Older Adults Newly Diagnosed with Symptomatic Myeloma Want to Participate in Treatment Decision Making

Abstract

Purpose/Objectives

The purpose of the study was to describe the preferences for participation in decision making of older patients newly diagnosed with symptomatic myeloma and to explore the association between sociodemographic variables and decisional role preferences.

Design

Descriptive, cross sectional design

Setting

Subjects’ homes and two large academic cancer centers.

Sample

The convenience sample consisted of 20 older adults (60 years of age and above) with symptomatic myeloma diagnosed within the past 6 months.

Methods

The Control Preferences Scale was administered followed by an in-person one-time semi-structured interview.

Main Research Variables

Role preferences for participation in treatment decision-making, age, gender, race, work status, personal relationship status, education, and income.

Findings

55% (n=11) of the subjects had preferred a shared role with the physician and 40% (n=8) had preferred to make the decisions after seriously considering the opinion of their physicians. Only one subject preferred to leave the decision to the doctor as long as the doctor considered the patient’s treatment preferences. Sociodemographic characteristics had no impact on preferences for participation in treatment decision-making.

Conclusions

The study findings indicate that older adults newly diagnosed with myeloma wanted to participate during treatment decision-making. Oncology nurses must respect the patient's desired role preference and oncology clinicians must listen to the patient and allow them to be autonomous in making treatment decisions if the patient so desire such control in the decision-making process. A culture of equipoise between the patient and the clinician during TDM must be cultivated in order to achieve the patient's desired level of participation. More studies that focus on supporting and involving patients diagnosed with myeloma in the decision-making process are needed in order to influence clinical practice and policy.

Practice Implications

Nurses and other oncology clinicians can elicit patient’s preferred level of participation in treatment decision-making. Oncology nurses can do the following nursing actions: make sure patients receive disease and treatment-related information; encourage patients to express their decisional role preference to the physician; develop a culture of mutual respect and value the patient's desire for autonomy for treatment decision making; acknowledge that the right to make a treatment choice belongs to the patient; 5) provide psychological support to the patient during treatment decision making throughout the care continuum.

In the past three decades, treatment decision making (TDM) studies have focused on decisional control preferences, with most studies conducted in breast and prostate cancer samples. A systematic review of decisional role preferences among cancer patients has shown an increasing trend of patients being interested in more participation during TDM (Tariman, Berry, Cochrane, Doorenbos, & Schepp, 2010). In order to meet and facilitate the patient’s preferred level of participation during TDM, interventional studies geared toward increasing a patient’s decisional satisfaction, reducing decisional conflict, and preventing anxiety and depression related to TDM have been steadily increasing in numbers (Allen et al., 2010; Caldon et al., 2010; Evans et al., 2010) with some studies targeting the elderly cancer patient population (Lewis et al., 2010; van Tol-Geerdink et al., 2008).

The growing interest in direct assessment of patient preferences in terms of control of decision making and changing landscape of information needs priorities (Beaver & Booth, 2007; Denberg, Melhado, & Steiner, 2006; Flynn, Smith, & Vanness, 2006; Mancini et al., 2007; Sabo, St-Jacques, & Rayson, 2007) coupled with a predicted shift away from paternalistic decision as baby boomers age (Pipe et al., 2005), present substantial opportunities for improving patient care and clinical outcomes in the area of treatment decision making. This is particularly true in the older adult cancer patient population, that is often understudied or underrepresented in clinical trials.

Increased patient participation in TDM has been associated with positive clinical outcomes such as a greater level of patient satisfaction with decisions and better psychological adjustment (i.e., less post-decision anxiety and depression) in patients (Gaston & Mitchell, 2005; Gattellari, Butow, & Tattersall, 2001). However, the preferred level of participation in TDM for older patients newly diagnosed with symptomatic myeloma (myeloma that requires immediate therapy) has not been previously studied. Moreover, the influence of sociodemographic factors in older myeloma patients’ level of participation during TDM is also unknown.

There are very few nursing research studies involving the MM patient population. With the rapidly growing number of treatment options for MM patients, this treatment decision making study is well suited for this patient population. Research findings from this study will help nurses in guiding their myeloma care practice.

