Parental-Led Sexual Abuse Education Amongst At-Risk Parents: Associations with Parenting Practices, and Parent and Child Symptomology

ABSTRACT

Considering the emphasis on parent-led sexual abuse education (PLSAE) in child sexual abuse (CSA) prevention, and the imperative of prevention in families living with demonstrated risk factors, it is important to understand the extent to which this group delivers PLSAE to their children, whether this is associated with any barriers or facilitators, whether parents are engaging in other protective behaviors (such as monitoring and involvement) and the relationship between these variables and other risk factors such as parent and child symptomology. We surveyed 117 parents, with children ranging in age from 25–89 months (67% boys), attending a parenting program for assistance with a range of parenting difficulties and child behavior problems from 2020–22. A large majority of parents reported not giving their children comprehensive prevention messages, discussing body integrity and abduction dangers to a greater extent. PLSAE was significantly positively associated with child internalizing and externalizing symptoms; parent and child age; and discussion of body integrity and abduction. However, PLSAE was not associated with any other measured variables (protective parenting; CSA knowledge; parenting self-efficacy; general and own-child risk appraisal; parent burnout, stress, depression or anxiety; child diagnosis; parental education level; employment or marital status; or income). The current findings suggest that investing resources into increasing parental knowledge, risk perception and confidence may be misguided. Future endeavors should consider helping parents be protective in other ways, for example, through the creation of safe environments and reducing the risks of CSA.

KEYWORDS:

• Child sexual abuse
• prevention
• parenting
• protection behaviors education
• sexual victimization

Educating children about child sexual abuse (CSA) and self-protective behaviors is thought to be protective, equipping children with the knowledge and skills to recognize, resist and report sexual advances (American Academy of Pediatrics, 2018; Bravehearts, 2021). CSA education is usually delivered to children within the school system; parents¹ are generally not included in this process, even though parent-led sexual abuse education (PLSAE) is recommended by CSA prevention campaigns (Bravehearts, 2021; Darkness to Light, n.d.). Within these campaigns, parents are encouraged to give their children comprehensive information about specific sexual abuse behaviors (e.g., that someone may try to touch the child’s genitals), the identity of potential offenders (e.g., family members, adults within the child’s social circle, and adolescents) and how to respond to an abusive encounter (American Academy of Pediatrics, 2018; Child Abuse Royal Commission, n.d.). Despite these efforts, parents continue to report that they do not give their children comprehensive CSA messages (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018; Prikhidko & Kenny, 2021; Zhang et al., 2020).

The goal of this paper is to replicate and extend previous research documenting rates of PLSAE and to explore potential PLSAE facilitators and barriers (Rudolph, Walsh, Shanley, et al., 2022; Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018) by analyzing the results of a survey of parents attending a parenting program in an Australian city. As CSA risk has been linked to certain features of domestic environments (such as parenting stress, parent mental ill-health, non-nuclear family structure), parenting behaviors (such as lack of involvement, lack of supervision/monitoring, neglect, and maltreatment) and child factors (such as mental ill-health, low social skills, and delinquency; see Assink et al., 2019 for a review), it is important to understand the rates of PLSAE and protective practices of diverse groups of parents. In view of this, the current sample represents parents living with some of these risk factors; specifically, those who have been referred to an intensive, individualized parenting program due to a history, or increased risk, of maltreatment and abuse, parental distress, or suspicion of clinical level child disruptive behavior problems (e.g., aggression, school refusal, impulsivity).

Parent-led sexual abuse education

Over the last four decades, research has demonstrated that parents are reluctant to give their children specific CSA information, preferring to educate their children about topics, such as body integrity (i.e., body ownership or private parts) and abduction dangers (Deblinger et al., 2010; Rudolph & Zimmer-Gembeck, 2018; Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018; Walsh et al., 2012). Rates of parents who warn their children about inappropriate genital touch have generally ranged from 23% to 64% (see Rudolph, Zimmer-Gembeck, Shanley, & Hawkins, 2018). Recent research confirms these statistics, – for example, 61% of a sample of Chinese parents told their 3–6-year-old children to say “no” if someone wanted to see or touch their private parts, and about half (48%) had told their child to tell a trusted adult if sexual abuse happens (Zhang et al., 2020). A study of parents from the US (youngest children in the family had a mean age of 6), found that a little over half of parents told their child/ren to leave if someone tries to touch their private parts (56%) and what to do if someone tries to keep private parts touching a secret (58%; Prikhidko & Kenny, 2021). Despite a substantial body of research documenting PLSAE dating back to the 1980s, no study has concentrated specifically on rates of PLSAE in families struggling with multiple stressors or living with CSA risk factors.

