**3.6 Study instrumentation used in this study**

The following study questionnaires were used to measure and monitor the effectiveness of the brace. Three questionnaires were used. The first questionnaire that was used to assess pain was the Oswestry Disability Index Questionnaire v2.1a (ODI) [26]. The second questionnaire that was used to assess the patients' quality of life was the EQ5D-5L [27]. The final questionnaire was used to assess the patients' mental and physical health, the Short Form-36 Version 2 (SF-36v2) [12]. The participants in the control group received the questionnaires together with the standard NHS treatment; this comprised a yearly follow-up appointment with their regular consultant to assess whether the curve had progressed to the point where a surgical approach was required.

*A Mixed Methods Study of the Experiences and Effectiveness of a Soft Brace for Adults… DOI: http://dx.doi.org/10.5772/intechopen.92387*

## **3.7 Oswestry Disability Index (ODI v2.1a)**

The ODI v2.1a questionnaire is one of the most commonly used questionnaires to assess lower back pain. The ODI is one of the most valid and reliable outcome measures with regard to the patient's perception of the 'pain they feel today' [28]. The ODI comprises of 10 sections; each section has six statements (see example below; **Table 1**). Responses are made on a five-point scale. As shown in the example below, the first statement is marked 0, and the last statement is marked 5. All additional statements are scored according to rank, with the highest score being taken if more than one box is marked in each section. Once the questionnaire is complete, the score is then calculated using the following formula: ODI % = Total score/5 × number of questions answered × 100. Therefore, if the ODI score is 40 based on the participant answering 4 for all 10 sections, the percentage of disability is determined: 40/5 × 100 = 80% disability. Since the initial publication, the ODI has been widely validated [29] and deemed to be a reliable measure for detecting small reductions in pain and disability before and after treatment [30].

The ODI correlates positively with the SF-36 and EQ5D questionnaires. In essence, a reduction in pain attained from the ODI correlates with an improvement in mental and physical health on the SF-36 questionnaire and an improvement in quality of life from the EQ5D for patients who had lower back pain [31, 32]. Wittink et al. [32] compared the outcomes of 424 patients with chronic pain who had been referred to a multidisciplinary pain centre where each patient was required to complete the SF-36, Multidimensional Pain Inventory (MPI) and the ODI before or on the date of their first appointment. They found that the MPI, SF-36 and ODI had good psychological measurements. The study also concluded that the ODI had the least amount of respondent burden and was easier to score, although the questionnaire does not provide as much detail as the MPI or the SF-36.

### **3.8 EQ5D-5L**

The EQ5D-5L is a widely used generic measurement of health-related quality of life that requires patients to self-report problems with regard to five elements: mobility, self-care, usual activities, pain/discomfort and anxiety/depression [33]. The questionnaire is split into two parts with the first part containing five questions that explore different elements of health, with each of the five questions containing three levels of severity, no problem, some problem or extreme problem, allowing patients to be classified into 1 of 243 states [34, 35]. An example question (see **Table 2**) from the EQ5D questionnaire is seen below:


**Table 1.** *Example ODI questions.*


## **Table 2.**

*Example EQ5D questions.*

From these 243 health states, a time trade-off (TTO) is calculated; this TTO is used to represent a person's quality of life. A value of 1 represents full health and a value of 0 represents being dead [36]. For this research study, the TTO scores were calculated using the software provided by EuroQual who designed the questionnaire. The second part of the EQ5D-5L requires the patient to indicate how 'good' or 'bad' their perception of their own general health is by marking a visual analogue scale (VAS). A VAS is a measurement tool that aims to measure an attribute of health that is believed to be on a continuous range across a continuous value. The scale ranges across a horizontal line that starts at 0 representing no pain and moves gradually to 100 representing extreme pain. The line is 100 mm in length and words are placed at the start of the continuum (0) representing no pain, and at the end of the line a100mm line representing extreme severe pain (see **Figure 1** below).

The validity and reliability of the EQ5D questionnaire have previously been confirmed by Brazier et al. [37] who compared the results of the EQ5D to the SF 6D (a development from the SF-36). Brazier's research study [37] found that on a data set of 2436 patients with a wide range of medical conditions ranging from lower back pain, chronic obtrusive pulmonary disease and irritable bowel syndrome, the questionnaire showed signs of being sensitive to change and having a strong coefficient of internal consistency.

