PubMed Health⌕ Search

PubMed · 10694730

What do we mean by validating a prognostic model?

Abstract

Prognostic models are used in medicine for investigating patient outcome in relation to patient and disease characteristics. Such models do not always work well in practice, so it is widely recommended that they need to be validated. The idea of validating a prognostic model is generally taken to mean establishing that it works satisfactorily for patients other than those from whose data it was derived. In this paper we examine what is meant by validation and review why it is necessary. We consider how to validate a model and suggest that it is desirable to consider two rather different aspects - statistical and clinical validity - and examine some general approaches to validation. We illustrate the issues using several case studies.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

D G Altman, P Royston. 2000-02-29. What do we mean by validating a prognostic model?. https://doi.org/10.1002/(sici)1097-0258(20000229)19%3A4%3C453%3A%3Aaid-sim350%3E3.0.co%3B2-5. https://pubmed.health/record/pubmed%3A10694730/what-do-we-mean-by-validating-a-prognostic-model

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite PubMed Health by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related citations

Escitalopram in the long-term treatment of major depressive disorder in elderly patients.

AIM: The primary aim was to investigate the long-term safety and tolerability of escitalopram (10 or 20 mg/day) treatment of elderly patients suffering from major depressive disorder. The secondary aim was to examine response to treatment, as measured by change in the Montgomery-Asberg Depression Rating Scale (MADRS) total score from study entry to each visit, using observed cases. METHOD: This extension trial included 225 patients who had completed an 8-week, double blind, placebo-controlled lead-in study, which was performed in outpatients in primary care and in specialist clinics. The intent-to-treat population comprised 223 patients. RESULTS: The overall withdrawal rate was 24%. The most common reason for withdrawal was adverse events (9%). The 5 most common adverse events were accidental injury, rhinitis, weight increase, arthralgia and coughing, with an incidence ranging from 8 to 13%. No new types of adverse events were reported in this extension study compared to the 8-week lead-in study. The mean weight increased from 69.7 kg at study entry to 70.3 kg at endpoint. The percentage of patients in remission (MADRS total score < or = 12) increased from 48% at study entry to 72% by week 52. CONCLUSION: Escitalopram demonstrated a favourable tolerability profile during 52 weeks of open-label treatment of elderly patients, with further improvement in depressive symptoms.

Accidental Falls↗

MRI of rupture of the spring ligament complex with talo-cuboid impaction.

The spring ligament complex is essential for the stability of the longitudinal arch of the foot and includes the ligaments between the calcaneus and the talus at the superomedial to inferoplantar aspect of the foot. Tears of the spring ligament complex are most commonly degenerative in etiology and secondary to concomitant abnormality of the posterior tibial tendon. We report MRI findings in a 30-year-old man who presented with traumatic rupture of the spring ligament complex, seen following dislocation of the talonavicular joint. We also describe the previously unreported MRI features of talo-cuboid impaction secondary to disruption of the spring ligament complex.

Accidental Falls↗