We read with great interest the article published by Mulraney et al^{Reference Mulraney, Hiscock, Sciberras, Coghill and Sawyer1} in the July 2020 issue. This was an extensive study and targeted a very vulnerable group of our society. The study was followed up longitudinally and pointed to the deficits in utilisation of mental health services. Although this study provided great insights, we had some reservations that we would like to discuss.

The authors have clearly described the take up of the data from the Longitudinal Study of Australian Children cohort. Clearly defined inclusion and exclusion criteria with a description of the recruitment process and how contact was made or how the questionnaire was applied, would have made the study design easier to follow. The recruitment process was difficult to infer especially for non-English-speaking readers.

The authors mentioned that children excluded from the analysis were likely to have less educated parents, more mental health problems and disturbed family dynamics but we think that these are one of the most vulnerable groups, who need a higher level of care.^{Reference Lyon, Ludwig, Stoep, Gudmundsen and McCauley2} The risk of underrepresentation of these subpopulations probably skewed the data, as those with higher unmet needs were left out of the study. Also, lack of mention of statistically significant values for the Vuong-Lo-Mendell-Rubin likelihood ratio test made it a little hard to interpret the results.

The authors have very rightly projected four trajectories – low symptoms, high decreasing, moderate increasing and high increasing – but while following the tables in the supplementary data with the trajectories and individual specialist contact or in different waves, it was difficult to follow through. The authors have reported an attrition of 3764 participants, which was not reflected in these tables. Also, the reason for drop-out of participants from 4348 at end of the wave 1 to 3764 at the end of wave 6 would have made it more informative. Other factors like dual or referred consultation, changes in outcome, ease of accessibility and presence of required services in the defined postcode could play a role in the proposed conclusions.^{Reference Reid, Stewart, Barwick, Carter, Leschied and Neufeld3} Probably the ratio of children who benefited with or without an adequate number of visits could have provided insights into the study.

What actually is of real concern is that psychiatrists and psychologists are the least consulted ones for mental health issues. A possible role of stigmatised views, of parents and society, towards mental health cannot be overlooked.

There are minor typographical errors that we noticed that were misleading like ‘Supplementary Fig. 1 shows’ in the results, while no such figure has been provided in the supplementary material or in Figure 1 of the article; age has been defined from 4 to 14 years, but in the graph that follows, age ‘10’ has been followed by ‘15’ and then by ‘14’.

It would be a great help if the authors could clarify these points. Thank you.