The Influence of Using a Footstool during a Prolonged Standing Task on Low Back Pain in Office Workers

1

Whilst the research question is of value,   there are currently areas of the paper that need addressing before being   suitable for publication. In particular, there seem to be an imbalance in the   literature review and discussion, with a focus on the probable benefits of   footstall use for LBP patients. The results do not conclusively support this   finding, and this should be reflected in the write up.

The   introduction and discussion have been edited to reflect the reviewer’s   concern about an imbalance in the literature review. The reviewer’s specific   comments with regards to this have been address below.

2

Given that   there is no calculation of ‘effect size’ the title of the paper could be   altered to more accurately reflect its contents.

The word   “effect” has been removed from the title of the paper. The title now reads “The   influence of using a footstool during a prolonged standing task on low back   pain in office workers”.

3

Abstract: Lines 14-16 Please make it clear in this section that   the single group performed the task both with a footstall and without with a   1 week separation. As a reader I was unsure as to whether this was a   comparison between two different groups or the same group repeated.

The abstract has been edited to more   clearly indicate that one group of participants performed the tasks one week   apart. The added text is underlined below.

Revised text: “This repeated   measures within subjects study aimed to determine whether office workers   with LBP are able to work at a standing workstation for one hour without   exacerbating symptoms, and whether using a footstool affects the LBP   severity. Sixteen office workers with   LBP performed computer work at a standing workstation for one hour under the   following two conditions one week apart: with a footstool and without   a footstool.”

4

Introduction: Lines 35-36 The end of this   sentence could be re-written, as it currently it not clear.

This   sentence has been re-written to read as follows:

“Having regular sitting breaks may help relieve the progression   of LBP, but the utility of this strategy and duration of sitting breaks   needed are unclear [4].”

7

Introduction: Lines 38-39 and 47 Whilst   there is some evidence for the degree of lordosis being related to LBP, the   majority of the literature to date is more inconclusive. This should be   reflected in the critique of the literature at this stage.

The   sentence at lines 38-39 has been rewritten to acknowledge inconsistencies in   the literature. It now reads as:

“While   research findings are often inconclusive  [6,7], the following risk factors   for LBP have been suggested: increased lumbar lordosis [6,7], a lack of movement [8], and reduced sagittal plane   postural variability [9].”

The   following reference has been added to support this statement:

Sadler SG, Spink MJ, Ho A, De Jonge XJ, Chuter VH. Restriction in   lateral bending range of motion, lumbar lordosis, and hamstring flexibility   predicts the development of low back pain: a systematic review of prospective   cohort studies. BMC Musculoskelet Disord. 2017;18:179.   https://doi.org/10.1186/s12891-017-1534-0.

The   sentence at line 47 has been edit to reflect that lumbar lordosis has been   proposed to be a risk factor for LBP. The sentence now reads: “As increased   lumbar lordosis is proposed to be a risk factor for the development of   LBP during standing and intermittent trunk flexion has been associated   with decreased standing-induced LBP, using a footstool may help to reduce   LBP in office workers who are using standing workstations.”

8

Introduction: Line 51 Office worker(s) Line   52 Remove the word ‘the’ after [13]?

This word   has been removed.

Introduction: Given that previous studies   have shown that ‘non LBP’ participants can also develop LBP during standing   tasks, why did you not choose to include such a control group?

The   majority of research investigating standing workstations in office worker has   been undertaken in pain-free populations.    In light of this, we specifically wanted to investigate the use of   standing workstation (with and without the addition of the footstool) in   officer workers with LBP. Further, previous research has investigated the use   of a footstool in individuals without LBP [21].

9

Methods: Lines 62-64 Is there any more   detail you could provide regarding the LBP participants? E.g. were there   participants with radicular symptoms, or was the group made up of   non-specific LBP participants?

Study   participants had non-specific LBP. This has been added to the eligibility   criteria.

The edited   text is underlined in the following sentence: “Participants were eligible for   inclusion if they were 18 years old or older, performed sitting computer work   for more than 30 hours per week, and had non-specific LBP for ³3 months that was rated in   severity as ³2/10 on an 11-point numeric   rating scale (NRS) anchored with "no pain" at 0 and "worst   pain imaginable" at 10.”

10

Methods:   Line 94 and Line 123 Given that over three quarters of participants in both   conditions experienced LBP during the exercise, and this was expected, did   you consider how this may this have affected the second task? Using the   methodology described (i.e. testing 1 week apart) it would seem likely that   there would be some sort of behavioural change in the second task due to fear   avoidance etc. Why therefore did you decide to repeat the task, and not   compare two separate groups?                                         

We choose   to use a repeated measures within subject design for this study to enable us   to compare using and not using a footstool within one participants. This   designed replicated that used previously to assess footstool use in   asymptomatic individuals [21] and accounted for likely   differences in standing-induced LBP between participants. To account for   possible effects of testing order between the footstool and no footstool   conditions, order was randomised for each participant.

