SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Objective To investigate and compare the demographic and psychosocial profiles of women with chronic pelvic pain, chronic pain in a different site, and those with no history of pain with specific reference to a history of sexual abuse.

Design A prospective comparative study.

Setting Pelvic Pain Clinic at Leicester General Hospital NHS Trust, Pain Management Clinic at Leicester Royal Infirmary NHS Trust and two General Practices.

Participants Thirty women with chronic pelvic pain, 30 women with chronic pain in a different site and 30 women attending their general practitioner with no history of pain.

Interventions A specifically designed patient profile questionnaire to identify and explore incidents of sexual and physical abuse was administered to each woman by a research psychologist for confidential self-completion. Data were also collected on other demographic, medical and psychosocial characteristics.

Results Women with chronic pelvic pain were found to have a higher lifetime prevalence of sexual abuse, involving penetration or other genital contact compared with the two comparison groups. The prevalence of physical abuse was the same in all groups. Women in the pelvic pain group were more likely to have approached their GP for symptoms not related to pelvic pain than women in the other two groups and the incidence of clinical anxiety was significantly higher in this group compared with the pain-free group. The prevalence of sexual problems was much higher in the group with pelvic pain compared with the other two groups.

Conclusion These findings indicate that women with chronic pelvic pain have a higher incidence of past sexual abuse compared with women in a comparison pain group and with women with no pain.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Chronic pelvic pain is a common and disabling condition, responsible for 10% of outpatient consultations1, 40%, of laparoscopies2 and is listed as the indication for 10% to 12% of the hysterectomies performed in the United States accounting for approximately 70,000 procedures annually3,4. The pathophysiology of chronic pelvic pain is poorly understood. The prevalence and type of pathology found at laparoscopy varies from study to study. Often there is little relationship between the intensity of the pain described and the extent of observed pathology. Women with chronic pelvic pain can have normal laparoscopies and women without chronic pelvic pain have a significant incidence of abnormal laparoscopic findings2.

Studies from the United States indicate a higher incidence of childhood and adult sexual abuse in women with chronic pelvic pain compared with those with no pelvic pain5,6,7 and also in women with pelvic pain, compared with women with chronic headache8. Two uncontrolled studies from Europe have found that the incidence of sexual abuse in women with pelvic pain is close to that estimated in normal populations9,10. However, a Norwegian study found a statistically significant association between adverse sexual events and a history of pelvic pain in women who did not have a history of pelvic inflammatory disease and who were living in a physically abusive relationship11.

The aim of this study was to compare the incidence of sexual abuse in a group of women with chronic pelvic pain with or without obvious pathology, with a comparable group of women with pain in another site and a pain-free comparison group. We sought to establish whether the incidence of sexual abuse in the women with pelvic pain was higher than in either of the other two groups.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

The study comprised 90 English-speaking women between the ages of 16 and 50 years, recruited between March 1993 and April 1994 from three different sources.

The index group consisted of 30 consecutive referrals to the Leicester Pelvic Pain Clinic which is a tertiary referral clinic. All the women had been investigated by gynaecologists and their pelvic pain was attributed to a variety of diagnoses following investigation by laparoscopy or ultrasound. One-third of the women had undergone a diagnostic laparoscopy which revealed no intra-abdominal pathology. In 10 women a diagnosis of pelvic inflammatory disease had been made following laparoscopic findings; four also had ovarian cysts and endometriosis. Five women had been given a single diagnosis of ovarian cysts and three women had been told that they had endometriosis. Two women were diagnosed as having adhesions and one as having fibroids.

This group was compared with two other groups of women. The comparison pain group consisted of 30 consecutive referrals to the Pain Management Clinic who matched the age criteria and who had experienced pain in a site other than the pelvis for at least six months. The majority of these women had musculoskeletal pain such as headache or backache, and a small number had complex regional pain syndrome (Type I and Type II). The pain-free group consisted of 30 women randomly selected from two general practitioner surgeries: one suburban and the other a county town practice. Women were excluded from this group if they did not meet the age criteria or if they had a history of pain. The consent of each woman's general practitioner was obtained before participation in the study.

