Mas et al. BMC Health Services Research (2019) 19:962 https://doi.org/10.1186/s12913-019-4810-x RESEARCH ARTICLE Open Access Effectiveness of a multidisciplinary BIOPSYCHOSOCIAL intervention for non- specific SUBACUTE low back pain in a working population: a cluster randomized clinical trial Romina Raczy Mas1,2,3, Tomàs López-Jiménez2, Enriqueta Pujol-Ribera2,3, María Isabel Fernández-San Martín1, Jenny Moix-Queraltó4, Elena Montiel-Morillo5, Teresa Rodríguez-Blanco2, Marc Casajuana-Closas2,3, M. J. González-Moneo6, Ester Núñez Juárez7, Montse Núñez Juárez8, Mercè Roura-Olivan9, Raquel Martin-Peñacoba10, Magda Pie-Oncins6, Montse Balagué-Corbella2, Miguel-Ángel Muñoz1,2,3, Concepción Violan2 and Anna Berenguera2,3* Abstract Background: Low back pain (LBP) is a multifactorial condition with individual and societal impact that affects populations globally. Current guidelines for the treatment of LBP recommend pharmacological and non- pharmacological strategies. The aim of this study was to compare usual clinical practice with the effectiveness of a biopsychosocial multidisciplinary intervention in reducing disability, severity of pain and improving quality of life in a working population of patients with subacute (2–12 weeks), non-specific LBP. Methods: Longitudinal cluster randomized clinical trial conducted in 39 Primary Health Care Centres (PHCC) of Barcelona, with patients aged 18–65 years (n = 501; control group = 239; 26 PHCC, intervention group = 262; 13 PHCC). The control group received usual clinical care. The intervention group received usual clinical care plus a biopsychosocial multidisciplinary intervention, which consisted of physiotherapy, cognitive-behavioural therapy and medication. The main outcomes were changes in the Roland Morris Disability Questionnaire (RMDQ), and the minimal clinically important differences. Secondary outcomes were changes in the McGill Pain (MGPQ) and Quality of Life (SF-12) questionnaires. Assessment was conducted at baseline, 3 and 12 months. Analysis was by intention- to-treat and analyst-blinded. Multiple imputations were used. (Continued on next page) * Correspondence: aberenguera@idiapjgol.org; https://www.idiapjgol.org 2 Institut Universitari d’Investigació en Atenció Primària (IDIAP Jordi Gol), Gran Via Corts Catalanes, 587, àtic, 08007 Barcelona, Spain 3 Department of Paediatrics, Gynaecology and Obstetrics and Preventive Medicine, Universitat Autònoma de Barcelona Faculty of Medicine-Bellaterra, 08193 (Cerdanyola del Vallès), Spain Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Mas et al. BMC Health Services Research (2019) 19:962 Page 2 of 9 (Continued from previous page) Results: Of the 501 enrolled patients, 421 (84%) provided data at 3 months, and 387 (77.2%) at 12 months. Mean age was 46.8 years (SD: 11.5) and 64.7% were women. In the adjusted analysis of the RMDQ outcome, only the intervention group showed significant changes at 3 months (− 1.33 points, p = 0.005) and at 12 months (− 1.11 points, p = 0.027), but minimal clinically important difference were detected in both groups. In the adjusted analysis of the RMDQ outcome, the intervention group improvement more than the control group at 3 months (− 1.33 points, p = 0.005) and at 12 months (− 1.11 points, p = 0.027). The intervention group presented a significant difference. Both groups presented a minimal clinically important difference, but more difference in the intervention group. The intervention group presented significant differences in the MGPQ scales of current pain intensity and VAS scores at 3 months. No statistically significant differences were found in the physical and mental domains of the SF-12. Conclusions: A multidisciplinary biopsychosocial intervention in a working population with non-specific subacute LBP has a small positive impact on disability, and on the level of pain, mainly at short-term, but no difference on quality of life. Trial registration: ISRCTN21392091 (17 oct 2018) (Prospectively registred). Keywords: Primary health care, Multidisciplinary biopsychosocial intervention, Non-specific subacute low back pain, Disability, Pain, Quality of life, Cognitive-behavioural therapy Background Compared to no treatment and to other guideline recom- Low back pain (LBP) is a common health problem that af- mendations, recent evidence-based studies support a multi- fects approximately 80–85% of the general population at disciplinary approach to ameliorate LBP [6]. Accordingly, it least once in their lifetime and has a global prevalence be- has been suggested that the timely integration of multidis- tween 17 and 32%, of which 11–12% are disabled by this ciplinary treatment strategies that include physiotherapy, condition [1, 2]. In a recent survey conducted in Spain, cognitive-behavioural therapy and medication for patients LBP was highly prevalent (50.9%) at all ages, but especially with non-specific subacute LBP, might reduce the individ- in the working age population (18–65 years old) [3]. ual and social impact [12]. Following a systematic review by The Global Burden of Disease 2010 ranks LBP Kamper et al. (2014), who adopted the term ‘multidisciplin- amongst the top ten causes of DALYs (disability-ad- ary biopsychosocial rehabilitation’ to integrate education justed life years) [2]. Consequently, LPB is associated and physiotherapy