Parents are generally in favor of CSA education (Rudolph, Walsh, Shanley, et al., 2022; Prikhidko & Kenny, 2021; Zhang et al., 2020), which makes the reasons for the low rates of PLSAE more mysterious. Various reasons have been postulated including parents’ unwillingness to, or discomfort with, discussing topics of a sexual nature or about CSA (Davis et al., 2013; Finkelhor, 1984) belief that their children are too young (Deblinger et al., 2010; Prikhidko & Kenny, 2021), knowledge deficiency (Tutty, 1993; Walsh et al., 2012; Wurtele, 2008), low confidence or self-efficacy (Burgess & Wurtele, 1998; Walsh et al., 2012; Wurtele, 2008), or the belief that their children are at low risk (Collins, 1996; Tutty, 1993). Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018) studied PLSAE in a sample of Australian and UK parents of 6–11-year-olds and found support for some of these with PLSAE rates related to discussion of other sensitive topics, parental appraisal of the risk of CSA to children in general, and parental experience of CSA (directly or indirectly). However, PLSAE rates were not found to be related to parental CSA knowledge; general parental self-efficacy; self-efficacy regarding CSA protection; appraisal of their child’s risk of experiencing CSA; child sex; or parental age, education, or income. Although not previously tested, it makes intuitive sense that struggling parents and parents of symptomatic children are less likely to engage in PLSAE. The present study assesses these associations for the first time.

Although parents’ ability to protect their children from CSA has mostly been equated with PLSAE, the effectiveness of education in keeping children safe from abuse is yet to be confirmed (Finkelhor et al., 1995, 2014; Rudolph, Walsh, & Zimmer-Gembeck, 2022; Pelcovitz et al., 1992). Only two previous studies have considered whether PLSAE is associated with less reported CSA experiences. Finkelhor et al. (1995) found that although children aged 10 to 16 who had received comprehensive PLSAE knew more CSA facts, used more self-protection strategies, and were more likely to disclose, they were not more able to prevent sexual assaults than their non-CSA-educated peers. However, Rudolph, Walsh, and Zimmer-Gembeck (2022) found that PLSAE was not associated with disclosure likelihood or CSA knowledge; and that it was protective parenting (i.e., parental involvement, care, monitoring, and supervision) – not PLSAE – that was associated with lower rates of CSA experiences when retrospectively reported by university students. The present study also seeks to determine the extent to which at-risk parents engage in protective parenting behaviors that may be important in protecting CSA, such as monitoring and involvement, and whether these parenting practices are related to PLSAE.

Parental protective behaviors and CSA risk factors

In view of these findings from Rudolph, Walsh, and Zimmer-Gembeck (2022), and the low rates of PLSAE reported in previous research (Prikhidko & Kenny, 2021; Zhang et al., 2020) even amongst parents who report high levels of education (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018), it is important to consider the part played by protective parenting behaviors in CSA prevention. Three qualitative studies have explored the behaviors parents engage in, aside from PLSAE, with the intention of reducing their children’s CSA risk. Strategies include supervision; monitoring; maintaining close, loving parent-child relationships; “being there;” involvement; vigilance; communication; fostering self-esteem; and, being alert to signs of CSA and making safe choices about contact/care by others (Babatsikos & Miles, 2015; Collins, 1996; Rudolph & Zimmer-Gembeck, 2018).

In previous research (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018), parents who reported more monitoring, involvement, and general communication also reported a higher frequency of, and level of comfort with, discussing CSA topics (i.e., PLSAE). These parents also reported higher self-efficacy regarding their ability to protect their children from CSA and believed their children to be less at risk of CSA. This study requires replication as the participating parents were generally highly educated and reported medium or higher income levels. Investigating whether these relationships are supported in other populations, such as those who are seeking assistance because of parenting difficulties and unstable family conditions, is needed to understand the generalizability of the results, and the implications this may have on future CSA prevention strategies.

The parenting behaviors deemed protective by parents in previous research (Babatsikos & Miles, 2015; Collins, 1996; Rudolph & Zimmer-Gembeck, 2018), and those found to be associated with CSA incidence (Rudolph, Walsh, & Zimmer-Gembeck, 2022), align with demonstrated family and parenting risk factors for CSA; such as a lack of supervision/monitoring (Finkelhor et al., 1997, 2007; Testa et al., 2011) and low care/involvement (Black et al., 2001; Fergusson et al., 1996; Laaksonen et al., 2011; Pérez-Fuentes et al., 2013). Other risk factors, which tend to be higher in families involved with departmental child protection services and those mandated (or strongly encouraged) to complete parenting programs, include high parenting stress; low sense of parenting competence; prior or concurrent forms of child abuse in the home environment; low family SES (socioeconomic status); parental unemployment and non-traditional family structure; child shyness or low levels of social skills; and child with a mental condition (see Assink et al., 2019 for a meta-analysis of CSA risk factors). These demonstrated risk factors highlight the need to explore their association with PLSAE.

The present study

Considering the emphasis on PLSAE for CSA prevention, and the importance of prevention in families living with demonstrated risk factors (especially those involved with statutory child protection agencies), it is important to understand the extent to which this group delivers PLSAE to their children, whether this is associated with any barriers or facilitators.

