Background: Non-specific low back pain is a growing problem in society. No treatment have shown satisfying results to reduce pain or disability for patients with non-specific low back pain, and 1-18% of these patients develop chronic low back pain. STarT Back Screening Tool (SBST) is an instrument for sub grouping patients with non-specific low back pain into low, medium or high risk of developing chronic low back pain and then modifying the treatment after the different needs of every patient. The purpose of this study was to do a systematic review, investigating validation of the SBST to evaluate the justification of its use by clinicians. Method: Pubmed, Cinahl and Medline was searched in February 2017 for studies investigating criterion validity, construct validity and content validity of the SBST. The author (JG) assessed risk of bias and extracted relevant data following the procedures of PRISMA-statement. Result: 15 articles were identified for inclusion in the review. 8 articles investigated criterion validity, 11 articles investigated construct validity and 1 article investigated content validity. Predictive validity showed heterogeneous statistical analysis and varying results, a narrative result was presented that showed marginal benefits for the use of SBST for prediction of future outcome. Concurrent validity was measured with Spearman’s rank correlation coefficient in all investigated articles, showing results between 0.34-0.802. Discriminant validity was measured with Area under the curve analysis in all articles, scoring between 0.69-0.92. Convergent validity showed a Pearson’s correlation between 0.708-0.811 and a Spearman’s rank correlation between 0.35-0.74. Conclusion: Because of heterogeneity of the results it is not possible to draw conclusive conclusions. However, results tend to show limited evidence for the use of SBST as a predictive instrument for patients with non-specific low back pain.

Fatores psicossociais não são rotineiramente identificados na avaliação fisioterapêutica e podem influenciar no prognóstico de pacientes com dor lombar. O questionário STarT Back Screening Tool (SBST) foi criado para distinguir subgrupos dentre pacientes com dor lombar em relação ao prognóstico no tratamento considerando fatores físicos e psicossociais, classificando-os em baixo, médio e alto risco de mau prognóstico. Objetivo: Traduzir e adaptar transculturalmente o SBST para Língua Portuguesa do Brasil, testar a confiabilidade e validade de construto. Métodos: A primeira etapa consistiu na tradução, síntese, retro-tradução, revisão pelo grupo de tradução, pré-teste e avaliação dos documentos pelo Comitê. A versão pré-final foi aplicada em duas amostras de 52 pacientes cada, com dor lombar, ambos os sexos, idade acima de 18 anos e níveis educacionais variados. Para verificação da confiabilidade intra-avaliador, foram realizadas duas entrevistas em outra amostra de 50 pacientes, e os resultados analisados pelo Kappa ponderado quadrado. Também foram calculados a consistência interna usando de Cronbach (n=105), o erro padrão da medida (n=50) e a validade de construto (n=151), correlacionando o SBST-Brasil com as versões brasileiras do Oswestry Disability Index (ODI), Rolland Morris (Brasil- RM) e Fear Avoidance Beliefs Questionnaire para trabalho (FABQ-Work) e atividade física (FABQ-Phys). Resultados: O consenso foi atingido nas etapas de tradução e retrotradução, apenas o item 6 não foi compreendido e depois de reformulado foi reaplicado em outros 52 pacientes que não tiveram mais dúvida. A confiabilidade foi considerada substancial (0,79 - 95% IC 0,63 0,95), a consistência interna aceitável (0,74 pontuação total e 0,72 da subescala psicossocial), e o valor do erro de padrão da medida muito bom (1,9%). Em relação à validação, a pontuação total do SBST- Brasil mostrou alta correlação com o Brasil-RM (r=0,61), moderada correlação com o ODI (r = 0,56) e FABQ-Work (r=0,3) e fraca correlação com o FABQ-Phys (r = 0,18). Já a subescala psicossocial mostrou correlação moderada com o