The study was approved by a Research Ethics Committee established by the NHS Health Research Authority (NRES Committee South West – Frenchay, REC reference 12/SW/0162 of 25 June 2012). This is a 6-month, prospective, observational study based on a convenience sample. We will recruit index participants from the pool of women referred to specialized domestic violence agency SURVIVE South Gloucestershire and Bristol for community outreach support or refuge accommodation by other agencies or self-referred. The SURVIVE agency is a charitable organization which provides a range of services to women who have experienced DVA, including three refuges for women with children, outreach support, community-based Freedom Program groups, Independent Domestic Violence Advisers, support phone line, and Information Sessions in South Gloucestershire, and Bristol UK.
Inclusion criteria for participants contain age ≥ 18 y.o. Exclusion criteria include inability to read English, current use of steroid-based medications, pregnancy, presence of pituitary and/or adrenal gland disorder, and symptomatic psychotic illness. No participant will be excluded on the basis of their disability, ethnicity, religion or sexual orientation.
Recruitment
Potential index participants will be identified by written invitation letters with invitation slip attached offered by researcher or agency staff to every woman referred or self-referred to domestic violence agency SURVIVE. Woman willing to be approached to discuss potential participation in the study can either send her contact details to the researcher by phone or e-mail, or put down her contact details on the invitation slip, tear it off and return to the researcher or agency staff who has given the invitation to her.
The researcher will collect contact details sent by phone, e-mail, and completed invitation slips and phone all women willing to be approach to hear more about the study. When initial telephone contact has been made with a woman, the researcher will assess her eligibility and arrange a face-to-face meeting with all women interested and eligible to take part, meeting the woman at a safe and convenient place. As part of the first face-to-face meeting, the researcher will give a detailed participant information sheet to the woman, explaining each point and answer any questions that the woman may have prior to seeking her written informed consent.
The researcher will ask every woman already consented to participate in the study to suggest up to three female friends or family members of the same age group who are aware of her current circumstances to be her potential controls. The researcher will emphasize that suggesting friend controls is completely optional. If a woman does not want to recommend friend controls it will not affect her participation in the study nor the services she receives. If a woman wishes to suggest friends as potential controls the researcher will ask her to pass ‘An invitation to the CEASE study’ to those people. ‘An invitation to the CEASE study’ and all documentation for friend controls present the study as biological research on the impact of family stress on women's health and wellbeing and do not mention SURVIVE agency. Although it does state that the research is about DVA, the researcher will ask the index participant only to pass the invitation to confidantes who know that she has experienced abuse. A friend control willing to discuss potential participation in the study will send her contact details to the researcher by post, phone, text, or e-mail. The researcher will contact potential friend controls by phone to give brief information about the study, assess their eligibility and arrange a face-to-face meeting at a safe and convenient place. As part of the first face-to-face meeting, the researcher will give a detailed participant information sheet to the friend control, explaining each point and answer any questions that the woman may have prior to seeking her written informed consent. In parallel with recruitment of friend controls through index research participants we will approach potential controls in local communities through invitation letters displayed in public places (e.g. post office, library, supermarket, community centre) in areas surrounding SURVIVE sites in South Gloucestershire and Bristol.
Measurements
This study will consist of one baseline and two follow-up measurements. Baseline evaluation will be carried out at the first face-to-face meeting between researcher and woman and during the following week. Women will be asked to undertake follow-up assessment at 3 and 6 months after joining the study. With regard to safety preferences participants will have a choice to receive and return follow-up questionnaire and saliva collection kit by post or at the follow-up face-to-face meeting with the researcher. Each assessment will last approximately 45 minutes and will include: (1) numerous standardized self-administered questionnaires to assess socio-demographics, mental and physical health, (2) weight and height measurement, and (3) self-completion of three saliva samples. The questionnaire and saliva samples will be completed by the participant at a place that is safe and convenient to her and returned by post or via collection by the researcher. Research plan and measurements are summarized in Additional file
2.
Socio-demographic information will be collected through use of a standardized self-report questionnaire and include age, ethnicity, children, qualifications, employment status, total annual income and relationship to perpetrator.
