The authors have declared that no competing interests exist.
Several organizational factors facilitate or hinder information transfer in palliative care teams. According to past research, organizational factors that reduce information transfer include the inconsistent use of shared electronic patient files, frequent changes of healthcare staff, a lack of opportunities for personal exchange, and a lack of evaluation of collaborative processes. Insufficient information sharing between professionals can negatively impact patient safety, whereas studies have shown that some organizational factors improve collaboration between professionals and thus contribute to improved patient outcomes. The main purpose of this study is thus to investigate whether, and if so how, organizational factors contribute to successful information exchange in palliative care teams in Switzerland, while also accounting for the different care contexts of primary and specialized palliative care. A nationwide survey was aimed at medical professionals working in palliative care. In total, 379 participants (mean age = 49.8 years, SD = 10.3) were included in this study. Two main outcome variables were examined: healthcare providers’ satisfaction with information transfer in their team and their overall satisfaction with communication in their team. Hypotheses were tested by employing structural equation modeling. Findings revealed that the strongest predictors for effective information transfer in palliative care teams were sufficient opportunities for face-to-face meetings and supervision alongside feedback tools to improve collaborative practices and the application of guidelines and standards for collaboration. Face-to-face meetings were an even greater contributor to information transfer in specialized settings, whereas sharing the same work-based values with colleagues was considered more important in primary settings. Results from this study contribute to the existing literature elucidating how information transfer is facilitated in the field of palliative care. If proposed measures are implemented, this could possibly improve patient outcomes in palliative care. Furthermore, the findings can be useful for healthcare organizations and associations to make more efficient resource allocation decisions with the aim to optimize information transfer within the workforce.
The topic of interprofessional collaboration in healthcare has received considerable attention in recent decades and has specifically gained importance in the interdisciplinary field of palliative care (PC). Due to the broad spectrum of PC patients’ needs, successful PC delivery relies on efficient collaboration between medical doctors, nurses, and a wide range of support services within and across different institutions and settings [
Palliative care teams can be described generally as complex, flexible, yet adaptive structures that shape the team, its members, and its environment [
There are several organizational barriers to information transfer. Often originating from a lack of structural resources and time pressure, organizational factors that can hinder information transfer in PC teams include the lack of standardized guidelines for collaboration, the inconsistent use of shared electronic patient files, and the lack of opportunities for personal exchange and feedback through meetings or supervisions [
The primary objective of this study is to test the influence of select organizational variables on the perceived quality of information exchange of PC providers. Moreover, this study exploratively assesses how the quality of information transfer affects PC providers’ perceived satisfaction with collaboration, as well as their satisfaction with job-related tasks.
The methods section in this paper is presented in three parts. First, the paper will examine the organizational variables that facilitate or hinder information exchange in the study sample of Swiss PC providers. Second, the paper investigates whether information transfer affects PC providers’ satisfaction with communication, and consequently, their satisfaction with job-related tasks. Third, the paper investigates if certain organizational determinants for information transfer are context dependent. For this purpose, the two settings of primary palliative care (PPC) and specialized care (SPC, test for moderation) are distinguished.
