Dataset on the effects of psychological care on depression and suicide ideation in underrepresented children

These underrepresented children and adolescents aged 6 to 18 years were initially enrolled by retrieving citywide digital unique identifiers (UIDs) from the resident social care system and school registration system, underscoring sampling gender equality as well (n = 313,201, 48.8% girls). To pursue equal access for all underrepresented individuals who are currently in underprivileged conditions, we screened these targeted cohorts outside family, school and social care systems by being aided from local civil-affair authorities. Children/adolescents were finally gathered into the dataset only if the informed consent for agreements on this study had been digitally signed by their parent(s) or legal caregiver(s). All the actions to carry out primary psychological healthcare for these underrepresented cohorts were additionally overseen by local civil-affair authorities. This study has been officially approved by the IRB of Nanchong Psychosomatic Hospital (No. NCPP 2022002).

Underprivileged conditions herein covered almost all the circumstances requiring legally mandated social care in China and were finally checked by local social-care authorities. The dfUCA has been legally defined for child/adolescent whose parent(s) meet the criteria of severe disability, severe illness, serving a prison sentence, mandatory isolation for drug rehabilitation, subject to other measures restricting personal freedom, missing, revoked guardianship qualifications, or being deported (expelled) from the country. The orphans were marked for receiving social care if they lost parents or were unable to locate to specifically biological parents. The CECD referred to an extremely disadvantageous living environment that children/adolescents were experiencing in the family, containing a main member suffering from old age, physical weakness, living alone without care, loss of working capacity, or low economic income that was less than the local minimum criteria that the National Bureau of Statistics of China published per year ($ 8 206 at 2022 in Sichuan). The “left-behind” children/adolescents were identified if they were left in the home without care from either parent. Finally, the “single-parent” ones would be included in the social care system if they were solely parented by a (biological) mother or father. In total, 249,772 children/adolescents with underprivileged conditions that were legally defined above were finally collected into this dataset. They are generally underrepresented from current healthcare system and even existing scientific studies.


The project was conducted by a total of 569 centers for carrying out primary psychological healthcare, with each school or community-based institute (e.g., orphanage, community-care site and children’s hospital) for building a data center. This primary psychological healthcare used a “2 + 2” workflow, with two rounds of psychological screening and two rounds of early psychological care (see Fig. 2b). All the included children/adolescents were initially screened for depressive symptoms by the Center for Epidemiological Studies-Depression Scale (CES-D) (i.e., 1st round to screening, see details underneath). Those who captured risks of depressive symptoms appearing by the last round were tested by both Zung’s self-report depression scale (SDS) and the Chinese version of the Positive and Negative Suicide Ideation Inventory (PANSI-C) (i.e., 2nd round to screening depression and suicide ideation, see details below). To reduce the risks of errors in manual statistics, all these measurements were performed by a purpose-built app that was feasible for multifarious cellphones from mainstreaming manufacturers (e.g., Android, iOS). For further psychological health-care rounds, children/adolescents who were identified with the presence of depression or suicide ideation underwent the first round of psychological care conducted by qualified psychological health-care specialists. A portion of these children/adolescents would be further transferred to a government-sponsored mental health center (hospital) to receive clinical medical treatments if this decision was made by their specialists, which was the second round of early psychological care. As primary health-care services, this “2 + 2” workflow used an open-loop framework to press these children/adolescents with high risks of depression and suicide for reaching full medical cares. Thus, we no long provided clinical care, or followed these children/adolescents if they needed inpatient or additional medical treatments.


As the initial screening, we used the CES-D to detect whether children/adolescents exhibited depressive symptoms. This scale consisted of 4-point-Likert-scoring 20 items and is a widely adopted tool for screening depressive symptoms in nonclinical populations, which requires children/adolescents to report the frequency of appearing these depressive experiences within one week, with 0 points for rating “rare days (<1 day)” and 3 points for rating “almost all the week (5–7 days)”26,27. Child/adolescent would be marked with the levels of risk of appearing depressive symptoms if her/his total score of all 20 items exceeded 16 points28. Details for all the measurements can be found in the Supplementary Table 1.

Both Zung’s SDS and PANSI-C were utilized for (pre) clinically identifying the presences of depression and suicide ideation in the cohorts that identified depressive symptoms from initial screening (i.e., CES-D scores were over 16 points), respectively. The SDS was widely validated for high reliability and validity in screening children/adolescents’ (pre)clinical depression, which consisted of 20 items scored on a 4-point Likert scale as well, with higher total scores for more severe depressive levels29,30. To estimate the risks of coexisting suicide ideation in these depressive children/adolescents, the two-dimensional 14-item PANSI-C was deployed, with one facet for testing the positive suicide ideation (6 items, such as “I feel hopeful for my future life”, “I am very confident for my future”) and with another one for testing the negative suicide ideation (8 items, such as “I am a loser and want to suicide”, “I want to suicide when I cannot address troubles myself”)31. This tool was demonstrated for high two-factor structural validity and high reliability in screening adolescents’ suicide and even ensuing suicide behavior32,33. It should be noted that the PANSI-C was adopted only for children/adolescents aged more than 12 years old given the local ethical restrictions. Despite such protective policy, this restriction could be exempted if the screening for suicide ideation to children outside this age limitation had been proactively required by their parents, with a legally signed disclaimer.

Sociodemographic data

The sociodemographic characteristics of each child/adolescent included the most fundamental facets, including gender (i.e., girl/boy), age and living area (i.e., urban/rural). This information was collected by digital records from resident/school registration systems and was confirmed by the children/adolescents themselves.

Socioeconomic status (SES)

This section measured their family socioeconomic status and even subjective SES by 5 items, such as parents’ educational degrees, occupations and total incomes. Categories of degrees and occupations were coded by benchmarks formulated from the National Bureau of Statistics of China. Full details can be found in the Supplementary Table. 1.

Family parenting data

We collected data regarding their parenting circumstances in the family, except for those with social parenting (e.g., orphan). This domain included 7 items (if applicable), such as the number of offspring, the experiences of separation from parents and subjective satisfaction for family (see Supplementary Table 1).

Health-related data

This section collected essential health-related data as legally required by the “Healthy China 2030 Plan” policy. Here, we asked children/adolescents to report their health status (i.e., chronic disease), long-term drug-dependent treatments, sleep and cellphone abuse (see Supplementary Table 1).

Life events data

This dataset contained 4 items to capture the effects of four major life events on themselves by a 5-point self-rating scale, including “unjustly accused”, “dispute with close friends”, “major traumatic diseases in the family (including yourself)” and “death in the family”. Notably, the life event data were collected from these children/adolescents aged 8–12 years only. Full information can be found in Supplementary Table 1.


Leave a Reply

Your email address will not be published. Required fields are marked *