Journal of the American Academy of Dermatology
Kriiger et aL risk for developing malignant melanoma in a German population. Int J Dermatol 1989;28:517-23. 7. Elder DE, Greene MH, Bondie EE, et al. Acquired melanocytic nevi and melanoma: the dysplastic nevus syndrome. In: Ackermann AB, ed. Pathology of malignant melanoma. New York: Masson, 1981:185-215. 8. Clark W, Elder D, Guerry D, et al. A study of tumor progression: the precursor lesion of superficial spreading and nodular melanoma. Hum Pathol 1984;15:1147-65. 9. Nie NH, HuH CH, Jenkins JG, et al. SPSS Statistical package for the social sciences. New York: McGraw-Hill, 1975. 10. Garbe C, Wiebe!t H, Orfanos CE. Change of epidemiologic characteristics of malignant melanoma during the years 1962-1972 and 1983-i 986 in the Federal Republic of Germany. Dermatologica 1989;178:131-5. 11. Crombie JK. Distribution of malignant melanoma on the body surface. Br J Cancer i981;43:842-9. 12. Holman C, Mulroney C, Armstrong B. Epidemiology of pre-invasive and invasive malignant melanoma in Western Australia. Int J Cancer 1980;25:317-23. 13. Pack G, Lenson N, Gerber D. Regional distribution of moles and melanomas. Arch Surg 1952;65:862-70. 14. MacKie RM, English J, Aitchinson TC, et al. The number
15. 16. 17. 18.
and distribution of benign pigmented moles (melanocytic naevi) in a healthy British population. Br J Dermatol 1985;113:167-74. Weinstock MA, Colditz GA, Willet WC, et aL Moles and site-specificrisk of nonfamilial cutaneous malignant melanoma in women. J Nati Cancer Inst 1989;81:948-52. MacKie RM, Aitchinson T. Severe sunburn and subsequent risk of primary cutaneous malignant melanoma in Scotland. Br J Cancer 1982;46:955-60. Green A, Sisldnd V, Hansen ME, et al. Melanocytic nevi in schoolchildren in Queensland. J AM ACADDERMATOL i989;20:1054-60. Gallagher RP, McLean DI, Yang P, et al. Suntan, sunburn, and pigmentation factors and the frequency of acquired melanoeytie nevi in children: similarities to melanoma: the Vancouver mole study. Arch Dermatol 1990; 126:770-6. Rampen FHJ, Fleuren BAM, de Boo ThM, et al. Prevalence of common "acquired" nevocytic nevi and dysplastic nevi is not related to ultraviolet exposure. J AM ACAD DERMATOL 1988;18:679-83. Nicholls EM. Development and elimination of pigmented mole and the anatomical distribution of primary malignant melanoma. Cancer 1973;32:191-5.
The psychological effects of androgenetic alopecia in men Thomas F. Cash, PhD NorJblk, Virginia Background: Despite the prevalence of androgenetic alopecia among men, little is known about its psychological effects. Objective: This investigation examined the psychosocial sequelae that balding men attribute to hair loss and compared balding and nonbalding men in personality functioning. Methods:Subjects included 63 men with modest balding, 40 men with more extensive balding, and 42 nonbalding controls. All anonymously completed a battery of standardized psychological measures. Results: Reported effects of balding reflected considerable preoccupation, moderate stress or distress, and copious coping efforts. These effects were especially salient among men with more extensive balding and among younger men, single men, and those with an earlier hairloss onset. Relative to controls, balding men had less body-image satisfaction yet were comparable on other personality indexes. Personality correlates of the psychological responses to hair loss were identified. Conclusion: Although most men regard hair loss to be an unwanted, distressing experience that diminishes their body image, balding men actively cope and generally retain the integrity of their personality functioning. (J AM ACAD DERMATOL 1992;26:926-31.)
From the Department of Psychology,Old DominionUniversity. Supportedby a grant fromManning, Selvage,and Lee,Inc.,NewYork, New York, and The Upjohn Company,Kalamazoo,Michigan. Accepted for publication Dec. 2, 1991. Reprint requests: Thomas F. Cash, PhD, Department of Psychology, Old Dominion University,Norfolk, VA 23529-0267.
