Diastasis of the rectus abdominis muscle (DRAM) affects a significant number of women during the antenatal and postnatal period [1], and has been related to reduced pelvic control [2] and potential pelvic and low back pain. Historically, DRAM measurement has been done using palpation or the ‘finger method’ to discriminate between women with and without DRAM, i.e., diagnose. Also calipers, tape measures, and more recently ultrasound are used to diagnose DRAM or monitor DRAM width [3]. Barbosa et al. [4] recently published a clinically important comparative measurement study in ‘‘Archives of Gynecology and Obstetrics’’. The authors compared measurements on ultrasound images with those from calipers. The two main aims were (1) to evaluate the measurement error between the two methods, and (2) to find the discriminative agreement or diagnostic value of calipers. The second objective is of great clinical importance for daily practice. Based on these outcomes, clinical decisions, management and follow-up might change. However, the authors interpreted the kappa statistic incorrectly, undervaluing the ability of calipers to identify women with DRAM. They state that ‘‘according to the Kappa test, 65 % of the diagnoses given in the clinical examination were confirmed by ultrasonography’’ (Epub p.3). Although the chance-corrected agreement reflected by the kappa coefficient is indeed 0.66 (r = 0.656), the percentage agreement, and hence proportion ‘confirmed by ultrasonography’ is 83 %. Table 1 provides additional diagnostic accuracy results, based on raw data presented by Barbosa et al. [4]. Keeler et al. [3] found in their survey that 96.6 % of women’s health physiotherapists used palpation to assess, and measure DRAM width. In 1987, Bursch [5] concluded that this method is ‘unreliable’, but used inappropriate statistics to evaluate this [6], and might not have sufficiently considered the method used for their purpose of measurement. It is important to think about the specific purpose of measurement before deciding which measurement method to use: do you want to diagnose (presence/ absence) or rank (small/moderate/large)? Do you want to monitor or evaluate DRAM width over time (using difference in cm or mm), or simply report on DRAM width of a patient? We would like to emphasise the importance of the purpose of clinical measurement and the importance of psychometric and diagnostic studies such as those of Barbosa et al. [4]. Currently, insufficient psychometric information is available for widely used measurement methods such as calipers or palpation. Additional psychometric studies are needed. Psychometric requirements differ depending on the measurement purpose. If discrimination between absence and presence of DRAM is the aim, using, for example, palpation or a tape measure, the positive predictive value or sensitivity is of more importance than absolute measures of reliability. However, when evaluating DRAM width as treatment outcome, small measurement error is desired for which ultrasound [7] or calipers [8] might be more suitable. In summary, the choice of the DRAM measurement method that is clinically feasible for a specific clinical A. T. M. van de Water (&) Musculoskeletal Research Centre and Department of Physiotherapy, School of Allied Health, La Trobe University, Bundoora, VIC 3086, Australia e-mail: a.vandewater@latrobe.edu.au
Read full abstract
Aug 10, 2014
Theodore Miclau + 19
Open Access
