Body positions and movement patterns in female patients with congenital adrenal hyperplasia

HORMONES

AND

BEHAVIOR

26, 441-456 (1992)

Body Positions and Movement Patterns in Female Patients with Congenital Adrenal Hyperplasia’ RALF W. DITTMANN Psychosomatic

Department,

Children’s Hospital, D 2000 Hamburg

University of Hamburg, 20, Germany

Martinistrasse

52,

Female patients with congenital adrenal hyperplasia (CAH; N = 33; 11-41 years), simple-virilizing (SV) patients (N = 19), salt-wasting (SW) patients (N = 13), and sister controls (N = 14) were compared with regard to their body positions and movement patterns. Data collection comprised both self assessments and mothers’ assessments using 20 sex-dimorphic items with corresponding “more masculine” and “more feminine” versions for each variable, represented in photographs (forced-choice approach). Primarily based on mothers’ assessments, single-item results suggested slightly more masculine positions and patterns for female CAH patients compared to sisters, for SW patients more distinct than for SV patients. Results from an II-item scale (*‘motor behavior”, LY = 0.59) revealed differences between SW (more masculine) and SV patients for self assessments (P, one-tailed, < 0.09); sisters were in an intermediate position closer to the SV patients. According to mothers’ assessments, the CAH patient group as a whole differed (more masculine) from sisters (P < 0.06); this finding was mainly accounted for by the SW group (P < 0.04). Complex analyses on the relationship of motor behavior and intervening variables (e.g., postnatal androgenization, onset of puberty, menarche, height, weight, sexual orientation) revealed very few significant results. Findings rather suggested organizational hormonal effects on body positions and movements prenatally; they are in line with main results from the interview section of the Hamburg CAH study (e.g., “Gender-related behavior”). An approach of this kind seems to be justified for investigating motor behavior in future psychoendocrine studies. Q 1~2 Academic PIGS, IW.

Congenital adrenal hyperplasia (CAH) is a group of inborn enzyme deficiencies the most common of which is the 21-hydroxylase deficiency (95%). This disorder results in a decreased adrenal cortisot synthesis, an increased production of ACTH, cortisol precursors, and adrenal androgens (New and Levine, 1981, 1984; New, 1985; Migeon, 1989; Kaplan, 1990). Two major subtypes of the classical CAH have to be considered: the simple-virilizing (SV) variant and the salt-wasting (SW) variant w&h 1 Preliminary data were presented as a poster at the 9th Congress of the European Society for Child and Adolescent Psychiatry, ESCAP, London, September 1992. 441 0018-506x/92 $4.06 Copyright D 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

442

RALF W. M-ITMANN

an additional deficiency in aldosterone production. Female CAH patients are born with different degrees of genital virilization (Prader stages). In general, they need both corrective surgery of the genitalia and lifelong replacement of cortisol (for SW patients: plus a mineralocorticoid). As an “experiment of nature” female CAH patients have been studied for more than 2 decades in order to demonstrate “masculinizing” effects of prenatal hormones on different aspects of behavior: gender-related (or gender-role) behavior, cognitive functioning, psychosexual development, and sexual orientation (e.g., Money and Ehrhardt, 1972; Ehrhardt and Meyer-Bahlburg, 1981; Money, Schwartz, and Lewis, 1984; Resnick, Berenbaum, Gottesman, and Bouchard, 1986; Dittmann, 1989; Dittmann, Kappes, Kappes, Biirger, Stegner, Willig, and Wallis, 1990a; Dittmann, Kappes, Kappes, Borger, Meyer-Bahlburg, Stegner, Willig, and Wallis, 1990b; Dittmann, Kappes, and Kappes, in press). Compared to healthy sisters as controls, female CAH patients (and their mothers)-in comprehensive interviews--reported, e.g., significantly “more masculine” gender-role behavior, and the patients showed differences in psychosexual development and sexual orientation (Hamburg CAH study; Dittmann, 1989; Dittmann et al., 1990a,b, in press). Earlier results were corroborated by these findings (camp. Money and Ehrhardt, 1972; Ehrhardt and Meyer-Bahlburg, 1981; Money et al., 1984). Jenni and Jenni (1976) reported sex differences in “carrying books” (camp. Table 3) becoming more distinct with development from preschool ages to adolescence. Barlow, Hayes, Nelson, Steele, Meeler, and Mills (1979) studied “sex role motor behavior” in a college population and demonstrated significant sex differences in body positions and movements; they developed and validated a behavioral checklist that discriminated in the expected direction between biological males and females as well as between masculine and feminine females. The objective of this part of the Hamburg CAH project was to investigate effects of prenatal hormones on motor behavior (body positions and movement patterns), sexually dimorphic behavior not directly related to reproductive or aggressive functions. So far, there are very few data on this topic both from animal research and on humans (camp. DeVries, DeBruin, Uylings, and Comer, 1984): Beach (1970) reported on sexual behavior in prenatally androgenized female beagle dogs additionally implanted with testosterone pellets at birth. These female animals showed urination in the masculine way, lifting one hind leg-as do the male dogsand, like the normal males, urinating more frequently than the control females. Money (1987, p. 389) mentioned an undated film of Short and Clarke that illustrated prenatally androgenized ewes showing mating behavior and “. . . (. . .urinating behavior also) . . . exactly like a ram, even though, at the same time, . . . own ovaries . . . secreting estrogen, not androgen.” Goy (1990) without having studied this aspect systemat-

