they can help clinicians and researchers to determine if there are meaningful changes in the performance of this test. Furthermore, the estimate of random variation in the data (SEM) and the minimum detectable change (MDC) are both lower when the Chronopic is used. Previous research has also observed this trend assessing test–retest reliability under the single condition in TUG and 30 s chair stand test [9,55]. Bland–Altman plots were also reported for a more comprehensive analysis of the results, showing bias and limits of agreement.

In our study, the TUG test was used to test the validity of the 3MBWT since it is a tool frequently used in clinical practice [56,57], showing high reliability in FM in both single [9] and dual-task conditions [11]. Although the TUG test does not contemplate walking backward, as previous studies did [26–29], in the present study we have used the TUG test to conduct the concurrent validity. Moreover, this test has obtained the highest relationship values in most cases [26,27,29]. Significant correlations between 3MBWT and TUG tests were obtained, which can be considered relevant since the TUG test comprises movements that can occur in activities of daily living, including walking, turning, and sitting [37]. In addition, correlation analyses showed a strong correlation in the single condition between the 3MBWT and TUG test measured with a stopwatch, as in previous studies in stroke [26], community-dwelling older adults [27], and adults [23]. In our study, the correlation in single conditions between 3MBWT and TUG measured through the automatic timer also showed strong correlation values. Similarly, the correlation in the dual-task conditions also obtained strong values in the test and retest measured by a stopwatch and a Chronopic.

Our results also found a positive correlation between the 3MBWT performance and the impact of FM obtained through the FIQR questionnaire. Strong to moderate correlations were found between FIQR total score and the performance obtained in single and dual-task conditions using manual stopwatch and Chronopic. Additionally, correlations between FIQR dimensions and 3MBWT have been reported. In this line, the strongest correlations were found in the symptoms dimension, followed by overall impact, where moderate correlations were also found. However, function dimension did not show significant correlation with 3MBWT performance in most cases. These findings could indicate that performance on the 3MBWT does not correlate well with the actions included in the function dimension by including activities that are not highly associated with walking backward (i.e., combing hair, preparing food, shopping) but do correlate with symptomatology and overall impact. Nevertheless, future studies should corroborate this hypothesis.

Previous studies have highlighted the importance of assessing backward gait in older adults [24,25], showing a reduction in performance in different parameters when comparing to young and middle-aged adults. In this regard, during backward gait, greater neuromuscular and proprioceptive control is required, in addition to faster and more frequent balance corrections due to the elimination of visual feedback [22,58]. Additionally, it has been shown that walking backward shows greater or equal sensibility than walking forward, and it is strongly related to the risk of falling [23]. In this sense, assessing backward gait, via the 3MBWT, in FM may be of interest to clinicians to observe changes in mobility and balance due to the balance impairment manifested in this population [59]. Thus, a previous study [60] conducted a physical exercise intervention based on walking backward in community-dwelling older adults. Given these reasons and the strong relationships between the impact of the disease and the performance obtained in the 3MBWT, this test can be used as a useful tool in FM populations to assess exercise-based interventions.

The present study has some limitations. In this regard, the relatively small sample size did not allow us to generalize results to all women with FM. Moreover, only women were included in this cross-sectional study. Thus, results cannot be extrapolated to men with this disease. Therefore, it would be interesting to extend the sample with different age ranges, which allow us to establish cut-off points.

#### **5. Conclusions**

The results obtained from this study show that the 3MBWT is a reliable tool under the single and dual-task conditions in women with FM. It shows higher reliability values when time is taken using a Chronopic. This test also shows high concurrent validity with the TUG test, and its performance is related to the impact of the disease. These results may help clinicians and researchers in the assessment of balance and functional mobility and to interpret the effect of interventions in this population.

**Author Contributions:** Conceptualization, J.L.L.-L., F.J.D.-M. and N.G.; Data curation, A.M.-G. and F.J.D.-M.; Formal analysis, J.L.L.-L., A.M.-G. and F.J.D.-M.; Investigation, S.V.; Methodology, S.V. and N.G.; Project administration, N.G.; Supervision, S.V.; Validation, A.M.-G. and N.G.; Visualization, J.L.L.-L.; Writing—original draft, J.L.L.-L. and S.V.; Writing—review and editing, A.M.-G., F.J.D.-M. and N.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** The author J.L.L.-L. was supported by a grant from the Spanish Ministry of Education, Culture, and Sport (FPU18/05655). The author S.V. was supported by a grant from the Universities Ministry of Spain and the European Union (NextGenerationUE) "Ayuda del Programa de Recualificación del Sistema Universitario Español, Modalidad de ayudas Margarita Salas para la formación de jóvenes doctores" (MS-03). The author A.M.-G. was supported by a grant from the Spanish Ministry of Education, Culture and Sport (FPU17/03130). In the framework of the Spanish National R+D+i Plan, the current study was co-funded by the Spanish Ministry of Sciences and Innovation (reference PID2019-107191RB-I00/AEI/10.13039/501100011033). This study was also funded by the Research Grant for Groups (GR21176) funded by the Junta de Extremadura and the European Regional Development Fund (ERDF/FEDER) "a way of doing Europe". This study was supported by the Biomedical Research Networking Center on Frailty and Healthy Aging (CIBERFES) and FEDER funds from the European Union (CB16/10/00477) "a way of doing Europe".

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Research Ethics Committee of the University of Extremadura (approval reference: 51/2021).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Data are available under reasonable request to the corresponding author.

**Acknowledgments:** The authors are very grateful to the Extremadura Association of Fibromyalgia (AFIBROEX) in Cáceres for helping recruit the participants for this study.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