The development of interprofessional collaboration in healthcare has been shown to vastly benefit from the formalization of rules and procedures [
Establishing standardized procedures is best achieved by the dissemination and application of guidelines and standards for best practices for interprofessional collaboration [
With respect to the healthcare setting, frequent transitions of healthcare providers hinder information flow within the team [
Thus, based on the theoretical framework of Bainbridge et al. 2010, the primary objective of this study is to examine the influence of certain organizational variables on the perceived quality of information exchange of Swiss PC providers. The following hypotheses concerning organizational determinants were tested in the first part of this study:
This part of the study investigates what factors affect PC providers’ satisfaction with communication in PC teams. Ultimately, team communication in healthcare is more than just accurate information transmission. Multidisciplinary PC teams are socially constructed groups that operate at the intersection of multiple institutional and professional cultures [
Moreover, healthcare research indicates that an open communication culture is facilitated by a clear delineation of roles and tasks among team members, alongside collective risk-taking [
The study also explores how satisfaction with communication affects providers’ satisfaction with job-related tasks. Impaired communication in PC teams can lead to increased misunderstandings at the workplace, which can trigger disputes within the workforce and lower job satisfaction for providers [
In order to investigate HCPs satisfaction with communication, the following hypotheses were tested:
This study investigates whether the importance of certain organizational variables for information transfer is moderated by the context of two different PC-settings, namely primary palliative care (PPC) and specialized palliative care (SPC), which are distinguished by the patients’ current condition [
Especially in PPC, general practitioners (GPs) and nurses face limited time and financial reimbursement for collaborative activities, which can result in gaps in information sharing and, consequently, in care shortages [
A greater degree of institutionalization can be expected in SPC, where different healthcare providers work together in proximity and where collaboration is often governed by existing guidelines and standards [
To date, explorations of the organizational factors that improve information transfer and the dissemination of patient information is rare in PC-related contexts. Therefore, the first and main objective of this study is to test the influence of the organizational variables mentioned above on the perceived quality of information exchange and the dissemination of patient information. Hypothesized predictors of information exchange in the team and hypothesized additive and interactive effects of information exchange in the team on satisfaction with communication and satisfaction with work-related tasks are depicted in
Hypothesized predictors of information exchange in the team and hypothesized additive and interactive effects of information exchange in the team on satisfaction with communication and satisfaction with work-related tasks (after Bainbridge et al., 2010).
To examine associations of organizational factors and information transfer on multiple levels, the study uses structural equation modeling [SEM,
The survey was aimed at healthcare professionals, primarily medical doctors and nurses, who were active in palliative care provision in Switzerland in 2017. The data collection via email was performed between September 19 and November 30, 2018. Three rounds of reminders, including informed consent, were sent out the following month. In Switzerland, we estimate the total number of palliative care providers, including GPs who regularly treat palliative care patients, at around 4500. We used a sample gathered from a wide range of healthcare units, including acute-care hospitals, nursing associations, hospices and nursing homes, which allows our research design to be informative without relying on transnational data. A total number of 1,111 healthcare providers who are actively involved in palliative care provision took part in the online study (f = 64.7%, m = 14.3%, mean age = 50.9 years, SD = 10.3). At around 24.5%, the response rate of this study can therefore be considered representative for the Swiss healthcare context.
In order to contact medical doctors and nurses, a two-step recruiting approach was carried out by identifying organizations of interest, which then recruited their employed or associated healthcare providers to complete the survey. The anonymity of responders was ensured at all times and the study data was handled in accordance with the Swiss law governing the use of scientific data.
The survey items were adapted from Bainbridge et al.’s (2010) tool for evaluation of healthcare provision [
Two items were selected as main outcome variables.
In order to assess information transfer in their care teams, PC providers were asked to evaluate the information exchange with their immediate team members (6-point Likert scale: “very good” to “very bad”).
To assess the satisfaction with overall communication in providers’ immediate work environments, the following item was selected: The communication within our organization/our institute is good (6-point Likert scale: “does not apply at all” to “fully applies”).
The following seven predictor variables for information transfer were measured: (1) the availability and use of internal guidelines and standards for collaboration in the providers’ immediate work environment (two dichotomy items: 0 = no, 1 = yes), (2) if a clear share of responsibility was present in the immediate work environment (4-point Likert scale: “yes,” “rather yes,” “rather no,” to “no”), (3) if regular opportunities for face-to-face meetings were present, (4) whether or not the team used electronic tools to manage patient files, (5) whether or not work processes were regularly evaluated with quality circles or feedback rounds, (6) whether or not PC providers had a case manager in their immediate work environment (all dichotomy, dummy-coded items: 0 = no, 1 = yes), (7) and whether or not there were frequent changes of caregivers in the immediate work environment (4-point Likert scale: “yes”, “rather yes”, “rather no”, and, “no”).