The following sentence was added to the start of the methods to   explain the study design: “A within subject repeated   measures study design was used to compared severity of   LBP experienced during a prolonged standing tasks with and without a   footstool. »

11

Results: Line 145 remove ‘or’ replace with   ‘of’. Line 152 insert the word ‘of’ after n=8

I believe   the change to line 145 has been made by the editor as there is no “or” in   this sentence.

The word   “of” has been inserted in line 152 as recommended.

12

Discussion: Paragraph 1 and Table 1 Given   that there were an equal number of participants who improved or worsened   under the footstool condition, it is felt that this should be reflected more   in the discussion. Currently the discussion focusses very much on the likely   positive impact of the footstall, but the results do not support this   unidirectional standpoint.

Text has   been added to the first paragraph of the discussion to emphasise the   variability in responses between participants and state that equal numbers of   participants reported worsening and improvement in LBP symptoms when using   the footstool.

  Added text: “However, it must also be noted that an equal number of   participants (n=4) reported worsening and improvement in their LBP when using   the footstool which suggests considerable variability between individuals.”

13

Discussion: Lines 185-187 It is also shown   that the use of a footstall may worsen LBP symptoms in this group.

Text has   been added to the previous paragraph to further emphasise that a proportion   of individuals using and not using a footstool experienced worsening of their   LBP symptoms. The text was added to this paragraph as it was thought to link   more closely with the content of this paragraph than the following paragraph,   and it allowed a balance presentation of worsening of symptoms in some   participants in both the footstool and no footstool conditions.

The   underlined text has been added: “However, it must be noted that a proportion   of individuals did experience exacerbations in their pain when using a   standing workstation both with and without a footstool.”

14

Discussion: Lines 188-200 You mention the   above at the start of this paragraph, but then describe only the likely   positive impact mechanisms of using a footstall. It may be suitable to also   highlight how the opposite may be the case and why? It may also be of value   to discuss possible sub-groups of LBP patients that may respond differently   during such tasks (i.e. O’Sullivan’s flexion/extension aggravation groups).

Text has   been added to this paragraph to provide suggestions for the positive impact   of not using a footstool, and also to identify that sub-groups of individuals   with LBP may respond differently to standing with and without a footstool.

Added   text: “However, alternatively, it is possible that not using a footstool may   facilitate greater variability in movement as individuals may shift their   weight differently when not perceived to be limited by lifting a leg on and   off a footstool. Due to the diversity of LBP presentations and suggestion of LBP   sub-groups with different aggravating factors and postures, it may not be   surprising that individuals respond differently to using and not using a   footstool.”

The   following reference has been added to support the suggestion of sub-groups:

Dankaerts W, O'Sullivan P, Burnett A, Straker L. Altered patterns of   superficial trunk muscle activation during sitting in nonspecific chronic low   back pain patients: importance of subclassification. Spine (Phila Pa 1976).   2006;31:2017-23. https://doi.org/10.1097/01.brs.0000228728.11076.82.

The final   sentence of this paragraph (which has not been altered) indicates the need   for further research in this area.

15

Discussion: Line 184 Given that previous   studies of 2 hours duration revealed similar results, do you feel that a 1   hour task is long enough to change symptoms? This could be commented on in   the discussion.

The   limitation of using a 1-hour standing task has been identified and the   selection of this duration has been justified in the limitations section of   the discussion.

Added   text: “Third, the prolonged standing task in this study was 1-hour; whereas,   many other studies have used a 2-hour standing task. This duration was chosen   as the majority healthy participants in our previous study developed LBP   within the first hour of the standing task, and guidelines for standing at   work do not recommend people stand for more than one hour at a time.”

The   following references have been added to support this justification:

Johnston V, Gane EM, Brown W, et al. Feasibility and impact of sit-stand   workstations with and without exercise in office workers at risk of low back   pain: A pilot comparative effectiveness trial. Appl Ergon.   2019;76:82-89. https://doi.org/10.1016/j.apergo.2018.12.006.

Waters, T.   R., and R. B. Dick. 2015. "Evidence of Health Risks Associated with   Prolonged Standing at Work and Intervention Effectiveness." Rehabilitation Nursing 40 (3). doi:   10.1002/rnj.166.

16

Discussion:    Lines 208-211 These are indeed the key   limitations, and they could be expanded upon. Given that the pain severity   rating was around 2-3 at baseline, then for any clinically meaningful change   to happen, there would need to be almost complete pain resolution in most   cases, which is perhaps unrealistic. Future studies may best recruit   participants with higher baseline scores.

