Skip to main content

Predictors of self-management in patients with chronic low back pain: a longitudinal study



Self-management (SM) is a key recommended strategy for managing chronic low back pain (CLBP). However, SM programmes generate small to moderate benefits for reducing pain and disability in patients with CLBP. The benefits of the SM programme can potentially be optimised by identifying specific subgroups of patients who are the best responders. To date, no longitudinal study has examined the predictive relationships between SM and biopsychosocial factors in patients with CLBP. The aim was to determine whether biopsychosocial factors predict SM and its change over time in patients with CLBP.


In this multi-centre longitudinal cohort study, we recruited 270 working-age patients with CLBP (mean age 43.74, 61% female) who consulted outpatient physiotherapy for their CLBP. Participants completed self-reported validated measures of pain intensity, disability, physical activity, kinesiophobia, catastrophising, depression and SM at baseline and six months. SM constructs were measured using eight subscales of the Health Education Impact Questionnaire (heiQ), including Health Directed Activity (HDA), Positive and Active Engagement in Life (PAEL), Emotional Distress (ED), Self-Monitoring and Insight (SMI), Constructive Attitudes and Approaches (CAA), Skill and Technique Acquisition (STA), Social Integration and Support (SIS) and Health Service Navigation (HSN). Data were analysed using General Linear Model (GLM) regression.


Physical activity and healthcare use (positively) and disability, depression, kinesiophobia, catastrophising (negatively) predicted (p < 0.05, R2 0.07–0.55) SM constructs at baseline in patients with CLBP. Baseline depression (constructs: PAEL, ED, SMI, CAA and STA), kinesiophobia (constructs: CAA and HSN), catastrophising (construct: ED), and physical disability (constructs: PAEL, CAA and SIS) negatively predicted a range of SM constructs. Changes over six months in SM constructs were predicted by changes in depression, kinesiophobia, catastrophising, and physical activity (p < 0.05, R2 0.13–0.32).


Self-reported disability, physical activity, depression, catastrophising and kinesiophobia predicted multiple constructs of SM measured using the heiQ subscales in working-age patients with CLBP. Knowledge of biopsychosocial predictors of SM may help triage patients with CLBP into targeted pain management programmes.

Trial registration

The study protocol was registered at on 22 December 2015 (ID: NCT02636777).

Peer Review reports


Low back pain (LBP) is a common condition (point-prevalence 18.3% ± 11.7%) [1], experienced by individuals of all ages globally [2]. It is the leading cause of disability, measured using the Years Lived with Disability (YLDs) [3, 4]. The high prevalence (18.3%), poor remission (54–90%) and high recurrence rates (24–80%) of low back pain [1, 5] result in chronic low back pain (CLBP) requiring higher health care needs such as general practitioner consultations found to be double that of matched controls without CLBP, and higher direct treatment cost estimated at £1000 per year per patient with CLBP [6]. In the United Kingdom (UK), national guidelines [7,8,9] recommend supported self-management (SM) as a management strategy in patients with CLBP.

The term SM is often inconsistently defined [10] as there is no agreed definition [11]. Nakagawa-Kogan [12] and colleagues defined SM as a combination of biological, psychological and social intervention techniques to alter long-term conditions by retraining self-regulating body processes to maximise disease management. This SM definition was based on the process model of therapy [13], which included role restructuring, forming the therapeutic alliance, developing commitment for change, analysing behaviour, negotiating treatment objectives, executing treatment, maintaining motivation, monitoring progress, and generalisation and termination of treatment. Clark defined SM as day-to-day home-management tasks to minimise the impact of disease as guided by healthcare providers [14], which highlighted both social and cognitive SM [15]. The UK National Health Service views SM as the ‘actions taken’ by individuals to recognise, treat and manage health and disease independently and in partnership with the healthcare system [16]. SM is advocated in the UK to manage long-term conditions, including low back pain [17].

