Indirect Victims of Sexual Offence Investigations: Exploring the Impact of “The Knock” on Partners’ Mental Health and Wellbeing

Abstract

“The knock” refers to the moment when the police first arrive at a house to investigate suspected sexual offending by a member of that family. This paper examines the levels of trauma, wellbeing and crisis support for partners, both at the time of the knock and at the current time. Forty-eight participants who had experienced the knock completed an online survey. All respondents were female; they had experienced the knock 0–20 years previously (M = 3.88 years). Participants provided demographics and completed the World Health Organisation (WHO) Wellbeing Index, the International Trauma Questionnaire (ITQ) and the Crisis Support Scale (CSS) for (i) retrospective (at the time of the knock) and (ii) current levels of social support. Participants reported lower levels of wellbeing and higher levels of trauma in comparison with general population norms. Approximately 40% of participants’ scores on the ITQ exceeded the criterion for post-traumatic stress disorder (PTSD), with a further 25% of participants meeting the criterion for Complex PTSD. Levels of trauma were negatively correlated with wellbeing and with both retrospective and current self-reported crisis support. Qualitative analysis of open-ended questions explored participants’ experiences of the knock, their perceptions of police conduct, and the personal, relational, and practical consequences that followed. The findings highlight substantial and enduring harm among partners and are discussed in relation to implications for current policy and practice.

1. Introduction

“The knock” refers to the moment police officers unexpectedly arrive at an individual’s home to conduct an arrest or investigation. For family members, especially partners of individuals under investigation for sexual offences, this event often heralds a catastrophic disruption to their lives, triggering emotional, social, and financial upheaval. The sudden realisation of a partner’s suspected criminal conduct shatters the family’s sense of normality, instigating what Armitage et al. (2023) describe as an “ontological assault,” a profound questioning of identity and previously held beliefs about their partner and themselves.

The psychological consequences of this traumatic event are severe and enduring. Partners commonly experience symptoms consistent with post-traumatic stress disorder (PTSD), including hypervigilance, intrusive memories, and avoidance behaviours (Brogden and Harkin 2000; Kavanagh et al. 2023a; Salter et al. 2024). Despite these substantial psychological impacts, clinical assessments of PTSD among this population remain limited, highlighting a critical gap in empirical research and underscoring the urgent need for targeted therapeutic interventions.

Moreover, partners of individuals accused or convicted of sexual offences face significant societal stigma through the influence of “secondary stigma,” whereby negative societal perceptions and blame extend to family members, resulting in social isolation, judgment, and discrimination (Evans et al. 2023; Kavanagh et al. 2024; Levenson and Tewksbury 2009; Wakeham et al. 2024). This stigma compounds partners’ distress, fostering a profound sense of shame and discouraging them from seeking necessary support (Kavanagh et al. 2024), and this is often accompanied by practical consequences such as employment loss, housing difficulties, and disrupted social networks (Cassidy and Ronis 2025; Kavanagh et al. 2023a; Salter et al. 2024).

These collateral consequences further exacerbate partners’ emotional and financial instability, intensifying their vulnerability. Consequently, partners often adopt various stigma management strategies, including withdrawal from social networks, relocation to new geographical areas, and selective disclosure of information to control societal reactions and minimise further trauma (Kavanagh et al. 2024). However, the systems intended to provide support frequently fall short. Initial interactions with the criminal justice system typically occur during arrest, a process partners often describe as disorganised and dependent on the attitudes of individual officers rather than structured institutional protocols (Armitage et al. 2024; Condry 2013; Kavanagh et al. 2023a, 2023b). The lack of clear guidance from agencies on navigating criminal justice and child protection processes further compounds feelings of abandonment and confusion (Hernandez et al. 2009; Loureiro 2016). Armitage et al. (2023) argue that treating these harms as inevitable collateral damage promotes institutional passivity and fails to adequately support affected families.

