Psychiatric Annals

An Aging Psychologist Assesses the Impact of Age on MMPI Profiles

Wendell M Swenson, PhD

Abstract

1958

It has been nearly a quarter of a century since this writer first ventured to use the MMPI with a normal but aged population.1 At that time there was a complete absence of literature regarding the use of the M MPI with gerontic subjects. Aaronson2 had attempted an analysis of a number of the cases given in An Atlas for the Clinical Use of the MMPI.* Categorizing the data from this publication by age decade, he found that only 2% of the 871 cases represented ages greater than 60 years. Thus, the opportunity of investigating MMPI profiles among the aged was indeed untapped to that time. Although structured personality assessment was a wellestablished activity in the science of psychology in those days, the population of older people in our society had been relatively untouched. No doubt this was due at least in part to the then common ageism stereotype - older people just would not want to cooperate with any kind of "new fangled" paper-and-pencil tools.

In an early attempt to destroy this stereotype, in 1958 we obtained MMPI profiles on 95 individuals over the age of 65 - primarily from the then-popular "Golden Age Clubs." In 195b I had gathered MMPI profiles on 66 female geriatric patients in a state mental hospital. The test results from these two groups are shown in Table 1. The profile for the so-called gerontic normals seems to be primarily a mildly neurotic one. with depression showing the most significant deviation followed closely by scales I and 5 (hypochondriasis and hysteria, respectively). In contrast to the 66 female geriatric patients, there was very little evidence of any significant deviation in scales 6 (paranoia). 7 (psychasthenia), and 8 (schizophrenia). At that time we were impressed with the fact that the only scale below the mean of the normals in the original standardization group was scale 9 (hypomania). Subsequent research has substantiated this observation as a common phenomenon in aging - namely, limitation of the number of interests and activities in individuals as they grow older.

Such were the results of our first excursion into the psychologic assessment of the gerontic sector of our population. Investigations similar to ours began to appear in the literature in ever-increasing numbers. It is not the purpose of this writing to review that literature. The reader is referred to the extensive review by Gynther4 of all significant research on aging and personality. He has outlined in an objective and lucid fashion most of the MMPI studies completed on psychiatric patients, medical patients, and normal subjects from 1960 to the present date.

1965

As indicated elsewhere in this issue, since 19b) we have been using computer technology at the Mayo Clinic for routine scoring and interpretation of the MMPI on large numbers of medical patients.1' In 1965fe we published data on age and sex differences related to MMPI response frequency in 25.000 medical patients. Here, for the first time, we began to analyze the significant age changes for specific MMPI items. Some of the results were rather striking. For example, in regard to the distribution of responses to the item "My daily life is full of things that keep me interested," it was our prediction that the percentage "true" response to this item would decrease with age. Such was not the case (Figure 1). Actually there is a small but consistent increase in the "true" response with advancing age up through the eighth decade.

Of still more interest was the discovery that two groups of items of rather homogeneous content demonstrated a significant curvilinear relationship with age. For example, in…

1958

It has been nearly a quarter of a century since this writer first ventured to use the MMPI with a normal but aged population.1 At that time there was a complete absence of literature regarding the use of the M MPI with gerontic subjects. Aaronson2 had attempted an analysis of a number of the cases given in An Atlas for the Clinical Use of the MMPI.* Categorizing the data from this publication by age decade, he found that only 2% of the 871 cases represented ages greater than 60 years. Thus, the opportunity of investigating MMPI profiles among the aged was indeed untapped to that time. Although structured personality assessment was a wellestablished activity in the science of psychology in those days, the population of older people in our society had been relatively untouched. No doubt this was due at least in part to the then common ageism stereotype - older people just would not want to cooperate with any kind of "new fangled" paper-and-pencil tools.

In an early attempt to destroy this stereotype, in 1958 we obtained MMPI profiles on 95 individuals over the age of 65 - primarily from the then-popular "Golden Age Clubs." In 195b I had gathered MMPI profiles on 66 female geriatric patients in a state mental hospital. The test results from these two groups are shown in Table 1. The profile for the so-called gerontic normals seems to be primarily a mildly neurotic one. with depression showing the most significant deviation followed closely by scales I and 5 (hypochondriasis and hysteria, respectively). In contrast to the 66 female geriatric patients, there was very little evidence of any significant deviation in scales 6 (paranoia). 7 (psychasthenia), and 8 (schizophrenia). At that time we were impressed with the fact that the only scale below the mean of the normals in the original standardization group was scale 9 (hypomania). Subsequent research has substantiated this observation as a common phenomenon in aging - namely, limitation of the number of interests and activities in individuals as they grow older.

