Methods
Families were randomly assigned to the intervention (five sessions PLH-YC parenting programme group) or the control group (one session lecture on parenting; https://rise-plh.eu/about-lecture-intervention-2/). The intervention trial was conducted during the COVID-19 pandemic and was shifted to mostly remote assessment and online intervention delivery via Zoom [26]. Families were recruited from December 2020 until February 2021.
Across the three countries, a total of 823 families were assessed and randomly allocated to the PLH-YC (n = 415) or lecture (n = 408) condition. Participating parents were mainly female (97% in both groups), about 35 years of age (PLH-YC group: M = 35.4, SD = 5.1; lecture group: M = 35.6, SD = 5.3). Child mean age was 5 years (PLH-YC group: M = 5.2, SD = 2.1; lecture group: M = 5.1, SD = 2.0). Baseline levels of the aggressive behaviour subscale of the Child Behavior Checklist (CBCL) [31] were approximately half a standard deviation above the mean (t-scores: PLH-YC group: M = 55.5, SD = 12.0; lecture group M = 55.1, SD = 11.0). Regarding clinical levels of child behaviour problems, 8.5% in the PLH-YC group met diagnostic criteria for ODD or conduct disorder (CD assessed with the MINI-KID-P) [32], compared to 7.4% in the lecture group. There were no baseline differences regarding any of the assessed outcomes and demographic data between the two groups (34).
AE assessment
Based on the experience in phase 2 and feedback from research staff at implementation sites, we further optimised the AE assessment procedures for phase 3. This included shortening the AE checklist (e.g. items were merged, for example back pain and headache were merged to one item assessing pain) and an adaptation of the adjudication form (e.g. specification of criteria for (S)AE; adding outcome). We also changed the assessment procedures to administer the AE checklist in individual assessments during the intervention (phone calls independent of the parenting programme sessions). This allowed for greater differentiation between the implementation and research components of the study, thus increasing the likelihood of reporting AE during intervention delivery. Also, this allowed the use of identical procedures in the intervention (five sessions) and control group (one session).
Following the recommendations by Linden (2013) [7], we considered all AE that occurred in parallel (with a time connection) with the intervention study using a standardised checklist, because all occurring negative events could potentially be caused by the intervention or study. In a second step, more details were gathered in an adjudication interview. This included an evaluation of whether or not the AE was related to the intervention or study and hence classified as an adverse reaction or not (Table 1).
AE checklist
All parents were asked to complete the self-report checklist five times: at pre-assessment, three times during the intervention, and at a 12-month follow-up. The checklist consisted of 12 items assessing physical problems (three items: accident, injury, pain), behavioural problems (three items: aggressive/violent behaviours, sleep disturbances, pain), emotional problems (one item: emotionally distressed), and other significant problems in daily life (five items: difficulties with personal relationships, unplanned hospitalisation, emergency room visit, death of a loved one, any other problems).
For each item, the parent reported whether this had happened to him/herself or the child (yes/no) since the last contact with the assessor or facilitator (usually 2 weeks; at pre-assessment we asked: ‘in the last four weeks’ before this baseline assessment). During the intervention and at the follow-up assessment (conducted about 6 months after the AE assessment following the fifth (final) session/4 weeks after the lecture) we asked ‘since the last interview’ so as to not miss any (S)AEs that happened during/after the completion of the programme and might be causally related to the study participation. If a parent reported a problem on the checklist, he/she then rated its severity (from 1 = mild to 4 = severe, answer-format for problems in daily life: yes/no; full checklist, ESM 2).
Adjudication interview
Based on the experiences of the previous phases of the RISE project where parents also reported mild events on the checklist that did not necessarily meet the AE criterion of a significant deterioration (noteworthy worsening of an existing problem or a significant new problem), we only followed up reports of behavioural, physical, and emotional problems with moderate or severe severity (3 or 4 on the checklist), or a significant problem in daily life. If this threshold was met, a semi-structured follow-up interview (ESM 2) was conducted to find out more about what had happened and to decide whether or not the definition of an (S)AE was met.
During the interview, parents provided a detailed description of the event and the research personnel recorded whether any actions were taken (e.g. offer additional 1:1 counselling), the outcome of this action (e.g. 1 = recovery to 5 = death, 6 = unknown), and to whom the AE happened (child/parent), whether the AE was (un-)expected and its severity (from 1 = mild to 4 = severe). Also, the causal relationship to the study was rated based on the detailed parent report (1 = not related to the study, 5 = definitely related to the study; for complete follow-up form, see ESM 2). If parents reported more than one problem (e.g. worsening of aggressive child behaviour, parents’ newly occurring emotional distress), the follow-up form was completed for each potential AE separately.
AE procedures
At pre- and follow-up assessment, the AE checklist was administered as part of the main outcome measure assessments. These assessment meetings were conducted in-person or over the phone/or video (depending on local COVID-19 restrictions) by local data assessors [26]. During the intervention, three phone calls were conducted to assess child and parent behaviour and the AE checklist was administered, after the first PLH-YC session (intervention group) or the lecture (control group); the third PLH-YC session (or for the control group 2 weeks after the lecture), and the last PLH-YC session (fifth session, or for the control group 4 weeks after the lecture). The data assessor entered the parents’ responses to the checklist into a tablet (via Open Data Kit software). The follow-up form was completed by the data assessor via paper–pencil.
Additional validation measures
Child mental health problems were assessed with the CBCL (total score; CBCL 1½–5 version: 100 items, CBCL 6–18 version: 113 items) [31]. Parents reported on a 3-point Likert scale; raw scores were transformed to age-adjusted standardised t-scores (based on multisocieties including Romania). Higher CBCL scores represented more mental health problems. Parent mental health problems were assessed using the DASS-21 [33]. Sum score ranges from 0 to 63 with higher scores indicating more depressive, anxiety, or stress symptoms. Social support was assessed with the emotional support subscale of the Medical Outcomes Study (MOS) Social Support Survey (eight items, mean score transformed to a 0 to 100 scale with higher scores indicating more perceived support) (Heinrichs N, Lachman, JM, Waller F, Müller M, Frantz I, Raleva M, et al: Multi-country randomised controlled trial of a systematically optimised, remotely-delivered five-session parenting intervention with parents of children ages 2–9 years with conduct problems in Eastern Europe, submitted). For full description of measures, see Taut (2021) and Heinrichs (2024) [26, 34].
Add Comment