File
Shashua A.1, Srulovicy E.2,3, Reges O.3, Hoshen M.3, Ghilai A.3, Levran I.1, Balicer R.3
1Clalit Health Services, Physical Therapy, Tel Aviv, Israel, 2University of Haifa, School of Nursing, Haifa, Israel, 3Clalit Health Services, Clalit Research Institute, Tel Aviv, Israel
Background: Low back pain (LBP) is a common and complicated complaint with significant health and economic implications. Within Clalit, the biggest health-care organization in Israel, approximately 12% of 700,000 LBP referrals to primary care providers (PCPs) were referred to physical therapy (PT) during 2014. Evidence from the literature indicates that the recommended treatments in the guidelines are not effective for all patients with LBP. Therefore, the need to identify subgroups that will benefit from PT treatment is essential.
Purpose: To identify clinical, demographic, and treatment related factors associated with improvement in functional status or pain level, and reduction in analgesic consumption or recurrent PCPs´ visits in patients with LBP.
Methods: A retrospective cohort study among Clalit members with LBP who completed a course of PT treatment between January 2013 and June 2014. Inclusion criteria: diagnosis of LBP (excluding specific low back pathologies); a continuous membership in Clalit six months prior and post PT treatment; and age 25 or older. Exclusion criteria: active cancer during the five-year period prior to PT treatment; patients who attended only one PT session. Information was obtained via a comprehensive data warehouse of Clalit, which is comprised of electronic medical records and administrative data. Univariate and multivariable regression models were performed to examine the association between clinical, demographic, and treatment related variables and functional status, pain level, analgesic consumption, and recurrent PSPs´ visits.
Results: The sample included 54,767 patients who were mostly female (60.6%) with a mean age of 57.5±16.8. Fifty percent and 63% of patients demonstrated clinical improvement in functional status and pain level, from a score of 45.7±16.9 to 22.9±15.2 (Pv 0.001), and 7.3±1.7 to 3.3±2.0 (Pv 0.001), respectively. Analgesic consumption during the 6 months following end of treatment decreased by 20% (Pv 0.001).
In the multivariable regression models, moderate and high physical activity, previous episodes of LBP, analgesic consumption pre-treatment, socio-economic level, ethnic sector, manual therapy, and electrotherapy were significantly associated with all outcomes. Body mass index and comorbidity were significantly associated with recurrent PCPs´ visits and analgesic consumption during the 6 months following end of treatment. Number of PT sessions was significantly associated with both functional and pain outcomes.
Patients with clinical significant improvement in functional and pain outcomes were at a reduced risk of 40% to utilize analgesic and PCPs´ visits within 6 months after the end of treatment (Pv 0.001).
Conclusion(s): This study provides data of a usual PT treatment process for LBP given in all community based clinics of Clalit, not following any specific protocol, which can reflect the real world of clinical practice. Some factors were found to be consistently significantly associated with improvement in functional and pain outcomes, and decrease in PCPs´ visits and analgesic consumption. Further intervention studies are required to offer appropriate treatments for the unimproved patients.
Implications: Our results suggest creating a decision support model for PCPs and physiotherapists. PCPs will be able to refer patients who will benefit from PT treatment, and physiotherapists will be able to offer those patients the appropriate treatment for better results.
Funding acknowledgements: The authors would like to thank the Clalit Research Institute stuff headed by Professor Ran Balicer. This work was unfunded.
Topic: Professional issues
Ethics approval: The study was approved by the ethical committee of Clalit.
All authors, affiliations and abstracts have been published as submitted.
