Journal of Gerontological Nursing

Feature Article 

Enabling Advance Directive Completion: Feasibility of a New Nurse-Supported Advance Care Planning Intervention

Michelle M. Hilgeman, PhD; Constance R. Uphold, PhD, ARNP, FAAN; Amber N. Collins, MS; Lori L. Davis, MD; Douglas P. Olsen, PhD; Kathryn L. Burgio, PhD; Classie A. Gordon, DNP, RN; Tranace N. Coleman, RN, ADN, BS; Jamie DeCoster, PhD; Whitney Gay, MA; Rebecca S. Allen, PhD, ABPP


Adults who complete an advance directive (AD) are not consistently offered information about the risks, benefits, or alternatives (RBA) of the life-sustaining medical procedures addressed on standardized forms. The current article describes a new patient-centered nurse-supported advance care planning (NSACP) intervention focused on providing information about RBA of life-sustaining procedures. Fifty participants (mean age = 50.26 years) at a Veterans Affairs medical center were randomized to the NSACP intervention or a comparison condition. Before randomization, 78% (n = 39) expressed interest in RBA information. Of participants in the NSACP group, 94% (n = 30) completed an AD. Participants who received NSACP made more decisions to decline life-sustaining treatment than those who were randomized to the comparison group. Promising feasibility data include brevity (mean = 46 minutes), high patient satisfaction, participant retention, and treatment fidelity. The NSACP holds promise as a brief, educational intervention to support patients in completing an AD. [Journal of Gerontological Nursing, 44(7), 31–42.]


Adults who complete an advance directive (AD) are not consistently offered information about the risks, benefits, or alternatives (RBA) of the life-sustaining medical procedures addressed on standardized forms. The current article describes a new patient-centered nurse-supported advance care planning (NSACP) intervention focused on providing information about RBA of life-sustaining procedures. Fifty participants (mean age = 50.26 years) at a Veterans Affairs medical center were randomized to the NSACP intervention or a comparison condition. Before randomization, 78% (n = 39) expressed interest in RBA information. Of participants in the NSACP group, 94% (n = 30) completed an AD. Participants who received NSACP made more decisions to decline life-sustaining treatment than those who were randomized to the comparison group. Promising feasibility data include brevity (mean = 46 minutes), high patient satisfaction, participant retention, and treatment fidelity. The NSACP holds promise as a brief, educational intervention to support patients in completing an AD. [Journal of Gerontological Nursing, 44(7), 31–42.]

Advance care planning is the platform from which a formal written advance directive (AD) evolves. Benefits and challenges of AD use have been extensively debated. ADs may serve as concrete aids to overcome individuals' aversion to considering decisions about dying (Perkins, 2007). Yet factors that limit their value include a potential poor match to real wishes; incomplete understandings of medical care, associated complications, and survival rates; preference changes that are not updated in writing; clinical care situations that are not addressed; and inaccessibility of the document when needed (Perkins, 2007).

Historically, only 18% to 36% of adults in the United States complete an AD (U.S. Department of Health and Human Services, 2008). In the United States, the highest completion rates, 72%, are observed among adults age ≥60 who die in the hospital (Silveira, Wiitala, & Piette, 2014). Community-dwelling adults are estimated to have AD completion rates as low as 26% (Rao, Anderson, Feng-Chang, & Laux, 2014) up from the 5% to 15% documented one decade ago (Kirschner, 2005). However, little is known about AD completion rates by Veterans. This is problematic because family composition in Veteran populations differs from civilian populations in ways that can impact the availability of surrogate decision makers (Clever & Segal, 2013). Once separated from active duty, Veterans are three times more likely to be divorced than non-Veterans (Kelty, Kleykamp, & Segal, 2010). Similarly, 56% of Veterans are 60 or older and thus more likely to need an AD (U.S. Department of Veterans Affairs [VA], n.d.).

Although progress has been observed from studies conducted with Veterans in the early 1990s, opportunities for improvement remain. Sugarman, Weinberger, and Samsa (1992) found that only 4% of participating Veterans had a living will; 43% had never heard of a living will; and 91% believed that signing a living will would not impact their treatment. One study (N = 324) included responses from 44 Veterans (Freeman & Berger, 2009), who reported higher rates of AD completion than non-Veterans (i.e., 44% and 36%, respectively). In a rural sample, 87% of participants (N = 201; mean age = 54.2 years) did not have an AD (Mahaney-Price et al., 2014). Of those, 40% expressed a desire for help completing an AD. Interest in help completing a living will was not associated with age or most measures of illness or disability, in contrast with many health care providers' beliefs (Freeman & Berger, 2009). In a 2017 study, significant discrepancies were observed in the number of Veterans who have thought about their preferred surrogate decision makers and treatment preferences (50% and 70%, respectively), and those who have an AD documented (26%) in their electronic medical record (Dubbert, Garner, Lensing, White, & Sullivan, 2017).

Nursing as a discipline emphasizes patient education as key to effective intervention (Bastable, 2009; Pyle, 2015; Strupei, Buß, & Dassen, 2013) and may be particularly well-positioned to facilitate seamless advance care planning discussions. Nurses already assume responsibility for provision of information to patients about their disease processes; elicitation, clarification, and documentation of patient preferences for medical care; facilitation of advance care planning and written ADs; and provision of emotional support to patients and families in many settings (Murphy et al., 2000). Several promising studies have examined strategies that facilitate nurses' role, including: (a) using an interview guide (Horne, Seymour, & Shepherd, 2006; Song et al., 2009); (b) reading an outline concerning end-of-life care, providing a fact sheet, and answering questions (and patient education by a physician) (Holliday, 2009); (c) conducting educational sessions with a workbook, discussion, and guidance in completing the AD form (Medvene et al., 2003); and (d) using a story telling approach to facilitate communication of treatment preferences (Chan & Pang, 2010). These studies boast completion rates from 49% to 70% in nursing home and community settings (Holliday, 2009; Medvene et al., 2003; Molloy et al., 2000; Song et al., 2009). Yet, outside of research studies, local policies may be devoid of a clear delineation of nursing authority in the provision of RBA information about common life-sustaining procedures as a component of advance care planning. A user-friendly manual and policies to guide nurses in AD role responsibilities could address remaining barriers (e.g., inconsistent physician support) (Lipson, Hausman, Higgins, & Burant, 2004).