Historically, research has predominantly focused on primary victims—those directly harmed by offences—rather than secondary victims such as the partners and families of the accused (Morgan and Zedner 1992). Early research into child sexual abuse similarly neglected the experiences of non-offending family members, particularly partners, resulting in significant gaps in policy, therapeutic intervention, and support structures (Tamraz 1997; Hernandez et al. 2009). Despite increasing recognition of their trauma, comprehensive research and structured support for these secondary victims remain inadequate.

The disruption caused by the knock severely undermines partners’ “assumptive world,” their foundational beliefs about themselves, their relationships, and their social environment (Duncan et al. 2022; Janoff-Bulman 1992; Wakeham et al. 2024). This event triggers cascading losses, including relational support, social roles, financial stability, and community belonging, as well as significant emotional repercussions such as guilt, shame, and a fear of judgment (Cahalane and Duff 2018; Bailey 2018). Partners frequently grapple with profound feelings of betrayal and mistrust, severely impacting their mental health and ability to function in their familial roles (Zimmer-Gembeck et al. 2022).

Despite these challenges, existing academic discourse often narrowly frames partners’ roles as primarily supportive, emphasising their potential contribution to rehabilitation and child safeguarding (Armitage et al. 2024; Cahalane et al. 2013). This approach, however, risks instrumentalising partners, neglecting their own trauma and support needs. Kamitz (2025) critiques this tendency, advocating a rights-based perspective that acknowledges partners’ intrinsic needs, independent of their utility for rehabilitation.

Recognising this gap, peer-to-peer support groups have emerged as a critical resource, providing emotional relief and reducing feelings of isolation (Sample et al. 2018; Kavanagh et al. 2024). These groups facilitate shared experiences, validation, and mutual understanding, significantly improving emotional resilience. Nonetheless, these supports remain inadequately funded and insufficiently publicised, limiting accessibility for many families in need (Loureiro 2016).

Given the complex psychological, social, and systemic impacts associated with the knock, there is a pressing need for a more comprehensive, trauma-informed approach to policy and practice. Enhanced recognition and targeted interventions should prioritise partners’ emotional and psychological wellbeing, acknowledging their experiences not merely as collateral damage but as central concerns requiring dedicated attention and support.

The Current Study

This paper explores the impact of the “Knock on the Door” event experienced by partners, ex-partners and their families. The research hypotheses were:

• Participants would have significantly lower levels of wellbeing than the general population and significantly higher levels of trauma.
• Participants’ levels of trauma would be negatively correlated with levels of wellbeing.
• Participants with higher levels of crisis support would report lower levels of trauma.
• Participants with dependent children would report higher levels of trauma than those without children.

The qualitative component of the study was designed to capture participants’ lived experiences of the knock and its aftermath, providing depth and context to the quantitative findings. Specifically, the open-text questions explored how participants experienced the police visit and subsequent investigation, how these events affected their emotional wellbeing, relationships, and daily lives, and what forms of support they found helpful or lacking. The aim was to identify recurring experiences and concerns across participants, offering a richer understanding of the psychological, social, and practical consequences of the knock.

2. Materials and Methods

A favourable ethical opinion was obtained from the Business, Law and Social Sciences Research Ethics Committee of Nottingham Trent University (2020/176, July 2020) and the Lucy Faithfull Foundation.

2.1. Sample

Participants were recruited to complete an anonymous online survey advertised on the Lucy Faithfull Foundation’s Stop It Now! forum for friends and family of people investigated for online child sexual abuse offences. Additional recruitment took place via social media platforms (Twitter and LinkedIn) and through snowball sampling.

Participants (N = 48) were partners or ex-partners of people investigated, arrested or convicted of sexual offences. At the time of data collection, three partners had not received any convictions, eleven were under investigation, two had been cautioned, seven were awaiting a court date, five received a community sentence, seven received a suspended sentence, six received a custodial sentence, and seven had been released following a prison term.

2.2. Measures

Participants provided basic demographic information to contextualise their responses (gender, age, education level, ethnicity and religion). They also shared details about their partner’s offence, their current relationship status with the partner, the presence of children, and both the participant’s and their partner’s present circumstances. Specific inquiries about the knock included its date, time, and individuals present in the house at that time, the behaviour and attitude of the police and social workers during the event, the children’s awareness of the offence, and any significant changes that occurred post the knock in areas such as relationships, employment, and relocation. Psychometric measures of wellbeing, social support, and levels of trauma were included to assess the current state of the participant and the potential impact of the offending and the knock event on their lives.

