PhilSPEN Online Journal of Parenteral and Enteral Nutrition

(Article 12 | POJ_0002.html) Issue February 2012 - December 2014: 1-7

Original Clinical Investigation

The results of the validation process of a Modified SGA (Subjective Global Assessment) Nutrition Assessment and Risk Level Tool designed by the Clinical Nutrition Service of St. Luke’s Medical Center, a tertiary care hospital in the Philippines

Abstract | Introduction | Methodology | Results | Discussion | Conclusion | References | Back to Total Names Codes

Submitted February 15, 2013 | Posted March 16, 2014


Leia Lacuesta-Corro MD (1,2), Grace Paguia MD (1,2), Angelica Lorenzo MD (1,2), Donabelle Faye Navarette RND (2), Luisito O. Llido MD (1,2)

Institution where research was conducted:

  1. Clinical Nutrition Fellowship Training Program, St. Luke’s Medical Center, 279 E. Rodriguez Sr. Ave., Quezon City, Metro Manila, Philippines 1102
  2. Clinical Nutrition Service, St. Luke’s Medical Center, 279 E. Rodriguez Sr. Ave., Quezon City, Metro Manila, Philippines 1102



Background: There is a need to validate the modified SGA nutritional assessment form designed by PhilSPEN (Philippine Society of Parenteral and Enteral Nutrition) for use in the clinical nutrition process.

Objectives: To validate the modified- SGA form in order to have a nutritional assessment tool that can be used in this institution

Methodology: The form was subjected to sensitivity and specificity tests which also include the positive and negative predictive values as well as the diagnostic accuracy. ROC (Receiver Operating Characteristic) curves were also created to show the visual quality of the diagnostic tool.

Results: Total subjects: 179 (True positive=90; True negative=80; False positive=4; False positive=5)

Sensitivity = 94.7%
Specificity = 95.2%

Positive Predictive Value = 95.7%
Negative Predictive Value = 94.1%
Diagnostic Accuracy = 95%

Conclusion: The modified SGA nutrition assessment and risk level tool is shown by the validation tests to be an acceptable tool for performing nutrition assessment.


KEYWORDS: Modified SGA form, nutritional assessment, PhilSPEN


Nutrition assessment is the process where a patient, who is identified to be nutritionally at risk through the nutrition screening process [1,2] will undergo a more intensive body composition analysis in order to determine if he is malnourished or not and if malnourished if he is mild, moderate, or severely malnourished. [1-3] The degree of malnutrition has a major influence in outcome from an injury or traumatic event especially on the quality of management whether medical or surgical, thus there is a need to have a reliable and accurate tool for nutritional assessment in order to provide a more rational and adequate treatment for the patient. There is a lot of nutritional assessment tools available worldwide, but one of the most commonly used tool is the Subjective Global Assessment developed by Dr. Jeejeeboy and his team. [4]

The St. Luke’s Medical Center , a tertiary care hospital in the Philippines, has a nutrition program and one of its goals for achieving optimum standards of patient care is to develop its own simple to use, but validated, nutritional assessment tool for the “nutritionally at risk” patient. Through its partnership with the Philippine Society of Parenteral and Enteral Nutrition (PHILSPEN) [5-7] it adopted the modified- SGA form developed for use by both the society and the major institutions of care throughout the country (Philippines). The Clinical Nutrition Service of this institution decided to validate the modified- SGA form in order to have a nutritional assessment tool that can be used in this institution. This is the report on the result of the validation process.



The modified SGA form (Figure 1) is filled up by members of the Clinical Nutrition Team of St. Luke’s Medical Center, Quezon City, Philippines, with the final reference diagnosis of nutritional status determined by the senior members of the team. The modified SGA table and BMI (Body Mass Index) components were the data evaluated and validated. The serum albumin and Total Lymphocyte Count were not included in the validation process – these are designated to be part of the nutrition risk assessment.


Figure 1: Modified SGA form

Since there is no “gold standard” as to the diagnosis of malnutrition it was decided to define the “reference standard” for malnutrition based on these criteria: a) the use of a formal SGA done by a member of the clinical nutrition team on the same patient, b) final diagnosis of malnutrition by an experienced member of the clinical nutrition service – a senior consultant who has been with the service for more than 10 years and c) diagnosis of malnutrition in a clinical nutrition service which has defined malnutrition in its overall context for the past 10 years. [8] The forms were examined and evaluated by one senior clinical nutrition physician (Dr. L.O.L.) and two junior clinical nutrition physicians, Dr. L.L-C and Dr. G.P.). Table 1 also lists the criteria as to the consideration of the finding as False Positive and Negative. The combined assessments or diagnosis is considered the “reference” value as to whether the patient is malnourished or not.


The validation process was done by the following procedures. The sensitivity and specificity of the modified-SGA form was determined using 2 x 2 tables. The tables contain the True Positive, True Negative, False Positive and False negative values. Receiver Operating Characteristic (ROC) curves were also created to show the strength of the sensitivity and specificity of the tool. Finally the Positive and Negative Predictive Values were also determined. The software used was the NCSS-PASS© software designed by J. Hintze [9] and the explanations of the process were taken from the 4th edition of Basic and Clinical Biostatistics. [10]

The SGA grades of the subjects were identified (A, B or C). Those with SGA grade C were designated as true positive for malnutrition. Those with SGA grade A were designated as true negative cases or not malnourished. Finally for those with SGA grade B, they were designated as false negative if with subcutaneous fat or muscle loss (+1/+2) as shown in Table 1. A false positive diagnosis was designated to those with SGA grade B without subcutaneous fat or muscle loss using the guide shown in Table 1. The scores from the BMI levels were added to the SGA scores with the maximum of 2 indicating moderate malnutrition and ≥3 as severe malnutrition, however, the final decision as to which category the patient belongs, whether false positive, false negative or true positive rests on the one doing the nutritional assessment. The final decision in classifying the patient as malnourished or not rests on the senior members of the clinical nutrition team who have either seen the patients assessed or have reviewed the accomplished forms – this is the “reference standard or diagnosis”. Only patients with complete records and fully accomplished modified SGA forms were included in the study.