The present study replicates and extends the work of Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018), who found protective parenting practices to be associated with PLSAE in a sample of high SES Australian and UK families. To extend on these past novel findings, we revisited these associations in a sample of at-risk parents, many of whom reported a low income and all of whom reported a range of parenting and family difficulties. The aims of the present study were to: (a) describe the rates of PLSAE in at-risk parents; (b) examine whether at-risk parents who report using more protective parenting behaviors (monitoring/supervision and care/involvement) are more likely to engage in PLSAE and feel more comfortable doing so; and (c) explore whether PLSAE in at-risk parents is associated with measured facilitators and barriers. Facilitators included parental CSA knowledge; parenting efficacy; and CSA risk appraisal for children in general, and their child specifically. Barriers included symptoms of parents’ stress, depression and anxiety, and child internalizing and externalizing symptoms.

Method

Participants

Participants included 117 parents of young children: 98 mothers, nine fathers, five grandparents, one stepparent, two foster parents, and two kinship carers, aged 21 to 62 (M = 37.2, SD = 7.9), referred to a government-funded service delivering Parent-Child Interaction Therapy (PCIT). Four additional parents attempted the survey but discontinued before answering most or all questions, so they were not included in the analyses. Focal children included 78 boys and 39 girls between the ages of two years one month and seven years five months (M = 4.45, SD = 1.19). Regarding marital status, 43% of participating parents reported being married, 32% single/de facto/never married, 12% divorced/separated, and 14% other. Regarding parent location of birth, 68% of parents were born in Australia, 10% New Zealand, 5% the UK, and 16% in other countries. Five parents (4%) identified as Aboriginal, Torres Strait Islander, or South Sea Islander. Regarding highest level of education, 10% of parents completed junior high school, 14% senior high school, 29% a certificate or diploma qualification, 10% some university study, and 37% a university degree. Most participants reported being employed for wages or self-employed (56%), others reported domestic duties or not looking for work (25%), and the remainder reported they were looking for work (7%), unable to work (7%) or students (6%). Household incomes (AU$) were reported as under $25,000 (14%), $25,001 to $50,000 (25%), $50,001 to $100,000 (27%), $100,001 to $150,000 (18%), and over $150,000 (16%).

Parents referred to the program had, or were at risk of having, contact with statutory child protection services. Most families (67%) were referred to the program by government or nongovernment agencies after reporting significant parent distress and child disruptive problems interfering with home or school (55% health/pediatrician, 8% statutory child protection service, 4% community agency), and 33% of parents self-referred to the service due to significant parenting difficulties with children who had high levels of disruptive behaviors at home and/or school. The parents in this sample were living with some CSA risk factors, for example, poverty (almost 40% of participants reported earning half the average annual Australian wage), parental problems and difficulties, child neglect or maltreatment (or risk thereof), and child behavior problems (Assink et al., 2019).

Procedure

Approval for the study was granted by the Griffith University Human Research Ethics Committee (Reference number: 2013/80). The study was conducted in an urban area of Australia within a university-based, government-funded service delivering PCIT (Campbell et al., in press). PCIT is an intensive parent support program designed for parents (or other caregivers) and their children. After referral to the clinic, families were assigned a therapist for an intake interview. During this interview, parents were informed about the research and were asked for their consent to participate. Participation involved completing three surveys: Immediately (T1: intake), within one week of starting treatment if duration of time from T1 was greater than 5 weeks (T2: pre-treatment), and one week after treatment was completed (T3: post-treatment). Parents were informed they could decline to participate or withdraw their participation at any time, and this would not affect their receipt of the no-cost parenting program. No parents declined research participation. The general purpose of the study from which these data were drawn was to inform therapists about children’s behaviors, and to examine the effectiveness of PCIT for reducing children’s behavior problems and improving positive, and reducing negative, parenting practices. Data were collected on child behavior, parenting practices, parent-child relationship quality, parental well-being, and (for this study) PLSAE practices. The data analyzed in this study were collected at T1 between August 2020 to August 2022.

Measures

All measures were taken at the time of client intake, before parenting sessions had begun.

Demographics

The parent demographic variables measured were caregiving role (i.e., mother, father, grandparent), age (in years), birth country, marital status, employment status, highest level of education completed, total household income bracket, and indigeneity. Child demographic variables reported by parents were child’s gender and age.