Each prospective participant gave informed written consent. The women recruited to the study were asked to complete an anonymous, confidential, self-report questionnaire reporting demographic details, gynaecological, medical and psychosexual history and any history of sexual and physical abuse. The questions about sexual abuse were adapted from the Sexual Life Events Inventory used by Palmer et al.12. The women were also asked to complete a somatisation scale, previously reported by Reiter and Gambone6. This consists of a checklist for 26 common somatic and emotional symptoms unrelated to the pelvic or abdominal pain for which they had been treated or seen a doctor (e.g. headache, mood swings, premenstrual syndrome and dizziness). A net somatisation score was determined for each woman on the basis of the total number of items endorsed on the scale.

The questionnaire was administered by a research psychologist who remained with the participant to answer any questions and provide counselling if necessary. All the women also completed the Hospital Anxiety and Depression questionnaire13. This questionnare is a standardised and widely used scale for assessing anxiety and depression in nonpsychiatric patient groups. The information was analysed using SPSS for Windows, version 6.1. Groups were compared using χ2 tests and ordinal data were analysed using Mann-Whitney U and Kruskall-Wallis tests.

The study was approved by the Leicestershire Ethics Committee.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

All the women approached agreed to participate in this study.

Demographic details

The three groups were comparable in terms of age, years in education and number of children (Table 1). Ethnic origin was also comparable, with only five women in the entire sample not describing themselves as white European. However, there were significantly fewer married women in the pain-free group compared with the other two groups. There was also a trend towards fewer women in the index group being in paid employment (41%), compared with the comparison pain group (62%) and the pain-free group (70%), although this failed to reach statistical significance.

Table 1.  Demographic details for the three groups. Values are given as mean [SD], median {range} or n (%). Group 1 = index group (n= 30); Group 2 = comparison pain group (n= 30); Group 2 = pain-free group (n= 30).
Demographic detailsGroup 1Group 2Group 3
Age (years)30.1 [5.23]33.6 [9.25]30.4 [8.10]
Years in education12.6 [2.58]12.5 [2.23]12.2 [2.23]
No. of children2.0 {0–3}0/1 {0–3}1 {0–4}
Married21 (70)20 (67)14 (47)
In paid employment12 (41)18 (62)19 (70)

Obstetric and gynaecological details

Menstrual and reproductive details for the three groups are summarised in Table 2. The groups were comparable in terms of the mean age of menarche and the duration of menses. A third of the women with pelvic pain were no longer menstruating. This was due to having had a hysterectomy in all but two: one was taking medroxy-progesterone acetate and the other had never menstruated because of congenital absence of the uterus. Laparoscopic examination had confirmed the presence of normal ovaries and a normal vagina in this woman. Women with pelvic pain were significantly more likely to experience problems with dysmenorrhoea than women in the pain-free group (73% vs 23%, P < 0.01). The women with chronic pelvic pain had also experienced more obstetric problems with almost twice as many having previously miscarried or been investigated for infertility compared with each of the other two groups. Three of the women in the index group had experienced a stillbirth compared with none in the other two groups.

Table 2.  Comparison of menstrual and reproductive details for the three groups. Values are given as mean [SD] or n (%). Key as for Table 1.
Menstrual & reproductive detailsGroup 1Group 2Group 3
  1. *Percentage of those women still menstruating.

  2. † P < 0.01 (compared with Group 3: χ2, test).

Menarche (years)13 [2.54]13 [1.54]13 [0.91]
Amenorrheic10 (33)0(0)2(7)
Dysmenorrhea15* (75)†14 (48)9* (32)
Menorrhagia13* (65)10 (33)11* (32)
Termination of pregnancy8(27)6 (20)3(10)
Miscarriage7 (23)4 (13)4 (13)
Stillbirth3 (3)0 (0)0 (0)
Infertility investigations6 (20)3 (10)0 (0)

Six women in the comparison pain group did not report chronic pelvic pain despite known gynaecological pathology (Table 3). Two women in the index group subsequently had their symptoms relieved by medroxy progesterone acetate; one of these women was formally diagnosed by ultrasound scan and magnetic resonance imaging as suffering from pelvic congestion. Women with chronic pelvic pain were more likely to have been sterilised compared with the pain free group. Six of the women in the index group (20%) had previously undergone a hysterectomy and bilateral salpingo oophorectomy, and two (7%) hysterectomy alone. The indication for surgery had been treatment of pelvic pain but there had been subsequently no apparent relief in pain symptoms.

