Intensity of Interprofessional Collaboration and related factors in Intensive Care Units. A descriptive cross-sectional study with an analytical approach

Abstract

Objective

To determine the Intensity of Collaboration between the intensive care professionals of a third level hospital.

Method

Descriptive cross-sectional study with an analytical approach. Setting: 6 intensive care units of a third level hospital. Sample: nurses and doctors. Consecutive type non-probabilistic sampling. Data collection: sociodemographic, economic, motivation and professional satisfaction variables, and the intensity of collaboration using the “Scale of Intensity of Interprofessional Collaboration in Health.”

Results

A total of 102 health professionals (91 nurses and 11 doctors) were included. The mean overall Intensity of Collaboration (IoC) was moderate. Men showed higher scores in all factors (p < .05). The IoC global score was higher in the group of professionals with ≤10 years of experience (p = .043) and those who were highly satisfied with the profession (p = .037). Physicians presented higher scores in the global IdC (p = .037) and in the Collaboration mean (p = .020) independently in the multivariate models. A negative linear relationship (rho: −0,202, p = .042) was observed between age and the overall IoC score. Professionals aged ≤30 years reported a higher perception of Shared Activities (p = .031). Negative linear relationships were observed between years of experience and total IoC score (rho: −0,202, p = .042) and patients' Perception score (rho: −0.241, p = 0.015). The research activity also showed to be a variable related to a greater degree of Collaboration at a global level and in some of the factors (p < .05). The scale of IoC obtained a Cronbach’s α of 0,9.

Conclusions

The intensity of interprofessional collaboration in ICUs is moderate. Professionals with experience of ≤10 years, a higher level of satisfaction and participation in research activities show a greater intensity of collaboration. Doctors perceive collaboration more intensely than nurses. All factors contribute equally to the internal consistency of the questionnaire.

Resumen

Objetivo

Determinar la Intensidad de Colaboración entre los profesionales de cuidados intensivos de un hospital de tercer nivel.

Método

Estudio descriptivo de tipo transversal con enfoque analítico. Ámbito: 6 unidades de cuidados intensivos de un hospital de tercer nivel. Muestra: enfermeras y médicos. Muestreo no probabilístico de tipo consecutivo. Recogida de los datos: variables sociodemográficas, económicas, de motivación y satisfacción profesional y, la intensidad de colaboración mediante la Escala de la Intensidad de la Colaboración Interprofesional en Salud (INCINS).

Resultados

Se incluyeron un total de 102 profesionales sanitarios (91 enfermeras y 11 médicos).

La media de la Intensidad de colaboración (IdC) global fue moderada. Los hombres mostraron puntuaciones superiores en todos los factores (p < 0,05). La puntuación global del IdC fue superior en el grupo de profesionales de ≤10 años de experiencia (p = 0,043) y los muy satisfechos con la profesión (p = 0,037). Los médicos presentaron puntuaciones mayores en el IdC global (p = 0,037) y en la media de Colaboración (p = 0,020). Se observó una relación lineal negativa (rho: −0,202, p = 0,042) entre la edad y la puntuación global del IdC. Se observaron relaciones lineales negativas entre los años de experiencia y la puntuación total del IdC (rho: −0,202, p = 0,042), y la puntuación de Percepción de los pacientes (rho: −0,241, p = 0,015). La actividad investigadora también mostró ser una variable relacionada con un mayor grado de Colaboración a nivel global y en algunos de los factores (p < 0,05). La escala de IdC obtuvo un α de Cronbach de 0,9.

Conclusiones

La intensidad de colaboración interprofesional en las UCI es moderada. Los profesionales con experiencia ≤10 años, mayor nivel de satisfacción y de participación en actividades de investigación muestran mayor intensidad de colaboración. Los médicos perciben la colaboración de forma más intensa que las enfermeras.

What is known/what does it contribute?

Inter-professional collaboration in the ICU is important for patient safety and quality of care, and it also has an impact on the work environment and the level of satisfaction of professionals.

The intensity of Inter-professional collaboration in intensive care has been little studied in our Spanish context and more specifically in intensive care units.

Study implications

The identification of the intensity of collaboration between ICU nurses and doctors, as well as the related variables, will allow us to identify areas for improvement and develop strategies to increase Inter-professional collaboration.

The results have shown that this tool has adequate reliability in the intensive care context (Cronbach’s α of .9).