The following covariates were included in the analysis to control for gender (0 = male; 1 = female), age, position (leading vs. no leading position), socio-geographic workplace (1 = large city, 5 = small village in rural area), job description (nurse, medical doctor), and additional training in palliative care.
All hypotheses were tested using a structural equation model via the SEM function of the package ‘Lavaan’ (latent variable analysis, version 0.6–4, in R: Development Core Team 2012) [
In total, 379 participants, aged between 24 and 76 years (Mean = 49.8 years, SD = 10,3) were included in this study. The detailed sample description is summarized in
PPC (n = 229) | SPC (n = 150) | Overall (n = 379) | |
---|---|---|---|
Mean (SD) | 50.7 (9.86) | 48.4 (10.9) | 49.8 (10.3) |
Median [Min, Max] | 53.0 [25.0, 75.0] | 50.0 [24.0, 76.0] | 52.0 [24.0, 76.0] |
Male | 31 (13.5%) | 36 (24.0%) | 67 (17.7%) |
Female | 198 (86.5%) | 114 (76.0%) | 312 (82.3%) |
Nurses | 196 (85.6%) | 111 (74.0%) | 307 (81.0%) |
Medical doctors | 33 (14.4%) | 39 (26.0%) | 72 (19.0%) |
Larger city | 69 (30.1%) | 96 (64.0%) | 165 (43.5%) |
Other | 160 (69.9%) | 54 (36.0%) | 214 (56.5%) |
None | 70 (30.6%) | 39 (26.0%) | 109 (28.8% |
Yes | 159 (69.4%) | 111 (74.0%) | 270 (71.2% |
No | 35 (15.3%) | 13 (8.7%) | 48 (12.7% |
Yes | 194 (84.7%) | 137 (91.3%) | 331 (87.3%) |
No | 182 (79.5%) | 93 (62.0%) | 275 (72.6%) |
Yes | 47 (20.5%) | 57 (38.0%) | 104 (27.4%) |
No | 17 (7.4%) | 2 (1.3%) | 19 (5.0%) |
Yes | 212 (92.6%) | 148 (98.7%) | 360 (95.0%) |
No | 29 (12.7%) | 11 (7.3%) | 40 (10.6%) |
Yes | 200 (87.3%) | 139 (92.7%) | 339 (89.4% |
Standard deviations (SD) and zero-order correlations are provided in
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 Information exchange in the team | 1 | ||||||||||||||
2 Guidelines available | 0.15 |
1 | |||||||||||||
3 Application of these guidelines | 0.26 |
0.59 |
1 | ||||||||||||
4 E-Tools to share patient files | 0.02 | 0.09 | 0.05 | 1 | |||||||||||
5 Case Manager | 0.15 |
0.06 | 0.10 | 0.04 | 1 | ||||||||||
6 Feedback Tools | 0.32 |
0.21 |
0.25 |
-0.03 | 0.17 |
1 | |||||||||
7 Opportunities for face to face meetings | 0.58 |
0.22 |
0.23 |
-0.00 | 0.10 |
0.35 |
1 | ||||||||
8 Frequent changes of caregivers | -0.31 |
-0.11 |
-0.11 |
0.02 | -0.04 | -0.15 |
-0.28 |
1 | |||||||
9 Good division of responsibility | 0.28 |
0.21 |
0.23 |
0.09 | 0.15 |
0.24 |
0.25 |
-0.22 |
1 | ||||||
10 Colleagues share the same values | 0.44 |
0.07 | 0.24 |
0.01 | 0.03 | 0.21 |
0.28 |
-0.28 |
0.25 |
1 | |||||
11 Satisfaction with communication | 0.43 |
0.11 |
0.21 |
-0.06 | 0.04 | 0.19 |
0.32 |
-0.31 |
0.26 |
0.46 |
1 | ||||
12 Satisfaction with work-related tasks | 0.29 |
0.04 | 0.15 |
-0.06 | 0.01 | 0.14 |
0.20 |
-0.18 |
0.18 |
0.47 |
0.40 |
1 | |||
13 Workplace demographics | 0.08 | 0.08 | 0.06 | 0.05 | 0.19 |
0.07 | 0.16 |
0.07 | -0.02 | 0.06 | 0.03 | 0.04 | 1 | ||
14 Additional training | -0.03 | -0.01 | 0.05 | -0.05 | -0.01 | 0.02 | -0.03 | 0.12 |
0.04 | -0.03 | -0.05 | 0.10 |
-0.10 | 1 | |
15 Gender | -0.09 | 0.05 | 0.07 | -0.03 | 0.01 | 0.05 | -0.13 |
0.14 |
-0.08 | 0.08 | -0.03 | 0.05 | -0.05 | 0.12 |
1 |
Pearson correlation coefficient (1-tailed),
* indicates p <0.05.
