nent in adolescents and young adults (Gould and Shaffer, 1986; Phillips, 1974; Phillips and Carstensen, 1986; Schmidtke and Hafner, 1988). However, some evidence from adoption studies suggests that imitation is unlikely to explain all familial transmission because suicides in biological relatives unknown to the adoptee increase his or her risk (Schulsinger et al., 1979). Two other studies examining familial concordance between suicide attempts, one looking at twin pairs of attempts, and the other at parent–child pairs has found wide variability in the timing of the pairs of attempts, not consistent with imitation (Brent et al., in press; Statham et al., 1998).
Family discord impacts suicidal behavior, particularly among adolescents (Brent et al., 1994; Kosky et al., 1986; Kosky et al., 1990; Taylor and Stansfeld, 1984) (see Chapters 5 and 6). However, it is unclear to what extent family discord is the cause or consequence of other difficulties that may lead to suicide. Both sexual and physical abuse (see Chapter 5) have been associated with suicidal behavior (Brent et al., 1999; Brown et al., 1999; Fergusson et al., 1996; Kaplan et al., 1997; Renaud et al., 1999) as well as changes in central serotonin metabolism.
I see now that I had been incubating this death far longer than I recognized at the time. When I was a child, both my parents had half-heartedly put their heads in the gas oven. Or so they claimed. It seemed to me then a rather splendid gesture, though shrouded in mystery, a little area of veiled intensity, revealed only by hints and unexplained, swiftly suppressed outbursts. It was something hidden, attractive and not for the children, like sex. But it was also something that undoubtedly did happen to grownups. However hysterical or comic the behavior involved—and to a child it seemed more ludicrous than tragic to place your head in the greasy gas oven, like the Sunday roast joint— suicide was a fact, a subject that couldn’t be denied; it was something, however awful, that people did. When my own time came, I did not have to discover it for myself (Alvarez, The Savage God: A Study of Suicide, 1971/ 1990:291-292).
It is difficult to know how genetics and abuse interact in their influence on suicide. First, abuse is more likely to occur in the presence of parental depression and substance abuse (Chaffin et al., 1996). Second, abusing parents have a greater rate of impulsive control disorders, including suicide attempt (Roberts and Hawton, 1980). Third, abuse may bring about conditions that interact with pre-existing genetic vulnerablities for other risk factors such as depression or substance abuse (c.f., Silberg et al., 1999). Finally, there is evidence that some effects of sexual