Most individuals in the United States (U.S.) who are seriously ill say they want to die at home 1. Yet, research has consistently shown that the majority of older U.S. adults 234, especially those with chronic diseases 235, die in the hospital. This is the case in other countries, too, according to studies conducted in the 1990’s 67.

More recent research has found that high rates of hospital deaths among older adults and those with chronic illnesses exists in most of the countries in which it has been studied, including Australia 8, Belgium 91011, Canada 12, Korea 13, Singapore 14, the U.K. 11, and the U.S. 15. Exceptions to this observation are The Netherlands 11 and Sweden 16.

A number of factors have been examined that contribute to the high rates of hospital deaths. U.S. studies consistently have found that hospital deaths are related to hospital bed availability and vary across states and regions of the country, with hospital death rates being lowest in the Western states 345. This association between hospital bed availability and rate of hospital deaths has been found in Belgium and the United Kingdom, as well 11.

Another geographically related variable that is associated with rates of hospital deaths is population density. Research in Belgium 10, Germany 17, and Taiwan 18 has shown that people living in urban areas are more likely to die in the hospital than those living in rural areas, although findings from other studies are inconsistent 27. The observed relationship between hospital deaths and population density, as well as the regional differences, may be partly attributable to the number of available hospital beds per capita, with areas that have a greater number of hospital beds per person having higher rates of hospital deaths 319.

Geographically based (e.g. census tracks) measures of socio-economic status have also been found to be associated with hospital death rates. For example, Belgian 9 and American researchers 19 have reported that individuals from poorer neighborhoods have higher rates of hospital deaths than those from wealthier neighborhoods.

Recent studies have begun to explore the degree to which different healthcare services are associated with different end-of-life outcomes. For instance, an Australian study reported that community nursing services, which included emotional support for patients and families, were associated with lower hospital death rates 20. Other Australian studies reported that community-based palliative care services were associated with lower rates of hospital deaths 821. The lower rates of hospital deaths were associated with higher rates of institutional hospice deaths in one study 21 and higher rates of home deaths in the other 8. A Belgian study also found that community-based palliative care services were associated with lower rates of hospital deaths and higher rates of home deaths 22.

U.S. research suggests that hospital-based palliative care programs can affect hospice enrollment and hospital deaths. One study, for example, reported that inpatient palliative care produced higher rates of referrals to home-based hospice programs than usual acute care 23, and another reported that palliative care teams successfully diverted patients into inpatient and home-based hospice programs 24. In a randomized control trial, patients who received palliative care services had fewer admissions to intensive care units (ICU) and longer hospice stays 25. Other research further indicates that palliative care teams may help to reduce hospital deaths, especially ICU deaths 26.

Chaplains have long been considered to be an integral part of the palliative care team 27 and they have routinely provided emotional and spiritual support to patients and family members even when palliative care services were not available 2829303132. However, only two studies to date have examined the effectiveness of chaplaincy care, one of which found chaplaincy care reduced patient anxiety 33 and the other of which reported that it helped patients contend with spiritual distress 34.

Although no previous research has examined the relationship between chaplaincy care and end-of-life outcomes, there is reason to believe that spiritual care, like that provided by chaplains, may affect end-of-life decision-making. Recent research has found that patients who say their spiritual needs were met by hospital staff were more likely to receive hospice care and less likely to receive aggressive treatment 3536.

We have noticed that one of the things that chaplains do for patients is help them understand and feel comfortable with their core values and how these can be synthetic with their health care goals. Since people generally prefer to die at home, chaplains may help patients and families to resolve issues about end of life care, so that patients are less likely to die in the hospital.

Table 1 presents the univariate statistics for the independent, dependent and control variables. Of the 3,585 hospitals in the sample, nearly 2,457 hospitals provided chaplaincy services (68.5 %), and 1,092 hospitals provided palliative care services (30.5 %). Just under 18 % were located in the 13 western states. The average rates of hospital deaths and hospice enrollment were approximately, 35 % and 29 %, respectively.

On a scale of 0-6, the mean population density of 2.2 indicates that the average hospital in the sample was located in a geographical area with a population of over 250,000 residents. The average size of hospitals in the study was between 100 and 200 beds, based on the AHA bed size code.

Table 2 presents the results of the OLS regression models predicting rates of hospital deaths and hospice enrollment. The values given in the table are standardized beta’s (β’s). which show the unique association of chaplaincy services and each of the control variables with the two dependent variables: hospital deaths and hospice enrollment.

* p < .05 ** p < .01 *** p < .001 Values in the table are standardized beta’s.

As hypothesized, chaplaincy services were associated with significantly lower rates of hospital deaths (β = -.04, p < .01) compared to hospitals that did not provide chaplaincy services, controlling for region of the country, population density, hospital beds, socio-economic status, palliative care and type of hospital (see Table 2). As expected, hospitals located in western states and those with palliative care teams had significantly lower rates of hospital deaths.

