Background
Recent studies on care quality improvement and patient safety show that health care is still not safe and that the number of adverse events is underestimated [
1‐
3]. Health care across organization levels is often poorly organized, complex, and uncoordinated. Furthermore, not all patients receive consistent, high-quality medical care. The organization of care can be more effective when multidisciplinary teams are involved in its organization and if care is organized around medical conditions and care processes [
4,
5].
Care processes contain key interventions that support the diagnosis or treatment of patients. These key interventions contain unique bundles of products and services, and temporary firms can effectively achieve their delivery. Temporary firms are health care providers who operate together when a specific well-defined patient group is admitted to a care facility or institution. These teams contain members from different health care professions, have a shared clinical purpose, and have direct care responsibilities [
6,
7]. They work in a complex environment, under interactive and dynamic conditions, and their membership frequently changes. Therefore, these teams are action teams [
7,
8]. The challenges for these teams are effective communication, coordination, and control over the care process [
7].
A care process has five key characteristics that affect the organization of care: coordination of the care process, patient-focused organization, communication with patients and family, collaboration with primary care, and follow-up of the care process [
9]. The Care Process Self-Evaluation Tool (CPSET) assesses these five key characteristics by using a 29-item Likert scale. It is based on the perceptions of team members involved in organizing a care process. The primary study assessing the validity and reliability of this tool was performed in 2007, and statistical analysis of the five factors produced Cronbach’s alpha values between 0.776 and 0.928 [
9]. In 2011, an international Delphi study was performed to identify indicators that affect multidisciplinary teamwork in care processes. This study showed that CPSET is a good tool for following up improvements in multidisciplinary teamwork [
10].
Since its original validation, CPSET has been used in different organizations and teams. The first aim of this study was to evaluate the stability of the psychometric properties of CPSET. The second aim was to calculate cutoff scores for the subscales and overall score, with the goal of helping health care managers rank the CPSET scores of their teams.
Discussion
This paper describes the psychometric properties of the CPSET and defines CPSET cutoff scores. The psychometric properties of the CPSET showed that this tool is valid and reliable for evaluating the organization of a care process as perceived by team members. The original five-factor structure with 29 items was confirmed by CFA. The reliability of the CPSET was measured by Cronbach’s alpha. The Cronbach’s alpha results in this study varied between 0.869 and 0.950 and were higher than those reported in the original 2007 study of Vanhaecht et al. (Cronbach’s alpha = 0.776-0.928) [
9]. This indicates that the scale is still reliable.
A multilevel analysis was performed at team and hospital levels. Our results showed that the ICCs of scores at the team level were higher than those at the hospital level. This means that there was less variance in CPSET scores within teams than within hospitals, which was expected. The ICCs in our study were low, and most of the variation could be explained by team and hospital variations. One possible reason for the low ICCs in our study was that teams were composed of professionals from different disciplines, with each team member having different perceptions about the organization of care. As shown in Table
3, profession, age, and gender significantly influenced CPSET scores. Medical doctors scored significantly higher than other health professionals on the overall CPSET scale and on the following subscales: ‘patient focused organization,’ ‘coordination of care,’ ‘communication with patient and family,’ and ‘collaboration with primary care.’ Paramedics scored significantly higher on the subscale ‘follow-up of care.’ Men scored significantly higher on the overall CPSET scale and the subscales ‘communication with patient and family’ and ‘collaboration with primary care.’ Team members between 20 and 39 years old scored significantly lower on the subscale ‘communication with patient and family’ than those in other age categories. Significant differences were found between younger ( < 40 years) and older (> 50 years) health care professionals on the subscale ‘coordination of care.’ Coordinators of care pathways scored lower on the subscales ‘coordination of care’ and ‘follow-up of care,’ perhaps because they tend to be more critical of the organization of care.
The differences we observed in the perceptions of medical doctors and nurses are consistent with those observed in previous research. In the present study, physicians perceived teams to be more organized than nurses in terms of teamwork, collaboration, and communication with nurses, which is consistent with the finding that physicians generally perceive teamwork to be better coordinated [
19‐
21]. A negative correlation exists between professional autonomy and the level of nurse-physicians collaboration [
22]. Different perspectives in communication can be caused by hierarchical factors, gender, different patient care responsibilities, different perceptions of requisite communication standards, and differences in training methods for nurses and doctors [
19]. Communication skills training can improve patient-nurse communication but not patient-doctor communication. Skills training that contains patient-centered communication can increase information exchange and continuity of care for patients [
23].
The lowest CPSET scores were observed on the subscale ‘communication with patient and family’ and ‘follow-up of care.’ Organizations need to improve on priorities, communication, and coordination of care, as suggested by Bates et al. [
24]. Relational coordination can be used to improve ‘follow-up of care.’ This framework can lead to better quality of care for patients, and health care providers reported fewer adverse events [
25]. Effective and safe hospital care depends on good teamwork. Greater teamwork results in higher patient satisfaction rates, higher nurse retention, and lower hospital costs [
26]. Multidisciplinary teamwork is essential for quality health care.
The 2012 review of Deneckere et al. showed that multidisciplinary teamwork can be supported by using care pathways [
7]. Care pathways can improve the work environment and organization of care, and can have a positive impact on the well-being of health care providers [
7,
27,
28]. Further research is needed on using the CPSET to study the effect of coordinating mechanisms, such as care pathways, on how health care providers perceive the organization of care.
Although the CPSET has been used for several years, health care managers have problems interpreting the CPSET scores of their team members. Therefore, we compiled a table of cutoff scores that will permit health care managers and team members to compare how they perceive the organization of their care process relative to other team members. By using the cutoff table as a starting point, team members can discuss how they can improve the organization of care. When different teams of the same care facility or institution complete the CPSET, the cutoff scores will help health care managers rank the teams in that facility or institution. However, this should be done carefully. The primary aim of the cutoff scores is to initiate discussions within teams. For example, they can look for possible reasons why they perceive the organization of care to be low and what they expect from other team members. They can also learn from actions taken by other teams. We hypothesize that teams that use care bundles, care pathways, or evidence-based protocols will have CPSET scores in the higher percentiles compared with teams that do not use quality improvement initiatives. But further research is needed to determine whether the structured care associated with quality improvement initiatives does indeed change the perception of health care providers according to the actual organization of care.
Some limitations of our study should be considered. One limitation is the risk of social desirability and selection bias. Team leaders and coordinators of care processes decided which specific team members would complete the CPSET. Hence, it is possible that not all team members or health care professionals involved in a specific care process were surveyed. Another concern is that the results of this study are based on data from two countries: Belgium and the Netherlands. Therefore, a comparable study should be conducted in additional countries. The validity of the CPSET is currently being tested in French, Norwegian, Italian, Portuguese, English, and German languages.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
KV, WS, and MP defined the design of the study and organized the data collection. SD, EVG, and DS coordinated the data cleaning and supported the participating organizations. LB, TM, RCB, and SK performed the statistical analysis. KV, SD, and DS prepared the first draft of the manuscript. All authors discussed the results and approved the final version of the manuscript.