Principal findings
As anticipated, based on prior analysis of barriers to implementing youth depression guidelines [
15], while practice is broadly in line with guideline recommendations, significant improvements are required in some areas to ensure recommendations are followed appropriately.
First, the current study shows less than 100% concordance with the assessment and recording of depression severity, despite treatment recommendations being largely determined by severity. Depression severity is often simply gauged by clinical judgement using DSM/ICD; however, this was seldom done and even in the context of a mandated tool (though not a specific depression scale), the lack of uniformity in establishing severity striking. In no case was there evidence that a depression scale measure was used.
Second, the majority of young people who were antidepressant naïve on entry to the service did not appear to have an adequate trial of psychological therapy before prescription of an antidepressant. Ours [
15] and other research [
16] has highlighted clinicians’ concerns about the applicability of evidence for those with severe presentations. The US primary care guidelines (Recommendation 3, pg 1319) [
1], practice parameters of the AACAP (Recommendation 9, pg 1511) [
2] and new Australian guidelines include a caveat allowing earlier prescription of medication in this case (Recommendation 5: pg 55) [
4]. However, the early prescription demonstrated in the current study was not entirely accounted for by the severity of depression.
It could be that medication was prescribed earlier for young people who declined to engage in psychological therapy. Our previous work has highlighted difficulties engaging young people with severe and complex presentations in psychotherapy. Again the US primary care guidelines (Recommendation 3, pg 1319) [
1] and practice parameters of the AACAP (Recommendation 9, pg 1511) [
2], the Australian guidelines (Recommendation 5, pg 55) [
4] as well as the NICE guidelines (Recommendation 7.10.1.3, pg 127) [
3] allow provision for medication in this case. This is somewhat paradoxical given the need for close monitoring in the case of medication prescription and the significant gaps between recommendations about monitoring and actual practice highlighted in this research.
Third, fluoxetine was the first medication prescribed for 92% for those 18 years and under, and 53.3% for those over 18 years. This is an encouraging result for those under 18 and suggests that unambiguously stated and potentially less complex recommendations are more likely to be carried out. The result for those over 18 was not unexpected given the audit was undertaken before the Australian guidelines [
4], which include ‘good practice points’ recommending extrapolation of evidence for those up to the age of 25, were released. These good practice points are based on the FDA data showing an increased risk of suicide related behaviours after prescription of an antidepressant for those up to the age of 25 [
18].
Finally, the Nice guidelines recommend “
careful monitoring of adverse drug reactions, as well as for reviewing mental state and general progress; for example, weekly contact with the child or young person and their parent(s) or carer(s) for the first 4 weeks of treatment” (Recommendation 7.10.1.3 pg 127) for those initiated on an antidepressant
. Our results show monitoring was undertaken far less frequently than recommended and was not done in a systematic nor precise way. This is consistent with previous US research [
19]. Our previous work regarding clinicians’ perceived lack of expertise and time to undertake systematic monitoring, particularly of medication induced adverse side effects [
15]. Together with research that has highlighted that a reliance on spontaneous report of side effects does not sufficiently identify all those at risk of suicide [
20], the current results provide further impetus for services to invest in systems that allow regular collection of standardized information about adverse effects, emergent suicidality and resolution of depression symptoms, for example, an easily accessible online tool that incorporates all elements requiring monitoring.
Strengths and weaknesses of the study
We did not formally assess inter-rater reliability; rather we relied on a high level of detailed instruction in the audit form and intensive training. The Principal author also undertook the majority of the audit.
Our results are based on medical files records rather than observation of what actually took place. This limited the set of behaviours that we could investigate; for example, it was not practical to assess the type of psychotherapy e.g. CBT or IPT being conducted as the assessment of this from case notes was felt to be highly inaccurate.
This is the first study to our knowledge assessing this range of key clinician behaviours in a public youth mental health service relative to guideline recommendations for the treatment of depression in young people. Some research has been undertaken to assess the gaps between guideline recommendations and clinical practice [
19,
21,
22], but each focus on only one or two recommendations. Our results are consistent with research on guideline implementation and the quality of depression care in adults. Adherence to guideline recommendations for adult depression is generally low in primary care e.g. [
23‐
25] even when interventions aiming to improve evidence based care are implemented e.g. [
26‐
28]. Of the few studies that have been undertaken in adult mental health specialist settings, adherence is better. For example in a national survey in the US, while appropriate care (defined as the respondent receiving medication or counseling consistent with guidelines) was only received by 19% of those who visited primary care providers, 90% of those visiting mental health specialists received appropriate care [
24]. One study looking at more specific indicators of guideline adherence by psychiatrists in private practice showed targeted training about providing guideline concordant care resulted in patients receiving more medication [
29]. Another study in psychiatric outpatient services showed adherence with all of a range of indicators of guideline adherence was seen in up to 55% of cases. Of note was the very low adherence to recommendations regarding routine outcome monitoring in the therapeutic phase [
30].