GENDER DIFFERENCES AND ADOLESCENT RISKS.

INTRODUCTION

This paper examines characteristics and behaviour that are construed as posing risks to the well-being of adolescents in New Zealand, concentrating on the differential incidence and impacts of these risks on males and females.

The term "at risk", applied to individuals, families and households or to larger sections of society, such as age or ethnic groups, has gained increasing currency in policy literature in New Zealand and elsewhere.(1) "Risk factors" are assumed to predispose an individual or group to some negative outcome and therefore to pose a threat to well-being. These factors may be personal characteristics (gender, race, intelligence or temperament) or external factors, inherent in the family, community, educational or peer group environment. They may also be choices relating to behaviour or lifestyle. Risk may be perceived in relation to a specific threat to well-being, such as susceptibility to an illness, or to a combined or cumulative set of risk factors, such as those which might lead to a "cycle of disadvantage". The concepts of "risk factors" and "at risk" status are complex and fluid. It is easy to over-simplify them and to assume linkages and causality, resulting in deterministic or prescriptive conclusions. Use of the concepts must also avoid the dangers of generalisation and stereotyping. Like the concept of poverty, risk may be assessed on a relative or on an absolute basis. Risk is potential -- it may or may not result in a negative outcome, so that an attempt can be made to assess and measure the level of risk inherent in any situation.

Bearing in mind the caveats expressed in the previous paragraph, this paper examines factors that may constitute risks to the well-being of adolescents in New Zealand. This group is defined as young people of secondary school age (13-17), although information from beyond this range is drawn in as relevant. Gender is highlighted in the analysis because it is clearly a major factor in social differentiation. This is not to imply that other characteristics, especially socioeconomic status and ethnicity, are irrelevant to patterns of risk. They also must be considered in analysis and in the development of policy.

The paper begins by presenting a typology of risk, listing factors commonly assumed to be associated with negative outcomes. It then reviews recent New Zealand research findings on a range of risk factors. It does not attempt to assess the proportions of young people at risk or to measure the degree or intensity of risk.

A TYPOLOGY OF RISK

Risks have been defined as threats to well-being, here considered on an individual basis. Well-being has many dimensions. It encompasses security: physical, psychological, social, and economic security. It also includes health in the widest meaning of the term -- mental, physical, social and spiritual health. Many threats to health and security arise from pathologies within the family, such as violence and abuse of various kinds, and in society - for example, crime victimisation. Others emanate from lifestyle choices -- smoking, alcohol and drug use, and dangerous behaviour. The high degree of injury and illness that is related to lifestyle highlights the interactions between health, well-being and social factors such as living arrangements, work status and income. Well-being is the result of real-world influences which are complex and interactive.

Table 1 classifies and lists significant threats to the well-being of young people and indicates whether (according to evidence from recent New Zealand research) males or females are more likely to be susceptible, or whether the risk affects both groups more or less equally. The table omits risks that apply only to one sex, such as pregnancy.

Table 1 Adolescent Risks - Gender Differences

                                                     Both     Mainly
Risk Type       Specific Risk                       Genders    Male

Mental Health   Behavioural and conduct problems                X
and Behaviour   Truancy                                X
                Suspension/expulsion                            X
                Mental health problems                          X
                Substance dependence                            X
                Serious offending/arrest                        X
                Delinquency (adolescence limited)      X

Health          Alcohol misuse                         X
                Smoking
                Cannabis use                           X
                Other drug use                                  X

Physical        Accidental injury and death                     X
                Intentional injury and homicide                 X
                Suicide                                         X
                Suicide attempt
                Family violence

Sexual          Abuse (sexual)
                Early sexual experience
                Unsafe sexual behaviour                         X
                STD

Economic        Financial hardship (family)            X
                Unemployment
                Low income
                Lack of educational                             X
                qualifications

                                                    Mainly
Risk Type       Specific Risk                       Female

Mental Health   Behavioural and conduct problems
and Behaviour   Truancy
                Suspension/expulsion
                Mental health problems
                Substance dependence
                Serious offending/arrest
                Delinquency (adolescence limited)

Health          Alcohol misuse
                Smoking                                X
                Cannabis use
                Other drug use

Physical        Accidental injury and death
                Intentional injury and homicide
                Suicide
                Suicide attempt                        X
                Family violence                        X

Sexual          Abuse (sexual)                         X
                Early sexual experience                X
                Unsafe sexual behaviour
                STD                                    X

Economic        Financial hardship (family)
                Unemployment                           X
                Low income                             X
                Lack of educational
                qualifications

MENTAL HEALTH AND BEHAVIOUR RISKS

This category ranges from behavioural and conduct problems to serious mental health disorders, which can be diagnosed clinically. It also includes offending against the law, from minor delinquency to serious criminal behaviour.

Behaviour and Conduct Problems

Both the Christchurch Health and Development Study (CHDS) and the Dunedin Multidisciplinary Health and Development Study (DMHDS) found that boys were more prone to behaviour and conduct problems in childhood than girls. The patterns are established before the age of 10 (Fergusson and Horwood 1995, Fergusson and Horwood 1997, Feehan et al. 1994). These problems are strongly associated with family disadvantage and instability and have their sequel in adolescent offending and mental health problems.