2.3. Crisis Support

The Crisis Support Scale (CSS) measures social support received from others (Joseph et al. 1992). The CSS consists of seven items. Respondents are asked to rate the items on a seven-point Likert scale ranging from 1 (Never) to 7 (Always). The CSS assesses the extent to which individuals receive core aspects of crisis support, including emotional and practical help. In most research using the CSS, all seven items are asked twice, so participants rate support at the time of the traumatic event, the knock (Time 1) and at the present moment, current (Time 2). Elklit et al. (2001) examined the scale’s psychometric properties, reporting robust internal consistency, stability, discriminatory power, and concurrent validity in their analysis of data from 11 studies involving 4213 participants. In all but one of these studies, scores were assessed retrospectively at Time 1, as is the case in the current study. Item 6 was reversed before summing all seven items to obtain a total CSS score for both Time 1 (retrospectively) and Time 2 (currently). A higher score indicates a greater level of social support. In the present study, Cronbach’s alpha for CSS was 0.81 at Time 1 and 0.88 at Time 2.

2.4. Wellbeing

Wellbeing was assessed using the WHO-5 Wellbeing Index (Staehr 1998), a generic global rating scale that measures subjective wellbeing (Heun et al. 2001). The scale comprises five positive statements, such as “I have felt calm and relaxed.” The respondent rates how well each statement applied to them in the past two weeks on a six-point scale from 0 (at no time) to 5 (all the time). Scores are summed to produce an overall score ranging from 0 to 25. According to standard WHO-5 scoring procedures, the raw total is multiplied by 4, resulting in a transformed score from 0 to 100. A higher score indicates better wellbeing. A raw score below 13, or a transformed score below 52, suggests poor wellbeing. The scale has demonstrated adequate validity as a screening tool for depression and as an outcome measure in clinical trials (Topp et al. 2015), as well as good internal consistency reliability (Saipanish et al. 2009). In this study, Cronbach’s alpha was 0.93.

2.5. Trauma

Trauma was assessed with the International Trauma Questionnaire (ITQ). This is a self-report measure of post-traumatic stress disorder (PTSD) and complex post-traumatic stress disorder (CPTSD) (Cloitre et al. 2018). The ITQ assesses a distressing event for the respondent (in the current study, the knock) and asks about the time that has passed since that event (<6 months, 6–12 months, 1–5 years, 5–10 years, 10–20 years, >20 years ago). Responses on all items are assessed with a five-point Likert scale ranging from 0 (Not at all) to 4 (Extremely). ITQ scores reflect current trauma symptoms relating to a past event.

PTSD: Six items measure PTSD symptoms in the past month, with the range of possible scores from 0 to 24. An additional three items measure functional impairment in the previous month for (i) social life, (ii) work, and (iii) all other important activities.

DSO: A further six items measure symptoms of disturbances in self-organisation (DSO) and how people feel about themselves and others, with scores ranging from 0 to 24. An additional three items measure functional impairment in the previous month for (i) social life, (ii) work, and (iii) all other important activities.

The ITQ has good internal reliability (Karatzias et al. 2019). In the current study, Cronbach’s alpha for the ITQ for PTSD symptoms (relating to the knock) in the past month = 0.84. The Cronbach’s alpha for DSO = 0.87, and for the ITQ total (12 items, six from PTSD and six from DSO) = 0.89.

Diagnostic criteria for PTSD require a score of ≥2 from one of the three symptom clusters of re-experiencing, avoidance, or sense of threat, plus a score of ≥2 from one of the three functional impairments. Diagnostic criteria for CPTSD include satisfying PTSD criteria, in addition to scoring ≥2 for at least one symptom from each of the DSO symptom clusters (affective dysregulation, negative self-concept and disturbed relationships) and a score of ≥2 for one of the functional impairments. Using the ICD-11 taxonomic structure, respondents are diagnosed with either PTSD or CPTSD.