** indicates p <0.01.
*** indicates p < 0.001. Correlations of binary variables should be interpreted with care.
Following best statistical practices, we report the measurement model on the full sample of n = 379 [
Internal guidelines and standards are relevant to information exchange; thus, their presence should improve information exchange within the team. We found little evidence in support of this hypothesis (β = -0.09, p = 0.08).
The use of those available internal guidelines and standards significantly explained the increase in information exchange within the team (β = 0.15, p<0.01).
The opportunity for face-to-face meetings (e.g. in the context of meetings, roundtables, and supervisions) significantly explained the increase of information exchange within the team (β = 0.48, p<0.001).
The use of electronic tools to manage patient files was not significantly correlated to an increase in information exchange within the team (β = 0.03, p = 0.50).
The regular evaluation of work processes with quality circles or feedback rounds predicted information exchange within the team (β = 0.10, p<0.041).
The presence of a case manager in the immediate work environment results did not significantly explain changes in information exchange within the team (β = 0.07, p = 0.09).
Frequent changes of caregivers in a team indeed predicted general information exchange in teams negatively (β = -0.15, p<0.001).
By applying a structural equation model, there was no support for hypotheses H1a, H1d, and H1f (for an overview, see
* indicates p < 0.05 ** indicates p < 0.01 *** indicates p < 0.005. Standardized effects are given. All effects are controlled for position (lead/no lead), type of caregiver (context), place of work (city vs. countryside), function (job description), additional training, and gender; n = 379.
The more frequently information is exchanged in the team, the more satisfied are care providers with communication (β = 0.27, p<0.001).
A clear division of responsibility within the team positively predicted providers’ satisfaction with communications (β = 0.10, p<0.032).
When colleagues felt that they shared the same values, this was positively associated with their satisfaction with communication (β = 0.32, p < 0.001) as well as their satisfaction with work-related tasks (β = 0.37, p < 0.001).
Exploratively, this study investigated the extent to which providers’ satisfaction with their communication affects their satisfaction with job-related tasks (H2d). Indeed, the results provide considerable evidence that providers’ satisfaction with communication positively predicts their satisfaction with work-related tasks (β = 0.24, p<0.001). Little support for hypotheses H2a-H2d were found (see
In order to test the moderating effect of care-giving context of primary care versus specialized care on select organizational factors, cross-group structural equalization modeling was employed. In both groups of PPC (n = 229) versus SPC (n = 150), the model explained a considerable amount of variance of information exchange in the team (34%, 46%), as well as providers’ satisfaction with work related tasks (31%, 28%).
Unsurprisingly, colleagues who share the same values reported higher satisfaction with communication in their team. Within PPC (β = 0.42, p < 0.001), individuals who reported sharing the same care-based values and ideals indeed showed higher predictive scores of satisfaction with communication as compared to SPC (β = 0.14, p = 0.06).
