Non-specific Low Back Pain (LBP) is a highly prevalent condition affecting a large part of the adult population with major consequences worldwide. It is considered one of the biggest economic burdens in western societies (1). The lifetime prevalence worldwide has been estimated to be 80–85% and the condition is often characterized by recurrent episodes (2). According to the Global Burden of Disease Study, LBP results in more years lived with disability than any other disease in the world (2).
For a highly disabling recurrent and costly condition like LBP, it seems logical to invest in preventive strategies to mitigate and minimize the consequences. However, the evidence for effective interventions aimed at preventing LBP is scarce and until now, only exercise, and exercise in combination with education, have been shown empirically to be effective (3).
Chiropractic MC is described as a long-term management strategy for musculoskeletal disorders, introduced when optimum treatment benefit has been reached after an initial care plan. The aim is to prevent future episodes and deterioration by treating the patient at regular intervals, regardless of symptoms (4–18). In an ambitious effort, researchers across the Scandinavian countries have systematically explored and investigated indications, content, and frequency of MC in a series of research projects (8–22). Commonly, MC patients are selected based on their previous history of pain and the effectiveness of the initial care plan. Selected MC patients are commonly scheduled with 1–3 months intervals and are treated with manual therapy, along with individual exercises and lifestyle advice (18).
Based on this knowledge, a pragmatic randomized clinical trial was designed to investigate the effectiveness of MC in patients with recurrent and persistent LBP (19). The trial found that the MC-group had 12.8 (95% CI: 10.1, 15.5; p: <0.001) fewer days with bothersome LBP over a year compared to the control group (20). Although more effective, the number of visits was higher in the MC-group with 1.7 (95% CI: 1.8, 2.1; p: <0.001) more treatments during the 52-week study period (20). There was a large variability in the data, suggesting that there may be subgroups of patients who experienced fewer days of pain and fewer visits than others.
Psychological (23, 24), behavioural (25) and social characteristics (26) of LBP patients are important in the transition from acute to recurrent and persistent pain states (27–33). In line with the bio-psycho-social model, the leading theoretical framework underpinning the management of LBP (26, 34–36) clinicians would be expected to consider cognitive processes, psychological and behavioural dimensions of the pain experienced when managing patients with pain. Based on the cognitive-behavioural conceptualization of pain, The West Haven-Yale Multidimensional Pain Inventory (MPI) was developed to capture the perceptions and consequences of living with chronic pain (37).
The original instrument has been used to identify three clusters/subgroups of patients (38) and has been shown to be reliable, valid, and useful in outcome-based research (39, 40). The three different subgroups are defined as Adaptive Copers (AC), Interpersonally Distressed (ID), and Dysfunctional (DYS). Individuals in the AC group are characterized by low pain severity, low interference with everyday life due to pain, low life distress, a high activity level, and a high perception of life control and have the best prognosis with the lowest risk for long-term sick leave (41). Individuals in the ID group are characterized by dysfunctional behaviours such as low levels of social support, low levels of solicitous and distracting responses from significant others, and high scores on punishing responses compared to the DYS and AC patients (41). Individuals in the DYS subgroup are characterized by having high pain severity, marked interference with everyday life due to pain, high affective distress, low perception of life control, and low activity levels and have the worst prognosis along with the highest risk of long-term sickness absence (41). The identification of these subgroups has been used in clinical settings to investigate a diversity of conditions such as neck pain and LBP (41–43), temporomandibular disorders (44), headaches (45), fibromyalgia (46), and cancer pain (47) and are associated with different clinical outcomes.
To suggest that the outcome of MC may be affected by psychological characteristics such as those identified with the MPI instrument is an appealing idea and could lead to a tailored approach and better outcomes in the prevention of LBP. In a secondary analysis of the data from the RCT (21, 22), it was found that patients with a less favourable psychological profile (DYS subgroup) reported better outcomes from the MC approach (30.0 days less with pain, 95% CI: -36.6, -23.4). Surprisingly, the effect of MC within the DYS subgroup was achieved at an equal number of visits compared to the control group. On the other hand, patients within the AC group who received MC reported more days with pain (10.7 more days, 95% CI: 4.0, 17.5) while also receiving a higher number of visits compared to the control group (3.5 more visits, 95% CI: 1.8, 5.3).
These results may change the way MC is delivered in clinical practice as we can now identify subgroups of patients where MC is most effective. At the same time, it becomes clear that for a specific group of patients, the AC group, MC should not be recommended. In the AC group, the effect of MC is, at best, equally effective or in a worst-case scenario, results in more days with LBP while at the same time it results in more visits to the chiropractor compared to the control group.
Although valid and reliable, the Swedish version of the MPI instrument (MPI-S) was not designed to be used in daily clinical practice but rather as a comprehensive research tool, and no short version aimed specifically at clinical application has been published in the scientific literature. To utilize the aforementioned research findings in clinical practice, a pragmatic and convenient instrument with fewer items needs to be developed.
The objective of the study was threefold: 1) To develop a new instrument for identifying dysfunctional patients in a clinical setting, with adequate sensitivity, specificity, and receiver operating characteristics. 2) To assess the instrument’s ability to reproduce the previously published effect estimates of MC, and 3) To test the sensitivity, specificity, and receiver operating characteristics in 3 other existing datasets to assess validity across populations.