Push and pull: resolving differences of opinion during meetings

In health care organizations, meetings often involve clinicians and administrators who have conflicting views about the issues facing them. Although participants will often talk about a good meeting or a meeting that was a waste of time, there is limited data about the specific interpersonal processes associated with effective decision making.

In the clinical arena, the importance of a supportive, trusting doctor-patient relationship is emphasized. Medical providers using a relationship-centered approach listen carefully, respond to patient clues and engage the patient as a partner. (1, 2) Differences of opinion are negotiated openly and respectfully to reach common ground. This approach to patient care has been associated with higher patient and provider satisfaction, (3, 4) improved clinical outcomes (5-8) and better adherence to treatment plans. (9)

While considerable data support the benefits of relationship-centered care in clinical settings, limited data exist about the possible benefits of a relationship-centered approach in organizations.

To investigate this, we analyzed interactions when differences of opinion occurred in business meetings. Our assumption was that any effective group needs to be able to identify and manage conflicting opinions. They then need to be able to build consensus about moving forward in a positive direction. The extensive work of Chris Argyris (10) and those who have studied his work describe two key processes to the effective management of differences of opinion:

* Stating opinions (advocacy)

* Asking questions to deepen understanding (inquiry)

An effective communication course, Positive Power and Influence, (11) uses different language to describe the same concepts. The course emphasizes strategies that push, similar to advocacy, and those that pull, similar to inquiry.

When differences of opinion are managed effectively, a blend of push and pull processes is believed to be important in producing the most positive outcomes.

Data for our study were obtained by audiotaping administrative meetings at Poudre Valley Hospital in Fort Collins, Colo, Rochester, N.Y. and Concord, N.H. Participants included physicians, nurses, clinic staff and business managers. Most of the participants were male (59%) and most of the meeting leaders were male (74%). A total of 45 meetings were recorded including outpatient clinics (38%), administrative buildings (35%) and hospitals (27%).

Push me, pull you

Forty-two cases of a difference of opinion were identified in the transcripts from 31 meetings. The differences of opinion between participants was openly acknowledged in 19 percent of the cases. The frequency of interpersonal processes and outcomes is shown in Table 2.

To assess the association between interpersonal processes during the discussion and the outcome, we clustered processes into two categories: push statements and pull statements.

A higher frequency of pull statements was associated with a group consensus outcome whereas push statements were associated with outcomes based on one person's opinion (individual solution) (Table 3).

The gender of participants was not associated with the frequency of push or pull statements and no significant differences were found in the average number of exchanges for the five types of outcomes.

When differences of opinion arose in this sample of meetings, participants were more likely to advocate their position (push) than attempt to understand others' positions (pull). Although decisions were reached in most cases, the most common method for making a final decision was an individual participant presenting a solution with no consensus or vote. However, when more pull statements occurred, the chance of a reaching a group consensus was increased.

The desired goal of negotiation is to reach a decision that is mutually satisfying to both parties. (12) When all group members commit to a plan of action, the potential of follow-through is enhanced and group cohesiveness is maintained.

The results of this study suggest that the potential for group consensus is increased when members make more attempts to understand and affirm others' positions. However, this style of interaction happens a minority of the time. The more common interaction style is for members to advocate for their position, an interaction style associated with individual solutions rather than group consensus.

Research from physician-patient interviewing indicates that treatment outcomes are improved when the physician specifically solicits and considers the patient's viewpoint. (5-8) Our results suggest that a similar phenomenon may occur in administrative meetings.

Interruptions occurred in the majority of meetings (see Table 2). Being unable to present one's ideas fully, without interruption, can lead to escalating advocacy statements or withdrawal from the conversation. We hypothesize that inquiry and group consensus will be increased if group leaders minimize interruptions during meetings.

Push statements dominate

If our results suggest that pull statements are associated with group consensus, why do push statements dominate in this sample of administrative meetings?

1. In a group setting participants may be concerned their view will not be heard if they spend time trying to understand others' opinions.

2. Participants and the leader may perceive the inquiry mode to be more time consuming. This possible concern is not supported by our data, which indicate that the number of exchanges is similar when the outcome is based on a group consensus or an individual solution.