Introduction

In Health Services, specifically in hospital care institutions, and particularly in intensive care units (ICU), the level of complexity has increased to such an extent that it permeates the organisation, communication, and quality of results. In this organisational paradigm, concern for the critically ill patient is key, since it places the person at the centre of the care process, and gives them an essential value where efforts are oriented towards achieving quality care, a fact that highlights the need to work in a coordinated and collaborative manner between teams of health professionals.¹

Since the appearance of the concept “nurse-doctor collaboration”, introduced by Zwarenstein and Bryant,² the binomial has been studied from various perspectives and methodologies. Thus, collaboration in the field of health has been defined in different ways. Weiss and Davis³ state that collaboration consists of interactions between a nurse and a doctor that enable the knowledge and skills of both professions, synergistically influencing the provision of care. On the other hand, at the institutional level, there is a consensus definition from the World Health Organization (WHO)⁴ that says: “When several health workers from different professional backgrounds work together with patients, families, caregivers and communities to provide the highest quality of care”.

D’Amour et al.⁵ state that the logic of Inter-professional collaboration is perceived as a process by which individuals from different professions structure a collective action in order to coordinate the services they provide to individual clients or groups. The interdisciplinary integration of care is the result of this coordination.” Additionally, the American Nurses Association (ANA),⁶ through its Credentialing Center, has indicated the importance of Inter-professional collaboration in the care delivery system.

It is precisely, from the definitions of the term collaboration provided by different authors^2,6–8 that it can be stated that they share the meaning of joint work and common achievement or purpose. The sense of action in motion stands out in the descriptions: working together, sharing information and decision-making and the willingness to collaborate on the basis of sharing power and authority.^9,10 Concepts such as sharing, relationship, interdependence and power are widely used when talking about collaboration.¹¹

The results of studies on Inter-professional collaboration have revealed a broad consensus in the scientific community on the impact that effective teamwork and a good level of communication have on achieving better quality of care.¹² Thus, the benefits of a positive relationship between nurses and doctors are well documented in the literature^13,14 demonstrating improvements in organisations in terms of increased quality of care, lower costs, economy in decision making, decreased morbidity and mortality of patients,¹⁵ reduction or adjustment of the stay of at-risk patients,¹⁶ together with that of recruitment and retention of nurses^17–19 and an increase in nurse satisfaction.^20–22 The latter issue is supported by other studies that report that nurses who experience a practical collaboration have lower levels of burnout.²³ For all these reasons, the results of these investigations indicate the need to promote positive changes in Inter-professional collaboration between nurses and doctors. A strong association has also been found between effective nurse-doctor collaboration relationships and the reduction or adjustment in hospital stay of at-risk patients.¹⁶

However, Gardner’s study²⁴ shows certain difficulties when applying a horizontal model of egalitarian relationships in a vertical and hierarchical structure, and its approach as a process or result. From a cultural perspective,²⁵ Inter-professional conflict has been identified as a stress factor in the work environment. The conflict would be explained by the cultural conditions in the paternalistic, hierarchical, dominance and submission relationships of the nurses, where even verbal and physical abuse²⁶ are employed. Studies based on this knowledge have been expanding and evolving towards the search for different models of cooperation and assertiveness,^27,28 a positive approach to conflicts²⁶ and approaches from non-clinical disciplines.²⁹ Other authors point out that educational factors can also influence the concept of mutual collaboration.³⁰ All of this would explain the discrepancies between the different perspectives and conceptualisations of collaboration between nurses and doctors³¹ with the profession, power, level of autonomy and with educational and cultural factors.^6,7,12,17 Elements that act as barriers have been identified^9,26 and the underlying structures have been explored.^25,27,28 This has made it possible to propose strategies to promote Inter-professional collaboration,^{15,16,19,22,32} although it has also been the subject of criticism and they cite the contextualization of the results in the international arena.

In the context of intensive care, Inter-professional collaboration is closer than in other services or hospital areas.³³ Studies report a higher degree of nurse-doctor collaboration³³ compared to other healthcare areas,^25,34,35 although they also determined the conflicts that exist in the ICU.³⁶

The common difficulty that the study of Inter-professional collaboration presents is being able to empirically demonstrate the effectiveness of this collaboration, since there is no valid and reliable measurement instrument to determine the degree of collaboration in the different Inter-professional contexts so that study results may be generalised. On the contrary, there is a variability of instruments that respond to different research objectives.^19,37

Tools have also been designed to evaluate different aspects of teamwork or the attitudes of professionals, and reviews have been carried out on the validity of these tools³⁸ and on the effects of applying Inter-professional Collaboration in clinical practice.³⁹ Furthermore, the association of collaboration has been measured with different elements^7,37 and descriptions of reference patterns for Inter-professional collaboration have been made.⁴⁰ Other studies have focused on the processes to achieve user satisfaction and the quality of care of nursing interventions through the creation of healthy work environments⁴¹ or on the social responsibility of the company.⁴² Studies show a better attitude towards collaboration among nurses,^17,19,32 although there are also studies that affirm the contrary.^31,43

The review carried out by Dougherty and Larson,³⁸ on five instruments used to measure nurse-doctor collaboration, shows that all of them are originally English and reproduce contexts which differ from Spanish ones. This has had an impact on a small number of studies on teamwork and collaboration between health professionals in our field since empirical demonstration becomes even more difficult due to the scarcity of original measurement instruments in our context and the limitation of being able to use only those adapted to other contexts and validated in Spanish, such as the Collaboration and Satisfaction About Care Decisions (CSACD) questionnaire.