Truancy is an indicator of school problems and has been linked to educational failure and under-achievement. The CHDS showed that 39% of males and 40% of females in the sample had played truant by the age of 16, so gender differences are not significant (Fergusson, Lynskey and Horwood 1995:30). Mild and occasional truancy is probably not pathological, but severe truancy is related to adjustment problems. Suspensions and expulsions from school are further measures of unacceptable behaviour. On a national basis the number of suspensions has doubled since 1990, and totalled more than 10,000 in 1998. Over 70% of the latter were of male students. The leading reasons for suspension, of all students, in 1998 were continual disobedience, physical assaults on other students and verbal assaults on staff (Ministry of Education data) (no breakdown given by gender).

Mental Health Problems

Estimates of the incidence of mental ill health among young people vary depending on how it is defined. McGeorge suggested that 5% of those aged 0-19 have mental health disorders requiring specialist assessment or treatment (O'Reilly 1996). DMHDS figures are higher (16% overall at age 13), but all agree that boys significantly outnumber girls in these statistics (Silva and Stanton 1996:155). There was again an association with family problems, such as poor maternal mental health and history of parental separation, and also with poorer language and literacy skills. These problems frequently originate in the pre- or early school periods, but tend to persist into adolescence.

The Christchurch study showed at age 15 a 26% prevalence of identifiable mental health disorders (Fergusson, Horwood and Lynskey 1993). In the Dunedin cohort, also at 15, the rate was 22% (McGee et al. 1990). This grew to 37% by age 18 (Feehan et al. 1994)(2). Alongside higher male incidence were gender differences in the conditions diagnosed. Females had higher rates for social phobia and major depression (the most common categories), but there was a male predominance in alcohol dependence, conduct disorder and marijuana dependence.

Much mental ill health goes untreated, with serious implications for well-being (see the later section on suicide). Increased rates of admission to psychiatric hospitals suggest a higher incidence of serious mental health problems among young adults. The rate of first admission is high for the 15-19 age group -- 22.5 per 1000 for males, and 18.9 per 1000 for females (Ministry of Youth Affairs 1994). The most common reason recorded for admission at this age is alcohol dependency, heavily weighted to males.

Offending/Delinquency

Adolescents, especially boys, report high levels of minor offending, such as occasional truancy and experimentation with alcohol, tobacco and cannabis. Much of this could be considered "normal" adolescent behaviour. Moffit and Harrington (1996) described it as "Adolescent Limited Delinquency". It accounts for the high proportion of males who have police contact for minor offences -- 25% of New Zealand boys aged 10 in 1967 had appeared in court before their 25th birthday (Lovell and Norris 1990). Maxwell and Morris (1993) studied juvenile offenders during the first year of the Children and Young Persons and their Families Act, 1989. Boys committed most of the more serious offences and accounted for 83% of those arrested.

Exposure to delinquent peers plus the onset of puberty, about the time of entry to secondary school, combine as important links to delinquency (Moffit and Harrington 1996:179). Girls may be involved in adolescent-limited delinquency (though not to the same extent as boys) if they experience early puberty and attend co-educational schools. Access to delinquent male role models, especially older boys, is a key factor in girls' delinquency.

Most adolescents lose their motivation for delinquency as adult roles become available and the consequences of bad behaviour threaten this status. A small proportion, however, become "Life Course Persistent" offenders and account for a high proportion of criminal offences. These are predominantly male and have poorer social skills, academic achievements and mental health than the general group. These characteristics, plus delinquent behaviour, may lead to outcomes such as drug dependency and imprisonment, which further trap the group in an antisocial life-path.

The Dunedin study compared the likelihood of delinquency behaviours between males and females (Moffit and Harrington 1996:184). The male to female ratios were 4.5:1 for childhood conduct disorder; 7:1 for adult antisocial personality and 2:1 for self-reported adolescent delinquency. Males committed a higher proportion of the more serious and violent offences, whereas for youth-oriented crime, such as using drugs, vandalism and theft, there was a more even incidence between the gender groups.

HEALTH RISKS

Health risks include smoking, cannabis, alcohol and other drug use.

Smoking

A large-scale study of fourth-formers in 1992 found that two-thirds had tried smoking and 24% were current smokers (more than one cigarette a month) (Ford et al. 1995). Not only is tobacco use high among children, but it may be increasing. A more recent survey of 1500 students aged 13-15 found 12% of the boys smoking every day (up from 8% in 1991) and 14% of the girls (up from 11%) (N.Z. Drug Foundation 1997).

As shown in Table 2, by the mid to late teenage years girls are more likely to smoke than boys, according to DMHDS figures (Silva and Stanton 1996:189). Current smoking was higher among females than males -- 27% and 22%, respectively -- in the fourth-form study cited above (Ford et al. 1995). Information from the 1996 Census also shows that more females are regular smokers (Davey 1998:103). This applies to both dependent and independent teenagers, although the latter group (living away from home and/or with their own incomes) have higher smoking rates than the former (who are more likely to be full-time students), as shown in Table 3.

