The use of general practice services by CSA survivors is of interest because of the potentially long-term adverse effects of CSA on physical and mental health. As with intimate partner violence (IPV), CSA is associated with diverse and long-term chronic illnesses that are difficult to treat, such as chronic pain syndromes,6 irritable bowel syndrome, chronic pelvic pain, back pain and headaches, as well as gynaecological problems, health risk behaviours and avoidance of preventive care.7–10 CSA and IPV are also associated with poorer long-term mental health, particularly depression, anxiety, post-traumatic stress disorder, eating disorders, suicide, somatisation disorders and substance abuse.11–13
GPs are ideally situated to respond to and manage the long-term effects of CSA because they provide ongoing care to women who have experienced CSA. Their practices are a potential location for intervention programs14 and potentially provide access to early intervention and counselling services that may lead to an improvement in the quality of life.15 Doctors believe that they can support and assist women,16,17 but there is a dissonance between what they think they can do and what is actually done.18 GPs’ perceptions of their roles, their confidence in screening and their knowledge of child maltreatment have been associated with both routine and targeted screening.19
The recently released World Health Organization (WHO) guidelines on responding to intimate partner and sexual violence recommends that healthcare providers should have in-service training that provides them with the knowledge and skills to better support women, and encourages supportive attitudes.20 The necessary context-specific skills include how to enquire, the best way to respond to women who have experienced violence, knowledge of existing support services and relevant laws.20
High levels of satisfaction have been reported in Australian general practice21 – more than 80% of a nationally representative sample of women reported their most recent GP visit was good, very good or excellent.1 Studies in Britain have shown the importance of patient-centred care, continuity of care and accessibility in patient satisfaction.22 The time GPs spent dealing with complex psychosocial issues with disadvantaged patients was positively related to the patient’s ability to cope with illness.23
CSA survivors describe the need to feel safe in their interactions with health professionals.16 Establishing trust and having health professionals who understand and are knowledgeable about trauma and its associated long-term problems are critical to the provision of good care.24 Consultations that are more patient-centred and collaborative improve information sharing, satisfaction with the consultation and physical examination and patient outcomes.16,24–26
North American studies have demonstrated associations between increased health service usage and child physical abuse (CPA) and CSA.27–32 The majority of these studies have been undertaken with clinical samples,27,29,30,32 and have concentrated on hospital-based, rather than primary care services. Chartier et al28 studied childhood abuse and health service utilisation in a community sample in Canada. Their study found an association between both CPA and CSA and high emergency room use but no association was found with higher general practice service use. Walker et al31 randomly sampled women members of a health maintenance organisation in the USA. They reported that women who experienced CSA had significantly higher annual healthcare costs of $245 (95% CI $132–381) above that of non-abused women. Median primary care costs were also increased by $32 per annum.
We could not identify any studies that investigated CSA and general practice service use in Australia. Most studies were from North America. None were identified that investigated CSA, general practice service use and patient service satisfaction with a community sample of women. The primary aim of our study was to explore whether general practice service use was higher and satisfaction lower for CSA survivors in an Australian representative community sample. Our hypothesis was that ongoing experiences of violence in adulthood would add to service use effects of CSA and further decrease service satisfaction.
Methods
Participants
The Australian Longitudinal Study on Women’s Health (ALSWH) has been described in detail elsewhere.33 To summarise, the ALSWH consists of three cohorts of Australian women aged 18–23 years (1973–78 cohort), 45–50 years (1946–51 cohort) and 70–75 years (1921–26 cohort) at the time of baseline survey in 1996. Participants were randomly sampled from the Medicare Australia database, which included all Australian citizens and permanent residents. Women from rural areas were deliberately oversampled to capture the health experiences of those women outside urban centres.33
The current study used data collected at Survey 4 (2006) of the 1973–78 cohort because it was the first and only survey to ask participants for information about their experiences of CSA. Of the 9145 participants (aged 28–33 years) who completed Survey 4, 7383 women answered questions that asked about CSA, IPV and adult physical and sexual assaults. ALSWH reported a retention rate of 67.5% at Survey 4.34
Measures
Demographics
Participants’ age, marital status (single, married, defacto, separated/divorced/widowed), place of residence (metropolitan or non-metropolitan, which consists of rural and/or remote areas), education (year 12 equivalent, no formal education, trade/apprentice/certificate/diploma, university degree/postgraduate degree) and income management (whether income is impossible/difficult or not too bad/easy to manage) were collected.
Abuse and violence
Three violence variables were created from the four single questions on CSA, IPV and adult physical and adult sexual violence in the original questionnaire. These variables were:
No violence
CSA only
Lifetime violence, which consisted of CSA with any of the 3 forms of adult violence listed above.
Women who experienced adult violence only were excluded from the analysis because our aim was to investigate the impact of CSA.