In this sense, the study by San Martín-Rodríguez et al.⁴⁴ measured the intensity of Inter-professional collaboration between health professionals and related it to the objectives of satisfaction and quality. This study was carried out in 2007, in a private Spanish university hospital, in which a population of 123 nurses belonging to the surgical, medical and intensive care hospitalisation services participated. In this study, the Intensité de la Collaboration Interprofessionnelle developed Sicotte et al.⁴⁵ and validated in Spanish by San Martín-Rodríguez et al.⁴⁴ was used.

Despite the studies that exist in Spain on nursing work in the ICU, only two studies have been found that directly address the knowledge of the level of collaboration and Inter-professional coordination between nurse-doctor in an ICU. One was carried out in a private hospital⁴⁶ and another in a public hospital.⁴⁷ Both used the INCINS scale to measure the intensity of collaboration and associated personal factors. However, more studies are required in the context of public hospitals, with different patient profiles and other associated factors to continue with the INCINS validation process.

The general objective of this study was to determine the intensity of collaboration (IoC) between health professionals working in the multipurpose ICUs of a tertiary hospital. Secondarily, the factors related to the degree of collaboration are analysed. Among the related factors, differences are observed depending on the type of professional (nurses and doctors). Also, to determine the reliability of the Intensity Scale of Inter-professional Collaboration in Health (INCINS for its initials in Spanish) in the context of intensive care.

Method

Design

A cross-sectional, descriptive study with an analytical focus was conducted.

Scope

The study was carried out in 6 ICUs of a tertiary public university hospital, each of them with a different specialty (Table 1). The staffing of professionals is the same for all units and consists of a nurse-patient ratio of 1:2, 1 head of service, 2 assistant physicians and a variable number of residents. Data collection took place between the months of April and June 2012.

Table 1.

Distribution of public hospital ICU beds.

Name of the ICU	ICU beds UCI	Intermediate beds
IMA (Internal Medicine Intensive Monitoring Area)	8	8
ACCU (Acute Coronary Care Unit)	8	8
LICU (Liver ICU and gastroenterology)	8	6
UCAR (Cardiac Surgery ICU)	8	8
SICU (Surgical ICU)	14	8
IRMA (Intensive Respiratory Monitoring Area)	6	6
Total beds	52	44

Population and sample

The study population included all nurses and doctors from the different ICUs of the hospital. The sample of nurses was made up of staff professionals, supervisory nurses or ICU coordinators or substitutes with more than one year of experience and the doctors were made up of head of service, assistant heads or contracted shift workers with more than one year of experience in the unit. All professionals who had not worked in the unit in the last year (sick leave, leave, or leave of absence) were excluded.

A total of 147 questionnaires were delivered and the response rate was 75% (n = 110). The sample consisted of 102 professionals, of which 90% (n = 91) were care nurses/managers and 10% (n = 11) doctors.

A convenience sampling technique was used to select the sample.⁴⁸ The sample calculation was carried out according to the main objective of the study: degree of intensity of collaboration between nurses and doctors working in the multipurpose ICUs of a tertiary hospital. Previous studies show that the mean collaboration intensity score was 61.68 (±6.84).⁴⁶ Accepting an alpha risk of 5% and a beta risk of 10% in a bilateral contrast, a minimum of 50 professionals were needed to detect a minimum difference of 10% in the score. The sample was adjusted taking into account 15% losses.

Variables and measurement tools

The variables were collected using the following self-administered instruments:

• 1.

  Ad-hoc questionnaire with a total of 75 items grouped into four blocks: i) sociodemographic variables: sex, age, and marital status; ii) labour and economic variables: profession, job position, schedule, shift, administrative situation, monthly salary in euros, years of experience as a professional and in the current service; iii) training variables: highest academic degree, in the last three years: number of training courses taken, number of articles published, number of research projects developed, number of scientific meetings attended and number of communications presented in scientific meetings; iv) motivation and professional satisfaction variables: reasons why you chose the profession, reasons why you remain in the profession and degree of satisfaction with the chosen profession.

• 2.

