Introduction
Faced with an aging population and the predicted rise in death rates, which will increasingly affect families and communities, the need to understand how the community copes with death is an important priority (Doka, Reference Doka2015; Kellehear, Reference Kellehear, Stillion and Attig2015). Understanding factors that are associated with more adaptive death attitudes and death competencies can inform the development of future health-promoting palliative care strategies that aim to encourage death conversations, death preparedness and planning, and acceptance of death as a normal part of life in the general community (Cox et al., Reference Cox, Bird and Arthur2013). It can also assist with service planning for end-of-life care and inform approaches to training and development for health professionals regarding improving competence to cope with death (Chan et al., Reference Chan, Tin and Wong KL2015; Claxton-Oldfield et al., Reference Claxton-Oldfield, Crain and Claxton-Oldfield2007; Sansó et al., Reference Sansó, Galiana and Oliver2015).
There is limited empirical knowledge about 21st century community attitudes toward death and dying (Collins et al., Reference Collins, McLachlan and Philip2017; Cox et al., Reference Cox, Bird and Arthur2013). In Australia, studies have generally been small and focused on death attitudes of health professionals or patients (Collins et al., Reference Collins, McLachlan and Philip2017; Hack et al., Reference Hack, McClement and Chochinov2010; Halliday & Boughton, Reference Halliday and Boughton2008; Peters et al., Reference Peters, Cant and Payne2013; Tranter et al., Reference Tranter, Josland and Turner2016), although two studies have examined death fear/anxiety in samples of Australian university students (Bath, Reference Bath2010; Davis et al., Reference Davis, Deane and Lyons2016). Similarly, previous research worldwide has focused predominantly on death attitudes in relation to anxiety and fear, with calls for a broader focus including strengths-based death competencies and coping skills (Neimeyer et al., Reference Neimeyer, Moser and Wittkowski2003; Robbins, Reference Robbins and Neimeyer1994; Wass, Reference Wass2004). Death competence can be defined as a construct representing a range of perceived knowledge and skills for coping with death, along with attitudes and beliefs about these abilities (Robbins, Reference Robbins and Neimeyer1994). International studies have found greater death competence to be associated with older age (Haig, Reference Haig2014; Robbins, Reference Robbins1991a, Reference Robbins1991b, Reference Robbins1992, Reference Robbins and Neimeyer1994; Seo et al., Reference Seo, Kim and Kim2013), but not gender (Camarneiro & Gomes, Reference Camarneiro and Gomes2015; Haig, Reference Haig2014; Reimer, Reference Reimer2007; Robbins, Reference Robbins1991a, Reference Robbins1992; Schmidt-RioValle et al., Reference Schmidt-RioValle, Montoya-Juarez and Campos-Calderon2012; Siracusa et al., Reference Siracusa, Cruz-Quintana and Pérez-Marfil2011), and the limited findings regarding the role of religiosity and spirituality (Haig, Reference Haig2014; Seo et al., Reference Seo, Kim and Kim2013) and educational level (Seo et al., Reference Seo, Kim and Kim2013; Siracusa et al., Reference Siracusa, Cruz-Quintana and Pérez-Marfil2011) have been inconsistent. The role of socioeconomic status and occupation remains unknown (Kim & Hwang, Reference Kim and Hwang2014; McNeely, Reference McNeely1998). Research with nurses, hospital volunteers, and family caregivers suggest that greater experience in providing care for terminally ill people is associated with greater death competence (McNeely, Reference McNeely1998; Reimer, Reference Reimer2007; Robbins, Reference Robbins1992; Seo et al., Reference Seo, Kim and Kim2013). Similarly, personal experience with death/bereavement has been found related to greater death competence in volunteers, and greater health-professional self-competence in death work (Chan et al., Reference Chan, Tin and Wong KL2015; Robbins, Reference Robbins and Neimeyer1994). Overall, research on correlates of death competence in the general community is scarce worldwide, and no studies on factors associated with death competence in Australian-based samples exist (Reimer, Reference Reimer2007; Robbins, Reference Robbins1991a; Siracusa et al., Reference Siracusa, Cruz-Quintana and Pérez-Marfil2011).
