Session VII - Tibia

Use of the NEO-Five Factor Inventory to Assess Personality of Patients with Lower Extremity Trauma: A Two-Year Prospective Study

Adil H. Haider, MD, MPH; Ellen J. MacKenzie, PhD; David Edwin, PhD, Johns Hopkins School of Public Health and Policy, Baltimore, MD; Michael J. Bosse, MD, Carolinas Medical Center, Charlotte, NC and the LEAP Study Group

Background: It is generally believed that personality traits of a patient can significantly influence clinical and functional outcomes following trauma. Experiences from other clinical specialties suggest that certain personality types often respond better to selected treatments. However the relationship between personality, treatment and outcomes following severe trauma has not been well documented. This is due in part to the fact that existing personality inventories are long and exhaustive, and often impractical for use in the trauma setting. Contemporary psychological research measures an individual's personality in the context of five identified "Super Traits" namely: Neuroticism (N), Extraversion (E), Openness to experience (O), Agreeableness (A) and Conscientiousness (C). The NEO-Five Factor Inventory © (NEO-FFI) is a self-administered test that quantifies an individual's personality within this context. Unlike the more popular Minnesota Multiphasic Personality Inventory­2© (MMPI-2) which consists of 567 questions, the short form NEO-FFI consists of only 60 items. It also has the added feature of a third person version whereby a patient's significant other may rate the patient's personality as a surrogate.

Objective: To assess the reliability, validity and utility of a brief, flexible personality measure in the context of acute trauma.

Methods: Patients with high-energy lower-extremity trauma (HELET) admitted to 8 Level 1 trauma centers were administered the NEO-FFI within the context of a larger study designed to asses the benefits of limb reconstruction versus amputation. A total of 557 patients (93% of those enrolled in the larger study) completed the NEO-FFI before being discharged >from the acute hospital stay. For 416 of these patients, a significant other was identified and agreed to provide an independent assessment of the patient's personality at discharge. At 2 years post-injury, 95% of patients who were followed and asked to complete the NEO-FFI again complied, yielding 396 patients for whom complete baseline and 2-year assessments were available for comparison. Statistical analyses were performed to determine: 1) agreement between the scores provided by the patients and their significant others; 2) the stability of the measured personality traits over a period of 2 years; and 3) the relationship of the measured NEO-FFI traits with other characteristics and health habits of the patients known to be influenced by personality. An exhaustive literature search was also carried out to determine if the relationships observed in the trauma population were similar to those of the general population.

Results: As a group, study patients tended to be more neurotic, more extraverted and less open when compared to the general population norms for the NEO-FFI. (P < .05).

Table 1 summarizes 1) the agreement between scores provided by the patient and their significant others (as measured by the intra-class co-relation co-efficient); and 2) the stability of the scores over 2 years (as measured by the score change averaged over those patients who completed the NEO-FFI at baseline and 2 years later.)

Patient and Significant Patient Scores at

Other Scores at baseline (N=416) 24 Months (N=396)

NEO-FFI Trait	Patients Score At Baseline Mean (S.D)	Sig. Other Score At baseline Mean (S.D)	Intraclass Correlation Co-eff (95%C.I)	Patients Score 2 years post injury Mean (S.D)	Mean Change in Scores between baseline and 2 yrs Mean (95% C.I.)
Neuroticism	18.63 (7.61)	19.35 (8.94)	0.51 (.43, .57)	18.66 (8.52)	0.03 (-.79, 1.39)
Extraversion	30.32 (5.77)*	30.87 (6.34)	0.46 (.39, .54)	27.86 (6.27)*	-2.46 (1.53, 3.21)
Openness to Experience	24.69 (5.65)	23.92 (5.64)	0.54 (.47, .60)	24.42 (5.27)	0.27 (-.24, 1.35)
Agreeableness	30.57 (5.73)	30.94 (7.66)	0.44 (.36, .51)	30.50 (5.77)	0.07 (-.82, 0.83)
Conscientiousness	34.01 (6.29)*	32.46 (8.81)*	0.44 (.37, .52)	33.14 (6.49)	0.87 (-.24, 1.53)

* Means significantly different at P < 0.001 level.

Table 2 demonstrates the observed relationships between personality traits and patient characteristics and health behavior. For example smokers (n=209), tend to be more neurotic (N) and less extraverted (E), agreeable (A) and conscientious (C), when compared to non-smoking patients. (n=348). (P < 0.0001)

Patient characteristics and Health behaviors.	Measure	Number of Patients	Personality  Scores	Significance  Level
Years of Education	=> 13 yrs vs < 13 yrs	173 vs 384	Higher E, A, C. Lower N	P<.0001
Marriage	Never Married vs Married	201 vs 356	Higher O & A. Less N.	P <.00001
Household Income	=>$25,000 vs < $25,000	250 vs 307	Higher E, O, A and C. Less N.	P <.001
Insurance	Private vs None or Public	254 vs 303	Higher A. Less N	P <.001
Presence of Emotional Support	High vs Medium & Low	437 vs 120	Higher E, A, C. Lower N	P <.0001
Presence of Assistive Support	High vs Medium & Low	319 vs 236	Higher E, A, C. Lower N	P<.0001
Presence of Guidance Support	High vs Medium & Low	369 vs. 188	Higher E, A, C. Lower N	P <.0001
Smoking	Smokers vs Non Smoker	209 vs 346	Less E, A & C. Higher N.	P <.0001
Alcoholism	SMAST + vs SMAST ­	161 vs 396	Less A & C. Higher N.	P <.001
Number of Drinks per week	> 58 drinks/wk vs < 58/wk	91 vs 466	Less A & C. Higher N.	P <.00001
Hard Drugs	Used in past 12months vs No Use in past 12 months	51 vs 506	Less E & A. Higher N.	P <.0001

Discussion: For a personality test to be useful in the trauma setting it must be practical, reliable and valid. The NEO-FFI assessed in this study was easily completed by the vast majority of our patients and their significant others. We found acceptable agreement between the scores of the patients and their significant others. This not only lends support to validity but it also suggests that there is potential to use a patient's significant other if the patient is unable to complete the personality questionnaire. Perhaps one of the most significant findings of the study is that personality as measured by the NEO-FFI is robust to the acuteness of the event and is relatively stable over time. The slight, albeit statistically significant, decrease in extraversion (and increase in introversion) may be explained by onset of depression associated with prolonged treatments and decreased quality of life. It is remarkable that even after severe trauma individuals' personality traits do not change. This finding suggests that it does not make a difference at which stage of patient care the instrument is administered and implies that the NEO-FFI measures something very fundamental about the patient. We also found that personality traits of trauma patients are related to patient characteristics and behaviors in the directions that were expected based on the literature available on the general population. Results indicate that, as in the general population, positive personality traits like increased extraversion; agreeableness and conscientiousness are positively correlated with years of education, marriage and household income, whereas people who smoke, abuse drugs or abuse alcohol are more emotionally unstable or neurotic.

Conclusions: The NEO-FFI is a brief, reliable measure of underlying personality traits that is practical for use in a trauma setting. Its use in both outcomes research and patient evaluation should be encouraged.