Brasil-RM (r =0,49) e ODI (r =0,47) e fraca correlação com o FABQ-Work (r =0,24) e FABQ-Phys (r =0,21). Além disso, a estratificação dos grupos de risco do SBST-Brasil foi compatível com os valores de classificação do ODI, Brasil-RM e FABQ-Phys para risco de mau prognóstico. Conclusão: O SBST-Brasil mostrou-se confiável e válido para avaliar a influência de fatores psicossociais em pacientes com dor lombar. / Psychosocial factors are not usually identified on physiotherapy assessment, what influences on low back pain prognosis. The questionnaire STarT Back Screening Tool\" (SBST) was created to identify patients with potentially treatment modifiable prognostic influenced by physical and psychosocial indicators which are relevant to initial decision making, classifying them into low, medium and high risk categories. Aim: Translation and cross-culturally adaptation of the SBST into Brazilian-Portuguese, assess its reliability and construct validity. Methods: The translation and cross-cultural adaptation process involved the steps of Translation, Synthesis, Back-translation, Committee Revision, Pretesting and Submission the documents to the Committee. The pre-final version was applied to two samples of 52 patients each, with low back pain, both sexes, aged over 18 years, from different education levels. After that, two interviews assessed intra-rater reliability on a different sample of 50 patients, and the quadratic weighted Kappa was used for statistical analysis. The Cronbach´s alpha was also calculated for internal consistency (n=105), standard error measurement (n=50) and construct validity (n=151) analyzing the correlation between the SBST-Brasil and the Brazilian version of Oswestry Disability Index (ODI), Rolland Morris (Brasil- RM) and Fear Avoidance Beliefs Questionnaire for work (FABQ-Work) and physical activity (FABQ-Phys). Results: Consensus was found on translations and back-translations processes, and only the item 6, which patients had some difficulty in understanding, was reviewed and re-applied to 52 additional subjects with no difficulties being reported. The reliability was classified as substantial (0.79 - 95% CI 0.63 0.95), the internal consistency acceptable (0.74 for total score and 0.72 for psychosocial subscale score) and a very good standard measurement error (1.9%). The correlation between the SBST-Brasil total score and Brasil- RM was high (r =0.61), moderate with ODI (r = 0.56) and with FABQ-Work (r = 0.3), and low correlation with FABQ-Phys (r = 0,18). Moderate correlation was found with SBST psychosocial subscale score and Brasil-RM (r = 0.49), also with ODI (r = 0.47), and low correlation with FABQ-Work (r = 0.24) and FABQ-Phys (r = 0.21). After that, the stratification of SBST-Brasil risk groups was compatible with ODI, Brasil-RM and FABQPhys classification for risk of poor prognosis. Conclusion: The SBST-Brasil has shown a reliable and valid tool to screen and identify low back pain patients influenced by psychosocial factors.

Fatores psicossociais não são rotineiramente identificados na avaliação fisioterapêutica e podem influenciar no prognóstico de pacientes com dor lombar. O questionário STarT Back Screening Tool (SBST) foi criado para distinguir subgrupos dentre pacientes com dor lombar em relação ao prognóstico no tratamento considerando fatores físicos e psicossociais, classificando-os em baixo, médio e alto risco de mau prognóstico. Objetivo: Traduzir e adaptar transculturalmente o SBST para Língua Portuguesa do Brasil, testar a confiabilidade e validade de construto. Métodos: A primeira etapa consistiu na tradução, síntese, retro-tradução, revisão pelo grupo de tradução, pré-teste e avaliação dos documentos pelo Comitê. A versão pré-final foi aplicada em duas amostras de 52 pacientes cada, com dor lombar, ambos os sexos, idade acima de 18 anos e níveis educacionais variados. Para verificação da confiabilidade intra-avaliador, foram realizadas duas entrevistas em outra amostra de 50 pacientes, e os resultados analisados pelo Kappa ponderado quadrado. Também