This Article Full Text (PDF) All Versions of this Article: 0272989X08315254v1 28/6/810    most recent References Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to Saved Citations Download to citation manager Request Permissions Request Reprints Add to My Marked Citations Citing Articles Citing Articles via Google Scholar Citing Articles via Scopus Google Scholar Articles by Fischer, T. Articles by Hummers-Pradier, E. Search for Related Content PubMed PubMed Citation Articles by Fischer, T. Articles by Hummers-Pradier, E. Social Bookmarking                         What's this?

Family Practitioners' Diagnostic Decision-Making Processes Regarding Patients with Respiratory Tract Infections: An Observational Study

Department of General Practice/ Family Medicine, Georg-August University Göttingen, Göttingen, Germany, tfische{at}gwdg.de

Susanne Fischer, MD

Department of General Practice/ Family Medicine, Georg-August University Göttingen, Göttingen, Germany

Wolfgang Himmel, PhD

Department of General Practice/ Family Medicine, Georg-August University Göttingen, Göttingen, Germany

Michael M. Kochen, MD, MPH, PhD, FRCGP

Department of General Practice/ Family Medicine, Georg-August University Göttingen, Göttingen, Germany

Eva Hummers-Pradier, MD, PhD

Hannover Medical School, Hannover, Germany

Background. The influence of patient characteristics on family practitioners' (FPs') diagnostic decision making has mainly been investigated using indirect methods such as vignettes or questionnaires. Direct observation—borrowed from social and cultural anthropology—may be an alternative method for describing FPs' real-life behavior and may help in gaining insight into how FPs diagnose respiratory tract infections, which are frequent in primary care. Objective. To clarify FPs' diagnostic processes when treating patients suffering from symptoms of respiratory tract infection. Methods. This direct observation study was performed in 30 family practices using a checklist for patient complaints, history taking, physical examination, and diagnoses. The influence of patients' symptoms and complaints on the FPs' physical examination and diagnosis was calculated by logistic regression analyses. Dummy variables based on combinations of symptoms and complaints were constructed and tested against saturated (full) and backward regression models. Results. In total, 273 patients (median age 37 years, 51% women) were included. The median number of symptoms described was 4 per patient, and most information was provided at the patients' own initiative. Multiple logistic regression analysis showed a strong association between patients' complaints and the physical examination. Frequent diagnoses were upper respiratory tract infection (URTI)/common cold (43%), bronchitis (26%), sinusitis (12%), and tonsillitis (11%). There were no significant statistical differences between ``simple heuristic'' models and saturated regression models in the diagnoses of bronchitis, sinusitis, and tonsillitis, indicating that simple heuristics are probably used by the FPs, whereas ``URTI/common cold'' was better explained by the full model. Conclusion. FPs tended to make their diagnosis based on a few patient symptoms and a limited physical examination. Simple heuristic models were almost as powerful in explaining most diagnoses as saturated models. Direct observation allowed for the study of decision making under real conditions, yielding both quantitative data and ``qualitative'' information about the FPs' performance. It is important for investigators to be aware of the specific disadvantages of the method (e.g., a possible observer effect).

Key Words: direct observation study • respiratory tract infection • diagnostic strategy • clinical decision making • simple heuristics.

This version was published on November 1, 2008

Medical Decision Making, Vol. 28, No. 6, 810-818 (2008)
DOI: 10.1177/0272989X08315254


CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    Twitter    What's this?