The interconnection between how we feel about death and how we live has been long-stated, with the assertion that confronting mortality existentially enables us to live more meaningfully (Cozzolino & Blackie, Reference Cozzolino, Blackie, Hicks and Routledge2013; Frankl, Reference Frankl2011; Robbins, Reference Robbins1991a, Reference Robbins1991b; Taubman-Ben-Ari, Reference Taubman-Ben-Ari2011; Yalom, Reference Yalom1980). The potential importance of meaning, purpose, quality, and values in life for promoting adaptive attitudes toward death has been highlighted, but there is a lack of research in this area (Davis et al., Reference Davis, Deane and Lyons2015; Taubman-Ben-Ari, Reference Taubman-Ben-Ari2011). Evidence of the interplay between meaning in life and death awareness has been shown in experimental studies manipulating mortality salience (Cozzolino & Blackie, Reference Cozzolino, Blackie, Hicks and Routledge2013; Taubman-Ben-Ari, Reference Taubman-Ben-Ari2011), and although studies have found greater meaning in life related to lower death anxiety/fear (Durlak, Reference Durlak1972; Kim, Reference Kim2015; Meyersburg & McNally, Reference Meyersburg and McNally2011; To & Chan, Reference To and Chan2016), very little is known about the strength or direction of its relationship with death competence. Two studies found a greater sense of purpose in life related to higher death competence in university students (Robbins, Reference Robbins1991a) and palliative care professionals (Galiana et al., Reference Galiana, Oliver and Sansó2017). In Hong Kong health professionals, greater presence of meaning in life was associated with greater self-competence in death work (Chan et al., Reference Chan, Tin and Wong KL2015). There is also some support for the association between death competence and health professional quality of life (Sansó et al., Reference Sansó, Galiana and Oliver2015). Nonetheless, much remains to be discovered regarding the relationship between these life-related factors and perceived death competence and its effect on individual and community behavior and capabilities.
Online technologies provide growing opportunities for opening up dialog about death. With this in mind, we developed Dying2Learn (Rawlings et al., Reference Rawlings, Miller-Lewis and Collien2017), a Massive-Open-Online-Course (MOOC), which aimed to provide a community platform for social discussion and connection regarding death, dying, and palliative care for the Australian public. The MOOC provided a rich environment to understand contemporary views and perspectives about death and dying and for identifying factors associated with various attitudes toward death such as death competence.
The purpose of this cross-sectional study was to develop an understanding of factors associated with perceived death competence. Specifically, in people who enrolled in a MOOC on death and dying, we aimed to determine whether meaning in life, quality of life, and sociodemographic characteristics (gender, age, occupation, education, death and caring experience, religiousness, spirituality, and location) were significantly associated with perceived death competence, and which factors had the strongest predictive ability. Given the limited existing research on predictors of death competence, no directional hypotheses were made for this exploratory study.
Method
The study methodology is described in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology statement (Vandenbroucke et al., Reference Vandenbroucke, von Elm and Altman2007).
Participants and procedure
Participants for this cross-sectional study were recruited from the Dying2Learn MOOC during course enrollment in June 2016. Details about the six-week interactive online course created by CareSearch (www.caresearch.com.au) are provided elsewhere (Miller-Lewis et al., Reference Miller-Lewis, Tieman and Rawlings2018; Rawlings et al., Reference Rawlings, Miller-Lewis and Collien2017; Tieman et al., Reference Tieman, Miller-Lewis and Rawlings2018). The MOOC was freely available through the OpenLearning platform. Students completed questions regarding their sociodemographic background and death attitudes at enrollment. Overall, 1,069 of the 1,156 MOOC enrollees were invited to participate in this optional research study (late enrollees were excluded because of having immediate access to course content). MOOC enrollees were sent study information and an invitation including a unique web link to an online survey via e-mail. This study was approved by Flinders University Social and Behavioral Research Ethics Committee (Project No. 7247). Data were deidentified following data-matching from the data sources and before data analyses.
Overall, 312 (29.2%) MOOC enrollees agreed to participate in the research study, and 277 provided complete responses on the measure of death competence (25.9% response rate). Based on enrollment records, the sociodemographic profile of the participants recruited for the research study was very similar to nonparticipating MOOC enrollees, with the exception of research study participants being significantly older (t [df = 1,146] = –5.42, p < 0.0005) than those not participating (M = 53 vs. M = 48 years). Research study participants also subsequently completed a greater percentage of the MOOC (M = 45.92 [SD = 31.86] vs. M = 23.61 [SD = 29.33], t [df = 433.4] = –10.35, p < 0.0005). Table 1 provides demographic information about the 277 research study participants. At enrollment, the mean age was 52.9 years (SD = 12.4, range = 19–84 years), and the vast majority were female (95.6%). The education level was high, with 70.4% having university education, and 64.3% identified themselves as a health professional.
