Malignant neoplasms can induce significant changes in the individual’s body as well as in their nutritional state, both by the direct action of the cancer and by the treatments used, such as surgery, chemotherapy or radiotherapy, either isolated or combined. As a consequence, malnutrition, often observed in this population, can compromise the expected therapeutic responses, requiring reduction of treatment doses, increasing levels of toxicity, and in some cases making the patient ineligible for surgical procedures. These factors can negatively impact the clinical outcome [1,2].

Malnutrition is predominantly characterized by a negative energy balance and loss of muscle mass, resulting from a complex interaction of factors such as reduced food intake, increased basal metabolic rate, insulin resistance, lipolysis, proteolysis, side effects associated with treatments, systemic inflammation and the presence of catabolic factors, which may be of individual origin or induced by the tumor. Early recognition of this condition is essential for the development of effective therapeutic strategies, with the aim of minimizing the impacts of malnutrition on the patient’s clinical state [2,3].

Most nutritional evaluation studies focus on hospitalized patients, which can introduce a bias of selection due to the controlled environment, with constant access to medical care, adequate nutrition and medications. However, in an outpatient context, it is possible to observe the actual conditions of the patient’s daily life, including their limitations and challenges. Factors such as the absence of a caregiver, adverse socioeconomic conditions, cognitive impairment, inadequate use of medications to manage side effects and the need to prepare their own diet can significantly influence, both positively and negatively, the patient’s response to treatment [4,5].

Several screening protocols are widely used for the early detection of nutritional risk, such as the Nutritional Risk Screening-2002 (NRS-2002), which has proven highly effective in hospitalized patients with various pathologies and is recommended by the European Society for Parenteral and Enteral Nutrition (ESPEN), the Instituto Nacional de Câncer (INCA, National Cancer Institute) and the Brazilian Society of Parenteral and Enteral Nutrition (BRASPEN). In addition, the Patient-Generated Subjective Global Assessment (PG-SGA) is a validated nutritional assessment tool for use in patients with any pathology. Recently, this tool has undergone a global revalidation, in which it was established that the first part, consisting of four questions and known as Short Form (Short Form), can be used as a nutritional screening tool for cancer patients, as indicated in the Brazilian guidelines mentioned above [3,6-10].

Since there is no validated questionnaire in Brazil for conducting nutritional screening in adult and elderly individuals, carrying solid tumors or hematological neoplasms, in outpatient care, it becomes necessary to develop an instrument that portrays the reality of the Brazilian population. This questionnaire should help to properly identify the patient, directing him to appropriate nutritional care and counseling.

The aim of this study is to propose a nutritional screening tool for patients with cancer in outpatient care, taking as a basis existing tools: Avaliação Subjetiva Global Produzida Pelo Paciente – versão reduzida (ASG-PPP, Global Subjective Assessment by the Patient - short form), Nutritional Risk Screening- 2002 (NRS-2002) and Nutritional Screening Tool for Oncological Outpatients (NUTRISCORE) and incorporating the specificities of this group of individuals. To minimize failures and resolve possible problems before their clinical implementation, potentially real situations will be simulated, allowing adjustments and adjustments in the tool before its definitive application to patients.

The tool will undergo semantic and technical validation by other professionals to ensure its practical applicability.

The purpose of this study is applied research, with observational nature, the form of the approach is quantitative, the objectives are of the explanatory type. The main technical procedure was the survey technique with transversal and prospective development.

A multicenter study was designed to validate the Triagem Nutricional em Oncologia para Adultos e Idosos (TriNOA, Nutritional Screening in Oncology for Adults and Elderly) questionnaire (Figure 1), which aims to measure the nutritional risk of oncological patients (adults and elderly) in outpatient care. The research was conducted in the Department of Clinical and Experimental Oncology of the Federal University of São Paulo with the collaboration of another seven specialized centers in oncology that also focus on the nutritional care of these patients.

Some of the study participants are located in different geographical regions of the country and with different forms of care (public or private). The sample of the research, prioritizes what can well represent the Brazilian oncological patient, which needs nutritional monitoring, regardless of ethnic variety, race, and economic condition.

All centers applied the questionnaire for a period of six months or until guaranteeing the repetition of at least two consultations per patient, the first consultation being called phase 1 and the second consultation being called phase 2. Eventually some patients could have more than two visits, but it was necessary that at least 10% of patients were treated in each of the centers. Ambulatory patients, aged 18 years or older, who were diagnosed with malignant neoplasms, including solid and hematological tumors, who were treated for oncological evaluation or treatment and who were able to answer the questionnaire were included. The exclusion criteria are patients in the terminal phase, unconscious, those who present difficulty orientation in time, space and memory failures. The questionnaire contains nine questions, the first seven are completed by the nutritionist in an interview with the patient and the final two are consulted in each individual’s record, which includes information on the location of the disease and active treatment. The protocol for this research was approved by the Ethics and Research Committee of the Federal University of São Paulo under the number 0574/2019.

Each of the questions contains a numerical value conditioned on the answer, and the sum represents the patient’s nutritional risk at three different levels, such as: non-significant risk, moderate risk and high risk, for each nutritional risk, there is a conduct to be carried out by the nutritionist. The total time to complete the questionnaire should not exceed two minutes for each patient under normal conditions.

