key: cord-0044900-h4yff3rs
authors: Stewart, Jonathan; Witte, Tricia H.
title: Secondary Trauma and Parenting Practices in Internet Crimes against Children Task Force Investigators
date: 2020-06-06
journal: Am J Crim Justice
DOI: 10.1007/s12103-020-09530-8
sha: d0c268f880afbb2f1f39d080cf4506afd641d242
doc_id: 44900
cord_uid: h4yff3rs

Investigating cases of child pornography requires daily exposure to sexually explicit material involving children and may have negative implications on the mental well-being of those in this line of work. This study aimed to identify whether secondary traumatic stress symptoms were associated with participants’ parenting behaviors and concerns about their own children’s use of the internet. Internet Crimes Against Children Task Force workers (n = 212) completed online questionnaires measuring work exposure to sexually explicit material, secondary traumatic stress symptoms, and parenting behaviors. Professionals in this field reported a wide range of secondary trauma symptoms, and their parenting behaviors were both directly and indirectly (via secondary trauma) affected by prolonged exposure to sexually explicit material involving children. Internet monitoring behaviors were more prevalent for parents of younger children, and mothers’ parenting behaviors were more strongly associated with secondary trauma symptoms than were fathers. Results have implications for mental health and parenting services for professionals in this field.

Investigators in this line of work are frequently exposed to sexually explicit material depicting abused children, and this indirect exposure to the child victim's trauma (via interviews, images, and audio/video files) has the potential to negatively impact their functioning, not only at their jobs but more broadlyat home when parenting their own children. The purpose of this study was to investigate symptoms of traumatic stress in a sample of professionals who investigate internet crimes against children, and to determine whether these symptoms were associated with the ways in which they parent their own children.

The term "secondary traumatic stress," coined by Charles Figley (1995) , encompasses the psychological and emotional toll that professionals in a caring or helping capacity experience due to the nature of caring for those who have experienced, or are currently experiencing, emotional distress. Secondary traumatic stress is defined as "the natural, consequent behaviors and emotions resulting from the knowledge about a traumatizing event experienced by a significant other" or "the stress resulting from helping or wanting to help a traumatized or suffering person" (Figley 1995, p. 10) . Essentially, secondary traumatic stress encapsulates the negative effects experienced by professionals due to indirect exposure to traumatic events (Figley 1995) . Secondary traumatic stress symptoms are similar to those seen in posttraumatic stress disorder (PTSD), which include avoidance of trauma reminders, compartmentalizing stress reactions, reexperiencing the trauma through intrusive thoughts, and increased arousal when reminders of trauma cases are present (Ludick and Figley 2017) . Secondary traumatic stress and PTSD are differentiated by the mechanism through which symptoms arise. With PTSD, symptoms manifest in response to directly witnessing or experiencing a traumatic event, whereas secondary trauma symptoms occur in response to indirect exposure, such as hearing accounts of another person's trauma.

Unfortunately, instances of secondary traumatic stress are not isolated to one specific profession. This construct has been investigated with numerous working populations, such as social workers and therapists (Bell 2003; Ben-Porat and Itzhaky 2009; Canfield 2005) . One study found that 37% of child protective service workers who interviewed child victims experienced clinical levels of secondary traumatic stress (Cornille and Meyers 1999) . Similarly, in another sample of child protective services workers, aggregate mean scores on a scale measuring levels of secondary traumatic stress were moderate, with 34% of participants meeting core clinical criteria for work-related PTSD (Bride et al. 2007 ). In the realm of healthcare, one study investigating the prevalence of secondary traumatic stress among emergency nurses demonstrated that 75% of the sample experienced at least one symptom of secondary traumatic stress within the week prior to the study (Morrison and Joy 2016) .

With respect to law enforcement officials, numerous hours have been devoted to studying the prevalence of work-related stress and burnout in individuals who work in a law enforcement capacity. In one study of federal law enforcement investigators, participants completed measures that gauged exposure to disturbing media (i.e. images, videos, and forensic interviews), burnout, and intentions of ending their career as a result of their work (Perez et al. 2010) . The participants' job-related requirements included searching computers and other technological devices for evidence of criminal activity and spent seven to eight hours every day searching for such files. Findings from this study indicated that employees were experiencing significant rates of burnout and were at greater risk for developing secondary traumatic stress symptoms (Perez et al. 2010) . Similar results were found in a study examining how work exposure to child pornography affects Internet Crimes Against Children (ICAC) Task Force agents and affiliates in a sample drawn from 511 agencies. Researchers found that 45% of respondents noticed problems arising from work exposure to child pornography, with approximately 40% believing that more mental health services were needed in their agencies (Wolak and Mitchell 2009) . Furthermore, in a study documenting the hazards of investigating internet crimes against children, it was found that digital evidence handlers who had more frequent exposure to materials involving sexual abuse of children reported increased levels of secondary trauma and more frequent initiation of coping behaviors such as distraction by focusing on unrelated tasks when at home (Burruss et al. 2017) .