Table 1. Description of measures used in study and their descriptive statistics
ABS, Australian Bureau of Statistics; AQoL, Assessment of Quality of Life; CDS, Coping with Death Scale; M, mean; MLQ, Meaning in Life Questionnaire; MOOC, Massive-Open-Online-Course; NA, not available; SD, standard deviation; SEIFA, Socio-Economic Index for Areas.
* In our study, the AQoL-4D, the 4-subscale dimension scores showed low internal consistency, perhaps indicative of these subscales only being derived from 3 items each. The AQoL subscales distributions also deviated from normality by showing negative skew and kurtosis, and therefore should be interpreted with some caution. Strong positive correlations between AQoL subscales and the total utility score were indicative of potential multicollinearity (Table 2). Regression analyses confirmed the presence multicollinearity between the total score and subscale scores on the AQoL, so the subscales were eliminated from further analysis.
† One participant was missing data on age; therefore, analyses with this variable include n = 276.
‡ SEIFA scores are only generated for participants residing in Australia with an Australian postal code, n = 257 of the 277 complete cases.
Measures
Table 1 also outlines the survey measures used in the current study and their descriptive statistics (including internal consistency where applicable). Sociodemographic information was measured with questions adapted from the Australian Bureau of Statistics (2013, 2015a, 2015b, 2015c) where possible. Questions regarding religiousness and spirituality (John E. Fetzer Institute and National Institute on Aging Working Group, Reference John2003), and death and caring experiences were also queried. Death competence was measured with Bugen's Coping with Death Scale (Bugen, Reference Bugen1981; Camarneiro & Gomes, Reference Camarneiro and Gomes2015; Neimeyer et al., Reference Neimeyer, Moser and Wittkowski2003; Robbins, Reference Robbins1991a, Reference Robbins and Neimeyer1994). Health-Related Quality of Life was measured with the Assessment of Quality of Life (AQoL) 4D instrument (Assessment of Quality of Life, 2014, 2015; Hawthorne et al., Reference Hawthorne, Richardson and Osborne1999, Reference Hawthorne, Richardson and Day2000, Reference Hawthorne, Richardson and Day2001, Reference Hawthorne, Korn and Richardson2013; Osborne et al., Reference Osborne, Hawthorne and Lew2003, Reference Hawthorne, Korn and Richardson2013), and meaning in life was measured with the Meaning In Life Questionnaire (Steger, Reference Steger2010; Steger et al., Reference Steger, Frazier and Oishi2006, Reference Steger, Kawabata and Shimai2008, Reference Steger, Oishi and Kashdan2009).
Statistical approach
Given the exploratory nature of this cross-sectional study and the adequate sample size, missing data were not imputed to increase statistical power. A statistical significance level of p < 0.05 was used, and effect sizes were interpreted based on standard recommendations (Cohen, Reference Cohen1988; Fergusson, Reference Fergusson2009). Descriptive statistics were produced for each variable. Tests for normality revealed scores on the death-competence scale were slightly negatively skewed, but did not deviate greatly from normality (Pallant, Reference Pallant2010) (Kolmogorov-Smirnov test = 0.077). The study variables generally did not show substantial deviation from normality based on skewness, kurtosis, or histogram examination. The direction and strength of bivariate associations between study variables were examined using Pearson's correlation coefficients, with focus placed on associations with death-competence scores as the outcome of interest. Following this, multiple linear regressions predicting scores on death competence from the group of predictor variables were conducted, enabling the identification of the variable with the strongest unique association with death-competence scores, as well as total variance in death-competence explained by the combined group of predictor variables. Data analyses were conducted in IBM-SPSS, version 23.