Parent-led sexual abuse education, and discussions about abduction and body integrity

Parents were presented with a list of topics, used by Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018), related to PLSAE, body integrity, and strangers/abduction risk, and indicated whether they had spoken to their children about each topic and, if yes, the degree of comfort they felt with the discussion. Response options to each item were 1 (never), 2 (yes, but I felt uncomfortable with the discussion) or 3 (yes, and I felt at ease with the discussion). The extent to which caregivers had engaged in PLSAE was assessed with nine items. Specific abusive behaviors, protective behavior, and possible perpetrators were all included in the risk questions (e.g., “that an adult/older person might touch him/her on his/her genitals/private parts,” “what to do if an adult/older person tries to touch him/her on his/her genitals/private parts,” “that it might be someone known to the child that does this”). Two questions asked about discussion of body integrity: “your body belongs to you” and “private parts are not ok to be seen or touched by others.” Finally, three questions were asked about discussions of abduction risk: “never to go with anyone unless it has been arranged with a parent,” “someone may tempt, lure or grab you” and “what to do if someone attempts to tempt, lure or grab you.” Responses per subscale were averaged to create total scores where higher scores indicated greater willingness and comfort with discussion of the relevant topic (score range, mean and standard deviation are displayed in Table 2). Cronbach’s α was .84, .85, and .76 for items measuring PLSAE, body integrity, and abduction risk, respectively.

Table 1. Proportion of Parents Reporting PLSAE, and Discussion of Body Integrity and Abduction with Their Children (n = 117).

	Proportion of Participants (%)
Topic	Never	Yes, but uncomfortable	Yes, and comfortable
Sexual abuse	84	3	13
Adult/older person may try to touch them (the child) or get child to touch the adult on genitals/private parts	69	9	22
What to do if this happens	65	9	26
Adult/older person might show them ‘rude’ pictures	95	2	3
What to do if this happens	96	0	4
Adult/older person may talk to child in a sexual way or about sexual topics	94	1	5
What to do if this happens	96	0	4
Perpetrator may be known to the child	93	2	5
Perpetrator may be a family member	93	1	6
Child’s private parts are not ok to be seen or touched by others	15	5	80
Body ownership – child’s body belongs to child	17	4	79
Never go with anyone unless arranged by parent	15	5	80
That an adult/older person might try to tempt, grab or lure child away	53	3	44
What to do if this happens	52	8	40

Table 2. Means SDs of All Measures and Partial Correlations Between All Measures (Adjusting for Child Age; N = 117).

	1.	2.	3.	4.	5.	6.	7.	8.	9.	10.	11.	12.	13.	14.
1. PLSAE	–
2. Disc body integrity	.22*	–
3. Disc abduction	.21*	.48***	–
4. Protective parenting	.07	.21*	.17	–
5. Involvement Subscale	.03	.16	.15	.95***	–
6. Monitoring Subscale	.11	.21*	.12	.60***	.31***	–
7. Parenting efficacy	.17	.03	.14	.43***	.34***	.41***	–
8. CSA Knowledge	−.01	.05	−.10	−.04	−.01	−.11	−.22*	–
9. Risk CSA General	.16	.23*	.21*	.04	.08	−.09	−.01	−.01	–
10. Risk CSA Specific	.10	−.02	−.001	−.30**	−.24**	−.27**	−.26**	−.04	.14	–
11. Child Internalizing	.31***	−.05	.08	.03	.02	.05	.14	−.13	.11	−.05	–
12. Child Externalizing	.19*	.03	.21*	.04	.09	−.12	−.04	−.10	.19*	.02	.66***	–
13. Parent Burnout	−.10	−.07	−.08	−.34***	−.28**	−.32***	−.57***	.12	.04	−.02	.09	.26**	–
14. DASS	.08	.03	−.01	−.26**	−.23*	−.19*	−.37***	.06	.05	−.01	.15	.16	.72***	–
15. Child Diagnosis	−.01	.03	.10	.01	.05	−.14	.06	.02	.02	−.15	.15	.26**	.07	.06
Possible range of scores	1–3	1–3	1–3	2–10	1–5	1–5	1–4	2–16	1–5	1–5	0–2	0–2	1–5	0–3
M	1.24	2.64	2.09	8.71	4.39	4.31	2.63	9.66	4.03	3.09	0.46	0.96	1.96	0.64
SD	0.42	0.69	0.77	2.09	1.73	0.72	0.67	2.84	1.07	1.30	0.29	0.38	0.74	0.54

Note. *p < .05, **p < .01, ***p < .001; PLSAE = parent-led sexual abuse education, CSA = child sexual abuse. DASS = Depression, Anxiety and Stress Scale, Child diagnosis was coded 0 = no, 1 = yes.

Protective parenting: monitoring and involvement

Parental monitoring and involvement were measured with 21 items. Response options for all items ranged from 1 (never or rarely) to 5 (very frequently or always). Eleven items measured parental monitoring, which were adapted from the Parental Knowledge Scale (Stattin & Kerr, 2000; e.g. “In the past year, there have been times when I did not know where my child was or who he or she was with.”), the Parental Monitoring Instrument (Cottrell et al., 2007; e.g. “I check on the specifics of planned activities”), and the Situational Prevention of CSA scale (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018, e.g. “My child spends time at home without adult supervision”). Ten items measured involvement, where six were adapted from the Alabama Parenting Questionnaire involvement subscale (Shelton et al., 1996; e.g. “I play games with my child”) and four from the Situational Prevention of CSA scale (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018, e.g. “I find it hard to be interested in the things my child wants to do.”). Negatively worded items were reverse-scored and items on each subscale were averaged so higher scores indicated higher levels monitoring or involvement (range, M and SD are displayed in Table 2). A total protective parenting score was created by calculating the average of the total pool of items in the two subscales, Cronbach’s α = .67.