Theoretical Framework

Degner and Beaton’s (1987) Patterns of Decision Making provided the theoretical framework for this study. These patterns encompass the various levels of patient participation in decision making and are patient-centered, directly eliciting decisional role preferences from the patient’s perspective. According to the model, four decision-making patterns may occur: a provider-controlled decision-making pattern emerges when patients decline to become involved in selecting their own treatment, even when urged to do so by the physician. In this case, the patient is essentially saying: It’s up to you, doctor; you’re the expert. On the other hand, a patient-controlled decision-making pattern occurs when patients make it clear that they are the ones who will make the decisions (Degner & Beaton, 1987). When patients want discussion of options with their physician and think about the options prior to making the final decision with their physician, a jointly-controlled decision-making pattern occurs. Finally, when patients are incapable of making treatment decisions and the family makes decisions for them, a family-controlled decision-making pattern emerges (Degner & Beaton, 1987). There were no cases of family controlled decision-making in the current study since the sample criteria excluded anyone in this category.

Table 1 illustrates the various roles patients can play during treatment decision making. This framework of decisional role patterns was developed based on a 4-year qualitative study into decision-making roles in life-threatening situations such as cancer (Degner & Beaton, 1987). The Control Preferences Scale (CPS) is a measure of decision-making preferences developed from a qualitative study by Degner and colleagues (Degner, Sloan, & Venkatesh, 1997).

Table 1.

Degner and Beaton’s Pattern of Decision Making (Degner, Sloan, & Venkatesh, 1997) (The Control Preferences Scale)

Active – Patient-Controlled ○I prefer to make the final treatment decision (Card A). ○I prefer to make the final treatment decision after seriously considering my doctor’s opinion (Card B).

Collaborative – Jointly controlled ○I prefer that my doctor and I share responsibility for deciding which treatment is best (Card C).

Passive – Provider-controlled ○I prefer my doctor to make the final treatment decision, but only after my doctor has seriously considered my opinion (Card D). ○I prefer to leave all treatment decisions to my doctor (Card E).

Family-Controlled ○The family makes final decisions for the patient, who is incapacitated.

The purpose of the study was to describe the preferences for participation in decision making of patients newly diagnosed with multiple myeloma and to explore the association between sociodemographic variables and decisional role preferences.

Methods

Design and sample

A descriptive, cross-sectional study was conducted involving administration of the Control Preferences Scale (CPS) followed by a one-time, semi-structured interview. The convenience sample consisted of 20 older adults referred to the Seattle Cancer Care Alliance (SCCA) or the Northwestern University Myeloma Program (NUMP) by several hematologists/oncologists in the greater Seattle or Chicago areas, respectively. Eligibility criteria included older adults who were (a) 60 years of age and above, (b) diagnosed with symptomatic myeloma within the past six months, (c) able to read and write English, and (d) able to give informed consent.

Instrument

The CPS is a measure of decisional role preferences using a card-sort technique that has two sets of five cards each. Each card describes a different role in decision making and is illustrated with a sketch of characters (physician/patient) representing their different roles in decision making. The first set of five cards illustrates possible roles that the patient could assume, ranging from the patient selecting his or her own treatment (cards A and B) through a collaborative role model (card C) to a scenario where the physician makes the decision (cards D and E). The process of administering the card sort was described extensively by Degner, Sloan, and Venkatesh (1997).

The CPS offers a simple and fast method to elicit a patient’s decisional role preference (Degner, Sloan, & Venkatesh, 1997). This scale has been found to be a valid tool in the measurement of decisional role preferences in patients newly diagnosed with various types of cancer (Degner & Sloan, 1992). Permission to use the CPS scale for this study was obtained prior to the study design.

