A Comparison of Two Self-management Programmes for Patients With Back Pain

Research Questions The overarching aim of this mixed methods study is to investigate whether the use of a structured self-management programme (the Pain Toolkit) results in better outcomes for patients with chronic back pain compared to usual care. There will be a quantitative and qualitative component to the work. In the quantitative component patients will complete outcome measure scores at baseline and then 6 and 12 months after receiving the Pain Toolkit or the standard programme for self-management. In the qualitative component a group of patients will be invited to participate in semi-structured interviews after using the Pain Toolkit. The interviews will explore their experience of using of the toolkit, its acceptability and ease of use.

The principle research question is:

1. To investigate the effectiveness of structured self-management (the Pain Toolkit) compared to unstructured advice/information on function for patients discharged to self-management from an evidence based back pain pathway. The secondary research question is:
2. To investigate the effectiveness of structured self-management (the Pain Toolkit) compared to unstructured self-management advice/information on pain for patients discharged to self-management from an evidence based back pain pathway.
3. To investigate the effectiveness of structured self-management (the Pain Toolkit) compared to unstructured self-management advice/information on quality of life for patients discharged to self-management from an evidence based back pain pathway
4. To investigate the effectiveness of structured self-management (the Pain Toolkit) compared to unstructured self-management advice/information on healthcare utilisation for patients discharged to self-management from an evidence based back pain pathway
5. To explore patients' experiences of using the Pain Toolkit for self-management of pain

Background There is a wealth of literature outlining the disease burden of chronic pain and in particular back pain, which forms the basis of the background to this study. Pain is widespread in the UK with almost 10 million people affected almost every day (British Pain Society 2015). More locally, for Durham Dales, Easington and Sedgefield Clinical Commissioning Group (DDES CCG) where the student researcher works, the highest cost prescribed drug is Pregabalin which is primarily used as a pain relieving medication. The CCG spends in excess of £2m per year on this one drug alone (DDES 2016). Focussing on the literature relating to back pain, the World Health Organisation state that: 'back pain is not a disease but a constellation of symptoms. In most cases, the origins remain unknown' (WHO 2013). In 2010 the Global Burden of Disease Study estimated that low back pain was among the top 10 diseases and injuries that account for the highest number of Disability Adjusted Life Years worldwide (Al Mazroa 2013). The impact of this is demonstrated in the UK, by the Trades Union Congress (TUC 2008) who reported that British businesses lose an estimated 4.9 million days per year to work related back pain, with the North East of England suffering more than most. Other authors concur that most people will experience low pain at some point in their lives (Hoy, Brooks et al 2010 p769; Yang et al 2016 p459).

The full cost to the NHS of chronic pain and its management is not known (BPS 2017). For the specific area of low back pain the World Health Organisation report back pain is one of 'seven high-burden conditions….for which the currently available treatment is inadequate in reversing or halting the progression of disease '(WHO 2013 page x). The evidence above shows that back pain is therefore such a widespread concern to patients and commissioners that it is important to look for an evidence-based approach to manage it.

Using the research question: 'Does using the pain toolkit improve outcomes for patients accessing the north of England regional back pain pathway?' the student researcher consulted the following databases and found no results referencing the Pain Toolkit: AMED, EMBASE, OVID on line, EBM, HMIC and CINAHL. This would indicate that the Pain Toolkit as a method for self-management of pain has not been well studied. The search terms were widened to include programmes of self-management of chronic pain. The same databases were searched using the extended terms. 135 articles were returned from the search which was then narrowed by removing duplicates and articles that were described as study protocols or descriptive studies. 2 papers describing randomised control trials or RCTs were removed because they related to a physical activity programme rather than self-management. One article found during the literature review reported a review of RCTs which gave advice within self-management programmes (Liddle, Gracey and Baxter 2007). This paper undertook a systematic review of 39 randomised control trials exploring various aspects of advice on self-management for patients with back pain. The authors concluded that 'No conclusions could be drawn as to the frequency and content of advice that is most effective for LBP [lower back pain] due to the limited number and poor quality of RCTs in this area' (Liddle, Gracey and Baxter 2007 p310). They conclude that 'more investigation is needed into the role of follow-up advice for chronic LBP patients' (Liddle, Gracey and Baxter p 327). They also found that a wide variety of outcome measures were used in the various studies making meaningful comparisons between the studies difficult.