Table 2 Recent Use (within a month) of Substances by Children under 15 (% of DMHDS sample)

Substance    Age   Males   Females

Tobacco      9     3       2
             11    6       7
             13    11      9
             15    21      33

Alcohol      9     47      45
             11    43      39
             13    46      45

Inhalants    13    2.5     1.2
             15    3.2     5.8
             18    2.1     1.8

Hard drugs   13    0.3     0.3
             15    3.4     2.6
             18    8.6     4.0

Source: Silva and Stanton 1996:192.

Table 3 Age Group 15-19, Percentage who are Regular Smokers, by Gender

              Male   Female   Total

Dependent     11.5   13.3     12.4
Independent   26.2   30.5     28.3
Total         18.0   20.8     19.4

Source: 1996 Census

The Dunedin study found an association between smoking before age 15 and delinquent behaviour, for both boys and girls. However, the likelihood of concurrent substance use was much higher for males, with a strong link between drinking and smoking by age 13, which did not appear for females (Silva and Stanton 1996:194).

Cannabis

Experimentation with cannabis is very common among teenagers, especially males (Black and Casswell 1992). As a threat to health it is acknowledged to be less significant than tobacco and alcohol (Abel and Casswell 1998). However, because it is illegal, cannabis use has legal consequences. At age 13, 0.8% of males and 1.3% of females in the DMHDS sample were recent users of cannabis (Silva and Stanton 1996:192). By age 15 at least 1015% had used it once or more, with no gender differences, but males were more likely to be users later in adolescence. The early use of cannabis (defined as first use before the age of 15) has been associated with behavioural problems and poor mental health, but may be a symptom rather than a cause (Fergusson, Lynskey and Horwood 1993).

Alcohol

A high proportion of New Zealand school children drink alcohol. At ages 9, 11 and 13, DMHDS figures for alcohol consumption by girls and boys were similar (Table 2). By age 18, however, alcohol consumption was much higher for males and at this stage girls reported fewer negative experiences (Silva and Stanton 1996:215-216). A 1997 survey of 14-18 year olds showed that more than half those identified as "heavier" drinkers (defined as having consumed five or more glasses the last time they drank) were boys, and these were more likely to be cigarette or cannabis users (ALAC 1997). These studies suggest that as many as 20% of teenagers are alcohol abusers or alcohol dependent, with evidence that alcohol leads to problem (and risk-taking) behaviour and violence. Nevertheless drunkenness is widely condoned throughout New Zealand society, possibly because a high proportion of adults also consume alcohol.

Papers from the CHDS show links between alcohol misuse and offending at age 14-15, applying to both males and females (Table 4). Even allowing for common risks, arising from family background, personal characteristics and peer affiliations, the association between alcohol misuse and violent offending remains (Fergusson, Lynskey and Horwood 1996). Being male was also a predictor of self-reported abusive/hazardous drinking at age 16 (Fergusson, Horwood and Lynskey 1995).

Table 4 Comparison of Rates (per 100) of Violent and Property Offending among Males and Females who Reported Alcohol Misuse and Remaining Sample Members (age 14-15)

                             Males
                    Alcohol misuse   No misuse

Violent offences         32.1           7.6
Property Offences        45.3          12.4
N                         53            42

                           Females
                    Alcohol misuse   No misuse

Violent offences         15.4           3.1
Property Offences        50.0           7.3
N                        126            453

Source: Fergusson and Lynskey and Horwood 1996.

Other Drug Use

Adolescents, both males and females, who continue to smoke are also more likely to try cannabis, inhalants and hard drugs regardless of whether they continue to drink alcohol (Silva and Stanton 1996:200). Table 2 shows that hard drug use by young teenagers was very low in the Dunedin study, but had increased by age 18 and was higher for males at that age (as was the use of inhalants).

PHYSICAL RISKS

Physical risks include accidental and intentional injury and death, suicide, attempted suicide, sexual risk, abuse and unsafe sexual behaviour.

Accidental and Intentional Injury and Death

Research involving discussion groups of teenagers came to the conclusion that behaviour which involves taking risks, and therefore encountering potential threats to well-being, is normal behaviour at this stage of life (Colmar Brunton Research 1993a, 1993b). Younger teenagers focus on thrill seeking -- like riding skateboards and bikes -- but by age 16 most had moved on to experimenting with alcohol and drugs (mainly cannabis). Boys are more susceptible to thrill seeking, but girls are also at risk. As a result, accidents are the primary cause of adolescent physical incapacity and death. They account for 46% of hospital attendance by males aged 15-19, and are the second largest category for females (after conditions associated with pregnancy) (Davey 1998:109).

In 1994, accidents accounted for over 80% of deaths of teenagers, with male rates exceeding female rates by a considerable amount. The figures are dominated by road deaths, and gender differences are very clear in the 15-19 age group (Table 5). Numbers killed and injured and rates per 100,000 are both higher for males than for females. It is possible that some road deaths involving young males may, in fact, be suicides -- see the later section on suicide.