3. A competitive culture in which the most verbal and assertive members win would impede the likelihood of members sincerely attempting to understand others' ideas.

4. When the complexity of the issue is not high, it may be more efficient to have one person decide. We did not code the interviews based on complexity; however, most of the issues discussed involved a complex mix of clinical, operational and/or financial dimensions.

5. Finally, managers and administrators may have succeeded due to their assertive, decisive styles. Indeed, physician training inculcates an ability to make efficient decisions as an individual. The training and career ascent of many managers and administrators may have selected those most adept at decision making, making the inquiry mode less utilized.

Limitations with this study warrant caution in interpretation of the results. First, this is a descriptive study, so cause-effect relationships cannot be assumed. It is possible, for example, that the occurrence of pull statements did not cause a group consensus outcome. Other variables, such as the leadership style, may have led to the decision outcome.

Also, the presence of a tape recorder may have altered the group interactions. In addition, non-verbal communication was not included. Facial expression and body language, such as head-nodding, are important modes of communication that cannot be measured with our data collection method. Finally, the sample size is small. Replication of these findings in other settings is needed to increase confidence in the results.

Recent findings in doctor-patient communication suggest that a collaborative style leads to better outcomes. Our preliminary findings suggest that physician administrators consider the possible benefits of decreasing the pushing and increasing the pulling that occurs during administrative meetings.

Table 1: Definitions and examples of the interpersonal processes and
decisions made during the 42 cases.

Acknowledgment of Conflict:
Acknowledgment: A statement that shows recognition that there is a
difference of opinion.
Examples:
"Well, however, let me argue the opposite side here ..."
"I have some problems with that, but we can talk about it."
"No, I disagree."
"And you may disagree with this, but I think ..."

Interpersonal Processes:
1. Advocacy: Expressing and/or explaining one's position; expressing an
opinion.
Examples:
"I think we ought to bid higher."
"I agree with John, because ..."
"I'm just trying to explain ..."
"I'm supporting that ..."
2. Inquiry: Explicit effort to understand more completely another's
position.
Examples:
"Why do you think we should bid higher?"
"Can you say more about that?"
"Any other comments about this topic?"
"But let me understand what the problem is for the nurses"
3. Affirmation/validation: Validating that one has understood another's
position.
Examples:
"I think those are valid comments."
"Oh, I see what you're saying."
"I understand that's your desire."
"That's a good alternative, too."
4. Self-disclosure: Sharing one's feelings/emotional reaction about an
issue.
Examples:
"I'm feeling uncomfortable. I don't want to keep pushing into this"
"I guess I don't ... I just ... it bugs me that they have so much access
to patient records."
"I don't know what the right answer is."
5. Summarizing: Providing a summary of the current discussion.
Examples:
"So what you're saying is ..."
"Let me see if I have this right ..."
"Let me just go back and reinforce a couple of things. We talked
about ..."
"Just to reiterate, if we ..."
6. Interruption: A statement that cuts off another person.
7. Humor: A remark followed by laughter of others.

Discussion Outcomes:
1. Individual solution: One participant offers/states a solution, but
opinions of others are not solicited.
2. Group consensus: An agreement is reached all participants are asked
if they agree.
3. Tabled: Explicit statement to postpone the issue due to lack of time
or more information is needed.
4. Vote: participants are asked to vote.
5. No decision: a decision is not reached due to sidetracking (change to
another topic), impasse (no resolution is apparent), or interruption
(out of time or some participants.)

Table 2: Summary of the frequency of interpersonal processes and
outcomes in the 45 cases.

Processes during discussions:

                        % of cases  Mean times per case

Advocacy                   100              5.7
Inquiry                     45              2.1
Affirmation/validation      33              1.6
Self-disclosure             10              1.3
Summarization               12              1.0
Interruption                79              4.8
Humor                       41              2.0

Outcomes of Discussion:

                     % of cases

Individual solution      41
Group consensus          19
Vote                      7
Tabled                   19
No decision              14

Table 3: The association between the frequency of "push" and "pull"
interpersonal processes and outcomes of discussions.

                            Mean "push"  Mean "pull"
Type of outcome             statements   statements

Individual solution (N=17)     4.2          1.0
Group consensus (N=8)          5.5          2.8

Acknowledgments
Funding for this project was provided by the Fetzer Institute.