MOTOR

BEHAVIOR

IN CAH FEMALES

TABLE

443

1

Areas of Investigation in the Hamburg CAH Study -Coping and compliance (with CAH) (interview) -Gender-related behavior (gender-role) (interview) ---Gender-related present attitudes and plans for the future (interview) -Psychosexual development and sexual orientation (interview) -Body image -Intelligence, socioeconomic status, age, etc. -Degree of pre- and postnatal androgenization -Motor

behavior (assessment scale)

ically had observed prenatally androgenized female rhesus monkeys moving in a (more masculine?) “sailor’s walk.” Mtiller, Kraus-Orlitta, Dirlich-Wilhelm, and F6rster (1983) compared the motor behavior of 12 CAH girls (8-14 years) with that of normal girls and boys using a video technique. Seven behaviors were studied (while sitting, standing, and pitching a ball). The findings revealed a characteristic mixed pattern of feminine and masculine gestures and movements for the CAH patients that could be described as “androgynous” (camp. Mtiller, 1981). Money and Lobato (1988) described the “body language” of a female patient with XY hermaphroditism (with insufficient prenatal androgenization) as neither conforming to “an ultramasculine nor an ultrafeminine stereotype.” Significant sex differences in motor behaviors have been shown (e.g., Jenni and Jenni, 1976; Barlow et al., 1979). Based on the available evidence both concerning masculinizing effects of sex hormones on, e.g., gender-role behavior in humans and from the aforementioned findings on sex hormone effects on motor behavior both in animals and in patients, we hypothesized that female CAH patients who suffered an excess of prenatal androgens show more masculine movement patterns and body positions compared to female controls. In addition, we intended to study relationships with variables such as postnatal androgenization, somatic parameters, and sexual orientation, within the patient group in particular. Having used a newly formed instrument for the assessment of body positions and movement patterns we will focus on first experiences and results.

METHODS Table 1 summarizes the topics and variables targeted in this study; the assessment of motor behavior was carried out between two interview sections.

444

RALF

W. DITTMANN

Subjects

Thirty-four female CAH patients (85% of a hospital sample), and 14 healthy sisters as controls (ages, 11-41 years, difference ns), as well as 31 mothers participated in the Hamburg CAH study. Data about the socioeconomic status could be obtained for 29 families (SES; Kleining and Moore, 1968); two families were classified as upper class, 18 as middle class, and nine as lower class. Overall, patient and sister groups did not differ significantly in SES. Since not all data could be collected from every single subject sample sizes vary accordingly, e.g., information on simplevirilizing or salt-wasting variant of CAH was not available for two subjects. Thus, for self assessments of motor behavior data were collected from 33 patients (SV, 19; SW, 12) and 14 sisters; for mothers’ assessments sample sizes were 29 patients (SV, 16; SW, 13) and 12 sisters. Instrument

Barlow et al. (1979) used videotapings to investigate sex-role-specific motor behaviors while subjects were sitting, standing, and walking; their checklist could be used reliably by the raters. Based on the authors’ detailed descriptions for those sex-dimorphic motor behaviors (16 items) we had ooze lCyear-old girl imitate both the more masculine (m) and the more feminine (f) version of each behavior. (This person was not a CAH patient and was not involved in the study otherwise.) One or more typical photographs were taken to illustrate/characterize one behavior (m or f). (For examples of the original photographs, see Dittmann, 1989.) Four items both from daily-life experiences and from the literature were added in the same way (e.g., item 18, Jenni and Jenni, 1976). (For the combined original 20 items, compare the “Answer sheet,” Table 2.) Procedure

Probands (patients, healthy sisters as controls; mothers reporting on their daughters) were asked to check-in a forced-choice way-which of the two behaviors (per item, m or f) was closest to their own typical behavior, or that of their daughter(s), respectively. The more masculine or more feminine versions of the behaviors had not been labeled as such, they had been arranged to “A and B” in the set of photographs and on the Answer sheet at random. Statistical Analyses

The scorings for single items were “0” for the more masculine answer, or “1” (point) for the more feminine answer. Single items were analyzed primarily by nonparametric Mann-Whitney U tests (group comparisons, one-tailed). For a second set of analyses-based on the available data from all subjects-an 11-item composite scale (motor behavior) could be

MOTOR

BEHAVIOR TABLE

Answer sheet Name Please, check A or B! No.