Table 5 People Aged 15-19 Killed and Injured in Motor Accidents, 1991-1996, by Gender

Year          Killed                  Injured
       Male   Female   Total   Male   Female    Total

1991    74      27      101    2183    1234     3420
1992    76      29      105    1940    1142     3087
1993    69      28      97     1764    1084     2851
1994    42      22      64     2009    1201     3215
1995    58      24      82     1913    1253     3169
1996    70      25      95     1663    1010     2675

Source: Land Transport Safety Authority.

After road crashes, sports injuries are the second leading cause of ACC claims for adolescents, followed by home and work-related injuries. Work-related injury rates for males are double those for females, related to the greater concentration of males in dangerous manual occupations, such as construction and forestry. The pattern is the same for sports injuries. Male rates were 9.9 per 1000 for 10-14 year olds and 23.9 for 1519 year olds, in contrast to the rates for females -4.9 and 8.4 respectively (Department of Health 1992). This again is related to the more dangerous nature of male sports.

Two-thirds of homicide victims in New Zealand are male and the age group most at risk is 15-24 (Injury Prevention Research Centre 1995). Young people, especially those aged 15-24, have the highest rates for crime victimisation, especially violent or sexual crime and are prone to repeat victimisation (Young et al. 1997:34). This age group also reports the highest level of violence between domestic partners. In the Dunedin cohort at age 21, 37% of the women and 22% of the men reported experience of partner violence (Magdol et al. 1997). Young women not only report higher levels of physical assault, but also more serious injury resulting from it (Langley et al. 1997).

Suicide

Suicide rates in New Zealand have grown rapidly in the last decade and the main age group at risk is 15-24. The majority of suicides in all age groups are males, accounting for 80% of the total. One in four young males who die between 15 and 24 years commits suicide. In 1994 the annual mortality rate from suicide in this age group was 39.9 per 100,000 for males and 9.7 for females, compared to 22.9 and 8.0, respectively, in 1986 (Ministry of Health 1997).

International comparisons highlight New Zealand's high rates of youth suicide (Injury Prevention Research Centre 1995:38-39, Drummond 1996). Various risk factors have been identified (Rivers 1995). A very high proportion of young people who commit suicide have a diagnosable mental disorder. Some personality traits -- withdrawal, perfectionism, aggressiveness and hopelessness -- may predispose people to suicide. There are also social factors -- violence, family disruption and dysfunction. Childhood sexual, physical or emotional abuse also seems to be a risk factor. Personal behavioural factors -- alcohol drug use, impulsive and anti-social behaviour, stressful life events -- also play a part.

Attempted Suicide

Along with completed suicides, hospitalisation rates for intentional self-injury or unsuccessful suicide attempts have also increased, although the rate per 100,000 has not changed greatly. Females account for almost two-thirds of individuals admitted to hospital after attempted suicide, with the leading age group for both sexes being 15-24 (Disley 1994, Injury Prevention Research Centre 1995). This is associated with different patterns of suicide methods. Males are more likely to use hanging or shooting, as well as deliberate vehicle crashes. Females are more likely to use overdoses of drugs or medication, from which they are more easily rescued.

Between the ages 14 and 16 years, 4.2% of girls and 1.9% of boys in the Christchurch sample reported a suicide attempt and 12% had considered suicide (Fergusson and Lynskey 1995a and b). These were mainly minor incidents, few requiring medical treatment, indicating a large degree of under-reporting in official figures. National figures also show that these attempts are more common among females. Adolescent suicide attempts have been linked to psychiatric disorders, especially depression and anxiety and to other adjustment problems -- offending, substance abuse and school drop out. Fergusson and Lynskey (1995a) suggest that a common pathway of childhood disadvantage, including poverty, parental substance abuse or offending, marital conflict and family instability, leads to adolescent suicide attempts and psychiatric and adjustment problems.

Sexual Risk

There is a trend towards earlier sexual intercourse since the 1970s. It is common for teenagers to have multiple partners, but the majority use contraception and teenage fertility has fallen. Concerns about young people's sexuality may be based more on morality than threats to well-being, although girls especially are vulnerable to sexual abuse, and risk-taking behaviour, such as alcohol and drug abuse, may lead to unwanted pregnancy and sexually transmitted diseases (STDs).

Several studies show that early involvement is most common among girls. In the CHDS sample 10% of the girls and 7% of the boys reported sexual activity by the age of 15 (Lynskey and Fergusson 1993). The DMHDS sample was 18 in 1991-1992. At this stage 58% of males and 68% of females had had sex in the previous 12 months (Silva and Stanton 1996:227). A higher percentage of females were in stable relationships and 36% had had only one sexual partner, compared to only 23% of males.

Factors associated with early sexual activity include leaving school early, lower school qualifications and living away from parents. These were more important than socioeconomic status (Silva and Stanton 1996:232). Fergusson and Lynskey (1996) found an apparent correlation between early onset of sexual activity and alcohol misuse. Boys and girls in the Christchurch sample who were classified as alcohol misusers were much more likely to have had sexual experience before 16 than non-misusers, and to have indulged in risky sexual behaviour, such as having multiple partners and unprotected intercourse. In addition to the disinhibiting effects of alcohol, this seems to arise through the influence of common family, individual and peer factors, including childhood adversity, novelty- seeking behaviour, and affiliation with delinquent peers.