The current article describes the feasibility of a single-session, patient-centered, nurse-supported advance care planning (NSACP) intervention delivered by a RN. Developed for the professional nurse to use across settings and the illness–wellness continuum, the intervention aims to: (a) be easy-to-use, (b) include RBA information, (c) require ≤1 hour, and ideally (d) enable completion of an AD within that 1 hour. In the sections that follow, the following are described: (a) intervention development and conceptual framework; (b) randomized, controlled, two group pre-/posttest feasibility design; (c) usual care comparison condition; and (d) key measures; before discussing (e) the preliminary results and conclusions. The intervention offers a standardized approach for discussing critical information about the RBA of commonly contemplated medical treatments and procedures to support more informed choices during advance care planning.

Nurse-Supported Advance Care Planning Intervention

Intervention Development

The NSACP intervention was developed based on the premise that enabling a sense of security is necessary before individuals are willing to consider threatening themes such as loss of autonomy and death (Perkins, 2007). The Theory for Enabling Safety framework describes three specific enabling processes: affirming, comforting, and guiding—the foundation for the educational and support components of the intervention session (Öhlén et al., 2007). The current authors' adapted conceptual model (Figure 1) places the Theory for Enabling Safety in the context of the professional nurse rather than a family support member. The resulting intervention manual provides RNs with scripted text that facilitates discussion of: (a) advance care planning education; (b) guidance in values identification; (c) RBA information about common life-sustaining procedures and the opportunity to ask questions about those procedures; (d) multiple structured opportunities to ask questions and discuss personal values and concerns; and (e) assistance in completing the AD form.

Nurses' role in enabling safety for advance care planning discussions. Modified from the Theory of Enabling Safety Framework (Öhlén et al., 2007).

Figure 1.

Nurses' role in enabling safety for advance care planning discussions. Modified from the Theory of Enabling Safety Framework (Öhlén et al., 2007).

Prototype Testing

Before the current study was initiated, the drafted intervention went through two rounds of prototype testing to finalize materials. The first round used 13 baccalaureate nursing student pairs in a role-playing methodology with one student playing the “nurse” and one the “patient.” All completed the intervention in ≤45 minutes, and all but one mock “patient” completed the AD accurately. A focus group was used to collect feedback (i.e., the intervention was easy to follow, training and content was easy to understand). Second, the intervention was tested with 10 Veteran stakeholders—without prior AD experience—in a small non-randomized study. Nine (90%) completed the AD during the study session, and all 10 reported the highest level of satisfaction (4 on a 4-point scale) on the Client Satisfaction Questionnaire (CSQ-8; Attkisson, 2012).

Summary of NSACP Content

The intervention comprises 50 PowerPoint® slides with a Flesch–Kincaid Grade Level 5 readability. The intervention is administered in one of two formats: depicted visually on the RN's computer as a slide show, or as a printed handout. The intervention text begins with information about the right to make medical care decisions in advance and why an AD is important. The text identifies the nurse as the ACP guide for the intervention session. The nurse guide follows the text and facilitates a focus on participant feelings and past participant experiences generally and in making important life decisions. Key terms, relationships, and considerations are introduced, including: AD, living will, and choosing a health care proxy. Next, the nurse encourages the participant to consider reasons for wanting to complete an AD and personal values that may influence future changes in health and abilities or desires for specific medical care. A series of discussion questions guide the interaction, asking Veterans for their thoughts and feelings about the topic and RBA material. For example, participants are asked, “What feelings are you having as we discuss these decisions?” RBA information about each of six procedures (i.e., cardio-pulmonary resuscitation, respirator, kidney dialysis, feeding tube, intravenous antibiotic, and hospitalization) follows. Finally, Veterans are provided page-by-page guidance in completing the VA form 10-0137, “Advance Directive Durable Power of Attorney for Health Care and Living Will” (access and an instruction to destroy old AD copies whenever a new AD is completed (VA, 2012). Corresponding NSACP slides would require only small modifications for pairing with alternate forms in other settings.

RN Training on NSACP Manual

Four VA staff RNs (C.A.G., T.N.C., and two others) served as interventionists. All worked full time in either clinical or non-clinical roles throughout the hospital, ranging from primary care to quality management to education; none were involved with the research department prior to this study. Training included study protocol procedures, theoretical foundations (Öhlén et al., 2007), and role play with feedback for improvement. Provision of the RBA information and solicitation and monitoring of Veterans' reactions to the information presented was rehearsed. The training included principles and techniques for managing challenges of the RN role and refresher training on cultural competency and sensitivity issues. The intervention was delivered during office appointments, typically to outpatients, although some participants were in long-term care or other residential treatment programs (e.g., homelessness, substance abuse, mental health recovery); none were in acute care.


Study Design

A randomized, controlled, two-group, pre-/posttest feasibility study was conducted as a first step toward the goal of establishing a new evidence-based intervention (Bowen et al., 2009). To gain more experience with the newly developed intervention, an unequal randomization ratio (2:1) was applied, as recommended in prior review articles (Dumville, Hahn, Miles, & Torgerson, 2006). Not every newly developed intervention should be tested for effectiveness; feasibility studies are used to evaluate whether an intervention warrants additional testing and is practical in a given setting or with a specific population, and can provide information on ways to modify the intervention or protocol (e.g., recruitment, randomization, data completion rate) (Green & Glasgow, 2006). With these goals, process measures such as recruitment, randomization, treatment fidelity, retention, and patient satisfaction were identified as indicators of success.