For the purpose of this study, SM defines a dynamic and continuous ability to manage the disease, its symptoms, treatment, physical, psychological, and lifestyle changes [11] when living with a chronic illness. SM encompasses behaviour, role and emotional management with managing the disease by solving day-to-day problems, making conscious decisions, using appropriate health care resources, forming patient and healthcare provider partnerships and taking appropriate actions towards a healthy lifestyle [18]. SM programmes commonly consist of interdisciplinary group education delivered through a wide range of learning strategies in face-to-face, group-based, or internet-based interventions delivered by professionals or expert patients [18, 19]. The primary aim of SM programmes is to enhance SM and self-efficacy (confidence in one’s ability for SM) by utilising adult learning principles, case management theory and individualised treatment [20], allowing and encouraging individuals to manage their long-term conditions [21].

SM programmes are successful in reducing pain intensity [standardised mean difference- (SMD) -0.29 immediate in nine studies, -0.20 in long-term in four studies] and disability (SMD -0.28 immediate in nine studies, -0.19 in long-term in four studies) in patients with CLBP [22]. A similar reduction of pain intensity (11 studies, SMD -0.10) and disability (eight studies, SMD -0.15) has been reported when SM programmes have been delivered by expert patients or lay leaders [21]. However, at best, the clinical benefits of SM programmes are small to medium and often short-term in managing pain, disability, and self-efficacy in patients with CLBP.

These minor benefits of SM programmes are potentially due to several factors, including the lack of targeted SM interventions for specific populations [23]. The effectiveness of a treatment strategy depends on causal and mediation effects [24] and treatment matching [25]. Therefore, further understanding of the predictive relationships between SM constructs and biopsychosocial constructs in patients with CLBP may help identify a specific sub-group of patients with CLBP that may benefit from SM programmes and enhance the overall programme effectiveness. However, to our knowledge, predictors of SM in people with CLBP have only been investigated in one study to date, [26], which demonstrated that age [β =  − 0.197, Standard Error (SE) = 0.074] and poor overall health (negatively) and education attained at college. SM support (positively) predicted SM in 230 patients with CLBP when measured using the Patient Activation Measure (PAM) (β = 2.292, SE = 0.965). Yet, these predictive associations did not include psychological characteristics as potential predictors in the previous study and little attention has been paid to biopsychosocial measures and SM of CLBP. The aim of this study was to investigate whether there is a predictive relationship between SM constructs and biopsychosocial measures in patients who were seeking outpatient physiotherapy treatment for their CLBP.


Study design and sample size

This multi-centre prospective (non-experimental) longitudinal cohort study obtained a favourable ethical opinion from National Health Service Research Ethics Committee (Ref No 15/ES/1067- November 2015) and was conducted in line with the registered protocol ( ID: NCT02636777) [27]. A priori sample size calculation (using G*Power version 3.1.5 software) estimated that at least 130 participants would be required to detect a change with an effect size of 0.5 with 80% power and 5% significance level using the Health Directed Activity (HDA) subscale because this sub-scale produced the largest sample size required [28].

Inclusion and exclusion criteria

Patients with CLBP were recruited from six UK National Health Service Trusts, including five acute care trusts and one community musculoskeletal service provider. For the study, low back pain was defined as pain in the posterior aspect of the body between the lower margins of the twelfth ribs and the gluteal folds with or without pain in one or both legs [5]. Patients, who were walking in the community without any aids, aged between 18 and 65 years, who attended outpatient physiotherapy for their chronic low back pain, and who could read, write, and understand English, were included in the study. Patients were excluded if they were diagnosed with cancer or other known or self-reported specific causes for their low back pain (major trauma, fracture, inflammatory condition, ankylosing spondylitis, Grade 3 and 4 spondylolisthesis, severe spinal canal stenosis, or lumbar intervertebral disc protrusion or extrusion, spinal deformity); had undergone spinal surgery in the last one year or scheduled for any major surgery in the coming six months; who were pregnant women or women who had childbirth in the previous one year; had cognitive impairment and neurological diseases; and had severely impaired vision and hearing hindering survey completion.


