Background Plantar heel pain (PHP) is a debilitating musculoskeletal condition from which only 65–75% recover within a year. Treatment has unsatisfactory results, partially due to poor understanding of PHP prognosis.

Methods We carried out a prospective cohort study with 1 year follow-up. 136 patients were recruited from five different countries using a validated and reliable online questionnaire battery that included 5-demographic, 13-biomedical, 8-psychological, 3-social and 8-activity-related factors. Kaplan-Meier survival estimates were used to estimate prognosis, and a multiple Cox-regression was used for the prediction model.

Results The survival analyses revealed that risk of having PHP was 97.1% after 3 months, 80.8% after 6 months and 52.5% 1 year from study entry. The risk was significantly greater for those with poor general foot health (HR: 1.02, 95% CI= 1.00–1.02), longer symptom duration (HR: 0.84, 95% CI= 0.64–0.91), not-having injection (HR: 2.61, 95% CI= 1.29–8.96) and believing to get worse in the future (HR: 1.69, 95% CI= 1.35–3.47) at baseline. The model provided accurate prediction of overall recovery (C-statistic 0.68; 95% CI 0.66–0.79) for PHP with acceptable discrimination and calibration.

Conclusion Patients with long duration symptoms who have poor foot health and negative future beliefs have a poorer prognosis, irrespective of demographic variables. When recommending interventions to patients, clinicians should consider patients’ conditions from broad psychosocial and biomedical perspectives. Importantly, PHP is not a self-limiting condition as previously claimed. Strategies designed to prevent chronicity may optimise prognosis. This is the first study to provide robust prognostic evidence about biopsychosocial factors for PHP.