As has been argued above, although back pain is a significant burden to patients and the NHS, the area of self-management has not been well studied and there is no clear guidance as to appropriate content for self-management programmes. The studies related to self-management programmes are hampered by use of diverse outcome measures and lack of rigor. Consequently an appropriately powered randomised control trial, using a validated outcome measure to assess the effectiveness of a specific intervention, such as the pain toolkit, is required to address these inadequacies in the literature.

To test the effectiveness of the pain toolkit the following experimental hypothesis has been developed: Patients using the Pain Toolkit will report an outcome of a 10 point reduction in the Oswestry Disability Index compared with patients using standard support.

Methods In the proposed study, the intervention of the Pain Toolkit is to be tested to see whether its use in a specific cohort of patients with back pain at the point of discharge to self-management, will improve health outcomes. In order to test this there will be 2 groups of patients who are randomly allocated to receive either the pain toolkit or defined standard treatment. Other study designs such as case control and cohort studies were considered, however, because the study is prospective, involves the use of a specific intervention and is time limited a randomised control trial was judged to be the most appropriate design. For the qualitative aspect of the study a smaller cohort of 8-12 patients will be selected from the group that receive the Pain Toolkit to participate in semi-structured interviews, loosely based around the readiness to change theory (Kerns et al., 1997, Lorig and Stewart 1996). This aspect of the study will explore participants' experiences of using the Pain Toolkit. A flowchart showing the process for recruitment is shown at appendix 11. The questions within the semi-structured interviews are designed to explore whether the participant had tried any other self-management programmes prior to starting to use the intervention for the study. This deductive approach had led to a conceptual framework that attitudes towards self-management may be enhanced by structured literature to support the participant. This concept will be explore during the qualitative aspect of the study and will give context to the findings of the quantitative study.

Sample - Quantitative Study The study group will be a convenience sample of patients accessing the back pain pathway at the point of discharge to self-management. Inclusion criteria are: patients with back pain who have been referred to the North of England regional back pain service who consent to take part in the study. Exclusion criteria are patients with red flag indicators where immediate referral to secondary is indicated (NICE 2016). In addition children less than 18 years and, people who do not speak English and people with reduced mental capacity will be excluded. Although easy read versions of the toolkit are available (Pain toolkit 2017b), the Back Book that will be used for the control group is only available in English.

Using the NQUERY software, we estimate that a sample size of 70 in each group will have 90% power to detect a difference in mean change of 10 points (this is considered by NICE to be a clinically relevant change) between the intervention and comparator groups assuming that the common standard deviation (SD) of change is 18 points using a two group t-test with a 0.050 two-sided significance level. The estimate of SD of change scores was obtained from some previously collected data involving 967 participants. Ultimately, the data will be analysed with a similar between-subjects model for comparison of change scores, but with covariate adjustment for baseline measurements, age and sex. Similar studies show a 30% drop out rate. Therefore to ensure that the study is appropriately powered, 100 participants will be recruited for each group, giving a total of 200 participants overall.

The study will also consider refusals, drop outs and loses to follow up. This may include patients who do not use the pain toolkit during the study period or who do not complete the outcomes measures. It would potentially impact on the interpretation of the study results if either of these groups were large in number, it will therefore be important to determine how their results will be reported at the end of the study. An anticipated dropout rate of 30% will be included in the power calculation.

Instruments and analysis are described elsewhere in the registration application.

Conclusion As shown previously the impact of reducing pain in patients will have significant personal and economic benefits for patients and the whole health economy. The pain toolkit is perceived as an easy to use, widely available self-management tool that could be given to patients at any point in their pathway. There is however little evidence to back its use and so research is needed in this area. Should the study prove that patients using the pain toolkit experience a reduction in their ODI score this may indicate that it would be cost effective to offer the pain toolkit more widely to patients. Conversely if the study shows that there is little or no benefit in the use of the pain toolkit then alternative self-management tools can be explored.