Recruitment procedures

After the researchers obtained approval from the University of Washington (UW) and Northwestern University (NU) Human Subjects Division, older adults recently diagnosed with symptomatic myeloma were recruited to participate in the study. The researcher made every effort to recruit from both university- and community-based practices in order to include a diversity of study subjects. Adults aged 60 and older who had an appointment at the hematology or transplant service of SCCA or NUMP and were found to be eligible for the study were recruited by mail using a recruitment flyer. Older patients diagnosed with asymptomatic myeloma disease not requiring therapy were not recruited in the study because treatment decisions are not needed at the time of diagnosis. At NUMP, one of the researchers personally approached potential subjects and introduced the study to gauge patient interest in participation based on IRB-approved protocol. A review of clinic schedules at SCCA and NUMP was conducted weekly to identify potential study subjects. One other UW-affiliated community clinic was checked weekly for potential study subjects.

The CPS was administered and an interview conducted in either the patients' homes or a designated research-related conference rooms at SCCA and NUMP. These rooms were strictly assigned for research use only and met the human subject division’s standard for patient privacy. If a patient wanted the interview to be conducted at a later time, a one-week period following the clinic appointment was allowed for re-scheduling. Moreover, if a patient wanted the interview to be conducted at his/her home, the researcher conducted the interview at the subject’s home, as requested. Patients were also asked to describe the decisional role that they preferred from the CPS card. Interviews were audio-recorded and transcribed verbatim by professional transcriptionists and verified by the principal investigator against the actual recording. The study subjects received $5 Tully's gift certificate immediately after completing the interview schedule.

Analysis

Coombs’ Unfolding Analysis

Data from the CPS scale were analyzed using the Statistical Analysis System^® computer program syntax developed by Sloan and Yeung (Sloan, 1994; Sloan & Yeung, 1994). This analysis was based on a scaling model developed by Coombs, termed the unfolding theory (Coombs, 1964). According to Degner, Sloan, & Venkatesh (1997), this psychological scaling model is based on the assumption that “an individual’s preference corresponds to an ideal point on a continuum, and that this ideal point can be derived by presenting successive paired comparisons of stimuli that fall along the continuum (p.25).” The reliability of the scale is established when 50% plus one of the experimental subjects’ preferences fall along the hypothesized scale. In this study, ABCDE and DCBAE competing scales met the reliability criterion of >50% of subjects belonging to the hypothesized scale as seen on Table 3.

Table 3.

Rank Ordering of the Two Competing Scale Models

CONTROL PREFERENCES SCALING
SUMMARY FOR SCORING
Scale NAME	VALID	VALPER	INVALID	INVALPER	CELL	REVERSAL
ABCDE	14	70	6	30	3	N
DCBAE	14	70	6	30	6	N
CBADE	9	45	11	55	8	N
ACBDE	7	35	13	65	8	N
DBCAE	7	35	13	65	8	N
ADCBE	6	30	14	70	8	N
BCADE	6	30	14	70	9	N
BCDAE	6	30	14	70	8	N
DABCE	5	25	15	75	9	N
DACBE	4	20	16	80	10	N

Note: This table shows the top two competing TDM scale models that meet Coombs' reliability criterion of 50% plus 1 (valid permutation greater than 11).

Coombs' unfolding analysis was based on a scaling model developed by Coombs, termed the unfolding theory (Coombs, 1964). The unfolding theory holds that for any given hypothesized scale, only 11 subsets of the 120 possible permutations of the five decisional role cards will be transitive. These 11 transitive responses range from ABCDE to EDCBA and they include ABCDE, BACDE, BCADE, BCDAE, CBDAE, CDBAE, CDBEA, CDEBA, DCEBA, DECBA and EDCBA. “Transitivity” means that subject’s response falls within the 11 possible valid permutations and the subject’s preference for each of the paired comparisons of stimuli is consistent with the hypothesized A to E psychological continuum. The reliability of the scale is established when 50% plus one of the experimental subjects’ preferences fall along the hypothesized scale (i.e. ABCDE metric). One could easily see that a patient's logic on taking control of the decision making process falls off the hypothesized scale when the order presentation of the card that the patient chooses is BCDEA or BCEDA as shown in Figure 1. In other words, card B represents active decision role as well as card A, and when these 2 card options are at extreme ends of the card sort arrangement, the respondents clearly did not understand the logic of taking control of the decision making process.

Figure 1.

