EULAR points to consider for the conduction of workforce requirement studies in rheumatology
| Number | Point to consider | LoA | LoE |
| 1 | Workforce models should integrate supply, demand and need of the respective geopolitical entity (eg, municipality, region, state, country) and should express results as full-time equivalents and as number of rheumatologists | 9.5 (0.9) 95% ≥8 | 5 |
| 2 | Workforce models should provide projections over a period of 5–15 years | 9.1 (1.1) 90% ≥8 | 5 |
| 3 | Workforce models should not assume a current balance between supply and need | 9.6 (0.7) 100% ≥8 | 5 |
| 4 | Workforce models should, where possible, rely on several data sources and include uncertainty analyses | 9.8 (0.4) 100% ≥8 | 5 |
| 5 | Workforce models should be regularly updated; updates should include an analysis of the actual performance (ie, prediction validity) of the previous model | 9.5 (0.6) 100% ≥8 | 5 |
| 6 | Workforce need for patient care should be based on the prevalence and referral rates of diseases managed by rheumatologists as well as on an estimation of time needed per patient | 9.7 (0.7) 100% ≥8 | 5 |
| 7 | Workforce need for patient care should consider current and future demographics, sociocultural characteristics of the population and disease patterns | 9.5 (0.9) 95% ≥8 | 5 |
| 8 | Workforce need and supply should consider work outside rheumatology patient care (eg, administrative tasks, research, teaching, non-rheumatological disease management), as well as patient care performed by other health professionals in rheumatology | 9.4 (0.9) 95% ≥8 | 5 |
| 9 | Workforce supply should account for demographic composition of rheumatologists, the number of rheumatologists entering and leaving the workforce, and generational attitudes of rheumatologists towards scope of practice and work–life balance | 9.1 (2.3) 85% ≥8 | 5 |
| 10 | Workforce models should consider the effects of medical developments, including new technologies, medications, artificial intelligence and e-health, on demand and supply | 9.4 (1.1) 85% ≥8 | 5 |
Numbers in column ‘LoA’ indicate the mean and SD (in parentheses) of the LoA, as well as the percentage of task force members with an agreement ≥8. None of the studies identified corresponded to any of the categories of Oxford Centre for Evidence-Based Medicine.64 Evidence level was therefore set as ‘5’, which is the lowest level of evidence.
LoA, level of agreement; LoE, level of evidence