with cognitive-behavioural psychology with a huge individual and societal burden and remains with the aim to improve disability and function, [9] the a frequent reason for medical consultation globally [4]. current NICE guidelines (2016) recommend early multidis- In Spain alone, LBP generates over 2 million annual con- ciplinary management [13]. sultations in primary care (ENSE 2011/12) [5]. Lastly, the main objective of the current study was to Despite the wide range of treatments and health-care re- evaluate the change in disability using the validated sources devoted to LBP, back-related disability and burden Spanish version of the Roland Morris Disability ques- have increased, in recent years [6]. A study carried out in tionnaire (RMDQ) across the intervention and its associ- 36 Primary Health Care Centres (PHCC) in Spain showed ation with minimal clinically important differences. The that despite guideline-based management, the pain con- second goal were to assess changes in pain intensity tinued in 37% and had worsened in 10% of patients after using the McGill Pain Questionnaire (MGPQ) and in two months [7]. The natural history of LBP can be quality of life as measured by the Short Form 12-Item extremely variable and recurrences are common, with, ap- (SF-12). proximately 65% of patients still experiencing pain one year after the onset of this condition [8, 9]. Methods It is generally accepted that subacute LBP occurs after Design a period of at least 6 months without LBP, and that it An analyst-blinded longitudinal cluster randomized con- has a duration between 2 and 12 weeks [10]. Research trolled clinical trial was conducted. Patients with non- conducted in Spain reported changes in disability, pain specific subacute LBP treated with a multidisciplinary and quality of life after 2 weeks of LBP [11]. Once the approach (intervention group) were compared with a subacute episode has been established, early interven- control group receiving only usual clinical care. (Current tions are recommended to avoid deterioration, even if it Controlled Trials identifier: ISRCTN21392091) (17 oct is considered that approximately one third of patients 2018) (Prospectively registred). The study protocol has have a favourable evolutions [6]. been previously published [12]. Mas et al. BMC Health Services Research (2019) 19:962 Page 3 of 9 Setting included between 6 and 12 participants and some The trial was conducted in the primary care setting. A PHCCs had more than one group receiving the same total of 39 PHCCs located in Barcelona and its greater intervention. To guarantee the standardisation of the metropolitan area participated in the project. group sessions, only one qualified psychologist and one physiotherapist with expertise in group interventions im- Study population plemented the intervention in all PHCCs. At the end of Patients were included if they presented LBP lasting be- the study, the control group also received the educa- tween 2 and 12 weeks, and if they did not have a history tional booklet and the audio-visual material. of LBP during the 6 months prior to the current episode [10, 14]. Participants were active workers, aged between Outcomes measures 18 and 65 years, they had to understand Catalan or The main outcome measurement was change in disability Spanish and were required to be contactable for at least as measured by the Roland Morris Disability Questionnaire twelve months after the onset of the study. Exclusion (RMDQ) [18], translated and validated into Spanish [19] criteria were as follows: patients unwilling to participate; (scale 0–24; lower scores indicate less disability). A min- LBP that coexisted with cognitive impairment or psychi- imal clinically important difference in disability under 2.5 atric disorders; other causes of disability which impeded RMDQ points compared to the baseline value for subacute responding to the questionnaires; pregnancy and breast- and chronic patients was considered negligible [20, 21]. feeding; physical problems in the preceding 3 months; Secondary outcome measurements were: intensity of and a diagnosis of fibromyalgia. In addition, the GP had pain, as evaluated with the Spanish version of the McGill to certify that no signs or symptoms frequently associ- Pain Questionnaire (MGPQ) (McGill Pain Questionnaire, ated with specific LBP or potentially severe illnesses Melzac, 1975) [22, 23], which assesses 3 parameters with 3 were present. Detailed information about the recruit- dimensions (sensorial, affective and evaluative): Total In- ment procedure has been published elsewhere [12]. tensity Score (scale 0–14), Current Intensity Score (scale 0–5) and Visual Analogical Scale (VAS, scale 0–10); and Randomization (see flowchart-Fig. 1) the mental and physical health-related quality of life, mea- Randomization was by cluster, and the randomization unit sured with the Spanish version of Short Form 12 version 1 was the PHCC. A cluster design was used because the [24] (SF-12, scale 0–100; lower scores indicate worse intervention was delivered to groups and to minimize con- health related quality of life). tamination. After the PHCC agreed to participate, they The main independent variable was the intervention were allocated either to the control or the intervention arm: biopsychosocial multidisciplinary intervention, or