Research in Gerontological Nursing

Empirical Research 

Modes of Decision Making Used by Nursing Home Residents and Their Families When Confronted With Potential Hospital Readmission

Ruth M. Tappen, EdD, RN, FAAN; Deborah Elkins, DNP, MBA, FNP-BC; Sarah Worch, MS, MA; MaryAnn Weglinski, RN


The purpose of the current study was to characterize the decision-making processes used by nursing home (NH) residents and their families when confronted with an acute change in condition and the choice of transfer to the hospital or treatment in the NH. Using cognitive task analysis, 96 residents and 75 family members from 19 NHs were asked how they would make this choice. Fifty-one residents (53%) and 61 family members (81%) used a deliberative mode characterized by seeking information and weighing risks and benefits. Ten residents (10%) and five family members (7%) used a predominantly emotion-based mode characterized by references to feelings and prior experiences in these facilities. Thirty-six residents (38%) and nine family members (12%) delegated the decision to a family member or provider. Age and resident/family status were associated with mode used; transfer choice, gender, religion, education, and ethnic group were not. Although classic theories of information processing posit two modes of decision making, deliberative and affective, the current data suggest a third mode, that of delegating the decision to trusted others, particularly family members and providers.

[Res Gerontol Nurs. 2016; 9(6):288–299.]


The purpose of the current study was to characterize the decision-making processes used by nursing home (NH) residents and their families when confronted with an acute change in condition and the choice of transfer to the hospital or treatment in the NH. Using cognitive task analysis, 96 residents and 75 family members from 19 NHs were asked how they would make this choice. Fifty-one residents (53%) and 61 family members (81%) used a deliberative mode characterized by seeking information and weighing risks and benefits. Ten residents (10%) and five family members (7%) used a predominantly emotion-based mode characterized by references to feelings and prior experiences in these facilities. Thirty-six residents (38%) and nine family members (12%) delegated the decision to a family member or provider. Age and resident/family status were associated with mode used; transfer choice, gender, religion, education, and ethnic group were not. Although classic theories of information processing posit two modes of decision making, deliberative and affective, the current data suggest a third mode, that of delegating the decision to trusted others, particularly family members and providers.

[Res Gerontol Nurs. 2016; 9(6):288–299.]

Hospitalization of nursing home (NH) residents can have serious repercussions, including increased risk of falls, delirium, nosocomial infections, confusion, and pressure ulcers (Creditor, 1993; Friedman, Mendelson, Bingham, & McCann, 2008; Leff et al., 2005; Ouslander & Maslow, 2012; Thomas & Brennan, 2000). Potentially preventable hospitalizations (PPHs) also increase the financial burden on an already-stressed health care system. Although regulatory efforts are underway to discourage preventable hospital transfers and the associated costs, little attention has been directed to the role of resident and family insistence on transfer (Brownell, Wang, Smith, Stephens, & Hsia, 2014). An early study found that residents and families together accounted for 16% of transfer decisions (High & Rowles, 1995). More recently, resident and/or family insistence on transfer to acute care was identified by NH staff as a particularly intractable source of preventable hospital readmission, accounting for 15% to 20% of PPHs (Lamb, Tappen, Diaz, Herndon, & Ouslander, 2011). Evidence suggests that providers acquiesce to resident and/or family insistence on transfer even when hospitalization may be unwarranted (Helton, van der Steen, Daaleman, Gamble, & Ribbe, 2006).


Although some health care decision-making processes have been well described (Reyna, 2004), a gap exists in the literature regarding an understanding of how older adults and their families make decisions concerning acute care transfers from NH settings. Understanding this process may give health care providers a clearer picture of the patient–family decision-making process in this context. To adequately address family and resident insistence when a hospital readmission may be preventable, how the decision is made needs to be understood. The current study addresses this gap by investigating the modes NH residents and their families actually used or reported they would use from transcripts of 171 interviews of NH residents and family members.

Modes of Thinking in Decision Making

Epstein's (1994) classic theory of information processing posits two parallel modes: a deliberative one that is rational and conscious, the other an affective mode that is experientially based and emotionally driven. The affective mode generates thoughts and feelings relatively spontaneously and effortlessly, whereas deliberative processing is reason-based, more effortful, and comparatively slow (Peters, Hess, Vastfjall, & Auman, 2007). Although affective and deliberative processes have been well described (Peters et al., 2007; Sloman, 1996; Slovic, Peters, Finucane, & MacGregor, 2005), they have yet to be thoroughly examined in older adults' health care decision making.

Age differences have been found in use of these modes of decision making (Peters et al., 2007). Older adults appear to experience some age-related decreases in use of deliberative processing and rely more on affective processing (Carstensen & Turk-Charles, 1994).

Trust and Delegation

An additional factor that may contribute to mode of health care decision making is the degree of trust in care providers (Trachtenberg, Dugan, & Hall, 2005). Trust may be conceptualized as a willingness to depend on another to fulfill a need in a vulnerable situation (Bell & Duffy, 2009; Hupcey, Penrod, Morse, & Mitcham, 2001). Trust is considered a primary factor in patients' preferred level of involvement in personal health care decision making (Trachtenberg et al., 2005). Particularized trust refers to trust toward specific groups (e.g., family members, personal acquaintances), whereas generalized trust refers to a broader sense of overall trust in others (Li & Fung, 2013).

Studies of age-related differences in trust suggest that levels of trust increase with age (Li & Fung, 2013). In a sample of 2,197 patients from the Medical Outcomes Study, analysis of responses to a query as to whether they preferred to leave decisions about their care to their provider, younger participants ages 35 to 44 preferred a more active role than those 75 or older (Arora & McHorney, 2000). Similarly, in a representative sample of 2,765 adults, the preference for an active role in health-related decision making increased up to age 45 but then declined, shifting toward preference for provider decisions by age 85. Those reporting poor health preferred more provider direction (Levinson, Kao, Kuby, & Thisted, 2005). In a large Canadian sample of clinic outpatients, it was found that older, less-educated individuals were more likely to prefer a more passive, as opposed to active or shared, role in decision making as well (Deber, Kraetschmer, Urowitz, & Sharpe, 2007).