Biopsychosocial factors

LBP duration, presence of related leg pain, age, gender, ethnicity, postcode, educational level, current employment status, annual household income, marital status, and living arrangements were recorded at the baseline. Additionally, the amount and nature of treatment received, and analgesics used were collected at baseline and follow-up. Other biopsychosocial measures utilised in this study included the 11-item Numeric Pain Rating Scale [29], 24-item Roland Morris Disability Questionnaire [30], International Physical Activity Questionnaire-Short Form (IPAQ-SF) [31], Tampa Scale of Kinesiophobia (TSK) [32], Pain Catastrophising Scale (PCS) [33], Patient Health Questionnaire-9 (PHQ-9) [34].


Self-management was measured using a multi-domain scale Health Education Impact Questionnaire (heiQ) version 3 [35]. The heiQ consists of 40 items, which measure eight different constructs of SM: Health-Directed Activities (HDA), Positive and Active Engagement in Life (PAEL), Emotional Distress (ED), Self-Monitoring and Insight (SMI), Constructive Attitudes and Approaches (CAA), Skill and Technique Acquisition (STA), Social Integration and Support (SIS) and Health Service Navigation (HSN). Each of the 40 items can be scored using four-point ordinal scale options from ‘strongly disagree’ to ‘strongly agree’ with no neutral option given. The heiQ has high internal consistency (Cronbach’s a 0.70–0.89) and discriminant validity in patients with chronic diseases [28, 35]. The heiQ scale has been chosen for its ability to capture multiple SM constructs across physical, psychological and social domains and low response bias [36].


Willing patients completed an expression of interest and were screened using the study selection criteria. Eligible and consenting patients were requested to complete the surveys at two-time points: baseline and follow-up after six months. Responses could be completed using paper, online, and telephone survey modes at participant preference to maximise patient convenience and the survey response rate [37]. The Bristol Online Survey (BOS) platform was used for the online survey, which ensured data integrity and adhered to high data security standards.

Data analyses

Data analyses were performed with a significance set at p < 0.05 in statistical software [IBM SPSS Statistics for Windows, Version 24.0 (Armonk, NY: IBM)]. Data were screened using stem-and-leaf plots and summaries to identify the presence of an impossible value. Scatter plots were visually assessed for any outliers. As the sample size was large (n > 100), normality was assessed using histograms and Q-Q plots. In the case of non-symmetrical or non-normal distribution, a Shapiro–Wilk test was utilised [38] for normality and Levene’s test for homogeneity of variance. The bootstrapped and accelerated intervals (n = 1000) were reported for all analyses.

The correlation between the model variables and each SM construct was estimated. Model variables that showed significant (p ≤ 0.05) correlation with the SM constructs were utilised in regression analysis. A multivariate regression analysis using a general linear model (GLM) was performed for each SM construct to identify predictors for baseline and change variables. Multi-collinearity was assessed using the Variance Inflation Factor (VIF < 10) for each independent variable.


Characteristics of the patients

A total of 434 patients with chronic low back pain (CLBP) expressed an interest in the study from six recruitment sites. Forty-nine (n = 49, 11.29%) patients were excluded at the screening stage for the following reasons: not meeting the inclusion criteria (n = 20, 4.61%), declined to participate (n = 15, 3.46%) and not contactable (n = 14, 3.23%). The remaining willing patients (n = 385) were invited to complete the baseline survey. Of these, 270 completed the baseline survey (Fig.1), and 153 completed the six-month follow-up survey. Recruitment of 270 participants was sufficient to detect a change of 0.4 (effect size) at baseline; completion of 153 six-month surveys was sufficient to detect a change of 0.5 (effect size) at follow-up.

Fig. 1
figure 1

Flow of the participants in the study

The mean age of recruited patients was 43.74 years (standard deviation 11.89, n = 270). 61% of patients were female, and 83.7% of patients were from White ethnic backgrounds. The demographic details and clinical characteristics are presented in Supplementary File 1, showing no significant difference between completers and non-completers of the follow-up survey, except for the highest level of education. There was no significant difference (p < 0.05) in SM constructs at baseline between the recruitment centres. The bivariate correlation (Spearman's rho) between the eight SM constructs ranged from 0.15 to 0.59, suggesting they are related but separate SM sub-constructs. Demographic characteristics of the participants at baseline and comparison between completers and non-completers of the follow-up survey are presented in the Supplementary File.