Usual Care Comparison Group

Within the VA, usual care for out-patient Veterans who are interested in completing an AD are referred to a social worker in the primary care clinic. This usual care support typically includes: (a) a copy of the AD document, (b) unscripted education on how to complete the AD document, and (c) encouragement of clarifying and goal-related questions. If the Veteran has medical questions, the social worker refers the Veteran to the primary care provider (PCP).

The extent to which information about RBA is spontaneously raised during usual care AD discussions is beyond the scope of the current study. To establish a relative baseline, participants were asked whether they were interested in receiving RBA information prior to AD completion. Those who indicated “yes” were asked to meet with their PCP before the AD session with their social worker. The study team facilitated the consultation with primary care via e-mail, letting them know that their patient was interested in meeting to discuss the RBA information. Veterans who did not express interest in RBA information proceeded directly to the social worker. The study authors acknowledge that this design decision placed emphasis on participants obtaining the requested information over a true comparison to usual care and considered calling the comparison group a “modified usual care” condition. Those who had not met with their provider after 2 months were scheduled for the post-test assessment and exited from the study.

Recruitment and Study Procedures

Recruitment was performed in in-patient, outpatient, and rural mobile health clinical areas at a southeastern U.S. VA medical center. Recruitment activities included informational outreach to care providers and care team members using the Institutional Review Board–approved study flyers.

A trained member of the research team (A.N.C.) met individually with interested Veterans to determine eligibility, obtain informed consent, and collect baseline measures. Veterans meeting the following inclusion criteria who were willing to provide consent were eligible: (a) age ≥19 years, (b) had not previously completed an AD, and (c) spoke and understood English. Those with: (a) psychotic symptoms that interfered with the ability to consent, (b) diagnoses of dementia, or (c) serious mental illness were excluded—as were any Veterans who reported actively considering suicide or homicide and those receiving hospice or palliative care.

After completing baseline measures, a randomization assignment envelope was opened, and the participant was notified immediately of group assignment. The principal investigator (PI) was blinded. Next, the RN (C.A.G., T.N.C.) or social worker met one-on-one with the Veteran to complete the AD session. Finally, the research assistant (A.N.C.) conducted a same-day, in-person, post-intervention satisfaction measure and a follow-up telephone assessment 8 to 30 days later. Participants received $20 for completing the baseline measures and $20 for completing the post-intervention assessment. No compensation was provided for the follow-up telephone interview.


Baseline Measures. Demographic data, past experience with advance care planning, desire for information about ADs, desire for help completing an AD, and desire for RBA information about common life-sustaining procedures were collected. Physical and mental health over the previous 4 weeks was also assessed using the self-reported Short Form-12, version 1 (SF-12) (Ware, Kosinski, & Keller, 1996) under an agency licensing agreement.

Feasibility and Process Measures. Data collection and retention targets were determined a priori based on a review of the literature and consensus of the investigators. Specifically: (a) enrollment rate (i.e., the number of Veterans who signed the informed consent form divided by the number who participated in the informed consent session) ≥60%; (b) randomization rate (i.e., the number of participants randomized divided by the number of participants consented and screened) ≥60%; (c) retention rate (i.e., the number of participants completing the initial follow-up assessment divided by the number who were randomized) ≥85%; and (d) data completion rate (i.e., the number of data values collected and entered in the database divided by the number of data values expected [i.e., 1,800 data values]) ≥90%.

Treatment delivery was estimated using time in minutes for each session. Mean session duration for each group was calculated for the analyses. Duration data allow for rough preliminary estimates of cost by group using mid-range local Veterans Health Administration (VHA) hourly salary data (RN = $36.40 per hour and social worker = $23.64 per hour). This rough proxy for cost includes only an estimate based on clinician time.

Treatment fidelity was assessed by an independent monitor using audio recordings and a 12-item fidelity checklist for the NSACP intervention sessions, and a five-item checklist for the social worker–delivered sessions. Items assessed adherence to the manual in the NSACP intervention group and that contamination had not occurred in the comparison condition (Howe, Keogh-Brown, Miles, & Bachmann, 2007). Each interventionist had his/her first session with a participant audited to ensure post-training adherence; after the first session, a randomly selected 50% of sessions were reviewed. Completed checklists were reviewed by the PI, and feedback was provided to prevent drift.

Satisfaction with the advance care planning session was assessed in two ways. (a) The CSQ-8 (Attkisson, 2012), which assesses perceived treatment value with established reliability, internal consistency, and validity across racial/ethnic groups. Summary scores range from 8 to 32, with higher scores indicating greater satisfaction. (b) Five study-specific items were also assessed using a 5-point Likert scale (very satisfied to very dissatisfied). For example, “How satisfied are you with your new AD decisions?” and “Does your new AD match your real wishes?”

Outcome Measure. AD completion rate was tracked based on participants' completion of VA Form 10-0137 (i.e., completed or not completed) as evidenced by their electronic health record. The form includes decisions about six different treatment scenarios that align with the RBA information provided in the NSACP intervention. Participants who were lost to follow up or did not complete post-intervention self-assessments could still be included in analyses related to AD completion rate (Figure 2). Responses to the six scenarios were also collected to further characterize the decisions made following the advance care planning sessions.

CONSORT flow diagram of the current study.Note. NSACP = nurse-supported advance care planning; ACP = advance care planning; RBA = risks, benefits, or alternatives; PCP = primary care provider; AD = advance directive; CSQ-8 = 8-item Client Satisfaction Questionnaire.

Figure 2.

CONSORT flow diagram of the current study.