Older adults' tendency toward increased trust has several implications for decision making in health care and, more specifically, for the decision whether to transfer from the NH to acute care or remain in the NH when an acute change in condition occurs. Older adults may be more susceptible to making acute care transfer decisions based primarily on trust, deliberating less about the potential outcomes and risks. They may also look to others, including friends and family members, who may not have pertinent knowledge to make decisions on their behalf. These tendencies may be reinforced by NH staff, particularly as the resident's physical health and/or cognitive function decline. For example, in case analyses from a 3-year anthropological study of decision making in four NHs, Shawler, Rowles, and High (2001) noted progressively diminishing support of NH staff for resident decision making and increasing resident decisional dependence, which they associated with a decline in health.

Cognitive Task Analysis

Cognitive task analysis is a set of methods for studying how individuals think, reason, and make decisions in a variety of situations (Crandall, Klein, & Hoffman, 2006). One approach, the Critical Decision Method (CDM), involves using a series of probing questions to elicit the participant's decision making during a specified incident (Hoffman, Crandall, & Shadbolt, 1998). The current study used CDM-based interviews to elicit NH resident and family members' decision-making processes when confronted with a real or hypothetical possibility of transfer to acute care when a change in the resident's condition occurred. A proposed descriptive model that emerged from the qualitative analysis of this data is also presented.


An integrated mixed-methods design was used. Qualitative data regarding decision-making mode using CDM was obtained through semi-structured interviews of NH residents and family members. Quantitative sociodemographic data were obtained from each participant and compared to decision mode used.


The interviews were conducted in 19 NHs in South Florida. Six were not-for-profit and 13 were for-profit. The number of beds ranged from 60 to 325 (mean = 156 beds, SD = 59.06 beds). The mean number of reported subacute patients per NH was 46 (SD = 25.89 patients; range = 15 to 121 patients).

Sample Development

In total, 96 NH residents and 75 family members were interviewed. NH staff approached potentially eligible residents and family members to ascertain interest before they were approached by members of the research team to explain the study and invite them to participate. Inclusion criteria were adult NH resident or adult family member of a current resident, and ability to answer the questions and consent to participate. Bilingual interviewers were made available to avoid exclusion based on language. Exclusion criteria were inability to participate in an interview due to intubation, lowered level of consciousness or dementia, and those who declined to participate. Residents were screened with the Mini-Cog (Borson, Scanlan, Brush, Vitaliano, & Dokmak, 2000) and excluded if their score indicated dementia because deficits in communication and episodic memory associated with dementia would make it difficult for them to recall and respond to the complex, often abstract, probing questions of CDM (Adlam, Patterson, & Hodges, 2009; Maki, Amari, Yamaguchi, Nakaaki, & Yamaguchi, 2012). This exclusion occurred infrequently because NH staff approached those residents they believed would be eligible prior to screening and consent. Family members were those individuals identified either as a close relative or significant other of a current NH resident. Thirty-six family members were adult children, 26 were spouses, and 13 fell under the category of other (i.e., sibling, grandchild, or parent). All participants provided written consent according to procedures approved by the university Institutional Review Board prior to interview.

Data Collection

A frequently used approach to CDM is the use of semi-structured interviews during which individuals are asked to relate their personal experience or discuss a significant event, in this case the hospital transfer decision (Crandall et al., 2006; Dionne-Odom, Willis, Bakitas, Crandall, & Grace, 2015; Hoffman et al., 1998). The questions or probes are sequenced to guide interviewees through the process that they used to make the decision; however, some respondents nevertheless respond in a more circular approach in their description or begin with their ultimate decision and then work backward to describe how they arrived at the decision, requiring some flexibility on the part of the interviewer. The probing questions are considered necessary to elicit this information as the respondents are typically “working at the edge of what people are able to articulate clearly” (Crandall et al., 2006, p. 196) when reporting their decision-making processes. The CDM-based questions used in the current study for a hypothetical change in condition were as follows:

  • Options: If you became ill, for example, developed pneumonia or a urinary tract infection, would it be your preference to be treated here in the NH or would you want to be sent to the hospital for treatment?
  • Experience: Have you given any thought to this question before today?
  • Goals and Priorities: What would be most important to you in making this decision?
  • Information Seeking: What information would you need to make a decision?
  • Guidance: With whom would you discuss the decision?

Length of the interviews ranged from 15 to 45 minutes. The interviews were audiorecorded if the participant agreed to being recorded. If the participant did not, the interviewer made notes during the interview. One hundred forty-four (84%) interviews were recorded. The interviews were transcribed and the transcriptions reviewed by the interviewers prior to analysis to ensure accuracy. Interviews conducted in Spanish were first translated into English by native speakers and the translation reviewed for accuracy by a second native speaker, both members of the investigative team. Sociodemographic information, including age, gender, religion, ethnic group, and years of education, was also obtained from all participants.

Data Analysis

Once reviewed for accuracy, the transcribed interviews were read and reread by the authors, all of whom had conducted a portion of the interviews. As this was done, several distinctive modes of decision making used by the responding participants emerged. A third reading of each transcript was performed by two coders to identify the predominant mode of decision making used by each participant. All coding of the predominant decision-making mode used was reviewed by the team. Instances of disagreement between two coders were identified and resolved by consensus.

Rigor of the qualitative analysis was addressed by maintenance of a detailed audit trail, use of a larger-than-usual sample for qualitative analysis, verification by those who conducted the interviews, and negative case analysis, which led to identification of a third predominant mode (i.e., delegation) and addition of a sub-theme (i.e., emotional overtones) that were found in some of the responses from those who used the deliberative or delegation modes (Tappen, 2016a).

