Letter to the Editors
Br J Sports Med
J Obstet Gynecol Canada
Oslo, Nov 23rd 2018
Comment and questions to Mottola et al (2019): 2019 Canadian guideline for physical activity throughout pregnancy
We have read the Canadian guideline for physical activity throughout pregnancy with great interest. We note that the guideline team have made their recommendation regarding pelvic floor muscle training (PFMT) based on evidence from a systematic review from the same research group (Davenport et al 2018). The main results of this review are in line with the latest Cochrane review (Woodley et al 2017) on the same topic; while there are some methodological differences and variations in which studies were included or not (two of the largest studies on PFMT was left out from the Davenport review; Mørkved et al 2003 and Stafne et al 2012), the findings in terms of size and precision of effect are similar, although Davenport et al used odds ratio and Woodley et al used risk ratio for their summary statistic. Davenport et al reported that PFMT gave a 50% reduction in prenatal UI and a 35% reduction in postnatal UI, but the guideline team concluded a “weak recommendation” for PFMT because UI was not rated as a "critical outcome" and the evidence was of "low quality". We find this conclusion at odds with the evidence and the interpretation of the evidence based on the guideline team’s own criteria.
The Canadian guideline grades evidence as either “strong or weak based on the (1) balance between benefits and harms, (2) overall quality of the evidence, (3) importance of outcomes (ie, values and preferences of pregnant women), (4) use of resources (ie, cost), (5) impact on health equity, (6) feasibility and (7) acceptability. A strong recommendation is one where “Most or all pregnant women will be best served by the recommended course of action” and a weak recommendation is one where “Not all pregnant women will be best served by the recommended course of action; there is a need to consider other factors such as the individual’s circumstances, preferences, values, resources available or setting. Consultation with an obstetric care provider may assist in decision-making.”
We disagree with how PFMT has been classified in relation to these criteria and would like to question and comment on the following:
1.Balance between benefits and harms: the effect size of antenatal PFMT for prevention of UI is moderate and there are no harms of PFMT, so this would be in favor of a strong recommendation. In addition, many of the studies in the Davenport et al (2018) review did not compare PFMT with no exercise/untreated controls, rather the control groups typically also had some advice or instruction in PFMT as part of ‘routine’ care. Such trials are likely to under- rather than overestimate the effect of PFMT, suggesting the true effect size may be larger than that calculated by Davenport et al (2018).
2.Overall quality of the evidence: There are three issues to consider here:
a. Research design. There are sufficient numbers of RCTs evaluating effect of PFMT on UI during pregnancy to do meta-analyses without including cohort studies with lower internal validity. The Davenport review (2018) referred to cohort studies and showed that general exercise (not PFMT) may increase the odds of developing UI. Aerobic exercise usually includes high impact activities (jumping and running). Numerous studies (Bø 2004, Nygaard & Shaw 2016) have shown that high impact activities are associated with UI, therefore this combination is likely to be provocative of UI and mask a stand-alone effect of PFMT on reduction of UI. Studies on aerobic exercise can therefore not be expected to have a positive effect on UI and should not be recommended to be combined with studies on PFMT. In all reports in this area general exercise /physical activity needs to be separated from specific PFMT, in order not to confuse the readers.
b. Risk of bias. All the trials are inevitably at risk of bias through an inability to blind participants and providers, yet this would be the same for all forms of physical activity (yoga) and should not affect the rating of PFMT more than other exercise. The included trials are also the usual ‘mixed bag’ of less and more robust trials. While most studies are small to moderate in size it seems likely that the true underlying effect is within the existing confidence limits of the effect estimate (Herbison et al 2011). Both in the Cochrane (Woodley et al 2017) and the Davenport et al (2018) reviews the upper limits of the confidence intervals suggest clinically important reduction in UI.
c. Statistical heterogeneity. For PFMT, a plausible explanation for statistical heterogeneity is the different training doses and supervision (Hay-Smith et al 2011). We agree that more work is needed to find a ‘cut off’ for effectiveness in PFMT delivery and dose, but in the meantime there are certainly robust trials with well described interventions demonstrating clinically significant effect that are suitable models for application in practice.