2.6. The Knock

A priori themes were identified to shape the open-ended text questions the authors developed to investigate the experience of the knock event. Five themes were identified: (1) Experience of the knock, (2) Police approach towards the person being investigated and arrested and their family, (3) Changes post-knock, (4) Social and psychological support needed, and (5) Children’s experience of the knock. Fifteen open-ended text questions were developed.

2.7. Data Analysis

Data were collated via Qualtrics (online survey programme), and the quantitative responses were imported into Excel. Numerically recoded qualitative data were added to the Excel file, which was imported into IBM SPSS Statistics (Version 30.0) for analysis. Qualitative responses were analysed using Nvivo (Version 15.1.2). Responses to the open-text questions were first read in full to gain familiarity with the data. Initial descriptive codes were then generated to capture recurring experiences and patterns across participants’ accounts. These codes were reviewed for frequency and relevance before being grouped into higher-order themes reflecting broader areas of shared experience. Once the candidate themes were established, they were checked against the original data to ensure they accurately represented participants’ responses, and were then reviewed by the wider authorship team. This process yielded five themes: Emotional Impact and Psychological Response, Police Approach and Procedural Experience, Practical and Material Repercussions, Social and Relational Consequences, and Coping and Support Strategies. The analysis was descriptive and exploratory, focusing on similarities and differences that emerged naturally from the data rather than following a formal coding framework. This approach enabled a clear overview of participants’ lived experiences and complemented the quantitative findings.

3. Results

3.1. Quantitative Findings

3.1.1. Sample Characteristics

Participant characteristics are shown in Table 1. Forty-eight partners completed the survey. Participants were predominantly women and all identified as White British. Most reported having children, and the majority of index offences were internet-mediated rather than contact or non-contact offences. Time since the knock varied widely, from the same year to almost two decades later.

Table 1. Sample characteristics of partners (n = 48).

Characteristic	Category/Statistic	n	%
Age (years)	M = 46.73 (SD = 11.65)	48	100
Gender	Female	47	97.9
	Male	1	2.1
Ethnicity	White British	48	100
Children	Yes	40	83.3
	No	8	16.7
Index offence type	Internet-mediated	40	83.3
	Contact	7	14.6
	Non-contact	1	2.1
Time since the knock (years)	M = 3.28 (SD = 3.54; range 0–19)	48	—

3.1.2. Wellbeing

As shown in Table 2, mean WHO-5 wellbeing scores were low. On average, partners scored substantially below the standard screening cut-off of 50, and around two-thirds fell below this threshold, indicating likely poor wellbeing. Time since the knock was not significantly associated with wellbeing (Table 3), suggesting that low wellbeing persisted irrespective of how long ago the initial police contact had occurred.

Table 2. Wellbeing and crisis support scores and paired comparisons (n = 48).

Measure	Time/Comparison	n	Mean	SD	Test vs. Comparator/Change	p	Effect Size/95% CI
WHO-5 wellbeing (0–100)	Current vs. cut-off 50	48	34.92	25.91	t(47) = −4.03 (vs. 50)	0.0002	Cohen’s d = −0.58; 95% CI for M [27.39, 42.44]
Crisis Support Scale (CSS)	Time 1 (at the knock)	48	22.98	8.36	—	—	—
	Time 2 (current)	48	26.96	10.08	—	—	—
CSS change (Time 2 − Time 1)	Paired difference	48	3.98	—	t(47) = −2.23	0.031	95% CI for ΔM [0.39, 7.57]

Note. Negative t values reflect the direction in which the difference was computed; the mean change (Time 2 − Time 1) is positive, indicating increased support over time.

Table 3. Associations between wellbeing, crisis support, diagnostic severity and time since the knock.