The decision-making mode used by each participant was entered into a SAS 9.3 database as a categorical variable. The modes used were compared with characteristics of the sample using analysis of variance and logistic regression. The modes used by those who had experienced an actual rehospitalization were compared with the modes used by those who were responding to a hypothetical situation as well as to participants' expressed preference to remain in the NH or be transferred to acute care if an acute change in condition occurred using the chi-square statistic.



Of the 96 residents interviewed, 75% were long-stay and 25% were short-stay (i.e., subacute) patients. Sixty-nine (72%) residents were women and 27 (28%) were men. Average patient age was 77.39 (SD = 11.84 years, range = 47 to 99 years), and average years of education was 12.89 (SD = 2.94 years, range = 5 to 20 years). Based on self-report, 75% were European American, 14% were African American or Afro-Caribbean, 9% were Hispanic, and 2% were Other. Family members were younger, with a mean age of 65.56 (SD = 14.63 years, range = 21 to 93 years) and had more years of education (mean = 14.5 years, SD = 2.56 years, range = 7 to 21 years). Sixty-nine percent of family members were women and 31% were men. Sixty-four percent of family members were European American, 21% were African American or Afro-Caribbean, 11% were Hispanic, and 4% were Other.

Decision-Making Modes

The most commonly used mode of decision making found was deliberation. The deliberative mode involved obtaining information about the resident's condition, what treatment was available in the NH versus in the hospital, what care providers recommended, what other family members thought, and weighing the pros and cons of the choices. Not all of these sources of information were mentioned by each participant but they were the most common sources. Example quotes included:

My prognosis would help me make the decision.

I would like to find out the pros and cons of both…. Well, I think the severity of the illness has a lot to do with it. In fact, I say that I would have to take into consideration just how well they could treat the illness at the NH.

I need to know what her condition is and if her condition is too severe to be treated here.

The diagnosis, what tests were done to come up with the diagnosis? What else? The treatment of it, what would the treatment consist of, an antibiotic, IV [intravenous] antibiotic.... Umm…I should probably go with the level of care that is needed. If it is like what I consider advanced level of medical training, I would want her treated in the hospital…. I just think that letting the family know, like making sure they are aware of the tests that are done, the results coming in and what is going to be done to their family member definitely helps the decision.

Fifty-one residents (53%) and 61 family members (81%) used the deliberative mode in their decision. The difference in the proportion of residents versus family members using the deliberative mode was significant (χ2 (1, N = 171) = 14.8256, p = 0.0001).

Also evident, although less commonly used, was the emotion-based response of some participants. Those whose decision mode was predominantly emotion-based typically described either very positive or very negative feelings about the NH, hospital, or both, as illustrated by the following statements:

Well, everybody knows that the hospitals are getting a bad reputation. Because they say you are going to come out dead. Unfortunately, I know a couple of people that this happened to. I have seen examples like that and I am really afraid of the hospital.

I would rather stay here [NH] under any circumstances, 100%. I feel better here, no other information would persuade me.

I love the hospital.

I think she should be in a hospital. She gets more care around the clock. I don't trust [these] places. I think I was freaked out when she had pneumonia in the NH.

I have to be honest. I think that the NH here has been a good experience. People are wonderful. They really are. I think she is happy here, too.

Only 10 residents (10%) and five family members (7%) used a predominantly emotional mode of decision making. The difference between the two groups was not significant (χ2 (1, N = 171) = 0.7399, p = 0.39).

A third mode was to delegate the decision to a trusted family member and/or care provider, in effect to place one's trust in the decision of others, as illustrated by the following statements:

Oh, I don't decide myself, they [the NH staff] do…they pick it up right away. They know how sick I was the first time…. Well, whatever is best for me, if they feel that going to the hospital will help my condition.

Actually, my daughter makes the decisions. She handles the finances and all. I would not be involved.

My older son…they contact him for everything that goes on with me.

If the doctor says I need to go, I need to go.

My daughter is really the quarterback of the family. And I have delegated that authority to her.

The doctors and the nurses…they have the decision making. No input. It is up to the doctor.

She doesn't need any information. She just goes if they tell her.

I wouldn't decide anything. I would talk to the doctor. To tell you the truth, I would tell them, if they feel that they can do it here that is alright or either carry her to the hospital. It's up to them. I wouldn't try to boss them too much.

Delegating the decision to others was used by 36 residents (38%) and nine family members (12%). This difference was significant (χ2 (1, N = 171) = 14.1197, p = 0.0002).

Some of the responses that indicated the use of deliberation or delegation also included emotional reactions that, although evident, were subtler and more subdued than those that were predominantly emotion-based decisions. These were coded as having emotional overtones:

I just don't like hospitals…I would have to be really sick and the doctors would have to tell me that.

If we hadn't had a good hospital experience, we might have been more influenced to stay here.

There was no difference in the proportion of resident and family responses that had emotional overtones. One third of residents' (n = 32) and one third of family members' (n = 25) responses had emotional overtones.

Decision-Making Mode and Preference for Nursing Home Care or Hospital Transfer

Thirty-six respondents (21%) expressed a clear preference that the resident remain in the NH, whereas 41 (24%) clearly preferred that the resident be transferred to the hospital if a change in condition occurred (the remainder were neutral and/or indicated it depended on the situation). Differences between the proportion of residents' compared to family members' preferences for remaining in the NH or transfer to the hospital tested with the chi-square statistic were not significant: χ2 (1, N = 171) = 2.53, p = 0.1115 for the NH preference; χ2 (1, N = 171) = 2.1031, p = 0.147 for hospital preference.