Abuse

Research on childhood sexual abuse (CSA) shows that females are more at risk. Among the CHDS cohort, interviewed at 18, 17.3% of the females and 3.4% of the males had experienced CSA before the age of 16 (Fergusson, Horwood and Lynskey 1996a). Nearly 6% of females in the total sample and 1.4% of the males reported abuse involving attempted or completed intercourse.

CSA was found to be a significant risk factor for later psychiatric disorder in the CHDS study, but social disadvantage, family dysfunction and adverse childhood events also contributed (Fergusson, Horwood and Lynskey 1996a). Young women recording CSA, especially severe CSA, had greater sexual vulnerability in adolescence. They had higher rates of early sexual activity, teenage pregnancy, multiple sexual partners, unprotected intercourse, and STDs. For example, the rate for more than five sexual partners by age 18 was 14% for girls who had not experienced child sexual abuse and 52% for those who had experienced serious abuse.

Unsafe Sexual Behaviour

Well over 80% of sexually active teenagers in the Christchurch and Dunedin samples had used a recognised form of contraception (Silva and Stanton 1996:230). Unsafe sexual behaviour - multiple partners and never using condoms - was reported by 9.3% of the DMHDS sample, more males than females (Silva and Stanton 1996:232). These were the less traditional, more forceful and dominant, more alienated teenagers, and resembled study members who engaged in other dangerous behaviour such as unsafe driving, drug use and crime. In the CHDS, at age 16, 59.8% of the males and 46.5% of the females (who were sexually active) were classified as having had multiple partners. Alcohol misuse may place teenagers at greater risk of unprotected intercourse. Among alcohol misusers, 28% of boys and 42% of girls reported unprotected intercourse, but for those who did not misuse alcohol the figures were 2% and 7% respectively (Fergusson and Lynskey 1996). There was an association between STDs and multiple partners for males and females (Fergusson, Horwood and Lynskey 1997). In Dunedin, STDs were reported by 12% of sexually active women and 2% of sexually active men at age 18 (Silva and Stanton 1996:230). There are, therefore, risks to both males and females from sexual activity.

ECONOMIC RISK

Economic risks include low income and unemployment.

Income Levels

Financial hardship constitutes a threat to the well-being of individuals and families and was the most prevalent risk factor identified by Denny, applying to both girls and boys in New Zealand secondary schools (Denny 1994). For younger adolescents, and those aged 15-19 who remain in the parental household, there are no significant differences in household income by gender (Davey 1998). But among "independent" teenagers, who either live away from home or have independent incomes, it is clear that females predominate in low-income households (quintiles 1 and 2) and a higher proportion of males are in households above the mean level (Table 6).

Table 6 Independent Teenagers aged 15-19, Percentage in Each Quintile of Household Income, by Gender

Quintile   Male   Female

1 (low)    12.9    19.7
2          16.1    19.7
3          17.8    17.6
4          22.7    19.0
5 (high)   30.1    24.0

Source: From Birth to Death database (Davey 1998)

The pattern is similar for personal incomes (Davey 1998). Taking the 15-19 age group as a whole, males and females together, incomes are generally low. In 1996, 20% had no income and over half had under $10,000 per annum (13% of this age group did not respond to the census question). Females predominate in the categories with no income and incomes to $5000, with males much more likely to have personal incomes from $15,000 upwards. In this case it is possible to explain the gender difference by higher levels of educational participation by young women, but gender differences in household incomes for independent teenagers are more difficult to explain.

Unemployment

Since 1981, full-time work involvement has fallen for both males and females in the 1519 age group, although participation levels remain higher for males (Table 7). Part-time work has increased, especially since 1991, and the female levels are only slightly higher than for males. In the 1980s, young women were more likely than young men to be out of the labour force altogether, but by 1996 the figures were similar for both groups, with a higher percentage of young women in the labour force and a lower percentage of young men (Table 7). This is despite higher female retention in education. It is likely that many in the part-time labour force are also students, seeking to meet some of the costs of study. Table 7 also shows a consistent trend for higher levels of unemployment among young women, compared to young men, according to census information (unemployment people are counted as part of the labour force).

Table 7 Age Group 15-19 Employment Status, 1981-1996, by Gender

Figures are percentage of age group, by gender

                      1981        1986        1991        1996
                    M     F     M     F     M     F     M     F

Full-time work      50    40    47    40    26    21    25    17
Part-time work       7     8     8     9    12    14    26    31
Unemployed           7     9    12    13    18    19    11    13
Non-labour force    36    43    33    38    44    46    39    40
Total              100   100   100   100   100   100   100   100

Source: From Birth to Death databases.

The association between unemployment and psychiatric disorder was examined in the CHDS at age 18 (Fergusson, Lynskey and Horwood 1997). The results showed increased risks of psychiatric disorder in young people who had been unemployed for six months or more. Family and personal factors, such as low socio-economic status, Maori or Pacific Island ethnicity, family change and parental conflict, lower IQ and substance abuse, appeared to explain most of this. It is possible, therefore, that individual vulnerability to psychiatric disorder may be established well before entering the labour force. The links between youth unemployment and rates of psychiatric disorder remain contentious.