  Intensity scale of Inter-professional Collaboration in Health (INCINS), Spanish version of the questionnaire “Intensité de la Collaboration Interprofessionnelle” by Sicotte et al.⁴⁵ translated into Spanish and validated by San Martín-Rodríguez et al.⁴⁴

The scale measures the intensity of Inter-professional collaboration and is composed of 16 items evaluable on a Likert-type scale that offers 5 response options for each item. Scores (1, 2, 3, 4, 5, or conversely, 5, 4, 3, 2, 1) are assigned according to the meaning of the response for the attitude they measure. Thus, it takes the value 1 for the categories “Very hierarchical, Little, Very Bad, Very Bad and Totally Disagree” and takes the value 5 for the categories “Very Equal, High, Very Much, Very Good, Very well and Totally Agree”. The maximum value of the scale is 80 points and the minimum is 16. In addition, three levels of IoT collaboration are established: high between 60 and 80 points, medium between 38–59.99 and low between 16−37.99. The 16 Items are distributed in four factors:

• • 1.

    Coordination. Collects information on the structuring of activities and the personal effort to reconcile clinical activities. With a total of 5 items (4, 6, 7, 8 and 9), with a minimum score of 5 and maximum of 25 points.

  • 2.

    Opinion on the patient’s perception. It shows the professional's opinion on the patient's perception. Both provide information about the holistic vision of the patient, the continuity of care and strategies that increase the quality of care. It corresponds to items 5 and 17, with a minimum score of 2 points and a maximum of 10.

  • 3.

    Shared activities. It provides information on the four clinical activities that are carried out as a team: data collection and problem identification, decision making, intervention planning and evaluation of results. With a total of 6 items (11, 12, 13, 14, 15 and 16) and a minimum score of 6 and a maximum of 30.

  • 4.

    Global Appreciation. They address issues of a global nature: level of hierarchy in Inter-professional relationships, daily planning of the common work of the team’s professionals, and the team's holistic vision of the patient's needs. It corresponds to items 1,3 and 10 and a minimum score of 3 and maximum of 15 points.

Following the recommendations of Bengoechea Calpe,⁴⁶ the factor scores are in turn categorised into high, medium and low as follows: 1) High_Coordination between 19 and 25, Medium_Coordination between 12 and 18.99 and Low_Coordination between 11.99−5 points, 2) Opinion on the patient's perception (Opp): Opp_high between 8 and 10, Opp_medium between 5 and 7.99 and Opp_low between 2 and 4.99 points, 3) Shared activities (SAc): SAc_high between 22 and 30, SAc_medium between 14 and 21.99 points and SAc_baja between 6 and 13.99 points; and 4) Global appreciation (GA): GA_high between 11 and 15, GA_medium between 7 and 10.99 and GA_low between 3 and 6.99 points.

The INCINS has a reliability based on Cronbach’s α coefficient⁴⁹ for the total scale of .9 and for each of the factors of .808 in Coordination, .611 Opinion on the patient's perception, .863 in Shared activities and .665 in Global appreciation.⁴⁴

Data collection

To access each of the ICUs, authorisation was requested from the centre’s Nursing Management through a letter of introduction and a personal interview to explain the project. Given the favourable attitude of the director, the same procedure was followed with the heads of Nursing Management, the Service heads and the Supervisors or Coordinators of each ICU.

The nursing supervisor or coordinator of each ICU was the person in charge of providing the questionnaires to the nurses and doctors, and assigning a mailbox within the same service where the completed questionnaires were deposited.

Once the questionnaires were administered (about a month), two rounds of scheduled collection were carried out, one week and fifteen days after delivery, and a third round depending on the degree of participant response.

Each potential participant was provided with an envelope containing: i) a cover letter containing the data of the main researcher, the objective of the study, and the importance of collaboration; ii) explanatory sheet with instructions for correct registration and an email address to consult possible questions; iii) informed consent iv) ad-hoc questionnaire of sociodemographic, economic and professional variables and motivation and professional satisfaction; v) INCINS Scale.

Data analysis

Firstly, a descriptive analysis of all the variables and the 16 items of the INCINS was carried out by calculating the distribution of frequencies and percentages for the categorical variables and, mean, standard deviation (SD) and other indices of central tendency and dispersion for the quantitative type. The normality of the quantitative variables was analysed using the Kolmogorov–Smirnov test.

Secondly, to study the relationship between the INCINS score and qualitative variables, statistical analysis was carried out using the non-parametric Mann–Whitney U test and the Kruskal–Wallis test in the case of variables with more than 2 categories. The linear relationship was analysed using Spearman's correlation and the relationship between qualitative variables using the Chi Square test (Fisher’s exact test in the case of frequencies <5).

Finally, multivariate linear regression models were built for the INCINS global score and for each of the factors. Those with clinical and bibliographic meaning (age, sex, profession, satisfaction, shift, schedule, years of profession and years of experience in the current workplace) and that showed a p value <.1 in the variables were included as explanatory variables. univariate analysis. The results were shown with the beta coefficient, p value and 95% confidence intervals (CI).