A total of 179 subjects were included in the study. Table 2 shows the 2 x 2 tables and Table 3 shows the validation results of the modified-SGA form.



Table 3 data also show high values indicative of the quality of the diagnostic capabilities of the form through its positive and negative predictive values and diagnostic accuracy.


Figure 2: ROC (Receiving Operating Characteristic) Curves

Figure 2 shows that the True Positive and True Negative curves were closest to the “superior characteristics of a diagnostic test” with the true negative result having a better quality compared to the true positive result [11]



It is shown by this validation process that the modified SGA tool is an effective tool in identifying the severely malnourished patient with a sensitivity of 94.7%, specificity of 95.2%, a positive predictive value of 95.7%, and a diagnostic accuracy of 95%. The ROC curves further show the value of the modified SGA form as a good diagnostic tool in the area of “ruling in malnutrition” which is the true positive curve and even better in “ruling out malnutrition” which is the true negative curve. Although the clinical nutrition service had to determine its own “reference standard” for the diagnosis of malnutrition, the use of the formal SGA and the expertise and experience of the clinical nutrition physicians deciding on the diagnosis of the presence or absence of malnutrition is considered adequate enough. It is also to be noted that the use of scores from the SGA and BMI category levels is not a requirement for absolute nutritional diagnosis, but the final decision still rests on the examiner and the final total score in the “Nutritional Status” box below the tables. The scores are meant to be guides in the final choice of the nutritional diagnosis.

When this nutrition assessment tool was compared with the NRS 2002 (Nutrition Risk Screening year 2002) which was developed by the Denmark group of ESPEN [12] and the MUST (Malnutrition Universal Screening Tool) developed for Great Britain [13], in the area of sensitivity, sensitivity and positive predictive value, both were validated using the formal SGA designed by Jeejeeboy et al [4], the results are as follows: [14]


The modified SGA version of this institution came up with better results as to sensitivity, specificity, and positive predictive values compared to these two already accepted nutrition screening and assessment tools. [14] Since the report by Young et al [14] were on the utilization of NRS 2002 and MUST as nutrition screening tools, it stands to reason that the higher value or score of the PhilSPEN modified SGA tool is due to its design as a nutrition assessment tool thus providing a more in-depth analysis of the nutritional status of the patient. The validation process thus shows that the PhilSPEN modified SGA nutrition assessment and risk levelling tool is a good and acceptable tool for the diagnosis of the presence or absence of malnutrition and its attendant risks. However, the ability of the tool to define the level of nutrition risk status of the patient is still to be evaluated.



The modified SGA nutrition assessment tool is an acceptable tool for nutritional assessment for adult and elderly patients in the Philippines, whether out-patient or in-patient.



  1. Pesce-Hammond K, Wessel J. Nutrition assessment and decision making. The ASPEN Nutrition Support Practice Manual 2nd ed; Merritt R, editor-in-chief; A.S.P.E.N., Silver Spring, MD; 2005: 3- 26.
  2. How does clinical nutrition run? The value of implementing a clinical nutrition program and nutrition support team (NST) to address the problem of malnutrition in the hospitals of the Philippines. PhilSPEN Online Journal of Parenteral and Enteral Nutrition. Available at POJ_PositionPaper.html#NT_mechanics2. Accessed January 30, 2013.
  3. A.S.P.E.N. Board of Directors and Task Force on Standards for Specialized Nutrition Support for Hospitalized Adult Patients. Standards for specialized nutrition support: Adult hospitalized patients. Nutr Clin Pract 2002; 17: 384-91.
  4. Detsky AS, McLaughlin JR, Baker JP, et al. What is Subjective Global Assessment of nutritional status? J Parenter Enteral Nutr 1987; 11 (1): 8-13.
  5. The history of clinical nutrition in the Philippines. The PHILSPEN website. Available at Accessed February 21, 2013.
  6. Ocampo R B, Camarse CM, Kadatuan Y, et al. Predicting post-operative complications based on surgical nutritional risk level using the SNRAF in colon cancer patients - a Chinese General Hospital & Medical Center experience. Phil J Surg Spec 2008; 63 (4): 147-53. Also available at; accessed February 26, 2013.
  7. Del Rosario DC, Inciong J, Sinamban R, Llido L. The effect of adequate energy and protein intake on morbidity and mortality in surgical patients nutritionally assessed as high risk and low risk: report from a tertiary care private hospital in the Philippines. PhilSPEN Online Journal of Parenteral and Enteral Nutrition. Available at Accessed March 1, 2013.
  8. Clinical nutrition program and services. Available at POJ_PositionPaper.html#role_Philspen; Accessed February 15, 2013.
  9. NCSS software. Available at Accessed March 1, 2014.
  10. Dawson B, Trapp R. Basic and Clinical Biostatistics 4th edition; McGraw-Hill 2004: 305-10.
  11. Lang T, Secic M. How to report statistics in medicine 2nd edition; American College of Physicians, Philadelphia 2005: 137-8.
  12. NRS 2002 - Velasco et al. Eur J Clin Nutr 2011; 65: 269-74 13. MUST - Kyle UG et al. Clin Nutr 2006; 25: 409-17
  13. YoungAM, Kidston S, Banks MD, Mudge AM, Isenring EA. Malnutrition screening tools: comparison against two validated nutrition assessment methods in older medical inpatients.

Abstract | Introduction | Methodology | Results | Discussion | References | Back to Total Names Codes