Results
Descriptive statistics
Descriptive statistics describing the 277 study participants are provided in the final two columns of Table 1. The majority (93.1%) of participants resided in Australia, and on average lived in areas slightly above the socioeconomic national average. Participants were more likely to identify themselves as being spiritual rather than religious, and on average considered themselves to be in good to very good health. Overall, 64% had experienced the death of a loved one in the past 5 years. Seventy percent had cared for someone terminally ill, likely indicative of the considerable number of health professionals in the sample. Scores on the presence and search for meaning in life were similar to those found in people aged 65+ (Steger et al., Reference Steger, Oishi and Kashdan2009), and suggested our participants feel satisfied that they've grasped what makes life meaningful, why they're here, and what they want to do with life (Steger, Reference Steger2010). Scores on quality of life were indicative of being in very good health and similar to those found in the general Australian population (Hawthorne et al., Reference Hawthorne, Korn and Richardson2013). The mean score on death competence was 155.0, higher than previously found in school parents (M = 125) (Siracusa et al., Reference Siracusa, Cruz-Quintana and Pérez-Marfil2011), nonpatient care volunteer groups (M = 151.5) and palliative care volunteers pretraining (M = 143.5) (Robbins, Reference Robbins and Neimeyer1994), and closer to means found in hospice volunteers in training (M = 154.8) (Claxton-Oldfield et al., Reference Claxton-Oldfield, Crain and Claxton-Oldfield2007).
Bivariate analyses
A correlation matrix of bivariate associations between variables is shown in Table 2. Presence of meaning in life held the strongest bivariate association with death competence, with higher levels of presence of meaning in life significantly positively associated with greater death-competence scores and indicating a moderate effect size of practical significance. Conversely, higher levels of searching for meaning in life held a small, significant, negative association with lower death-competence scores. Greater age, religiousness, spirituality, total quality of life (and especially its Relationships subscale) held significant positive associations with greater death-competence scores, indicative of small effect sizes. Better self-assessed health was also associated with greater death-competence scores. Having experience of death was significantly associated with higher death-competence scores, as was carer experience. There was no significant relationship with death-competence scores for gender, residing in Australia, birthplace, socioeconomic-disadvantage, education level, or health professional status. Significant intercorrelations between the study predictor variables were not unexpected. Particularly of note was the significant moderate positive association between the presence of meaning in life and the AQoL relationships subscale and total score, and negative associations between the AQoL scores and self-assessed health.
Table 2. Bivariate correlations between variables (n = 277)
To allow comparison of magnitude across all variables, associations were analyzed using Pearson correlations. Associations between death competence and the dichotomous demographic variables were also analyzed using Independent samples t-test (not shown), and the conclusions were the same.
AQoL, Assessment of Quality of Life; SEIFA, Socio-Economic Index for Areas.
* p < .05 two-tailed significance.
** p < .01 two-tailed significance.
*** p < .001 two-tailed significance.
§ On self-assessed health status, higher scores indicate poorer health.
|| One participant was missing data on age, therefore analyses with this variable include n = 276.
¶ SEIFA scores are only generated for participants residing in Australia with an Australian postal code, n = 257 of the 277 complete cases.
Multivariable analyses
A multiple linear regression was used to assess the ability of the group of sociodemographic and meaning-in-life and quality-of-life variables to predict scores on death competence (Table 3). In the final model, a total of 34.4% of the variance in death-competence scores was accounted for by the group of variables (F [14,261] = 9.76, p < 0.005). After adjusting for all other variables in the model, greater presence of meaning in life, total quality of life, age, death experience, and carer experience were each statistically significant unique predictors of death-competence scores, with generally small effect sizes. The presence of meaning in life obtained the highest beta value (β = 0.350, p < 0.0005), indicating a moderate effect size. The remaining variables did not hold unique associations with death competence after controlling for the effects of all other variables in the model. The bivariate associations of religiousness and spirituality with death competence were no longer significant in the multivariable model.
Table 3. Simultaneous multiple linear regressions predicting death competence
CI95%, 95% confidence interval; AQoL, Assessment of Quality of Life; SEIFA, Socio-Economic Index for Areas.
* Collinearity diagnostics indicated the presence of multicollinearity between AQoL subscales and AQoL Total Utility scores; therefore, the reported multivariable analysis excluded the AQoL subscales.
† Of the 277 cases with data on death competence, n = 1 was missing data on age; therefore, the multiple regression had a total n = 276.
‡ On self-assessed health status, higher scores indicate poorer health.
§ Valid data on the SEIFA disadvantage index only exist for participants who reside in Australia (n = 256). We tested a model including n = 256 Australian residents only to assess the association of SEIFA disadvantage to death competence.