Predictor	Outcome	Statistic	Value	p
Time since the knock (years)	WHO-5 wellbeing	r	0.22	0.142
CSS Time 1 (at the knock)	WHO-5 wellbeing	r	0.19	0.187
CSS Time 2 (current)	WHO-5 wellbeing	r	0.52	<0.001
Diagnostic severity (No PTSD < PTSD < CPTSD)	WHO-5 wellbeing	ρ	−0.57	<0.001
CSS Time 1	Diagnostic severity	ρ	−0.35	0.014
CSS Time 2	Diagnostic severity	ρ	−0.31	0.030

Note. r = Pearson correlation coefficient; ρ = Spearman rank-order correlation. Higher WHO-5 scores indicate better wellbeing. Higher diagnostic severity scores indicate progression from no PTSD to PTSD and then to CPTSD.

3.1.3. Crisis Support

Descriptive and inferential statistics for crisis support are also presented in Table 2. Reported crisis support increased from Time 1 (retrospective rating at the knock) to Time 2 (current), with a statistically significant mean improvement. As shown in Table 3, current crisis support (Time 2) was moderately and positively associated with wellbeing, whereas retrospective support at Time 1 showed only a weak, non-significant association. This pattern suggests that partners who currently experience greater support tend to report better wellbeing, over and above the support they recall receiving at the time of the incident.

3.1.4. PTSD and Complex PTSD (ICD-11, ITQ)

ITQ diagnostic outcomes are summarised in Table 4. Almost nine in ten participants met criteria for either PTSD or CPTSD, with similar proportions meeting each diagnosis and only a small minority meeting neither. As shown in Table 4, diagnostic severity (ranging from no PTSD, to PTSD, to CPTSD) was strongly associated with lower wellbeing. Higher levels of crisis support at both time points were associated with lower diagnostic severity, indicating that those who perceived more support tended to show less severe trauma-related symptomatology.

Table 4. ICD-11 PTSD and Complex PTSD prevalence overall and by subgroups (ITQ; N = 48).

Subgroup	Diagnostic Category	n	%
Overall ITQ diagnostic categories	No PTSD/CPTSD	6	12.5
	PTSD	22	45.8
	CPTSD	20	41.7
	Any PTSD/CPTSD	42	87.5
Children in household—Yes (n = 40)	No PTSD/CPTSD	4	10
	PTSD	19	47.5
	CPTSD	17	42.5
Children in household—No (n = 8)	No PTSD/CPTSD	2	25
	PTSD	3	37.5
	CPTSD	3	37.5
Internet-mediated offence (n = 40)	No PTSD/CPTSD	5	12.5
	PTSD	19	47.5
	CPTSD	16	40
Contact offence (n = 7)	No PTSD/CPTSD	1	14.3
	PTSD	3	42.9
	CPTSD	3	42.9
Non-contact offence (n = 1)	No PTSD/CPTSD	0	0
	PTSD	0	0
	CPTSD	1	100

Note. Percentages are within each row subgroup. Caution is warranted when interpreting offence-type comparisons due to small and uneven group sizes, particularly for the non-contact group (n = 1). χ²(2) = 1.39, p = 0.500.

3.1.5. Children in the Household

Table 4 also presents diagnostic outcomes by the presence of children in the household and by offence type. Presence of children was not significantly associated with ITQ diagnostic category, and the proportions meeting PTSD and CPTSD were high in both groups. Similarly, wellbeing did not differ significantly by offence type. Descriptively, WHO-5 scores were low across internet-mediated, contact, and non-contact offences, and PTSD/CPTSD prevalence was high in all offence categories. Given the small and uneven group sizes, particularly for the non-contact group, these comparisons should be interpreted with caution.

3.1.6. Summary

Taken together, the quantitative findings indicate substantial psychological burden among partners following the knock. Wellbeing scores were markedly below clinical screening norms, and the vast majority met ICD-11 criteria for PTSD or CPTSD. Greater current crisis support was linked to better wellbeing and lower diagnostic severity, whereas time since the knock and family or offence-type characteristics showed no reliable associations with outcomes.