Androgenetic alopecia affects the majority of men during their lifetime. 1,2 Despite a voluminous literature on the psychology of physical appearance, 3, 4 there exists little scientific study of the psychosocial impact of male-pattern baldness (MPB). One recent experiment examined the effects of
Volume 26 Number 6 June 1992 MPB on initial social impressions and found that men and women perceive visibly balding men as older and less physically and socially attractive than their nonbalding peers. 5 Thus balding may constitute a socially disadvantageous condition for men. Another study suggests that balding men may feel less physically attractive than men without hair loss and that balding men with public self-consciousness m a y be especially bothered by hair loss. 6 However, the few remaining published studies offer conflicting findings. 7 Because of the prevalence of MPB and its potentially detrimental social effects, this research was conducted to discern whether MPB affects men's psychosocial experiences and self-image. METHOD
Subiects Thirty-one barber shops and hair salons in metropolitan South Hampton Roads, Virginia, referred male clientele to participate in the research for a nominal fee. The shops received packets that included step-by-step procedures and forms for recruitment of qualifying clientele (i.e., at least 18 years old, with no alopecia caused by medical conditions or treatments). To minimize selfselection biases, the study was described to clients in general terms, and stylists approached clients without considering their expressed or inferred attitudes about hair (loss). More than 80% of the solicited men volunteered and were scheduled for a 90-minute appointment at the university laboratory. The sample of 145 participants was 96% white, 31% married, and ranged in age from 18 to 70 years (median 30 years) and in educational attainment from 11 to 21 years (median 14 years), with considerable occupational diversity. None had ever worn a hairpiece or had received medical or surgical treatment for MPB.
Assessments of hair loss Subjects' stylists and ultimately the subjects themselves used an adapted Norwood-Hamilton chart8 to classify their hair (loss) patterns. Balding men also indicated their age when they first noticed their hair loss, the amount of MPB since onset, the amount of expected hair loss during the next decade, and the social notieeability of their hair loss.
Measures of specific psychosocial effects of hair loss The Hair Loss Effects Questionnaire (HLEQ) listed 70 possible effects of MPB, including emotional, cognitive, and behavioral events. MPB subjects indicated how their hair loss affected them on a 5-point rating of each item (i.e., - 2 = happened much less;-1 = happened
Androgenetic alopecia 927 somewhat less; O = no change; + l = happened somewhat more; and +2 = happened much more). On the basis of rational and statistical procedures, four compositeHLEQ subscales were derived: (1) Negative Socioemotional Events (20 items; e.g., feeling self-conscious,beingteased); (2) Positive Soeioemotional Events (15 items; e.g., feeling self-confident); (3) Cognitive Preoccupation (12 items; e.g., thinking about hair loss, noticing bald men); and (4) Behavioral Coping (14 items; e.g., seeking reassurance, doing things to conceal hair loss or improve appearance). Nine of the 70 items did not contribute to any subscale. Reliability coefficients from 0.80 to 0.93 indicate a high degree of internal consistency of these subscales.
Measures of personality and psychosocial well-being Standardized measures were included to assess bodyimage satisfaction,9-u social serf-esteem, 12 social anxiety,t3 public self-consciousness,J3sexualself-eonfidenee,l° and locus of control.14Locus of control refers to a person's belief system about his extent of control over eventsin his life. All measures possessed satisfactoryreliabilityfor this sample.
Data collection procedures On arrival at the laboratory, each subject viewed a videotape emphasizing the anonymity and confidentiality of his data and then completed materials alone in a private room. To avoid reactivity in measurement, subjects answered personality inventories before questionnaires specific to hair loss. RESULTS Three hair-loss groups ware constituted: (1) No hair loss (nonbalding controls; n = 42); (2) low hair loss (either frontal recession or loss at the crown or some overall thinning; n = 63); and (3) high hair loss (more extensive hair loss at both hairline and crown; n = 40). Given the acceptable reliability of self-classifications in relation to stylists' judgments (Cohen's x = 0.80) and because subjects completed questionnaires based on their self-perceived baldness, self-classification was used to define groups for data analyses. In support of the groups' validity, only 15% of the men in group 2 (low hair loss) versus 70% of men in group 3 (high hair loss) believed that their MPB was moderately or very noticeable to others (p < 0.001). Men with high hair loss also reported a great er amount of hair loss since onset (p < 0.001). Groups did not differ in current age or duration of hair loss (median 6.5 years). Table I summarizes comparisons of the low-hairloss and high-hair-loss subjects on the HLEQ. The mean score per item was tested by t tests (/9 < 0.05)