Distribution of Respondents On and Off of the Hypothesized ABCDE Decision Making Scale

Association Analyses

The differences in decisional role preferences were examined using dichotomous categories of gender (male versus female), age (<70 years versus ≥70 years), education (less than 4 years of college versus ≥4 years of college), marital status (single versus married), income (≤$55,000 versus >$55,000), and work status (retired versus non-retired). The comparisons were made using simple analysis of variance (ANOVA). As often as possible, the cut-off for each category was based on equal distribution of the number of subjects on each category. Spearman rank-order correlations were used to determine the relationship between ordinal CPS score and the respondent’s age, education, and income variables. Given the results of the bivariate analysis and the small sample size, multivariable analyses were deemed unwarranted by our statisticians.

Triangulation of Qualitative and Quantitative Data

Across-method triangulation (Waltz, Strickland, & Lenz, 2005), a form of methods triangulation was employed for cross validation of the data obtained pertaining to patient’s decisional role preferences. The subjects’ verbal descriptions of their desired level of participation as elicited using the interview and the patients’ preferences for participation, as measured by the CPS card sort, are compared. This approach collected rich detailed information from the subjects regarding their perspectives on decisional role preferences which were then compared to the original description of the CPS cards.

Results

Seventy-nine potential subjects were contacted by mail at SCCA from October 2009 through July 2010. Of these, 14 (17.7 %) subjects responded and all participated in the study. At NUMP, the researcher identified six potential subjects and one of the investigators approached them in person about participation, which was allowed by NU IRB review committee. All six potential subjects agreed to participate. Informed consent was obtained from all study subjects. Fourteen out of twenty interviews were conducted in the subjects’ homes.

Table 2 presents sociodemographic characteristics of the study subjects. The subjects’ mean age was 67.45 years (age range: 60–82 years). The majority of subjects in this sample were Caucasian (90%); female (60%), married (60%), retired (65%), had at least a 2-year college education (75%) and had an income >$35,001 (75%).

Table 2.

Sociodemographic Characteristics of the Sample

Variable	N	Percentage
Age (mean, 67.45 years)
60–70	14	70
71–82	6	30
Gender
Male	8	40
Female	12	60
Race
White	18	90
Asian	1	5
American Indian	1	5
Work Status
Full time	2	10
Working on medical leave	2	10
Not working	2	10
Retired	13	65
Student	1	5
Personal Relationship Status
Single	2	10
Married or partnered	12	60
Divorced	5	25
Widowed	1	5
Highest Level of Education
9th – 12th grade	5	25
2 years of college	2	10
4 years of college	10	50
Graduate degree	3	15
Annual Household Income
$18,000 or less	3	15
$18,001 to $35,000	2	10
$35,001 to $55,000	5	25
$55,001 to $85,000	5	25
$85,001 and above	5	25

Unfolding Analysis

As illustrated in Table 3, the hypothesized ABCDE decision making scale comprised more than 50% of the respondents’ answers (14 out of 20 subjects). This means that the data show support for an underlying dominant dimension of control, ranging from keeping control (active - cards A and B) through collaboration (sharing- card C) to giving away control (passive-cards D and E). Moreover, these results show that the 50% plus 1 criterion of reliability had been met. Interestingly, the DCBAE metric also had more than 50% of the respondents’ answers (14 out of 20 subjects). This means that a second, competing model of dichotomous preference (shared, card C, and active, cards B and A) is seen in this group of older adults newly diagnosed with symptomatic myeloma. All other competing scales had 9 subjects or less out of 20 subjects, and therefore did not meet the eligibility criterion for validity of underlying theory of varying degree of control preferences. Figure 1 shows the distribution of respondents’ preferences on and off the hypothesized decision making control scale.

Decisional Role Preferences

An examination of the distribution of preferences based on the first card in the preference order indicated that 55% (n=11) of the subjects preferred a shared role with the physician and that 40% (n=8) of the subjects preferred to make the decisions after seriously considering the opinion of their physician (see Table 4). Only one subject preferred to leave the decision to the doctor as long as the doctor considered the subject’s treatment preference. No individual chose card A or E, the extremes of decision making preference choices. Overall, the percentage of the subjects wanting to have some kind of control over the treatment decision was very high at 95% (n=19).