Very little research has focused on the ways in which job-exposure to child abuse materials during investigations may carry over to negative parental interactions. However, there is a sizeable research base on the spillover effects of a range of work-related activities for law enforcement officials and their families, such as how shift scheduling can alter family functioning (i.e. day and night shift and marital stress) and the psychological effects of being in the line of duty, such as officer hypervigilance both on-and off-duty (Craun et al. 2015) . Although these studies may offer insight into how work in the field of child exploitation affects these professionals' families, few studies have explicitly explored the implications this kind of work may have between these professionals and their own children. Craun et al. (2015) analyzed the impact of investigating internet crimes against children on interpersonal relationships and found that higher levels of secondary traumatic stress increased respondents' reported discomfort levels in expressing intimacy with their own children. This same study (using a predominantly male sample) revealed that several respondents felt their interpretation of expressing intimacy with their children (i.e. hugging, tickling, etc.) changed after being exposed to sex offenders engaging in these same behaviors in order to groom their underage victims (Craun et al. 2015) . Specifically related to police work, research has indicated that female police officers are likely to engage in "danger-protection" parenting practices that arise from their own experiences as police officers in an attempt to prevent their own children from becoming victims or offenders of various crimes. Reported practices included setting strict, authoritarian rules, omniscient monitoring and supervision of their children (i.e. use of GPS tracking devices, keystroke loggers on computers etc.) and carefully monitoring self-disclosure about work to their children (Agocs et al. 2015) . These studies highlight the stressors law enforcement officials and their families experience as a result of the traumatic nature of this line of work; however, these studies do not provide detailed information about the association between secondary traumatic stress and parenting behaviors among investigators of child abuse or how the association may differ between mothers and fathers.

While there is limited information in the literature on the association between secondary traumatic stress and parenting practices among investigators of child abuse, there is a wealth of information on the connection between parenting practices and PTSD which may help bridge the gap. Parental exposure to traumatic experiences and the presence of parental PTSD have been linked to heightened levels of parental distress and decreased parenting satisfaction as well as problematic parenting behaviors, such as emotional distance towards children and a higher risk of child abuse, as well as increased incidence of PTSD and anxiety in children (Berz et al. 2008; Cross et al. 2018; Gewirtz et al. 2010; Ruscio et al. 2002) . Similarly, National Guard members who were deployed during the Iraq and Afghanistan conflicts who experienced greater PTSD symptoms were more likely to self-report poorer parenting behaviors such as greater likelihood of utilizing harsh disciplinary actions (Chesmore et al. 2018) . PTSD symptomatology in deployed and returning veterans has also been found to be associated with negative parenting outcomes. Low satisfaction in parenting, increased aggression toward children, low warmth and engagement, and emotional distance from children have all been found to be related to the emotional numbing symptom of PTSD; with poorer, disengaged parent-child relationships being associated with the avoidance symptom of PTSD (Davidson and Mellor 2001; Duranceau et al. 2015; Lauterbach et al. 2007; Samper et al. 2004; Sherman et al. 2016 ). Related to gender differences, a study examining the moderating role of PTSD on a parenting intervention for military families yielded results that indicated paternal PTSD significantly moderated the effects of parenting interventions, such that the interventions were less effective for fathers with PTSD (Chesmore et al. 2018) . Findings from this study suggest that fathers with clinical levels of PTSD displayed smaller improvements in effective parenting skills compared to fathers without PTSD, possibly because fathers with PTSD were avoiding managing child problems because of the emotional arousal associated with parenting practices (Chesmore et al. 2018) . The body of literature addressing the ways in which PTSD affects parenting behaviors of veterans offers insight into the mechanisms of PTSD symptomatology and how it is implicated in the manifestation of negative outcomes in the children of traumatized parents. Although the same mechanisms can be hypothesized to apply to law enforcement officials who investigate and protect child sexual abuse victims, the literature is slim on data providing such associations, especially concerning the mechanisms of secondary trauma.