CSA knowledge

Parents’ knowledge of CSA was measured with eight items based on Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018) and Tutty (1993). Items included facts regarding victims, perpetrators, and the nature of abuse that are commonly misunderstood or underestimated. They were presented in a multi-choice format with response options that were uniquely relevant to each item. For example, one item was “If a child has been sexually abused there will be physical evidence,” where responses were (a) almost always, (b) about half the time, (c) rarely, and (d) never. Parents were asked to select the answer they believed was correct. Correct answers were coded as 2 while incorrect answers were coded as 0, and total knowledge scores were calculated via adding together the eight items, such that higher scores indicate higher knowledge (range, M and SD are displayed below).

Parenting efficacy

Four items based on the Parenting Sense of Competence Scale (Johnston & Mash, 1989) were completed by parents to measure general parenting efficacy: “I feel I am a good parent,” “I know what I am doing as a parent,” “I am a much better parent than most” and “I feel I have the skills needed to be a good parent.” Response options ranged from 1 (not true at all) to 4 (very true) and total score was calculated by averaging items, such that higher scores indicated a greater sense of efficacy (range, M and SD are displayed below). Cronbach’s α = .81.

Parental risk appraisal

Parents’ perceived level of risk of CSA for children in general, and their children specifically, were assessed with one item each (general: “I believe all children are at risk of sexual abuse,” specific: “My child is less likely than other children to be sexually abused”). These items have been used in previous research (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018). Response options ranged from 1 (strongly disagree) to 5 (strongly agree) and the second item was reverse scored, so a higher score indicated greater perceived risk in both areas (range, M and SD are displayed below).

Parenting symptomology – depression, anxiety, and stress

Parents’ levels of depression, anxiety, and stress symptomatology was measured using the 21-item Depression, Anxiety and Stress Scale (DASS−21; Lovibond & Lovibond, 1995). Each subscale was measured using seven items, where participants rated how strongly each item related to them over the past week (e.g., depression: “I felt that life was meaningless,” anxiety: “I felt I was close to panic,” stress: “I felt that I was using a lot of nervous energy”). Response options ranged from 0 (did not apply to me at all) to 3 (applied to me very much or most of the time). Total DASS score was calculated by averaging all items, such that higher scores indicated higher depression, anxiety, and stress (range, M and SD are displayed below). Cronbach’s for items on the depression, anxiety, and stress subscales were .92, .82, and .89 respectively; and .94 for all DASS items.

Parental burnout

Parental burnout was assessed using 20 items from the 23-item Parental Burnout Assessment (PBA; Roskam et al., 2018). Participants rated how often each item on the scale was relevant for them (e.g., “I feel so completely run down by my role as a parent”), where response options ranged from 1 (never or rarely) to 5 (very frequently or always). Total scores were calculated by averaging all items, such that higher scores indicate higher parental burnout (range, M and SD are displayed in Table 2), Cronbach’s α = .94.

Child internalizing and externalizing symptoms

Child internalizing and externalizing symptoms were assessed using the Child Behavior Checklist (CBCL: Achenbach & Ruffle, 2000). The CBCL for ages 1.5–5 years was used, as most children attending the clinic (90% in the current sample) are aged five and under. The full scale consists of 99 items, where parents rated how often each item was true for their child now or within the past 2 months. Twenty-five items assessed internalizing symptoms (e.g., “Avoids looking others in the eye”) and 26 assessed externalizing symptoms (e.g., “Destroys their own things”). Response options ranged from 0 (not true) to 2 (very or often true). Total internalizing and externalizing symptoms scores were calculated by averaging all items relevant to each subscale, such that higher scores indicate higher symptomatology (range, M and SD are displayed below). Cronbach’s α for items on the internalizing and externalizing subscales were .87 and .90, respectively.

Child diagnosis

Parents reported (with yes/no) whether the focal child had been diagnosed with any of the following: autism spectrum disorder (ASD), attention deficit disorder (ADD), attention deficit hyperactivity disorder (ADHD), fetal alcohol syndrome disorder (FASD), oppositional defiant disorder (ODD). A child diagnosis indicator variable was created to indicate no diagnosis (0) or at least one diagnosis (1).

Results

Overview of data analyses

All analyses were conducted using IBM SPSS (version 28). Descriptive analyses were conducted to summarize the percentages of parents who engaged in PLSAE and discussed body integrity and abduction risk. To test the associations of demographic measures with PLSAE and discussion with children about body integrity and abduction risk, correlations with child and parent age were calculated, and t-tests and analysis of variance (ANOVA) were used to compare demographic groups (parents’ partner status, employment status, education level, income level, and child gender). Given significant associations of child age with some measures (see below), correlations adjusted for child age were estimated (i.e., partial correlations) between all other primary measures to address the research aims.