Exposure to abuse through the lifespan will be measured using two psychological instruments. First, cases of childhood abuse and neglect will be detected using the Childhood Trauma Questionnaire (CTQ). CTQ is a 28-item inventory that provides brief, reliable and valid screening for histories of abuse and neglect. The CTQ inquires about five types of maltreatment – emotional, physical, and sexual abuse, and emotional and physical neglect – with five items representing each type. Participants respond to a series of statements about childhood events, which are endorsed on a 5-point frequency scale (0 = “never true”, 4 = “very often true”). Item scores are then summed to produce scores that quantify the severity of maltreatment in each area and that can be compared to clinical data. Cut-off scores for detecting likely cases of abuse and neglect are also provided. The CTQ was validated with data from over 2,000 respondents, including both clinical and non referred groups [
33]. Second, exposure to DVA from intimate partner and other family members will be measured using the Composite Abuse Scale (CAS). The CAS is a widely used self-report of behaviors that women describe as abusive by their partners [
34]. It is an easily administered measure that provides standardized sub scale scores on four dimensions of DVA. It consists of 30 items presented in a six point format requiring respondents to answer “never”, “only once”, “several times”, “monthly”, “weekly” or “daily” in a twelve month period. The Severe Combined Abuse sub scale has 8 items that represent severe physical abuse items, all sexual abuse items, and physical isolation aspects of emotional abuse. The Emotional Abuse subscale has 11 items that include verbal, psychological, dominance and social isolation abuse items. The Physical Abuse subscale has 7 of the less severe physical abuse items and the Harassment subscale has 4 items that are about actual harassment. CAS is scored by summating the frequency scores of the 30 items. The 0–5 for each item thus gives a possible score for each sub scale. Cut-off scores to exclude women labeled abused incorrectly are provided. The CAS can give not only prevalence figures, but also associations with other physical and emotional co-morbidities. It has been used in specialized domestic violence agencies, general practice, antenatal clinics, emergency departments, and drug and alcohol clinics. The CAS has demonstrated good acceptability and reliability, with some evidence of criterion and external validity. To measure DVA duration an item is added to the CAS asking participants the length (“one occasion”, “up to 6 months”, “up to 1 year”, “up to 3 years”, “up to 5 years”, and “more than 5 years”) of abusive relationship.
Levels of stress and mental health of the participants will be evaluated using five psychological instruments for detecting the most common mental disorders alongside with standardized questionnaire on use of alcohol and drugs. First, the Generalized Anxiety Disorder Scale (GAD-7) will be used to screen on probable cases of generalized anxiety disorder and to assess symptom severity. The GAD-7 is based on the most prominent diagnostic features of the DSM-IV diagnostic criteria for generalized anxiety disorder and has excellent reliability, as well as criterion, construct, factorial, and procedural validity. The 7 items assess the frequency of core symptoms of generalized anxiety disorder within the past two weeks. Items are rated on a 4-point frequency scale (0 = “not at all”, 3 = “nearly every day”) with a total score ranging from 0 to 21. A score up to 4 indicates the absence of generalized anxiety disorder, scores of 5–9 represent mild, scores of 10–14 represent moderate and scores of 15 and higher represent severe anxiety symptoms levels [
35].
Second, Patient Health Questionnaire Depression Scale (PHQ-9) will be used to diagnose major depression and assess symptom severity. The PHQ-9 is based on the DSM-IV diagnostic criteria for major depressive disorder and has excellent reliability, as well as criterion, construct, factorial and procedural validity. The 9 items assess the frequency of depressive symptoms within the past two weeks. Items are scored on a 4-point frequency scale from 0 = “not at all” to 3 – “nearly every day” with a total score ranging from 0 to 27. A cut-off point of >9 is recommended for the screening of any depressive disorder. A score up to 4 indicates the absence of depression, scores of 5–9 represent mild, scores of 10–14 represent moderate and scores of 15 and higher represent severe depression [
35,
36].
Third, PTSD will be diagnosed using the PTSD Symptom Scale: Self-Reported version (PSS-SR). The PSS-SR is a 17-item measure that assesses the 17 DSM-IV symptoms of PTSD. The severity over the last two weeks of each item on the PSS-SR is rated using a 4-point scale (0 = “not at all”, 4 = “every day”). The total score is calculated as the sum of severity ratings for the 17 items. The total score > 13 indicates on likelihood of PTSD. The PSS-SR has satisfactory internal consistency, high test-retest reliability, and good concurrent validity [
37].
Fourth, the Sheehan Disability Scale (SDS) will be used to assess functional impairment in three inter-related domains; work/school, social and family life. The participant rates the extent to which work/school, social life and home life or family responsibilities are impaired by her symptoms on a 10-point visual analog scale. Score 5 or greater on any of the three scales is associated with significant functional impairment. The SDS has high sensitivity and good specificity for patients with most frequent mental disorders [
38].
Fifth, the Perceived Stress Scale (PSS) will be used for measuring the perception of stress [
39]. This widely used 10-item psychological instrument with good internal reliability and moderate construct validity measures the degree to which situations in one’s life are appraised as stressful over the past month. It assesses the amount of stress in one’s life rather than response to a specific stress and had been used in studies of both mental and physical health. The questions in the PSS ask about feelings and thoughts during the last month. In each case, respondents are asked how often they felt a certain way. Answers are rated on a 5-point frequency scale (0 = “never”, 4 = “very often”). PSS scores are obtained by reversing responses to the four positively stated items and then summing across all scale items. Scores can range from 0 to 40, with higher scores indicating greater stress.