group. During recruitment, the GPs of the PHCC, who usual clinical care. knew about the allocation (intervention or control), identi- fied the patients consulting for new episodes of subacute Data collection and follow-up LBP. The patients who met the inclusion criteria were in- All participants were invited to attend the PHCC for vited to participate, without knowing the allocation of their outcome assessments. They were assessed at baseline PHCC. All patients signed the informed consent form. and at 3 and 12 months. To maximise patient’s adher- ence and to avoid loss of participants, patients received a Intervention design phone call at 6 months. Detailed socio-demographic and Table 1 shows the treatment components of the inter- clinical variables have been published elsewhere [12]. vention and the control groups. Both groups received For each assessment, the same two expert psychologists guideline-based pharmacological treatment. Participants made up to three phone calls at different times during the in the control group received usual clinical care, based day to book the appointments and performed the outcome on the Clinical Guidelines for Lumbar Spine Disorders measures by interviewing the participant, collected infor- in Adults published by the Catalan Institute of Health mation by reviewing medical records, contacted the pa- [15]. Patients allocated to the intervention group re- tient’s GP to inquire about their development (compliance ceived the same care described for the control group, and factors associated with low back pain) and answered plus the Spanish version of the educational booklet “The questions about the study. A senior psychologist specia- Back Manual” [16, 17] and some audio visual materials. lised in pain management conducted the intervention. The intervention was conducted by a GP and/or nurse, a psychologist and a physiotherapist. The programme Sample size lasted a total of 10 h, as explained in Table 1. Sessions The sample-size was calculated based on change in took place during the week and lasted between 90 and RMDQ at three months of follow up. To allow for the 120 min. To maximise participant adherence to the cluster randomization by PHCC, we considered an intra- group sessions, different times were offered. Each group class correlation coefficient of 0.05. In order to detect a Mas et al. BMC Health Services Research (2019) 19:962 Page 4 of 9 Fig. 1 Study Flow chart. Notes: PHCC = Primary Health Care Centres; GP = General practitioner; ICS = Catalan Institute of Health. LBP = Low Back Pain difference of 2.5 points between the two intervention based on an intention-to-treat principle. The analysis arms with a standard deviation of 5.7, an alpha error of was performed at the individual level using cluster data 0.05, a beta error of 0.20, and a 20% dropout rate, a sam- methods [25]. ple size of 348 subjects was required per intervention The intervention effect at each follow-up was assessed arm, with a the total number of PHCCs of 36. PASS 15 using the change (follow-up minus baseline) in the inter- “Test fot Two Means in a Cluster-Ramndomized Design” vention group minus the change in the control group in module (Utah, USA, ncss.com/software/pass) was used the outcomes. to calculate sample size. To address potential biases due to incomplete follow- up, multiple imputation by chained equations with 100 Statistical analysis imputed datasets was applied to outcomes and covari- Data were analysed in accordance with CONSORT ates [26–28]. Estimates from each imputed dataset were guidelines, extension to cluster randomized trials, and combined following the rules outlined by Rubin [29]. Mas et al. BMC Health Services Research (2019) 19:962 Page 5 of 9 Table 1 Components of the biopsychosocial multidisciplinary intervention and usual care OBJECTIVE THEORY PROGRAM PRACTICAL PROGRAM INTERVENTION GP + Nurse Answer queries, demystify concepts about LBP Basics of anatomy and Pain mechanisms, types, causes and GROUP 2h and promote adherence to the intervention biomechanics of the spine susceptibility factors. Healthy life habits, concerns and beliefs about LBP. Physiotherapist Provide tools on exercises/postures to avoid Body posture, ergonomics Relaxation exercises (breathing), 4h pain and improve quality of life and benefits body awareness and postural control Psychologist Provide participants with cognitive-behavioural Influence of cognition, Relaxation guidelines, cognitive 4h therapy techniques. emotions and behaviour in restructuring and time pain management. Assertiveness and problem solving, life values. CONTROL Clinical Practice Guidelines for Lumbar Spine Disease in Adults GROUP • Patient education, give reassuring and positive information about the benign nature of LBP, offer written information including specific advice. • Advise avoiding bed rest and encourage the person to be physically active and continue with normal activities as far as possible. • Consider offering a structured physical exercises program tailored to personal preferences. • Physical exercise should be introduced gently at first (walking, cycling and swimming) and progressively increased in intensity. • Recommend attendance to the “Back school” after six weeks to those patients who have resumed their daily tasks. • Prescribe pharmacological treatment according to established