Plsae

Table 1 displays a detailed breakdown of parents reports of PLSAE. Notably, a large majority of parents reported they had never spoken to their child about sexual abuse. Most parents reported that they did not discuss prevention topics and behaviors specific to CSA perpetration. For example, more than two-thirds of parents reported they had not told their children that an adult/older person may try to touch the child’s private parts, and almost all parents had not told their child that the perpetrator could be known to the child or a family member. In contrast, a high proportion of parents reported they had discussed abduction dangers, body integrity, and private parts.

Correlations between demographics and CSA discussion topics (PLSAE, body integrity, abduction)

Child age (in months) was positively correlated with PLSAE (r = .21, p = .02), communication about body integrity (r = .25, p = .006), and communication about abduction (r = .45, p < .001). Child gender (boy = 1, girl = 2) was not associated with PLSAE, meaning there was no difference in level of parent communication with girls compared to boys, t(115) = −1.76, p = .08.Parent age (in years) was significantly positively related to PLSAE (r = .20, p = .03). Group comparisons revealed that there were no significant differences in PLSAE or in discussions about body integrity or abduction based on groups defined by parents’ partner status, employment status, level of education, or household income.

Descriptive statistics and correlations between measures

The means and SDs for all primary measured variables are presented in Table 2. Protective parenting, parenting efficacy, and appraisal of general CSA risk were rated by parents to be moderately high to high. However, CSA knowledge and appraisal of own-child risk indicated parents’ knowledge was average and parents rated their own children to be at lower risk of CSA than children in general. According to cutoffs by Lovibond and Lovibond (1995), the mean scores of the present sample for depression, anxiety, and stress were all within normal range; although, all extreme scores fell in the extremely severe range.

Correlations, also shown in Table 2, indicate that PLSAE was not significantly associated with protective parenting (total, involvement only, or monitoring only) or any other measured variables except child internalizing and externalizing symptoms, and discussions of the related child protection topics (body integrity and abduction). Discussion about body integrity was significantly positively associated with protective parenting (a combination of monitoring and involvement, and monitoring alone), but discussion of abduction was not. Discussion of body integrity and abduction were both positively associated with parents’ perception of general risk of CSA. Discussion of abduction was also associated with child externalizing symptoms. However, PLSAE and discussion about body integrity or abduction were not significantly associated with parenting efficacy, parent knowledge, or parents’ perception of their own child’s risk of CSA. The other significant correlations were between parents’ perception of their own child’s CSA risk, measures of parenting practices and level of self-efficacy. Specifically, parents who indicated more risk of CSA for their child also reported fewer protective parenting practices and lower parenting efficacy.

Discussion

Parent-led child sexual abuse education (PLSAE) is encouraged by CSA prevention campaigns, with a view to helping children respond appropriately to risks for sexual abuse. Available knowledge suggests that although parents are aware of the risks and are very supportive of protective behaviors education both at school and at home, few parents report providing their children with accurate and detailed information about CSA. Supporting parents to engage in these discussions with their children requires a better understanding of PLSAE correlates and facilitators, in addition to a more nuanced understanding of the contribution of situational factors, such as high levels of parental stress and socio-economic pressures.

The current study extends on previous research by Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al (2018, 2022). exploring associations between PLSAE and discussion of related topics (body integrity and abduction), and parent and child factors, in at-risk parents of young children who completed a survey prior to attending a government-funded parenting program. Parent factors included protective parenting (monitoring and involvement), parenting symptomology (depression, anxiety, stress and burnout), parenting efficacy, parental CSA knowledge, and CSA risk appraisal (for children generally and for their own child specifically). Child factors included internalizing and externalizing symptomology, and diagnoses.

Results show few parents reported PLSAE and, on average, PLSAE scores were low. Instead, parents were more likely to discuss body integrity and abduction dangers with their children. PLSAE was not associated with any of the measured parent or child variables with the exception of child internalizing and externalizing symptomology, and discussion of related topics (body integrity and abduction).

Parents’ reports of PLSAE

Parents reported on the extent to which they had discussed certain CSA-related topics with their children. Less than one in six parents spoke to their children about “sexual abuse” (self-defined). Consistent with previous research, parents were much more likely to discuss private parts, body integrity, and abduction (Deblinger et al., 2010; Rudolph & Zimmer-Gembeck, 2018; Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018; Walsh et al., 2012). For children to enact taught protective behaviors, the information they receive must be clear and explicit (i.e., naming specific sexual behaviors, possible perpetrator identities, and unambiguous courses of action). The results presented here suggest that at-risk parents of 2–7-year-olds are reluctant to give their children this information. Less than one-third of parents told their child that an adult or older person may try to touch the child’s genitals, only 7% of parents told their child that the perpetrator may be a family member, or someone known to the child, and just over one-third told their child what to do if the child finds themselves in a sexually abusive scenario. Despite the vulnerability to CSA of this age group, they will not receive school-based CSA education until the ages of 5 to 7. This suggests that, either more resources should be devoted to assisting parents to educate their children (notwithstanding the results presented below), or the protective potential of parents should be harnessed in alternative ways.