Table 4.

Distribution of Decisional Role Preferences

Preferred Role	Frequency	Percent	Valid Percent	Cumulative Percent
B	8	40.0	40.0	40.0
C	11	55.0	55.0	95.0
D	1	5.0	5.0	100.0
Total	20	100.0	100.0

Sociodemographic Variables and Decisional Role Preferences

There were no statistically significant differences in decisional role preferences across dichotomous categories of gender (male versus female), age (<70 years versus ≥70 years), education (less than 4 years of college versus greater than 4 years of college), marital status (single versus married), income (≤$55,000 versus >$55,000), and work status (retired versus non-retired) using ANOVA. There were also no statistically significant correlations found between CPS score and the respondent’s age, education, and income variables.

Quantitative and Qualitative Decisional Role Preference

Table 5 illustrates the study subjects decisional role preferences using the CPS card, the decisional category, and the patient’s own description of preferred role. The majority of study subjects’ (n=17; 85%) descriptions of their preferred roles have similar or exact description with the decision categories originally described in the CPS card by Degner and Beaton (1987). Only three (15%) subjects had different personal meaning or interpretation of preferred role when compared to the original CPS description of the three decision categories.

Table 5.

Patient’s Decisional Role Preference, Category, and Description

Subject ID	First Card in CPS	Decisional Category	Patient’s Description of Preferred Decisional Role	Description Matches with Decisional Category
01	B	Active	So it was neutral at first, and then as some of the shock wore off and some of the reality came in, I started participating more in my treatment.	Yes
02	C	Shared	The fact that before treatments are started, I know what it is going to be, and if there is for some reason a drug that I don’t feel I could take, then I still have a right to say no to that.	Yes
03	D	Passive	I definitely want to be involved in the decision, but knowing that the doctor knows more than I do about the treatments that are available and which one I’m best suited for, I would go with the doctor’s opinion after I’ve heard what the options are and discussed them.	Yes
04	B	Active	I would like to have full involvement. I will listen to what the doctor says or what he feels, because I feel he has that knowledge. And I probably would take his recommendation, but I would make the decision myself.	Yes
05	C	Shared	My oncologist seems to think that acupuncture and massage are all fine, but those are things that I’ve explored myself. So I take his advice, and then I do my other kinds of things that are alternative sorts of things, too.	Yes
06	C	Shared	I take the input from what I've gotten back from the tests that my doctor sends me. But that wasn't good enough for me because I wanted a second opinion.	Yes
07	C	Shared	Now that I’ve kind of had a chance to step back and have a more sober, view of it and more objective view, I feel that I’m in a better frame of mind, if you would, to maybe look at the options and discuss this in a more objective manner.	Yes
08	C	Shared	I want to know about things. I'm curious. I want to know as much as I can. And then with the help of my husband and my kids, make a decision.	Yes
09	B	Active	Well, I will make my decision along with my husband at that point on which is best for us as a family, and we rely upon our doctor's medical advice to lead us to a conclusion.	Yes
10	B	Active	I ask her [my doctor] everything I can think of when we meet. I listen to what she has to tell me. If the decision is something clear enough that I can make it immediately or if I need to make it immediately, I do.	Yes
11	C	Shared	I want her [my doctor] to see how I’m doing and see what condition I am and how I’m progressing, you know; worse or better, then, make her decision based on that.	No
12	C	Shared	I want to know. I am very, very nosey that way. I want to know.	No
13	B	Active	My wife was heavily involved in making the decision, too. Since she was very much affected by it, so together we made the decision. I probably had a little more influence than her, but her opinion was considered as well.	Yes
14	C	Shared	Well, my preference is and it has been so far is shared involvement with the doctor and also a key to it has been the ability to be able to get a second opinion, so I can have two experts look at the situation.	Yes
15	B	Active	I write things down and I try to get as many opinions as possible. And of course I think the doctor’s opinion about what to do is 90% of what it is. But I want to understand, if decisions are being made on my behalf, why they're being made.	Yes
16	B	Active	I want to know, I like to understand, really, why something is being done. Not that I truly would understand it as a lay person, but I want to understand the logic for doing something and I do want to understand the potential for being successful and risk factors, you know, because I want to use that to gauge how fast I want to live my life.	No
17	C	Shared	I would like to go over the pros and cons of therapy. I see them in my life for me and then make that decision with the doctor as far as what the doctor feels after the doctor heard where I was coming from. The doctor knows what would be the best for me in this situation because the doctor has the overall picture and I only have pieces.	Yes
18	C	Shared	Well, I like to have the best possible treatment plans to cure my disease, and knowing my doctor, I was confident at that time to listen to her opinion, and we made a decision collectively to further my treatment.	Yes
19	C	Shared	I’d like to be well informed of my choices, on both the pros and cons of those choices, in language I can understand, so that I can help participate in making the choice.	Yes
20	B	Active	Well, we knew that it was incurable, and the medical community nationwide seemed to be taking the transplant approach to improve the possibilities of longevity. So we were impressed with the facility and all of the staff and have decided to go ahead in that direction.	Yes