3. Importance of outcome: UI is a prevalent (>30%) and bothersome condition reducing QoL and especially participation in physical activity (Nygaard et al 2005, Hamid et al 2015), and therefore important to prevent. In the Canadian guideline it is stated that prior to convening the panel, "10 pregnant women were recruited by convenience sampling and invited to provide input on the perceived benefits and harms of physical activity, and to identify pregnancy outcomes that were most important to them." Studies consistently finds women perceive UI as stigmatizing and at the same time ‘normal’ for parous women, and UI is a topic they are reluctant to talk about (Hamid et al 2015). Postpartum the dominant view of women is that of ‘if only I’d known then what I know now’, and ‘I wish someone had told me about UI, and taught me how to do PFMT properly’ (Mason et al 2001, Mason et a 2001, Neels et al 2016). If the 10 pregnant women in the expert group were continent, not aware that they might develop incontinence after birth and were more concerned about other common maternity conditions or the health of the babies, UI may not have reached their attention. We are surprised that specialized women's health physiotherapists who are the experts in this field both in high quality research and clinical practice, were not included in the panel nor as experts.
4. Use of resources: PFMT is already part of ante-and postnatal health care in most developed countries. PFMT has proved to be effective as part of group training for women and can therefore be administered at low cost to the health system.
5. Impact of health equity: Not informing or providing PFMT to pregnant women creates inequity as failure to prevent UI in pregnancy means that women are potentially set up for many years of UI symptoms with all the consequent effects on self-esteem, withdrawal from physical activity, not playing with their children, the cost of buying products and laundry and the cost of physiotherapy and surgery.
6. Feasibility: PFMT has successfully been incorporated in comprehensive exercise classes since 1986 (Bø et al 1990, Mørkved et al 2003, Stafne et al 2012).
7. Acceptability: PFMT research is firmly on the side of acceptability. Studies show that women want to do PFMT as first line treatment, but they must be informed about why and how they should do it (Mason et al 2001, Mason et al 2001). The long-term effect of PFMT is, as for all exercise interventions, dependent on maintenance of training. There are challenges with long term adherence/attrition from all forms of exercise/physical activity programs, and this is NOT a specific nor more pronounced problem for PFMT. Again, this is not an argument for assigning PFMT a weak recommendation.
In summary, it appears the guideline panel has, perhaps in deciding on a weak recommendation, over-emphasized concerns about quality of evidence (in which other areas of exercise in medicine are there more RCTs showing clinically relevant effect?), and may not have ‘heard’ how bothered women are about the problem of UI. Most, or all, pregnant women would benefit from PFMT during pregnancy to prevent UI because: PFMT does prevent UI in late pregnancy, postpartum, and potentially for life (as well as preventing pelvic organ prolapse), it does no harm, women would do PFMT if they knew why it was important (but the system fails them by not giving them this information), women who do leak experience significant bother, and the training can be incorporated with other physical activity to maximize gains from time spent in exercise. The Canadian guidelines' weak recommendation appears inconsistent with the evidence and positive impact of existing research.
Strong recommendation for yoga?
We further question the evidence for the "strong recommendation" and "high quality evidence" that adding yoga and gentle stretching is beneficial. For which conditions is yoga beneficial during pregnancy? It would seem that yoga/ gentle stretching classes would indeed have the same cost and feasibility/ equity/ acceptability concerns as group training of the PFM.
Diastasis recti abdominis
Why is diastasis recti abdominis considered a critical outcome? There is no scientific evidence that this causes any harm. The guidelines refer to the systematic review of Davenport et al (2018 b). They conclude that there is no relationship between prenatal exercise and diastasis. However, the guideline states that continuing aerobic exercise (walking) is associated with less odds of development of diastasis. Based on which studies? Neither Sperstad et al (2016) nor Fernandes de Mota et al (15) found such associations.
The guideline recommends that women with a diastasis postpartum should avoid curl-ups and refer to Mota et al (2015) to support this statement. However, Mota et al (2015) found the contrary; abdominal crunch narrows the inter-rectal distance and indrawing opens the gap. This has now been supported by several studies.
In conclusion, we are concerned that how the Canadian guideline group have defined quality of evidence and weak and strong recommendations, may mislead pregnant women and heath care providers to believe that there is weak scientific evidence for prevention and treatment effects of PFMT for UI. This may discourage pregnant women from starting or continuing PFMT in a very important period to prevent and treat the condition. The guideline group may have - inadvertently – through issuing a weak recommendation for PFMT in pregnancy, put the onus onto individual health care professionals and women to make decisions about teaching or doing the exercises without regard to all the factual information. Women ‘don't know what they don't know’. Are we truly accepting that one third of women will experience stress urinary incontinence by mid-age when this could potentially be prevented through ante-natal PFMT? We urge the panel to re-consider their recommendation and are happy to supply any further evidence as required to guide the evidence grading.