445

IN CAH FEMALES 2

DOB --__I I

(instruction not presented here) Date -1 I %

A

Item

Standing 1 Feet placement 2 Arm movements 3 Wrist action 4 Hand(s) in pocket, etc. Walking (lower body) 5 Strides 6 Hip movements 7 Feet placement Waling 8 Arm 9 Wrist i0 Arm

--

(upper body) movements action position

Sitting (lower body) 11 Position of buttocks 12 Feet placement 13 Position of legs Sitting (upper body) 14 Arm movements 1.5 Wrist action 16 Hand motions Other movements/positions 17 Resting hands on hips 18 Carrying of books 19 Taking off a sweater 20 Looking at fingernails

formed that showed an acceptable internal consistency (Cronbach’s alpha = 0.59). The items chosen and their more masculine and more feminine characteristics are summarized in Table 3. (For all self assessments alpha for this scale was 0.37, for all mothers’ assessments alpha was 0.73.) So far, there are no further data available on the reliability of this scale. Barlow et al. (1979) for their checklist reported an overall interrater agreement of 0.8. The vast majority of these 11 items were derived from their checklist that had proven validity in adequately discriminating in the expected direction between biological males and femates as well as between masculine and feminine females. For this scale, thus, a maximum scoring of 11 points was possible (range, O-11; “femininity” in motor behavior). Case-control group comparisons

446

RALF W. DIITMANN

TABLE 3 Items from Assessment Scale for Motor Behavior, More Masculine (m), and More Feminine (f) Characteristics (According to Barlow et al., 1979; “Jemri & Jenni, 1976) Characteristics No.

m

Item

Standing 2 Arm movements 4 Hand(s) in pocket, etc.

f

From shoulder In pocket, under belt, etc.

From elbow Others

6 Hip movements 7 Feet placement

Long, smooth; free knee action Minimum Straddling a line

Small steps; controlled knee action Pronounced Stepping on a line

Walking (upper body) 8 Arm movements

From shoulder

From elbow

Walking (lower body) 5 Strides

Sitting (lower body) 11 Position of buttocks 12 Feet placement 13 Position of legs

Away from back of chair Close to back of chair Uncrosses, apart Uncrossed, together Legs crossed, ankle on knee Legs crossed, knee on knee

Sitting (upper body) 14 Arm movements

From shoulder

Other movements/positions 18 Carrying of books” At the side of the body; arms long

From elbow In front of the body; arms bent

for scale scorings were also carried out by Mann-Whitney U tests. Spearman rank correlations (one-tailed) were computed to study the relationship between motor behavior and both indicators of sexual orientation and several other factors. (Group means (x) and standard deviations (SD) were added for illustrative purposes only; P levels of significance up to ~0.10, one-tailed, were included to show tendencies.) For further methodological details, see Dittmann, 1989; Dittmann et al., 1990a. (Because of a more comprehensive statistical approach in this paper (e.g., more subjects included) sample sizes and, thus, results do not correspond completely to the (preliminary) findings presented in Dittmann, 1989, 1991.) RESULTS

The most important single-item results for self assessments are depicted in Table 4: The patient vs sister group comparisons revealed only two items (Nos. 2 and 15) with lower mean rankings (closer to the masculine pole) for the patient group (Mann-Whitney U tests, one-tailed, P s 0.10;

3

0.82 0.39 1.00 0.00 0.58 0.50 0.29 0.47

0.61 0.50 0.15 0.29

Other movements/positions 18 Carrying of hooks” Pt Sr 20 Looking at fingernails Pt Sr it