Study aim Previous research has been devoted to investigating such topics as compassion fatigue, workplace burnout, and secondary traumatic stress across a variety of helping professions. Pertaining to law enforcement officials, previous research has explored the development of secondary traumatic stress and significant burnout as a result of their experiences in the workplace (e.g., Perez et al. 2010 ). There has also been research on PTSD and the resulting negative implications on parenting practices, particularly with fathers (Chesmore et al. 2018) , and parental monitoring behaviors, particularly with mothers (Agocs et al. 2015) . However, the present literature is slim on secondary traumatic stress among child abuse and child pornography investigators and whether the development of secondary traumatic stress carries over to parenting behaviors for mothers and fathers. Due to this gap in the present literature, the goal of the present study was to analyze whether working in a profession that involves investigating internet crimes against children influences how law enforcement officials parent their own children, through the mechanism of secondary traumatic stress. Specifically, the present study examined whether secondary traumatic stress in child abuse investigators was associated with parenting behaviors, such as warmth, involvement, internet monitoring, and being concerned about their children's use of the internet. In the present study, employees of both state and federal law enforcement agencies across the United States who serve various job roles related to legal cases involving child pornography completed measures that assessed secondary traumatic stress, general parenting behaviors such as warmth toward their children and involvement in their children's lives, and parental monitoring behaviors relating to their children's internet use. Based on the review of the literature, the present study hypothesized that secondary traumatic stress symptoms would be negatively associated with warmth and general involvement with their children, and this relationship would be stronger for fathers compared to mothers. The present study also hypothesized that secondary traumatic stress would be positively associated with parental internet monitoring behaviors and concern about information disclosure online, and this relationship would be stronger for mothers.

Participants Participants for the current study included law enforcement officials working agential or forensic examination capacities from across the United States and were affiliated with state and federal Internet Crimes Against Children (ICAC) Task Forces. The sample consisted of both sworn law enforcement agents and civilian employees involved in the collection, interaction and analysis of disturbing media related to child pornography and protecting child sexual abuse victims. Estimates from the State of Alabama's ALEA (Alabama Law Enforcement Agency) Office estimate that there are approximately 4000 to 5000 professionals across the country currently working in this field. Data were collected from 359 law enforcement officials across a period of seven months from initial survey dissemination. Of the 359 responses collected, 147 were removed from analysis as they did not meet inclusion criteria. The final sample size was 212.

Inclusion Criteria Participants in the current study were included regardless of how long they have held their current job position, and no age restrictions were placed on the current study's participants. Respondents were required to meet the following criteria: a) have at least one child between the ages of five (5) and eighteen (18), and b) have at least some involvement in the process of child pornography investigations.

Sociodemographic Data Data were collected from 359 participants. After sorting participant responses that did not meet the inclusion criteria, 147 data points were removed from analysis, yielding n = 212. Of the 147 participants removed from analysis, 103 participants were removed as they reported they did not have any children between the ages of five (5) and eighteen (18). Several participants (n = 44) were removed because they reported they were not in any way involved in the investigation process, and thus were not exposed to explicit material. Participants were majority Caucasian (88.7%; n = 188), and majority men (65.1%; n = 138). A majority of the sample had at least a bachelor's degree (66%; n = 140), with 83% (n = 176) of the sample having an annual household income above $75,000. Related to demographics of the children, child gender was almost evenly split between boys (48.6%; n = 103) and girls (49.1%; n = 104) (see Table 1 ). 

Demographics Participants were asked to report the age of their youngest and oldest child, as well as how many children they had between the ages of five (5) and eighteen (18). If participants had more than one child between the ages of five (5) and eighteen (18), they were asked to think of the child who had the most recent birthday in order to answer the questions that followed. Participants were then asked to report this child's age and gender. Participants were also asked to report their current job/work duties, including search warrants, searching electronic devices, interviewing child victims, and interviewing alleged perpetrators, as well as how frequently they were involved in these duties (1 = "Daily", 2 = "2-3 times a week", 3 = "Once a week", 4 = "Never"). At the end of the survey, participants were asked to report their own gender, ethnicity, educational background, and annual household income.

Internet Parenting Style Instrument Parenting behaviors relating to children's internet usage were assessed using the Internet Parenting Style Instrument (IPSI) (Álvarez et al. 2013 ). The Internet Parenting Style Instrument consists of 25 items and utilizes a 5point Likert-scale with the following anchor points: 1 = never; 5 = always. Items within the measure are broken down into five categories: supervision, stopping internet usage, internet usage rules, communication about internet use, and supporting the child with internet use. Sample supervision items include "I am around when my child surfs on the internet" and "I use special software to block certain internet sites for my child." Sample stopping internet usage items include "I stop my child when they visit a less suitable website" and "I stop my child when I see they are chatting online." Items about internet usage rules include "I limit the time my child is allowed on the internet" and "I limit the type of websites my child is allowed to visit." Items relating to communication include "I define internet rules together with my child" and "I talk with my child about the dangers related to the internet." Items gauging support include "I sit together with my child at the computer to surf on the internet" and "I show my child how to surf safely on the internet." Respondents were asked to rate each item on the 5-point scale with how true to them they believed each item to be. Internal consistency for each subscale of the Internet Parenting Style Instrument ranged from acceptable to excellent, with Cronbach's α coefficients from 0.65-0.91.