General practice service use
The frequency of consultation with a GP was assessed with a single item: ‘How many times have you consulted a family doctor or another GP for your own health in the last 12 months?’ General practice service use was dichotomised to low service use (1–4 times) and high service use (5 or more times) based on the average number of GP visits/person in Australia being 5.1.35
GP satisfaction
GP satisfaction was assessed using the modified version of the Group Health Association of America (GHAA) consumer satisfaction survey.36 The ALSWH modified GHAA (GP satisfaction scale) consists of five items assessing patients’ satisfaction level on the:
amount of time spent with the doctor
doctor’s explanation of the problem and treatment
doctor’s interest in how the patient felt about tests, treatment or advice given
opportunity to ask questions
personal manner of the GP (courtesy, respect, sensitivity, friendliness).
Responses were on a 5-point Likert scale that ranged from 1 (poor) to 5 (excellent). Cronbach’s alpha for the GP satisfaction scale calculated for this sample was 0.95 The GP satisfaction scale was dichotomised to 0=Not Satisfied (consisting of Fair and Poor responses) and 1=Satisfied (consisting of Excellent, Very Good, Good responses).1
Analyses
Data analyses were conducted using IBM SPSS Statistics 21.37 Descriptive analyses, such as frequencies, percentages and means, were used to show cumulative responses of participants with different abuse/violence experiences. Chi-square tests for independence were conducted to explore demographics of participants, general practice service use and GP satisfaction scale by different experiences of abuse/violence experiences.
Two direct logistic regressions were conducted to further explore differences in GP service use by abuse/violence experiences and differences in GP satisfaction of women with different abuse/violence experiences. The abuse/violence measures were first entered alone in Model 1 (unadjusted model), and then with demographic factors such as age, marital status, location of residence, education and income management in Model 2 (adjusted model). In the second regression, frequency of general practice service use was additionally controlled for when exploring differences in GP satisfaction by abuse/violence experiences.
Ethics
ALSWH has ethics clearance from Human Research Ethics Committees at the Universities of Newcastle and Queensland, Australia. The Monash University Human Research Ethics Committee (CF11/3391 – 2011001811), Victoria, Australia provided approval for the protocol of this study.
Results
Demographics and characteristics of participants by abuse experience(s) and their general practice service use and satisfaction are presented in Table 1. Chi-square tests for independence indicated significant associations between abuse and violence experiences with marital status, place of residence, education, income management, general practice service use and the GP satisfaction scale (Table 1).
Table 1. Demographics and characteristics of participants by abuse experiences and their GP service use and satisfaction
Variables
|
No abuse
|
CSA only
|
Lifetime violence
|
Age mean (SD)
|
N = 6264
30.57 (0.2)
|
N = 889
30.60 (0.05)
|
N = 230
30.76 (0.10)
|
Marital status
Single
Married
Defacto
Separated/divorced/widowed
|
N = 6243
1421 (22.8)
3567 (57.1)
1085 (17.4)
170 (2.7)
|
N = 885
186 (21.0)
505 (57.1)
159 (18.0)
35 (4.0)
*P <0.001
|
N = 226
62 (27.4)
62( 27.4)
64 (28.3)
38 (16.8)
|
Place of residence
Metro
Non-metro
|
N = 6178
3828 (62.0)
2350 (38.0)
|
N = 878
486 (55.4)
392 (44.6)
*P <0.001
|
N = 228
113 (49.6)
115 (50.4)
|
Education
Year 12 or equivalent
No formal qualifications or Year 10 equivalent
Trade/apprenticeship/certificate/diploma
University degree/Postgraduate degree
|
N = 6244
994 (15.9)
427 (6.8)
1567 (25.1)
3256 (52.1)
|
N = 884
171 (19.3)
100 (11.3)
266 (30.1)
347 (39.3)
*P <0.001
|
N = 230
34 (14.8)
50 (21.7)
86 (37.4)
60 (26.1)
|
Income management
Unmanageable
Manageable
|
N = 6234
2299 (36.9)
3935 (63.1)
|
N = 889
427 (48.0)
462 (52.0)
*P <0.001
|
N = 228
147 (64.5)
81 (35.5)
|
GP service use
Low (0–4 times)
High (5 or more times)
|
N = 6250
4460 (71.4)
1790 (28.6)
|
N = 888
610 (68.7)
278 (31.3)
*P <0.001
|
N = 230
125 (54.3)
105 (45.7)
|
GP satisfaction scale
Not satisfied
Satisfied
|
N = 5485
382 (7.0)
5103 (93.0)
|
N = 777
71 (9.1)
706 (90.9)
*P = 0.004
|
N = 192
23 (12.0)
169 (88.0)
|
Note: N varied due to missing responses
*Pearson Chi-square test for independence exploring associations between abuse categories and socio-demographics and GP variables
|
Of the women surveyed, 6264 (84.8%) reported no experiences of violence; 889 (12%) reported CSA; and 230 (3.1%) reported lifetime violence (CSA and IPV/adult sexual violence).