For the construct validation process of the instrument, an exploratory factor analysis (EFA) was carried out on the questionnaire and also on the four factors that make up said instrument. This analysis allowed us to contrast the results of the construct validation of the Spanish version of the instrument carried out by San Martín-Rodríguez et al.⁴⁴ Convergent validity was analysed by calculating the Pearson correlation between the results obtained in the validation of the instrument by San Martín-Rodríguez et al.⁴⁴ and, the results obtained in the present study in the described context. The reliability of the instrument was carried out through internal consistency by calculating Cronbach’s α coefficient of the items (correlation between the items of the instrument) and also between the four factors that make up the instrument.

The statistical packages SPSS (version 23.0) and R Studio (V2.5.1) were used, considering the differences statistically significant with p < .05.

Ethical considerations

The participants were informed about the nature of the study, its voluntary nature and the confidential treatment of the data collected. All participants signed an informed consent. The study was approved by the Clinical Research Ethics Committee (CEIC) of the Hospital Clínico de Barcelona, Registry 2012/7653 and was carried out in accordance with the principles of the Declaration of Helsinki and current regulations. The confidentiality of the participants was maintained at all times in compliance with the data protection law 2016/679.

Results

Sample description

The sample consisted of 102 professionals: 91 nurses (90%) and 11 doctors (10%).

Table 2 shows the sociodemographic and professional characteristics. The average age of the sample was 41.7 (±9.9) years, with 84.3% being women. The average age at the start of the profession was 22.6 (±3.6) years, with an average of 18 (±9.7) years of experience and 9.8 (±8.1) years in the current position. 37.3% of the professionals worked mainly on the night shift and 84.3% had a permanent contract. 70.6% had post-university training, the most common being a Master's degree or specialty in 30.4%.

Table 2.

Sociodemographic and professional characteristics.

		n (%)
Age
Mean (SD)		41.7(9.9)
Sex	Man	16 (15.7%)
	Woman	86 (84.3%)
Profession and job characteristics	Coordinator/Supervisor	4 (3.9%)
	Nursing assistant	87 (85.3%)
	Associate physician	9 (8.8%)
	Doctor with on-call contract	2 (2.0%)
Shift	Morning	32 (31.4%)
	Afternoon	18 (17.6%)
	Weekend	7 (6.9%)
	Night	38 (37.3%)
	Others	7 (6.9%)
ICU length of service
Mean (SD)		9.77 (8.1)
Professional length of service
Mean (SD)		18.03 (9.7)
Start of profession
Mean (SD)		22.64 (3.6)
Timetable	Fixed	89 (87.3%)
	Shifts	6 (5.9%)
	Others	7 (6.9%)
Administrative situation	Permanent contract	86 (84.3%)
	Interim contract	1 (1.0%)
	Temporary contract	15 (14.7%)
Salary	Under 1500 €	13 (12.7%)
	From 1500 to 2000 €	55 (53.9%)
	From 2000 to 2500 €	26 (25.5%)
	From 2500 to 3000 €	4 (3.9%)
	From 3500 to 4000 €	2 (2%)
	Over 4000 €	2 (2%)

SD: standard deviation.

In the last 3 years, 15.7% (n = 16) of professionals had not taken any training course, 55.9% (n = 57) had taken between 1 and 4 courses and 28.4% % (n = 29) had participated in more than 4 training courses. Furthermore, 69.6% (n = 71) did not have published articles, 55.9% (n = 57) had not participated in research projects and 61.8% (n = 63) did not have communications presented in scientific meetings.

The main reasons for choosing the profession were the influence of the family in 29.4% (n = 30), professional recognition in 30.4% (n = 31) and vocation in 29.4% (n = 30). Finally, 69.6% (n = 71) of the professionals reported continuing with the profession for personal satisfaction and 81.3% (n = 83) of the professionals were between very satisfied and moderately satisfied with the chosen profession.

Intensity of inter-professional collaboration in health

The psychometric evidence in the study context showed a Cronbach’s α for the total scale of .9 and for each of the factors it was .9 in Coordination, .9 Opinion on the patient’s perception, .9 in Shared activities and .9 in Overall appreciation. These 4 factors explain 61.5% of the total variance according to the factor analysis in principal components and with Varimax rotation.

The overall mean of Inter-professional Collaboration Intensity was 50.9 (±9.1). 15.1 (±3.5) points were obtained in the Coordination factor, 7.3 (±1.7) in Opinion on the patient’s perception, 19.4 (±4.3) points in Shared activities and 8.8 (±1.8) points in Global appreciation. Table 3 shows the descriptives of the items that make up the INCINS, grouped into its four factors: Coordination, Shared Activities, Opinion on the patient's perception and Global appreciation.

Table 3.

Relationship between the INCINS and the sociodemographic and professional variables.