guidelines. GP General practitioner, LBP Low Back Pain After imputation, the distribution of observed and im- variables, with no statistically significant differences be- puted values was practically equal. tween groups. Multivariate regression analysis of each outcome vari- Table 3. In the adjusted analysis of the RMDQ outcome, able was performed for the imputed datasets, taking into the intervention group improvement more than the con- account the cluster effect in the models. We conducted trol group at 3 months (− 1.33 points, 95% CI: − 2.22 to − linear or logistic mixed-effects model and linear or logis- 0.45, p = 0.005) and at 12 months (− 1.11 points, 95% CI: tic regression adjusting the standard error for the cluster − 2.08 to − 0.13, p = 0.027). The intervention group effect of the PHCC. The final models were adjusted for presented a significant difference. A minimal clinically im- age, gender, baseline outcome measurement, and the sig- portant difference was achieved in both groups, with a nificant confounder and significant interaction variables. difference over 3.5 points in the intervention group com- We used mixed models and the function “mi estimate” pared with baseline at each time-point (3.8 RMDQ points in Stata. In these models, we added the variable PHCC at 3 months and 5.1 RMDQ points at 12 months). as a cluster/multilevel effect. The linear mixed model Regarding the level of pain in the adjusted analysis, a was used in the cluster data, with two models for each marginal difference was observed at 12 months in total time-point comparing changes at 3 months and at 12 intensity, in the intervention group (− 0.69 points; 95% months with baseline data. CI: − 1.41 to 0.02; p = 0.058). However, the intervention Statistical significance was set at P < 0.05 (2-tailed). group presented a significant differences at 3 months for The analyses were performed using Stata/SE version current intensity score (− 0.32 points; 95% CI: − 0.63 to 14.2 for Windows (Stata Corp. LP, College Station, − 0.02; p = 0.040) and for VAS score (− 0.77 points; 95% TX, USA). CI: − 1.53 to − 0.01; p = 0.046). The outcome of SF-12 increased in both groups during Results the follow-up period, but no statistically significant dif- A total of 501 subjects were included in the study; 262 ferences between groups on the physical and mental subjects (13 PHCC) were allocated to the intervention health domains were observed. group and 239 subjects (26 PHCC) to the control group. After 3 and 12 months, 421 (84%) and 387 (77.2%) par- Discussion ticipants provided data, respectively. The losses were The aim of this study was to evaluate the effectiveness of a due to work incompatibility, caregiving duties and lack multidisciplinary biopsychosocial intervention in an active of interest in the study (see Flowchart, Fig. 1). In general population with non-specific subacute LBP. The results patients who dropped out were significantly younger. shown in Table 3 were obtained with multiple imputation, Mean age of participants at baseline was 46.8 (SD: although similar values were obtained without multiple 11.5) years and 64.7% were women. Table 2 shows base- imputation (data not shown). The main results indicate line socio-demographic characteristics and clinical statistically significant differences regarding disability and Mas et al. BMC Health Services Research (2019) 19:962 Page 6 of 9 Table 2 Baseline socio-demographic characteristics and clinical variables Total Control group Intervention group No. of PHCC / No. of patients 39/501 26/239 13/262 Socio-demographic characteristics: Age (years), mean (SD) 46.8 (11.5) 46.4 (11.1) 47.2 (11.9) Sex (female), n (%) 324 (64.7) 145 (60.7) 179 (68.3) Educational level, n (%) (n = 499) ▪ Illiterate or primary school only 122 (24.4) 61 (25.6) 61 (23.4) ▪ Secondary school 274 (54.9) 134 (56.3) 140 (53.6) ▪ University 103 (20.6) 43 (18.1) 60 (23.0) Paid job (yes), n (%) 369 (73.7) 188 (78.7) 181 (69.1) Clinical variables: Body mass index (kg/m2) classification (n = 500), mean (SD) ▪ Normal weight 222 (44.4) 105 (43.9) 117 (44.8) ▪ Overweight 195 (39.0) 94 (39.3) 101 (38.7) ▪ Obesity 83 (16.6) 40 (16.7) 43 (16.5) Abbreviations: SD standard deviation, LBP Low Back Pain, PHCC Primary Health Care Centre. Data are mean (SD) or n(%) pain intensity, with a small effect in the intervention moderate short-term improvement in disability and pain group. Although greater in the intervention group, min- reduction, but no statistically significant improvement imal clinically important differences in disability were after 1 year [32]. achieved in both groups. No differences were observed re- We believe that the perceived improvement of disabil- garding quality of life. ity may be sufficient for some participants but not all. The results of this trial agree with some studies on Considering that over 50% of participants in this subacute and chronic LBP, where moderate quality intervention were women, some studies indicated that evidence showed efficacy in contrast with a non- patients with greater disability and worse quality of life multidisciplinary rehabilitation [8, 10]. Our findings pro- were frequently women who suffered also from somatic vide new information on the role of multidisciplinary and mood co-morbidities and perceived higher levels of biopsychosocial interventions