A network analysis approach to the needs of oncological patients.

Abstract

e24128 Background: In recent years, an ever greater importance is given to the needs of cancer patients which could impact medical treatments, adherence and compliance, and patients’ quality of life (Teo et\xa0al. 2019; NCCN, 2015). However, recognizing and addressing the needs of patients may not be enough. Indeed, on one hand, needs that clinicians might consider as important could play a marginal role, and – on the other hand – needs that clinicians may perceive as unimportant may be central for patients. An innovative approach – psychometric network analysis (PNA; Epskamp, 2017) – was used to assess the network among needs ( nodes). Notably, the more a central node (need) is modified (addressed), the more a cascade change will occur in all the other nodes (needs). This study aimed to evaluate: (A) the structure of relationships among needs ( edges) of cancer patients, (B) which needs ( nodes) are the most relevant ( central), and (C) play a key role in the network. Methods: Patients ( n = 511; mean age = 65.95, SD = 12.72; 280 males) were enrolled at the Oncology Day Hospital at the “Presidio Ospedaliero” of Saronno, ASST Valle Olona, Italy. Patients were tested with the Need Evaluation Questionnaire (NEQ) which is composed of 23 items that investigate as many needs – divided into 4 areas: (1) information about diagnosis/prognosis, (2) information about exams and treatment, (3) communicative, and (4) relational needs. Results: Preliminary analysis revealed that all of the items were informative (SDitem < 2.5SDall_itemsSD) and there was no redundancy between items (redundancy index < 0.25). An Ising model (5,000 nonparametric bootstraps) with LASSO regularized nodewise logistic regression was performed. PNA showed a high accuracy: CS-coefficient = 0.56. On one hand, PNA showed that item#2 ( “I need more information about my future condition”; z = 2.039), item#17 ( “I need to speak with a psychologist”; z = 1.209), and item#13 ( “I need to be reassured more by the doctors”; z = 1.201) were the strongest (central) nodes. On the other hand, item#14 ( “ I need the hospital to provide better services”; z = -2.261) was the weakest node in the network. Conclusions: These findings show useful implications for clinical practice. Clinical interventions should address the needs showing the strongest connections in the network. These central nodes can influence all the other connected needs, thus representing important needs to be targeted by clinicians – allowing to tailoring more targeted and efficient therapeutic approaches to meet patients’ needs, with beneficial effects for medical treatments and quality of life.