Implementation of practice-based projects promoting apporiate opiod use in Dutch primary care: a mixed methods evaluation.

ABSTRACT:

Background
Opioids are frequently used for chronic non-malignant pain. However, long-term use lacks efficacy and is associated with risks, such as side-effects and opioid use disorder. Despite updated guidelines, general practitioners (GPs) struggle to reduce opioid prescribing.

Aim
The aim of this study is to investigate the relationship betweenthe beliefs of GPs towards opioid prescribing for chronic non-malignant pain and the actual opioid prescribing behaviour of GP practices.

Method
Guided by the health belief model (HBM), a survey was designed to examine the GPs beliefs [1]. Prescribing data of GP practices of the corresponding GP was available from The Nivel Primary Care Database. Factor analysis was used to classify survey statements into the HBM-domains. Prescribing data over 2019 was used to determine the number of oral morphine equivalents (OME) prescribed, and the number of long-term opioid users (LTOT:[4  prescriptions and [600 OME, within 365 days) per practice, per 1000 registered patients. Linear regression was used to investigate the correlation between the beliefs (HBM-domain scores) and the prescribing behaviour (OME and LTOT).

Results
Both prescribing and survey date were available of 84 unique GP practices. Based on the factor analysis, three HBM-do mains were defined: benefits, threats and self-efficacy. The fourth domain, barriers, was excluded due to invalidity. The mean number of prescribed OME per practice per 1000 registered patients in 2019 was 85,860 mg (SD 51,644 mg). It was not significantly associated with HBM-domain scores of the GP, but it was positively associated with the mean age of registered patients per practice (p = 0.01). The mean number of patients with LTOT per 1000 registered patients per practice was 8.5 (SD 3.4). As for OME, the number of LTOT patients was not significantly associated with the beliefs domain scores of GPs, but a positive association with the mean age of registered patients per practice was found (p = 0.001).

Conclusion
No association was found between the attitude of GPs towards the use of opioids in the treatment of chronic non-malignant and the prescribing behaviour within the practice they work. The influence of practice characteristics like the mean age of the registered patients may have a greater influence on the variance in opioid prescribing behaviour of practices than the attitude of the GP. 

References/Acknowledgments
[1] Jansen-Groot Koerkamp, E. A. W., Weesie, Y. M., Heringa, M., et al. (2024). General practi tioners’ and community pharmacists’ beliefs and practices on opioids for non-malignant pain. European Journal of Pain. https://doi.org/10. 1002/ejp.2304. Disclosure of Interest: None Declared.