Table 3.  Comparison of gynaecological histories and surgery for the three groups. Values are given as n (%). Key as for Table 1; PID = pelvic inflammatory disease; D&C = dilatation and curettage; TAH = total abdominal hysterectomy; BSO = bilateral salpingo-oophorectomy.
Previous gynaecological historyGroup 1Group 2Group 3
  1. *P leqslant R: less-than-or-eq, slant 0.001; **P leqslant R: less-than-or-eq, slant 0.01; † P, leqslant R: less-than-or-eq, slant 0.001 (Fisher's exact test); †† P re;0.01 compared with Group 2 and P leqslant R: less-than-or-eq, slant 0.001 compared with Group 3; ‡ P leqslant R: less-than-or-eq, slant 0.05 compared with Group 3; § P, leqslant R: less-than-or-eq, slant 0.05; ¶ P leqslant R: less-than-or-eq, slant 0.001 compared with Group 3 (Pearson values).

Pathology
 PID10 (33)*1 (3)1 (3)
 Ovarian cyst8 (27)**3(10)3(10)
 Endometriosis5 (17)†0 (0)0 (0)
 Adhesions3 (10)1 (3)0 (0)
 Fibroids2(7)1 (3)0(0)
 Other problems8 (27)3 (10)7 (23)
Surgery
 D&C16 (53)††6 (20)3 (10)
 Sterilisation7 (23)‡4 (13)1 (3)
 TAH & BSO6 (20)§0 (0)1 (3)
 Hysterectomy2 (7)0 (0)1 (3)
 Caeserean section4 (13)4 (13)3(10)
 Ectopic pregnancy1 (3)1 (3)0(0)
 Other gynae surgery8 (27)¶5 (17)1 (3)

Details of sexual history

Table 4, gives details about the age at which the women first had sexual intercourse and the number of sexual partners. No statistically significant differences were found between the groups for either of these two variables. Three of the women in the comparison pain group and one woman in the pain-free group had never had sexual intercourse. All the index group had had intercourse.

Table 4.  Comparison of, sexual histories for the three groups. Values are, given as mean [SD] or n (%). Key as for Table 1.
Sexual historyGroup 1Group 2Group 3
  1. *P leqslant R: less-than-or-eq, slant 0.001; **P leqslant R: less-than-or-eq, slant 0.01; § P leqslant R: less-than-or-eq, slant 0.05 (χ2 test)

Mean age at first sexual intercourse17 [2.47]18 [2.01]17 [2.56]
No. of partners
 None0 (0)3 (10)1 (3)
 15 (17)11 (37)7 (23)
 2-522 (73)12 (40)18 (60)
 6-92 (7)3 (10)2 (7)
 leqslant R: less-than-or-eq, slant101 (3)0 (0)1 (3)
Sexual problems
 Dyspareunia22 (73)*8(27)7(23)
 Loss of interest in sex19 (63)*11 (37)9 (30)
 Post-coital pain17 (53)**4 (13)2 (7)
 Vaginismus12 (40)§4 (13)3 (10)
 Other1 (3)2 (7)3 (10)

The majority of women were currently involved in a sexual relationship. There was a much higher prevalence of current sexual problems among the women with pelvic pain compared with either of the other two groups (Table 4)., Correspondingly, these women's ratings of their current sexual relationship were much poorer than the ratings given by the women in the two comparison groups. Only eight women (32%), in the index group rated their sexual relationship as good or very good, compared with 16 (66%) in the comparison pain group and 20 (87%) in the pain-free group (P, = 0.01). The majority of women, however, in each of the three groups rated their general relationship as good or very good (index group 21 (84%); comparison pain group 21 (87%); pain-free group 22 (96%)).

The sexual abuse histories for the three groups are shown in Table 5. Twelve women with chronic pelvic pain (40%) reported a total of 28 different types of sexual abuse as described in Table 5. This figure compares with five women (1 7%) from the comparison pain group reporting 15 different types of abusive experiences, and five women (1 7%) from the pain-free group reporting 10 different types of experience.