Number of hospital beds was positively and significantly related to hospital deaths, but population was not. However, these two variables were positively correlated, and the association between hospital deaths and population density (β = 0.14) was statistically significant (p < .001) when the model was re-run without number of beds. This supports previous findings that population density is positively related to hospital deaths and shows that the inclusion of number of beds in the model captured a substantial portion of the variation in hospital deaths that would otherwise have been attributable to population density.

As hypothesized, chaplaincy services also were associated with significantly higher rates of hospice enrollment (β = .06, p < .001). Population density and number of beds were positively related to hospice enrollment, while socio-economic status was negatively related to enrollment.

The findings support both hypotheses: that chaplaincy services are significantly associated with (1) lower rates of hospital deaths and (2) higher rates of hospice enrollment. Although these associations are relatively small, they can have profound consequences for patients and families. One would not expect appreciably larger net effects of specific services, since many unknown variables also contribute to the outcomes being examined. Nevertheless, the fact that the regression model for rates of hospital deaths replicated the effects of variables known to be related to hospital death rates from previous research lends credence to the generalizability of the findings. While palliative care exhibited the expected negative association with hospital deaths, it is not apparent why it did not have a significant positive association with hospice enrollment.

Chaplains often help patients and family members to feel comfortable with their core values and to see how these can be congruent with their health care goals. Since most people prefer to die at home 14243, discussions of values and goals with chaplains might partially account for the observed association of chaplaincy services with decreased hospital deaths and increased hospice enrollment.

The present results are consistent with recent research that patients who said their spiritual needs were met by hospital staff were more likely to receive hospice care and less likely to receive aggressive treatment 3536. Related research has found that families of dying patients who felt supported 44, and were able to discuss spiritual issues when making end-of-life decisions were more satisfied with those decisions 44.

Other research has found that meeting the spiritual needs of patients is associated with higher levels of patient 45 and family 4647 satisfaction with the quality of overall care. Meeting spiritual needs is particularly important at the end of life in order to achieve spiritual and psychological healing when physical healing is no longer possible, and palliation is the goal 4849. At this time of life, it is essential to align treatment with the patient’s goals 50, which are often influenced by spirituality and religion 51.

A recent study of patients in a palliative care unit with end-stage cancer found that 88 % of patients expressed the desire to work with a chaplain 52. Roughly half of the patients indicated they would like the chaplain to provide a sense of “presence,” listen to them, visit with them, or accompany them on their journey. The more religious patients also desired various spiritual interventions from the hospital chaplain.

Surveys of hospital administrators in the U.S. have found that professional chaplains are expected to play a number of roles 5354. In addition to providing spiritual and emotional support to patients and families, generally, chaplains are also expected to help patients and family members address end-of-life issues and deal with difficult decisions. Related U.S. research indicates that hospital staff refer patients and families to chaplains for such reasons 32, and that chaplains spend a considerable amount of time dealing with these kinds of issues 5556. Moreover, Australian studies have found that the majority of hospital chaplains helped patients and families make treatment decisions 57, and that many chaplains assisted families and staff dealing with pain control and life support issues 585960.

This study has a number of limitations, including the fact that the AHA survey only indicated whether or not a hospital provided chaplaincy services. It did not provide information about the specific nature, scope, and quality of the services. Another limitation is that the patient outcome data are aggregated across a 5-year period (2001-2005), and there is evidence that the provision of chaplaincy services increased somewhat during this time 37. This introduces some degree of error to the models, which are based on the hospital services that existed in fiscal-year 2005 (2004-2005).

Although the analyses controlled for a number of variables that have been shown to be associated with end-of-life outcomes, the presence of chaplaincy services could still be a proxy for some other variable or variables. It could, for example, be an indication that a hospital is committed to providing patient-centered and family-focused care in which all members of the team are dedicated to listening to the patient and aligning care with patient and family wishes. Since the evidence indicates that this stance generally leads to less use of hospital resources at the end-of –life, these hospitals might have fewer hospital deaths even if they did not have chaplains. Nonetheless, this type of care, whether provided by a chaplain or another discipline, appears to have an impact.

Its limitations notwithstanding, we believe the present study provides support for the possible role of chaplains in reducing hospital deaths and increasing hospice enrollment. This could be attributable to the fact that chaplains assist patients and families in making decisions about care at the end-of-life and aligning their values and wishes with actual treatment plans.

The findings suggest that chaplaincy services may play a role in decreasing hospital deaths and increasing hospice enrollment. This may be attributable to chaplains’ assistance to patients and families in making decisions about care at the end-of-life, perhaps by aligning their values and wishes with actual treatment plans. Additional research is warranted.

The authors declare that they have no competing interests.

KJF designed the study with KG, NRS and MC, all of whom wrote the initial draft. KJF, LLE and GFH were the primary authors. KJF conducted the statistical analyses. All the authors reviewed drafts and approved the final content.