A multiple linear regression with Australian residents only allowed the assessment of the unique association of socioeconomic disadvantage to death competence, and an opportunity to investigate potential differences between people in and outside of Australia on predictors of death competence. This model accounted for 33.8% of the variance in death competence (F [14,241] = 8.80, p < 0.0005). The significant unique predictors for people residing in Australia were the same as those found for all participants, and there was no significant association between socioeconomic-disadvantage and death competence.
Discussion
Main findings
The findings of this exploratory study have highlighted an important association between peoples’ feelings about life and about death. A sense of a presence of meaning in life was most strongly related to higher levels of competence in coping with death, even more than having experience with the death of a loved one or caregiver experience. The presence of meaning in life was more important than searching for meaning in life in explaining death-competence scores; with greater searching for meaning in life related to lower death-competence scores in bivariate analyses but not maintained in multivariable analyses. When people feel their life holds meaning, do they feel more comfortable with the idea of their mortality and more competent to deal with death? Or is mortality awareness a necessary precursor for facilitating the pursuit of meaning in life? Can helping people build the presence of meaning in one's life lead to greater death competence? Indeed, the meaning of life and death may be interrelated concepts that shape each other (Cozzolino & Blackie, Reference Cozzolino, Blackie, Hicks and Routledge2013; Davis et al., Reference Davis, Deane and Lyons2015; Robbins, Reference Robbins1991a; Taubman-Ben-Ari, Reference Taubman-Ben-Ari2011). The potential transactional interplay among mortality awareness, meaning in life, and death competence requires longitudinal research to untangle. One possibility is that awareness of our mortality is required for a true sense of meaning in life to develop (Cozzolino & Blackie, Reference Cozzolino, Blackie, Hicks and Routledge2013), which can in turn lead to the development of competence coping with death.
Our correlational findings within the Australian setting were generally consistent with the limited amount of preexisting research on death competence (Camarneiro & Gomes, Reference Camarneiro and Gomes2015; Galiana et al., Reference Galiana, Oliver and Sansó2017; Haig, Reference Haig2014; McNeely, Reference McNeely1998; Reimer, Reference Reimer2007; Robbins, Reference Robbins1991a, Reference Robbins1991b, Reference Robbins1992, 1994; Schmidt-RioValle et al., Reference Schmidt-RioValle, Montoya-Juarez and Campos-Calderon2012; Seo et al., Reference Seo, Kim and Kim2013; Siracusa et al., Reference Siracusa, Cruz-Quintana and Pérez-Marfil2011), indicating similarities between people residing in Australia and other countries. We found life-related variables more strongly associated with death-competence than sociodemographic variables. Older age (and perhaps the experience that comes with it) was the only demographic variable independently related to greater death competence. The bivariate associations with religiousness and spirituality disappeared once other factors were adjusted for, which may indicate the presence of a mediated effect that could explain some of the mixed findings in previous studies (Haig, Reference Haig2014; Seo et al., Reference Seo, Kim and Kim2013). In addition, we have demonstrated an association between greater health-related quality of life and greater perceived death competence. Results were particularly indicative of a role of quality-of-relationships within quality of life, which needs further investigation with psychometrically strong social support measures. The interplay between these variables as people age and their health deteriorates would be an interesting focus for future longitudinal research.
In our sample, we found that bereavement experience and experience caring for the dying was associated with greater death competence, but there were no differences on death competence between health professionals and the general public. The self-selected nature of the sample already interested in the topic of death (as demonstrated by enrollment in a death course) may have concealed differences in death competence normally expected between health professionals and the general public. Some health professionals may have sought out the course because of limited exposure to caring for the dying or for palliative care learning opportunities, and some participants who didn't identify as health professionals may have had considerable experience in caring for the dying, either within their personal life or through working as death doulas or funeral celebrants, and therefore potentially not truly representative of the general community. Understanding death competence and its explanatory factors in different types of health professionals and community groups with varying levels and types of caring experience (both clinical and personal) is an important avenue for future research (Chan et al., Reference Chan, Tin and Wong KL2015). Investigating the interplay between meaning in life, types of caring experiences, and death competence in these different occupational and community groups is also required. It may be that certain life tasks, such as altruistic service (Cozzolino & Blackie, Reference Cozzolino, Blackie, Hicks and Routledge2013; Sinclair et al., Reference Sinclair, Raffin and Pereira2006) or caring for a dying person you know personally (i.e., personal caring rather than professional caring), can induce the development of meaning in life, which in turn can lead to greater death competence.