INCINS		Total	Coordination	Opp	sA	gA
		p-Value^1,2	p-Value^1,2	p-Value^1,2	p-Value^1,2	p-Value^1,2
		Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
		.011*^,^1	.076^1	.059^1	.031*^,^1	.072^1
Age	23−30 years	52.9 (8.1)	14.9 (3.9); 14 (4.5)	7.9 (1.2)	20.7 (3.6)	8.8 (1.7)
	31−40 years	53.9 (8.8)	16.3 (3.6); 16 (4.0)	7.7 (1)	20.36 (4.4)	9.18 (1.7)
	41−50 years	46.9 (9.2)	14.1 (3.2); 14 (4.0)	6.7 (1.8)	17.6 (4.7)	8.1 (1.7)
	51−63 years	51.1 (8.2)	15.0 (3.4); 16 (4.0)	7.1 (2.5)	19.7 (3.1)	9.1 (1.8); 9 (3.0)
		.006*^,^2	.003*^,^2	.043*^,^2	.126^2	.019*^,^2
Sex	Man	49.8 (8.9)	14.7 (3.5)	7.2 (1.8)	19.1 (4.2)	8.6 (1.9)
	Woman	56.5 (7.9)	17.3 (2.8)	8.1 (1.5)	21.1 (4.9)	9.4 (1.0)
		.003*^,^2	.001*^,^2	.672^2	<.001*^,^2	.009*^,^2
Profession	Nurse	49.9 (8.8); 50 (12.0)	14.8 (3.5); 15 (5.0)	7.3 (1.7); 8 (1.0)	19.0 (4.3); 19 (6.0)	8.6 (1.8); 9 (3.0)
	Doctor	58.8 (7.7); 59 (4.0)	18.2 (2.6); 18 (2)	7.5 (2.1); 8 (2.0)	22.7 (2.6); 23 (2.0)	9.8 (1.2); 10 (1.0)
		.687^1	.316^1	.384^1	.915^1	.584^1
Shift	Morning	51.1 (8.8)	15.0 (3.3)	7.2 (1.5)	19.8 (4.2)	8.8 (1.6)
	Afternoon	49.4 (7.2)	14.4 (2.9)	7.1 (1.2)	18.6 (3.3)	9.2 (1.8)
	Weekend	53.9 (4.2)	15.7 (2.9)	7.7 (1.0)	21.0 (2.4)	9.0 (1.3)
	Night	49.8 (10.9)	15.1 (4.1)	7.1 (2.3)	18.8 (5.3)	8.4 (2.1)
	Others	55.0 (6.9)	16.5 (3.2)	8.4 (1.3)	20.6 (2.9)	9.2 (0.9)
		.182*^,^2	.674*	.024*	.148*	.916*
	<10 years	52.0 (9.5)	15.3 (3.8)	7.7 (1.4)	20.0 (4.5)	8.8 (1.8)
Length of service in ICU	≥10 years	49.5 (8.4)	15.0 (3.2)	6.9 (2.1)	18.7 (4.1)	8.7 (1.7)
		.043*^,^1	.139^1	.226^1	.099^1	.104^1
Professional length of service	Up to 10 years	54.0 (8.2)	15.6 (4.1)	8.0 (1.3)	21.0 (3.4)	9.0 (1.6)
	From 11 to 20 years	51.6 (9.1)	15.8 (3.6)	7.4 (1.0)	19.3 (4.5)	8.9 (1.9)
	From 21 to 30 years	47.2 (9.5)	13.9 (3.1)	6.9 (2.2)	18.1 (4.8)	8.1 (1.7)
	From 31 to 40 years	51.6 (7.3)	15.4 (2.8)	7.0 (2.5)	19.5 (3.4)	9.4 (1.7)

INCINS: intensity of inter-professional collaboration in health; Opp: opinion on patient perception; sA: shared activities; gA: global appreciation; SD: standard deviation.

*

p < .05.

1

Kruskal–Wallis test.

2

Mann–Whitney U test.

Variables related to the intensity of Inter-professional collaboration

Table 3 shows the differences in the INCINS total scores and by factors depending on the characteristics of the professionals. The 2-to-2 multiple comparisons performed are not shown in the table.

There is a negative linear relationship (rho: −.202, p = .042) between age and the INCINS total score. Overall, differences are observed in the INCINS total score (p = .011) as well as in the SAc factor (p = .031) depending on age. In comparisons 2 to 2, it is observed that professionals in the 23−30 year old group and in the 31−40 year old group present statistically higher scores in the Opp factor compared to the 41−50 year old group (p = .043 and p = .025, respectively). Furthermore, the 31−40 year old group presents higher Ag factor values (p = .029) compared to the 41−50 year old group. The mean of the SAc factor in the 23−30 years and 31−40 years groups was higher than in the 41−50 years group (p = .021; p = .009 respectively). In the sex variable, the men in the sample showed higher scores in all factors (p < .05) except for the SAc factor.