Preference for NH or hospital treatment was also not related to decision-making mode: χ2 (1, N = 171) = 1.4053, p = 0.236 for deliberative mode; χ2 (1, N = 171) = 1.0218, p = 0.312 for emotional mode; and χ2 (1, N = 171) = 0.5298, p = 0.467 for delegation mode. Presence of emotional overtones also was not related to preference for NH or hospital.

Sociodemographic Characteristics and Decision Mode

No significant gender differences were found in the use of the three modes or presence of emotional overtones. There were, however, significant age differences in the use of these modes of decision making. Younger respondents were more likely to use the deliberative mode but exhibited less use of delegating the decision to others. The age of those who did and did not primarily use emotion in their decision making, whether resident or family member, did not differ significantly nor did the presence of emotional overtones (Table 1).

Average Age by Decision-Making Mode Used: Analysis of Variance for Residents and Family Members

Table 1:

Average Age by Decision-Making Mode Used: Analysis of Variance for Residents and Family Members

Logistic regression was used to identify characteristics of respondents associated with each of the modes of decision making used. Respondent age and whether the respondent was a resident or family member had a significant association with use of the deliberative mode; only age had a significant association with use of the delegation mode. None of the factors evaluated (i.e., age, resident or family member status, gender, religion, years of education, years of residence in the United States as a proxy for acculturation or ethnic group membership [as African American, Hispanic, or Afro-Caribbean with European American as the comparison group]) were significantly associated with use of the predominantly emotional mode (Table 2).

Logistic Regression: Factors Associated with Decision Mode Used
Logistic Regression: Factors Associated with Decision Mode Used

Table 2:

Logistic Regression: Factors Associated with Decision Mode Used

Hypothetical Versus Actual Rehospitalization Experience

The majority (79%) of the sample responded to the hypothetical question posed: “If you became ill, for example, developed pneumonia or a urinary tract infection, would it be your preference to be treated here in the NH or would you want to be sent to the hospital for treatment?” A smaller number (21%) spoke of an actual experience that had occurred.

There were some differences in the distribution of decision mode used in these two subsamples. A larger proportion of those responding to a hypothetical situation used the deliberative mode, 83% (n = 115/135) versus 54% (n = 19/35) of those describing an actual rehospitalization. On the other hand, a larger proportion of those responding about an actual situation used the delegation mode (69% [n = 24]) than those responding to a hypothetical situation (15% [n = 20]). Both of these differences were statistically significant. The difference in the proportions of actual versus hypothetical situation using a predominantly emotional mode was not significant (Table 3).

Decision Mode Used in Hypothetical Versus Actual Change in Condition Situations (N = 170)

Table 3:

Decision Mode Used in Hypothetical Versus Actual Change in Condition Situations (N = 170)

Descriptive Model

The Figure depicts the model that emerged from the current analysis. In the first level of the model, NH residents and family members are confronted with a decision (i.e., If you became ill, would it be your preference to be treated here in the NH or sent to the hospital for treatment?). The second level contains the approaches (modes) to decision making. The three modes identified from the analysis were deliberative, emotional, and delegation. The deliberative mode was characterized as information seeking and weighing the pros and cons of each choice. The emotional mode was affective in nature, leading to decisions based predominantly on subjective reactions and past experiences. The delegation mode was typified by the expressed preference to leave the decision to others. Emotional overtones were found to be present in a portion of the interviews classified as the deliberative or delegation modes of decision making.

Decision-making modes used by nursing home residents and their families.


Decision-making modes used by nursing home residents and their families.

Potential moderators identified in the current study were age, illness, fatigue, a sense of vulnerability or increasing incapacity (physical or psychological), and prior experience in the NH and hospital. Indications of these moderators were found during the coding of the interviews and data analysis and in the literature. For example, older NH residents were more likely than their younger family members to use delegation when faced with the decision to stay in the NH or transfer to acute care if a change in condition occurred. Conversely, younger family members were more likely to use the deliberative mode when confronted with the same question. Therefore, age was the first potential moderator of the approach to decision making identified; this has been noted in a number of previous studies (Arora & McHorney, 2000; Deber et al., 2007; Levinson et al., 2005; Li & Fung, 2013).

It was observed that some of the NH residents fatigued easily during the interviews, usually due to illness and/or multiple chronic conditions. Illness and fatigue are, therefore, identified as a second possible moderator when looking at the model. Severity of illness and feeling overwhelmed by the illness have been noted to affect preferences related to decision-making mode (Belcher, Fried, Agostini, & Tinetti, 2006; Levinson et al., 2005). Third, feeling of vulnerability was noted to be a potential moderator. Belcher et al. (2006) noted that some older adults were reluctant to participate in decision making regarding medications due to a perceived lack of knowledge, fear, and anxiety related to that illness. Schröder-Butterfill and Marianti (2006) argue that assessing older adult vulnerabilities is an important component in decision making. Several residents reported that family members had warned them not to make any decisions until they were consulted. Others seemed uncertain of their ability to adequately weigh risks and benefits and make important decisions. Awareness of cognitive decline and fear of making the wrong decision may also be a factor (Smebye, Kirkevold, & Engedal, 2012). Other participants said they had not realized they could be involved in the transfer decision, a situation noted in other studies (Nota, Drossaert, Taal, & van de Laar, 2016). Fourth, experiences in the NH and/or in the hospital prior to transfer to the NH were often referenced in participants' reports of their decision making. Although Sjoberg (2007) observed that negative emotions play a more important role in decision making, in some cases participants referred to positive feelings about the NH or hospital as the driver of their decision. The potential influence of these moderators should be further tested in future studies.


NH resident and family member responses to a real or hypothetical change in condition that raised the question of hospitalization were thoroughly examined and categorized as having used a predominantly deliberative process, emotion-based process, or delegated to trusted others, primarily family members or providers, process. Family members were significantly more likely to use the deliberative mode, seeking information and weighing the risks and benefits; residents were more likely than family members to trust others to make the decision.

