Variation of the apparent viscosity of thickened drinks.

Background: In dysphagia care, thickening powders are widely added to drinks to slow their flow speed by increasing their viscosity. Current practice relies on subjective evaluation of viscosity using verbal descriptors. Several brands of thickener are available, with differences in constituent ingredients and instructions for use. Some thickened fluids have previously been shown to exhibit time-varying non-Newtonian flow behaviour, which may complicate attempts at subjective viscosity judgement.

Aims: The aims were to quantify the apparent viscosity over time produced by thickeners having a range of constituent ingredients, and to relate the results to clinical practice.

Methods & Procedures: A comparative evaluation of currently available thickener products, including two which have recently been reformulated, was performed. Their subjective compliance to the National Descriptors standards was assessed, and their apparent viscosity was measured using a rheometer at shear rates representative of situations from slow tipping in a beaker (0.1 s⁻¹) to a fast swallow (100 s⁻¹). Testing was performed repeatedly up to 3 h from mixing.

Outcomes & Results: When mixed with water, it was found that most products compared well with subjective National Descriptors at three thickness levels. The fluids were all highly non-Newtonian; their apparent viscosity was strongly dependent on the rate of testing, typically decreasing by a factor of almost 100 as shear rate increased. All fluids showed some change in viscosity with time from mixing; this varied between products from -34% to 37% in the tests. This magnitude was less than the difference between thickness levels specified by the National Descriptors.

Conclusions & Implications: The apparent viscosity of thickened fluids depends strongly on the shear rate at which it is examined. This inherent behaviour is likely to hinder subjective evaluation of viscosity. If quantitative measures of viscosity are required (for example, for standardization purposes), they must therefore be qualified with information of the test conditions.