delivered in groups in the pain [33–35]. In addition, Chow and colleagues pointed primary care setting. at factors that had been associated with persistent disab- The initial study sample consisted of 696 participants ling LBP, namely maladaptive pain coping behaviour, and 36 PHCCs. Eventually, 501 participants from 39 high baseline functional impairment, and low general PHCCs were recruited, since the period of recruitment health status [36]. was not extendable. Recruitment bias was detected, ap- According to the MGPQ, short-term pain reduction parently the GPs of the intervention PHCC were more was observed to be slightly better in the intervention motivated to recruit. To balance the number of partici- group. However, some research indicates that minor im- pants, more PHCCs were included in the control group. provements might be underestimated by patients with The minimal clinically important difference for disabil- LBP that return to their activity or work when non- ity was over 2.5 RMDQ points. Kovacs et al. (2007) disabling pain persists [37, 38]. Other studies confirm showed that an improvement in disability below 2.5 positive results in pain reduction within six weeks, and RMDQ points compared to their baseline in each group emphasize the benefits of a multidisciplinary interven- was clinically irrelevant in patients with subacute and tion [39]. For instance, Kamper and colleagues (2015) chronic LBP. When analyzing each time-point, differ- found that multidisciplinary biopsychosocial rehabilita- ences in disability were greater in the short term in both tion for subacute LBP was more effective than other groups. A meta-analysis of LBP showed noticeable physical interventions. In contrast, a systematic review short-term improvement during the first six weeks with conducted by the Cochrane Database 2017 did not find multiple treatments, but beyond this time improvement any evidence that this type of intervention was more slowed [30]. Some other studies reported that compared effective than other treatments for subacute LBP [4]. with usual care, multidisciplinary rehabilitation reduces The positive effect on disability and pain intensity was pain intensity and disability, mainly short-term (< 3 not corroborated by the results regarding quality of life. months) [31]. Specifically, Fritz (2015) evaluated the out- The lack of improvement in quality of life might be ex- comes of early physiotherapy versus usual care, with plained different levels of disability or pain at baseline. Mas et al. BMC Health Services Research (2019) 19:962 Page 7 of 9 Table 3 Changes in the Roland-Morris Disability, McGill Pain and SF-12 questionnaires between groups at follow-up (N = 501) Control Group (n = 239) Intervention Group (n = 262) Difference (95%CI) between group (IG - CG) Value Difference*(95% CI) Value Difference*(95% CI) change IG - CG P-value Adjusted difference** P-value RMDQ, mean (SD) Baseline 9.9 (5.3) 10.0 (5.2) three months 7.4 (5.5) −2.3 (−3.1 to −1.6) 6.2 (4.9) −3.8 (−4.5 to −3.2) −1.5 (− 2.5 to −0.5) 0.003* −1.33 (− 2.22 to − 0.45) 0.005* 12 months 6.0 (5.7) −3.8 (− 4.8 to − 2.9) 5.1 (4.9) −5.1 (−5.8 to − 4.3) − 1.2 (− 2.4 to − 0.0) 0.043* −1.11 (− 2.08 to − 0.13) 0.027* MPQ Total intensity score, mean (SD) Baseline 6.5 (3.1) 6.7 (3.1) three months 4.6 (3.6) −1.8 (−2.3 to − 1.3) 4.0 (3.6) − 2.7 (− 3.2 to − 2.2) − 0.9 (− 1.6 to − 0.1) 0.022* −0.49 (− 1.39 to 0.42) 0.294 12 months 3.6 (3.6) −2.8 (− 3.3 to − 2.2) 3.1 (3.2) −3.6 (− 4.1 to − 3.0) − 0.8 (− 1.6 to 0.0) 0.040* −0.69 (− 1.41 to 0.02) 0.058** Current Intensity score mean (SD) Baseline 2.6 (1.1) 2.5 (1.2) three months 1.7 (1.5) −0.9 (− 1.1 to − 0.7) 1.3 (1.4) −1.2 (− 1.4 to − 1.0) −0.3 (− 0.6 to 0.0) 0.083 −0.32 (− 0.63 to − 0.02) 0.040* 12 months 1.6 (1.4) −1.1 (− 1.3 to 0.8) 1.4 (1.3) −1.1 (− 1.3 to − 0.9) 0.0 (−0.3 to 0.3) 0.854 −0.18 (− 0.43 to 0.08) 0.162 VAS, mean (SD) Baseline 5.9 (2.3) 5.8 (2.3) three months 4.1 (3.3) −1.8 (−2.2 to − 1.3) 3.2 (3.2) − 2.7 (−3.1 to − 2.2) −0.9 (− 1.6–0.3) 0.004* −0.77 (− 1.53 to − 0.01) 0.046* 12 months 3.9 (3.2) −2.0 (− 2.5 to − 1.5) 3.6 (3.0) −2.3 (− 2.7 to − 1.9) − 0.3 (− 0.9 to 0.4) 0.404 −0.27 (− 0.88 to 0.34) 0.374 SF-12 Physical health, mean (SD) Baseline 40.7 (9.3) 41.9 (9.0) three months 45.3 (9.8) 4.2 (2.7 to 5.6) 46.5 (8.7) 4.5 (3.2 to 5.8) 0.4 (− 1.6 to 2.3) 0.716 0.55 (− 1.19 to 2.29) 0.520 12 months 46.2 (9.5) 5.0 (3.3 to 6.7) 47.0 (8.9) 4.9 (3.5 to 6.3) −0.1 (−2.3 to 2.1) 0.922 0.53 (−1.20 to 2.27) 0.532 SF-12 Mental health, mean (SD) Baseline 42.3 (12.4) 43.4 (12.8) three months 45.0 (13.2) 2.6 (0.7 to 4.6) 48.8 (12.0) 5.1 (3.4 to 6.9) 2.5 (−0.1 to 5.0) 0.061 2.56 (− 0.33 to 5.45) 0.082 12 months 47.0 (11.9) 5.0 (2.9 to 7.1) 48.9 (11.2) 5.5 (3.6 to 7.5) 0.5 (−2.3 to 3.4) 0.707 1.48 (−0.86 to 3.83) 0.206 Abbreviations: SD standard deviation, CI confidence interval, RMDQ Roland-Morris Disability Questionnaire (scale 0–24; lower scores indicate less disability), MGPQ McGill pain questionnaire; 3 dimensions (sensorial, affective and evaluative) with Total Intensity Score (scale 0–14), Current Intensity Score (scale 0–5) and Visual Analogical Scale (VAS, scale 0–10); SF-12 = 12-item short-form health survey version 1 (scale 0–100; lower scores indicate worse health related quality of life). * Differences were calculated between follow-up and baseline