There was no evidence to suggest that women with CSA experiences alone had higher general practice service use (AOR = 1.06, 95%CI: 0.91–1.24), compared with those without any experiences of violence. Women with lifetime violence experiences had almost twice the odds of higher general practice service use, compared with those without any violence experiences, after adjusting for demographic factors (AOR = 1.82, 95% CI: 1.37–2.40; Table 2).
Table 2. Odds of reporting high general practice service use by abuse experiences
Variables
|
Model 1 (unadjusted)
|
Model 2 (adjusted for demographics)*
|
OR (95% CI)
|
P
|
OR (95% CI)
|
P
|
Abuse/violence experiences (reference category: no violence)
|
CSA
|
1.14 (0.98–1.32)
|
0.101
|
1.06 (0.91–1.24)
|
0.450
|
Lifetime violence
|
2.09 (1.61–2.73)
|
<0.001
|
1.82 (1.37–2.40)
|
<0.001
|
Age
|
|
|
0.96 (0.93–1.00)
|
0.036
|
Marital Status (reference category: single)
|
Married
|
|
|
1.25 (1.10–1.43)
|
0.001
|
Defacto
|
|
|
1.01 (0.86–1.20)
|
0.898
|
Separated/divorced/widowed
|
|
|
1.04 (0.40–0.77)
|
0.810
|
RRMA (reference category: metro)
|
Non-metro
|
|
|
0.965 (0.08–0.87)
|
0.519
|
Education (reference category: Year 12 or equivalent)
|
No formal qualifications or Year 10 equivalent
|
|
|
1.05 (0.85–1.31)
|
0.643
|
Trade/apprenticeship/ certificate/ diploma
|
|
|
1.14 (0.98–1.34)
|
0.091
|
University degree/Postgraduate degree
|
|
|
0.84 (0.72–0.97)
|
0.018
|
Income management (reference category: unmanageable)
|
Manageable
|
|
|
0.68 (0.61–0.76)
|
<0.001
|
*Adjusted for demographic variables age, marital status, place of residence, education and income management
CI, confidence interval; CSA, childhood sexual abuse; OR, odds ratio; RRMA, rural, remote and metropolitan areas
|
The odds of women with CSA being satisfied with their GP visits was approximately 0.31 times lower than those without violence experiences, after adjusting for demographic factors (AOR = 0.69, 95% CI: 0.53–0.91; Table 3). When frequency of general practice service use was added to the model, there was no significant change in the odds of satisfaction with general practice services. (AOR = 0.69, 95% CI: 0.53–0.90; Table 3).
Table 3. Odds of reporting GP satisfaction by abuse and violence experiences
Variables
|
Model 1 (unadjusted)
|
Model 2 (adjusted for demographics)*
|
Model 3 (fully adjusted)†
|
OR (95% CI)
|
P
|
OR (95% CI)
|
P
|
OR (95% CI)
|
P
|
Abuse/violence experiences (reference category: no violence)
|
CSA
|
0.74 (0.57–0.97)
|
0.029
|
0.69 (0.53–0.91)
|
0.008
|
0.69 (0.53–0 .90)
|
0.007
|
Lifetime violence
|
0.55 (0.35–0.86)
|
0.009
|
0.69 (0.42–1.13)
|
0.141
|
0.67 (0.41–1.09)
|
0.106
|
Age
|
|
|
1.00 (0.94–1.07)
|
0.973
|
1.00 (0.94–1.07)
|
1.00
|
Marital status (reference category: single)
|
Married
|
|
|
1.96 (1.55–2.47)
|
<0.001
|
1.95 (1.54– 2.46)
|
<0.001
|
Defacto
|
|
|
0.87 (0.67–1.13)
|
0.286
|
0.87 (0.67–1.13)
|
0.308
|
Separated/divorced/widowed
|
|
|
1.15 (0.68–1.94)
|
0.610
|
1.16 (0.68–1.96)
|
0.590
|
RRMA (reference category: metro)
|
Non-metro
|
|
|
1.28 (1.04–1.58)
|
0.019
|
1.29 (1.04–1.59)
|
0.019
|
Education (reference category: Year 12 or equivalent)
|
No formal qualifications or Year 10 equivalent
|
|
|
1.15 (0.74–1.82)
|
0.523
|
1.16 (0.74–1.82)
|
0.527
|
Trade/apprenticeship/certificate/diploma
|
|
|
0.99 (0.73–1.36)
|
0.967
|
0.98 (0.72–1.35)
|
0.920
|
University degree/postgraduate degree
|
|
|
0.85 (0.64–1.14)
|
0.273
|
0.86 (0.64–1.15)
|
0.303
|
Income management (reference category: unmanageable)
|
Manageable
|
|
|
1.14 (0.93–1.40)
|
0.217
|
1.16 (0.94–1.42)
|
0.160
|
Frequency of GP visit (reference category: low use)
|
High use
|
|
|
|
|
1.28 (1.02–1.59)
|
0.031
|
*Adjusted for demographic variables age, marital status, place of residence, education and income management
†Adjusted for all variables listed in Model 2 and GP service use
CI, confidence interval; CSA, childhood sexual abuse; OR, odds ratio; RRMA, rural, remote and metropolitan areas
|
Despite low adjusted odds ratios there was no evidence to suggest an association between lifetime violence and satisfaction with GP visits, after adjusting for demographic variables (AOR = 0.69, 95% CI: 0.42–1.13) and after adjusting for demographic variables and frequency of GP visits (AOR = 0.67, 95% CI: 0.41–1.09).