Kari Bø, Professor, PhD. The Norwegian School of Sport Sciences, Dept of Sports Medicine, Oslo, Norway.
Chantale Dumoulin, Professor, PhD. University of Montreal, School of Rehabilitation, Montreal, Canada
Cristine HJ Ferreira, Associate Professor, PhD, University of Sao Paulo, Ribeirao Preto Medical School, Ribeirao Preto, Brazil
Helena Frawley, Associate Professor, PhD, Monash University, Dept of Physiotherapy, Melbourne, Australia
Jean Hay-Smith, Associate Professor, PhD, University of Otago, Wellington, New Zealand
Siv Mørkved, Professor, PhD, Norwegian University of Science and Technology, Dept of Public Health and Nursing, Trondheim, Norway
Ingrid Nygaard, Professor, MD, MS, University of Utah, School of Medicine, Salt Lake City, USA
Margaret Sherburn, PhD, Physiotherapy, School of Health Sciences, The University of Melbourne, Melbourne, Australia

References
Bø K. Urinary Incontinence, Pelvic Floor Dysfunction, Exercise and Sport. Sports Med 2004; 34: 451-464.
Bø K,Hagen RH, Kvarstein B, Jørgensen J, Larsen S. Pelvic floor muscle exercise for the
treatment of female stress urinary incontinence: III.Effects of two different degrees of pelvic floor muscle exercises. Neurourol Urodyn; 1990: 9,5:489-502.
Davenport MH, Nagpal TS, Mottola MF et al. Prenatal exercise (including but not limited to pelvic floor muscle training) and urinary incontinence during and following pregnancy; a systematic review and meta-analysis. Br J Sports Med 2018; 52: 1397-1404.
Davenport MH (B), Ruchat SM, Sobierajski F et al. Impact of prenatal exercise on maternal harms, labour and delivery outcomes: a systematic review and meta-analysis. Br J Sports Med. 2018 Oct 18. pii: bjsports-2018-099821. doi: 10.1136/bjsports-2018-099821. [Epub ahead of print]
Fernandes da Mota PG, Pascoal AG, Carita AI, Bø K. Prevalence and risk factors of diastasis recti abdominis from late pregnancy to 6 months postpartum, and relationship with lumbo-pelvic pain. Man Ther 2015; 20:200-2005
Hamid TA, Pakgohar M, Ibrahim R, Dstjerdi MV. "Stain in life": The meaning of urinary incontinence in the context of Muslim postmenopausal women through hermeneutic phenomenology. Archieves Geront Geriatr 2015; 60: 514-521.
Hay-Smith EJC, Herderschee R, Dumoulin C, Herbison GP. Comparisons of approaches to pelvic floor muscle training for urinary incontinence in women. Cochrane Database of Systematic Reviews 2011, Issue 12. Art. No.: CD009508. DOI: 10.1002/ 14651858. CD009508.
Herbison P, Hay- Smith J, Gillespie WJ. Meta-analyses of small numbers of trials often agree with longer-term results. J Clin Epidem 2011; 64: 245-153.
Mason L, Glenn S, Walton I Hughes C. Do women practice pelvic floor exercises during
pregnancy or following delivery? Physiotherapy 2001, 87; 662-670.
Mason L, Glenn S, Walton I, Huges C. The instruction in pelvic floor exercises provided to women during pregnancy or following delivery. Midwifery 2001; 17: 55-64.
Mota PI, Pascoal AG, Carita AI, Bø K. The Immediate Effects on Inter-rectus Distance of Abdominal Crunch and Drawing-in Exercises During Pregnancy and the Postpartum Period. J Orthop Sports Phys Ther. 2015;45:781-8.
Mørkved S, Bø K, Schei B, Salvesen KA. Pelvic floor muscle training during pregnancy to prevent urinary incontinence: a single-blind randomized controlled trial. Obstet Gynecol. 2003;101: 313-9.
Neels H, Wiebren A,Tjalma A, Wyndaele, De Wachter S, Wyndaele M, Vermandel A. Knowledge of the pelvic floor in menopausal women and in peripartum women. J. Phys. Ther Sci. 2016; 28: 3020–3029.
Nygaard IE, Shaw JM. Physical activity and the pelvic floor. Am J Obstet Gynecol. 2016; 214: 164-171.
Nygaard I, Girts T, Fultz NH, Kinchen K, Pohl G, Sternfeld B. Is urinary incontinence a barrier to exercise in women? Obstet Gynecol. 2005;106: 307-14.
Sperstad JB, Tennfjord MK, Hilde G, Ellström-Engh M, Bø K. Diastasis recti abdominis during pregnancy and 12 months after childbirth: prevalence, risk factors and report of lumbopelvic pain. Br J Sports Med. 2016;50: 1092-1096.
Stafne SN, Salvesen KÅ, Romundstad PR, Torjusen IH, Mørkved S. Does regular exercise including pelvic floor muscle training prevent urinary and anal incontinence during pregnancy? A randomised controlled trial. BJOG. 2012 Sep;119(10):1270-80. doi: 10.1111/j.1471-0528.2012.03426.x. Epub 2012 Jul 17.
Woodley SJ, Boyle R, Cody JD, Mørkved S, Hay-Smith EJC. Pelvic floor muscle training for prevention and treatment of urinary and faecal incontinence in antenatal and postnatal women. Cochrane Database of Systematic Reviews 2017, Issue 12. Art. No.: CD007471. DOI: 10.1002/14651858.CD007471.pub3.