CONCLUSION AND POLICY IMPLICATIONS

Table 8 brings together material from the CHDS, the DMHDS and other New Zealand studies, summarised above. It lists the linkages that have been observed in this literature between risk factors and other characteristics, especially those with notable gender differences. The findings show, firstly, that a set of individual, social and family conditions appear to increase individual susceptibility to a wide range of adolescent disorders and adjustment problems. Secondly, these teenage problems are frequently correlated or "co-morbid".

In reviewing recent New Zealand research findings, this paper has described a wide range of risk factors that confront young people in their teenage years. Most are encountered by both males and females, but often the risk falls predominantly on one group. Males are certainly more prone to behavioural and conduct problems and to other risks which may arise from these, such as mental ill health, substance dependence and criminal offending. Greater risk-taking among males brings them into physical danger. On the other hand, sexual risks and family violence pertain especially to females and they are subject to some types of economic risk to a greater extent than males.

Again and again in the above analysis, the influence of home and family circumstances has been pinpointed. "Seriously disadvantaged, dysfunctional and disorganised home environments" is a code phrase for a set of factors that include economic circumstances, parental behaviour and personal characteristics, and family history. No evidence is presented to show that either sex is disproportionately exposed to such adverse environments. A further factor or factors must be present to explain gender-based differences in behaviour. The answers may lie not only within the young people themselves, in terms of psychological and personality factors, but also in their family backgrounds and in society at large.

Exploration of gender-based differences in behaviour and susceptibility to risk factors is therefore a fruitful area for further research, focusing on the young people themselves and their family backgrounds. This should bring in factors other than gender, such as socio-economic status and ethnicity. Gender differences in educational performance and attainment have also been brought to the fore recently. An Education Review Office report, released in July 1999, noted that girls out-perform boys in educational performance in most countries, but that New Zealand gender differences are among the largest found internationally (Minister of Education Press Release 29 July 1999, Wagemaker 1993). Further evidence of gender differences comes from school retention rates, qualifications and grades achieved, all showing higher levels of female achievement compared to males (Davey 1998, Hughes et al. 1996). The suggestion that males may now be at an educational disadvantage contrasts with the situation ten years ago when there was concern about educational opportunities for girls (RCSP 1988:81).

Many factors have been associated with educational performance, including socioeconomic influences, ethnicity and type of school (Nicholson and Gallienne 1995). This paper asks whether the relative educational under-achievement of males may be linked to their exposure to risk factors which influence their behaviour and hence their school performance. At the same time, however, higher educational participation and attainment does not appear to be delivering improved outcomes for young women in terms of economic well-being. Their income levels in the late teenage years are lower than for males on average. They are more likely to be unemployed. In later life women are more likely than men to become dependent on welfare benefits. In order to throw light on these distinctions, wider societal factors, such as sex-role stereotyping and gender discrimination, need to be examined.

Gender analysis is a basic tool in policy work and disaggregation by gender is fundamental to an understanding of social trends. This paper has shown that for a key group in society - adolescents, who will be the adults of tomorrow - gender differences in terms of susceptibility to risk are significant and relevant to policy. However, we should not assume that the risk factors that have been identified systematically disadvantage either males or females. The pattern is much more complex and requires closer scrutiny in the context of society as a whole.

Table 8 Linkages between Risk Types (from New Zealand literature)

                Mental Health and Behaviour

Early           Later substance use and abuse (alcohol,
conduct         cigarettes, illicit drugs)
problems        Attempted suicide and suicide ideation (because
                of common risk factors)
                Academic difficulties and failure,
                Increased risk of juvenile offending
                Mental health problems
                Seriously disadvantaged, dysfunctional and
                disorganised home environments

Truancy         Disadvantaged and unstable family background,
                family history of drug/alcohol problems
                Juvenile offending, substance abuse, mental health
                problems (severe truancy)

Delinquency     Substance abuse
                Mental health problems
                Exposure to delinquent peers
                Unemployment at 15
                Seriously disadvantaged, dysfunctional and
                disorganised home environments

Mental          Early conduct problems
health          Delinquency
problems        Socio-economic disadvantage, marital dysfunction,
                impaired parenting, exposure to abuse, parental
                psychopathology (for conduct disorders and
                substance abuse)
                Socio-economic disadvantage, family history of
                depression, exposure to childhood adversity,
                predisposing factors within child, such as
                hyperactivity (for depression, affective disorders).

                Health Risk

Alcohol         Property and violent offending
misuse          Early sexual activity, unprotected intercourse
                Cannabis use

Smoking         Parental and peer group smoking
                Other and subsequent substance use (cannabis,
                inhalant, hard drugs)

Cannabis        Other substance abuse, juvenile offending,
use             delinquent and substance abusing peers, mental
                health problems
                Unemployment
                School drop-out

                Sexual Risk

Childhood       Increased risk of psychiatric disorder
Sexual          Greater sexual vulnerability during adolescence
Abuse           (female)
                Early sexual activity (before 16), early pregnancy
                Social disadvantage and family instability
                (families with high levels of marital conflict,
                impaired parenting/bonding, parents with
                adjustment problems - alcohol, drugs, criminal
                offending)

Premature       Seriously disadvantaged, dysfunctional and
sexual          disorganised home environments
experience

Unsafe sexual   Early school-leaving, lower school qualifications
behaviour       and living away from parents
                Multiple sexual partners
                Less traditional, more forceful and dominant and
                more alienated personalities
                Other dangerous behaviour such as unsafe driving,
                alcohol of drug problems and involvement in crime.