Note. NSACP = nurse-supported advance care planning; ACP = advance care planning; RBA = risks, benefits, or alternatives; PCP = primary care provider; AD = advance directive; CSQ-8 = 8-item Client Satisfaction Questionnaire.

Data Analysis

Prior to descriptive and inferential analyses, each variable was evaluated for normality, missing data, and outliers to ensure all assumptions were met for each statistical test. Descriptive statistics were used to report aggregate participant characteristics. The chi-square statistic was used to determine whether statistically significant differences occurred across group assignment for nominal and ordinal scale demographic variables (e.g., gender, race/ethnicity, marital status). Independent samples t tests were used to assess group differences on interval and ratio scale data, including age and scores on the physical and mental health composite scores of the SF-12. Chi-square tests were also used to examine potential differences between participants in the NSACP group versus the comparison group regarding: (a) desire for information about and help with completing an AD and desire for information about RBA; (b) desire for each of six different life-sustaining treatments (yes or no); and (c) the number of Veterans electing to complete the AD form.


Participant Characteristics

Fifty participants completed informed consent and were randomized in a 2:1 ratio using a grid with 66% (n = 33) of participants assigned to the NSACP intervention and 34% (n = 17) randomized to the comparison group (Figure 2). Table 1 and Table 2 summarize participant characteristics and tests of differences between groups at baseline. Participants were 50.26 years old (SD = 12.68 years), predominantly non-Hispanic Black, urban-dwelling, and educated beyond a 12th grade level. The sample had more men than women. There were no between-group differences except for total household income (p = 0.04), such that those randomized to the NSACP intervention reported a lower income.

Participant CharacteristicsParticipant Characteristics

Table 1:

Participant Characteristics

Comparison of Age and Sf-12 Health Characteristics at Baseline (N = 50)

Table 2:

Comparison of Age and Sf-12 Health Characteristics at Baseline (N = 50)

Only four participants reported prior discussions about advance care planning with their providers or a lawyer; none had AD documents. At baseline, 86% desired information about ADs; 78% (n = 39) desired information about RBA of common life-sustaining procedures; and 82% desired help completing an AD. Table 3 summarizes baseline desire for information and help. A between-group difference (p = 0.02) was noted for desire for information about RBA such that those randomized to the NSACP intervention were more likely to report a desire for this information at baseline. In small sample pilot studies, it is expected that randomization may not fully equate groups (West & Thoemmes, 2008).

Frequency of Yes Responses for Information or Help at Baseline (N = 50)

Table 3:

Frequency of Yes Responses for Information or Help at Baseline (N = 50)

Feasibility and Process Outcomes

Indicators of successful data collection and retention met or exceeded a priori acceptability levels, such that study enrollment, randomization, and retention of Veterans randomized to the NSACP intervention group all reached 100%. Retention across all participants was 94% (i.e., 47 of 50 participants completed the post-intervention assessment). Observed data completion rates also exceeded targets, at 99% (i.e., 1,773 items present/1,800 items possible), with no detectable patterns in omitted or missing data.

The mean duration of the NSACP intervention was 46 minutes and 29 minutes for the comparison group, including education, discussion, and completion of the AD form with witnessed signatures. Therefore, the approximate cost per session was $27.59 for NSACP and $11.23 for the comparison session.

Treatment fidelity was promising for the NSACP intervention, with a mean adherence rating of 89.58%, with all but one session falling in the 75% to 100% range. Sessions in the comparison group achieved a mean adherence of 73.33%, reflecting perfect adherence across all reviewed sessions except for two sessions by the same interventionist, who met only one of five required elements. After consultation with the PI, this social worker interventionist ended participation in the study. For this pilot study, all data are retained.

Of a possible 32 points, mean participant satisfaction with the advance care planning sessions was 31.03 (SD = 1.38) on the CSQ-8 for the NSACP group and 28.00 (SD = 4.13) for the comparison group, which was a nonsignificant difference. Participant satisfaction was also assessed over telephone by the research coordinator 8 to 30 days after the intervention visit. Most NSACP participants reported that they were very satisfied (n = 21, 91%) with the intervention session, and the remaining 9% (n = 2) reported that they were somewhat satisfied. Furthermore, 87% (n = 20) indicated that they were very satisfied with the AD decisions made during the session, with the remaining 13% (n = 3) reporting that they were somewhat satisfied. Individuals assigned to the comparison group also reported high levels of satisfaction, such that all six who completed the telephone follow up reported being very satisfied with the social worker session. In addition, three of four individuals (75%) in the comparison condition reported being very satisfied with the resulting AD decisions and one (25%) reported being neither satisfied nor dissatisfied.

Preliminary Outcomes

Thirty of 32 (94%) NSACP participants completed an AD including the living will scenarios section of the VA Advance Directive Form. Their AD decisions across six scenarios are summarized in Table 4. Five of 17 (29%) participants in the comparison group completed an AD form. A higher than anticipated number of Veterans in the comparison group who expressed interest in medical information relevant to the RBA of life-sustaining treatment decisions before completing an AD were not able to connect with their PCP within the 2-month study window, which contributed to the lower rates in the comparison group.

Advance Directive Decisions by Group (N = 35a)

Table 4:

Advance Directive Decisions by Group (N = 35)

Although statistical tests were not performed, Veterans who received the NSACP intervention appeared to be more confident in their decisions to decline life-sustaining treatments. Those in the comparison group selected “it depends” 50% of the time (i.e., 24 of 48 unique decisions), whereas those in the NSACP group selected “it depends” only 30% of the time (i.e., 55 of 180 total decisions). In the NSACP group, approximately 50% of the decisions documented were to decline the treatment described (i.e., 48% [87 of 180 decisions]). Interestingly, preferences for initiating life-sustaining treatment were similar across groups (21% and 23%, respectively) (Table 4).


