The validity of the Patient-Generated Subjective Global Assessment Short-form© in cancer patients undergoing chemotherapy

Abstract

The high prevalence of malnourished cancer patients highlights the importance of sensitive and specific tools for nutritional risk and status assessment screening, namely the Patient-Generated Subjective Global Assessment (PG-SGA®). This study aimed to assess whether the short-form version of the PG-SGA® (PG-SGA© SF) would be appropriate to identify the nutritional risk of patients when compared with the final global score of PG-SGA© (long-form version). This transversal and observational study comprised a convenience sample of cancer patients undergoing chemotherapy at the Champalimaud Clinical Centre between December 2016 and February 2018. Clinical data and anthropometric parameters were collected in order to apply PG-SGA® and PG-SGA© SF. The data was statistically analysed through SPSS version 22 (SPSS Inc, Chicago, IL, USA). In this study 355 patients were enrolled and PG-SGA© SF results showed that 69.3% (n=246) of the population presented at least one risk factor for malnourishment (Σ (box A) ≥1). Additionally, PG-SGA® revealed that 50% of patients (n=177) have a risk of developing malnourishment or are already malnourished (B and C classification). The concordance of results showed to be high (coefficient k=0.62; p<0.001), meaning that PG-SGA SF© has a good sensibility (95%) and specificity (67%) for the identification of nutritional risk and assessment of nutritional status when compared with the complete version of PG-SGA©. According to our results, PG-SGA© SF is a useful and sufficient tool, representing an easier and faster way to identify at-risk or malnourished patients.