                Physical Risk

Suicide and     Diagnosable psychiatric disorder
suicide         Substance abuse problem
attempt         Problems of personal adjustment
                Juvenile offending
                Seriously disadvantaged, dysfunctional and
                disorganised home environments

Abuse in        Juvenile offending
childhood       Substance abuse
                Mental health problems
                Violent behaviour

                Economic Risk

Unemploy-       Increased risk of psychiatric disorder (anxiety
ment            disorder and substance use)
                Increased risk of offending, arrest and convictions

(1) For example, in their post-election briefing papers in 1999 the Ministry of Social Policy suggested that about half of New Zealand families are at little or no risk of poor outcomes for their children (examples of poor outcomes for children included school failure and unemployment). Forty-five per cent of families are at some risk and another 5% exhibit "persistent, multiple and serious disadvantage", which considerably heightens their "at risk" status (Ministry of Social Policy 1999:7). These estimates were derived mainly from the work of David Fergusson, using the Christchurch Health and Development Study.

(2) These studies used the DSM-III-R set of criteria for measuring mental health, but the results may not be strictly comparable.

REFERENCES

Abel, Sally and SallyCasswell (1998) "Cannabis Policy in New Zealand: Policy and Prospects" Social Policy Journal of New Zealand, Issue 10:70-85

ALAC (1997) Youth and Alcohol Survey. Alcoholic Liquor Advisory Council Occasional Paper No. 1, Wellington.

Black, S. and S. Casswell (1993) Drugs in New Zealand. A Survey, 1990. Alcohol and Public Health Research Unit, University of Auckland.

Colmar Brunton Research (1993) Qualitative Research on Teenage Alcohol and Drug Related Injuries for Injury Prevention Communications, ARCIC, Wellington.

Colmar Brunton Research (1993a) Literature Review on Adolescent Alcohol/Illicit Drug Related Injuries, ARCIC, Wellington.

Davey, J.A. (1998) Tracking Social Change in New Zealand: From Birth to Death IV, Institute of Policy Studies, Victoria University of Wellington.

Denny, Gillian (1993) Indicative Survey of At Risk Students in New Zealand Secondary Schools, unpublished paper held in the National Library, Wellington.

Department of Health (1992) Adolescent Health; Potential for Action, Department of Health, Wellington.

Disley, B. (1994) "Suicide: An overview of Research in New Zealand" Community Mental health in New Zealand, 8 (2):5-11.

Drummond, W. (1996) Suicide New Zealand: Adolescents at Risk, Nagare and BCU Press, Palmerston North.

Feehan, M., R. McGee, S.N. Raja, and S.M. Williams (1994) "DSM-111-R disorders in New Zealand 18 year olds" Australian and New Zealand Journal of Psychiatry, 28 (1):87-99.

Fergusson, D.M., and L.J. Horwood (1995) "Early Disruptive Behavior, IQ, and Later School Achievement and Delinquent Behaviour" Journal of Abnormal Child Psychology 23 (2):183-199

Fergusson, D.M., and L.J. Horwood (1997) "Gender Differences in Educational Achievement in a New Zealand Birth Cohort" New Zealand Journal ooEducational Studies, 32 (1):83-96.

Fergusson, D.M., L.J. Horwood and M.T. Lynskey (1993) "The prevalence and comorbidity of DSM-III-R diagnoses in a birth cohort of 15 year olds" Journal of the American Academy of Child and Adolescent Psychiatry, 32: 1127-1134.

Fergusson, D.M., L.J. Horwood and M.T. Lynskey (1995) "The prevalence and risk factors associated with abusive or hazardous alcohol consumption in 16-year olds" Addiction 90:935-946.

Fergusson, D.M., L.J. Horwood and M.T. Lynskey (1996a) "Childhood Sexual Abuse and Psychiatric Disorder in Young Adulthood: Part 1: The Prevalence of Sexual Abuse and Factors Associated with Sexual Abuse" Journal of the American Academy of Child and Adolescent Psychiatry, 35:1355-1364.

Fergusson, D.M., L.J. Horwood and M.T. Lynskey (1996b) "Childhood Sexual Abuse and Psychiatric Disorder in Young Adulthood: Part 2: Psychiatric Outcomes of Childhood Sexual Abuse" Journal of the American Academy of Child and Adolescent Psychiatry, 35:1365-1374.

Fergusson, D.M., L.J. Horwood and M.T. Lynskey (1997) "Childhood Sexual Abuse, Adolescent Sexual Behaviors and Sexual Revictimization" Child Abuse and Neglect, 21 (8):789-803.