Initial findings from the current feasibility study suggest that this single-session NSACP intervention has potential as an efficient, nurse-delivered intervention for supporting individuals in the completion of an AD. It is promising that newly trained RNs achieved satisfaction ratings that are comparable to the more experienced social workers who had been providing advance care planning support for an average of 10 years. Information about RBA of life-sustaining treatment decisions are folded into the discussion—meeting a need expressed by approximately 80% of participants in the current sample. The manual offers a standardized approach with built-in opportunities for participant questions and discussion, and step-by-step guidance through completion of the AD within 1 hour. The high AD completion rate observed for the NSACP intervention group suggests that this intervention may be an effective vehicle for advance care planning discussions by RNs. And perhaps most important is the observed pattern in the decisions made across groups on the completed AD forms (Table 4), which may indicate increased safety and comfort when completing an AD—consistent with the Theory for Enabling Safety—without dissuading those who want to initiate life-sustaining treatment.

The current feasibility study was a necessary initial step for the novel intervention developed, but conclusions can not be made about safety, efficacy, effectiveness, or cost (Leon, Davis, & Kraemer, 2010). More sophisticated cost analysis data should be collected before conclusions are drawn. Acceptable feasibility was supported, as evidenced by meeting or exceeding each predetermined element for the intervention and design. The high enrollment rate was likely related to the brief discussion with each interested Veteran before proceeding to an informed consent visit. The opportunity to hear about eligibility, the steps of the study, and compensation details allowed uninterested Veterans to decline immediately. The high randomization rate was likely related to the immediacy of eligibility determination and group assignment after same-day consent. The high retention rates likely were achieved because of the brevity of the intervention and “warm hand-offs” of the participant to the interventionist. In addition, the study had: (a) a tightly contained duration (maximum of 90 days for each participant), (b) enrolled those who were interested in AD information or help, and (c) provided monetary compensation.

An understanding of the benefits and burdens of life-sustaining treatments is necessary for optimal advance care planning (Skår et al., 2014). The current study complements other important work in gerontological nursing focused on family decision making and agreement between surrogates and patients on goals of care (Patient-Centered Advance Care Planning [PC-ACP], Briggs, Kirchhoff, Hammes, Song, & Colvin, 2004; Sharing Patients' Illness Representations to Increase Trust [SPIRIT], Song, Metzger, & Ward, 2017). Although there is nothing preventing a surrogate decision maker from participating in the NSACP session, this intervention is designed for a more targeted one-on-one dialogue between the nurse and patient. The manual maps onto the VA AD form, ensuring understanding of each element included. The intervention aligns well with the VHA Life-Sustaining Treatment Decisions Initiative, which is a national quality improvement project led by the VHA National Center for Ethics in Health Care (VA, 2017). This initiative provides standards, tools, resources, and education to support nurses and other providers in discussing and documenting goals of care and life-sustaining treatment decisions with high-risk patients. This VHA initiative promotes patient-centered care that honors patients' values, goals, and preferences related to life-sustaining treatments and is consistent with the adapted Theory for Enabling Safety framework guiding the intervention. The intervention is also well-suited for adaptation to other health care settings where discussions are likely to occur one-on-one, and for patients without available surrogates who may be at increased risk of being unrepresented or unbefriended (Pope, 2013).

Research Implications

Several considerations for future research emerged. First, decisional conflict, confidence in decisions, or knowledge of life-sustaining treatments were not measured. Future studies could measure these items directly, which may shed light on participants' rationale for declining life-sustaining treatments more consistently than those in the comparison condition. Second, the current study included a heterogeneous group of Veteran participants. Future studies could examine the NSACP intervention in specific subgroups who are likely to need AD documents, such as those with end-stage renal disease, congestive heart failure, or chronic obstructive pulmonary disease, as is common in advance care planning research (Abdul-Razzak et al., 2017; Briggs et al., 2004).

Finally, selection of an appropriate comparison condition should be weighed carefully. For the current feasibility study, the researchers were interested in knowing how many individuals desired additional information about the treatment decisions. Use of an untouched “usual care” condition without influencing whether participants had access to desired medical information was considered. However, because decisions made during the study could impact clinical decision making in the lives of participants (i.e., they completed real forms that were scanned into their electronic health records), the investigator team felt an ethical obligation to facilitate access to requested information in both groups.

A number of participants in the comparison condition discontinued their participation because of not following through with their PCP. Based on informal information collected during the study, it appears that it is rare for these discussions with PCPs to occur around the time of AD completion. This is not the first study to note that coordination with providers can be a critical step in successful AD completion (Murphy et al., 2000). In the current study, barriers included: (a) patients without specific questions and providers who were unprepared to lead such a discussion; (b) providers who offered to accommodate the request at the next appointment, which might not be for 3 to 6 months; and (c) PCPs at other facilities or in the community without electronic communication supports available. Positioning the NSACP intervention with a RN case manager within the VA's patient aligned care team would be one way to coordinate care closely with PCPs without relying exclusively on the availability of the physician or nurse practitioner.

Finally, the current authors believe nurses are uniquely positioned to guide patients in advance care planning discussions and recommend caution when transferring the NSACP intervention to other professionals to administer in future clinical trials. Many nurses have experience with patients as they are resuscitated, intubated, placed on respirators, dialyzed for kidney failure, and endure multiple hospitalizations—each with associated emotional and physical risks. Nurses are with patients 24/7. They provide support during the preparation, anticipation, procedures themselves, and consequences—both positive and disappointing—of those procedures. This is the history of nursing; it is at the very heart of patient-centered gerontological nursing.


The NSACP intervention warrants further testing. The results of this feasibility study suggest that (a) recruitment, randomization, retention, assessment, and data collection are feasible; (b) the NSACP intervention may impact the types of decisions made; and (c) the AD completion rate and satisfaction scores for the NSACP intervention are promising. Although it is premature to make direct comparisons between groups, the NSACP intervention was associated with an AD completion rate three times that of the comparison condition (94% vs. 29%). The intervention was completed in <1 hour, cost less than $28 to deliver by a RN, and addressed sensitive topics (e.g., RBA associated with life-sustaining treatments). The NSACP intervention has the potential to add to the growing literature that is positively shaping and expanding nurses' roles in guiding advance care planning discussions in clinical practice.