0.36 0.47

+ f

1J.52

0.50

0.45 0.50 0.39 0.50

+

+

+

0.10

co.04


0.74

0.45

0.47 051

0.89 0.32

SV Sr SV Sr

0.21 0.42

0.88 0.32

SV 0.84 0.37 Sr SV 0.21 0.42 Sr

sv Sr

SV Sr sv Sr

+ +

+ +

+

+

+

dl.007

co.10

co.02

SW 0.17 0.39 Sr SW 0.00 0.00 Sr

SW 0.58 0.51 Sr SW 0.00 0.00 Sr

SW 0.25 0.45 Sr

SW 0.75 0.45 Sr SW 0.67 0.49 Sr

+

+

+

+

f

+

+

co.03

<0.05

<0.02

<0.03

<0.03

Note. LMR, lower mean rank (closer to masculine pole), indicated by + . P, level of significance, Mann-Whitney U tests (one-tailed, corrected for ties), values up to 0.10 included to show tendencies. Pt (patients) = 33; SV (Simple-virilizing patients) = 19; SW (salt-wasting patients) = 12; Sr (sisters) = 14. ” Item became part of the Il.-item scale (motor behavior).

15 Wrist action

0.73 0.64 0.18 0.36

Pt o.s.5 0.51 Sr 0.36 0.50

Pt Sr Pt Sr

Pt Sr Pt Sr

Sitting (upper body) 14 Arm movements”

Walking (upper body) 10 Arm position

3 Wrist action

Standing 2 Arm movements”

TABLE 4 Results, Single Items, Self Assessments, Means (x), Standard Deviations, and Significance Levels for Group Comparisons with P G 0.10 (MannWhitney U tests, one-tailed): Pt vs Sr, SW vs Sr, SV vs Sr -.-___MannMannMannWhitnev U Whitnev I/ Whitney U No. Item X SD LMR P. X SD LMR P’ X SD LMR P

Item

13 Position of legs”

Sitting (lower body) 12 Feet placement”

10 Arm position

Walking (upper body) 9 Wrist action

Walking (lower body) 5 Strides”

0.21 0.58 0.41 0.58 0.69 0.42 0.45 1.00

X

0.41 0.51 0.50 0.51 0.47 0.51 0.51 0.00

Pt Sr Pt Sr

Pt Sr Pt Sr

0.45 0.83 0.31 0.67

0.51 0.39 0.47 0.49

0.69 0.47 0.42 0.51 0.28 0.45 0.58 0.51

+

+

+ +

+

+ +

+ +

+

SD LMR

Pt 0.48 0.51 Sr 0.25 0.45

Pt Sr 2 Arm movements’ Pt Sr 3 Wrist action Pt Sr 4 Hand(s) in pocket, etc.” Pt Sr

Standing 1 Feet placement

No.

CO.02

co.02

co.04

<0.06

<0.09

0.0005

CO.06

co.01

MannWhitney U P.

0.56 0.51

0.62 0.50

0.25 0.45

0.25 0.45

SV 0.56 0.51 Sr sv 0.44 0.51 Sr

SV 0.69 0.48 Sr sv 0.19 0.40 Sr

+

+

+ +

+

+ +

+

+

SD LMR

SV 0.50 0.52 Sr

SV Sr SV Sr SV Sr SV Sr

X

co.07

<0.02

CO.08

co.10

<0.005

<.04

c.04

MannWhitnev U P”

0.31 0.48

0.77 0.44

0.62 0.51

0.15 0.38

0.38 0.51

0.69 0.48

SW 0.31 0.48 Sr SW 0.15 0.38 Sr

SW Sr SW Sr

+

+

t +

+

+ +

+

+

SD LMR

SW 0.46 0.52 Sr

SW Sr SW Sr SW Sr SW Sr

X

-co.005

co.005

10.09

0.0002

co.04

co.02

MannWhitnev U P-

TABLE 5 Results, Single Items, Mothers’ Assessments, Means (X), Standard Deviations, and Significance Levels for Group Comparisons with P 6 0.10 (Mann-Whitney lJ tests, one-tailed): Pt vs Sr, SV vs Sr, SW vs Sr

;

0.45 0.45 0.49 0.39 0.41 0.49

co.10

+ -t

co.003

+

+

co.03

Sr

SV

0.31

0.48

SV 0.31 0.48 SK sv 0.75 0.45 Sr

Sitting (upper body) sv 0.44 0.51 Sr

+

+

f

+ 0.02

SW Sr SW Sr SW Sr

0.08 0.28

0.46 0.52

0.23 0.44

SW 0.77 0.44 Sr

+

+

+

+

0.06

co.03

co.006


--me-.__ Note. LMR, lower mean rank (closer to masculine pole), iudicated by + P, level of significance, Mann-Whitney LI tests (one-tailed, corrected for ties), values up to 0.10 included to show tendencies. Pt (patients) = 29; SV (simple-viritizing patients) = 16; SW (salt-wasting patients) = 13; Sr (sisters) = 12. ” ltem became part of the 11-item scale (motor behavior).