In relation to the profession, due to the limited number of doctors included, these weighted associations offer only a first approximation. The results of the INCINS applied to the group of nurses and doctors working in the 6 ICUs show that the scores for the intensity of total collaboration (p = .001), coordination (p = .002), SAc (.004) and Ag (.013) are statistically higher in the group of doctors compared to that of nurses (Fig. 1). No statistically significant differences were observed between nurses and doctors in relation to the Opp factor.

Figure 1.

Relationship between profession and INCINS factors.

Regarding the relationship of INCINS scores with professional variables, negative linear relationships were observed between years of experience and the overall INCINS score (rho: −.202, p = .042), and the Opp factor score (rho: −.241, p = .015). The results showed statistically significant differences between the groups of years of experience stratified by categories and the total INCINS score (p = .043), with the average being higher in the group of professionals with ≤10 years of experience. Comparisons 2 to 2 show in this case that the values of the Ag factor in the group 21–30 years old were lower compared to the group ≤10 years old and the group >30 to 40 years old (p = .035, p = .037), respectively. The relationship between years in the current position and the 4 factors showed only negative linear relationships in the case of the SAc and the Opp (rho: −.211, p = .033; rho: −.244, p = .014) respectively.

In relation to the professional satisfaction variable, statistically significant differences were observed between the total INCINS and the degree of satisfaction, with the average being higher in the group of very satisfied professionals (p = .037). Likewise, a higher score was observed in the Coordination factor (p = .033) in the group of very satisfied professionals. In the rest of the factors, no statistically significant differences were shown. Comparisons 2 to 2 only show higher Ag levels in the “Very satisfied” group compared to the “Somewhat satisfied” group (p = .032).

Finally, statistically significant differences have been observed in the total INCINS score and the variables related to research activity. Professionals with an active research activity, participating in projects, scientific publications and other related activities, had higher values in the total INCINS score (p < .05). These results were replicated for all the variables analysed also in the SAc factor (all p < .05). In the rest of the factors, higher scores were obtained in the group of professionals with research activity only in some of the variables analysed, especially in the case of the Coordination factor. Furthermore, in the Coordination factor, professionals with training courses, published articles and participation in projects presented higher scores in the total INCINS (p < .05).

Multivariate models for IoC

Five multivariate models have been built for the total INCINS score and the score for each of the 4 factors, including in the linear regression models as explanatory variables those that have shown a p value <.1 in the univariate analysis (age, sex, type of professional, years of experience, satisfaction and the number of articles published ≥1 of the research activity.

The multivariate models show that the group of doctors have a higher overall score in the INCINS questionnaire compared to the group of nurses independently of the rest of the included variables (b: 6.882, p = .037, 95% CI: .425–13.210). Similarly, medical professionals have a higher score in the coordination factor compared to nurses independently of the rest of the variables (b: 2.999, p = .020, 95% CI: .521–5.524). Finally, no variables independently related to higher scores in the Opp, SAc and Ag factors were identified.

Discussion

The average level of Inter-professional collaboration in our study coincides with a large part of other studies carried out in ICU and in different cultural contexts.^3,31,50,51

The results shown have made it possible to confirm, like other publications, that, in the study of Inter-professional collaboration, there are two constants regarding the characteristics of its participants: the participation of nurses is higher than that of doctors and the participation is maintained. predominance of women in the nursing profession,^{46,47,50–52} and that of men, traditionally associated with the medical profession, is no longer representative in this profession.^53–55 In our context, the study by the Barcelona College of Physicians⁵⁶ corroborates this trend in the three indicators it studies: doctors under 65 years of age (54.7% women, 45.3% men), resident doctors (64.7% women, 35.3% men), and medical students (68% women, 32% men).

Also, the percentage of participation of nurses in our study is higher (85.3%) than that of doctors (14.7%), this percentage being similar to those presented by Del Barrio Linares et al.⁵² in their study on 6 ICUs in two tertiary hospitals where nursing participation (78.7%) also exceeded that of doctors (21.3%). International multicentre studies in the ICU context also share the same distribution despite geographical diversity^57–59 However, in other studies^51,60 carried out in different Latin American countries, doctors participated in a higher percentage than nurses. This difference could be related to the geographical area, which leads us to think about cultural factors that facilitate or encourage participation depending on the profession.

The average age in our study coincides with that of other studies^{46,47,51,52,61} with the exception of Friganović and Selič's study⁶² where the participants are younger (between 26–35 years) and have less seniority in the ICU. At the national level, the characteristics of public or private entities of the centres can influence the average age of ICU professionals and the years of experience of ICU nurses, being lower in the private centre,⁴⁶ than in the public.⁴⁷ This circumstance does not occur in the study by Friganović and Selič,⁶² carried out in 5 public university hospitals in Zagreb and that leads us to think about other factors related to the cultural context and socioeconomic development in a health system that requires a large demand for nurses in specialised services such as ICUs.

