Outcome prediction for plantar heel pain
Abstract
Introduction: Plantar Heel pain (PHP) is a common, problematic disorder with unsatisfactory treatment outcomes. The physical impairments associated with PHP are commonly documented but there is insufficient data to develop multi-variable biopsychosocial models that explain presentation, differences to other conditions, severity or prognosis. In this thesis, I investigated the demographic characteristics and biopsychosocial factors in persons with PHP to develop a prediction model for recovery. Methods: There were four main parts to the thesis. A systematic review explored the prognostic factors for recovery or successful treatment of plantar heel pain. A feasibility study was performed to investigate the feasibility of data collection procedures and establish equivalence to usual procedures for the questionnaire battery. A case-control study was conducted to identify associated factors of PHP severity. Finally, a prospective cohort study was implemented to develop a prediction model for recovery of PHP. Results: There are limited biomedical factors which can be used to predict PHP outcome, with a notable absence of high quality prospective cohort studies that consider multiple variables including psychosocial or psychological factors. Questionnaire administration by our online method was valid and reliable. The case control study showed the factors associated with foot health severity were overall quality of life (QoL) (β=0.35; p<0.001), education level (β=−0.22; p=0.003), sex (β=−0.20; p=0.007), morning pain duration (β=−0.18; p=0.01) and disease duration (β=−0.15; p=0.04) in the context of a comprehensive model. The cohort study revealed the risk of still having PHP was 52.5% after 1 year. People who have better general foot health, a shorter symptom duration and have had an injection at any time have a higher chance of recovery. The model provided accurate prediction of the overall recovery (C-statistic 0.68; 95% CI 0.66 to 0.79) for PHP with acceptable discrimination and calibration. Conclusion: There are assumptions in the literature that prognostic factors for plantar heel pain recovery are mainly physical. My online questionnaire considering a wide range of biopsychosocial variables was valid for remote monitoring of patients for clinical and research purposes, including the cohort study. The developmental models showed severity and recovery are not just determined by physical features of the presentation. Patients presenting with PHP of long duration who score worse on the foot health of FHSQ have a poorer prognosis, irrespective of age, sex and other demographic variables. My results suggest that strategies aimed at preventing chronicity of more severe PHP may optimise prognosis.
Authors
Gulle, HalimeCollections
- Theses [4213]