0.28 0.75 0.62 0.83 0.21 0.33

Pt 0.59 0.50 Sr 0.25 0.45

Other movements/positions 17 Resting hands on hips Pt Sr 18 Carrying of books’ Pt Sr 20 Looking at fingernails Pt Sr

I6 Hand motions

450

RALF W. DITTMANN

one item ~0.05). One item (No. 3) showed a lower ranking for the sister group (P level ~0.04). Comparing the subgroup of SV patients to the sister group three items (Nos. 10, 14, and 18) reached P levels co.10 (two items ~0.05); however, all showed lower rankings for the sisters. Considering the group comparisons of SW patients vs sisters four items (Nos. 2, 15, 18, and 20) revealed more masculine rankings for SW patients with P levels ~0.05; but, sisters appeared more masculine again in item No. 3 (P < 0.03). Accordingly, for the six items Nos. 9, 10, 14, 15, 18, and 20, SW patients compared to SV patients showed lower rankings (P < 0.10, five items CO.05; not illustrated in Table 4), most obvious for items No. 10 (P < 0.01; “arm movements while walking”) and No. 18 (P < 0.0001; “carrying of books”, item from Jenni and Jenni, 1976). The corresponding results for mothers’ assessments are summarized in Table 5: The CAH patient vs sister differences were more clearcut; for seven items (Nos. 1, 4, 10,12,13,17, and 18) patients revealed the lower mean ranks (P < 0.10; six items ~0.05). However, there were also four items (Nos. 3, 5, 9, and 16) with lower rankings for sisters (P < 0.10; one item ~0.05). With regard to the SV patient vs sister differences six items (Nos. 1,2,4, 10, 12, and 17) revealed lower rankings for SV patients with P levels co.10 (five items ~0.05). However, in contrast, for two items (Nos. 5 and 9) there were lower mean ranks for sisters (P < 0.10). Salt-wasting patients, according to mothers’ assessments, showed seven items (Nos. 1, 4, 12, 13, 17, 18, and 20) with lower mean ranks (P < 0.10, six items ~0.05); for the sister group three items (Nos. 3, 9, and 16) revealed lower rankings (one item cO.05). SV/SW comparisons, based on mothers’ data, showed SW patients in a position closer to the masculine pole (P < 0.10) for five items (Nos. 4, 6, 12, 13, and 18; one item ~0.05; not illustrated in Table 5); SV patients were rated more masculine in items 2 and 16 (P < 0.05). Summarizing these single-item findings and taking into account the multiple statistical comparisons, self assessments did not reveal (significant) differences between the entire CAH patient group and the sister controls; however, considering the SW patient results there seemed to be a tendency to more masculine assessments for some aspects in body positions and movement patterns in particular (P < 0.05, items Nos. 2, 15, 18, and 20). The patient/sister findings from mothers’ assessments were slightly more distinct; the SW patients were assessed as significantly more masculine (P < 0.05) in items 1, 4, 12, 13, 17, and 18. Table 6 illustrates the results for the scale scorings on motor behavior from both self assessments and mothers’ assessments: Considering the mean scorings in self assessments, SW patients showed the relatively most masculine value while SV patients had a mean slightly more feminine than the sisters; however, all the given group comparisons were found beyond P = 0.10. The SW/SV difference reached P < 0.09. As in the

el e +

+

LMR

<0.06

P

SV Sr

SV Sr 5.12

5.95

X

2.73

1.61

SD

+

t

LMR P

MannWhitney U

Sr

SW

4.08

SW 5.00 Sr

X

2.60

1.76

SD

+

+

LMR

<0.04

P

MannWhitney 0

Note. LMR, Lower mean rank (closer to masculine pole), indicated by + . P, level of significance, Mann-Whitney U tests (one-tailed, corrected for-ties), values up to 0.10 included to show tendencies. Self assessments: Pt (patients) = 33; SV (simple-virilizing patients) = 19; SW (salt-wasting patients) = 12; Sr (sisters) = 14. Mothers’ assessments: Pt (patients) = 29; SV (simple-virilizing patients) = 16; SW (salt-wasting patients) = 13; Sr (sisters) = 12.