measurements. Mean differences are shown for quantitative outcomes and percentage differences for dichotomous outcomes. **All models were adjusted for the score at baseline, significant confounders and significant interaction variables. Intervention group minus usual care group, mean differences are shown for quantitative outcomes and odds ratios for dichotomous outcomes. Intervention group minus usual care group, mean differences are shown for quantitative outcomes and odds ratios for dichotomous outcomes. Total Intensity Score, VAS Pain Score and Mental Health were estimated with a mixed model considering the PHCC as random effect For some patients, the effect at follow-up might not Further research achieve their expectations of improvement. In patients Further research on complex interventions in LBP with levels of disability or pain that are moderate or low should consider the UK Medical Research Council at baseline, the effect of the intervention will have a bet- (MRC) framework, which consists of several phases that ter perceived impact. Other findings suggest that early can be iterated and that use qualitative and quantitative active physiotherapy can lead to improved outcomes in methods. This methodology includes the perspective of global health perception [34]. Interestingly, some authors the patient throughout the study in order to design the define the relationship between physical activity and the intervention based on the population needs, preferences risk of chronic LBP as a U-shaped distribution, i.e., both and experiences. The MRC methodology aims to design too little and excessive activity presented increased risks feasible, effective and sustainable interventions for the of chronic LBP and worse quality of life [36, 40, 41]. primary health care setting. Other factors that intervene in the evolution of LBP in- Finally, we must encourage the collaboration of pri- clude explaining the natural history of this condition mary care professionals and the community in everyday and the implementation of the guidelines that seek to re- clinical practice and in large scale, multidisciplinary lieve or minimize pain. interventions. Mas et al. BMC Health Services Research (2019) 19:962 Page 8 of 9 Limitations of the study Health and the “Beca ICS per a la capacitació en investigació i realització del The number of missing data was similar for both groups doctorat a l’Atenció Primària”, by the IDIAP Jordi Gol together with the Catalan Institute of Health. The project received a research grant from the at three and 12 months of follow-up. However, the follow- Carlos III Institute of Health, Ministry of Economy and Competitiveness ing were more likely to drop out of the study: younger (Spain), through the Network for Prevention and Health Promotion in people; people with a lower consumption of analgesics; Primary Care, reference RD16/0007/0001, co-funded with European Union ERDF funds (European Regional Development Fund). and at 12 months follow-up, patients with lower family burden and better jobs. Availability of data and materials Some confounding factors might limit the conclusions The datasets used during the current study are available from the corresponding author on reasonable request. of this study. For instance, the differences in the profile of patients, since they were allocated by PHCC and so- Ethics approval and consent to participate cioeconomic status was not considered. This study was approved by the Clinical Research Ethics Committee of the One of the most important limitations of the study Institute of Research in Primary Health Care IDIAP Jordi Gol (P07/25). Confidentiality and anonymity of the data were ensured in accordance with was recruitment, since more PCHH were allocated to the 15/1999 Data Confidentiality Law. the control group than the intervention group. Informed Consent: The information was provided orally as well as written. Study subjects had sufficient opportunity to ask questions regarding study details. Informed consent followed the guidelines of the Helsinki Declaration. Conclusions Data confidentiality: Confidentiality and anonymity of the data was ensured, The main conclusion of this study is that a multidiscip- both in the implementation phase of the project and in presentations or linary biopsychosocial intervention in a working popula- publications resulting there from. Individual data was codified to ensure anonymity. Only researchers and monitors had access to the data. tion with non-specific subacute LBP has a small positive effect on disability and intensity of pain. Although Consent for publication greater in the intervention group, minimal clinically im- Not applicable. portant differences were achieved in both groups. The Competing interests results did not show any differences on quality of life. None declared. The results of this biopsychosocial multidisciplinary Author details intervention agree with previous studies, which also 1 Gerència Territorial de Barcelona, Catalan Institute of Health, c/ Sardenya, show limited effectiveness. The main advantage of deliv- 375, Entl, 08025 Barcelona, Spain. 2Institut Universitari d’Investigació en ering an intervention from a PHCC is geographical Atenció Primària (IDIAP Jordi Gol), Gran Via Corts Catalanes, 587, àtic, 08007 Barcelona, Spain. 