A predominately emotion-based decision and presence of emotional overtones in family members' and residents' responses were found to be less frequently used when compared to the deliberative and delegation modes. Interestingly, there were no differences found in the proportion of residents and family members using an emotion-based decision mode or emotional overtones. Gender, ethnicity, level of education, religion, and years of residence in the United States were not associated with the mode used. Only age and status (i.e., resident or family member) were found to be significantly associated with any of the modes used: respondents who delegated the decision to others were older than those who used the deliberative mode, and residents were more likely to leave the decision to those they trusted than family members. The mode used was also not significantly related to expressed preference for NH versus hospital treatment when a change in condition occurs. However, those who responded to a hypothetical situation were more likely to use the deliberative mode, whereas more respondents who related an actual experience reported having delegated the decision to others.

Family members and residents clearly differed in approach to making this important decision, with families more often being deliberative and residents more frequently delegating the decision to others. There are several possible reasons for this:

  • The resident may feel ill, fatigued, and/or no longer confident of his/her cognitive abilities, particularly the ability to make important decisions.
  • Family members may have directed the resident not to make decisions without consulting with them.
  • Residents are older: More are members of the “Greatest Generation” that had higher levels of trust in authorities, the law, and medical professions in particular (Hughes & Goldie, 2009; Jennings & Stoker, 2004).

This difference may also be related to their status as NH residents. Several studies have noted a diminishing decisional independence and increased delegation of decision making to families in NH residents (Hughes & Goldie, 2009; Reinardy & Kane, 1999) either willingly and consciously or as a gradual removal by others from decision making (Shawler et al., 2001). High and Rowles (1995) termed this progressive surrogacy, which occurs more often in long-stay residents and those with increasing incapacity and/or nearing the end of life. Recently, Chuang, Abbey, Yeh, Tseng, and Liu (2013) noted a feeling of powerlessness in Taiwanese NH residents they interviewed. Although this powerlessness was not verbalized in the current sample, it could be an underlying factor contributing to use of the delegation mode.

The failure to find any association between decision mode and preference for NH or hospital-based treatment was unexpected. It was thought that the insistence on hospital transfer might be a more emotion-based response but this was not evident. It seemed reasonable to expect that in the event of an actual emergency, it may be more difficult to be deliberate and that emotion may outweigh deliberation at that time. Nevertheless, the emotion-based mode did not seem to be strongly associated with either preference or type of event discussed, whether actual or hypothetical. In fact, many emotion-based responses were favorably inclined toward remaining in the NH, with some residents and families reporting negative hospital experiences and more familiarity, comfort, and trust in NH staff (Tappen, 2016b).

An additional question raised by the results is whether the use of delegation and putting trust in others could be considered a variation of the emotion-based mode of decision making. However, there were several distinct differences between the delegation and emotional modes:

  • Most responses reflecting use of delegation were relatively unemotional either in the words chosen or the tone of voice used.
  • Most respondents who used delegation had already selected the decision maker and named him/her; some indicated that they had deferred to this individual in the past.
  • When using the emotion-based mode, the respondent's preference was based on his/her own feelings, whereas this choice was delegated to another by those using the delegation mode.

Historically, the research literature has proposed a dual information processing approach to decision making (Epstein, 1994; Reyna, 2004), using either a deliberative or affective process. More recently, it has been suggested that cognitive information may also be a part of the input that results in emotion-based decisions (Glöckner & Witteman, 2010). The current analysis of the decision to be made in this specific situation (i.e., whether to be treated in the NH or transferred to the hospital for a change in condition) revealed an additional mode, that of delegating the decision to another. The specific situation to which residents and families responded, whether to be treated in the NH or transferred to the hospital when a change in condition occurs, may help explain their reliance on three, as opposed to two, modes of decision making. Another consideration is the need for an often quick decision regarding where to receive treatment for a change in condition, which may not always allow time for extensive information gathering and weighing of choices. Illness, fatigue, and feeling overwhelmed or unprepared to make such a decision may also be influencing factors.

Implications for Practice

Making decisions regarding hospitalization or receiving treatment in the NH when a change in condition occurs can be challenging for residents and family members. Three modes—deliberative, emotional, and delegation—were used in this specific area of decision making. Giving medical providers, nurses, social workers, and NH administrators some insight into how individuals process and arrive at these decisions may be useful in developing approaches to supporting residents and family members to make informed choices. In some cases, the decision is made in the midst of a crisis. Residents or their families may insist on transfer to acute care without forethought. In a meta-synthesis of studies related to end-of-life experiences, Fosse, Schaufel, Ruths, and Malterud (2014) observed that residents and family members found making a decision “in the heat of the moment” painful (p. 7). A limitation of the current study, exclusion of those with dementia, suggests that these findings cannot be applied to these residents without further study.

Appreciating the extent to which NH residents may place their trust in family members or providers to make decisions on whether to transfer to acute care from the NH is another striking finding of the current study. When this is the case, providers should involve those to whom the decision has been delegated at the earliest opportunity. Gjerberg, Lillemoen, Førde, and Pedersen (2015) noted that residents differ in the amount of information they want and that some expressed confidence in their providers in regard to end-of-life decisions. This finding was apparent in the current study as well. Participants who used the delegation mode indicated that they had confidence in their family or providers to make the right decision for them. On the other hand, those using the deliberative mode wanted more information than those who used the delegation or affective mode.

Decision-making guides are becoming increasingly common in health care (Courtney, Spivey, & Daniel, 2014). For those designing such a guide for NH residents and families faced with the decision to transfer, it is important to recognize that individuals' approaches to making this decision may be different. Addressing each of the three modes of decision making in constructing a decision guide may make it more salient to all readers than it would be if only one mode of decision making were addressed.

Implications for Research

Future studies should evaluate the decision modes used by older adults confronted with other important health-related decisions and the effects of the suggested moderators, particularly cognitive ability and physical condition, on the decision-making mode used.