3.2. Qualitative Findings

3.2.1. Emotional Impact and Psychological Response

This theme captures the immediate and ongoing emotional, cognitive, and psychological reactions to the knock and arrest, including acute shock responses and longer-term distress. Across the dataset, partners described a profound and often overwhelming psychological shock following the knock and arrest, frequently using trauma-related language to convey the intensity of the moment and its aftermath. Nearly all respondents reported acute reactions such as disbelief, panic, “numbness,” physiological symptoms (shaking, sickness), and intrusive thoughts, with many describing weeks or months of persistent anxiety and dread. Several accounts evoke a sense of temporal dislocation or dissociation, feeling as though life had stopped or the world had “turned upside down,” followed by a long “waiting game” for information and outcomes. As one partner put it, “It was literally as though the floor had vanished from beneath me and I was free-falling.”

3.2.2. Police Approach and Procedural Experience

This theme encompasses partners’ perceptions of how officers conducted the knock, search, and arrest, including the manner of communication, the degree of intrusion, and any information or aftercare provided. Most participants described police attendance and searches as invasive and poorly explained, with mixed or negative experiences of communication and aftercare. Common features included teams in plain clothes, limited or delayed explanations, lengthy custody periods with few updates, and little consideration for the partner left at home. Some reported courteous conduct from individual officers, while a minority received leaflets or helpline details. However, the dominant narrative was one of being marginalised in the process during a moment of crisis. One respondent highlighted this dynamic sharply: “They rang me a couple of hours later… No one was concerned for me.”

3.2.3. Social and Relational Consequences

This theme reflects the impact of the arrest on intimate relationships, family dynamics, and social interactions, including experiences of stigma, shame, and altered trust. Stigma and social withdrawal were frequently reported. Many partners limited disclosure to a very small circle or concealed the situation altogether, fearing or experiencing judgment from friends, family, colleagues, and the wider community. Relationships were often strained; some ended, while others continued but under conditions of diminished trust and ongoing stress. Several respondents described feeling ashamed by association and feeling unsafe to seek understanding from their usual networks. As one participant wrote: “It’s not something I felt it was safe to discuss with anyone. The stigma of the whole thing is crushing.”

3.2.4. Practical and Material Repercussions

This theme covers the concrete, day-to-day consequences of the arrest, including disruptions to employment, housing, finances, and childcare arrangements. The events frequently caused immediate and lasting disruptions to housing, employment, and finances. Many partners reported job loss, reduced hours, extended sick leave, or the need to change roles. Several faced new childcare responsibilities and safeguarding-related restrictions. A notable minority relocated (sometimes with little notice) due to concerns about community visibility, media coverage, or personal safety. One account illustrates these overlapping pressures: “Yes, I had to sell my house and move to a new town… I had to quit my job because of stress… I haven’t worked since Jan 2018.”

3.2.5. Coping and Support Strategies

This theme describes the measures partners took to manage distress and seek help, including informal support networks, professional services, and suggestions for improving practice. Most respondents found some form of support over time, though pathways were rarely straightforward, and professional help was inconsistent. Partners often depended on informal networks, peer support groups, and, where available, specialist or generalist counselling; several highlighted that peer spaces felt especially validating, while others faced long waits, costs, or unhelpful responses from services. Many suggested practical improvements, such as an information pack, proactive signposting, or a liaison officer for partners. One participant captures the value of peer connection: “I am part of a WhatsApp group… They are so supportive and are a much-needed lifeline.”

4. Discussion

The results of the current study provide essential insights into the experiences and wellbeing of partners of individuals investigated, arrested and convicted for various types of sexual offences, exploring both the immediate and long-term aftermaths.

4.1. Demographics and Offence Type

All participants in this study were white British females, which limits the generalisability of the results to other demographic groups. This highlights the need for further studies with a more diverse sample to determine whether these experiences are consistent across cultural, racial, and gender groups. Most offences were internet-mediated, so further investigation would be needed to understand the impact of contact offences.

