Defining the rehabilitation adherence curve and adherence phases of stroke patients: an observational study

BACKGROUND Patient adherence is a crucial determinant of rehabilitation in the long term after stroke. However, adherence is inconstant and fluctuates along a time course, and the underlying regular pattern of adherence variation remains to be clarified. OBJECTIVE We aimed to describe the longitudinal pattern of adherence to rehabilitation exercises in stroke patients and to determine different adherence phases based on formulated rehabilitation adherence curve. PATIENTS AND METHODS Rehabilitation adherence levels were prospectively collected using the Questionnaire of Exercise Adherence (EAQ) among patients diagnosed with first-onset stroke since the second week of stroke onset, with a follow-up of 24 weeks. SPSS19.0 was used to formulate a fitting curve based on a scatter diagram. Possible causal factors for the different adherence phases are also discussed from the psychological, socioeconomic, and behavioral aspects. RESULTS A total of 98 patients were included in this study. General adherence of the included subjects was classified as low to medium during follow-up. The adherence fitting curve was an "S" curve, with the fitting function y =0.005x3-0.211x2+1.963x+52.345. Three phases, namely, rapid increase phase, slow decrease phase, and stable phase, were identified based on the adherence curve, and relevant theories are explored. CONCLUSION Rehabilitation adherence of stroke patients is a dynamic behavioral process that continuously changes along a time course, with a regular pattern of an "S" curve and includes a rapid increase phase, a slow decrease phase, and a stable phase.