(http://www.amreg.org/). The main data elements include information of medical centers and acupuncturists, characteristics of patients with CLBP, acupuncture operation details, outcomes and safety information. Informed consent requires scanning and uploading. The names of diseases related to chronic back pain are identified by International Classification of Diseases (ICD-10,2010 Version) [37]; TCM syndromes are confirmed by Classification and codes of diseases and ZHENG of traditional Chinese medicine (GB/T 15657–1995) [38]. The acupuncture point vocabulary adopts the WHO Standard Acupuncture Point Locations in the Western Pacific Region (A WPRO Publication, 2008) [39]. All data elements will be collected using the established terminology specifications, data dictionaries. Most of the data use structured information entry, allowing free text entry to a certain extent. Platform support Pinyin code retrieval and input and fuzzy matching input to ensure the standardization of registration information and facilitate the statistical analysis of data. Under the premise of ensuring the necessary information, entries should be as little as possible to minimize the burden on clinicians. Specific entries and variables are available in Table 1. Considering the characteristics and clinical practice of acupuncture therapies, we referenced Standards for Reporting Interventions in Clinical Trials of Acupuncture (STRICTA) [27] to building data dictionaries and Wei et al. BMC Musculoskeletal Disorders (2019) 20:488 Page 4 of 10 frameworks. Structured, standardized, and logical metadata will be accumulated to reflect the real clinical process of acupuncture treatment for CLBP. Baseline demographic data, CLBP characteristics, patientcentered outcomes of the effectiveness and safety will be truthfully recorded. Specific data element entry and minimum dataset in the register can be found in Table 2. Acupuncture treatment and combined treatment should be recorded at each visit. Changes in symptoms, ODI, and NRS scores were recorded before and after treatment for effectiveness evaluation. Statistical methods Missing data Although we have internal quality control methods to ensure data integrity, follow-up nonresponse shall be an inescapable challenge in a prospective observation study. Fortunately, we will recruit experienced physicians with stable patient population (at least 5 years of acupuncture experience or middle and senior titles), so the bias for lack of follow-up could be controlled partially. Real-time monitoring of the electronic network platform shall control missing data less than 20%. According to the classical rules of missing classification, follow-up nonresponse belongs to missing not at random (MNAR) [40]. Multiple imputation invented by Rubin shall be adopted to process missing data [41]. Analytic approach Data with poor compliance, non-compliance with the study program or missing important information are not included in the analysis. Data processing and statistical analysis will use SPSS version 24 (IBM Corp., Armonk, NY, USA) and SAS Version 9.4 (SAS Institute Inc., NC 27513–2414, USA). We will select 1-week,4-week,12-week, assessments respectively at evaluation time point on the ODI and NRS index. Firstly, baseline characteristics comparability needs to be assessed between the propensity-matched groups. Model adjustment suggested by Rosenbaum and Rubin will be adopted to conquer residual covariance imbalances [42]. Secondly, multivariate linear regression Table 1 Metadata of main items and variables Major data sources Main items Property Main variables and labels Medical Centre Institutional background Indispensable Medical Organizations, Location and hospital attributes, Medical institution level Registration of contribution Indispensable Number of recruited practitioners, Number of enrolled patients Physicians Demographic Information Indispensable Name, Age, Gender, Place of residence, Contact information, Educational background Practitioner background Indispensable Medical Organizations, Years of employment, Professional qualification (senior or intermediate professional title) Patients Demographic Information Indispensable Name, Age, Gender, Nationality, ID number, Residence, Occupation, Marriage status. Educational Qualifications, Contact information CLBP Information Indispensable Visit date, Chief complaint, Course of disease, Location of pain, Characteristics of pain, causes of pain aggravation, Medical history, Causes of pain relief, Back function, Concomitant symptoms, History of acupuncture and Reception, auxiliary examination Diagnosis information Indispensable Disease diagnosis, Syndrome differentiation of TCM etiology and meridians, Complicating disease Lumbar imaging Optional X ray, CT examination, Magnetic Resonance Imaging Informed Consent Indispensable NA Acupuncture treatment process Acupuncture rationale Indispensable Acupuncture date, Acupuncture style, Acupoint selection principle, Operation of acupoints Acupuncture details Indispensable Response elicited, Needle retention time, Needle stimulation, Needle type (gauge, length), Adjunctive therapy, Needle parameter, Co-intervention Acupuncture regimen Indispensable Number of treatment session, Frequency of treatment Combined therapy Indispensable