Patient-Generated Subjective Global Assessment

February 1, 2003

Traditionally, dietitians have relied on objectiveparameters (such as anthropometric, biochemical,and immunologic measures) to assess nutritionalstatus. The usefulness of these parameters has beenquestioned in view of the many non-nutritional factorsaffecting the results. Hence, subjective assessmentof nutritional status has been used to overcomethese difficulties.

Traditionally, dietitians have relied on objective parameters (such as anthropometric, biochemical, and immunologic measures) to assess nutritional status. The usefulness of these parameters has been questioned in view of the many non-nutritional factors affecting the results. Hence, subjective assessment of nutritional status has been used to overcome these difficulties.

Subjective global assessment (SGA) assesses nutritional status based on the features of a history (weight change, dietary intake change, gastrointestinal symptoms that have persisted for greater than 2 weeks, and functional capacity) and physical examination (loss of subcutaneous fat, muscle wasting, ankle/sacral edema and ascites).[1-3] Features are combined subjectively into an overall or global assessment in which patients are rated as being well nourished, moderately (or suspected of being) malnourished, or severely malnourished.[1,3] A healthcare professional such as a nurse, dietitian, or physician completes the whole assessment.

Subjective global assessment has been found to have a high degree of inter-rater reliability.[1,4,5] It has also been found to have predictive and convergent validity, correlating well with measures of morbidity (incidence of infection, use of antibiotics, length of stay), a number of traditional objective parameters (anthropometric, biochemical, functional, and immunologic), and quality of life.[1,4-8]

Subjective global assessment has been applied successfully as a method of assessing nutritional status in a variety of patient populations, including: oncology patients, gastrointestinal surgical patients, dialysis patients, AIDS patients, patients with human immunodeficiency virus, lung transplant recipients, liver transplant candidates, geriatric patients, and post-Whipple procedure patients.

There are a number of advantages for using SGA as a nutrition assessment tool. These include its subjectivity, its use of clinicians' experience and expertise, and the fact that it does not rely on laboratory tests, uses available data, is quick, simple, reliable, and inexpensive, and can be performed by nonprofessional staff at the bedside. A disadvantage of SGA is that it lacks the sensitivity to detect shortterm improvements in nutritional status, for example, during a hospital admission.

A patient-generated subjective global assessment (PG-SGA) for oncology patients has been developed by Ottery.[9] This tool has two sections-a medical history section that is completed by the patient, and a physical assessment section that is completed by nursing, medical, or dietetic staff. The medical history section includes additional questions regarding the presence of oncology nutrition impact symptoms.

A further development by Ottery was a scored version of the PG-SGA, which includes a numerical score as well as the overall global rating.[10] The scoring system allows patients at risk for malnutrition to be identified and triaged for nutritional intervention. The scored PG-SGA may also be useful in monitoring short-term changes in nutritional status. The scored PG-SGA was recently validated against SGA and has been accepted as the standard for nutrition assessment for patients with cancer by the Oncology Nutrition Dietetic Practice Group of the American Dietetic Association.[10,11]

The scored PG-SGA offers several advantages over the SGA tool. These include the following: (1) the patient completes the medical history component, thereby decreasing the amount of time involved; (2) it contains more nutrition impact symptoms, which are important to the patient with cancer; and (3) it has a scoring system that allows patients to be triaged for nutritional intervention.

Therefore, the scored PG-SGA can be used as a nutrition assessment tool as it allows quick identification and prioritization of cancer patients with malnutrition.


Editor's Note: All workshop participants contributed to this discussion, highlights of which are presented here.

Screening for Nutritional Risk and Psychosocial Distress

Discussants established that the patient-generated subjective global assessment (PG-SGA) instrument used by Maree Ferguson can be used to screen and/or assess the nutritional status of patients with cancer. Discussion focused on how to screen cancer patients for nutritional risk factors so that they can receive appropriate nutrition referrals and intervention. Dr. Ferguson stated that the malnutrition screening tool comprised two questions regarding weight and appetite (Ferguson et al: Nutrition 15:458-464, 1999). She indicated that those questions are significantly predictive of the entire subjective global assessment. Discussants agreed that screening tools used in the clinical setting must be short and simple. More complex tools can then be used to evaluate the nutritional status and needs of patients who are flagged as "at risk" during the initial screen.

Evaluating and Measuring Psychosocial Distress

Discussants considered how to measure subjective factors such as patients' sense of wellbeing and patient and caregiver psychosocial distress. Dr. Steinbaugh pointed out that wellbeing is a constellation of factors that includes, for instance, nutrition and pain control; thus, measurement may be difficult. Dr. Ferguson agreed that this might be difficult in a clinical setting but that such quality-of-life instruments as the EuroQol EQ-5D and the European Organization for Research and Treatment of Cancer's (EORTC) QLQ-C30 can provide useful information.