Discussion

In this study of decisional control preferences in older adults newly diagnosed with symptomatic myeloma, 19 out of 20 subjects indicated a preference for some control or full control of the treatment decisions, with only one subject expressing a preferred passive role. This finding is contrary to previous reports that older adults with various types of cancers such as breast, prostate, and colorectal cancers are passive recipients of medical care (Deber, Kraetschmer, Urowitz, & Sharpe, 2007; Elkin, Kim, Casper, Kissane, & Schrag, 2007; Singh et al., 2010). Perhaps the impact of advanced age on relinquishing decisional control to physicians is moderated by higher education and higher income profile of the study subjects; these are variables previously reported as having strong correlation with more decisional control preference (Degner & Sloan, 1992; Janz et al., 2004; Ryan & Sysko, 2007; Wallberg et al., 2000). Alternatively, researchers could theorized older adults newly diagnosed with symptomatic myeloma may have a different profile of decisional role preferences because of their extensive previous and varied life experiences with the health care system as most older adults have at least one or more co-morbidities. It is unknown whether the number of co-morbidities and multiple exposures to health care system could increase or decrease decision making control preferences. This could be an area of future research on treatment decision making experience. Anecdotally, one subject in this study shared that when she experienced adverse effects of her first chemotherapy, it made her more involved in the decision making process; she asked more questions to her oncologist prior to agreeing to the next line of chemotherapy for her myeloma.

We have documented that the subjects in this study demonstrated a strong desire to participate in the decision-making process, though they may not have a full understanding of myeloma due to the complexity of the disease. During the interview, the subjects reported seeking information from various sources as we have reported in another paper and some subjects identified their physicians as the primary source to explain the different treatment options available to them (Tariman, Doorenbos, Schepp, Singhal, & Berry, 2014, in press). These findings have strong implications for physicians to provide the information that the study subjects want and need during treatment decision making clinic encounters.

The findings of the study are consistent with the most recent findings reported in studies from the United Kingdom (Caldon, Walters, & Reed, 2008) and Canada, where research showed increasing numbers of cancer patients wanting to have some control of treatment decisions—as high as 92–93% in some studies (Davison et al., 2003; Davison et al., 2007; Singh et al., 2010). It should be noted, however, that we did not measure the degree of congruence between subjects’ desired role and actual role during the TDM process though this is a persistent issue in published TDM studies (Tariman et al., 2010) because it is not part of the study objectives.

The emergence of a second valid metric (i.e. DCBAE scale in Table 3) in Coombs’ unfolding analysis found in this current study warrants further exploration. The subjects had a nearly 50/50 distribution between shared and active decisional role preferences. This is clearly a trend seen in the Western societies, where healthcare consumerism is on the rise (Fronstin & Collins, 2006). One could theorize that the former paternalistic model of physician-patient relationship is losing ground, as suggested by Rosenstein (1986) in the mid-1980s. Patient preferences have always tended to fall into a dichotomy of preferences. The shift in decisional preference toward either a shared or active role for patients has not been reported before. In the past, patient preferences tended to fall into either the “active” or “passive” category (Blanchard et al., 1988; Cassileth et al., 1980) particularly in the older adult patients with cancer (Singh et al., 2010); not shared or active as revealed in this current study.