Comparisons between the rates of PLSAE found by previous studies is complicated by study differences such as child age and definitions of sexual abuse. However, the rates of PLSAE found in the current study fall short of those reported previously, with most research finding that about half of parents educate their children about sexual abuse. For example, recent research from Tanzania (Mlekwa et al., 2016; Discussion item: “do not let someone touch your genitals;” child age: under 18), Australia (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018; Discussion item: “an adult/older person might try to touch you on your genitals;” child age: 6–11 years), China (Zhang et al., 2020; Discussion item: “that if someone wants to see or touch your private parts, you should say ‘No’ and leave at once;” child age: 3–6), and the USA (Prikhidko & Kenny, 2021; Discussion item: “leave a situation where someone is trying to touch your private parts;” child age: M = 6) reported rates of 46 to 61%. Although the focal children in the current sample were mostly boys, and some studies have found that parents discuss sexual abuse more with girls than boys (Prikhidko & Kenny, 2021), we did not find child gender to be associated with PLSAE in our sample.

Correlates, barriers, and facilitators oF PLSAE

Not much is known about why parents do not engage in PLSAE with their children; however, several reasons have been proposed. Evidence to support some of these postulations comes from the precursor to the current study (Rudolph et al., 2018), which found parental discussion about CSA to be associated with discussion of other sensitive topics, general CSA risk appraisal, and protective parenting in a sample of highly educated parents. Contrary to these findings, more protective parents did not report more PLSAE or general risk appraisal in the current study. Instead, PLSAE was only associated with other safety discussions with children (about body integrity and abduction) and was higher when parents reported their children had more internalizing and externalizing symptoms. It is not clear why children’s symptoms relate to at-risk parents greater PLSAE, but it is possible that parents caring for children with internalizing or externalizing symptoms may recognize that their child is at greater risk of CSA and attempt to counter this by giving their child CSA education.

Research on the association between parental CSA knowledge and PLSAE is limited and mixed, with Zhang et al. (2020) reporting conflicting results to Guo et al. (2019) and Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018), and parents themselves reporting the need for more information (Prikhidko & Kenny, 2021). The results of the current study support findings by Guo et al., and Rudolph et al., with parental CSA knowledge not associated with the likelihood of parents engaging in PLSAE.

Self-efficacy and risk appraisal were also not associated with PLSAE in this study. This finding is relevant to theories of health behavior change such as the Health Belief Model (Janz & Becker, 1984) and Protection Motivation Theory (PMT: Rogers, 1975) which propose risk appraisal and self-efficacy as prerequisites to positive behavior modification. As we found no significant association between PLSAE, risk appraisal, and self-efficacy, it is possible that parents may not believe that the response (PLSAE) is the most functional or appropriate. Or perhaps there were other reasons for parents not performing the prevention response, such as discomfort, and concern about the effectiveness of the approach or possible harms associated with it (Rudolph & Zimmer-Gembeck, 2018; Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018). To the best of our knowledge, these associations have only been researched previously by Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018), however, historical research found both response efficacy and self-efficacy promoted parental intentions to deliver CSA education (Burgess & Wurtele, 1998; Campis et al., 1989).

As research demonstrates a link between parental mental health and parenting and child outcomes (Russell et al., 2020; Seymour et al., 2015), it is no surprise that parental symptomology (depression, anxiety, and burn-out) was negatively associated with protective parenting and parenting efficacy in the current study. However, it is interesting that parental symptomology was not related to PLSAE. This suggests that the parents who choose to engage in PLSAE follow-through with this irrespective of their emotional state.

Taken together these results suggest that interventions designed to increase parental CSA knowledge, enhance parental awareness of CSA risk, and bolster parental self-efficacy may be misplaced, as increases in these potential discussion facilitators have not been shown to be associated with PLSAE in either highly educated or at-risk parents. Perhaps resources might be better directed to engaging parents in protection that does not require direct discussion with their children about the specifics of CSA.

Of interest is the associations found here between own child risk appraisal, protective parenting, and parenting efficacy, with parents who engaged in less protective parenting believing their child to be at greater risk and feeling less efficacious as a parent. This directly replicates findings with the previous high SES sample (Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018). Causality is unknown, however, it is possible that parents who are less protective are aware that this may place their child at higher risk of CSA, leading to negative evaluations of themselves as parents.

Study limitations, implications, and future directions

This paper presents findings which could move the CSA prevention field forward and improve CSA initiatives, however, several limitations of the current study can be noted. Firstly, utilizing self-reports of parenting and PLSAE is liable to socially desirable reporting, recall biases, and reporting errors. Supporting future research findings with observational data and reports from multiple sources would enhance the reliability of results.