### **3.9 SF-36v2**

The last questionnaire used in this study was the SF-36v2. This questionnaire is a multidimensional, non-diagnostic-specific measure of pain that consists of eight health scales: physical functioning (10 items), role limitations—physical (4 items), bodily pain (2 items), general health (5 items), vitality (4 items), social functioning (2 items), role limitations due to emotional problems (3 items) and mental health (5 items) [38]. The scale is directly transformed into a 0–100 scale based on the assumption that each question carries equal weight. The lower the score, the greater the disability; in essence, a score of 0 is equivalent to a maximum disability, and a score of 100 is equivalent to no disability. The SF-36v2 has been reported to be both a reliable and valid measurement of health and has been shown to be sensitive to changes in health conditions [39].

**Figure 1.** *Example of a visual analogue scale.* *A Mixed Methods Study of the Experiences and Effectiveness of a Soft Brace for Adults… DOI: http://dx.doi.org/10.5772/intechopen.92387*

### **3.10 Procedure**

Medical staff including consultants, physiotherapists and spinal nurses were the primary recruiters for participants for the study. Participants were recruited when they attended their regular orthopaedic clinics. Once consent was attained, the participants were then randomised into either the treatment or control group by a computer randomisation program. As part of the standard NHS procedure, on arrival the participants received a routine clinical examination. This included the measurement of the Cobb angle and pelvic/shoulder tilt, in addition to a posterior/ anterior X-ray using the soft brace X-ray protocol. The soft brace X-ray protocol was used for all clinical visits as it provided a standardised foot position, which in turn kept the patient's posture natural and standardised from visit to visit. The protocol also allowed for direct comparison of frontal and lateral X-rays and increased the inter-session reliability. If the participant fulfilled the study criteria, they were offered the opportunity to participate in the trial; if they declined, they then continued along the standard NHS treatment pathway.

If the patient chose to participate in the study, they were given the participant information sheet (PIS) together with the series of questionnaires described above to complete (ODIv2.1a, EQ5D-5 L and SF-36v2). Further information about the trial and details of future clinical appointments were also included in the PIS. The information sheet gave details about future clinical appointments, how the trial would run and what was expected of the individual with regard to their participation (such as being able and willing to attend the clinical appointments, adhere to the 8 hours wearing time and complete the questionnaires on time). The lead researcher then explained to the patients that they would be randomly allocated by a computer program into either the treatment group where they would receive the brace and questionnaires or the control group where they would follow the standard NHS treatment and also receive the study questionnaires. If participants were placed in the treatment group, they were given an appointment where the brace was fitted by a trained soft brace clinician.

The baseline questionnaire data were then administrated and collected by the lead researcher. A follow-up appointment was also made 1 month into treatment where an additional in-brace X-ray was taken as specified by the study protocol. This additional X-ray was not part of the standard NHS treatment; however, it was deemed necessary to assess whether the brace was having any immediate effect on the shape and position of the spine. This extra dose of radiation that the patients received as a result of the additional X-ray was deemed to be acceptable by the research ethics committee. Patients were also informed that they would be required to attend future appointments at 1, 3 and 6 months. This was in addition to their standard clinical appointments to see their regular consultant at 12 months to assess the long-term effects of the brace. Upon attending these routine clinic visits, the brace was adjusted by the soft brace clinicians if it was deemed to be too loose or ill-fitting as a result of any changes in their shape and/ or posture.

The brace is recommended by the manufacturer to be worn for 8–12 hours per day for the first 3 months and then adjusted to the individual's requirements after this. These adjustments included changing the wearing time and/or adapting the position of the bands if they had become loose or ill-fitting. Apart from the first appointment, which lasted approximately 90 minutes, each follow-up clinical visit lasted no longer than 40 minutes. The first appointment was slightly longer as this was where the participants were fitted and measured for their brace, whilst also providing them with the opportunity to ask the chief investigator (CI) or clinicians any further questions regarding their participation in the trial.

The participants in the control group received the ODIv2.1a, EQ5D-5L and the SF-36v2 questionnaire at home at months 1, 3 and 6 months, in addition to their standard treatment pathway. A follow-up yearly X-ray together with an appointment with their consultants was also made to see if surgery would be required due to curve progression. Both groups of participants were asked to complete and return all documents to the first author's University in a prepaid envelope. If individuals decided they did not want to participate in the trial or if they decided to have surgical treatment, they were followed up at 12 months by their consultant with an X-ray to assess if any curve progression was apparent.