Predictors of self-management

Figure 2 summarises the regression results for the eight SM constructs at baseline. These analyses met the normality and homogeneity assumptions except for minor heteroscedasticity for HDA. For example, IPAQ was a significant predictor of HDA, and HDA increased by 0.04 for each Kilo metabolic equivalent (MET) increase in physical activity [F (7,260) = 7.70, p < 0.01] with an adjusted R2 of 0.15. Figure 3 summarises regression results for change in the eight SM constructs. For example, change in kinesiophobia and physical activity predicted change in HDA [F (6,139) 6.18, p < 0.01] with an adjusted R2 of 0.18.

Fig. 2
figure 2

Predictors of self-management constructs at baseline

Fig. 3
figure 3

Predictors of change in self-management constructs at the follow up

Descriptive statistics of and correlations between the SM constructs, sensitivity analyses using mean substitution and baseline observed carried forward data imputations for lost to follow-up cases are presented in the Supplementary File.


We found that physical disability, physical activity levels, depression, kinesiophobia and catastrophising are the main modifiable biopsychosocial predictors of SM and its change in patients with CLBP. Further, we also found that age, pain intensity and pain duration do not predict SM and its change. Pain intensity and duration not predicting SM agrees with previous research [26]. However, our finding that age is not a predictor of SM contrasts with a cross-sectional study [26], where age correlates with SM negatively.

In our study, perceived physical disability negatively predicted three out of the eight SM constructs. However, physical disability measured using the Oswestry Disability Index was not a significant predictor of SM measured using the PAM in 230 patients with CLBP [26]. This difference in the findings could be due to the populations and different scales to measure SM and disability. For example, Kawi measured SM using PAM, which measures only patients’ activation and engagement from primary care and specialist pain centre in the USA.

We found that baseline depression had a significant negative predictive association for five out of eight SM constructs, suggesting that lower mood (i.e., symptoms of depression) was associated with poorer self-management outcomes. Although depression has not been examined as an explanatory variable in previous research investigating predictors of SM in patients with CLBP [26], depression is common in patients with other long-term conditions and has been found to impact negatively on SM. For example, depression is common in diabetes mellitus [39] and is an established negative predictor of diabetes SM in children [40] and adults [41, 42]. Depression has also been identified as a predictor of SM in older adults (n = 3292) in the UK, albeit using a different outcome measure [43]. Depression significantly predicted SM, measured using the Skill and Techniques Acquisition (STA) subscale of the German version of the heiQ, in patients with chronic conditions (n = 580), including rheumatism, asthma, orthopaedic disorders and inflammatory bowel disease [44].

Therefore, our results suggest that depression is a key predictor of certain constructs of SM in patients with CLBP, which agrees with broader research in patients with diverse long-term conditions.

Kinesiophobia and catastrophising have not previously been investigated as predictors of SM in patients with CLBP. However, distress and/or anxiety were investigated as a predictor of SM in patients with diabetes [45, 46]. An earlier study by Albright et al. [46] found stress had a significant negative predictive association with exercise and diet SM in 392 type II diabetes patients. Similarly, Schinkus et al. [45] found distress (measured using Diabetes Distress Scale) and anxiety (measured using the State-Trait Anxiety Inventory) were significant predictors of overall diabetes SM (measured using the Diabetes Self-Management Questionnaire) in 146 patients with type-I and type-II and gestational diabetes. These studies highlight the importance of measuring distress or anxiety or related variables as an explanatory variable in SM predictor studies.

In the present study, healthcare use, measured using the self-reported number of sessions attended at the general practitioner, physiotherapist, specialist, and other practitioners for CLBP, significantly predicted the SMI construct of SM. Further, education, income, living arrangements, being employed, being married, high annual income (> £30,000) and white ethnicity had significant associations in univariate GLM analysis. These results agree with the previous cross-sectional study [26], where age, education and income were significant predictors of SM in patients with CLBP. However, no significant predictive association was found in the multivariate GLM analysis for demographic and socioeconomic factors, suggesting that change in SM does not depend on age, education, and income.

