This study adhered to the Declaration of Helsinki, and we assured that the data would be used for scientific research only. Participants were informed in detail about the nature of the study. Spontaneous written consent was taken from the participants and Bangla informed consent form was attached with the questionnaire. Each participant enjoyed the right to participate or refuse to participation. They could withdraw their participations from the study at any stage by contacting the principal investigator and citing their ID number or name. Data taken from the participants were regarded as confidential and kept locked under the principal investigator. The participant’s data were kept anonymous in datasets and were tracked by using unique ID numbers. If any study subject became sick during the survey, proper advice and clinical management was given. Ethical clearance was obtained from the Ethical Review Committee of Bangabandhu Sheikh Mujib Medical University (BSMMU) in 2011 (Memo number: BSMMU/2011/6045) before starting data collection.

This case control study was carried out during the period of June to September 2011 at Sonargaon Upazila, Narayanganj district, consisting of 19 small villages located approximately 35 km from the capital city. The data collection period was from July to August 2011. Due to the presence of some industries nearby this predominant rural area had some influence of urbanization. There were a diverse range of occupations represented, including agricultural workers, weavers, garment workers, salesmen, and so on. All adults aged 18 years old or above from both sexes were considered as eligible population.

All the eligible adults of the study area were requested to attend the satellite Community oriented program for control of rheumatic diseases (COPCORD) camp according to a roster. Out of 5,005 eligible adults aged 18 years old and above, 4,850 participated in the camp and the other 155 were non-responsive. At the camp, a Bangla version of the World Health Organization (WHO)-International League of Associations for Rheumatology (ILAR) COPCORD Core English questionnaire was administered by the field enumerators. ^{14 , 15} By using WHO-ILAR COPCORD questionnaire, 1,315 patients were identified who had any rheumatic condition and the rest of the 3,535 patients had no rheumatic conditions. Those who were identified as rheumatic patients, of them 494 had LBP and they were sent to the rheumatologists. The rest of the 821 patients had rheumatic conditions other than LBP and were excluded from the study. Among the 494 LBP patients, 343 were clinically diagnosed as NSLBP patients using Asia Pacific League of Associations for Rheumatology (APLAR)-COPCORD questionnaire for NSLBP ¹⁶ by rheumatologists and considered as cases for the study. NSLBP is defined as low back pain not attributable to a recognizable, known specific pathology ( e.g., infection, tumor, osteoporosis, lumbar spine fracture, structural deformity, inflammatory disorder, radicular syndrome, or cauda equina syndrome). ¹⁷ To avoid any potential source of bias the diagnosis of NSLBP was made by exclusion of specific pathology with proper history taking, physical examination, imaging and sometimes opinion of expert rheumatologists. In case of any diagnostic difficulties investigators discussed among themselves to decide. Sometimes assistance of radiological investigations was taken. Those who had NSLBP but also had co-morbid illnesses such as symptomatic bronchial asthma, ischemic heart disease and intellectual disability were excluded. The rest of the 151 patients had LBP due to secondary causes and were excluded from the study.

From the patients who had no rheumatic conditions as confirmed by WHO-ILAR COPCORD questionnaire (n=3,535), 343 were recruited as controls of the study by matching with sex and age with ±5 years. A second recall of those who did not report to the camp was done to finally declare them as non-respondents. Details of the selection process is given in Figure 1. Questionnaires, blank consent forms and study information can be found as Extended data. ¹⁸

We used community specific APLAR-COPCORD questionnaire validated by Siddiqui et al. ¹⁶

History of chronic disease: Respondents were asked whether they are currently taking any medication for diabetes and/or hypertension, or have they ever been said to have raised blood glucose or blood pressure by any physician or qualified health care worker. If they said ‘yes’ to any of the above two then they were considered as having a history of chronic disease. ¹⁶

History of trauma: Respondents were asked whether they had any major accident/physical trauma over the last 12 months. ¹⁶

Prolonged continuous sitting: Respondents were asked whether they have to sit for ≥ 2 hours/day continuously irrespective of their daily activities or occupation over the last 12 months. Such as desk workers, driver, tailors. ¹⁶

Squatting: Respondents were asked whether they have to squat or sit bending knee for ≥ 1 hours/day continuously irrespective of their daily activities or occupation over the last 12 months. Such activities include sitting without stool/floor or yard scraping/wiping/cooking etc. ¹⁶

Bending of waist: Respondents were asked whether they have to bend their waist for ≥ 1 hours/day continuously irrespective of their daily activities or occupation over the last 12 months. Such as agricultural work (without lifting weight), manual weeding, gardening, fishing, laundering, cobbling, potter, blacksmith, weaving, manual brick crushing, carpenter, plasterer, corpse worker, manual brick maker, coppersmith etc. ¹⁶

Regular lifting or carrying heavy load: Respondents were asked whether they have to lift or carry heavy load regularly irrespective of their daily activities or occupation over the last 12 months. Such as porters, day laborer, construction worker, agriculture work, lifting water filled jar, cultivation in hilly land, peddler, tea plucking, industrial worker, brick crushing etc. ¹⁶

Prolong standing: Respondents were asked whether they have to stand for ≥ 2 hours/day continuously irrespective of their daily activities or occupation over the last 12 months. Such as seals man, nurse, street vendor, barber, security guards, bus helper/conductors, traffic police, teacher, receptionist, health worker, etc. ¹⁶

Depression: Depression in respondents was measured using the General Health Questionnaire (GHQ-12) ¹⁹ consisting of 12 items, each of which is evaluated by four indices. The GHQ-12 has satisfactory reliability with good sensitivity and specificity. Four-point Likert scoring method was used. Scoring ranged from 0 to 36. A score of >15 was considered as depressed.

Data were analyzed using Epi Info 7 software (RRID:SCR_021682). The sociodemographic characteristics and risk variables were presented in terms of numbers and percentages. Age and body mass index were described using the mean and standard deviation. A total of 12 factors such as, ‘age’, ‘sex’, ‘history of chronic disease’, ‘history of trauma’, ‘prolonged continuous sitting ≥ 2 hours/day’, ‘squatting ≥ 1 hour/day’, ‘bending of waist ≥ 1 hour/day’, ‘regular lifting or carrying heavy load’, ‘prolong standing ≥ 2 hours/day’, ‘Strenuous physical activity’, ‘overweight (body mass index ≥ 25 Kg/m ²)’, and ‘depression’ were examined for their relationship with NSLBP. The risk factors were identified using conditional logistic regression analyses and are shown as odds ratios with 95% confidence intervals. All missing data were dropped from the analysis. STROBE checklist for case-control study was followed to prepare the manuscript. ²⁰

To the best of our knowledge, so far, our study is the first community-based case-control study on NSLBP conducted in Bangladesh. This study attempted to cover all possible aspects of etiological factors. Age and sex matched cases and control subjects were enrolled. Culturally adapted and validated tools were used. However, this study also has weaknesses. The study area was selected purposively and its proximity to the BSMMU may not represent rural Bangladeshi population at large. Furthermore, the population was selected purposively. Performing imaging was very difficult because the diagnostic centers were far away from the study area, and we were limited by fund constraints. Thus, X-rays were only performed in a few of the cases.

Zenodo: Risk Factors of Non-specific Low Back Pain in a Rural Community of Bangladesh. https://doi.org/10.5281/zenodo.6709850. ²¹

Zenodo: Risk Factors of Non-specific Low Back Pain in a Rural Community of Bangladesh (extended data). https://doi.org/10.5281/zenodo.6824159. ¹⁸

This project contains the following extended data: -

Questionnaire Part 1.pdf