Second, the internal consistency of the created protective parenting scale, measured with Cronbach’s alpha as 0.67, is uncertain. Although an acceptable Cronbach’s alpha is generally agreed to be in range of 0.70–0.90 (Tavakol & Dennick, 2011), some alternative cutoffs have been utilized (e.g., reasonable 0.67–0.87, adequate 0.64–0.85, moderate 0.61–0.65, satisfactory 0.58–0.97: Taber, 2018).

Third, women make up the majority of participants in parenting programs, and in this study 83% of the participants were mothers. This under-representation of fathers reflects most CSA research (see Rudolph et al., 2023), and future research should prioritize recruiting fathers and other male carers (as was done by Smasal, 2006).

And finally, we were unable to control for the possible confound of personal CSA experiences. Previous research has demonstrated a link between parental CSA history and PLSAE (Rudolph et a., 2018), with parents who experienced sexual abuse in childhood reporting more CSA risk discussion with their children; however, due to ethical constraints, we were unable to measure either parent or focal child CSA experiences.

Furthermore, we used the term “older person” to refer to perpetrators who were not adults, but were older than the child, in order to capture the age differential that exists in the majority of CSA cases. However, in hindsight, the phrase “peer or older child” should have been included in order to capture parents talking to their children about peer sexual abuse and sexual abuse perpetrated by “older children,” rather than the more ambiguous “older person.”

A growing body of research is demonstrating that parents may be ineffectual educators of their children on topics related to sexual abuse and protective behaviors (Deblinger et al., 2010; Prikhidko & Kenny, 2021; Walsh et al., 2012), with even the majority of knowledgeable and educated parents not engaging in comprehensive PLSAE (Rudolph, Walsh, Shanley, et al., 2022; Rudolph & Zimmer-Gembeck, 2018; Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al., 2018). The implications of this are significant, as the predominate way to involve parents in prevention has been as educators. The current findings support those from Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al. (2018) and suggest that increasing parental knowledge may be inadequate to effect change in rates of PLSAE. Attempting to help parents accurately assess levels of CSA risk or assisting them to feel more confident with this aspect of parenting may also be ineffectual in increasing PLSAE. Given this, it is necessary to develop new ways in involve parents in prevention. Consideration of research on CSA risk factors, and evidence from survivors and perpetrators on the process of abuse, suggests parenting is critical in CSA prevention. Rudolph and colleagues’ two pathways model (e.g., Rudolph, Zimmer-Gembeck, Shanley, & Hawkins, 2018), based on situational crime prevention and CSA risk factors, may be a suitable way forward. The model proposes that parents can be protective via the creation of safe environments and child well-being (through positive parenting).

Conclusion

This study is the first to explore PLSAE practices in a sample of at-risk parents. It extends past research on PLSAE and parenting by Rudolph, Zimmer-Gembeck, Shanley, Walsh, et al (2018, 2022) to examine the association between PLSAE and child and parent symptomology. The results presented here highlight the oft reported finding that parents are reluctant to engage in PLSAE. Although CSA prevention campaigns attempt to increase parental CSA knowledge, including parental perception of risk, the current finding suggest that these factors are not associated with increased PLSAE, and that investing resources into increasing parental knowledge, risk perception and confidence may be misguided, and future endeavors should consider helping parents be protective in other ways.

Ethical standards and informed consent

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation [Griffith University Human Research Ethics Committee (Reference number: 2013/80).] and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all patients for being included in the study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was partly supported by a Griffith University Postdoctoral Fellowship to Julia I. Rudolph. Support for the larger program from which these data were drawn was provided by the Queensland Department of Children, Youth Justice and Multicultural Affairs to Melanie J. Zimmer-Gembeck (the Family Interaction Program).

Notes on contributors

Julia I. Rudolph

Julia I. Rudolph PhD., MA, BPsych (Hons), is a post-doctoral research fellow at the Institute for Lifecourse Development, University of Greenwich, in London, UK.

Melanie J. Zimmer-Gembeck

Melanie J. Zimmer-Gembeck, PhD, is a Professor in the School of Applied Psychology, Deputy Director and Foundational Research Lead in the Griffith Centre for Mental Health, and the Director of the Family Interaction Program at Griffith University in Gold Coast, Australia.

Drew Straker

Drew Straker, BPsych (Hons), is a student in the School of Psychology, University of New England, Armidale, Australia.

Victoria Hambour

Victoria Hambour, BPsych (Hons), is a registered psychologist and PhD candidate at the School of Applied Psychology, Griffith University, Gold Coast, Queensland, Australia.

Tanya Hawes

Tanya Hawes, BPsych (Hons), is a psychologist and researcher at Griffith University, Gold Coast, Australia.

Kellie Swan

Kellie Swan, BPsychSci(hons), is a PhD candidate, a Senior Research Assistant and a Psychologist at Griffith University, Gold Coast, Australia.

Notes

¹ In this paper, the term “parent” is used to encompass all types of carers, including kinship and foster carers.