Such were the results of our first excursion into the psychologic assessment of the gerontic sector of our population. Investigations similar to ours began to appear in the literature in ever-increasing numbers. It is not the purpose of this writing to review that literature. The reader is referred to the extensive review by Gynther4 of all significant research on aging and personality. He has outlined in an objective and lucid fashion most of the MMPI studies completed on psychiatric patients, medical patients, and normal subjects from 1960 to the present date.

1965

As indicated elsewhere in this issue, since 19b) we have been using computer technology at the Mayo Clinic for routine scoring and interpretation of the MMPI on large numbers of medical patients.1' In 1965fe we published data on age and sex differences related to MMPI response frequency in 25.000 medical patients. Here, for the first time, we began to analyze the significant age changes for specific MMPI items. Some of the results were rather striking. For example, in regard to the distribution of responses to the item "My daily life is full of things that keep me interested," it was our prediction that the percentage "true" response to this item would decrease with age. Such was not the case (Figure 1). Actually there is a small but consistent increase in the "true" response with advancing age up through the eighth decade.

Of still more interest was the discovery that two groups of items of rather homogeneous content demonstrated a significant curvilinear relationship with age. For example, in regard to percentage "true" responses to the item "I wake up fresh and rested most mornings." the change with age among females is of great significance (Figure 2). Here we find what may be evidence that there is a real low point in the lives and energies of women between 50 and 49 years of age. This curvilinear change no doubt coincides with complaints in clinical psychiatry of frustration related to the chores of child-rearing and other duties of the housewife in our culture. In Figure 3 is shown the percentage "true" responses to the item "I work under a great deal of tension." Here males ages 30 through 59 have a markedly greater frequency of "true" responses than do males in any other decade and far more than do women of any age.

On the basis of these observations, two new MMPI scales - 54 items for males and 45 items for females - were assembled in which this kind of curvilinear pattern was apparent. These scales were subsequently labeled the "worried breadwinner scale" and the "tired housewife scale," respectively. It was at this time that we saw clearly the desirability of age correction for scores on scales commonly used in clinical applications.

It is of some parenthetic interest to note that, in an analysis of age-related changes in MMPI responses on items concerning religion, the same type of curvilinear relationship was found in several items.7 For example, the percentage "true" responses to the statement "I feel sure that there is only one true religion" was the highest at the early and late age decades, and the lowest scores appeared in the middle age range. This relationship also appeared in response to the statement "I am very religious (more than most people)" and the item "I go to church almost every week." These results would certainly support the somewhat common assumption that religion plays a more significant role in individuals at earlier and later stages of their lives.

1973

In 1975 we examined the responses of 50.000 medical outpatients to the MMPI to obtain norms for a medical setting.8 One of the many by-products of this accomplishment was the opportunity to look at the overall changes in MMPI profiles in groups of individuals by age. Table 2 summarizes the T-score means and standard deviations on the 14 standard clinical and validating scales for these 50.000 medical patients grouped by age and sex. Only certain of the highlights need be identified. There is a very slight increase in somatic preoccupation (Hs) and a small but significant increase in depression (D) in both sexes with advancing age. Social conformity (Pd) shows a consistent decrease with advancing age in both sexes. There is a tendency for both sexes to have a more masculine-oriented interest pattern with advancing age, especially in the women who seem to show a loss of feminine identity to some degree in the older age group.

In scales allegedly measuring a more pathologic type of personality deviation (Pa, Pt, and Sc) there is a clear-cut decrease in deviance with advancing age, on almost a linear basis. This observation might lend some credence to the "survival of the fittest" hypothesis on a psychologic as well as on a physiologic basis. As noted previously in other research, there is a clear-cut linear decrease of elevation of the hypomania scale (Ma) with advancing age, the most significant change identified. We are now clearly convinced that these data support clinical observations that, as people grow older, they tend to disengage from or reduce the number of activities and interests that attracted them at an earlier age,

On the basis of the data analyzed above, it certainly is reasonable to assume that there is no general increase in significant psychopathology with advancing age. In fact, for the most part the reverse has been found to be true. With the exception of depression, it is reasonable to conclude that even among those involved in medical examination, individuals in our culture become more psychologically stable with advancing age.