2.68 2.79

Mothers’ assessmems Pt 4.66 Sr 6.17

SD

1.94 1.78

X

MannWhitney U

TABLE 6 Scale Motor Behavior, Self and Mothers’ Assessment, Means (/I’), Standard Deviations, and Significance Levels for Group Comparisons with P G 0.10 (Mann-Whitney U Tests, One-Tailed): Pt vs Sr, SV vs Sr, SW vs Sr

Pt 5.55 Sr 5.64

Self assessments

Results, ll-Item

452

RALF

W. DI’ITMANN

single-item analyses results for mothers’ assessments were more clearcut: both SW patients’ and sisters’ means appeared more distinct, with SV patients being in an intermediate position. The CAH patient/sister difference reached P < 0.06; this basic result was mainly accounted for by the SW group (P < 0.04). Finally, group means plus further statistical comparisons revealed the tendency of the mothers to assess their patient daughters relatively more masculine (Pt, P < 0.04; SV, P < 0.09) and their healthy daughters relatively (ns) more feminine compared to self assessments. SW and SV patients in this study did not differ significantly (MannWhitney U tests, two-tailed) in the “degree of genital virilization” at birth/first diagnosis (Prader stages; P = 0.13); the same hold true for both “onset of pubertal development” and “age at menarche” (P 3 0.13; smaller sample sizes), as well as for “body height” and “body weight” at interview (P > 0.20). However, compared to SV patients, SW patients in this study had been treated significantly “earlier” (P < 0.05) and “better” throughout life (P < 0.05), i.e., they had suffered less postnatal androgenization (Dittmann, 1989; Dittmann et al., 1990b). In order to study relationships and effects of important somatic variables on motor behavior within the CAH patient group additional analyses were carried out (groups dichotomized; Mann-Whitney U tests, two-tailed). According to self assessments, female CAH patients with low (N = 14) vs high degree (N = 12) of prenatal androgenization (genital virilization) differed significantly in scale scoring for motor behavior (P = 0.03; means, 6.1 vs 4.4). However, there were no significant differences between patient groups dichotomized with regard to measures of postnatal androgenization (onset of steroid replacement, P = 0.60; quality of treatment throughout life, P > 0.11). Corresponding results from mothers’ assessments did not reveal significant differences (all P > 0.14). Accompanying correlations (Spearman rank, one-tailed) turned out significant only between motor behavior scorings (self assessments) and degree of prenatal androgenization: all patients (N = 26), Y = -0.49, P = 0.01 (SV patients (N = 15), Y = -0.47, P < 0.08; SW patients (N = lo), r = -0.49, P < 0.15). Neither the correlations with respect to postnatal androgenization nor the relevant correlations from mothers’ assessments revealed any significant results (all P 3 0.20). There was a relationship between more feminine motor behavior and late onset of pubertal development (or vice versa) in the SW patient subgroup (N = 9, self assessments; I = 0.64, P = 0.06). However, all remaining correlations for motor behavior with the somatic variables onset of pubertal development, age at menarche, body height, and body weight at interview, done separately for all patients, SV patients, and SW patients, for self assessments and mothers’ assessments, respectively, did not reveal any significant relationships (all P > 0.15).

MOTOR

BEHAVIOR

IN CAH

FEMALES

453

Additionally, Spearman rank correlations were computed for motor behavior scorings with both results for scales “HOM” and “HEY (reflecting homosexual and heterosexual interests) and for an index (HOM - HET), all from the interview section on psychosexuality in the Hamburg CAH study (self report data only; Dittmann et al., in press). Correlations were separately done for the groups of all subjects, all patients, SV patients, SW patients, and sisters. None of the results reached statistical significance (vast majority, P 3 0.20). DISCUSSION The constellation of single-item findings points to somewhat more masculine body positions and movement patterns for female CAH patients compared to sister controls, for SW patients results were slightly more distinct than for SV patients. This summing of results was primarily based on mothers’ assessments. The attempt to form an easy to use paper-and-pencil version of an instrument for the assessment of sex-dimorphic body positions and movement patterns also resulted in an 11-item scale (motor behavior; overall Cronbach’s (Y = 0.59). With regard to self assessments, female CAH patients did not differ from sisters in their mean scale scorings; SV patients described themselves close to the controls while SW patients showed a tendency to more masculine mean scores (SW/SV; P < 0.09). According to mothers’ assessments (with a Cronbach’s (Y = 0.73) CAH patients were reported to show relatively more masculine motor behavior compared to healthy sisters (P < 0.06); this difference was mainly accounted for by the patient subgroup with the SW variant of CAH (P < 0.04). Miiller et al. (1983)-although for a younger age group--had found an ‘“androgynous” pattern of gestures and movements in CAH females and discussed the potential effectiveness of this pattern for interpersonnel behaviors in varying situations (SW/SV comparisons had not been carried out). However, more masculine motor behavior might also have anunfavorable-impact on social contacts and acceptance by peers, and thus, indirectly affect the psychosexual development of the patients (see Money et al., 1984; Dittmann et al., in press). Our correlative analyses with two scales and an index all reflecting “sexuai orientation,” though, did not suggest any important relationship with motor behavior. Unfortunately, during the course of this study we had no opportunity to collect detailed anthropometric data on patients and controls. However, the hndings on significantly less postnatal androgenization for the SW vs SV patients (Dittmann et al., 1990b) do not suggest “physical masculinity” to be a major factor for “more masculine body positions and movement patterns.” Both onset of pubertal development and menarche and height and weight at interview were as well not significantly related to motur behavior. The basic constellation of results-with CAH patients showing