The editorial by Zadro and Colleagues' calls for caution in marketing physical therapy services, and focuses on a lack of high-quality evidence to support all current claims made for PT First. Of this, we 100% agree. The purpose of our response is to highlight considerations we feel may be beyond the research-based concentration outlined in the authors’ editorial.

First and foremost, effective marketing strategies are influenced by many factors and vary depending on targeted end users, policy makers, and payers. These factors account for variations in the delivery of medical care, payment models, and the role of enforcing organizations. In Australia, it has been reported that a majority of patients receive appropriate evidence-based care for challenging and costly conditions such as low back pain.[1] In countries such as the United States, which boasts high rates of unnecessary imaging,[2] and high percentages of opioid prescriptions as initial treatment choices for nonspecific low back pain (>50% of patients), care is less guideline based[3], and heavily influenced by direct-to-consumer marketing strategies. The United States is immersed in a situation in which many high-risk, low-value treatments are easier to obtain, with insurance policies that comprehensively cover low-value care earlier (opioid prescriptions and steroid injections[4]); whereas low-risk, high-value interventions, such as those available from physical therapy, often require more out-of-pocket costs and a referral from another care provider. Referrals to physical therapy remain overall very low (7-16.3%). We would argue that the PT First strategy as a direct to consumer marketing approach is necessary and that this viewpoint likely varies based on the unique and complex interactions of healthcare delivery that are different between each society.

Our second point involves the use of physical therapy services as a substitute for traditional methods that have not shown benefit. Recently, the Centers for Disease Control and Prevention (CDC) released guidelines that recommended nondrug approaches such as physical therapy over long-term or high-dosage use of addictive prescription painkillers.[5] They made this recommendation despite the current lack of high-quality research to support physical therapy; they did so because the current approach was not working. Although Zadro and Colleagues' state that early access to harmful or ineffective physical therapy treatments is unlikely to improve patient outcomes or the opioid crisis, harms associated with PT are lower than those associated with other common first-line choices. In addition to harm, it’s important to also consider other outcomes that are valuable to patients. Porter et al.[6] argue for the capture of outcomes such as prevention of inappropriate downstream care, burden of receiving care, and delay in receiving care, which are not routinely collected in clinical trials. Promising findings from the PT-First studies suggests that downstream use of opioids, injections, and surgeries decreases when PT is accessed early.

Our third point is bound around the concept of high-value care. As Traeger and colleagues have stated, “Most commentators accept that the solution is not simply to stop providing low-value care, but rather that high-value healthcare requires the replacement of inappropriate care with appropriate care”.[7] We argue that replacement of inappropriate care is the backbone of the PT First initiative. Whereas further work is needed to optimize “appropriate care”, inappropriate care options are now well established. Reducing first line high stakes, high-risk options should at a minimum be a priority. Strategies designed to change system level issues are a priority. Insurance companies are paying attention and considering alternate payment and compensation models. Mass media campaigns have been called for to address the overemphasis of imaging, medication, and surgery,[8] and we feel that the PT First approach is a step in the right direction.