Table 5.  Comparison of sexual abuse histories for the three groups. Values are given as n (%). Key as for Table 1.
 Group 1Group 2Group 3
  1. Pleqslant R: less-than-or-eq, slant 0.05 compared with *Groups 1 & 2; †Group 2; ‡Group 3. §e.g. violent sex.

No. of women abused12 (40)*5 (17)5 (17)
Age at which abuse began (years)
 <166 (20)2 (7)4 (13)
 leqslant R: less-than-or-eq, slant166 (20)3 (10)1 (3)
No. of incidents
 16 (20)4 (13)0 (0)
 >16 (20)1 (3)5 (17)
Intercourse8 (27)†2 (7)4 (13)
Touching genitals6 (20)3 (10)2 (7)
Showing genitals4 (13)4 (13)2 (7)
Fondled sexually4 (13)3 (10)1 (3)
Oral contact1 (3)1 (3)1 (3)
Other experiences§5 (17)‡2 (7)0 (0)

The incidence of severe sexual abuse (i.e. that involving sexual intercourse) was much higher in the index group (n = 8) compared with the comparison pain group (n= 2). Four of the women in the pain-free group had experienced severe sexual abuse. Five women in the index group reported sexual experiences associated with violence compared with none of the women in the pain-free group. The majority (80%) of abused women affected in each group stated that they had found the experience very or extremely upsetting at the time and that the experience had left lasting effects on their lives. The median age at which the first experience of sexual abuse occurred was higher for the index group compared with the other two groups (index group 16 years, range 10-29 years; comparison pain group 9 years, range 4-18 years; pain free group 13 years, range 5-17 years).

Physical violence

Physical violence was reported by women in all three groups, but there was no statistical difference in the incidence of this between the three groups (index group 8 (27%); comparison pain group 6 (20%); pain free group 8 (27%))

Physical and psychological symptoms

The women with pelvic pain had approached their general practitioner with a greater number of other symptoms unrelated to pelvic pain, than each of the two comparison groups. A net somatisation score was obtained for each woman on the basis of the total number of items endorsed on the somatisation scale. The data obtained from measuring the number of symptoms experienced was not normally distributed. The median number of symptoms reported was 7.5 for the index group (range 1-23). Women in the comparison pain group scored a median of 4, (range 0-23), and the women in the pain free group scored a median of 3 (range 0-9). This difference between the index group and each of the two other groups was highly significant (Kruskall-Wallis: P < 0.001).

Table 6 shows the Hospital Anxiety and Depression scale scores. There was a significant difference between the median anxiety scores for each group (Kruskall Wallis: P < 0.05). Women in the index group were significantly more likely to experience a clinical level of anxiety (P < 0.01), than women in the pain-free group but not compared with women in the comparison pain group. The incidence of clinical depression was higher in the index group compared with both of the comparison groups, although this was not significant.

Table 6.  Comparison of Hospital Anxiety and Depression scale (HAD) scores for the three groups. Values are given as n (%). Key as for Table 1.
HAD scoreGroup 1Group 2Group 3
  1. *P, leqslant R: less-than-or-eq, slant 0.01 compared with Group 3.

Anxiety
 0-7 (not clinically significant)6 (20)13 (43)17 (57)
 8-10 (borderline ‘anxiety’)7 (23)6 (20)6 (20)
 11-21 (clinically significant)17 (57)*11 (37)7 (23)
Depression
 0-7 (not clinically significant)16 (53)19 (63)24 (80)
 8-10 (borderline ‘depression’)7 (23)7 (23)2 (7)
 11-21 (clinically significant)7 (23)4 (13)4 (13)

The women in the chronic pelvic pain group without any identified pathology (n= 9) were compared with the remaining women in the group who had identified pathology (n = 21). These two groups were very closely matched in their sexual abuse histories and their anxiety and depression scores. The prevalence of current sexual problems was also very similar between the two groups for all problems except dyspareunia which was more common in the women with identified pathology (81%vs, 56%). These women also endorsed proportionally more symptoms on the somatisation scale (median 10) when compared with the women with chronic pelvic pain without obvious pathology (median 5). The women in the chronic pelvic pain group who still had a uterus (n= 21) were compared with the remaining nine women who did not have a uterus to determine whether the absence of a uterus influenced symptom reporting (Table 7). The two groups were very closely matched in their histories of sexual abuse.