Journal of the American Academy of Dermatology
Table I, Psychosocial effects attributed to hair loss by low loss and high loss groups ,Extent of hair loss Nature of analysis
% Effects significant 60 (of 70 HLEQ items) % Effects 16 reported by majority of the group HLEQ factors: Negative socioemotional events % Subjects 31 who increased Group mean +0.24a comparisons Positive soeioemotional events % Subjects 6 who decreased Group m e a n +0.20a comparisons Cognitive preoccupation % Subjects 54 who increased Group mean +0.68a comparisons Behavioral coping % Subjects 48 who increased Group mean +0.57a comparisons
59 +0.65b 8 +0,05a
69 +0.96b 69 +0.77a
Noa'~: The group mean comparisons (F tests) are designated by subscripts,Row means not sharing a common subscript are significantly different (p < 0.05),
against a no-change population value of zero. Among low-hair-loss men, 60% of the 70 events had significant effects, and 16% were reported by at least half of these men. The majority of low-hairloss men were more preoccupied as a result of MPB, about one third experienced an overall negative socioemotional impact, and about half increased their behavioral coping strategies. Among high-hair-loss men, 79% of the potential effects were reported to a significant degree, with 36% reported by half or more of the men. More than two thirds reported increased cognitive preoccupation and behavioral coping as a result of hair loss, and the majority indicated that MPB increased negative socioemotional events. ANOVAs compared low-hair-loss and high-hair-
loss groups on the HLEQ's four types of effects that subjects attributed to their MPB. As Table I summarizes, the groups were significantly different (p < 0.01) with regard to the negative socioemotional impact of MPB. Negative effects were reported at almost twice the rate by high-hair-loss as by low-hair-loss men. Moreover, high-hair-loss men were more preoccupied with MPB (,p < 0.05) and were slightly more involved in coping activities (p < 0.10). To provide key descriptive highlights of the HLEQ results, Table II lists some of the most salient effects of hair loss reported by balding men. The table describes the percentage of low-hair-loss and highhair-loss men reporting these specific effects. What is the ultimate impact of MPB on the personality functioning of balding men? ANOVAs compared the no-hair-loss (controls), low-hair-loss, and high-hair-loss groups on the measures of personality and well-being. The findings are consistent. Although men report that MPB has an adverse, stressful impact, its impact does not apparently alter the personality functioning of the majority of men. The only reliable effects were on the men's body image. One effect is a circumscribed one; namely, men without MPB were significantly more satisfied with their hair than were balding men, and high-hair-loss men were significantly less satisfied with their hair than were low-hair-loss men (ps < 0.05). In addition, high-hair-loss men had greater overall body image dissatisfaction than nonbalding control subjects (ps < 0.05). Five characteristics of balding men were targeted for further analysis. The strategy for statistical analysis by ANOVAs was first to test for differences on the specific effects attributed to MPB (i.e., the four HLEQ scores) and on the level of body-image satisfaction with hair. If differences were found for any of these measures, then differences in psychosoeial functioning were evaluated. The first question concerns the importance of the age at onset of balding. Men whose MPB had begun by age 21 reported somewhat greater preoccupation (ps < 0.10), more negative events, and diminished positive events than men in the later onset groups. Men with early-onset hair loss had come to feel less satisfied with their hair and less satisfied with appearance in general (ps < 0.05). The next comparisons considered the current age of the balding men--those younger than 26 years of
Volume 26 Number 6 June 1992
Androgenetic alopecia 929
Table II. Percentage of men attributing specific effects to the occurrence of M P B Extent of hair loss Reported experience
Cognitive preoccupation Wish for more hair Notice bald/balding men Spend time looking in mirror at hair/head Wonder what others think Negative socioemotional events Get teased by peers Feel self-conscious Look older than actual age Worry that others will notice Feel helpless about MPB Worry about aging Feel less attractive Envy good-looking men Behavioral coping Try to improve hairstyle Try to improve physique Dress more nicely Wear hats or caps Seek reassurance about looks Grow a beard or mustache
I . 1-I/g.h
Table III. Psychological correlates of the extent of stress/distress attributed to M P B by balding men Personality/Well-being variable
Hair satisfaction Appearance evaluation Body areas satisfaction Appearance orientation Sexual self-confidence Social self-esteem Social anxiety Public self-consciousness Locus of control--internality Locus of control--chance Locus of control--powerful others
-0.51' -0.22t -0.35* +0,19t -0,21t -0.33* +0.17 +0.36* +0.01 +0.26t +0.21t
*p < 0,01 (two-tailed). tP < 0.05 (two-tailed).