Changes in depression, kinesiophobia, catastrophising, and physical activity level similarly predicted SM changes (R2 10% and 32%). Change in catastrophising predicted change in five out of eight SM constructs (HDA, PAEL, CAA, SIS and HSN). Catastrophising is a negative predictor for patients with CLBP and might contribute to the delayed recovery [47]. Patients with CLBP who had higher catastrophising were associated with a significantly higher disability using Roland Morris Disability Questionnaire in a UK population at a 12-month follow-up [48]. Further, patients with CLBP reported fluctuating negative pain-related thoughts affecting their coping and pain-related meta-cognition in a recent qualitative study [49], which could potentially influence HDA, PAEL, CAA, SIS and HSN. Change in depression predicted change in PAEL and CAA. Similarly, change in depression predicted SM in patients with diabetes [42] and long-term conditions [44].

Theoretical underpinning

According to the Social Cognitive Theory, one of the critical theories underpinning SM, cognitive factors and psychological states modify self-judgement and the SM [50, 51]. Therefore, depression, excessive negative pain-related emotions or catastrophising and fear related to pain or re-injury or kinesiophobia may influence one's SM ability. Similarly, physiological states, including depression, kinesiophobia and catastrophising, influence self-efficacy and SM [51]. Therefore, along with promoting healthy living and physical activity [52], psychological and behavioural factors should be targeted to enhance SM in patients with CLBP. From a behaviourist point of view, capability, opportunity, and motivation interact to generate behaviour, in which capability includes one's physical and psychological abilities to engage in (SM) activity. So, SM programmes can utilise the Behaviour Change Wheel to create opportunities using interventions and policies to motivate individuals to change their capability [53].

Strength and limitations

To our knowledge, this is the first prospective multi-centre longitudinal cohort study investigating predictors of SM in patients with CLBP. A strength of the study is the use of a multi-construct SM scale which provides a comprehensive assessment of SM constructs, and multi-component measures have not been used in previous studies of CLBP or other chronic conditions. The study has some limitations. The attrition rate was relatively high, with 117 patients lost to follow-up (43.33% attrition, compared with an anticipated 30%). However, there was no difference in baseline disease-related and SM outcomes between patients that completed follow-up and those who did not. The study had a poor representation of the Asian and male gender. However, it has been found that women are more likely to participate in survey research [54]. The lack of ethnic diversity may be due to excluding patients without good English language abilities due to a lack of funding for interpreters. However, the figures for ethnicity are comparable to the UK 2011 Census data [55] proportion of White and working citizens. Future studies may focus on ways to improve SM in low-income, non-White populations with CLBP, particularly individuals with poor health literacy.


Conducting the first prospective longitudinal study investigating biopsychosocial predictors of SM, we conclude that disability, physical activity, depression, catastrophising, and kinesiophobia predict SM and its change in working-age adults who attended physiotherapy for their CLBP. We recommend utilising these modifiable biopsychosocial factors in future research and clinical practice to triage and match patients into targeted SM programmes.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.


95% CI:

95% Confidence Interval


Bias Corrected and Accelerated


Bristol Online Survey


Chronic low back pain


Generalised Linear Model


Health Directed Activity


Health Education Impact Questionnaire


Health Service Navigation


International Business Machines Corporation


Intraclass Correlation Coefficient


International Physical Activity Questionnaire- Short Form


National Health Service


Numeric Pain Scale


Pain Catastrophising Scale


Patient Health Questionnaire-9


Roland Morris Disability Questionnaire


Standard Deviation


Social Integration and Support




Self-Monitoring and Insight


Years Lived with Disability


  1. Hoy D, et al. A systematic review of the global prevalence of low back pain. Arthritis Rheum. 2012;64(6):2028–37.

    Article  Google Scholar 

  2. Hartvigsen J, et al. What low back pain is and why we need to pay attention. The Lancet. 2018;391(10137):2356–67.

    Article  Google Scholar 

  3. Vos T, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 310 diseases and injuries, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015. The Lancet. 2016;388(10053):1545–602.