3Department of Paediatrics, Gynaecology and Obstetrics proximity, which should result in higher adherence. In and Preventive Medicine, Universitat Autònoma de Barcelona Faculty of patients with pain, travelling long distances has usually a Medicine-Bellaterra, 08193 (Cerdanyola del Vallès), Spain. 4Department of negative physical and financial impact. Finally, the con- Basic, Evolutionary and Educational Psychology, Universitat Autònoma de Barcelona Faculty of Psychology, Building B. Campus de la UAB, Bellaterra, stellation of symptoms presented by patients with LBP 08193 Barcelona, Spain. 5Hospital de la Santa Creu i de Sant Pau, 08041 still constitutes a challenge for medical and surgical Barcelona, Spain. 6Primary Care Centre Sant Martí, SAP Litoral, 08020 decision making. Barcelona, Spain. 7SAP Support to Diagnosis and Treatment, Gerència Territorial de Barcelona, Catalan Institute of Health, 08001 Barcelona, Spain. 8 Abbreviations Unit of Functional Rehabilitation, Department of Rheumatology, Hospital GP: General practitioner; LPB: Low back pain; MGPQ: McGill pain Clínic Universitari, 08036 Barcelona, Spain. 9Primary Care Centre Sant Andreu, questionnaire; NSAIDs: Non-steroidal anti-inflammatory drugs; PHCC: Primary SAP Muntanya, 08030 Barcelona, Spain. 10Primary Care Centre Sant Antoni, health care centre; QoL: Quality of life SF12; RMDQ: Roland morris disability SAP Esquerra, 08015, Barcelona, Spain. questionnaire; VAS: Visual analogical scale Received: 23 January 2019 Accepted: 4 December 2019 Acknowledgements We thank all patients and healthcare professionals who participated in the study for taking the time to share their experiences. References We also thank FM Kovacs for sharing invaluable experience and suggestions. 1. Zale EL, Lange KL, Fields SA, Ditre JW. The relation between pain-related fear and disability: a meta- analysis. J Pain. 2013;14:1019–30. Authors’ contributions 2. Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M, et al. Years TR, MIF, EM, JM, MR, MJG, EN, MN, MP, RM, MB, CV and EPR designed the lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries study. The study was coordinated by AB and EPR. TRB, MIF, EM, JM, MP, RM 1990-2010: a systematic analysis for the global burden of disease study and EPR conducted the study. EM, JM, MJG and MP implemented the 2010. Lancet. 2012;380:2163–96. intervention. RR, AB, TR, MC, EPR and TLJ analysed and interpreted the data. 3. Generalitat de Catalunya, Departament de Salut. Informe 2015 de l’Enquesta RR, AB, TR, MAM, EPR, CV and TLJ drafted the manuscript and incorporated de salut de Catalunya (ESCA). ESCA. 2016;:99. http://salutweb.gencat.cat/ the amendments suggested by the co-authors. All authors have read and web/.content/_departament/estadistiques-sanitaries/enquestes/Enquesta- approved the final version of the manuscript. de-salut-de-Catalunya/Resultats-de-lenquesta-de-salut-de-Catalunya/ documents/esca_2015.pdf. Accessed 26 Aug 2019. Funding 4. Marin TJ, Van Eerd D, Irvin E. Cocuban R. Malmivaara A, et al. This study was supported by the Marató TV3 and a grant from the Primary Multidisciplinary biopsychosocial rehabilitation for chronic low back pain. Care Management Barcelona Ciutat-ICS. The funding body had not any role Cochrane Database Syst Rev: Koes BW; 2017. in the design of the study, the collection, analysis, and interpretation of data 5. Fernández-De-Las-Peñas C, Alonso-Blanco C, Hernández-Barrera V, Palacios- and in writing the manuscript. Ceña D, Jiménez-García R, Carrasco-Garrido P. Has the prevalence of neck RR was awarded the following two grants: “Beca Xb, per a la Realització de pain and low back pain changed over the last 5 years? A population-based Tesis Doctorals”, by the Gerència territorial de Barcelona, Catalan Institute of national study in Spain. Spine J. 2013;13:1069–76. Mas et al. BMC Health Services Research (2019) 19:962 Page 9 of 9 6. Foster N, Anema J, Cherkin D. al. Et. Prevention and treatment of low disability index (ODI), progressive Isoinertial lifting evaluation (PILE), Quebec back pain: evidence, challenges, and promising directions. Lancet. 2018; Back pain disability scale (QBPDS), and Roland-Morris disability 6736:1–16. questionnaire. Arthritis Care Res. 2011;63(SUPPL. 11):158–73. 7. Kovacs FM, Fernandez C, Cordero A, Muriel A, Gonzalez-Lujan L. Gil del Real 31. van Middelkoop M, Rubinstein SM, Verhagen AP, Ostelo RW, Koes BW, van MT. Non-specific low back pain in primary care in the Spanish National Tulder MW. Exercise therapy for chronic nonspecific low-back pain. Best Health Service: a prospective study on clinical outcomes and determinants Pract Res Clin Rheumatol. 2010;24:193–204. of management BMCHealth ServRes. 2006;6:57. 32. Fritz JM, Magel JS, McFadden M, Asche C, Thackeray A, Meier W, et al. Early 8. Richmond H, Hall AM, Copsey B, Hansen Z, Williamson E, Hoxey- physical therapy vs usual care in patients with recent-onset low back pain: a Thomas N, et al. The effectiveness of cognitive behavioural treatment randomized clinical trial. JAMA - J Am Med Assoc. 2015;314:1459–67. for non-specific low back pain: a systematic review and meta-analysis. 33. Stefane T, dos Santos A, Marinovic A. Chronic low back pain : pain intensity PLoS One. 2015;10:1–20. , disability and quality of life. Acta Paul Enferm. 2013;26:14–20. 