26 23 23
45 41 39
age, men 26 to 35 years old, and men 36 years of age or older. The youngest men reported the most intense preoccupation and coping efforts in response to their hair loss (ps < 0.05). Comparisons were made between men who regarded their MPB as minimally versus highly socially noticeable. Significant results (ps < 0.05) in-
dicated higher perceived noticeability to be related to more negative socioemotional effects, more discontent with hair and overall appearance, and more externality in locus of control. M e n who expected little further M P B during the next 10 years were compared with men expecting much more hair loss. The prediction that the expectation of continued balding carries with it more adversity was confirmed--for the experience of negative events, cognitive preoccupation, and behavioral coping (ps < 0.05). Moreover, balding men who expected more (versus minimal) future M P B had less satisfaction with hair and with overall appearance; they were also less internal in locus of control (p < 0.05). Do experiences with M P B depend on a man's marital or relationship status? Comparisons indicated that single men who were not dating reported significantly greater preoccupation and more negative socioemotional effects, especially when compared with married men (ps < 0.05). A final analysis assessed whether there is a relation between how stressful MPB is and a person's current personality functioning. For the 103 balding men, Pearson correlations were calculated between reported stress/distress (i.e., H L E Q Negative Events scores) and the measures of personality functioning. The results in Table III convey a consistent picture of the relation between the stressfulness of hair loss and current functioning. Nine of the 11 correlations were statistically significant. The greater the per-
Journal of the American Academy of Dermatology
ceived stressfulness of MPB, the less satisfactory was the men's psychosocial functioning. Specifically, greater distress is associated with a poorer appearance-related body image albeit greater psychological investment in appearance (Appearance Orientation), a poorer sexuality self-concept, lower social self-esteem, higher social self-consciousness (as Franzoi et al.6 also found), and a more external Iocus of control. DISCUSSION The present investigation of the psychosocial impact of androgenetic alopeeia in men produced several conclusions. Among the most significant findings was that most men experience hair loss as a moderately stressful process. Balding men experience adverse effects that include self-conscious preoccupation with current and future alopecia, social teasing, worries about aging and about others' reactions to them, and feelings of diminished attractiveness. As a result, these men actively search for adjustive strategies to cope with, compensate for, or conceal their hair loss. The most distressed balding men are those with more extensive MPB and those who are younger, have an earlier MPB onset, are romantically unattached, and deem their MPB as progressive and socially noticeable. The experience of balding does not necessarily impair overall psychosocial functioning. Only bodyimage satisfaction seems to be globally affected. Consistent with other evidence,6 balding men feel less satisfied not only about their hair but also about their physical appearance in general. Whereas MPB is unlikely to exert widespread effects on personality, 7 the men most distressed by their hair loss display less adaptive psychosocial functioning. Causality cannot be inferred from correlational data. It may be true that men most stressed by MPB are ultimately adversely changed by it. Alternatively, men with poorer levels of well-being may initially be more predisposed to stress. They may lack the resources of a strong, positive self.concept to withstand the adversity of hair loss. In either case, the data imply that for some but not all men, androgenetic alopecia may diminish the quality of life. In some instances, hair loss may be the focal concern of patients with a body dysmorphic disorder, is Ours is a culture that places a premium on physical appearance. In this context, appearance-altering conditions can be psychosocially insidious, especially conditions like androgenetic alopecia with an uncertain course and a negative social meaning. 16 Medi-
cal treatments of persons with such body-image vulnerabilities have been discussed by Pruzinsky and Cash. 17Contrary to prevalent assumptions that only women have body-image problems, 4, 18 the present findings implicate MPB as contributing to men's body-image concerns. Still, the psychological effects of androgen'eric alopecia on women are likely to be even more deleterious than those observed among men, a hypothesis I recently investigated.* One important implication of the present results is that persons seeking remedies for androgenetic alopecia are often anticipating or experiencing losses beyond the loss of hair per se. An empathic understanding of these patients' concerns is essential to effective management. 19, 20 *Cash TF. Psychological effects of androgenetic alopecia among women: comparisons with female controls and with balding men. Unpublished technical report to The Upjohn Company, Kalamazoo, Mich., August 1991.