The   limitation of low baseline pain levels in our study has been emphasised in   the limitations and we have suggested that future studies recruit individuals   with higher baseline pain levels.

The edited   text is underlined in the following sentence in the discussion:

“Second,   the small sample size (n=16), low level of baseline pain and high   level of functioning (mean Oswestry score of 5.1/50) of our study   participants may require replication of this study with larger samples and individuals   with higher baseline pain levels and lower functioning to confirm the   generalisability of our findings.”

17

Conclusions: Please address the conclusions   to provide a more balanced reflection of the study’s results as per the   comments above.

We have   added text to the conclusion to indicate that some individuals experience   worsening of their LBP in both the footstool and no footstool conditions.

The edited   text is underlined in the following sentence: “This study identified that   most office workers with a history of LBP are able to use a standing   workstation without significant exacerbation of symptoms; however, a   proportion of individuals experience a clinical significant increase in pain irrespective   of using or not using a footstool.”

1

This paper compares the development of pain in office workers with   mild backpain during working at a standing workplace for one hour with or   without using a footstool. Generally, this manuscript is well written.   However, there are two main concerns:

Firstly, how valid is the NRS in measuring the intensity of   spontaneous back pain at intervals of 10 minutes? The provided reference 17   reported reliability of NRS for measurement intervals of 1 and 4 weeks.   Reference 18 stated that single NRS values recorded at hourly intervals show   high variation between consecutive days, but their average value was stable   and reproducible. The pain provoked by prolonged standing might be more   reliable described by subtracting the mean of the measurements at 10 to 60   minutes from baseline score.

The NRS   has been used to record changes in pain at 10-15 minute intervals during a   prolonged standing tasks in a number of previous studies.

The   following sentence has been added to the methods to indicate this:

“The 11-point   NRS has been used to record LBP at 10-15 minute intervals in numerous previous   studies using a prolonged standing task.”

And the   following references to support this statement have been added:

1.      Sorensen, C.J.; Johnson, M.B.; Callaghan, J.P.; George, S.Z.;   Van Dillen, L.R. Validity of a Paradigm for Low Back Pain Symptom Development   During Prolonged Standing, The Clinical journal of pain 2015, 31, 652-659.

2.      Nelson-Wong, E., Gregory, D.E., Winter, D.A., Callaghan, J.P.,   2008. Gluteus medius muscle activation patterns as a predictor of low back   pain during standing. Clin. Biomech. 23, 545–553.

3.      Nelson-Wong, E., Callaghan, J.P., 2010a. Changes in muscle   activation patterns and subjective low back pain ratings during prolonged   standing in response to an exercise intervention. J. Electromyogr. Kinesiol.   20, 1125–1133.

4.      Johnston V, Gane EM, Brown W, et al. Feasibility and impact of   sit-stand workstations with and without exercise in office workers at risk of   low back pain: A pilot comparative effectiveness trial. Appl Ergon. 2019;   76:82-89.

2

How was this pain recording organised? Did the study instructor   verbally remind the participants every 10 minutes to score their back pain?   Was a computer programme used for that purpose? Was an acoustic signal used   to indicate that the next back pain assessment is due? Who recorded the pain score:   the participants in writing on paper or their computer or the study   instructor? How did pain scoring interfere with their computer work?

Pain   recording was undertaken by a study investigator on a paper data collection   sheet. The investigator verbally asked the participant to rate their pain in   the NRS. If the indicated the presence of any pain (by stating a number   greater than 0), a paper body chart that was placed on the desk in front of   them for them to mark their location of pain. The body chart was removed as   soon as it was completed so it did not interfere with the participants work.  

The following text has been added to   the methods to explain this:

“The investigator verbally asked the participant to rate their   pain on a NRS. If the participate rated their pain >0 (indicating the   presence of pain), they were asked to mark the location of pain on a paper   body chart that was placed in front of them.”

3

Secondly, it remains unclear where the study was conducted? On   site, where the participants were employed as office workers? In a laboratory   of the University?  How many of the participants were self-employed? If   the study was conducted in the workers office, did it interfere with their   working duties, was it scheduled at the beginning, the middle or the end of   their working day?  Did the participants perform their routine computer   work? Were the computer tasks compatible with interruptions every 10 minutes?

The study   was conducted in a research laboratory at a University. The laboratory   contained a private room with a sit-stand workstation. All participants were   University employees and they undertook   their usual computer work during the study. The study was scheduled to   accommodate the participant’s work schedule.

The following text has been added to   describe where the study was conducted:

“The   1-hour testing sessions were conducted in a private room in a laboratory at   the participants’ workplace. »

The   following text in the methods section explains that the particpants completed   their usual computer work during the study :

“Participants completed a prolonged standing task [15] while   performing their usual computer-based work duties for an hour under two   different conditions….”