The current data suggest three primary modes of decision making used by NH residents and their families when confronted with a change in condition that might result in a hospital readmission: deliberative, emotional, and delegation. Appreciation of the differences in these modes and the factors that may moderate the resident's or family member's choice of decision-making mode should serve as a guide to discussions with residents and their family members about this important decision.


  • Adlam, A.L., Patterson, K. & Hodges, J.R. (2009). “I remember it as if it were yesterday”: Memory for recent events in patients with semantic dementia. Neuropsychologia, 47, 1344–1351. doi:10.1016/j.neuropsychologia.2009.01.029 [CrossRef]
  • Arora, N.K. & McHorney, C.A. (2000). Patient preferences for medical decision making: Who really wants to participate?Medical Care, 38, 335–341. doi:10.1097/00005650-200003000-00010 [CrossRef]
  • Belcher, V.N., Fried, T.R., Agostini, J.V. & Tinetti, M.E. (2006). Views of older adults on patient participation in medication-related decision making. Journal of General Internal Medicine, 21, 298–303. doi:10.1111/j.1525-1497.2006.00329.x [CrossRef]
  • Bell, L. & Duffy, A. (2009). A concept analysis of nurse-patient trust. British Journal of Nursing, 18, 46–51. doi:10.12968/bjon.2009.18.1.32091 [CrossRef]
  • Borson, S., Scanlan, J., Brush, M., Vitaliano, P. & Dokmak, A. (2000). The mini-cog: A cognitive “vital signs” measure for dementia screening in multi-lingual elderly. International Journal of Geriatric Psychiatry, 15, 1021–1027. doi:10.1002/1099-1166(200011)15:11<1021::AID-GPS234>3.0.CO;2-6 [CrossRef]
  • Brownell, J., Wang, J., Smith, A., Stephens, C. & Hsia, R.Y. (2014). Trends in emergency department visits for ambulatory care sensitive conditions by elderly nursing home residents, 2001 to 2010. JAMA Internal Medicine, 174, 156–158. doi:10.1001/jamainternmed.2013.11821 [CrossRef]
  • Carstensen, L.L. & Turk-Charles, S. (1994). The salience of emotion across the adult life span. Psychology and Aging, 9, 259–264. doi:10.1037/0882-7974.9.2.259 [CrossRef]
  • Chuang, Y.H., Abbey, J.A., Yeh, Y.C., Tseng, I.J. & Liu, M.F. (2013). As they see it: A qualitative study of how older residents in nursing homes perceive their care needs. Collegian, 22, 43–51. doi:10.1016/j.colegn.2013.11.001 [CrossRef]
  • Courtney, M.R., Spivey, C. & Daniel, K.M. (2014). Helping patients make better decisions: How to apply behavioral economics in clinical practice. Patient Preference and Adherence, 8, 1503–1512. doi:10.2147/PPA.S71224 [CrossRef]
  • Crandall, B., Klein, G.A. & Hoffman, R.R. (2006). Working minds: A practitioner's guide to cognitive task analysis. Cambridge, MA: A Bradford Book.
  • Creditor, M.C. (1993). Hazards of hospitalization of the elderly. Annals of Internal Medicine, 118, 219–223. doi:10.7326/0003-4819-118-3-199302010-00011 [CrossRef]
  • Deber, R.B., Kraetschmer, N., Urowitz, S. & Sharpe, N. (2007). Do people want to be autonomous patients? Preferred roles in treatment decision-making in several patient populations. Health Expectations, 10, 248–258. doi:10.1111/j.1369-7625.2007.00441.x [CrossRef]
  • Dionne-Odom, J.N., Willis, D.G., Bakitas, M., Crandall, B. & Grace, P.J. (2015). Conceptualizing surrogate decision making at end of life in the intensive care unit using cognitive task analysis. Nursing Outlook, 63, 331–340. doi:10.1016/j.outlook.2014.10.004 [CrossRef]
  • Epstein, S. (1994). Integration of the cognitive and the psychodynamic unconscious. American Psychologist, 49, 709–724. doi:10.1037/0003-066X.49.8.709 [CrossRef]
  • Fosse, A., Schaufel, M.A., Ruths, S. & Malterud, K. (2014). End-of-life expectations and experiences among nursing home patients and their relatives—A synthesis of qualitative studies. Patient Education and Counseling, 97, 3–9. doi:10.1016/j.pec.2014.05.025 [CrossRef]
  • Friedman, S.M., Mendelson, D.A., Bingham, K.W. & McCann, R.M. (2008). Hazards of hospitalization: Residence prior to admission predicts outcomes. The Gerontologist, 48, 537–541. doi:10.1093/geront/48.4.537 [CrossRef]
  • Gjerberg, E., Lillemoen, L., Førde, R. & Pedersen, R. (2015). End-of-life care communications and shared decision-making in Norwegian nursing homes: Experiences and perspectives of patients and relatives. BMC Geriatrics, 15, 103. doi:10.1186/s12877-015-0096-y [CrossRef]
  • Glöckner, A. & Witteman, C. (2010). Beyond dual-process models: A categorisation of processes underlying intuitive judgement and decision making. Thinking & Reasoning, 16, 1–25. doi:10.1080/13546780903395748 [CrossRef]
  • Helton, M.R., van der Steen, J.T., Daaleman, T.P., Gamble, G.R. & Ribbe, M.W. (2006). A cross-cultural study of physician treatment decisions for demented nursing home patients who develop pneumonia. Annals of Family Medicine, 4, 221–227. doi:10.1370/afm.536 [CrossRef]
  • High, D.M. & Rowles, G.D. (1995). Nursing home residents, families, and decision making: Toward an understanding of progressive surrogacy. Journal of Aging Studies, 9, 101–117. doi:10.1016/0890-4065(95)90006-3 [CrossRef]