The use of e-tools to share patient files is expected to play a stronger role for information exchange in SPC. The use of e-tools to exchange patient records showed no significant effect on information exchange in teams, neither in PPC (β = 0.01, p = 0.85) nor in SPC (β = 0.06, p = 0.31).
Opportunities for interprofessional exchange, such as face-to-face meetings, are expected to contribute to greater information sharing in SPC. Indeed, interprofessional exchange in the form of face-to-face meetings and supervisions had a strong effect for the setting of SPC (β = 0.60, p < 0.001), compared to PPC (β = 0.40, p < 0.001).
A moderation analysis including X2 difference tests were performed to test whether the group differences of the paths are statistically significant. First, the researchers tested for measurement invariance across the groups by comparing the unconstrained multi-group model with a constrained multi-group model where the respective factor loadings and measurement intercepts were set equal for both groups. A difference test on χ2 showed no difference between the two models (Δχ2 [
Models | χ2 | df | CFI | RMSEA | Δχ2 | Δdf |
---|---|---|---|---|---|---|
Unconstrained model | 76.98 | 56 | 0.95 | 0.044 | ||
86.60 | 57 | 0.93 | 0.052 | 9.61 |
1 | |
77.44 | 57 | 0.95 | 0.044 | 0.46 | 1 | |
84.03 | 57 | 0.94 | 0.050 | 7.05 |
1 | |
82.50 | 57 | 0.94 | 0.048 | 5.50 |
1 |
Note: * indicates p < 0.05
** indicates p < 0.01
*** indicates p <0.001.
Nr | Hypotheses | Value | True/False |
---|---|---|---|
H1a | The availability of internal guidelines and standards increases information exchange within the team. | Not confirmed | |
H1b | The use of those available internal guidelines and standards increases information exchange within the team. | Confirmed | |
H1c | The opportunity for face-to-face meetings (e.g. in the context of meetings, roundtables and supervisions) increases information exchange. | Confirmed | |
H1d | The use of electronic tools to manage patient files increases information exchange within the team and fosters continuous exchange of patient information. | Not confirmed | |
H1e | The regular evaluation of work processes with quality circles or feedback rounds increases information exchange within the team. | Confirmed | |
H1f | The presence of a case manager in the immediate work environment results in increased information exchange within the team | Not confirmed | |
H1g | Frequent changes of caregivers in a team hinder general information exchange. | Confirmed | |
H2a | The extent of information exchange in the team predicts providers`satisfaction with communication. | Confirmed | |
H2b | A clear division of responsibility within the team increases information exchange within the team and fosters continuous exchange of patient information. | Confirmed | |
H2c | Sharing the same values increases providers`satisfaction with communication | Confirmed | |
H2d | Provider`s satisfaction with communications affects their satisfaction with work-related tasks. | Confirmed | |
H3a | Colleagues who share the same values report higher satisfaction with communication especially in the primary palliative care setting. | Confirmed | |
H3b | The use of E-tools to share patient files is expected to play a stronger role for information transfer and satisfaction with communication in specialized settings. | Not confirmed | |
H3c | Opportunities for interprofessional exchange, such as face- to face meetings are expected to contribute to better information sharing in both settings, but especially in specialized settings. | Confirmed |
Note: * indicates p < 0.05
** indicates p < 0.01
*** indicates p < 0.001.
A vital aspect of quality of care in PC is the extent to which information is shared between HCPs who work together closely in a team. To optimize the quality of PC services provided, identifying organizational factors that enable explicit collaboration between coworkers is of utmost importance. Using a survey instrument, this study investigates the extent to which information transfer affects PC providers’ satisfaction with collaboration and, ultimately, their satisfaction with job-related tasks.