Fergusson, D.M., and M.T. Lynskey (1995a) "Childhood Circumstances, Adolescent Adjustment, and Suicide Attempts in a New Zealand Birth Cohort" Journal of the American Academy of Child and Adolescent Psychiatry 34(5):612-622.

Fergusson, D.M., and M.T. Lynskey (1995b) "Suicide Attempts and Suicide Ideation in a Birth Cohort of 16-year-old New Zealanders" Journal of the American Academy of Child and Adolescent Psychiatry, 34 (10):1308-1317.

Fergusson, D.M., and M.T. Lynskey (1996) "Alcohol Misuse and Adolescent Sexual Behaviours and Risk Taking" Pediatrics, 98 (1):91-96.

Fergusson, D.M., M.T. Lynskey, and L.J. Horwood (1993) Patterns of cannabis use among 13-14 year old New Zealanders New Zealand Medical Journal 106:247-250

Fergusson, D.M., M.T. Lynskey, and L.J. Horwood (1995) "Truancy in Adolescence" New Zealand Journal of Educational Studies, 30 (1):25-38.

Fergusson, D.M, M.T. Lynskey and L.J. Horwood (1996) "Alcohol Misuse and Juvenile Offending in Adolescence" Addiction. 91 (4):91-96.

Fergusson, D.M, M.T. Lynskey and L.J. Horwood (1997) "The Effects of Unemployment on Juvenile Offending" Criminal Behaviour and Mental Health, 7: 49-68.

Ford, D.J., R. Scragg, J. Weirand and J. Gaiser (1995) "A national survey of cigarette smoking in fourth form school children in New Zealand" New Zealand Medical Journal, 108, (1011):454-457.

Hughes, D., H. Lauder and R. Strathdee (1996) "First-Year University Performance as a function of type of Secondary School Attended and Gender" New Zealand Journal of Educational Studies, 31 (1):113-28.

Injury Prevention Research Centre (1995) Intentional Injury in New Zealand, Public Health Commission, Wellington.

Langley, J., J. Martin and S. Nada-Raja (1997) "Physical Assault among 21-year-olds by Partners" Journal of Interpersonal Violence, 12,(5):675-684.

Lovell, R. and M. Norris (1990) One in Four: Offending from Age Ten to Twenty-four in a Cohort of New Zealand Males, Wellington: Study of Social Adjustment Research Report No.8, Research Section, Department of Social Welfare, Wellington.

Lynskey, M.T. and D.M. Fergusson (1993) "Sexual Activity and Contraceptive Use amongst Teenagers under the age of 15 years" Journal of the New Zealand Medical Association, 106, (969): 511-514.

Magdol, L., T. Moffitt, A. Caspi, D. Newman, J. Fagan and P. Silva (1997) "Gender Differences in Partner Violence in a Birth Cohort of 21 year-olds: Bridging the Gap between Clinical and Epidemiological Approaches" Journal of Consulting and Clinical Psychology, 95 (1):6878.

Maxwell, G. and A. Morris (1993) Family, Victims and Culture: Youth Justice in New Zealand, Social Policy Agency and Institute of Criminology, Victoria University of Wellington, Wellington.

McGee, R., M. Feehan, S. Williams, F. Partridge, P.A. Silva and J. Kelly (1990) "DSM-III disorders on a large sample of adolescents" Journal of the American Academy of Child and Adolescent Psychiatry, 29:611-619.

Ministry of Health (1997) Mortality and Demographic Data 1994, New Zealand Information Service, Wellington.

Ministry of Social Policy (1999) Post Election Briefing Papers: Supporting Families and Whanau, Ministry of Social Policy, Wellington.

Ministry of Youth Affairs (1994) 15 to 25: A Youth Statistical Profile, Ministry of Youth Affairs, Wellington.

Moffit, T. and H. Harrington (1996) "Delinquency: The Natural History of Antisocial Behaviour" Chapter 10 in P.A. Silva and W. Stanton (eds.) From Child to Adult, Oxford University Press, Auckland.

Nicholson, T., and G. Gallienne (1995) Struggletown meets Middletown: A survey of Reading Achievement Levels among 13-year-old pupils in two contrasting socio-economic areas, New Zealand Journal of Educational Studies, 30 (1):15-23.

O'Reilly, L.M. (1996) "Students at Risk" Children 21, 5-8, Office of the Commissioner for Children, Wellington.

Rivers, L. (1995) Young Person Suicide: Guidelines to Understanding, Preventing and Dealing with the Aftermath, Special Education Service, Wellington.

RCSP (1998) The April Report, Vol.3, Part 2, Royal Commission on Social Policy, Wellington.

Silva, Stanton (1996) From Child to Adult: The Dunedin Multidisciplinary Health and Development Study, Oxford University Press, Auckland.

Wagemaker, H. (1993) Achievement in Reading Literacy: New Zealand's Performance in a National and International Context, Research Section, Ministry of Education, Wellington.

Young, W., A. Morris, N. Cameron and S. Haslett (1997) New Zealand National Survey of Crime Victims, Victoria University of Wellington.

Judith Davey Department of Sociology and Social Policy Victoria University of Wellington