This new tool measures diagnostic indicators such as: sensitivity, specificity, positive, negative predictive values using the tools: ASG-PPP short form and NRS-2002 as analysis reference. Although the Nutriscore tool is not validated in Brazil (as of the study date), it has also been used for comparison purposes.

The tool ASG-PPP is a good nutritional screening test but is applied to patients with any pathologies, however its summary form ASG-PPP short form is indicated for oncological patients, however, by our experience its implementation in outpatients dedicated to oncological patients does not present the benefits we need, due to a high rate of false positives, reducing the ability of better nutritional evaluation to patients due to the limitations of resources for care. Simply changing the criteria of the ASG-PPP short form also did not meet our needs and due to the limited questions addressed, so we decided to create a questionnaire that could give us a better understanding of the actual nutritional state of each patient. The same line of reasoning was applied to the questionnaires of the NRS-2002 and Nutriscore, the latter being not yet validated in Brazil.

Both the elaboration of the nine questions and the weight of each answer to each question was elaborated after consensus meetings of specialists who collaborated on the main aspects of the condition of these oncological patients which contributes to better meeting the need during screening of these individuals, depending on the socioeconomic, tumor location and treatment conduct.

Similar to the method adopted by: ASG-PPP short form, NRS-2002, and Nutriscore, were used as nutritional screening reference, patients who presented no significant nutritional risk with patients who presented some significant risk. A different point of our method is for the group that presents significant risk, which was subdivided into two groups, being moderate risk and high risk, based on the patient’s age, social conditions, feeding frequency, variation of body mass depending on the time, actual conditions of physical activities, being the most important, the location of the primary disease and the status of the current treatment.

Patients receive up to 8 points (adults) or 10 points (elderly), are those who present a non-significant risk and have as conduct the absence of need for intervention at that time and a new assessment required at each return. Patients who receive above these limits are considered individuals with significant risk, those who reach up to 19 points for adults or 25 points for the elderly, have a moderate risk and have as conduct the referral to the nutritionist for evaluation and nutritional guidance with monthly reassessment. Patients with a score equal to or greater than 20 points for adults or 26 points for the elderly, are classified with high risk, have the same behavior, but with a frequency of five-year re-evaluation. This reference was obtained in pilot studies with real data from patients treated in six oncology clinics of the Universidade Federal de São Paulo – Hospital São Paulo: neuro-oncology, gastro-oncology, lymphomas, multiple myeloma and two general oncology clinics, where the Receiver Operating Characteristic (ROC) curve was used as a reference to find the best cutting points that separate each of the groups.

The TriNOA questionnaire was applied by nutritionists during the outpatient care in the form of a face-to-face interview, in a first phase, the ASG-PPP short form questionnaires, NRS-2002 and Nutriscore were also applied for comparison purposes and in a second phase, only the TriNOAI questionnaire. In each evaluation the weight and height of each patient is measured, with clothes and barefoot and subtracted by an estimate of the weight of the clothes. The recorded values were compared with the previous data to evaluate whether there was weight variation depending on the time between each visit, where a critical rate of weight variation was considered, which comprised different weights in nutritional evaluation.

The sample size was based on the pilot study which presented a standard deviation of 8.30, we assumed an error between the true mean value (unknown) and the obtained in the pilot study of up to 15% and considered as a general error of the research, the standard adopted by the scientific community, in 5%, which resulted in a sample of approximately 330 patients (or 42 patients per center participating). Qualitative variables were used for the qui-square test, quantitative variables were used for measurements of descriptive statistics. The data showed adhesion to normal, so central trend (average) and variance (standard deviation) measurements were used. The inferential evaluation took place in the diagnostic indicators with the obtaining of sensitivity, specificity, positive and negative predictive values, ROC curve, ROC curve area. The general agreement was obtained by the kappa coefficient and all data was calculated in Excel using the Real Statistics plugin.

Nutritional screening protocols are the beginning of the patient’s risk detection process, and when properly conducted, they can and should identify situations in which the nutritionist can act promptly with the intention of minimizing the impacts on prognosis [12].

The questionnaires currently used to measure the nutritional risk of sick individuals, present characteristics that differ from the profile of the population of Brazil. These instruments are based predominantly on hospitalized populations, while the situation of outpatient care is distinct. Such disparity can compromise the accuracy of the results obtained, not adequately reflecting the reality of the population treated in outpatient contexts.

Mayo and Gomes [13], suggest in their study that health services should implement a nutritional assistance protocol for patients in chemotherapy treatment taking into account the type of tumor and the antineoplastic therapy; in this context, one can extrapolate to the need for an instrument that can perform the correct nutritional diagnosis considering the suggested aspects.

A nutritional screening tool was developed specifically for individuals undergoing oncological treatment in outpatient settings, with the goal of providing greater specificity, time efficiency, and safety in its application. The tool aims to support accurate diagnosis and enhance the assertiveness of nutritional assessments.

This instrument should be applied to real patients, randomly, respecting the inclusion criteria, and should present a representative sample of the Brazilian population. In addition, the instrument must be continuously evaluated and adjusted to meet the requirements necessary for the care of this group of individuals.