4.2. Wellbeing and Trauma

Partners of those who have undergone the knock reported wellbeing scores significantly lower than those of the general population, supporting the first hypothesis. This suggests profound psychological effects on partners, regardless of the offence type. Additionally, a substantial number of participants met the criteria for PTSD and complex PTSD, indicating a high level of trauma experienced by this group. Specifically, Armitage et al. (2023) found that, of 123 family members of individuals investigated for CSAM offences, most experience symptoms indicative of post-traumatic stress (72%) and some experience symptom levels that could suppress immune system functioning (67%). This somatic dimension is mirrored in the qualitative accounts in the current study, where participants described physical reactions such as uncontrollable shaking and sickness at the moment of the knock. Similarly, Salter et al. (2024) in a study of 45 partner and female relatives of CSAM offenders, found most participants (91%) reported negative impact on their mental health and approximately more than half (55.6%) experienced impact on their physical health, corroborating the markedly low WHO-5 scores in the present study and suggesting that this pattern of enduring psychological harm replicates across different studies.

Interestingly, although the time elapsed since the knock did not show a significant connection with wellbeing or trauma levels, there was a clear negative correlation between trauma levels and wellbeing, suggesting that as trauma increases, wellbeing decreases. This emphasises the potential ripple effect that a partner’s arrest and conviction could have on their significant other. The qualitative data generated in the current study give texture to this quantitative pattern. Partners described the moment of the knock in terms consistent with trauma, even with acute dissociation and derealization in some of them (“as though the floor had vanished from beneath me and I was free-falling”) and reported persistent psychological symptoms such as intrusive thoughts, panic, and numbness. These accounts align with the re-experiencing and avoidance symptom clusters captured by the ITQ, suggesting that the high rates of PTSD and CPTSD identified here reflect profound psychological disruption.

These findings are consistent with and extend emerging evidence in this area. Quantitative and survey-based studies document comparable levels of psychological harm among partners of sexual offences (see Armitage et al. 2023; Salter et al. 2024), while qualitative research confirms the profound nature of distress from the perspective of those who have lived it (Duncan et al. 2022; Wakeham et al. 2024).

4.3. Crisis Support

Notably, the link between crisis support and wellbeing was stronger closer to the time of the knock, indicating that immediate help might be more important in buffering against the initial shock and potential trauma. Over time, other factors might influence this relationship, weakening its strength. It also shows that timely interventions can be crucial in reducing the negative effects on wellbeing. The structural reason for the low level of crisis support at Time 1 can be explained by qualitative research in this field. Kamitz (2025) argues that the non-offending partners have historically been framed as instruments of rehabilitation rather than individuals with their own support needs, systematically deprioritising their wellbeing. Furthermore, Kamitz and Gannon (2024) found that non-offending partners frequently described interactions with professionals as inconsistent, ranging from positive (supportive and empathetic) to negative (dismissive and hostile).

The qualitative account in the current study aligns with the above: the same institutional absence and/or mixed experiences, with participants reporting being marginalised in the process during the knock and crisis, and limited and delayed explanations. Taken together, these findings suggest that the low Time 1 crisis support score is not simply a function of individual circumstances but reflects a systemic and structural gap in the response to this particular population.

However, by Time 2, it seems that increased crisis support is associated with reduced trauma, emphasising the continued importance of support even years after the incident. Importantly, the qualitative data from the current study suggest that this improvement in crisis support over time is not driven by structured professional intervention but by participants finding their own pathways to peer support groups and specialist counselling. Several participants highlighted the importance of peer spaces; they were spaces both for validation and for information. This aligns with Kavanagh et al. (2024) observation that peer-to-peer support groups have emerged as a critical resource precisely because services remain largely inaccessible to this population, providing emotional relief and reducing feelings of isolation where agency support has failed. Similarly, Kavanagh and Levenson (2022) found that group support interventions were effective for this population, providing emotional support and a sense of belonging—consistent with the improvement observed at Time 2.

