X‐ray fluorescence (XRF) is a widely used method for in vivo elemental analysis. Particularly for bone, it is a non‐invasive technique that provides information on composition without significant risk to the patient. XRF contributes a capability for measuring elements beneficial to human health, such as strontium. This is a proposed supplement that has been shown in clinical trials to reduce fracture risk in people diagnosed with osteoporosis. Although XRF is a viable method for quantifying bone strontium, there are still factors that constrain its effectiveness. X‐ray attenuation through overlying soft tissue decreases the signal, consequently requiring correction before estimating the true concentration of strontium in bone. A correction factor can be applied to account for the reduced signal, but an accurate measurement of overlying soft tissue thickness is required. It has been shown that using the correlation between Compton peak count rate and overlying thickness can be used as an estimation of overlying tissue. Lucite is commonly used as a soft tissue substitute; however, its mean atomic number is appreciably lower than soft tissue, somewhat limiting its applicability. This study tests the feasibility of using cellulose filter papers as a substitute for overlying soft tissue to perform XRF analysis of strontium‐doped hydroxyapatite bone phantoms. Mass attenuation coefficients are shown to be closer to those of soft tissue (International Commission on Radiation Units' four‐component) than Lucite, and the Compton correlation is used to estimate thickness as a correction factor to quantify true strontium concentration.