Artus, Majid (2011) The association between non-specific factors and response to primary care treatments for low back pain: a synthesis of evidence. Doctoral thesis, Keele University.

[thumbnail of Artus PhD 2011.pdf]
Artus PhD 2011.pdf

Download (3MB) | Preview


Findings from randomised controlled trials (RCTs) on primary care treatments for non-specific low back pain (NSLBP) often show modest or non-significant differences in responses to treatments. The overall response to treatment within arms, however, is often large. This raises the question of the non-specific effects associated with using the treatments and whether the size of these non-specific effects is much larger than the size of effects associated with the specific components of treatments. Non-specific effects in clinical trials, defined in this thesis as the effects on the overall improvement of symptoms (i.e. response to treatment) that is not attributed to the treatment itself, contribute to the clinical course of symptoms and can be related to the patient, the symptoms, the healthcare practitioner, the communication between the patient and practitioner, the nature of treatment provided and the setting and environment of the clinical encounter.

The objectives of this study were examining: 1) the pattern of within-arm overall responses to treatments in RCTs on non-specific low back pain; 2) sources of variation in responses to treatments by investigating the association of non-specific factors with overall responses to treatments; 3) the influence of patient characteristics on responses to treatments using individual patient data from RCTs; and 4) whether merely participating in RCTs adds to the size of response to treatments (the ‘trial effect’).

The findings suggest that responses to treatments for NSLBP follow a pattern of an early large improvement in symptoms within 13-27 weeks of starting treatment followed by a smaller further improvement. This pattern was common to arms of RCTs regardless of the type of treatment. There was evidence that participants who had back pain episodes of less than 13 weeks showed larger responses to treatments than those with longer duration. There was weak inconclusive evidence for the association with age, gender, history of back pain, overall trial quality, adequacy of patient blinding and adequate compliance. There was no evidence that participating in RCTs led to larger improvement in back pain symptoms compared with participating in cohort studies. In conclusion, there is some evidence for the association of factors that are not related to the treatments with responses to treatments for NSLBP. Insufficient data hindered the assessment of other non-specific factors that were considered to be important, such as practitioner-patient communication.

Item Type: Thesis (Doctoral)
Subjects: R Medicine > RM Therapeutics. Pharmacology
Divisions: Faculty of Medicine and Health Sciences > Primary Care Health Sciences
Contributors: van der Windt, DA (Thesis advisor)
Depositing User: Lisa Bailey
Date Deposited: 26 Jul 2017 13:25
Last Modified: 16 Aug 2022 08:13

Actions (login required)

View Item
View Item