Discussion
In this paper our purpose was to explore association of CSA with general practice service utilisation and service satisfaction.
CSA was not associated with higher general practice service use for Australian women who reported CSA when compared with women who did not report violence. However, before and after adjusting for demographic characteristics, the use of services was significantly higher for those who experienced both CSA and adult violence than for those who did not experience violence.
To better understand the general practice service utilisation by CSA survivors, this study went on to explore service satisfaction and particular areas of service that may be problematic. General practice service satisfaction was significantly lower among women who had experienced CSA, compared with women who had not experienced CSA, both before and after adjusting for demographics and frequency of service access. It would seem that these women are accessing general practice but their needs are not being met as well as those of women who have not experienced CSA.
Previous work has reported that women who have experienced CSA may find it difficult to disclose unless asked directly by their doctor. Reasons for this vary and include shame, fear of the doctor’s response, coping by not thinking of past traumatic experiences and re-experiencing trauma.16 A recent Australian study found that over 58% of women who had been asked about their childhood maltreatment experiences by their GP were ‘relieved’ or ‘hopeful’.38 Physical examinations, particularly gynaecological examination, breast examination and pap smears, may also be more difficult.26,39,40 These are common examinations for women in the age group (28–33 years) studied, and may explain why women in this study reported being less satisfied with their GP. Issues of trust, safety, anxiety, stress and even dissociation during the clinical encounter have been identified in qualitative studies of the perinatal period with CSA survivors.26
This study is relevant to GPs because many Australian women experience CSA and its health impacts of poorer long-term mental and physical health. GPs are potentially the first professional responders, educating women about the associations between their health and their experiences of childhood violence, asking sensitive questions, exploring their patients’ childhood experiences and supporting their patients’ access to counselling and specialist sexual violence services.
Practising ‘trauma-informed care’ could assist GPs to tailor their practice to benefit not only women who have experienced CSA, but also those with chronic and complex mental health issues and other complex traumas.41,42 Patient-centred care can be extended to become trauma-informed practice. The key principles involved are physical and emotional safety, developing trust over time with the GP and other clinic staff, developing an understanding of trauma and its effects on the patient behaviour, and the clinical encounter, including sometimes difficult and complex coping behaviours by patients.41 Shifting to a more collaborative management style maximises the patient’s feelings of choice and control. Discussing management options and offering choices are important ways that GPs can empower and help patients to develop new skills and coping strategies.
GPs may remain challenged by childhood maltreatment, particularly sexual violence, feeling they can help but lacking the education, resources and confidence to do so.38 A broader qualitative exploration of patient and GP perspectives is required so that general practice responses to CSA can be improved for the patient and the doctor.
Strengths and limitations
The strength of this study is that it is a cohort study of a nationally representative sample and focuses on child and adult violence. However, the study focuses on young Australian women and the findings may not be applicable to other age groups.
There are methodological limitations inherent in the study. The survey asked a single non-behavioural question about CSA, adult violence and adult sexual violence, and as such it is likely to result in underreporting of CSA, adult violence and adult sexual violence. Other forms of child abuse were not included in the survey.
Implications for practice
CSA survivors were found to be less satisfied with their visits to their GPs. Understanding more about the impact of trauma on patient behaviour and using trauma-informed practice guidelines may assist GPs to improve the clinical services for women who have experienced violence. Further research is required to better understand the issues that confront consumers and their GPs this area.
Competing interests: None.
Provenance and peer review: Not commissioned, externally peer reviewed.
Acknowledgements
The research on which this paper is based was conducted as part of the Australian Longitudinal Study of Women’s Health at the University of Newcastle and the University of Queensland. We are grateful to the Australian Government Department of Health for funding and to the women who participated in the survey.
Financial support for the analysis was provided by a Monash University Faculty of Medicine, Nursing and Health Sciences Small Grant in 2012 ($15,000).