  • Abdul-Razzak, A., Heyland, D.K., Simon, J., Ghosh, S., Day, A.G. & You, J.J. (2017). Patient-family agreement on values and preferences for life-sustaining treatment: Results of a multicentre observational study. BMJ Supportive & Palliative Care. Advance online publication. doi:10.1136/bmjspcare-2016-001284 [CrossRef]
  • Attkisson, C.C. (2012). The Client Satisfaction Questionnaire (CSQ) scales reprint portfolio. Retrieved from
  • Bastable, S. (2009). Nurse as educator: Principles of teaching and learning for nursing practice (3rd ed.). Sudbury, MA: Jones and Bartlett Learning.
  • Bowen, D.J., Kreuter, M., Spring, B., Cofta-Woerpel, L., Linnan, L., Weiner, D. & Fernandez, M. (2009). How we design feasibility studies. American Journal of Preventive Medicine, 36, 452–457. doi:10.1016/j.amepre.2009.02.002 [CrossRef]
  • Briggs, L.A., Kirchhoff, K.T., Hammes, B.J., Song, M.K. & Colvin, E.R. (2004). Patient-centered advance care planning in special patient populations: A pilot study. Journal of Professional Nursing, 20, 47–58. doi:10.1016/j.profnurs.2003.12.001 [CrossRef]
  • Chan, H.Y. & Pang, S.M. (2010). Let me talk: An advance care planning programme for frail nursing home residents. Journal of Clinical Nursing, 19, 3073–3084. doi:10.1111/j.1365-2702.2010.03353.x [CrossRef]
  • Clever, M. & Segal, D.R. (2013). The demographics of military children and families. The Future of Children, 23(2), 13–39. doi:10.1353/foc.2013.0018 [CrossRef]
  • Dubbert, P.M., Garner, K.K., Lensing, S., White, J.G. & Sullivan, D.H. (2017). Engagement in steps of advance health care planning by homeless veterans. Psychological Services, 14, 214–220. doi:10.1037/ser0000147 [CrossRef]
  • Dumville, J.C., Hahn, S., Miles, J.N. & Torgerson, D.J. (2006). The use of unequal randomisation ratios in clinical trials: A review. Contemporary Clinical Trials, 27(1), 1–12. doi:10.1016/j.cct.2005.08.003 [CrossRef]
  • Freeman, S.L. & Berger, A.M. (2009). Nebraska Veterans' preferences for end-of-life care. Clinical Journal of Oncology Nursing, 13, 399–403. doi:10.1188/09.CJON.399-403 [CrossRef]
  • Green, L.W. & Glasgow, R.E. (2006). Evaluating the relevance, generalization, and applicability of research: Issues in external validation and translation methodology. Evaluation & the Health Professions, 29, 126–153. doi:10.1177/0163278705284445 [CrossRef]
  • Holliday, S. (2009). Equipping patients for a time of helplessness: An educational intervention. Australian Journal of Rural Health, 17, 232–235. doi:10.1111/j.1440-1584.2009.01083.x [CrossRef]
  • Horne, G., Seymour, J. & Shepherd, K. (2006). Advance care planning for patients with inoperable lung cancer. International Journal of Palliative Nursing, 12, 172–178. doi:10.12968/ijpn.2006.12.4.21014 [CrossRef]
  • Howe, A., Keogh-Brown, M., Miles, S. & Bachmann, M. (2007). Expert consensus on contamination in educational trials elicited by a Delphi exercise. Medical Education, 41, 196–204. doi:10.1111/j.1365-2929.2006.02674.x [CrossRef]
  • Kelty, R., Kleykamp, M. & Segal, D.R. (2010). The military and the transition to adulthood. The Future of Children, 20, 181–207. doi:10.1353/foc.0.0045 [CrossRef]
  • Kirschner, K.L. (2005). When written advance directives are not enough. Clinics in Geriatric Medicine, 21, 193–209. doi:10.1016/j.cger.2004.08.006 [CrossRef]
  • Leon, A.C., Davis, L.L. & Kraemer, H.C. (2011). The role and interpretation of pilot studies in clinical research. Journal of Psychiatric Research, 45, 626–629. doi:10.1016/j.jpsychires.2010.10.008 [CrossRef]
  • Lipson, A.R., Hausman, A.J., Higgins, P.A. & Burant, C.J. (2004). Knowledge, attitudes, and predictors of advance directive discussions of registered nurses. Western Journal of Nursing Research, 26, 784–796. doi:10.1177/0193945904266528 [CrossRef]
  • Mahaney-Price, A.F., Hilgeman, M.M., Davis, L.L., McNeal, S.F., Conner, C.M. & Allen, R.S. (2014). Living will status and desire for living will help among rural Alabama veterans. Research in Nursing and Health, 37, 379–390. doi:10.1002/nur.21617 [CrossRef]
  • Medvene, L.J., Wescott, J.V., Huckstadt, A., Ludlum, J., Langel, S., Mick, K. & Base, M. (2003). Promoting signing of advance directives in faith communities. Journal of General Internal Medicine, 18, 914–920. doi:10.1046/j.1525-1497.2003.20351.x [CrossRef]
  • Molloy, D.W., Guyatt, G.H., Russo, R., Goeree, R., O'Brien, B.J., Bédard, M. & Dubois, S. (2000). Systematic implementation of an advance directive program in nursing homes: A randomized controlled trial. JAMA, 283, 1437–1444. doi:10.1001/jama.283.11.1437 [CrossRef]
  • Murphy, P., Kreling, B., Kathryn, E., Stevens, M., Lynn, J. & Dulac, J. (2000). Description of the SUPPORT intervention: Study to understand prognoses and preferences for outcomes and risks of treatment. Journal of the American Geriatrics Society, 48(Suppl. 5), S154–S161. doi:10.1111/j.1532-5415.2000.tb03125.x [CrossRef]
  • Öhlén, J., Andershed, B., Berg, C., Frid, I., Palm, C.A., Ternestedt, B.M. & Segesten, K. (2007). Relatives in end-of-life care—Part 2: A theory for enabling safety. Journal of Clinical Nursing, 16, 382–390. doi:10.1111/j.1365-2702.2006.01474.x [CrossRef]
  • Perkins, H.S. (2007). Controlling death: The false promise of advance directives. Annals of Internal Medicine, 147, 51–57. doi:10.7326/0003-4819-147-1-200707030-00008 [CrossRef]
  • Pope, T.M. (2013). Making medical decisions for patients without surrogates. New England Journal of Medicine, 369, 1976–1978. doi:10.1056/NEJMp1308197 [CrossRef]
  • Pyle, J.J. (2015). A motivational interviewing education intervention for home healthcare nurses. Home Healthcare Now, 33(2), 79–83. doi:10.1097/NHH.0000000000000184 [CrossRef]
  • Rao, J.K., Anderson, L.A., Feng-Chang, L. & Laux, J.P. (2014). Completion of advance directives among U.S. consumers. American Journal of Preventive Medicine, 46, 65–70. doi:10.1016/j.amepre.2013.09.008 [CrossRef]
  • Silveira, M.J., Wiitala, W. & Piette, J. (2014). Advance directive completion by elderly Americans: A decade of change. Journal of the American Geriatrics Society, 62, 706–710. doi:10.1111/jgs.12736 [CrossRef]
  • Skår, Å., Juvet, L., Smedslund, G., Bahus, M.K., Pedersen, R. & Fure, B. (2014). End-of-life care—How to find the appropriate level and intensity of medical treatment of seriously ill and dying patients. Retrieved from
  • Song, M.K., Metzger, M. & Ward, S.E. (2017). Process and impact of an advance care planning intervention evaluated by bereaved surrogate decision-makers of dialysis patients. Palliative Medicine, 31, 267–274. doi:10.1177/0269216316652012 [CrossRef]
  • Song, M.K., Ward, S.E., Happ, M.B., Piraino, B., Donovan, H.S., Shields, A.M. & Connolly, M.C. (2009). Randomized controlled trial of SPIRIT: An effective approach to preparing African-American dialysis patients and families for end of life. Research in Nursing & Health, 32, 260–273. doi:10.1002/nur.20320 [CrossRef]
  • Strupei, S., Buß, A. & Dassen, T. (2013). Effectiveness of nurse-delivered patient education interventions on quality of life in outpatients: A systematic review. Applied Nursing Research, 26, 232–238. doi:10.1016/j.apnr.2013.08.004 [CrossRef]
  • Sugarman, J., Weinberger, M. & Samsa, G. (1992). Factors associated with Veterans' decisions about living wills. Archives of Internal Medicine, 152, 343–347. doi:10.1001/archinte.1992.00400140089020 [CrossRef]
  • U.S. Department of Health and Human Services. (2008, August1). Advance directives and advance care planning: Report to Congress. Retrieved from
  • U.S. Department of Veterans Affairs. (n.d.). Veteran population. Retrieved from
  • U.S. Department of Veterans Affairs. (2012). VA advance directive: Durable power of attorney for health care and living will. Retrieved from
  • U.S. Department of Veterans Affairs. (2017). Life-sustaining treatment decisions initiative. Retrieved from
  • Ware, J.E. Jr.. , Kosinski, M. & Keller, S.D. (1996). A 12-item short-form health survey: Construction of scales and preliminary tests of reliability and validity. Medical Care, 34, 220–233. doi:10.1097/00005650-199603000-00003 [CrossRef]
  • West, S.G. & Thoemmes, F. (2008). Equating groups. In Alasuutari, P., Bickman, L. & Brannen, J., (Eds.), The SAGE handbook of social research methods (pp. 414–430). Los Angeles, CA: Sage. doi:10.4135/9781446212165.n24 [CrossRef]