The Concern about Information Disclosure Scale is a scale that was formulated specifically for this study and was adapted from the Actual Information Disclosure subscale of the Restrictive Mediation Measure (Shin and Kang 2016) . The Actual Information Disclosure subscale was originally developed for children to self-report on their own tendency to disclose personal information on the internet. For the purposes of the present study, this subscale was modified to assess parents' perceptions and concerns about their children's disclosure of information on the internet. This revised scale includes three (3) items on a 5-point Likert-scale with the following anchor points: 1 = never, 5 = always. Items include: "I am concerned that my child may send personal information to someone they have not met face-to-face," "I am concerned that my child may send photos or videos of themselves to someone they have not met face-to face," "I am concerned that my child is sharing their passwords with friends," and "I am concerned that my child will be the victim of an internet predator" in order to tailor this scale to the population being surveyed. Participants were asked to rate each item on the 5-point scale based on their level of agreement with this item. Internal consistency for the Concern about Information Disclosure scale was good, with a Cronbach's α coefficient of 0.85.

Alabama Parenting Questionnaire (APQ) The Alabama Parenting Questionnaire assesses important aspects of parenting practices that have been shown to be related to disruptive behavior problems in children (Shelton et al. 1996) . The APQ consists of six separate scales. For the purposes of the present study, only the Involvement and Positive Parenting subscales were utilized. The Involvement portion of the APQ consists of 10 items on a 5-point Likert-scale with the following anchor points: 1 = never; 5 = always. Sample items from this scale include "I have friendly talks with my child" and "I talk to my child about their friends." Participants were asked to rate each item on the 5-point scaled based on how true to them they believed each item to be. Internal consistency for the Involvement portion of the APQ scale was good, with a Cronbach's α coefficient of 0.80. The Positive Parenting portion of the APQ consists of six (6) items on a 5-point Likert-scale with the following anchor points: 1 = never; 5 = always. This scale was used as an assessment of parental warmth towards their children. Sample items include "I let my child know when they are doing a good job with something" and "I praise my child if they behave well." Participants were asked to rate each item on the 5-point scale based on how true to them they believed each item to be. Internal consistency for the Positive Parenting portion of the APQ scale was good, with a Cronbach's α coefficient of 0.82.

Secondary Traumatic Stress Scale Trauma related to the participants' profession was assessed using the Secondary Traumatic Stress Scale (STSS) (Bride et al. 2004 ). The STSS measures facets of secondary trauma in relation to workplace stress and consists of 17 items on a 5-point Likert-scale with the following anchor points: 1 = never; 3 = occasionally; 5 = very often. Minor wording changes were made to STSS items in order to tailor each item to the present study's population (e.g. scale items refer to "cases" instead of "clients"). Sample items from the STSS include "It seemed as if I was reliving the trauma(s) experienced by the child victims" and "I thought about my work/ job (or certain cases) when I didn't intend to." Participants were asked to rate each item on the 5-point scale based on how often each item occurred to them in the seven (7) days prior to participating in the study. Internal consistency for the Secondary Traumatic Stress Scale was excellent, with a Cronbach's α coefficient of 0.93.

Institutional Review Board (IRB) approved of the study and all related amendments was gained at the authors' university prior to data collection. The local ICAC Task Force agency served as the initial point of contact, with the agency's commander disseminating an email link to access an anonymous survey through the federal and state ListSERV (an application devoted to distributing emails and pertinent workrelated information to ICAC task force agents and related agencies across the country through an electronic mailing list). The first page of the survey outlined the study's procedure and was used to gain implied consent. Data were collected for a period of seven months from the date the survey was first administered, and participants were not given any compensation for their participation in this study.

All data analyses were completed using IBM's SPSS Statistics Software, v.24. Means, standard deviations, and bivariate correlations were conducted with all study variables prior to testing hypotheses. Linear regression analyses were performed to test the study hypotheses.