  • Hoffman, R.R., Crandall, B. & Shadbolt, N. (1998). Use of the critical decision method to elicit expert knowledge: A case study in the methodology of cognitive task analysis. Human Factors, 40, 254–276. doi:10.1518/001872098779480442 [CrossRef]
  • Hughes, C.M. & Goldie, R. (2009). “I just take what I am given”: Adherence and resident involvement in decision making on medicines in nursing homes for older people: A qualitative survey. Drugs & Aging, 26, 505–517. doi:10.2165/00002512-200926060-00007 [CrossRef]
  • Hupcey, J.E., Penrod, J., Morse, J. & Mitcham, C. (2001). An exploration and advancement of the concept of trust. Journal of Advanced Nursing, 36, 282–293. doi:10.1046/j.1365-2648.2001.01970.x [CrossRef]
  • Jennings, M.K. & Stoker, L. (2004). Social trust and civic engagement across time and generations. Acta Politica, 39, 342–379. doi:10.1057/palgrave.ap.5500077 [CrossRef]
  • Lamb, G., Tappen, R., Diaz, S., Herndon, L. & Ouslander, J.G. (2011). Avoidability of hospital transfers of nursing home residents: Perspectives of frontline staff. Journal of the American Geriatrics Society, 59, 1665–1672. doi:10.1111/j.1532-5415.2011.03556.x [CrossRef]
  • Leff, B., Burton, L., Mader, S.L., Naughton, B., Burl, J., Inouye, S.K. & Burton, J.R. (2005). Hospital at home: Feasibility and outcomes of a program to provide hospital-level care at home for acutely ill older patients. Annals of Internal Medicine, 143, 798–808. doi:10.7326/0003-4819-143-11-200512060-00008 [CrossRef]
  • Levinson, W., Kao, A., Kuby, A. & Thisted, R.A. (2005). Not all patients want to participate in decision making. Journal of General Internal Medicine, 20, 531–535. doi:10.1111/j.1525-1497.2005.04101.x [CrossRef]
  • Li, T. & Fung, H. (2013). Age differences in trust: An investigation across 38 countries. Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 68, 347–355. doi:10.1093/geronb/gbs072 [CrossRef]
  • Maki, Y., Amari, M., Yamaguchi, T., Nakaaki, S. & Yamaguchi, H. (2012). Anosognosia: Patients' distress and self-awareness of deficits in Alzheimer's disease. American Journal of Alzheimer's Disease and Other Dementias, 27, 339–345. doi:10.1177/1533317512452039 [CrossRef]
  • Nota, I., Drossaert, C.H., Taal, E. & van de Laar, M.A. (2016). Arthritis patients' motives for (not) wanting to be involved in medical decision-making and the factors that hinder or promote patient involvement. Clinical Rheumatology, 35, 1225–1235. doi:10.1007/s10067-014-2820-y [CrossRef]
  • Ouslander, J.G. & Maslow, K. (2012). Geriatrics and the triple aim: Defining preventable hospitalizations in the long-term care population. Journal of the American Geriatrics Society, 60, 2313–2318. doi:10.1111/jgs.12002 [CrossRef]
  • Peters, E., Hess, T.M., Vastfjall, D. & Auman, C. (2007). Adult age differences in dual information processes: Implications for the role of affective and deliberative processes in older adults' decision making. Perspectives in Psychological Science, 2, 1–23. doi:10.1111/j.1745-6916.2007.00025.x [CrossRef]
  • Reinardy, J. & Kane, R.A. (1999). Choosing an adult foster home or a nursing home: Residents' perceptions about decision making and control. Social Work, 44, 571–585. doi:10.1093/sw/44.6.571 [CrossRef]
  • Reyna, V.F. (2004). How people make decisions that involve risk: A dual-processes approach. Current Directions in Psychological Science, 13, 60–66. doi:10.1111/j.0963-7214.2004.00275.x [CrossRef]
  • Schröder-Butterfill, E. & Marianti, R. (2006). A framework for understanding old-age vulnerabilities. Ageing and Society, 26, 9–35. doi:10.1017/S0144686X05004423 [CrossRef]
  • Shawler, C., Rowles, G.D. & High, D.M. (2001). Analysis of key decision-making incidents in the life of a nursing home resident. The Gerontologist, 41, 612–622. doi:10.1093/geront/41.5.612 [CrossRef]
  • Sjoberg, L. (2007). The distortion of beliefs in the face of uncertainty. International Journal of Management and Decision Making, 8, 1–29. doi:10.1504/IJMDM.2007.012148 [CrossRef]
  • Sloman, S.A. (1996). The empirical case for two systems of reasoning. Psychological Bulletin, 119, 3–22. doi:10.1037/0033-2909.119.1.3 [CrossRef]
  • Slovic, P., Peters, E., Finucane, M.L. & MacGregor, D.G. (2005). Affect, risk, and decision making. Health Psychology, 24(4 Suppl.), S35–S40. doi:10.1037/0278-6133.24.4.S35 [CrossRef]
  • Smebye, K.L., Kirkevold, M. & Engedal, K. (2012). How do persons with dementia participate in decision making related to health and daily care? A multi-case study. BMC Health Services Research, 12, 241. doi:10.1186/1472-6963-12-241 [CrossRef]
  • Tappen, R.M. (2016a). Advanced nursing research: From theory to practice. Burlington, MA: Jones and Bartlett.
  • Tappen, R.M. (2016b, January31). They know me here: Patients' perspectives on their nursing home experience. Retrieved from
  • Thomas, E.J. & Brennan, T.A. (2000). Incidence and types of preventable adverse events in elderly patients: Population based review of medical records. BMJ, 320, 741–744. doi:10.1136/bmj.320.7237.741 [CrossRef]
  • Trachtenberg, F., Dugan, E. & Hall, M.A. (2005). How patients' trust relates to their involvement in medical care. Journal of Family Practice, 54, 344–352.