Table 7.  Comparison, of women with and without a uterus, in the index group, (Group 1). Values are given as n (%) or median {range} unless otherwise indicated. HAD, = Hospital Anxiety and Depression scale.
 With uterus (n = 21)Without uterus (n = 9)
  1. *Clinically significant.

Somatisation score10 (1-23)4 {1–20}
Sexual problems  
 Dyspareunia13 (45)6(67)
 Loss of interest8 (28)4 (44)
 Post-coital pain13 (45)4 (45)
 Vaginismus12 (41)7(78)
HAD scores
Anxiety 11-21*10 (48)6(67)
  Median score1011
Depression 11-21*5 (24)2(22)
  Median score69

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

This study is the first conducted on a British population, comparing the incidence of previous sexual abuse in women with chronic pelvic pain, chronic pain in a different site and no pain. The results confirm the hypothesis that there is a significantly higher incidence of previous sexual abuse in women with chronic pelvic pain compared with women in either of the other two groups. The abuse histories were complex and the abuse often severe, with eight of the women with chronic pelvic pain reporting abuse involving sexual intercourse.

Seven previous controlled studies have examined the prevalence of sexual abuse in women with chronic pelvic pain1,5,7,8,14–16. However, only two, both from North America, have incorporated women with chronic pain in a site other than the pelvis and women with no chronic pain complaints as comparison groups8,16. Rapkin et al.8 studied 31 women with chronic pelvic pain, 142 women with chronic nonpelvic pain and 32 women without chronic pain. No significant differences were found between the groups for childhood or adult sexual abuse. However, a significantly higher prevalence of physical abuse was found in the chronic pelvic pain group compared with either of the other two groups. Walling et al.8 studied 64 women with chronic pelvic pain, 42, women with chronic headache and 46 pain-free women and found that women with chronic pelvic pain had a higher lifetime prevalence of major sexual abuse, involving penetration or other genital or anal contact (53%), compared with the headache group (33%), and the pain-free group (28%). A, higher incidence of physical abuse was reported in both chronic pain groups compared with the pain-free group.

Fry et al.10 studied a British sample of 164 women referred to a tertiary pelvic pain clinic and found a reported incidence of combined childhood and adult sexual abuse of between 38%, and 48%, depending on the definition used. However, no control group data was reported. Studies investigating the incidence of sexual abuse have used different methods to elicit information about abuse (e.g. telephone interviews8, postal questionnaires and face to face interviews). The type of method used may influence the results. Wyatt and Peters17 found that inverted-funnel questioning (i.e. asking questions in a pattern proceeding from most severe abuse to least severe abuse and then excluding any abuse reported in a previous question from successive questions) yielded higher prevalence rates than self report questionnaires. Caution therefore needs to be applied when making comparisons between studies which have relied upon different methods of obtaining information about sexual abuse. A, further problem in this area of research is that definitions of sexual abuse also vary and some studies have excluded isolated incidents of abuse16.

Women in the pelvic pain group were polysymptomatic compared with the other two groups. Somatisation disorder in adult women has been shown to be associated with sexual abuse in childhood18 and the combination of sexual abuse and chronic pain may result in high levels of somatic preoccupation19,20. Twelve month follow up data from the outpatient pelvic pain clinic has revealed that 69% of women continue to present to their GP with pain complaints but in different parts of their body (unpublished data).

One possible explanation for these findings is that women who somatise (i.e. experience and communicate somatic distress and symptoms with no clear underlying pathology) are more likely to exaggerate or fabricate histories of sexual trauma. Morrison21, however, concluded that there was no objective basis for this and, that the childhood sexual abuse histories in his sample of 60 women with somatisation disorder were characteristically consistent, detailed and verifiable.

This study supports evidence from other countries that there may be an association between sexual abuse and chronic pelvic pain, but this does not imply causality. However, we suggest that an awareness of this possible association is important in the assessment and management of the polysymptomatic patient with chronic pelvic pain.

Acknowledgements

The authors would like to thank Ms A., Hull for her secretarial support.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. References