454

RALF

W. DIlTMANN

a tendency to more masculine patterns for motor behavior (primarily in mothers’ assessment)-had also been found in the comprehensive interview part of this study, e.g., for the domains “Gender-related behavior” (self and mothers’ assessments) as well as for “Current attitudes/plans for the future” (self reports). SW patients presented most clearcut results, too, while SV patients were in an intermediate position (Dittmann et al., 1990a,b). The interviews on “Psychosexual development and sexual orientation” (self reports) revealed corresponding findings (Dittmann et al., in press). Additional data from this study plus complex statistical analyses did not show effects of prenatal factors on behavior in its various aspects (e.g., gender-role, motor behavior). Thus, patient/sister plus SW/SV difference-onsistent across several behavioral domains (including motor behavior, although to a lesser degree)--rather suggest organizational, possibly differentiating hormonal effects on the development of brain and behavior prenatally. (This view was partly corroborated by the finding that motor behavior (self assessment) was significantly correlated with genital virilization at birth (one indicator for the degree of prenatal androgenization); in addition, there was a significant group difference between patients with low vs high genital virilization.) As to the remaining behavioral domains, prenatal hormonal effects seemed to be relatively independent of genital virilization (e.g., no significant SW/SV difference in this study); similar results, an independence of genital and behavioral masculinization, were reported both for prenatally androgenized ewes (Clarke, 1977) and prenatally androgenized female rhesus monkeys (Goy, Bercovitch, and McBrair, 1988). Apart from SW/SV differences in the adrenal production of androgens prenatally, several potential mechanisms might be involved in hormonal effects on brain development and behavior, e.g., SW/SV differences in cortisol levels and indirect cortisol effects via prenatal liver functions (on enzymes metabolizing sex hormones, on sexhormone-binding globulin; for a broader discussion, see Dittmann, 1989; Dittmann et al., 1990a,b, in press). Since sample sizes were relatively small in this study, and group differences were inconsistent, and since statistical results were small in magnitude we recommend replication studies for the domain of movement patterns and body positions in new CAH samples. Based on our experiences with this instrument, both on direct feedback from the subjects and on the findings presented here, further use of an approach of this kind seems to be justified in psychoendocrine studies. In addition, data from larger samples of “normal, healthy” subjects would reveal important information on the characteristics of the items (represented in photographs) and on this or modified scales, in particular.

MOTOR

BEHAVIOR

IN CAH FEMALES

455

ACKNOWLEDGMENTS This study was supported in part by a grant from the VW-Foundation, Hannover, Germany, to the Psychosomatic Department, Children’s Hospital, University of Hamburg, Az. I/37 302, and by a grant from the German Research Foundation (DFG), Bonn, to the author, Az. Di 302/3-l. The author thanks Drs. D. Borger, M. H. Kappes, M. E. Kappes, and D. Klusmann for their support in data collection and with statistical analyses, and Mrs. H. Hinzmann, E. Peters, and P. Somorjai for secretarial assistance.

