Child and adolescent mental health care in Dutch general practice: time trend analyses

To investigate the extent to which GPs identify child and adolescent mental health problems, prevalence rates of recorded mental health problems (ICPC-codes within the P-chapter) were determined for 2004-2008. The effects of patient age and gender on recorded prevalence rates, rates of referral and prescriptions in 2008 were analysed using multivariate multilevel logistic regression analyses with two-level hierarchical structured data (patients within general practices), using MLWin 2.02. Time trends in prevalence rates were analysed using multilevel repeated measures logistic models with three-level hierarchically structured data (occasions nested within patients, and patients nested within general practices). Models were adjusted for patient gender and the practice's length of recording. Time trends in prescriptions and referrals were analysed by multilevel logistic regression, using a compound-symmetry model with three-level hierarchically structured data. Models were adjusted for patient gender. Analyses were performed for children (0-12 years) and adolescents (13-18 years) separately.

Rates of children and adolescents who were recorded as having a mental health problem are reported in Table 2. In 2008, 6.6% of children and 7.5% of adolescents were recorded as having a mental health problem (ICPC-code P01-P99). Of the separate diagnoses selected for this study, overactivity was recorded most frequently over the years. Apart from these selected diagnoses, other frequently recorded diagnoses within the P-chapter were enuresis (P12), other concern with behaviour of child (P22), and specific learning problems/delay in development (P24).

The effects of age and gender on the prevalence of recorded mental health problems are reported in Table 3. Girls had an increased chance of being recorded as having anxiety or depression, and a decreased chance of being recorded as having overactivity. Compared to children, adolescents had an increased chance of being recorded as having anxiety or depression. After adjusting for gender, age did not significantly affect the prevalence of recorded overactivity.

The prevalence of recorded mental health problems (P01-P99) showed a significant linear increase between 2004 and 2008, for both children and adolescents (Table 2). A similar increase was found for the prevalence of anxiety in children and alcohol abuse in adolescents. The prevalence of recorded anxiety and depression in adolescents showed a significant quadratic trend, whereas no significant trend was found for the prevalence of depression in children and the prevalence of overactivity in both age groups.

Percentages of children and adolescents who were prescribed psychotropic medication are reported in Tables 4 and 5. In 2008, 15.2% of children and 29.4% of adolescents with mental health problems were prescribed psychotropic medication. In 2008, the majority of children (60.4%) and adolescents (74.7%) who were recorded as having overactivity received a prescription for psychotropic medication. Prescription rates were lower for anxiety and depression (Table 5). Compared to children, adolescents had an increased chance of receiving any prescription for psychotropic medication (see additional file 1).

The overall percentage of children with mental health problems receiving a psychotropic prescription did not change significantly over the years, whereas a slight decrease was found for adolescents (Table 4). The percentage of children with mental health problems who were prescribed psychostimulants increased significantly, from 8.0% in 2004 to 12.2% in 2008. When only children with a primary care diagnosis of overactivity were considered, a similar significant increase in prescriptions for psychostimulants was found (chi-square: 5.53; p = 0.02). No significant linear trend in prescriptions of psychostimulants was found for adolescents. Prescriptions for antidepressants decreased over time in both age groups. Prescriptions for hypnotics and sedatives increased over time for children, but not for adolescents, while prescriptions for anxiolytics decreased over time for adolescents, but not for children.

Percentages of children and adolescents who were referred to primary or secondary mental health care are reported in Tables 6 and 7. In 2008, 18.9% of children and 22.9% of adolescents with mental health problems were referred to any kind of primary or secondary mental health care (Table 6). Compared to children, adolescents had an increased chance of being referred to primary care (see additional file 1). Over the years, children and adolescents who were referred to other health care providers, were significantly more often referred to secondary than to primary mental health care (p < 0.05).

Between 2004 and 2008, the percentages of children and adolescents who were referred to any mental health care provider increased significantly (Table 6). In primary care, this increase could mainly be ascribed to an increase in referrals to primary care psychologists for children. Over the years, adolescents were increasingly referred to psychotherapists, and decreasingly to organisations for ambulatory mental health care.

Psychotropic medication and mental health care referrals were mainly offered separately. Only a minority of children and adolescents (in 2008: 2.1% of children and 4.5% of adolescents) received both types of treatment. A considerable number of youths with mental health problems (in 2008: 66.5% of children and 50.3% of adolescents) received neither psychotropic medication nor were referred to primary or secondary mental health care.

In 2008, 6.6% of children and 7.5% of adolescents were recorded by their GP as having a mental health problem. This is in sharp contrast with prevalence rates of 14-22% reported previously for mental health problems in the general child and adolescent population [2‐7]. A recent study in an adult population (18-65 years) [Verhaak et al., unpublished data] showed a much smaller difference between the prevalence of mental health problems in general practice (12.5%) and the general population (17.5%). In line with previous research [18, 21, 24, 26], these results may indicate that a number of youths with mental health problems are not identified by their GP. Although not all children and adolescents with mental health problems will need additional care, it is important to identify the ones who do.

The difference between the prevalence rates of child mental health problems recorded in general practice and in the general population may partly be explained by the fact that not all youths with mental health problems and/or their parents experience a need for care and/or consult their GP. Parents and youths may not be aware that GPs are an appropriate source of care for mental health problems [25]. Particularly adolescents may be reluctant to seek help for mental health problems [37] and to mention their mental health problems to their GP, for instance because of fear about a lack of confidentiality [25]. This may affect adolescents' chance of being identified as having mental health problems by their GP. However, our data did not confirm this, as the prevalence of recorded mental health problems was not found to be lower for adolescents than for children. This may be an indication of parents' continuing influence on the help-seeking process [19].