Descriptive Statistics Participants' job requirements, which were grouped and analyzed as the manner in which participants were exposed to explicit material (see Table 2 ) included executing search warrants (93.4%; n = 198), searching electronic devices (85.4%; n = 181), interviewing child victims (62.7%; n = 133), and interviewing alleged perpetrators (81.1%; n = 172). As evidenced by the sample values for each type of exposure to explicit material, a majority of the study's participants were required to perform multiple duties throughout the process of these investigations as a part of their work routine. Frequency of exposure to explicit material was also assessed, and of the participants surveyed, a majority of the sample (52.4%; n = 111) were involved in the investigation process and exposed to explicit material on a daily basis (see Table 2 ). Average scores on the STSS were approximately 35 points (minimum possible score = 17; maximum possible score = 85) with a standard deviation of 12.4 points. Six (6) participants reported no secondary traumatic stress symptoms, and the remaining 206 participants reported STSS scores between 18 and 80. STSS scores were not correlated with work exposure or with any of the parenting measures. The two subscales in the APQ were significantly correlated with each other (r = 0.50). Subscales within the IPSI were all significantly correlated with each other, ranging from 0.56 to 0.68. Lastly, concern about information disclosure was significantly correlated with parents defining internet rules with their children (r = 0.19; see Table 3 ). Note: Parent Gender and Child Gender were coded as male = 1 and female = 0 p < .10 *p < .05 **p < .01

Parental Warmth and Involvement To test the first study hypothesis, that STSS would be negatively associated with warmth and involvement in the child's life, particularly for fathers, a linear regression analysis was conducted with STSS (centered), participant gender (1 = male, 0 = female), and the interaction between participant gender and STSS entered into the model along with the following covariates: child gender (1 = male, 0 = female), child age, and work exposure to explicit material. The overall regression model for parental warmth was marginally significant, R = .29, F (6, 116) = 2.92, p = < .10, and accounted for 8% of the variance in the dependent variable. However, inconsistent with the hypotheses, STSS was not significantly associated with parental warmth and there was no evidence of moderation with parent gender. Covariates of child age and frequency of workplace exposure were significantly associated with parental warmth such that participants were more likely to display warmth towards younger children and were also more likely to display warmth towards their children if they reported higher frequency of exposure to explicit material at work. See Table 4 for regression coefficients.

The overall regression model for parental involvement was significant, R = .37, F (6, 184) = 4.77, p = < .01, and accounted for nearly 14% of the variance in the dependent variable; however, inconsistent with the hypotheses, STSS was not significantly associated with parental involvement and there was no evidence of moderation. In terms of the covariates, there was a significant association between participant gender and parental involvement, such that mothers were more likely to report being involved in their children's lives compared to fathers. In addition, child age was significantly negatively associated with parental involvement such that parental involvement was higher for younger children. See Table 4 for regression coefficients.

Internet Monitoring Each regression model predicting the internet monitoring subscales was significant and accounted for 36 to 50% of the variance (see Table 5 ). STSS did not significantly predict any of the internet monitoring subscales, though, with the exception of one: stopping internet usage. Interestingly, participants were less likely to stop their children from using the internet if their secondary traumatic stress levels were elevated, which was opposite than expected. There were no significant associations between participant gender and stopping behaviors, but there was a significant interaction between participant gender and STSS in predicting stopping behaviors. See Table 5 for regression coefficients. To understand this moderation, a test of simple slopes was performed. For women, there was a significant negative association between trauma symptoms and stopping behaviors: (β = −.35, p = .006). For men, there was no significant association: (β = .11, p = .206).

In terms of the covariates, child age was negatively associated with internet supervision, stopping, internet rules, internet communication, and internet support. Child gender was significantly associated with internet stopping, such that parents were more likely to stop their daughters from using the internet. Work exposure significantly predicted internet supervision, internet stopping, internet rules, and internet communication, such that the more work exposure to explicit materials the more internet monitoring. See Table 5 for regression coefficients.

Concern about Information Disclosure The overall regression model for concern about information disclosure was significant, R = 0.27, F (6, 176) = 2.27, p = <.05, and accounted for 7% of variance. In support of study hypotheses, there was a significant association between STSS and concern about information disclosure, such that participants were more likely to report being concerned about their children's disclosure of personal information on the internet when their STSS levels were elevated. There was no significant association between participant gender and concern about information disclosure. However, there was a marginally significant interaction between STSS and participant gender. See Table 5 for regression coefficients. To understand this interaction, a test of simple slopes was performed. For women, there was a significant positive association between STSS and concern about information disclosure (β = .26, p = .006). For men, there was no significant association (β = −.03, p = .778). None of the covariates (child gender, child age, work exposure) were found to be significantly associated with concern about information disclosure.

