Exercise for pregnant women with low back and pelvic pain (LBPP)
Posted on 3rd March 2017 by Rachel Sutton
This blog is a critical appraisal ofÂ a prospective randomized controlled trial evaluating the efficacy of an exercise program for pregnant women with low back and pelvic pain (LBPP).
Pregnancy-related low back and pelvic pain (LBPP) are very common, with 45% of woman affected to some extent (Vermani et al. 2010).Â Concerns for potential harm to the foetus mean that pharmacological pain management may be limited, and with increasing pain there may beÂ some degree of functional disability. Prior to and during pregnancy, it has been found that the use of exercise programs can provide effective pain management for LBPP. However, the quality of the evidence available is lacking due to many limitations with the research (Pennick and Liddle 2013).
The study in questionÂ (Ozdemir et al. 2015) was carried out to provide a more robustÂ investigation into whetherÂ individualised exercise programs increase functional capacity and relieve LBPP in pregnant woman.
What was the study?
The was a well-designed randomized control trial (RCT). 209 participants were assessed for eligibility and 113 were either excluded or declined to participate. This left 96 participants who consented and met the eligibility criteria. The intervention included education, training in protection methods and a standardised exercise program on top of usual care. A control group received usual care.
Participants were randomized into one of two groups using sealed, opaque envelopes and a simple randomisation method. To assess the primary outcomes of study, the Visual Analogue Scale (VAS) was used to measure change in participants’ pain level and the Oswestry Disability Index (ODI) was used to determine participants’ functional status.
Exercise levels were tracked, withÂ participants required to keep exercise diaries. All data was collected through telephone calls once a week. AfterÂ 4 weeks, a face to face interview was carried out with each participant, with their results being compared to baseline.
This study found that over 4 weeks, using the exercise program, participants with LBPP had a statistically significant improvement (p< 0.05) in functional capacity and a statistically significant reduction in pain intensity compared to the control group.
For a study to have a focused question it must address the population, intervention, control group and outcome measures (Santos et al. 2007). The research question addressed in this study clearly addressed each of these aspects.
- To ensure the RCT had sufficient statistical power to detect a differenceÂ in the outcomes measured, it was calculated that a sample size of 48 participants for each group was needed, which the study met.*
- Having strict inclusion criteria is important for ensuring groups are similar before the start of a trial. It increases the validity of a study because, if there is an imbalance between the groups, a positive result may be observed as a result of the differing characteristics of the participants rather than a result of the different interventions being comparedÂ (Roberts and Torgerson 1999). However, in this study, the eligibilityÂ criteria did not take into account whether or not it was the participantsâ€™ first pregnancy. This means previous injuries or structural damage due to pregnancy are not taken into account. During pregnancy there may be a detrimental effect on the structure and function of the pelvic floor muscles which meansÂ the pain experienced by woman who have already been pregnant could be higher, and their functional ability could also be at a lower baseline (Nursing Times 2009).
- Randomization of participants was achievedÂ using opaque, sealed envelopes. Additionally, a nurse who was independent from study carried out this process to further minimize bias. Ideally, an electronic randomization method would be used, as this limits any chance of human bias. After randomization, the characteristics in two groups were similar, showing that the randomization was effective in creating two comparable groups. Due to study design, it was not possible to blind participants and researchers asÂ group allocation was obvious. This could create expectations regarding the respective interventions and therefore could potentially be one reason for variance between the groups.
- The researchers used outcome measures that were valid, reliable and appropriate for what they were wanting to measure. This in turn increases the validity and reliability of the results and allows for the study to be replicated.
The design and results of this study support the researchersâ€™ aim to provide stronger evidence for use of exercise programs to relieve severity of LBPP and to promote functional ability. The intervention group showed a statistically significant improvement in both primary outcomes up to 4 weeks afterÂ ongoing treatment.
However, this study is focused on the short term effects; further research is needed to testÂ whether exercise programs can help woman suffering with LBPPÂ in the long term. Additionally, replication in more heterogeneous samples of womenÂ is advisable, toÂ ensure greater external validity and generalizability.
ABDULATIF, M., MUKHTAR, A. and OBAYAH, G., 2015. Pitfalls in reporting sample size calculation in randomized controlled trials published in leading anaesthesia journals: A systematic review.Â British Journal of AnaesthesiaÂ [online]. June, vol. 115, no. 5, pp. 699â€“707 [viewed 4 January 2017]. Available from: http://bja.oxfordjournals.org/content/115/5/699.full
CRITICAL APPRAISAL SKILLS PROGRAMME (CASP)- MAKING SENSE OF EVIDENCE., 2013.Â CASP Randomised Controlled Trial ChecklistÂ [online]. [viewed 5 January 2017]. Available from: http://www.casp-uk.net/casp-tools-checklists
NURSING TIMES., 2009.Â Pregnancy and childbirth: The effects on pelvic floor musclesÂ [online]. Nursing Times [viewed 4 January 2017]. Available from: https://www.nursingtimes.net/clinical-archive/continence/pregnancy-and-childbirth-the-effects-on-pelvic-floor-muscles/1999381.article
OZDEMIR, S., BEBIS, H., ORTABAG, T. and ACIKEL, C., 2015. Evaluation of the efficacy of an exercise program for pregnant women with low back and pelvic pain: A prospective randomized controlled trial.Â Journal of Advanced NursingÂ [online]. March, vol. 71, no. 8, pp. 1926â€“1939 [viewed 5 January 2017]. Available from: http://onlinelibrary.wiley.com/doi/10.1111/jan.12659/abstract;jsessionid=AF7E8556F5ADCB67210A31CF4DB4C642.f04t03
PENNICK, V. and LIDDLE, S. D., 2013. Interventions for preventing and treating pelvic and back pain in pregnancy.Â Cochrane Database of Systematic ReviewsÂ [online]. August, no. 8, [viewed 21 December 2016]. Available from: http://uir.ulster.ac.uk/26553/1/LBPP_in_pregnancy_Cochrane_review_2013.pdf
ROBERTS, C. and TORGERSON, D. J., 1999. Understanding controlled trials: Baseline imbalance in randomised controlled trials.Â BMJÂ [online]. July, vol. 319, no. 7203, p. 185 [viewed 4 January 2017]. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1116277/
SANTOS, C. M. da C., PIMENTA, C. A. de M. and NOBRE, M. R. C., 2007. The PICO strategy for the research question construction and evidence search.Â Revista Latino-Americana de EnfermagemÂ [online]. June, vol. 15, no. 3, pp. 508â€“511 [viewed 28 December 2016]. Available from: http://resolver.ebscohost.com/openurl?sid=EBSCO%3arzh&genre=article&issn=15188345&ISBN=&volume=15&issue=3&date=20070501&spage=508&pages=508-511
UNDERSTANDING HEALTH RESEARCH., 2016.Â Review a studyÂ [online]. [viewed 5 January 2017]. Available from: http://www.understandinghealthresearch.org/
VERMANI, E., MITTAL, R. and WEEKS, A., 2010. Pelvic girdle pain and low back pain in pregnancy: A review.Â Pain PracticeÂ [online]. January, vol. 10, no. 1, pp. 60â€“71 [viewed 13 December 2016]. Available from: http://www.salempainandspine.org/wp-content/uploads/2013/Pregnancy_Lowbackpain.pdf
*According to Abdulatif et al. (2015) there are four main components that are required for a sample size calculation. 1) The expected minimal clinically relevant difference between the groups; 2) the expected standard deviation of measurements for primary outcomes; 3) the desired power of the study and 4) the desired type 1 error probability.