foram calculados a consistência interna usando de Cronbach (n=105), o erro padrão da medida (n=50) e a validade de construto (n=151), correlacionando o SBST-Brasil com as versões brasileiras do Oswestry Disability Index (ODI), Rolland Morris (Brasil- RM) e Fear Avoidance Beliefs Questionnaire para trabalho (FABQ-Work) e atividade física (FABQ-Phys). Resultados: O consenso foi atingido nas etapas de tradução e retrotradução, apenas o item 6 não foi compreendido e depois de reformulado foi reaplicado em outros 52 pacientes que não tiveram mais dúvida. A confiabilidade foi considerada substancial (0,79 - 95% IC 0,63 0,95), a consistência interna aceitável (0,74 pontuação total e 0,72 da subescala psicossocial), e o valor do erro de padrão da medida muito bom (1,9%). Em relação à validação, a pontuação total do SBST- Brasil mostrou alta correlação com o Brasil-RM (r=0,61), moderada correlação com o ODI (r = 0,56) e FABQ-Work (r=0,3) e fraca correlação com o FABQ-Phys (r = 0,18). Já a subescala psicossocial mostrou correlação moderada com o Brasil-RM (r =0,49) e ODI (r =0,47) e fraca correlação com o FABQ-Work (r =0,24) e FABQ-Phys (r =0,21). Além disso, a estratificação dos grupos de risco do SBST-Brasil foi compatível com os valores de classificação do ODI, Brasil-RM e FABQ-Phys para risco de mau prognóstico. Conclusão: O SBST-Brasil mostrou-se confiável e válido para avaliar a influência de fatores psicossociais em pacientes com dor lombar. / Psychosocial factors are not usually identified on physiotherapy assessment, what influences on low back pain prognosis. The questionnaire STarT Back Screening Tool\" (SBST) was created to identify patients with potentially treatment modifiable prognostic influenced by physical and psychosocial indicators which are relevant to initial decision making, classifying them into low, medium and high risk categories. Aim: Translation and cross-culturally adaptation of the SBST into Brazilian-Portuguese, assess its reliability and construct validity. Methods: The translation and cross-cultural adaptation process involved the steps of Translation, Synthesis, Back-translation, Committee Revision, Pretesting and Submission the documents to the Committee. The pre-final version was applied to two samples of 52 patients each, with low back pain, both sexes, aged over 18 years, from different education levels. After that, two interviews assessed intra-rater reliability on a different sample of 50 patients, and the quadratic weighted Kappa was used for statistical analysis. The Cronbach´s alpha was also calculated for internal consistency (n=105), standard error measurement (n=50) and construct validity (n=151) analyzing the correlation between the SBST-Brasil and the Brazilian version of Oswestry Disability Index (ODI), Rolland Morris (Brasil- RM) and Fear Avoidance Beliefs Questionnaire for work (FABQ-Work) and physical activity (FABQ-Phys). Results: Consensus was found on translations and back-translations processes, and only the item 6, which patients had some difficulty in understanding, was reviewed and re-applied to 52 additional subjects with no difficulties being reported. The reliability was classified as substantial (0.79 - 95% CI 0.63 0.95), the internal consistency acceptable (0.74 for total score and 0.72 for psychosocial subscale score) and a very good standard measurement error (1.9%). The correlation between the SBST-Brasil total score and Brasil- RM was high (r =0.61), moderate with ODI (r = 0.56) and with FABQ-Work (r = 0.3), and low correlation with FABQ-Phys (r = 0,18). Moderate correlation was found with SBST psychosocial subscale score and Brasil-RM (r = 0.49), also with ODI (r = 0.47), and low correlation with FABQ-Work (r = 0.24) and FABQ-Phys (r = 0.21). After that, the stratification of SBST-Brasil risk groups was compatible with ODI, Brasil-RM and FABQPhys classification for risk of poor prognosis. Conclusion: The SBST-Brasil has shown a reliable and valid tool to screen and identify low back pain patients influenced by psychosocial factors.