4.4. Presence of Children and Trauma

Contrary to the fourth hypothesis, the presence of children did not show a significant positive correlation with trauma levels. This suggests that having children might not directly increase the trauma experienced by partners after the knock. However, this null finding should be taken with caution. While the presence of children did not significantly increase partners’ trauma score in the current study, having children in the household could introduce a distinct and additional layer of stressors that general trauma measures such as the ITQ may not fully capture. General trauma instruments are designed to capture individual symptom severity and may not be sensitive to the role-based burden specific to primary caregivers, such as the pressure to remain functional for the children while simultaneously managing their own crisis. Participants in the qualitative results described specific pressures arising from their role as the remaining caregivers, including safeguarding-related restrictions. This is consistent with what Armitage et al. (2024) describe as a burden of responsibility, in which non-offending parents manage the demand of the investigations and their own personal crisis while maintaining their caregiving role. Similarly, Kilmer and Leon (2017) documented that families with school-age children face a distinct set of challenges following a partner’s sexual offence, including managing disclosure, navigating restrictions on family activities and absorbing the social consequences for children. Future research should incorporate caregiving-specific measures alongside trauma instruments.

4.5. Differences by Offence Type

The study provides nuanced insights based on the type of offence:

• Contact Offences: These partners had lower immediate crisis support (T1) but saw a rise over time. They had a relatively better wellbeing score than non-contact offences but worse than internet offences. Their average trauma score met the criteria for PTSD.
• Internet Offences: Partners in this category seemed to fare the best, with the highest wellbeing and the lowest trauma scores. Their crisis support remained moderate at both time points.

These distinct patterns suggest that the nature of the offence significantly influences the partner’s psychological trajectory. Non-contact offences, despite not involving physical interactions, appear to elicit the most considerable distress and trauma.

4.6. Limitations

Several limitations should be acknowledged when interpreting these findings. First, the sample was relatively small and homogeneous, comprising almost exclusively White British women. As a result, the study may not reflect the experiences of partners from different ethnic backgrounds, genders, or cultural contexts. Recruitment through the Stop It Now! forum and social media may also have introduced self-selection bias, attracting participants who were already connected to support networks and potentially underrepresenting those who remain more isolated or distressed.

Second, the study relied on retrospective self-report measures, which are subject to recall bias, particularly in assessing perceived crisis support at the time of the knock. Similarly, the International Trauma Questionnaire captured current trauma symptoms relating to an event that may have occurred many years earlier, and therefore, the temporal relationship between cause and effect cannot be fully established.

Third, although both quantitative and qualitative methods were employed, integration between these strands was limited. Future studies could usefully explore how specific qualitative experiences, such as police conduct or social stigma, interact with quantitative outcomes such as wellbeing or post-traumatic stress.

Finally, the study’s cross-sectional design precludes causal inference. The relationships identified between crisis support, trauma, and wellbeing should therefore be interpreted as associations rather than evidence of directionality. Longitudinal designs would allow stronger conclusions to be drawn about the persistence and progression of these effects over time.

4.7. Implications and Future Directions

This study has examined the levels of trauma experienced by partners in response to ‘the knock’. While previous studies have explored trauma from a qualitative perspective, assessing experiences against validated scales, this study makes a significant contribution to the existing evidence on the harms to families caused by partners (and children) during an arrest (and subsequent investigation). These data quantify harms that will help influence policy and practice to meet those needs.

These findings highlight the importance of tailored interventions according to the offence type. Immediate crisis support seems essential for all groups; however, maintaining this support may be especially important for partners of individuals investigated and arrested for non-contact offences.

Furthermore, the study suggests that even if partners do not show immediate signs of trauma, the effects can emerge later. Ongoing assessments and interventions may be crucial for addressing and preventing these delayed reactions.

Finally, the lack of a connection between having children and trauma levels suggests that having children does not necessarily increase the trauma experienced. However, since most of the sample had children, further research could explore the indirect effects on the children and whether they influence the partner’s psychological wellbeing, as well as incorporate measures sensitive to caregiving-related stress and role burden.

The current study did not capture variables shown to predict adjustment in this population. Cassidy and Ronis (2025) identified self-esteem, instrumental support, and interpersonal relationship quality as significant predictors of adjustment among non-offending partners. Future studies should incorporate these variables to build a more comprehensive model of wellbeing in this group

In conclusion, the knock has significant and enduring effects on the partners of those investigated, arrested and convicted for sexual offences. A comprehensive, sustained support system that addresses both immediate and long-term needs is essential for this vulnerable, largely overlooked group. Future research should aim to include a more diverse sample and broaden its scope to better understand the broader familial impacts.