Perceived parental care and control among Israeli female adolescents presenting to emergency rooms after self-poisoning

Adolescence, Summer, 2005 by Gary M. Diamond, Hila Didner, Ariela Waniel, Beatriz Priel, Jack Asherov, Shosh Arbel

Deliberate self-harm, broadly defined as any conscious act designed to damage, ruin or cause physical pain to the self, is surprisingly prevalent among adolescents worldwide. Rates among community samples of 14- to 18-year-olds range from 6%-9% (Centers for Disease Control, 1991; Friedman, Asnis, Boeck, & DiFiore, 1987; Merrick, 2000; O'Sullivan & Fitzgerald, 1998; Ponizovsky, Ritsner, & Modai, 1999; Smith & Crawford, 1986; Spirito, Brown, Overholser, & Fritz, 1989), with rates being higher among females than among males (Burgess, Hawton, & Loveday, 1998; O'Connor, Sheehy, & O'Connor, 2000; Patton et al., 1997). Estimates for clinical samples are as much as three to five times greater (Pfeffer, Conte, Plutchik, & Jerrett, 1980; Pfeffer, Zuckerman, Plutchik, & Misruchi, 1984).

The overwhelming majority of self-harming acts involve self-poisoning of low lethality (Spirito, Brown, Overholser, & Fritz, 1989). Such acts may be best understood in interpersonal terms. According to attachment theory (Bowlby, 1988), a child's capacity to regulate distress is determined, in large part, by his or her perception of attachment figures as being accessible and responsive. Even during adolescence, teenagers continue to value and depend on relationships with their parents, turning to them in times of stress (Kobak & Sceery, 1988; Steinberg, 1990). When adolescents are unsure of their parents' physical, psychological or emotional availability, distress may be accompanied by dramatic and sometimes unregulated behavior, such as angry outbursts, sad withdrawal, and non-lethal self-harming acts, such as self-poisoning (Adam, Sheldon-Keller, & West, 1996; de Jong, 1992; Lessard & Moretti, 1998; West et al., 1999). In these instances, self-poisoning can be understood as an urgent, hopeful appeal for care and protection (Adam, 1994). In fact, from 20% to 50% of adolescents report that their self-harming acts were intended, at least in part, to make others understand how desperate they were feeling, to elicit help, and/ or to find out how much they were loved (Bancroft et al., 1979; Boergers, Spirito, & Donaldson, 1998; Hawton et al., 1982; Kienhorst, DeWilde, Diekstra, & Wolter, 1995).

Self-poisoning acts of low lethality may also represent teenagers' attempt to attain behavioral autonomy vis-a-vis their parents. The development of behavioral autonomy--"the capacity to make one's own decisions, to be less influenced by others, and to be more self-governing and self-reliant" (Holmbeck & Updegrove, 1995, p. 18)--is an important task of adolescence. Increased behavioral autonomy facilitates secondary processes such as role exploration, identity formation, and individuation (Erikson, 1968; Holmbeck & Updegrove, 1995; Steinberg, 1990). Ideally, behavioral autonomy evolves gradually in the context of the parent-adolescent relationship (Youniss & Smollar, 1985; Grotevant & Cooper, 1986). Parents grant their adolescent increasing responsibilities and freedom in accordance with their adolescent's cognitive, behavioral, and emotional abilities (Hauser, Powers, Noam, & Jacobson, 1984; Ryan & Lynch, 1989). Problems develop when there is a significant discrepancy between the degree to which a parent is willing to grant autonomy and the adolescent's capacity to manage autonomy (Holmbeck & Updegrove, 1995). When parents exert excessive control, adolescents may use acts of non-lethal self-poisoning in an attempt to compel parents to yield. Prior studies have shown that anywhere from 7% to 26% of adolescents report that their self-harming behavior was intended to influence somebody or get them to change their mind (Bancroft et al., 1979; Boergers et al., 1998; Hawton et al., 1982; Kienhorst et al., 1995).

As would be expected, research has shown that adolescents who report a history of self-harming behavior or ideation perceive their parents as less caring and more controlling than do their peers (Adam, Keller, West, Larose, & Goszer, 1994; Martin & Waite, 1994). Such adolescents are two to three times more likely to characterize at least one of their parents as low in care and high in control--a style termed "affectionless control" (Parker, 1984) or "authoritarian" (Baumrind, 1991)--than as high in care and low in control, or what has been termed "optimal" parenting (Parker, Tupling, & Brown, 1979). However, previous studies have relied on adolescents' retrospective self-report of self-harming behavior and ideation, raising questions about the accuracy of such accounts. Furthermore, previous studies assessed perceived parental care and control using a face valid, self-report, closed-ended questionnaire, which allowed subjects to easily identify the constructs being measured. Such a procedure is subject to various response biases, including demand characteristics.

This case-comparison study compared perceived parental care and control among 24 Israeli adolescent females admitted to pediatric emergency rooms immediately after a self-poisoning act of low lethality to perceived parental care and control among 23 matched, community control adolescents reporting no history of self-harming behavior. Three aspects of this study make it unique. First, in contrast to previous studies, the occurrence and lethality of the self-injurious behavior were objectively determined based on emergency room documentation. Second, parental care and control were assessed using not only a face valid, closed-ended questionnaire, but also via independent, observer ratings of adolescents' unconstrained descriptions of each parent. When describing their parents, the adolescents were unaware of the constructs of interest in this study, thus minimizing the threat of response biases. Third, this is the first such study conducted on an Israeli sample. We hypothesized that based on both measurement procedures, self-poisoning adolescents would report lower levels of maternal and paternal care and higher levels of maternal and paternal control, than adolescents in the comparison group. Furthermore, we hypothesized that this effect would not be moderated by levels of psychological symptoms.