Finally, use of alcohol and drugs will be measured by asking questions about quantity and frequency of substance use.
Physical health of the participants will be evaluated by body mass index (BMI) and a standardized questionnaire. To calculate BMI the researcher will follow standard clinical protocol on taking standing height and weight. Height will be measured to the nearest 1 mm (KaWe person-check, Germany), weight to the nearest 0.1 kg (Seca 803 Digital Personal Scale, Germany). Physical health problems will be measured using a modification of the Miller Abuse Physical Symptom and Injury Scale (MAPSAIS). This self-reported scale lists 9 injuries and 44 symptoms and illnesses related to DVA and asks if the woman was hospitalized, visited emergency department or undertaken surgery in the past 12 months. The scale was designed specifically for measuring long-term health consequences of DVA and demonstrated good content validity and test-retest reliability [
40].
Upon completion of questionnaire and anthropometry study participants will be given a saliva sample collection kit that includes a saliva collection device (Salivette tubes) with cotton pledget, an illustrated instruction sheet written in plain language explaining the procedure and providing helpful hints (such as putting the tube next to the bed for morning sample), a saliva collection diary in which participants can record experiences on the day of sample collection and a prepaid packaging addressed to the accredited laboratory compliant with current postal regulations. The participants will be asked to collect three saliva samples as soon as possible after questionnaire survey:
1.
Sample 1 “bedtime” – right before getting into bed.
2.
Sample 2 “awakening” – next morning as soon as the woman wakes up and is ready to get up for the day.
3.
Sample 3 “post-awakening” – 30 minutes after taking awakening sample.
Saliva will be collected by either passive drooling into the tube or by chewing (for 2 minutes) the cotton pledget provided with each tube. Participants must have waited at least 30 minutes after smoking, consuming any food or drink or teeth brushing before collecting the saliva sample. Tubes will be labeled with patient identifier and returned to the laboratory by post or via collection by the researcher. Upon arrival to the laboratory tubes with saliva will be centrifuged to remove the upper saliva layer for subsequent measurement of cortisol. The primary tube including the cotton pledget with individual cells will be then discarded in routine clinical waste. The upper saliva layer will be frozen until analysis. The saliva samples will be analyzed in the Department of Clinical Biochemistry at the Bristol Royal Infirmary using Ultra performance liquid chromatography – tandem mass spectrometry (UPLC-MSMS). The analysis also simultaneously measures cortisone a metabolite of cortisol as an additional measure of circulating cortisol [
41,
42].
To collect data on known covariates that have been related to diurnal cortisol secretion participants will complete saliva collection diary [
22]. Data collected on the day of saliva sampling include: schedule (whether weekend vs. weekday), time of awakening, lifestyle (smoking, alcohol use, caffeine, exercise/activity level, perceived stress and negative mood, recent meal, medications) and menstrual timing.
Statistical analysis
For the sample size calculation we used findings from similar studies in comparable populations [
24,
43,
44]. Thus, the number of participants required to detect a standardized difference in cortisol levels of 0.5 between abused women and non-abused controls with 80% power using a cut-off score for statistical significance of 0.05 is 128 (64 per group). Assuming 40% drop out rate before the study ends, around 213 eligible subjects would have to be approached. Assuming 60% response rate we will have to approach approximately 355 women, so 213 would consent to join the study, and of these 128 would be expected to complete the study.
Data will be coded and entered into research database during data collection by the two researchers responsible for recruitment. Stata software will be used for statistical analysis. All data will be checked for expected ranges, presence of outliers and abnormal values.
Sample will be analyzed by comparison of participant exposed to DVA and participants without DVA exposure on the basis of cortisol and mental health state data. Primary study outcome will be reported as diurnal cortisol variation, the degree of decline in cortisol across the day. Secondary study outcomes include CAR, size of post-awakening surge in cortisol that occurs in 30 minutes after awakening and mean salivary cortisol concentration across the day. As an additional measure of tissue cortisol the three saliva samples, awakening, post awakening, and bedtime, will also be assayed for cortisone.
Following descriptive statistics for all the cortisol and mental health outcomes, relationship between them will be investigated using appropriate linear (and, for binary outcomes, logistic) regression models. Using the CAS and CTQ data, these techniques will also be used to investigate the extent to which cortisol measures act as potential mediator between the type, duration and cessation of abuse and mental health condition for this group of women.
Missing data from women who dropped out of the study will be handled first by comparing their baseline variables with those who completed the study, and secondly using imputation techniques [
45].