Accepting or Rejecting Quality-of-Life Data

Anne Coble Voss indicated that many healthcare professionals consider quality-of-life data "soft" and that several validated measures of quality of life might be needed to convince them of the value of such data. She expressed the concern that many agencies in the United States (including the Food and Drug Administration [FDA]) will not make decisions about such issues as efficacy and reimbursement based on quality-oflife data. Jeff Salon pointed out, however, that such data have been used by the FDA in decisions about pharmaceuticals for cardiac and renal care. Aminah Jatoi indicated that the FDA also relied on some quality-of-life data in considering drugs for pancreatic cancer.

Bridging the Gap Between Patients' Perceptions and Health-Care Instrument Scores

Discussants agreed that patients' perceptions of subjective parameters such as pain, distress, and nausea often are inadequately measured by existing instruments. Dr. Benedetti reported that a frequently used pain scale on which patients rate pain levels from 0 to 10 may be viewed differently by patients and health-care researchers: Researchers see the scale as linear, while patients may perceive it as logarithmic. Therefore, most of the statistical data obtained from the analysis of this scale may be misleading. The point was made that while validated instruments have made pain "visible" to clinicians, nausea is still relatively "invisible" and difficult to measure. Methods of measuring pain (now part of Joint Commission on Accreditation of Healthcare Organizations [JCAHO] evaluations) could be models for developing measures of nausea. Since nausea contributes to loss of appetite in patients with cancer, such an instrument could help screen for patients who are at increased risk for weight loss.

Linking Subjective and Objective Instruments and Data

Discussants agreed that subjective data appear stronger when they correlate with objective data, in the eyes of many health-care professionals. Kathleen Murphy-Ende pointed out, however, that objectively measuring nausea might be difficult because it has multiple etiologies. Dr. Gallagher- Allred also pointed out that what a patient says is pain is pain, regardless of what healthcare professionals think. Perhaps the same is true for nausea.

Determining Whether Costs Outweigh Benefits of Validating These Instruments

Research in both nutritional and psychosocial research is difficult to do because investigative control is difficult to achieve. Furthermore, the number of patients and research sites that would be required for validation makes the cost of such research very high. In addition, money would probably be made available for developing and validating these instruments if institutions were penalized or subject to litigation if they did not use them. Discussants indicated that such a large study would need to be funded by a major agency such as the National Institutes of Health or the National Cancer Institute.

Investigating Hormonally Shaped Outcomes of Cancer and Cancer Treatment

Dr. MacDonald observed that among people with lung cancer, men lose more weight than women do and tend to not fare as well. He asked whether any studies had linked changes in testosterone levels with weight loss. Dr. Voss reported a similar finding from a large study of pressure ulcers in long-term care: men tend to lose more weight than women do, and when they lose weight, they are sicker and have poorer outcomes. Dr. Steinbaugh pointed out that in weight-loss programs for obesity, men consistently lose weight faster and more effectively than women, probably because men have higher energy requirements.

Financial Disclosure:The author has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.



Detsky AS, Baker JP, O'Rourke K, et al: Predicting nutritionassociatedcomplications for patients undergoing gastrointestinal surgery.JPEN J Parenter Enteral Nutr 11:440-446, 1987.


Detsky AS, Smalley PS, Chang J: Is this patient malnourished? J AmMed Assoc 271(1):54-58, 1994.


Detsky AS, McLaughlin JR, Baker JP, et al: What is subjectiveglobal assessment of nutritional status? JPEN J Parenter Enteral Nutr11:8-13, 1987.


Baker JP, Detsky AS, Wesson DE, et al: Nutritional assessment: Acomparison of clinical judgment and objective measurements. N Engl JMed 306(16):969-972, 1982.


Baker JP, Detsky AS, Whitwell J: A comparison of the predictivevalue of nutritional assessment techniques. Hum Nutr Clin Nutr 36c:233-241, 1982.


Detsky AS, Baker JP, Mendelson R et al: Evaluating the accuracy ofnutritional assessment techniques applied to hospitalised patients: Methodologyand comparisons. JPEN J Parenter Enteral Nutr 8:153-159,1984.


Ferguson M, Capra S, Bauer J, et al: Comparison of subjective globalassessment with objective nutrition parameters and outcome variables.Proc Nutr Soc Aust 21:164, 1997.


Ferguson M, Capra S, Bauer J, et al: Quality of life in patients withmalnutrition. J Am Diet Assoc 98(suppl 1):A-22, 1998.


Ottery FD: Definition of standardized nutritional assessment andinterventional pathways in oncology. Nutrition 12(1 suppl):S15-S19,1996.


McCallum PD, Polisena CG (eds): The Clinical Guide to OncologyNutrition. Chicago, The American Dietetic Association, 2000.


Bauer J, Capra S, Ferguson M: Use of the scored patient-generatedsubjective global assessment (PG-SGA) as a nutrition assessment tool inpatients with cancer. Eur J Clin Nutr 56:779-785, 2002.