Background

Non-specific low back pain (NSLBP) is one of the most prevalent conditions in the world and identifying patients at risk for developing chronic NSLBP is key to effective treatment.  The STarT back screening tool is a validated, prognostic screening tool that identifies subgroups of patients with NSLBP, and the risk factors associated with each subgroup, which can guide the treatment decision-making process.  The objective of this study was to translate and validate the English version of the STarT back screening tool into isiZulu.

Relevance

Make the STarT back screening tool useful and accessible to the isiZulu population of South Africa, and clinicians working with this population.

Participants

One medically trained and one university trained professional proficient in isiZulu to forward translate, two language professionals proficient in isiZulu to backward translate and an expert committee of healthcare and language professionals from Wits University for translation. Thirty isZulu patients with acute or chronic NSLBP for validation.

Method

Translation was done in four phases, which included forward translation, forward translation synthesis, backward translation and expert review.  Validation included expert review for content validity and testing and retesing of the preliminary translated tool on a sample of 30 participants

Analysis

Descriptive statistics analysed socio-demographic data.  Content validity was determined by calculating item and scale level rating.  Test-retest reliability was measured using Spearman's correlation coefficient.  Internal consistency was evaluated by calculating Cronbach's alpha coefficients for the total score (Questions 1-9) and for the psychosocial subscale (Questions 5,6,7,8,9).  Descriptive statistics were used to analyse a post-retest questionnaire evaluating the translated tool's utility and efficiency.

Results

Minor linguistic differences occurred with the translating process.  The expert committee came to agreement around the challenging translation of the terms "more slowly" in question 4, "physically active" in question 5, "terrible" in question 7 and "bothersome" in question 9.  A total of 30 patients were recruited, with 77% female and 23% male.  Item content validity for relevance was excellent (I-CVI = 1.00), and satisfactory (I-CVI = 0.94) for clarity, simplicity and ambiguity.  Scale content validity was acceptable at 0.955.  Spearman's CC for test-retest reliability was acceptable at 0.73.  Cronbach's Alpha for internal consistency for the total score for test one was 0.68 and 0.77 for test two, whist the psychosocial scale for test one was 0.62 and test two was 0.77.  Thirty three percent reported the translated tool very easy to understad and 40% reporte it very easy to complete.

Conclusion

The isiZulu STarT back screening tool showed excellent validity, acceptable reliability and internal consistency and can therefore be used by an isiZulu population of patients with NSLBP.  Testing of the clinical and cost-effectiveness of the English and isiZulu STarT back tool in the South African context is necessary.

Implications

Translation and validation of outcome measures is an important part of effective treatment in developing countries

Key words

STarT back, Translation, Validation

Funding source acknowledgement

South African Society of Physiotherapy

Ethics approval

Human Research Ethics Committee of the University of the Witwatersrand (M170717)