    Article  Google Scholar 

  4. Vos T, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. The Lancet. 2012;380(9859):2163–96.

    Article  Google Scholar 

  5. Hoy D, et al. The global burden of low back pain: estimates from the Global Burden of Disease 2010 study. Ann Rheum Dis. 2014;73(6):968–74.

    Article  Google Scholar 

  6. Hong J, et al. Costs associated with treatment of chronic low back pain: an analysis of the UK General Practice Research Database. Spine. 2013;38(1):75–82.

    Article  Google Scholar 

  7. Bernstein IA, et al. Low back pain and sciatica: summary of NICE guidance. BMJ. 2017;356:i6748.

  8. NICE Guideline, Low Back Pain and Sciatica in Over 16s: Assessment and Management. 2016, National Guideline Centre UK.

  9. SIGN, Scottish Intercollegiate Guidelines Network. Management of chronic pain. Edinburgh: SIGN; 2013., in SIGN Guideline 136. 2013, Scottish Intercollegiate Guidelines Network: Edinburgh.

  10. Audulv A, et al. Coping, adapting or self-managing - what is the difference? A concept review based on the neurological literature. J Adv Nurs. 2016;72(11):2629–43.

    Article  Google Scholar 

  11. Barlow J, et al. Self-management approaches for people with chronic conditions: a review. Patient Educ Couns. 2002;48(2):177–87.

    Article  Google Scholar 

  12. Nakagawa-Kogan H, et al. Self-management of hypertension: Predictors of success in diastolic blood pressure reduction. Res Nurs Health. 1988;11(2):105–15.

    Article  CAS  Google Scholar 

  13. Kanfer FH, Grimm LG. Managing clinical change: A process model of therapy. Behav Modif. 1980;4(4):419–44.

    Article  Google Scholar 

  14. Clark NM, et al. Self-management of chronic disease by older adults a review and questions for research. J Aging Health. 1991;3(1):3–27.

    Article  Google Scholar 

  15. Corbin, J.M. and A. Strauss, Unending work and care: Managing chronic illness at home. 1988: Jossey-Bass.

  16. NHS England. Supporting self-management/ self-care. 2018 16/12/2018]; Available from:

  17. Department of Health, Self Care–A Real Choice. Self Care Support–A Practical Option. 2005, Department of Health London.

  18. Lorig KR, Holman H. Self-management education: history, definition, outcomes, and mechanisms. Ann Behav Med. 2003;26(1):1–7.

    Article  Google Scholar 

  19. Marks R, Allegrante JP. A review and synthesis of research evidence for self-efficacy-enhancing interventions for reducing chronic disability: implications for health education practice (part II). Health Promot Pract. 2005;6(2):148–56.

    Article  Google Scholar 

  20. Alderson M, et al. The program for rheumatic independent self-management: a pilot evaluation. Clin Rheumatol. 1999;18(4):283–92.

    Article  CAS  Google Scholar 

  21. Foster, G., et al., Self‐management education programmes by lay leaders for people with chronic conditions. Cochrane Database of Systematic Reviews, 2007: p. Issue 4. Art. No.: CD005108. DOI:

  22. Du S, et al. Self-management program for chronic low back pain: a systematic review and meta-analysis. Patient Educ Couns. 2017;100(1):37–49.

    Article  Google Scholar 

  23. Taylor S, et al. Novel three-day community-based, non-pharmacological, group intervention for chronic musculoskeletal pain (COPERS): a randomized clinical trial. PLoS Med. 2016;13(6):e1002040.

    Article  Google Scholar 

  24. Mansell G, Kamper SJ, Kent P. Why and how back pain interventions work: What can we do to find out? Best Pract Res Clin Rheumatol. 2013;27(5):685–97.

    Article  Google Scholar 

  25. Turk DC, Okifuji A. Psychological factors in chronic pain: evolution and revolution. J Consult Clin Psychol. 2002;70(3):678.

    Article  Google Scholar 

  26. Kawi J. Predictors of self-management for chronic low back pain. Appl Nurs Res. 2014;27(4):206–12.

    Article  Google Scholar 

  27. Banerjee, A., P. Hendrick, and H. Blake, Predictors of self-management in patients with chronic low back pain: a longitudinal cohort study protocol. International Journal of Physiotherapy and Rehabilitation, 2016(April): p. 14–24.