Clinical implications
As the population ages, knowledge of factors associated with poorer death coping can be useful for identifying who might be a greater risk of distress when they or a loved one is facing death, and might prove useful additions to bereavement risk assessments. Being able to identify people who may struggle more in preparing for death or its aftermath can help health professionals target early intervention efforts toward those likely to be in greatest need. Understanding factors associated with greater death competence in health professionals is also important given that some have regular death exposure and for whom struggles in death competence may lead to burnout and compassion fatigue (Chan et al., Reference Chan, Tin and Wong KL2015; Sansó et al., Reference Sansó, Galiana and Oliver2015). Indeed, one study of end-of-life care nurses found greater meaning in life was associated with less emotional exhaustion and a greater sense of accomplishment (Gama, Barbosa, & Vieira, Reference Gama, Barbosa and Vieira2014). Screening health professionals or volunteers on meaning in life and death competence might help identify those who may be better suited to palliative-related roles or for whom additional training and support would be beneficial (Claxton-Oldfield & Banzen, Reference Claxton-Oldfield and Banzen2010; Gama et al., Reference Gama, Barbosa and Vieira2014). Interventions focused on increasing meaning in life have shown multiple psychosocial benefits for patients with advanced disease (Guerrero-Torrelles et al., Reference Guerrero-Torrelles, Monforte-Royo and Rodriguez-Prat2017; Vos et al., Reference Vos, Craig and Cooper2015), community-dwelling older adults (Chippendale & Boltz, Reference Chippendale and Boltz2015) as well as promise with palliative care staff and volunteers (Guerrero-Torrelles et al., Reference Guerrero-Torrelles, Monforte-Royo and Rodriguez-Prat2017; Sinclair et al., Reference Sinclair, Raffin and Pereira2006; Wasner et al., Reference Wasner, Longaker and Fegg2005). There is also suggestions that these interventions may be especially useful for professionals working in palliative care who exhibit lower meaning in life (Fillion et al., Reference Fillion, Duval and Dumont2009).
Limitations and future directions
This exploratory study has provided a beginning basis for understanding correlates of death competence, but must be interpreted with the following limitations in mind. First and foremost, the cross-sectional nature of this correlational study means that we cannot determine possible causal sequences or directions of effects on death competence. Without longitudinal data, we are unable to demonstrate temporal precedence of our predictor variables or examine potential transactional relationships among variables. Future longitudinal research following people over time is important to elucidate whether high presence of meaning in life, quality of life, and death and caregiving experiences temporally precede subsequent death competence. Second, the study sample comprised self-selected individuals who chose to enroll in a course about death, many of whom were health professionals and/or had experience caring for the dying or bereaved. This preexisting interest in the topic of death possibly skewed death-competence scores toward more adaptive levels than what might be seen in a representative sample of the general public. Third, a deeper investigation of death-competence levels held by people with different health professional and carer roles is warranted. To fully articulate the community versus health professional perspectives, clarification on various occupational roles and experience caring within a clinical or personal context is required. Future research could determine if health professionals working in palliative care have higher levels of death competence than other health professionals or people involved in death-related work (e.g., death doulas, funeral industry, volunteers). Finally, using the Coping with Death Scale (Bugen, Reference Bugen1981; Robbins, Reference Robbins and Neimeyer1994) allowed for a larger sample size, but relied on self-reported subjective perceptions of “likely” behaviors rather than objective behavioral observation, which may have provided a different picture.
Conclusions
In conclusion, the findings of this study suggest that our attitudes toward life and death are inextricably linked. What role adaptive attitudes toward meaning in life can potentially have on bolstering subsequent adaptive coping and competence regarding death needs further exploration in longitudinal studies including both health professionals and the general community. Future research is also needed to extrapolate the influence of certain life tasks or experiences, such as caring for a dying person, on the development of both meaning in life and competence coping with death.
Author ORCIDs
Lauren Miller-Lewis 0000-0001-6013-130X, Jennifer Tieman 0000-0002-2611-1900, Christine Sanderson 0000-0001-5423-5778
Acknowledgments
The Dying2Learn MOOC is part of CareSearch, which is funded by the Australian Government Department of Health. The funding body had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.
The authors thank the study participants and the many people who contributed their time and expertise to the CareSearch project, including members of the National Advisory Group. The authors also thank OpenLearning for their contributions, including providing the MOOC platform and assisting with data extraction.