Participant Characteristics

Characteristicn (%)χ2dfp Value
NSACP Group (n = 33)Comparison Group (n = 17)Total (N = 50)
  Male26 (79)15 (88)41 (82)
  Female7 (21)2 (12)9 (18)
  Black non-Hispanic22 (67)9 (53)31 (62)
  White non-Hispanic8 (24)5 (29)13 (26)
  Black Hispanic01 (6)1 (2)
  Two or more races non-Hispanic3 (9)1 (6)4 (8)
  Other: Israeli non-Hispanic01 (6)1 (2)
Marital status3.2540.52
  Divorced14 (42)7 (41)21 (42)
  Separated7 (21)1 (6)8 (16)
  Married6 (18)5 (29)11 (22)
  Never married5 (15)4 (24)9 (18)
  Widowed1 (3)01 (2)
Educational level5.3170.62
  Less than GED2 (6)1 (6)3 (6)
  GED2 (6)1 (6)3 (6)
  High school diploma9 (27)3 (18)12 (24)
  Some tech school or college11 (33)7 (41)18 (36)
  Tech school/associate degree5 (15)1 (6)6 (12)
  College degree2 (6)1 (6)3 (6)
  Some postgraduate02 (12)2 (4)
  Postgraduate degree2 (6)1 (6)3 (6)
Incomea (U.S. dollars)11.5850.04*
  None9 (28)1 (6)10 (20)
  ≤ 19,99913 (41)9 (56)22 (44)
  20,000 to 39,9997 (22)1 (6)8 (16)
  40,000 to 59,99903 (19)3 (6)
  60,000 to 99,9992 (6)2 (13)4 (8)
  100,000 to 149,9991 (3)01 (2)
Sub-setting of care4.3040.37
  Residential treatment19 (58)12 (71)31 (62)
  Outpatient clinic12 (36)3 (18)15 (30)
  Selma Rural Health Clinic1 (3)01 (2)
  Rural mobile clinic1 (3)1 (6)2 (4)
  Home-based primary care01 (6)1 (2)
Prior ACP experienceb1.8910.17
  None31 (94)14 (88)45 (90)
  One or more sessions with physician, NP, or other2 (6)2 (12)4 (8)
Is home zip code rural?0.2410.63
  No19 (58)11 (65)30 (60)
  Yes14 (42)6 (35)20 (40)