In summary, we agree with the editorial that more research is needed to identify appropriate triaging for MSK pain. We argue that it is also needed to support a primary care first strategy, a surgeon first strategy, a chiropractor first strategy and essentially every other provider. We felt that the editorial points fell short in its assessment of the potential immediate benefits of a PT First approach. Yes, we must continue to improve and promote utilization of evidence-based treatments within physiotherapy but let’s optimize on the opportunities that are currently available and that are already known (e.g., reducing opioids, inappropriate imaging, and low-value surgical options). We do not feel that “PT First” marketing is anti-collaborative, “failing to be part of team-based care and wanting to create a hierarchy among providers” as the majority of PT-First studies took place in multi-disciplinary and collaborative care settings. The focus of “PT First” in marketing was intended to promote an alternative to opioids, injections, or advanced imaging as first-care choices. In other words, “choose PT first, before choosing other higher risk treatment strategies.”

References
1 Runciman WB, Hunt TD, Hannaford NA, et al. CareTrack: assessing the appropriateness of health care delivery in Australia. Med J Aust 2012;197:100–5.
2 Gidwani R, Sinnott P, Avoundjian T, et al. Inappropriate ordering of lumbar spine magnetic resonance imaging: are providers Choosing Wisely? Am J Manag Care 2016;22:e68–76.
3 Friedman BW, Chilstrom M, Bijur PE, et al. Diagnostic testing and treatment of low back pain in United States emergency departments: a national perspective. Spine 2010;35:E1406–11.
4 Lin DH, Jones CM, Compton WM, et al. Prescription Drug Coverage for Treatment of Low Back Pain Among US Medicaid, Medicare Advantage, and Commercial Insurers. JAMA Netw Open 2018;1:e180235–e180235.
5 CDC Guideline for Prescribing Opioids for Chronic Pain. 2018.https://www.cdc.gov/drugoverdose/prescribing/guideline.html (accessed 11 Nov 2018).
6 Porter ME, Larsson S, Lee TH. Standardizing Patient Outcomes Measurement. N Engl J Med 2016;374:504–6.
7 Traeger AC, Moynihan RN, Maher CG. Wise choices: making physiotherapy care more valuable. J Physiother 2017;63:63–5.
8 O’Keeffe M, Maher CG, Stanton TR, et al. Mass media campaigns are needed to counter misconceptions about back pain and promote higher value care. Br J Sports Med 2018;:bjsports – 2018–099691.

If it's not too late. The authors may wish to correct their referencing throughout the paper. I noticed that the 3rd paragraph in the Introduction provides references that do not support the statements made. e.g. the Biswas et al paper did NOT assess the impact of PROLONGED sitting. Evidence on bouts of sitting is still very unclear, and none of these interventions have shown is has a meaningful impact.

It is also unclear how the authors can make their statement about 30 min/day being “likely to be clinically meaningful” – when it probably depends on what the sitting was replaced with (i.e. with standing vs. movement, etc) and the isotemporal substitution paper that is cited to support this assertion is based on a ‘theoretical’ shift of sitting to light activity from a cross sectional study (with risk of reverse causation). This seems to be selling a story that really isn't there.

A more reasoned conclusion might be that VERY SMALL reductions in TOTAL sitting per day seem possible (a drop in the bucket?) with interventions that require significant resources (notably, not too dissimilar to PA interventions), but whether or not such shifts in sitting per se would make any meaningful difference for health outcomes/biomarkers remains very unclear. Not much good if there is no efficacy for outcomes. The reducing sitting story comes across as more of an hypothesis and 'feel good' story, but the evidence upon which it is based (in terms of efficacy) for health outcomes remains VERY shaky indeed (or mostly null) at present. More critical probing is needed, particularly if it begins to detract from messages around physical activity per se - which probably make any purported effects of sitting go away pretty quickly (particularly when measured objectively, unlike the Ekelund paper). There is already ample evidence that physical inactivity is bad for health, etc. Much of the research on sitting is simply a good sales exercise (good for getting grants) and is splitting hairs, when the goal is really just to get people to move more. We'd really welcome more critical and reflective discussion on this as a whole.