Previous research has indicated that law enforcement officials who investigate cases of child sexual abuse may be at a higher risk of experiencing secondary traumatic stress as a result of the requirements of their job. However, only a few studies have documented this finding and even fewer have investigated the potential carryover effects of secondary traumatic stress on parenting behaviors in this population. Using a sample of 212 Internet Crimes Against Children Task Force investigators, this study tested the hypothesized association between secondary traumatic stress symptoms and the potential spillover effect regarding parenting behaviors (i.e. warmth, involvement, internet monitoring behaviors, and concern about online disclosure of information). It was expected that participants who reported higher levels of secondary traumatic stress symptoms would also report less warmth and involvement with their children. Based on the literature indicating that fathers in this particular field report more discomfort in expressing intimacy with their children, the present study also expected the association between secondary traumatic stress, warmth, and involvement to be moderated by gender, such that the relationship would be stronger for fathers compared to mothers. Contrary to expectations, secondary traumatic stress symptoms were not significantly related to parental warmth nor involvement. However, workplace exposure to explicit materials was positively associated with warmth. This finding hints to a possible protective mechanism that parents in this field may utilize in response to commonalities that typically land children in situations that make them prime targets for child predators. Similar to sex trafficking victims, runaways and those lured by false promises tend to be prime targets for human traffickers (National Center for Missing and Exploited Children 2018). As such, children who do not receive proper care or attention at home tend to search for it through other avenues; typically online. In light of this, these findings may be reflective of a protective parenting mechanism, in that parents in this field may display more warmth towards their children to ensure their children receive the care and attention they need at home rather than feeling the need to satisfy these needs from outside sources.

It was hypothesized that participants who reported higher levels of secondary traumatic stress would also report engaging in more frequent internet monitoring behaviors. Based on the study findings, one form of internet monitoring was significant related to secondary traumatic stress symptoms: stopping internet usage. However, contrary to predictions, the association was negative; the higher the traumatic stress symptoms, the less internet stopping behaviors. These results were perplexing at first, in that the expectations were for participants who reported higher levels of secondary traumatic stress to also report higher frequency of stopping their children from using the internet. Finding that participants who report higher secondary traumatic stress were less likely to stop their children from using the internet may be indicative of the avoidance mechanism of secondary traumatic stress (similar to that found in PTSD; Ludick and Figley 2017) playing an active role in the participants' thought processes when confronted with their children's internet use. As such, prior research has focused on anxiety sensitivity and cognitive avoidance in patients with PTSD, with common findings indicating that individuals with PTSD report an aversion to situations that may lead to physiological arousal, with examples of those being associated with anxiety or fear (Simpson et al. 2006) . Connecting this research to the findings of the present study, it would make sense for participants who report greater secondary traumatic stress levels to be less likely to intervene and stop their children from using the internet. Specifically, participants with elevated secondary trauma levels may actively avoid stopping their children from using the computer in order to completely avoid inserting themselves into a situation that may serve as a PTSD trigger since computers and other technological devices are the main focal points in the child pornography field.

The association between secondary traumatic stress and stopping behaviors was stronger for mothers, which also follows previous research findings on gender differences in PTSD symptom expression which indicate that females are more likely to have greater reactivity to emotional cues (Murphy et al. 2018) . As such, the indications of moderation may be reflective of this research in that mothers may purposely disengage from stopping their children from using the internet, thus not inserting themselves into a situation that would involve them confronting an emotionally charged trigger for their trauma symptoms. These interpretations are speculative, though; further research is needed to support the role of avoidance in parenting as it relates to secondary traumatic stress in this population.

Also in support of the hypothesis, secondary traumatic stress was significantly associated with concern about online information disclosure. Participants with elevated secondary traumatic stress levels reported greater concern about their children disclosing personal information online. There was also a marginally significant interaction between secondary traumatic stress and participant gender, indicating that the association between secondary traumatic stress and concern about information disclosure was stronger for mothers than fathers. Participants being concerned about their children's disclosure of information online may reflect an active role of the anxiety/arousal component of PTSD (Simpson et al. 2006) playing into their children's use of the internet. It is common within the realm of child pornography for predators to engage in online solicitation while posing as an adolescent in order to gather information from potential victims, with 97% of cases originating with a law enforcement entity involving online solicitation to an undercover agent (The National Center for Missing and Exploited Children 2016). Additionally, finding that women were more likely to report concern about their children's disclosure of information online follows the trend highlighted when discussing stopping internet usage. As the research states, it can be expected for mothers to be hypersensitive to fearful or anxiety provoking situations (Simpson et al. 2006) . This may give further support for the active role that the anxiety/ arousal component of PTSD plays in the thought processes of mothers within this field, though more research is needed to test these hypothesized links.