Average Age by Decision-Making Mode Used: Analysis of Variance for Residents and Family Members

Age of Those Who Used ModeAge of Those Who Did Not Use Mode
Decision ModeYears, Mean (SD)Years, Mean (SD)Fp Value
Deliberative69.00 (14.13)77.98 (13)16.24<0.0001
Delegation79.48 (10.85)69.47 (14.6)17.51<0.0001
Emotional73.80 (17.48)72.03 (14.07)0.210.651
Emotional overtone74.21 (13.71)71.16 (14.64)1.680.1968

Logistic Regression: Factors Associated with Decision Mode Used

Mode/FactorBSEBβOdds Ratio95% CIWald Chi-Squarep Value
Age0.040.02160.38321.04[1.005, 1.094]4.83640.0279*
Years of education−0.020.0772−0.04690.97[0.834, 1.129]0.14850.7000
Gender−0.720.4020−0.18590.483[0.220, 1.063]3.26770.0707
Years of U.S. residence00.0145−0.05780.99[0.967, 1.024]0.12070.7282
  Catholic vs. not Catholic0.520.72840.12791.68[0.403, 7.014]0.50980.4752
  Protestant vs. not Protestant−0.010.7365−0.00420.985[0.232, 4.170]0.00040.9831
  Jewish vs. not Jewish0.390.77020.09151.49[0.330, 6.747]0.26920.6038
  African American vs. not African American0.730.72140.13362.08[0.508, 8.588]1.04220.3073
  Hispanic vs. not Hispanic0.620.71670.10841.87[0.460, 7.631]0.76630.3814
  Afro-Caribbean vs. not Afro-Caribbean1.541.13810.17704.66[0.502, 43.448]1.83340.1757
Participant status (resident or family member)−0.920.4349−0.25670.395[0.168, 0.925]4.57260.0325*
Age−0.010.0277−0.13470.983[0.931, 1.038]0.36250.5471
Years of education−0.070.1122−0.11090.93[0.748, 1.161]0.39280.5308
Gender0.610.58420.15701.84[0.588, 5.805]1.10420.2933
Years of U.S. residence0.010.01810.21531.01[0.983, 1.056]1.06700.3016
  Catholic vs. not Catholic−0.151.2114−0.03760.85[0.588, 5.805]0.01600.8945
  Protestant vs. not Protestant−0.761.5555−0.20780.46[0.048, 4.473]0.44020.5070
  Jewish vs. not Jewish−0.341.2585−0.07900.70[0.060, 8.343]0.07520.7839
  African American vs. not African American0.621.18680.11311.86[0.18, 19.092]0.27580.5995
  Hispanic vs. not Hispanic0.531.21800.09301.73[0.157, 18.644]0.19530.6585
  Afro-Caribbean vs. not Afro-Caribbean0.341.39790.04011.41[0.092, 21.946]0.06220.8630
Participant status (resident or family member)0.530.65820.14651.70[0.468, 6.175]0.64950.4203
Age−0.060.0256−0.49540.94[0.894, 0.989]5.74570.0165*
Years of education0.090.08800.15721.10[0.928, 1.315]1.25910.2618
Gender0.460.45120.11881.59[0.657, 3.853]1.06020.3032
Years of U.S. residence00.0159−0.00191.00[0.969, 1.032]0.00010.9914
  Catholic vs. not Catholic−0.650.8406−0.16080.52[0.100, 2.700]0.60560.4365
  Protestant vs. not Protestant0.310.87770.08621.37[0.246, 7.679]0.13140.7170
  Jewish vs. not Jewish−0.480.8885−0.11050.61[0.108, 3.521]0.29520.5869
  African American vs. not African American−1.380.8566−0.25180.25[0.047, 1.338]2.62470.1052
  Hispanic vs. not Hispanic−1.280.7876−0.22290.27[0.059, 1.289]2.68170.1015
  Afro-Caribbean vs. not Afro-Caribbean−1.671.2364−0.19280.18[0.017, 2.106]1.84340.1746
Participant status (resident or family member)0.920.510.25632.53[0.929, 6.894]3.29540.0695

Decision Mode Used in Hypothetical Versus Actual Change in Condition Situations (N = 170)

n (%)
Decision ModeHypothetical Situation (n = 135)Actual Situation (n = 35)χ2p Value
  Not used23 (17)16 (46)12.920.0003
  Used112 (83)19 (54)
  Not used102 (76)23 (66)1.380.2396
  Used33 (24)12 (34)
  Not used115 (85)24 (69)5.140.0233
  Used20 (15)11 (31)

Dr. Tappen is Eminent Scholar and Professor, and Dr. Elkins is PhD Candidate, Florida Atlantic University, Christine E. Lynn College of Nursing, Boca Raton; Ms. Worch is Marriage and Family Therapist, Stable Place Inc., Davie, and PhD Student, Nova Southeastern University, College of Humanities, Arts and Social Sciences, Fort Lauderdale; and Ms. Weglinski is Clinical Supervisor of Sterile Processing, Boca Raton Regional Hospital, Boca Raton, Florida.

The authors have disclosed no potential conflicts of interest, financial or otherwise. This research was supported by the Patient-Centered Outcomes Research Institute (PCORI; grant 1IP2PI000281-01, Principal Investigator, R.M. Tappen). All statements in this article, including its findings and conclusions, are solely those of the authors and do not necessarily represent the views of the PCORI, its Board of Governors or Methodology Committee.

Address correspondence to Ruth M. Tappen, EdD, RN, FAAN, Eminent Scholar and Professor, Florida Atlantic University, Christine E. Lynn College of Nursing, 777 Glades Road, Building 84, Room 307, Boca Raton, FL 33431; e-mail:

Received: May 27, 2016
Accepted: August 24, 2016
Posted Online: September 27, 2016


Sign up to receive

Journal E-contents