In the workplace, in our study we can speak of a stable and expert work group where more than 80% of the professionals have an employment contract with a permanent position, have a fixed shift/schedule and with little rotation. However, in other studies^46,47,61 we find a notable variability in the distribution of shifts and weekly hours, which makes a comparative analysis difficult and leads us to think, on the one hand, that the organisation of working conditions responds to the interests and realities of each centre as a reflection of the social health policies motivated in recent years by the lack of nurses,^63–65 and the reduction of costs^66,67 and on the other hand, according to the published evidence we find disagreements on the impact of schedules and shifts of work on the well-being of professionals and on the quality and safety of the care provided.

With regard to postgraduate training, the most reported in the studies^46,47,51,61 was the master's degree, with our study being the one with the highest percentage of professionals with this training (30.4%), followed by the study of San-Martín et al.⁵¹ (25.5%), while Bengoechea Calpe⁴⁶ (17%), Saldaña⁶¹ (6.2%) and Friganović and Selič⁶² (8.7%), present considerably lower percentages of professionals with master's degrees. The number of studies,^46,61 that report on the doctoral training of professionals is smaller, less than 1% in the case of nurses.

Regarding the Intensity of Inter-professional Collaboration (IoC), the results of the study show that doctors have a greater perception of the total IoC and the coordination factor compared to nurses. This result would be in line with other studies carried out^68–70 but contrasts with others that demonstrate a greater willingness of nurses to Inter-professional collaboration.^60,71 This fact leads us to think that nurses and doctors have different perceptions of Inter-professional Collaboration and that contexts, in the case of nurses, influence when valuing collaboration, as shown by the results of Collette et al.’s study⁷⁰ where doctors from a hospital perceived greater collaboration than nurses in all areas of care, while the perception of nurses varied depending on the area of care (emergency, hospitalisation or ICU).

With respect to the relationship between sociodemographic variables and IoC, our results show a relationship between total IoC, the Opp factor and age. Younger professionals (<40 years old) show a better perception of the opinion that patients have of their work. In this sense, studies focused on nurse-patient relationships agree that the perceptions of nurses with an average age close to or greater than 40 years^72–77 have a lower perception of their care in relation to that declared by patients. whether in the context of the ICU^73,74 of a chronic patient,^72,75 or of hospitalisation.^76–80 Nurses with more years of experience in the ICU show lower total IoC scores, results that coincide with those of other studies.^46,47

Training has proven to be a strong ally of the IoC. In this sense, Bengoechea Calpe⁴⁶ identifies its specialised training programme for critical care nurses as a possible cause of the high level of IoC results. Our results allow us to affirm that nurses with some training activity, publication of articles or participation in projects, presented higher scores in the total INCINS and in the Coordination factor. Numerous studies highlight this same correlation between specific training in the ICU and a higher level of IoC,⁸¹ which also turns out to be a protective factor of the perception of non-beneficiary treatment⁵⁸ and a key element of teamwork.^82,83

Professional satisfaction is shown as an important element in the IoC study. In our study we found a relationship between total IoC and the group of professionals who declare themselves “Very satisfied”. This group also scores above the average in the Coordination factor and presents higher levels of the Ag factor in relation to the “Somewhat satisfied” group. These results are in line with other studies that quantify the degree of satisfaction of professionals,^51,52,60,84 and with studies where professional satisfaction is presented as one of the categories emerging from content analysis.⁸⁵

Limitations

Our study poses several limitations, mainly the small size of the group of doctors in the case of therefore being able to conclude and generalise the results in. However, this was not the main objective of the study, so the statistical power proposed a priori has been assumed with the total of 102 professionals included. Another point to take into account is that the study is carried out in the context of the ICU and this aspect can influence a more unified conception and interpretation of each of the items that have been measured. Future research is necessary in order to analyse the differences depending on the health profession, including other professionals as well.

Conclusions

The intensity of Inter-professional Collaboration between nurses and doctors working in the multipurpose ICU of a tertiary hospital is moderate. Doctors have a higher level of Collaboration Intensity compared to that of nurses independently at a global level and, in particular, they present a greater perception of coordination. Younger professionals, with fewer years of professional experience, those who are satisfied with their profession and have an active scientific activity obtain higher values of intensity of Inter-professional collaboration. The results of this study allow the possibility of detecting areas for improvement, especially regarding the continued training of nurses and their scientific activity.

Funding

No funding.

Conflict of interests

The authors declare that no conflict of interest has existed or exists.

Acknowledgements

We thank Dr. San Martín for her authorisation to use the INCINS questionnaire and for her availability for its proper functioning and understanding.