Limitations

There are limitations of this study primarily related to sample size and demographics. The small sample primarily consisted of Caucasians who were college-educated with relatively high-incomes. In addition, the majority of subjects were receiving care at a university-based comprehensive cancer center. The small sample size and lack of a diversity limit the generalizability of study findings. We were unable to make more meaningful comparisons of differences in decisional role preferences by subgroups and we were also unable to examine associations of decisional role preferences with multiple sociodemographic variables. Moreover, since this is a cross-sectional study, the findings may not be applicable to symptomatic myeloma patients who are beyond 6 months of diagnosis. These limitations should be addressed in future research. Additionally, further study using a longitudinal approach is needed to better describe the stability or change in study subjects’ decisional role preference over time in older adults diagnosed with cancer. Patients diagnosed with myeloma are excellent study population for studies involving older adults since the incidence of myeloma peaks at the 7^th decade of life (Kyle et al., 2004), but recruitment of a large number of subjects remains a major challenge because myeloma remains a rare form of cancer accounting for only 1% of all cancers diagnosed each year (Siegel, Naishadham, & Jemal, 2012). A direct approach in study recruitment is a very effective way of recruiting subjects; far better than the mail approach. The direct approach should be utilized if the local IRB committee still allows this approach in study subject recruitment. Lastly, myeloma affects a minority of younger patient population who are under 60 years of age. Examining the treatment decision-making patterns of this younger patient subgroup should be done in future studies and the differences in role preferences and influential treatment decision factors should also be explored.

Implications for Practice

The study findings suggest that study subjects diagnosed with symptomatic myeloma do want to participate in the treatment decision-making process. Although myeloma is complex and not easy to understand for laypersons, these findings indicate that subjects still want to learn as much as they reasonably can about the disease and treatment so as to understand the reason why certain treatment options might be better for them than others. The majority of subjects also wanted to share the treatment decision with their physicians and/or want to make the decision themselves. Therefore, it is critical for physicians, nurse practitioners, and physician assistants to practice full disclosure of treatment options to their patients so patients can make a truly informed decision. Since a patient’s level of preference for participation is highly variable and could personally mean differently for each patient, physicians and oncology nurses must also elicit the patient’s preference, explore what participation truly means for him/her, and facilitate the patient’s decision process. Because more cancer patients now want to participate in TDM, physicians, nurse practitioners, nurses, and policy makers must support more studies that can enhance patient involvement in treatment decision-making. Oncology nurses can do many things to help cancer patients achieve the level of participation they desire: 1) make sure patients receive disease and treatment-related information; 2) encourage patients to express their decisional role preference to the physician; 3) develop a culture of mutual respect and value the patient's desire for autonomy for treatment decision making; 4) oncology nurses must acknowledge that the right to make a treatment choice belongs to the patient; 5) provide psychological support to the patient during treatment decision making from the time of diagnosis to end of life care decision making.

Conclusion

Older adults newly diagnosed with symptomatic myeloma want a role in the treatment decision-making process. More studies that focus on supporting and involving patients in the decision-making process are needed in order to influence clinical practice and policy in this direction. It is essential that oncology nurses are cognizant of the differences in decisional role preferences in symptomatic myeloma patients in order to meet their individual decisional needs and preferences. Oncology nurses must respect the patient's desired role preference and oncology clinicians must listen to the patient and allow them to be autonomous in making treatment decisions if the patient so desire such control in the decision-making process. A culture of equipoise between the patient and the clinician during TDM must be cultivated in order to achieve the patient's desired level of participation.

Knowledge Translation.

• Contrary to the findings from the most recent meta-analysis where older patients (>50 years of age) preferred a passive role than younger patients (32% vs. 21%, p= <.001), this study showed only 5% of study participants preferred a passive role.

• In the past, the dichotomy for decisional role preference was between active and passive. The emergence of active and shared dichotomous roles found in this study require further investigation.

• Oncology nurses and other oncology practitioners must elicit a patient's decisional role preference and respect the patient's desire for autonomy during treatment decision making.