  28. Elsworth GR, Nolte S, Osborne RH. Factor structure and measurement invariance of the Health Education Impact Questionnaire: Does the subjectivity of the response perspective threaten the contextual validity of inferences? SAGE Open Medicine. 2015;3:2050312115585041.

    Article  Google Scholar 

  29. Jensen MP, Karoly P, Braver S. The measurement of clinical pain intensity: a comparison of six methods. Pain. 1986;27(1):117–26.

    Article  Google Scholar 

  30. Roland M, Morris R. A study of the natural history of back pain: part I: development of a reliable and sensitive measure of disability in low-back pain. Spine. 1983;8(2):141–4.

    Article  CAS  Google Scholar 

  31. Kim Y, Park I, Kang M. Convergent validity of the international physical activity questionnaire (IPAQ): meta-analysis. Public Health Nutr. 2013;16(3):440–52.

    Article  Google Scholar 

  32. Miller RP, Kori SH, Todd DD. The Tampa Scale: a Measure of Kinisophobia. Clin J Pain. 1991;7(1):51.

    Article  Google Scholar 

  33. Picavet HSJ, Vlaeyen JW, Schouten JS. Pain catastrophizing and kinesiophobia: predictors of chronic low back pain. Am J Epidemiol. 2002;156(11):1028–34.

    Article  Google Scholar 

  34. Smarr KL, Keefer AL. Measures of depression and depressive symptoms: Beck Depression Inventory-II (BDI-II), Center for Epidemiologic Studies Depression Scale (CES-D), Geriatric Depression Scale (GDS), Hospital Anxiety and Depression Scale (HADS), and Patient Health Questionnaire-9 (PHQ-9). Arthritis Care Res. 2011;63(S11):S454–66.

    Article  Google Scholar 

  35. Osborne RH, Elsworth GR, Whitfield K. The Health Education Impact Questionnaire (heiQ): an outcomes and evaluation measure for patient education and self-management interventions for people with chronic conditions. Patient Educ Couns. 2007;66(2):192–201.

    Article  Google Scholar 

  36. Banerjee A, et al. A systematic review of outcome measures utilised to assess self-management in clinical trials in patients with chronic pain. Patient Educ Couns. 2018;101(5):767–78.

    Article  Google Scholar 

  37. Dillman DA, et al. Response rate and measurement differences in mixed-mode surveys using mail, telephone, interactive voice response (IVR) and the Internet. Soc Sci Res. 2009;38(1):1–18.

    Article  Google Scholar 

  38. Razali NM, Wah YB. Power comparisons of shapiro-wilk, kolmogorov-smirnov, lilliefors and anderson-darling tests. Journal of Statistical Modeling and Analytics. 2011;2(1):21–33.

    Google Scholar 

  39. Whitworth SR, et al. Depression symptoms are persistent in Type 2 diabetes: risk factors and outcomes of 5-year depression trajectories using latent class growth analysis. Diabet Med. 2017;34(8):1108–15.

    Article  CAS  Google Scholar 

  40. Guo J, et al. Diabetes self-management, depressive symptoms, quality of life and metabolic control in youth with type 1 diabetes in China. J Clin Nurs. 2013;22(1–2):69–79.

    Article  Google Scholar 

  41. Schinckus L, et al. When knowing is not enough: Emotional distress and depression reduce the positive effects of health literacy on diabetes self-management. Patient Educ Couns. 2018;101(2):324–30.

    Article  Google Scholar 

  42. Oh H, Ell K. Associations between changes in depressive symptoms and social support and diabetes management among low-income, predominantly hispanic patients in patient-centered care. Diabetes Care. 2018;41(6):1149–56.

    Article  Google Scholar 

  43. Blakemore A, et al. Patient activation in older people with long-term conditions and multimorbidity: correlates and change in a cohort study in the United Kingdom. BMC Health Serv Res. 2016;16(1):582.