9. Anthony Delitto, PT, PhD. Steven Z. George, PT, PhD. Linda Van Dillen, PT, 34. Klemenc-Ketis Z. Predictors of health-related quality of life and disability in PhD. Julie M. Whitman, PT, DSc. Gwendolyn A. Sowa, MD P. Low Back Pain. patients with chronic non-specific low back pain. Zdr Vestn. 2011;80:379–85. J Orthop Sport Phys Ther. 2012. 35. Korovessis P, Repantis T, Zacharatos S, Baikousis A. Low back pain and 10. Qaseem A, Wilt TJ, McLean RM, Forciea MA. Noninvasive treatments for sciatica prevalence and intensity reported in a Mediterranean country: acute, subacute, and chronic low back pain: a clinical practice guideline ordinal logistic regression analysis. Orthop. 2012;35:e1775–84. from the American College of Physicians. Ann Intern Med. 2017;166:514–30. 36. Chou R, Shekelle P. Will this patient develop persistent disability low back 11. Kovacs FM, Abraira V, Zamora J, Fernández C. The transition from acute to pain? Jama. 2010;303:1295–302. subacute and chronic low back pain. Spine (Phila Pa 1976). 2005;30:1786–92. 37. Cherkin DC, Sherman KJ, Balderson BH, Cook AJ, Anderson ML, Hawkes RJ, 12. Rodriguez-Blanco T, Fernandez-San-Martin I, Balague-Corbella M, Berenguera et al. Effect of mindfulness-based stress reduction vs cognitive behavioral A, Moix J, Montiel-Morillo E, et al. Study protocol of effectiveness of a therapy or usual care on back pain and functional limitations in adults with biopsychosocial multidisciplinary intervention in the evolution of non- chronic low back pain: a randomized clinical trial. JAMA - J Am Med Assoc. specific sub-acute low back pain in the working population: cluster 2016;315:1240–9. randomised trial. BMC Health Serv Res. 2010;10:12. 38. Choi BKL, Verbeek JH, Tam WWS, Jiang JY. Exercises for prevention of 13. National Institute for Health and Clinical Excellence (NICE). Low back recurrences of low-back pain. Occup Environ Med. 2010;67:795–6. pain: Early management of persistent non-specific low back pain. Heal 39. Kindermans HPJ, Huijnen IPJ, Goossens MEJB, Roelofs J, Verbunt JA, Vlaeyen San Fr. 2009. JWS. “Being” in pain: the role of self-discrepancies in the emotional 14. Kovacs FM, Abraira V, Zamora J, Fernandez C. The transition from acute to experience and activity patterns of patients with chronic low back pain. subacute and chronic low back pain: a study based on determinants of Pain. 2011;152:403–9. quality of life and prediction of chronic disability. Spine. 2005. 40. Heneweer H, Vanhees L, Picavet HSJ. Physical activity and low back pain: a 15. Bordas JM, Forcada J, GarcÍa JA, Joaniquet FX, Pellisé F, Mazeres O, et al. U-shaped relation? Pain. 2009;143:21–5. https://doi.org/10.1016/j.pain.2008. Guies de práctica clínica i material docent. Patologia de la columna lumbar 12.033. en l’adult. Institut Català de la Salut: Barcelona; 2004. 41. Lin CWC, McAuley JH, MacEdo L, Barnett DC, Smeets RJ, Verbunt JA. 16. Kovacs F. El Manual de la Espalda. Palma de Mallorca: Fundació Kovacs; Relationship between physical activity and disability in low back pain: a 2002. systematic review and meta-analysis. Pain. 2011;152:607–13. 17. Kovacs FM, Royuela A. The Efficacy of a Short Education Program and a Short Physiotherapy Program for Treating Low Back Pain in Primary Care A Publisher’s Note Cluster Randomized Trial The Efficacy of a Short Education Program and a Springer Nature remains neutral with regard to jurisdictional claims in Short Physiotherapy Program for Treating Low Back Pain in Pri. Spine (Phila published maps and institutional affiliations. Pa 1976). 2010;35 July 2016:483–496. 18. 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;8:141–4. 19. Kovacs FM, Llobera J, Gil del Real MT, Abraira V, Gestoso M, Fernandez C, et al. Validation of the spanish version of the Roland-Morris questionnaire. Spine. 2002. 20. Bombardier C, Hayden J, Beaton DE. Minimal clinically important difference. Low back pain: outcome measures. JRheumatol. 2001;28:431–8. 21. Kovacs FM, Abraira V, Royuela A, Corcoll J, Alegre L, Cano A, et al. Minimal clinically important change for pain intensity and disability in patients with nonspecific low back pain. Spine. 2007;32:2915–20. 22. Melzack R. The McGill pain questionnaire: major properties and scoring methods. Pain. 1975;1:277–99. 23. Bejarano PF, Noriego RD, Rodriguez ML, Berrio GM. Evaluación del dolor: adaptación del cuestionario de Mc Gill. Rev Col Anest. 1985;13:321–51. 24. Luo X, George ML, Kakouras I, Edwards CL, Pietrobon R, Richardson W, et al. Ovid: Reliability, Validity, and Responsiveness of the Short Form 12-Item Survey (SF-12) in Patients With Back Pain. Spine (Phila Pa 1976). 2003;28: 1739–45. 25. Ukoumunne OC, Gulliford MC, Chinn S, Sterne JA, Burney PG. Methods for evaluating area-wide and organisation-based interventions in health and health care: a systematic review. Heal TechnolAssess. 1999;3:iii–92. 26. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med. 2011;30:377–99. 27. Graham JW. Missing data: Analysis and design. New York, NY: Springer. In: Missing Data Analysis Design. 2016. p. C2012;2012. 28. Lee KJ, Simpson JA. Introduction to multiple imputation for dealing with missing data. Respirology. 2014;19:162–7. 29. Wiley J. Multiple imputation for nonresponse in surveys. 1987. 30. Smeets R, Köke A, Lin CW, Ferreira M, Demoulin C. Measures of function in low Back pain/disorders: low Back pain rating scale (LBPRS), Oswestry