4

If conducted at the university, how long were participants off   duty on the date of the experiment?  All this information is essential   for understanding, whether the use of a standing workstation is a valid   alternative in the real working life of patients with mild back pain.

Participants   performed their regular computer work during the experimental session.  Thus, participants were not “off duty” to   attend the experiment. Further, as the testing occurred at in a laboratory at   the participants’ workplace, interruption to attend testing was minimal.

This has   been address with the added text described in the comment above.

5

Student´s t-test requires a normal distribution of data. Was   this checked?

Thank you   for this comment and drawing our attention to this. Data were not normally   distributed and as such Wilcoxon Paired Signed Rank tests were used instead   of Paired T-tests and data are presented as median and inter-quartile range   instead of mean and standard deviation.

The   following text has been added to the methods:

“As data   were not normally distributed, Wilcoxon Paired Signed Rank tests were used to   compare changes in pain severity between conditions and data are presented as   median and inter-quartile range (IQR).”

The   presentation of data and= p-values have been changed in the results to   represent this change in statistical analyses and presentation of data.

6

Whilst pain worsening is clearly defined as increase of NRS   score equal or greater than 2 points, a definition of improvement (= pain   reduction?) is missing. 

The text   in the methods has been modified to indicate that participants who report   their pain to be 2 points lower on the NRS at the end of the standing tasks is   also reported.

Edited   text: “The number (percentage) of participants who reported their pain to   be 2 points higher or lower on the NRS at the end of the standing   tasks (which suggests a clinically meaningful change in pain [20]) are   reported.”

The following text has been added to the results to present data for   the number of participants who reported an improvement in pain of ³2 points on the numerical   rating scale.

Text added   to the results: “Two out of 16 participants (13%) reported an improvement in   pain of ³2 points on the NRS for each of   the conditions.”

7

The body mass index should be added to anthropometric   characteristics of participants.

Body mass index has been added to the anthropometric   characteristics of participants.

1

The manuscript has improved, and all points criticised in the   previous version have been addressed. The setup of the experiment is now   clear, and the method of the statistical analysis appropriate. The course of   pain and their variation between individuals is better described.

Thank you   for your feedback and assistance in improving this paper. 

3

However, the exchange of mean and standard deviation with median   and interquartile range is incompletely performed, upper and lower quartile   should be separated by a semicolon. The text in  lines 148-152, page 4   must be corrected accordingly “There was   no significant difference in the change in maximal   severity of LBP pain experienced throughout the prolonged standing task   between the footstool (mean median (SDIQR):   1.3 (1.71.5)) and no footstool (median (IQR): mean   (SD): 1.3 (1.72)) conditions (p=0.8068). The change in the   severity of LBP experienced by participants from the start to the end of the   60-minutes standing task was also similar between conditions (median (IQR) footstool:   0 .1 (2.10.75),;   no footstool: 0.81 (2.32); p=0.2614).

The presentation of inter-quartile   range (IQR) has been changed to present the upper and lower (25% and 75%)   quartiles, rather than the range between these quartiles, as suggested by the   reviewer.

Previous   text: “There was   no significant difference in the change in maximal severity of LBP pain   experienced throughout the prolonged standing task between the footstool   (median (IQR): 1 (1.5)) and no footstool (median (IQR): 1 (2))   conditions (p=0.68). The change in the severity of LBP experienced by   participants from the start to the end of the 60-minutes standing task was   also similar between conditions (median (IQR) footstool: 0 (0.75), no   footstool: 1 (2); p=0.14).”

Revised text: “There was no significant difference in the change   in maximal severity of LBP pain experienced throughout the prolonged standing   task between the footstool (median (IQR): 1 (0; 1.25)) and no   footstool (median (IQR): 1 (0; 2)) conditions (p=0.68). The change in   the severity of LBP experienced by participants from the start to the end of   the 60-minutes standing task was also similar between conditions (median   (IQR) footstool: 0 (-0.13; 0.25), no footstool: 1 (0; 2);   p=0.14).”

4

Missing punctuation marks (line 37), word spacing (lines 38, 52,   58) and spelling (upper case after comma, line 41; “should read “compare”   instead of “compared”, line 64) must be corrected. Some extra words remaining   after amending the first version such as “using   paired t-tests” page 3, line121, or “Standing using a” page 5, line122, must finally be   deleted.

These   corrections have been made where possible in the paper. The extra words from   the previous revision were not observed to be present.

5

On page 5, lines 213-214, the sentence “Moving the foot onto and   off of a footstool would be expected to mayfacilitate increased movement and   having the foot position on or off of the footstool would may add   variability to sagittal plane posture” requires rephrasing.

These   corrections have been made.

6