    Article  Google Scholar 

  44. Musekamp G, et al. Improved self-management skills predict improvements in quality of life and depression in patients with chronic disorders. Patient Educ Couns. 2016;99(8):1355–61.

    Article  Google Scholar 

  45. Schinckus L, et al. The role of trait emotional intelligence in diabetes self-management behaviors: The mediating effect of diabetes-related distress. Pers Individ Differ. 2018;131:124–31.

    Article  Google Scholar 

  46. Albright TL, et al. Predictors of self-care behavior in adults with type 2 diabetes: an RRNeST study. Fam Med. 2001;33(5):354–60.

    CAS  Google Scholar 

  47. Wertli MM, et al. Catastrophizing—a prognostic factor for outcome in patients with low back pain: a systematic review. Spine J. 2014;14(11):2639–57.

    Article  Google Scholar 

  48. Grotle M, et al. Are prognostic indicators for poor outcome different for acute and chronic low back pain consulters in primary care? Pain. 2010;151(3):790–7.

    Article  Google Scholar 

  49. Schütze R, et al. ‘I call it stinkin’thinkin’’: A qualitative analysis of metacognition in people with chronic low back pain and elevated catastrophizing. Br J Health Psychol. 2017;22(3):463–80.

    Article  Google Scholar 

  50. Bandura A. Health promotion from the perspective of social cognitive theory. Psychol Health. 1998;13(4):623–49.

    Article  Google Scholar 

  51. Bandura A. Health promotion by social cognitive means. Health Educ Behav. 2004;31(2):143–64.

    Article  Google Scholar 

  52. Buchbinder R, et al. Low back pain: a call for action. Lancet. 2018;391(10137):2384–8.

    Article  Google Scholar 

  53. Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6(1):42.

    Article  Google Scholar 

  54. Mulder J, de Bruijne M. Willingness of online respondents to participate in alternative modes of data collection. Surv Pract. 2019;12(1):8356.

    Article  Google Scholar 

  55. 2011 Census: Key Statistics for England and Wales, March 2011. 2012, Office for National Statistics, UK.

Download references


Sincere thanks to Prof David Walsh (Director, Pain Centre Versus Arthritis, University of Nottingham), Eleanor Dunston (Consultant Spinal Physiotherapist, Nottingham University Hospitals NHS Trust, UK), Dr Rob Goodwin (Clinical Lead, Nottingham CityCare Partnership CIC.), Laura Houten (Senior Physiotherapist, Nottingham CityCare Partnership CIC.), Purba Bhattacharjee (Advanced Practitioner Physiotherapist, Nottingham CityCare Partnership CIC.), and all physiotherapists and participants in the six recruitment sites.


The corresponding author (AB) received an International Vice-Chancellor Scholarship from the University of Nottingham, but the funder had no direct role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.

Author information

Authors and Affiliations



All authors (AB, PH, and HB) contributed to the conception, obtaining ethical approval, data analysis and writing of this manuscript. AB collected the data. All authors read and approved the final manuscript.

Corresponding author

Correspondence to A. Banerjee.

Ethics declarations

Ethics approval and consent to participate

The study received a favourable ethical opinion and research governance approval from the Health Research Authority (HRA) in the United Kingdom [Reference: East of Scotland Research Ethics Service- 15/ES/1067]. All participants had provided informed consent (written or using an online link) before taking part in the study.

Consent for publication

Not Applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1:

Table 1. Demographic characteristics of the participants at baseline and comparison between completers and non-completers of the follow-up survey. Table 2. Characteristics of the participants at baseline. Table 3. Descriptive statistics for self-management constructs at baseline. Table 4. Spearman (rho) correlation for the self-management constructs at baseline. Table 5. Descriptive statistics after loss to follow up data imputation. Figure 4. Predictors of change in self-management constructs at follow up after mean substitution of the lost to follow up cases. Figure 5. Predictors of change in self-management constructs at follow up after last observation carried forward substitution of the lost to follow up cases.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Banerjee, A., Hendrick, P. & Blake, H. Predictors of self-management in patients with chronic low back pain: a longitudinal study. BMC Musculoskelet Disord 23, 1071 (2022).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: