Complexity and intensity of care: Key elements in determining health needs

In recent years, there has been an increasing interest in measuring the population's care needs in order to develop policies that establish safe nursing staffing levels in different healthcare settings. This approach is based on a solid foundation of studies that link nurse-to-patient ratios with adverse health outcomes, such as increased mortality, hospital readmissions, and prolonged hospital stays.^1,2 Scientific evidence has driven legislative reforms and regulatory frameworks for safe nursing staffing, as in the cases of California (2004), Queensland-Australia (2016), Wales (2016), Ireland (2018), and Scotland (2024), among others. All these experiences share a key element: nursing staffing must be based on an objective estimate of care needs, usually expressed in terms of nursing hours required per patient per day.

In the Spanish context, a structural shortage of nurses persists, accompanied by high territorial variability in nurse-patient ratios.^3,4 This scenario fuels the debate on how to measure care needs to improve population health outcomes and inform human resource strategies and future legislation that guarantees equitable care. Along these lines, the report prepared by the Ministry of Health on the “Employment Situation and Needs Perceived by Nurses in Spain” indicates that practically two-thirds of professionals believe there are not enough nurses to guarantee quality care for the patients they treat.³ It is also worth remembering that in the 1970s and 1980s, different systems emerged to quantify care needs with the aim of allocating nursing staff levels more closely aligned with the healthcare reality. Since then, various tools have been developed to measure workloads or levels of dependency and, more recently, patient classification systems aimed at assessing the complexity and intensity of care.^5–7

Against this backdrop, this leader aims to define and clarify the concepts of complexity and intensity of care, as well as briefly review their impact on health outcomes and their implications for future policy strategies.

Patient classification systems emerged as a response to the need to quantify nursing workload in order to adjust staffing levels to care needs. Some of the instruments, such as the Therapeutic Intervention Scoring System-28 (TISS-28), the Nine Equivalents of Nursing Manpower Use Score (NEMS), the Nursing Activities Score (NAS), and the Assessment of Nursing Workloads and Time (ANWT), have been used in critical care settings, while the Oulu Patient Classification (OPC) has been used in acute hospitalisation settings. However, traditional workload identification systems have proven insufficient to reflect the complexity of nursing care, as they primarily focus on quantifying the time associated with clinical interventions and procedures and on patient severity.^5,8 Therefore, some researchers have attempted to explore new ways to bring visibility to nursing practice, considering other variables such as the influence of the organisational context on care delivery, the impact of neglect, and the complexity and intensity of care.⁹

Currently, there is no consensus definition of the term "care complexity," although in recent years there has been a growing interest in analysing this concept, traditionally associated with the progressive aging of the population and the increase in chronic and multimorbid conditions, which has led to an increased demand for healthcare.¹⁰ However, in addition to age and comorbidities, there are other determinants of health that influence health outcomes and that have not been as widely studied, such as personal, economic, cultural, and environmental circumstances.¹¹

According to previous studies, care complexity is made up of different dimensions and includes: therapeutic or procedural complexity (nursing interventions, severity, level of acuity), professional expertise (experience, skills, critical reasoning, mental or emotional load), organisational complexity (work climate, characteristics of the organisational context) and individual or patient complexity.^6,12 In relation to individual complexity, Safford (2007) proposed a conceptual approach to evaluate it using the Vectorial Model of Complexity, which describes the interactions between biological, socioeconomic, cultural, environmental and behavioural forces as determinants of health, presenting the person as a dynamic and complex system.¹¹ Subsequently, Juvé-Udina (2010) defined “Individual Care Complexity Factors” as a set of specific characteristics in each person related to the different determinants of health and with the potential to increase the difficulty in the nursing care delivery process. Based on a participatory action research methodology study, she developed a tool where Individual Complexity Factors are identified and classified in a total of five domains: (i) mental and cognitive; (ii) psycho-emotional; (iii) socio-cultural; (iv) evolutionary and (v) comorbidity and complications.⁷ This tool is part of ATIC as a set of knowledge tools for generating data and information on the nursing care delivery process and its health outcomes.¹³

Of note is that the “Annual Report of the Spanish National Health System” already reported data related to some of these heath determinants, such as dependency ratios, educational level, poverty risk rates, and population lifestyles.¹⁰ However, the set of “Individual Care Complexity Factors” is not systematically recorded in the clinical and healthcare information systems of the National Health System. Recent research conducted in hospitals of the Catalan Health Institute has shown that more than 80% of hospitalised patients present at least one individual care complexity factor, and that a greater number of these factors is associated with poorer health outcomes (adverse events, mortality, readmissions, or emergency department visits). Thus, the various published studies indicate that “Individual Care Complexity Factors” are a useful tool for measuring patient characteristics that are related to increased nursing resources and poorer health outcomes. Among the most relevant findings is the fact that the lack of a caregiver is associated with adverse events or in-hospital mortality. Likewise, the presence of fear or anxiety and adjustment disorders have also been linked to unfavourable health outcomes. Therefore, these studies conclude that the early identification of "Individual Care Complexity Factors" could help detect those patients at higher risk of developing complications during their hospital stay.^14,15

Currently, different organisations use other tools to determine care needs. Some of these are Diagnosis-Related Groups (DRG-APR) or Adjusted Morbidity Groups (AMG), which are used as a healthcare management tool and incorporate metrics to stratify the population into different groups according to their level of morbidity-complexity, severity, or risk of mortality.¹⁶ However, none of these tools specifically addresses the provision of nursing care. Therefore, it is necessary to implement and evaluate instruments that allow for the determination of care needs in terms of nurse-hours per patient or the required nurse-patient ratio.

At this point, differentiation should be made between the concepts of complexity and intensity of care. Care complexity describes the multidimensional profile of needs arising from the patient, the organisational context, the required interventions, and even professional expertise. In contrast, intensity of care refers to the quantification of nursing activity—both direct and indirect—necessary to provide care, usually expressed in nurse-hours per patient per day or as a nurse-patient ratio.⁷ In recent years, various patient classification systems have been developed to determine the intensity of care. Some current systems use data from electronic health records, which include not only aspects related to nursing interventions or activities but also other elements recorded in nursing care plans.^7,17,18

In our context, Juvé-Udina (2019) developed the ATIC (Acute to Intensive Care) Patient Classification System, implemented in various healthcare providers, which measures the intensity of care required by hospitalised patients. This validated system assigns a weighted value to the primary nursing diagnosis, classifying patients into 10 levels of care intensity, with their equivalent in nursing hours required per patient per day and, therefore, in the nurse-to-patient ratio. This classification system demonstrated the discriminatory capacity of nursing diagnoses using ATIC Terminology (Architecture, Terminology, Interface, Information, Nursing, and Knowledge), showing near-excellent predictive capacity. Various studies conducted in hospital settings show that, according to the ATIC Patient Classification System, approximately two-thirds of patients require a nurse-to-patient ratio of 1:6 or lower.^4,7 This data is consistent with international standards established in different legislative frameworks that place safe ratios in medical-surgical units generally between 1:4 and 1:6 patients per nurse.¹⁹ These studies have also determined that nursing coverage (understood as the ratio between the hours of nursing offered and required per patient per day) in the hospital setting was 65%, observing a direct relationship between low levels of nursing coverage and health outcomes. This allowed us to determine that safe nursing coverage is around 90%.⁴

In this regard, evidence shows a clear need to assess the complexity and intensity of care, as this has demonstrated its association with health outcomes, making it a key tool for determining safe nurse-patient staffing levels. Furthermore, recent studies indicate that improved nursing staffing is associated with a reduction in mortality and other adverse events, as well as greater professional satisfaction.^2,20 Therefore, the complexity and intensity of care are key elements for establishing strategies to improve staffing levels.^4,21

Consequently, the need to systematically assess patients' health needs in different care settings is evident, with the aim of improving nursing staffing and establishing a regulatory framework that ensures equity of care within the National Health System. In this regard, Spain is making significant progress through the implementation of the Strategic Framework for Nursing Care (MECE 2025–2027 for its initials in Spanish) and the creation of the Healthcare Committee,²² which represent an historic opportunity to incorporate the assessment of care complexity and intensity into healthcare policy strategies.

Notwithstanding, much remains to be done to improve care delivery outcomes in different healthcare settings, since the majority of studies assessing care complexity and intensity have been conducted in hospital environments. This highlights the need to extrapolate and adapt these findings to other areas such as primary, intermediate, and social care. In this context, interoperability between systems and programming languages, along with the use of generative artificial intelligence tools, offers an opportunity to address the different dimensions of care complexity and intensity and generate useful indicators for healthcare practice, management, and policy.

To date, our studies have shown that the complexity and intensity of care impact population health outcomes. However, future research should evaluate prevention strategies targeting those complexity factors that can be modified. It would also be necessary to explore how the different dimensions of complexity and intensity of care interact with individual health outcomes, and to analyse their economic impact to ensure the sustainability of the healthcare system.