Comparison of Age and Sf-12 Health Characteristics at Baseline (N = 50)

CharacteristicMean (SD)tdfp Value
NSACP Group (n = 33)Comparison Group (n =17)
Age (years)51.24 (11.48)48.35 (14.95)0.76480.45
SF-12 scoresa
  PCS42.06 (11.67)38.37 (9.39)1.13480.27
  MCS47.74 (12.37)42.79 (11.13)1.38480.17

Frequency of Yes Responses for Information or Help at Baseline (N = 50)

Do you want…n (%)χ2dfp Value
NSACP Group (n = 33)Comparison Group (n = 17)
1. information about advance directives?29 (88)14 (82)0.2910.59
2. help completing an advance directive?a29 (91)12 (71)3.2610.07
3. information about risks, benefits, and alternatives of common life-sustaining medical treatments and procedures to help you make informed decisions?29 (88)10 (59)5.5210.02*

Advance Directive Decisions by Group (N = 35a)

Scenario% (n)
NSACP Group (n = 30)Comparison Group (n = 5)
YesNoIt DependsYesNoIt Depends
1. If I am unconscious, in a coma, or in a vegetative state and there is little or no chance of recovery.13 (4)43 (13)43 (13)80 (4)20 (1)0
2. If I have permanent, severe brain damage that makes me unable to recognize my family or friends (e.g., severe dementia).10 (3)50 (15)40 (12)20 (1)20 (1)60 (3)
3. If I have a permanent condition where other people must help me with my daily needs (e.g., eating, bathing, toileting).43 (13)23 (7)33 (10)40 (2)20 (1)40 (2)
4. If I need to use a breathing machine and be in bed for the rest of my life.20 (6)60 (18)20 (6)40 (2)60 (3)0
5. If I have pain or other severe symptoms that cause suffering and can't be relieved.27 (8)47 (14)27 (8)40 (2)60 (3)0
6. If I have a condition that will make me die very soon, even with life-sustaining treatments.13 (4)67 (20)20 (6)080 (4)20 (1)
Summary across all six scenariosb21 (38)48 (87)31 (55)23 (11)27 (13)50 (24)

Dr. Hilgeman is Clinical Psychologist, Ms. Collins is Health Science Specialist, Dr. Davis is Associate Chief of Staff for Research, Dr. Gordon is Quality Management, Ms. Coleman is Nurse Manager, and Ms. Gay is Health Science Specialist, VA Medical Center, Tuscaloosa, Alabama; Dr. Uphold is Associate Director of Implementation and Outcomes Research, North Florida/South Georgia Healthcare System, Geriatric Research Education and Clinical Center, and Associate Professor, Department of Aging and Geriatrics Research, College of Medicine, University of Florida, Gainesville, Florida; Dr. Olsen is Associate Professor, Michigan State University College of Nursing, East Lansing, Michigan; Dr. Burgio is Associate Director for Research, Birmingham/Atlanta Geriatric Research, Education, and Clinical Center, Birmingham VA Medical Center, and Professor of Medicine, University of Alabama at Birmingham School of Medicine, Birmingham, Alabama; Dr. DeCoster is Assistant Research Professor, Center for Advanced Study of Teaching and Learning, University of Virginia, Charlottesville, Virginia. Dr. Hilgeman is also Associate Adjunct Professor, and Dr. Allen is Professor of Psychology, Alabama Research Institute on Aging, and Department of Psychology, University of Alabama, Tuscaloosa, Alabama.

The authors have disclosed no potential conflicts of interest, financial or otherwise. This work was funded by the United States Department of Veterans Affairs (VA), Veterans Health Administration, Office of Research and Development, Health Services Research and Development, Nursing Research Initiative. The funding organization had no role in any part of the study. This study was approved by the Tuscaloosa VA Medical Center Institutional Review Board. The views in this article are those of the authors and do not necessarily reflect the views of the VA. This study was registered at, Identifier: NCT01746368.

The authors acknowledge the significant contributions of the principal investigator of this funded work, Ann F. Mahaney-Price, who obtained the funding; led the conception, design, and data collection; and supervised the intervention delivery and the study execution before her retirement. She wrote early drafts of the current article but has fully transitioned to pursuing other interests and chose not to be an author on the current article. The authors are also grateful for the support of Maria Andrews, MS, FACHE, and Alan Tyler, MS, MPA, FACHE, former Tuscaloosa VA Medical Center Directors. The authors thank Marietta Stanton, PhD, University of Alabama, for her contribution in cultural competency training of the professional nurse interventionists for this study. Finally, the authors thank the social workers and nurse interventionists at the Tuscaloosa VA Medical Center who generously donated their time and contributions to this project, including Ms. Anastasia Reddick and Ms. Cheryl Dickey.

Address correspondence to Michelle M. Hilgeman, PhD, Clinical Psychologist, VA Medical Center, 3701 Loop Road, TVAMC Research (151), Tuscaloosa, AL 35404; e-mail:

Received: August 21, 2017
Accepted: April 23, 2018


Sign up to receive

Journal E-contents