When youths or parents do consult their GP, mental health problems are rarely being presented as the main reason for the consultation [41‐43]. Even if patients do express their concerns, they generally do this in the context of physical problems, instead of directly relating to the actual problems [22, 44]. The limited length of consultations may contribute to this reluctance, as more time is needed to discuss psychological issues [45‐47].

Alternatively, GPs may be reluctant to diagnose a child or adolescent with mental health problems because of the stigma associated with the diagnosis. As in many countries, demand for child and adolescent mental health care in the Netherlands exceeds supply, which results in considerable waiting lists [48]. Particularly when the likelihood of obtaining specialist mental health care following referral is limited due to these waiting lists, GPs may not see the benefit of recording youths as having a mental health problem.

The prevalence of child and adolescent mental health problems recorded in general practice increased between 2004 and 2008, which may indicate that the identification of these problems by GPs has improved. It may also indicate that a previously found increase in the prevalence of Dutch children's and adolescents' mental health problems between 1993 and 2003 [3, 49] has persisted in later years, or that society's attitude towards mental health problems has changed, decreasing the stigma being attached to these problems and to seeking professional help.

In 2008, 15.2% of children and 29.4% of adolescents with mental health problems were prescribed psychotropic medication. Over the years, psychostimulants were prescribed most frequently. Previous research showed an increase in the use of psychostimulants by children and adolescents during the last two decades [33]. Although these previous results were based on pharmacy dispensing records, which include prescriptions from GPs and psychiatrists, we found a similar increase in the prescriptions for psychostimulants by GPs for children aged 0-12 years. This increase was not reflected by an increase in overactivity recorded in this age group, which may indicate that psychostimulants are increasingly being prescribed to overactive children. Our results further indicate that a previously found decrease in GPs' prescriptions of antidepressants to children and adolescents between 2001 and 2005 [50] persisted in later years. This may be due to the increasing consensus that antidepressant medication should not be used as first-line treatment for children and adolescents with mild to moderate depression, because of its limited - or in some cases even adverse - clinical effects [51, 52].

In 2008, 18.9% of children and 22.9% of adolescents with mental health problems were referred to any kind of mental health care. Despite recent reforms in Dutch mental health care, which aimed to make Youth Welfare Work Offices the only entrance point into child mental health care [36], we found an increase in the overall percentage of youths being referred to mental health care by their GP, which indicates that GPs still function as gatekeepers to child mental health care.

Over the years, only a minority of children and adolescents with mental health problems were referred to other health care providers. A considerable number of youths with mental health problems (in 2008: 66.5% of children and 50.3% of adolescents) received neither psychotropic medication nor were referred to primary or secondary mental health care. This may indicate that child and adolescent mental health problems are being undertreated. However, not all youths with mental health problems will need additional care. Watchful waiting may be the appropriate strategy for youths with mild mental health problems [29]. Moreover, these children and adolescents may have been offered some kind of psychological management (e.g. counselling or short-term therapy) in general practice. Data about psychological treatment provided in general practice are lacking, however. Children and adolescents, who were referred to other professionals, were mainly referred to secondary instead of primary care. Although this may be appropriate for a minority of youths with severe mental disorders, less severe mental health problems should preferably be treated in primary care [29].

The percentage of children being referred to primary mental health care providers increased over the years. This could be a result of Dutch policy measures to strengthen primary mental health care. However, an increase was also found for referrals of children to secondary mental health care. The latter finding is in line with time trend studies in other European countries, which showed an increase in the use of specialist child mental health care services over the years [53, 54].

Although differences between health care systems require caution to be exercised in generalising our results to other countries, we believe that our study provides valuable insights into the role that GPs can play in the identification and treatment of children and adolescents with mental health problems, which may be useful for other countries in which GPs function as gatekeepers. Many of the issues covered here are also relevant for countries with different health care systems. For instance, in the US, community child health services are provided by paediatricians, who are based within primary care. Children and adolescents with mental health problems have been shown to be frequent attenders in primary care [25, 27], which underlines the importance of the primary care setting in the identification and treatment of child and adolescent mental health problems.

Our time trend analyses provide insight into changes in the role of GPs in child mental health care over time, which may be a result of efforts to strengthen primary mental health care. However, to evaluate whether these time trends can actually be ascribed to the Dutch measures to strengthen primary mental health care, a more rigorous study design (e.g. an interrupted time series design with multiple measurements before and after the implementation of the policy measures) would be required. Conducting comparative studies in countries in which similar initiatives have been taken, may provide valuable suggestions about ways to strengthen primary mental health care.

Our results are based on a large dataset, which is representative for the Dutch population and for Dutch general practices regarding the degree of urbanisation and region [38]. GPs in LINH are trained in coding and data are checked for completeness and reliability. GPs traditionally did not record mental health referrals in their clinical work, but are asked to record these referrals as part of LINH. Therefore, data regarding mental health referrals may have been underestimated. Moreover, recording of referrals to social psychiatric nurses, psychotherapists and organisations for addiction treatment is optional, which may have caused these referrals to be underestimated.

Our results regarding the treatment of mental health problems in general practice refer to prescriptions for psychotropic medication and referrals to primary and secondary care. They do not indicate whether any kind of psychological management (e.g. counselling or short-term therapy) has been offered in general practice. With regard to the prescription data, it is unknown whether the GP was the initiator of the prescription or whether the GP repeated a prescription that was initiated by a secondary care health care provider.

We did not evaluate the accuracy of GPs' diagnoses of child and adolescent mental health problems by comparing these diagnoses with a gold-standard diagnostic test. Such comparisons would be useful to investigate whether the prescriptions and referrals matched the severity of the child's problems.