1. Hamilton JB. Patterned loss of hair in man: types and incidence. Ann N Y Acad Sci 1951;53:708-28. 2. Shear N, ed. Facts about healthy hair. Toronto: Grosvenor House, 1989. 3. Bull R, Rumsey N. The social psychology of facial appearance. New York: Springer-Verlag, 1988. 4. Cash TF. The psychology of physical appearance: aesthetics, attributes, and images. In: Cash TF, Pruzinsky T, eds. Body images: development~ deviance, and change. New York: Guilford, 1990:51-79. 5. Cash TF. Losing hair, losing points?: the effects of male pattern baldness on social impression formation. J Appl Soc Psycho[ 1990;20:154-67. 6. Franzoi SL, Anderson J, Frommclt S. Individual differences in men's perceptions of and reactions to thinning hair. J Soc Psycho| 1990;130:209-18. 7. van der Donk J, Passchier J, Dutree-Meulenberg ROGM, et al. Psychologic characteristics of men with alopecia androgenetica and their modification. [nt J Dermatol 1991; 30:22-28. 8. Norwood O. Hair transplant surgery, Springfield, Ill: Charles C Thomas, 1973. 9. Brown TA, Cash TF, Mikulka PJ. Attitudinal body-image assessment: factor analysis of the Body-Self Relations Questionnaire. J Pets Assess 1990;55:135-44. 10. Cash TF, Winstead BW, Janda LH. The great American shape-up: body image survey report. Psychol Today 1986; 20(4):30-7. 11. Cash TF. Body-image affect: gestalt versus summing the parts. Percept Mot Skills 1989;69:17-8, 12, Helmreich R, Stapp J. Short forms of the Texas SocialBehavior Inventory (TSBI): an objective measure of selfesteem. Bull Psyehonomic Soc 1974;4:473-5. 13. Fenigstein A, Scheier MF, Buss AH. Public and private self-consciousness: assessment and theory. J Consult Clin Psychol 1975;45:522-7. 14. Levenson H. Multidimensional locus of control in psychiatric patients. J Consult Clin Psychol 1973;41:397-404.
Volume 26 Number 6 June 1992
15. PhiUips KA. Body dysmorphic disorder: the distress of imagined ugliness. Am J Psychiatry 1991;148:1138-49. 16. Cash TF, Pruzinsky T, eds. Body images: development,deviance, and change. New York: Guilford, 1990. 17. Pruzinsky T, Cash TF. Medical interventions for the enhancement of adolescents' physical appearance: implications for social competence. In: Gullotta TP, Adams GR, Montemayor R, eds. Developing social competency in adolescence. Newbury Park, Calif."Sage, 1990:220-42.
18. CashTF, Brown TA. Gender and bodyimages: stereotypes and realities. Sex Roles 1989;21:357-69. 19. Cash TF, Butters JW. Poor body image: helping the patient to change. Med Aspects Hum Sexuality 1988;22(6):67-70. 20. Van Moffaert M. Training future dermatologists in psychodermatology. Gen Hosp Psychiatry !986;8:115-8.
The prevalence of acne on the basis of physical examination Robert S. Stern, MD Boston, Massachusetts
Background: It has been suggested that cystic acne is rare in women 15 to 44 years of age and infrequent in men. Objective: To determine the prevalence of acne, we analyzed the primary data from the National Health and Nutrition Examination Survey (NHANES) that included a cutaneous examination of a stratified random sample of 20,749 noninstitutionalized U.S. residents. Methods: We calculated the prevalence of various diseasestates based on NHANES primary data and the NHANES population weights. Prevalence estimates and male/female ratios of these estimates were calculated. Results: From 1971 to 1974, the projected prevalence of ache conglobata (grade IV acne) for women and men 15 to 44 years of age in the United States was 250,000 and 570,000, respectively. At the time of examination, an additional 582,000 women and 749,000 men were projected to have moderate acne with cysts and scars. Therefore the prevalence of ache conglobata and ache of at least a moderate degree with cysts and scars was 832,000 for women and 1,319,000 for men 15 to 44 years of age. The male/female ratio for acne with cysts and scars is approximately 1.6:1. Conclusion: The NHANES examination data demonstrate that ache with cysts and scars is common in both men and women. (J AM ACADDERMATOL1992;26:931-5.) Recently, there has been debate about the frequency of more severe forms of ache and the relative prevalence of this disease in men and women.t The National Health and Nutrition Examination Study ( N H A N E S ) was designed to measure the health and nutritional status of the U.S. population. A cutaneous examination was one element of a comprehensive examination. Detailed medical and social history data were also obtained. 2 Because of its sampling design, it is possible to project findings
From the Departmentof Dermatology,Beth IsraelHospital,Harvard Medical School. Accepted for publication Dec. 2, 1991. Reprint requests:Robert S. Stern, MD, Beth IsraelHospital,Harvard Medical School, 330 BreoklineAve., Boston,MA 02215. 16/1/35292
from this sample to calculate prevalence for the entire U.S. population. Past publications on the prevalence of ache have been based on the "significant diagnosis" and "complaint" sections of the examination form and not on the results of the primary examination. These earlier estimates suggested the prevalence of more severe forms of acne in women is low.l, 3, 4 To provide more objective estimates of the prevalence of ache of varying degrees of severity and other physical findings related to acne, we undertook an analysis of primary data from the N H A N E S cutaneous examination.
METHODS NHANES study The NHANES methods are detailed in earlier publications.R,3 Briefly, 20,749 persons who were noninstitu931