If patients chose to enter the trial, they were told that they could withdraw at any stage without giving a reason and it would not affect their future treatment. Participants were then provided with a participant information sheet and offered the opportunity to discuss the study with their consultant or a member of the research team. If patients agreed to participate in the trial, the consultants obtained signed consent. These completed forms were then passed on to the researcher who contacted the individual to verify they were still interested in participating in the trial (with at least a 24-hour gap after completing the slip). Once the patients confirmed they were still interested in participating, it was at this stage of the trial that they were informed which group they had been randomised into. At this point if the participants decided not to participate in the trial, it was explained they would follow the standard NHS treatment pathway. Each participant was also informed that if they did participate, all data they provided would be kept confidential and anonymised.

### **3.11 Sample size**

A sample size of 102 for this study was originally calculated based on literature from previous back pain studies. An attrition rate of 10% was also added to the sample size [40], meaning a total sample size of 112 participants was required. Due to recruitment issues, this number of participants was not achieved. In total only 15 participants were recruited to the study. However, as randomisation took place before the research team had the opportunity to meet and screen the potential participants, two participants who were included by the consultants did not meet the inclusion criteria and were therefore not eligible, ultimately resulting in only 13 participants being involved. Given the low sample size, it is important to establish that the appropriate analyses (and therefore appropriate conclusions) were applied to the data. Research [41] has shown that even with a sample size of two people in a series of fake t-tests, the Type 2 error that occurred did not surpass the acceptable value of 5%; this has also been confirmed in studies by [42, 43]. Therefore, the small numbers in this study should not have any effect on the validity of the results generated from the questionnaires, although of course the small study numbers must be recognised in the interpretation of the results, as it is possible that the participants may not form a typical cross section of patients with degenerative scoliosis.

### **3.12 Statistical analysis**

All study data were anonymised and analysed using a two-way repeated measure analysis of variance in SPSS [44]. The two-way analysis of variance (ANOVA) was used as it allows the differences between group means and their treatment methods to be compared over several time points [45]. The treatment group was compared against the mean data from the control group for each of the three time points to determine whether any significant differences were present. In-group comparisons were also calculated for both groups to see if any significant in-group changes occurred.

*A Mixed Methods Study of the Experiences and Effectiveness of a Soft Brace for Adults… DOI: http://dx.doi.org/10.5772/intechopen.92387*

Any missing data or data belonging to participants who withdrew from the study were analysed based on an intention-to-treat (ITT) basis. It was important to use ITT as the main problem researchers find when using an RCT study design is that participants do not always follow the instructions or they drop out of the trial [46]. The benefits of using ITT are that the analysis still reflects clinical situations and it gives unbiased estimations on the effectiveness of the proscribed treatment [47]. Furthermore, the use of ITT maintains the original sample size; if dropouts were excluded from the overall data set, a reduction of statistical power may occur.

### **3.13 Outcome measures**

The primary outcome measure for this study was a change in the participants' ODI scores over the 6-month duration. Secondary outcome measures were any changes in the EQ5D-5L and SF-36v2 scores.

### **3.14 Results**

### *3.14.1 Oswestry Disability Index*

The baseline data in **Table 3** show that the treatment group had a higher mean ODI than the control group upon initial consultation, although this difference was not statistically significantly different. Furthermore, a greater standard deviation was also present which indicated that a larger variation from the mean score was present within the treatment group. After 3 months, the mean ODI calculated from the treatment group's data showed a reduction of 9% in the 'pain they feel today' from baseline, in comparison to a 6.25% increase from the control group, although the standard deviation remained similar for both groups.

At the 6-month stage of the trial, the treatment groups' mean ODI decreased by a further 1.6%; this indicated a decrease in pain. This is in contrast to the control group, who as a group had a mean ODI change from data collected between months 3 and 6 of the study of 19%. The scores represent an overall decrease of 10.6% from the treatment group over the 6-month trial compared to an overall increase of 25.25% over the course of the 6-month trial from the control group.


### **Table 3.**

*Analysis of ODI scores over 6 months.*

1M0 represents the baseline mean ODI scores, M3 represents month 3 mean scores and M6 represents month 6 mean ODI scores. The +/− column represents score changes between months. Treatment represents the treatment group and control represents the control group. Furthermore, there was no significant difference between the treatment and control group over the course of the 6-month study (see **Table 3**). Also, there was no significant difference or in-group interaction amongst the groups between the treatment time points.