References

1. Tresolini, CP and the Pew-Fetzer Task Force. Health Professions Education and Relationship-Centered Care. San Francisco, Calif.: Pew Health Professions Commission and the Fetzer Institute. 1994.

2. Suchman AL, Botelho RJ, Hinton-Walker P (eds). Partnerships in Healthcare: Transforming Relational Process. Rochester, N.Y.: University of Rochester Press. 1998.

3. Kaplan SH, Greenfield S, Gandek B, Rogers WH, Ware JW. "Characteristics of physicians with participatory decision-making styles." Annals of Intern Med. 1996, 124:497-504.

4. Matthews OA, Suchman AL, and Branch WT. "Making 'connections': Enhancing the therapeutic potential of patient-clinician relationships." Ann Intern Med. 1993, 118:973-977.

5. Stewart M, Brown J, Donner I, McWhinney R. Oates J, Weston WW, Jordan J. "The impact of patient-centered care on outcomes." J Fam Pract, 2000, 49:796-804.

6. Headache Study Group of the University of Western Ontario. "Predictions of outcome in headache patients presenting to family physicians." Headache, 1986, 26:285-294.

7. Greenfield S, Kaplan S, Ware J. "Expanding patient involvement in care: effects on patient outcomes." Annals of Intern Med. 1985, 102:520-528.

8. Starfield B, Wray C, Hess K, Gross R, Birk PS, D'Lugoff BC. "The influence of patient-practitioner agreement on outcome of care." JAMA, 1981, 71:127-132.

9. Steele, DJ, Jackson TC, and Guttman, MC. "Have you been taking your pills? The adherence monitoring sequence in the medical interview." J Fam Pract, 1990, 20:294-299.

10. Argyris C. Knowledge for Action: A Guide to Overcoming Barriers to Organizational Change. Josey-Bass, San Francisco, 1990.

11. Positive Power and Influence. SMS Systems, Nashua, N.H.

12. Fisher R, Ury W, Patton B. Getting to Yes: Negotiating Agreement Without Giving In., 2nd ed. New York: Penguin Books. 1991.

RELATED ARTICLE: Data Details: How the study was done Data for this study were analyzed in three phases:

1. Audiotapes from the 45 meetings were transcribed. The transcripts were read to identify specific discussions that involved a difference of opinion (a series of interactions in which two or more participants disagreed). Of the 45 meetings, a difference of opinion was found in 31 meetings, resulting in 42 cases (some meetings had more than one case of a difference of opinion). These 42 cases were the corpus of the remainder of the study.

2. The second phase was a qualitative analysis of the cases. Two investigators read a sample of 15 cases to identify common patterns in the interactions. During this phase, we identified interactions that fit into three general categories: a) whether the difference of opinion was openly acknowledged, b) seven interpersonal processes that occurred during the discussions, and c) five outcomes that followed the discussions. The definitions and examples of the three categories are shown in Table 1.

[ILLUSTRATION OMITTED]

3. The third phase was a quantitative analysis of all 42 cases. For each case, each statement was coded to identify:

* If the difference of opinion was acknowledged

* The frequency with which each type of interaction occurred

* The outcome of each case.

Finally, to identify associations between the interpersonal processes during a discussion and the outcome of each case, we clustered the interpersonal processes into two categories: "push" and "pull" statements.

By Kim Marvel, PhD, William Gunn, PhD and Kristen L. Brezinski, MS

Kim Marvel, PhD, is educational associate director at Fort Collins Family Medicine residency program in Fort Collins, Colo. He can be reached at 970-495-8840 or [email protected].

[ILLUSTRATION OMITTED]

William Gunn, PhD, is director of primary care behavioral health at New Hampshire/Dartmouth Family Medicine residency program in Concord, N.H. He can be reached at 603-225-2711 x4786 and [email protected]

[ILLUSTRATION OMITTED]

Kristen L. Brezinski, MS, is the project evaluation assistant at Fort Collins Family Medicine residency program in Fort Collins, Colo. She can be reached at 970-495-8879 and [email protected].

[ILLUSTRATION OMITTED]

COPYRIGHT 2004 American College of Physician Executives
COPYRIGHT 2004 Gale Group