REFERENCES Barlow, D. H., Hayes, S. C., Nelson, R. O., Steele, D. L., Meeler, M. E., and Mills, J. R. (1979). Sex role motor behavior: A behavioral checklist. Behav. Assess. 1, 119138. Beach, F. (1970). Hormonal effects on socio-sexual behavior in dogs. In H. Gibian and E. J. Plotz (Eds.), 21. Colloquium der Gesellschaft fiir Biologische Chemie. Springer, New York. Clarke, I. J. (1977). The sexual behavior of prenatally androgenized ewes observed in the field. Reprod. Fertil. 49, 311-315. DeVries, G. J., DeBruin, J. P. C., Uylings, H. B. M., and Corner. M. A. (Eds.) (1984). Sex differences in the brain. In Progress in Brain Research, Vol. 61. Elsevier, Amsterdam/New York/Oxford. Dittmann, R. W. (1989). Priinatal wirksame Hormone und Verhaltensmerkmale van Patientinmen mit den beiden klassischen Varianten des 21-Hydroxylase-Defektes. Peter Lang, Frankfurt/Bern/New York/Paris. Dittmann, R. W. (1991). Motor behavior and congenital adrenal hyperplasia (CAH) in females. Abstract Booklet, p. 87. 9th Congress of the European Society for Child and Adolescent Psychiatry, London. Dittmann. R. W., Kappes, M. H., Kappes, M. E., Borger, D., Stegner, H., Willig, R. H., and Wallis, H. (1990a). Congenital adrenal hyperplasia I: Gender-related behavior and attitudes in female patients and sisters. Psychoneuroendocrinology 15, 401-420. Dittmann, R. W., Kappes, M. H., Kappes, M. E., Borger. D., Meyer-Bahlburg, H. F. L., Stegner. H., Willig, R. H., and Wallis, H. (1990b). Congenita! adrenal byperplasia. II. Gender-related behavior and attitudes in female salt-wasting and simple-virilizing patients. Psychoneuroendocrinology 15, 421-434. Dittmann, R. W., Kappes, M. E., and Kappes, M. H. (in press). Sexual behavior in adolescent and adult females with congenital adrenal hyperplasia. Psychoneuroendocrinology.

Ehrhardt. A. A., and Meyer-Bahlburg, H. F. L. (1981). Effects of prenatal sex hormones on gender-related behavior. Science 211, 1312-1318. Goy, R. W. (1990). Personal communication. Goy. R. W., Bercovitch, F. B., and McBrair, M. C. (1988j. Behavioral masculinization is independent of genital masculinization in prenatally androgenized female rhesus macaques. Harm. Behav. 22, 552-571. Jenni, D. .4., and Jenni, M. A. (1976). Carrying behavior in humans: Analysis of sex differences. Science 194, 859-860. Kaplan, S. A. (Ed.) (1990). Clinical Pediatric Endocrinology. Saunders. Philadelphia/ London/Toronto. Kleining, G., and Moore, H. (1968). Soziale Selbsteinstufung. Kolner 2. Soziol. Sozialpsychof.

20, 502-552.

Migeon. C. J. (1989). Diagnosis and treatment of adrenogenital disorders. In L. J. DeGroot, G. M. Besser, G. F. Cahill, Jr., J. C. Marshall, D. H. NeIson, W. D. Qdell, J. T.

456

RALF W. DITIMANN

Potts, Jr., A. H. Rubinstein, and E. Steinberger (Eds.), Endocrinology, 2nd ed., Vol. 2. Saunders, Philadelphia/London/Toronto. Money, J. (1987). Sin, sickness, or status? Homosexual gender identity and psychoneuroendocrinology. Am. Psychol. 42, 384-399. Money, J., and Ehrhardt, A. A. (1972). Man and Women. Boy and Girl. Johns Hopkins Univ. Press, Baltimore. Money, J., and Lobato, C. (1988). Matched pair of siblings concordant for 46,XY hermaphroditism with female sex assignment, and discordant for erotosexual outcome. Psychiatry

51, 65-79.

Money, J., Schwartz, M., and Lewis, V. G. (1984). Adult erotosexuai status and fetal hormonal masculinization and demasculinization: 46,xX congenital virilizing adrenal hyperplasia and 46,XY androgen-insensitivity syndrome compared. Psychoneuroendocrinology 9, 405-414. Miiher, M. (1981). Motorisches Geschlechtsrollenverhalten bei Miidchen mit dem adrenogenitalen

Syndrom.

Eine

Beobachtungsstudie

zur

geschlechtstypischen

Kiirperhaltung.

Psychological Dissertation, Univ. of Munich, Germany. Miiller, M., Kraus-Orlitta, U., Dirlich-Wilhelm, H., and Forster, C. (1983). Motorisches GeschlechtsroIIenverhaIten bei Mldchen mit adrenogenitalem Syndrom. Eine Beobachtungsstudie zur geschlechtstypischen Kiirperhaltung. Z. Kinder-Jugendpsychiut. 11, 100-115. New, M. I. (Ed.) (1985). Congenital adrenal hyperplasia. Ann. NY Acud. Sci. 458. New, M. I., and Levine, L. S. (1981). Adrenal hyperplasia in intersex states. In N. Josso (Ed.), The Zntersex Child, pp. 51-64. Karger, Basel. New, M. I., and Levine, L. S. (1984). Congenital Adrenal Hyperplasia. Springer, Berlin. Resnick, S. M., Berenbaum, S. A., Gottesman, I. I., and Bouchard, T. J., Jr. (1986). Early hormonal influences on cognitive functioning in congenital adrenal hyperplasia. Dev. Psychol. 22, 191-198.