Examining the variation in GPs’ referral practice: a cross-sectional study of GPs’ reasons for referral

Abstract

BACKGROUND: There is a large variation in referral rates to secondary care among GPs, which is partly unexplained. AIM: To explore associations between reasons for referral to secondary care and patient, GP, and healthcare characteristics. DESIGN AND SETTING: A cross-sectional study in Northern Norway. METHOD: Data were derived from 44 (42%) of 104 randomly selected GPs between 2008 and 2010. GPs scored the relevance of nine predefined reasons for 595 referrals from 4350 consecutive consultations on a four-level categorical scale. Associations were examined by multivariable ordered and multivariable multilevel logistic regression analyses. RESULTS: Medical necessity was assessed as a relevant reason in 93% of the referrals, 43.7% by patient preference, 27.5% to avoid overlooking anything, and 14.6% to reassure the patient. The higher the referral rates, the more frequently the GPs referred to avoid overlooking anything. Female GPs referred to reassure the patient and due to perceived deficient medical knowledge significantly more often than male GPs. However, perceived easy accessibility of specialists was significantly less frequently given as a reason for referral by female GPs compared with male GPs. When the GPs scored the referrals to be of lesser medical necessity, male GPs referred significantly more frequently than female GPs to reassure the patient due to patient preference and perceived deficient medical knowledge. CONCLUSION: There are striking differences in reasons for referral between Norwegian male and female GPs and between GPs with high and low referral rates, which reflects difficulties in handling professional uncertainty. Referring to reassure the patients, especially when referrals are less medically necessary, may reflect consideration and acquiescence towards the patients.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 70 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.