This paper contributes in two main ways to the existing literature on how information transfer is facilitated in the field of palliative care. First, we demonstrate the need for personal, face-to-face information exchange for PC providers who work in a team. Although it would be expected that electronic patient records in particular are essential for successful information sharing in the healthcare sector, this specific sample of palliative care providers highlights the fact that to date, opportunities for face-to-face meetings are paramount for successful information exchange in PC. Face-to-face meetings may be useful to support the social functions of healthcare teams, improving mutual respect in the care team, allowing team members to solve problems more efficiently, and facilitating the transmission of organizational culture [
Second, this research underlines that the success of interprofessional collaboration in PC is partially care-context dependent. This is due to the fact that primary and specialized care has evolved in isolation historically, with SPC showing higher levels of institutionalization and regulatory pathways for collaboration than PPC [
The results point to striking evidence that some organizational aspects affect successful information exchange between PC providers more drastically than others. Sufficient opportunities for face-to-face meetings and supervisions, feedback-tools to improve collaborative practices, and the application of guidelines and standards for collaboration were strong predictors for information exchange in PC teams. Based on our results, it is recommended that whenever institutes (hospitals, hospices, retirement homes, ambulant nursing organizations etc.) are establishing new collaborative processes in PC provision, they should aim to grant sufficient time for personal exchanges among the PC providers. Further, collaborative processes should be regularly evaluated in order to maintain and improve a sustainable social network between suppliers. Staff should be involved as early as possible in the improvement process to help ensure that changes correspond with their philosophy of collaboration [
Aligned with prior expectations, the study also found that colleagues who share the same work-related values reported significantly improved information transfer. High-functioning teams in healthcare settings should not only integrate principles of team-based care, but also agree on shared goals and values for delivery of patient-centered care [
Furthermore, the study found considerable evidence that providers’ shared values, as well as their satisfaction with communication, positively predict their satisfaction with work-related tasks. This is a key finding, as the satisfaction of team members is linked to staff retention, which is a critical element for team functioning, as well as a predictor for good healthcare provision [
The findings also suggest that frequent changes to caregiving negatively impact the information exchange in the team, as loss of information and misunderstandings occur easily when care teams are fluctuating. Much information is lost when health professionals change teams, therefore each PC team member should be trained to maintain the flow of information. Furthermore, making available written records of standards and guidelines on work procedures to all team members is recommended.
Unexpectedly, the study found little evidence that e-tools used to share patient files facilitate information transfer among PC team members. This is partly due to the fact that in Switzerland, e-tools for managing patient files are not yet mandatory for all healthcare providers and are far from being universally established [
This research contributes to a growing body of knowledge pointing to organizational differences between SPC and PPC, which are important to understand when considering future interventions to meet patients’ palliative care needs.
Given the diversity of organizational enablers for information transfer and collaboration presented above, we recommend further investigation into which variables affect information transfer while specifically distinguishing for PC teams in different care facilities and care contexts.
As with any research, we recognize the following limitations of our study. First, some of the dependent variables should be better operationalized in future research. This applies, for example, to the impact of CMs on information transfer in PC teams. Because the fields of activity and applications of CMs in palliative care in Switzerland remain largely unclear, we suggest that the role of CMs in palliative care be clarified in future investigations before definitive statements can be made about their impact on information transfer. The same caution applies to e-tools to share electronic patient files, which may depend on user characteristics, and user interface and user-friendliness; both of which contribute to successful communication in certain environments.
Perhaps the main limitation of this study is that it lacks the attributes of a standardized questionnaire to assess information transfer and organizational aspects of care. Future studies are advised to use the Care Process Self-Evaluation tool (CPSET), as seen in in the work of Seys and colleagues [
Particularly in the field of palliative care, institutions, employers, and other stakeholders, such as the federal administrations, desire to be informed about organizational factors that improve the exchange of information between PC providers. The present study is intended to serve as a basis for recommendations as to how information transfer and communication can be improved by the establishment of certain organizational enablers in interdisciplinary PC teams.
(DOCX)
(DOCX)
The authors would like to thank Emily Reeves, Eveline Degen, and Birgit Schmid from the University of Applied Sciences and the University of Basel, Switzerland, for their contribution to the research project.