Table

TABLE 1MEANS AND SIGMAS OF MMPI T-SCORES FROM 66 FEMALE GERIATRIC STATE HOSPITAL PATIENTS AND 95 NON-HOSPITALIZED ELDERLY SUBJECTS

TABLE 1

MEANS AND SIGMAS OF MMPI T-SCORES FROM 66 FEMALE GERIATRIC STATE HOSPITAL PATIENTS AND 95 NON-HOSPITALIZED ELDERLY SUBJECTS

Figure 1 . Distribution of "true" responses, by age and sex, for MMPI item 8. "My daily life is full of things that keep me interested." (From Pearson JS. Swenson WM. Rome HP: Age and sex differences related to MMPI response frequency in 25.000 medical patients. Am J Psychiatry 1 965. 1 2 1 .988-995 By permission of the American Psychiatric Association.)

Figure 1 . Distribution of "true" responses, by age and sex, for MMPI item 8. "My daily life is full of things that keep me interested." (From Pearson JS. Swenson WM. Rome HP: Age and sex differences related to MMPI response frequency in 25.000 medical patients. Am J Psychiatry 1 965. 1 2 1 .988-995 By permission of the American Psychiatric Association.)

1983

As indicated elsewhere in this issue, we recently completed a ^standardization of the MMPI, some 40 years after its original development.4 The details of that study and its results have been described by, Colligan in this issue. In that volume the restandardization sample was subdivided by age decade, just as was previously done in the Source Books in 1973. Pages 340 to 352 of that volume set forth in graphic form the age changes on the 10 basic clinical scales, as well as on the 5 validating scales. When one compares those results with the 1973 data, the similarity is striking. One could almost duplicate here the comments made in describing the results of the 50.000 medical patients' profiles 10 years previously. Again, the scores in the neurotic triad scales I (Hs), 2 (D), and 5 (Hy) show slight increases, with depression showing the greatest change. Again, scale 4 (Pd). the measure of social conformity, shows a consistent reduction in intensity of scores in both men and women. The reduction is almost completely linear in quality. Scale 5 (Mf) shows a slightly greater trend toward traditional masculine interests in both sexes, with a slightly more intense trend for women to become less traditionally feminine in their interest pattern. Scales b through 8 (Pa, Pt. and Sc) show relatively little change but what changedoes take place suggests a trend to greater normality with advancing age. Finally, as has been referred to twice previously in this article, scale 9 (Ma) shows a consistent and almost linear reduction in intensity from the 18- to 19-year-old level to the >70 age level.

Table

TABLE 2T-SCORE MEANS (M) AND STANDARD DEVIATIONS (SDJ ON 14 STANDARD CLINICAL AND VALIDITY SCALES FOR 50,000 MEDICAL PATIENTS, BY AGE AND SEX

TABLE 2

T-SCORE MEANS (M) AND STANDARD DEVIATIONS (SDJ ON 14 STANDARD CLINICAL AND VALIDITY SCALES FOR 50,000 MEDICAL PATIENTS, BY AGE AND SEX

Figure 2. Distribution of "true" responses, by age and sex. for MMPl item 3. "I wake up fresh and rested most mornings." (From Pearson JS. Swenson WM. Rome HP: Age and sex differences related to MMPI response frequency in 25.000 medical patients Am J Psychiatry 1 965. 1 2 1 :988-995. By permission of the American Psychiatric Association )

Figure 2. Distribution of "true" responses, by age and sex. for MMPl item 3. "I wake up fresh and rested most mornings." (From Pearson JS. Swenson WM. Rome HP: Age and sex differences related to MMPI response frequency in 25.000 medical patients Am J Psychiatry 1 965. 1 2 1 :988-995. By permission of the American Psychiatric Association )

CONCLUSIONS

The MMPI has withstood the impact of aging very well during the past 4 decades - perhaps better, or at least with more consistency, than has the aging psychologist writing this article. Two broad conclusions seem to evolve from this quarter-century analysis of MMPl testing experience.

