Objectives Timely referral of patients with inflammatory arthritis (IA) and inflammatory low back pain to rheumatology is crucial. This study designed and validated a referral algorithm for non-rheumatologists in Arab countries and identified gaps in correct referrals and referral parameters. Methods Sixteen rheumatologists adopted referral algorithms, based on validation studies and clinical expertise, and summarized them in educational material (EM), including a PDF and video module. Five illustrative clinical vignettes were electronically sent to non-rheumatologists, to make referral decisions (phase 1). Two weeks later, the EM and the same vignettes were emailed again (phase 2). The primary outcome was the change in correct referral decisions using McNemar’s test. Improvement in correct referral parameters was evaluated and factors associated with correct referrals were analysed using logistic binary multivariable analysis. Results Eight hundred participants (general practitioners, family physicians, fellows, surgeons and others) from 13 countries completed phase 1 and 136 phase 2, with no demographic differences between them. Rates of phase 1 correct referral decisions were 72%, 72%, and 86% for referral-requiring cases, and 22% and 19% for others. In phase 2, correct referral decisions improved to 79%, 82%, and 88% for referral-requiring cases and 42% and 25% for others. Rates of most correct parameters improved significantly; factors associated with correct referrals included correct phase 1 referrals, duration of consultation, and fewer years practicing medicine. Conclusion Minor improvements in hypothetical referral decisions were observed after education, while significant improvements were noted for referral parameters. These results will help shape future referral campaigns’ design. <Table Float="No" ID="Taba"> <tgroup cols="2"> <colspec align="left" colname="c1" colnum="1" /> <colspec align="left" colname="c2" colnum="2" /> <tbody> <row> <entry align="left" nameend="c2" namest="c1"> Key points • Inflammatory arthritis and low back pain referrals to rheumatologists need boosting. • The referral algorithm improved the rate of correct referral decisions. • The referral algorithm improved the rate of correct parameters guiding the decision. </entry> </row> </tbody> </tgroup> </Table>

Design and validation of a referral algorithm for inflammatory arthritis and low back pain: identifying gaps in primary care practices

期刊论文
发表日期：

2025年11月19日

查看全文

Nelly Ziade,
Karen Mechleb,
Chafika Haouichat,
Fatemah Baron,
Nizar Abdulateef Al-Ani,
Asal Ridha Adnan,
Lina El Kibbi,
Manal Mashaleh,
Bassel Elzorkany,
Sherif M. Gamal,
Avin Maroof,
Manal El Rakaawi,
Fatima Alnaimat,
Mariam Erraoui,
Krystel Aouad,
Basel Masri,
Ihsane Hmamouchi

摘要

Objectives

Timely referral of patients with inflammatory arthritis (IA) and inflammatory low back pain to rheumatology is crucial. This study designed and validated a referral algorithm for non-rheumatologists in Arab countries and identified gaps in correct referrals and referral parameters.

Methods

Sixteen rheumatologists adopted referral algorithms, based on validation studies and clinical expertise, and summarized them in educational material (EM), including a PDF and video module. Five illustrative clinical vignettes were electronically sent to non-rheumatologists, to make referral decisions (phase 1). Two weeks later, the EM and the same vignettes were emailed again (phase 2). The primary outcome was the change in correct referral decisions using McNemar’s test. Improvement in correct referral parameters was evaluated and factors associated with correct referrals were analysed using logistic binary multivariable analysis.

Results

Eight hundred participants (general practitioners, family physicians, fellows, surgeons and others) from 13 countries completed phase 1 and 136 phase 2, with no demographic differences between them. Rates of phase 1 correct referral decisions were 72%, 72%, and 86% for referral-requiring cases, and 22% and 19% for others. In phase 2, correct referral decisions improved to 79%, 82%, and 88% for referral-requiring cases and 42% and 25% for others. Rates of most correct parameters improved significantly; factors associated with correct referrals included correct phase 1 referrals, duration of consultation, and fewer years practicing medicine.

Conclusion

Minor improvements in hypothetical referral decisions were observed after education, while significant improvements were noted for referral parameters. These results will help shape future referral campaigns’ design.

Key points

• Inflammatory arthritis and low back pain referrals to rheumatologists need boosting.

• The referral algorithm improved the rate of correct referral decisions.

• The referral algorithm improved the rate of correct parameters guiding the decision.