It is important to note that exposure to explicit materials at work was predictive of increases in a variety of internet monitoring practices, independent from secondary traumatic stress. Specifically, internet communication, internet supervision, and internet rules were significantly related to increased work exposure even in the absence of any relation to secondary traumatic stress. These three parenting constructs are not necessarily anxiety-driven and may be different from internet stopping (avoidance) or concern for online disclosure (anxiety), as discussed above. Thus, it seems that being in this profession, and not necessarily the traumatic stress that may result from work in this profession, is directly related to certain proactive internet monitoring behaviors. Through the requirements of their work, parents may learn how to protect their children and ensure their children are familiar with the possible dangers of using the internet while simultaneously preventing them from engaging in risky behaviors while using the internet.

Additional findings regarding internet monitoring revealed that parents monitor their younger children more than their older children, and they monitor their daughters more so than their sons. Neither of these findings come as a surprise as published statistics indicate that 83% of offenders arrested for child pornography possessed images of young children between the ages of 6 and 12, and 58% possessed images of children ages 5 and younger (The National Center for Missing and Exploited Children 2016). Additionally, data from The National Center for Missing & Exploited Children indicated that 62% of offenders arrested for child pornography possessed pictures that were comprised of mostly female victims, offering further support for these findings.

Somewhat surprisingly, results from the current study showed no bivariate correlation between secondary traumatic stress and the amount of exposure to distressing material in the workplace. However, this finding is also consistent with the literature, which points to certain factors that make individuals more or less susceptible to secondary traumatic stress and other mental health outcomes, such as lower professional self-efficacy (Finklestein et al. 2015) and personal trauma, work support, and social support (Hensel et al. 2015) , and not simply the amount of exposure to traumatic materials. In fact, increased exposure to distressing materials may even lead to desensitization, a process that has been known to lessen symptoms of traumatic stress in therapeutic environments (Foa et al. 2000) .

Consistent with previous research findings in related fields (Cornille and Meyers 1999; Menashe et al. 2014; Pistorius et al. 2008 ), professionals who investigate child pornography may be impacted in their personal lives and their parenting behaviors. The present study offers additional support to this body of research with the association between secondary traumatic stress and participants' (particularly mothers) decision to stop their children's use of the internet, as well as their concern about their children disclosing information online.

Regarding the themes that appeared in these findings, it is important to acknowledge the need for additional mental health services to be made available to professionals in this field; especially when considering the variability of secondary traumatic stress symptoms, internet stopping behaviors, and concern about information disclosure reported by the sample. This need becomes more apparent with previous research on this population yielding similar conclusions. When studying a sample of ICAC Task Force agents from 511 agencies, 45% of respondents reported problems arising from their exposure to child pornography, with 40% reporting a belief that more mental health services were needed in their agencies (Wolak and Mitchell 2009) .

Another important finding in this study was that work exposure alone, regardless of secondary traumatic stress symptoms, had significant direct effects on professionals' parenting behaviors, particularly for younger children. From repeated exposure to forensic materials at work pertaining to child sexual abuse, parents in this profession likely construct boundaries around their child's use of the internet by: supervising their children when they are using the computer, communicating with their children about the benefits and dangers of the internet, and establishing rules for safe internet practices. These supervisory behaviors are likely appropriate and indicative of proactive and responsible parenting. It is important to consider the need for parenting instruction and/ or parenting classes for parents who investigate child pornography, as these parents may be interested in this material. While the availability of such classes tailored to this population is unclear, research on general parenting classes indicates such learning opportunities can yield stronger parenting skills, greater knowledge on child development, and lowered parenting stress levels for those who complete them (Scudder et al. 2014) . For professionals who investigate child pornography, classes that teach them how to properly separate and healthily manage the family and work realms could be beneficial to the mental well-being of these professionals, especially when acknowledging the relationship between their own children/parenting to the child victims they work to protect.