I. The gerontic population is by no means reluctant or shy when confronted with the challenge of submitting to lengthy structured personality test assessment. We experienced no significant obstacles in 1961 and we certainly had no difficulties in 1983. In fact, it was common experience in the most recent data-gathering process to note that older individuals were much more cooperative in this scientific venture than were the adolescents.

Table

TABLE 2 (continued)T-SCORE MEANS (M) AND STANDARD DEVIATIONS (SD| ON 14 STANDARD CLINICAL AND VALIDITY SCALES FOR 50,000 MEDICAL PATIENTS, BYAGE AND SEX

TABLE 2 (continued)

T-SCORE MEANS (M) AND STANDARD DEVIATIONS (SD| ON 14 STANDARD CLINICAL AND VALIDITY SCALES FOR 50,000 MEDICAL PATIENTS, BYAGE AND SEX

Figure 3. Distribution of "true" responses, by age and sex. for MMPi item 13. "I work under a great deal of tension." (From Pearson JS, Swenson WM, Rome HP; Age and sex differences related to MMPI response frequency in 25.000 medical patients. Am J Psychiatry 1 965. 121 988-995. By permission of the American Psychiatric Association.)

Figure 3. Distribution of "true" responses, by age and sex. for MMPi item 13. "I work under a great deal of tension." (From Pearson JS, Swenson WM, Rome HP; Age and sex differences related to MMPI response frequency in 25.000 medical patients. Am J Psychiatry 1 965. 121 988-995. By permission of the American Psychiatric Association.)

2. There is certainly no objective evidence familiar to me to show that individuals admit to more psychopathology as they age. With very few exceptions, the gerontic end of the continuum holds up very well in terms of stability, conformity, and lack of psychopathology. This long-range discovery has given me both personal and professional assurance that reassessment of our population a decade or two in the future will produce strikingly similar findings.

REFERENCES

1. Swenson VVM: Structured personality testing in the aged: An MMPI study of the gerontic population. J Clin Psychol 1961; 17:502-304.

2. Aaronson BS: Age and sex influences on MMlM prolilc peak distributions in an abnormal population. J Consult Clin Psychol 1958; 22:203-206.

3. Hathaway SR, Meehl PE: An Atlas for the Clinical Use of the MMPL Minneapolis, University of Minnesota Press, 1951.

4. Gynther MD: Aping and personality, in Butcher IN (ed): New Developments in the Use of the MMPI Minneapolis, University of Minnesota Press. 1979. pp 54-68.

5. Rome HP, Swenson WM, Malaya P, et al: Symposium on automation technics in personality assessment. Mayo Clin Proc 1962: 57:61-82.

6. Pearson IS, Swenson WM, Rome HP: Age and sex differences related to MMPI response frequency in 25,000 medical paiients. Am J Psychiatry 1965:121:988-995.

7. Swenson WM: Approaches to the study of religion and aging, in Religion and Aging. The Behavioral and Social Sciences Look at Religion and Aging Los Angeles, Rossmour-Cortese Institute for the Study of Retirement and Aging. University of Southern California, 1967. pp 59-84.

8. Swenson WM, Pearson IS. Osborne D An MMPI Source Book: Basic Item. Scale, and Pattern Data on 50,000 Medical Patients. Minneapolis. University of Minnesota Press, 1975.

9. Colligan RC, Osborne D, Swenson WM, et al: The MMPI: A Contemporary Normative Study. New York. Praeger Publishers, 1985.

TABLE 1

MEANS AND SIGMAS OF MMPI T-SCORES FROM 66 FEMALE GERIATRIC STATE HOSPITAL PATIENTS AND 95 NON-HOSPITALIZED ELDERLY SUBJECTS

TABLE 2

T-SCORE MEANS (M) AND STANDARD DEVIATIONS (SDJ ON 14 STANDARD CLINICAL AND VALIDITY SCALES FOR 50,000 MEDICAL PATIENTS, BY AGE AND SEX

TABLE 2 (continued)

T-SCORE MEANS (M) AND STANDARD DEVIATIONS (SD| ON 14 STANDARD CLINICAL AND VALIDITY SCALES FOR 50,000 MEDICAL PATIENTS, BYAGE AND SEX

10.3928/0048-5713-19850901-09

Sign up to receive

Journal E-contents