Regarding the present study, a few limitations must be acknowledged. Firstly, due to participants working in federal and state law enforcement capacities, requirements for the study's continuation included anonymity of respondents due to security clearances and regulations. So, due to these restrictions, demographic questions were not able to include a location that would offer insight to geographical differences. Additionally, initial data collection was to be completed with agencies contained within United States provinces. In light of this, the ListSERV through which the surveys were distributed unknowingly contained email contacts of affiliate agencies outside of the United States, with surveys reaching entities in Norway, Brazil, Great Britain, and Canada. As such, the prevalence of child sexual abuse crimes vary by country, as well as the availability of governmental assistance in combatting these crimes. Similarly, parenting practices and beliefs vary widely by country due to societal and cultural influences as well as geographical location. Thus, due to the anonymity of respondent data, respondents' country of origin may have implications on the data collected and the study's results. Also, due to the nature of the survey's distribution, the calculation of survey response rates was not possible as the survey was distributed throughout both federal and state agencies both within and outside of the United States. Lastly, that the survey did not assess respondents' marriage or child custody status is another important limitation to note given that approximately 15% of law enforcement professionals are divorced (McCoy and Aamodt 2010). As such, parenting responsibilities and time spent with their child(ren) are ultimately divided among divorced individuals and thus supervision and other monitoring behaviors may be skewed.

In light of the previously mentioned limitations, this study brings additional findings to the slim body of literature assessing the relationship between investigating child sexual abuse and the parenting behaviors of the professionals who perform these investigations. As such, the present study offers several important contributions to the literature. The evidence from the present study suggests that professionals in this field report a wide range of secondary trauma symptoms, and their parenting behaviors are both directly and indirectly affected by prolonged exposure to sexually explicit material involving children. Additional research analyzing the parenting behaviors of professionals in this field is suggested, with a focus on parent-child dynamics to assess the effects of this type of work on children of those who work in this field.

Data Availability The data and study materials are available upon request.

Conflicts of Interest The authors report no conflicts of interest.

Police mothers at home: Police work and danger-protection parenting practices

Internet Parenting Style Instrument

Strengths and secondary trauma in family violence work

Implications of treating family violence for the therapist: Secondary traumatization, vicarious traumatization, and growth

Associations between PTSD symptoms and parenting satisfaction in a female veteran sample

Correlates of secondary traumatic stress in child protective services workers

Development and validation of the Secondary Traumatic Stress Scale

The hazards of investigating internet crimes against children: Digital evidence handlers' experiences with vicarious trauma and coping behaviors

Secondary traumatization, burnout, and vicarious traumatization: A review of the literature as it relates to therapists who treat trauma. Smith College Studies in Social Work

PTSD as a moderator of a parenting intervention for military families

Secondary traumatic stress among child protective service workers: Prevalence, severity and predictive factors

The impact of internet crimes against children work on relationships with families and friends: An exploratory study

Trauma exposure, PTSD, and parenting in a community sample of low-income, predominantly African American mothers and children

The adjustment of children of Australian Vietnam veterans: Is there evidence for the transgenerational transmission of the effects of war-related trauma?

The home front: Operational stress injuries and veteran perceptions of their children's functioning

Compassion fatigue: Coping with secondary traumatic stress disorder in those who treat the traumatized

Compassion fatigue: Psychotherapists' chronic lack of self-care

Posttraumatic stress disorder and vicarious trauma in mental health professionals

Effective treatments for PTSD: Practice guidelines from the International Society for Traumatic Stress Studies

Posttraumatic stress symptoms among National Guard soldiers deployed to Iraq: Associations with parenting behaviors and couple adjustment

Meta-analysis of risk factors for secondary traumatic stress in therapeutic work with trauma victims

Secondary traumatic stress in police officers investigating childhood sexual abuse

Quality of parental relationships among persons with a lifetime history of posttraumatic stress disorder

Toward a mechanism for secondary trauma induction and reduction: Reimagining a theory of secondary traumatic stress

A comparison of law enforcement divorce rates with those of other occupations

Between the maternal and the professional: The impact of being a child welfare officer on motherhood

Secondary traumatic stress in the emergency department

Sex differences in PTSD symptoms: A differential item functioning approach

The National Center for Missing & Exploited Children 2016 Annual Report

Secondary traumatic stress and burnout among law enforcement investigators exposed to disturbing media images

Working with sexually abused children

Male war-zone veterans' perceived relationships with their children: The importance of emotional numbing

Posttraumatic stress disorder symptoms and parenting satisfaction among a national sample of male Vietnam veterans

Evaluation of an existing parenting class within a women's state correctional facility and a parenting class modeled from parent-child interaction therapy

Assessment of parenting practices in families of elementary school-age children

Veterans' perceptions of the impact of PTSD on their parenting and children

Restrictive Mediation Measure

Fear and avoidance of internal experiences among patients with substance use disorders and PTSD: The centrality of anxiety sensitivity

Work exposure to child pornography in ICAC task forces and Affliates