A 10-yr Analysis of Chronic Pelvic Pain and Chronic Opioid Therapy in the Women Veteran Population

A 10-yr Analysis of Chronic Pelvic Pain and Chronic Opioid Therapy in the Women Veteran Population Abstract Introduction Chronic pelvic pain (CPP) affects an estimated 30% of women Veterans. Previous research shows high rates of narcotic abuse in the women Veteran population. Narcotics are not recommended for the treatment of CPP. Understanding how CPP impacts narcotic prescribing in the women Veteran population is critical to addressing the public health crisis of opioid abuse. Our objective was to compare chronic opioid therapy (COT) prescribed 5 yr prior to and following CPP diagnosis and to identify predictors of COT as well as adverse events associated with COT. We choose to look at 10 yr of data because we thought this time period would provide unique insight into the longitudinal associations of CPP and COT and was available in the database. Materials and Methods Women with non-cancer CPP were included for analyses from the Veteran’s Affairs Corporate Database Warehouse. COT was defined as 90 d of opiates/calendar year for each of the 5 yr proceeding and following the diagnosis of CPP. Patient characteristics and potential variables influencing COT were collected. We compared baseline demographics between the women who received COT to the women who did not receive COT to find additional demographic predictors of COT in association with CPP. Multivariable analysis identified predictors of COT in this population of women with CPP. We utilized an interrupted time series analysis to understand the impact of the diagnosis of CPP on COT. Results A total of 49,601 women met inclusion criteria with an average age of 40.1 ± 11.5 yr; 37.3% self-characterized as being a racial minority and 24% had a history of military sexual trauma. Chronic use increased significantly (p < 0.001) in the 5 yr preceding the diagnosis of CPP from 6.3% (n = 3124) of women at time −5 to 13.6% (n = 6746) at time 0. In the first year following the diagnosis of CPP, 16.8% (n = 8,333) of women with CPP met the criteria for COT (p < 0.001) and 15% (n = 7440) of women with CPP remained in the COT group for the remaining 5 yr following the diagnosis. On average women in the COT group had 250–292 d of opioids/year. When comparing women who received chronic narcotics following the diagnosis of CPP versus those who did not receive chronic narcotics, women who received COT were older, more likely to smoke and more frequently diagnosed with other pain conditions such as back pain, headaches, and fibromyalgia. (All p < 0.001). In the multivariable model, predictors of COT following CPP diagnosis included prior COT (OR = 10.0 (95% CI 9.4, 10.6), a positive history of military sexual trauma, smoking, and other chronic pain conditions. Conclusions The distinct pattern of prescribing shown in this cohort may mean COT is prescribed for CPP and this prescribing pattern contributes to the adverse events associated with COT. As COT is not recommended for CPP, physicians need more education on the therapies available to help CPP patients. INTRODUCTION In 2014, the National Institutes of Health convened a workshop to address the “silent epidemic” of chronic pain and concomitant opioid use.1 Decreasing the use and abuse of narcotic medications is a national priority adopted by the Veterans Health Administration (VA).1,2 Veterans are twice as likely to die from accidental overdose compared with non-Veterans.3 Nevertheless, physicians continue to prescribe opioids to treat debilitating pain without proven efficacy, and with significant risk.4 The narcotic epidemic’s startling cost was $737 million 2008 U.S. dollars5 and is presently responsible for more deaths than heroin and cocaine combined.6 Recent prescribing data show U.S. physicians are more likely to prescribe opioids to women as compared with men.7 Seal et al8 retrospectively reviewed a cohort of Iraq and Afghanistan veterans with non-cancer related pain and found that both mental health diagnoses and female sex were associated with increased risk of a minimum 20-d opioid prescription. It is unclear why women receive more opioids than men but one chronic pain condition that disproportionately affects women is pelvic pain. Chronic pelvic pain (CPP) is defined as persistent pain of 6 wk to 6 mo duration, which arises from the pelvis and causes dysfunction.9 Ten percent of women referred to a gynecologist come for CPP.10 The estimated prevalence of CPP in the USA is 38 per 1,000 women aged 15-73.11 In studies from around the world, CPP population prevalence varies between 5.7% and 26.6%.12 Women Veterans disproportionately suffer from CPP at an estimated prevalence of 30%.13 Based on expert opinion, narcotics are not recommended as long-term, primary treatment for CPP.14,15 A systematic review evaluated over 60 risk factors in 122 studies for CPP and found that the strongest associations were between CPP and the “presence of pelvic pathology, history of abuse and coexistent psychological morbidity.”16 This association of CPP, sexual abuse and psychological morbidity could explain the increased risk of CPP in women Veterans given the high prevalence of military sexual trauma (MST) which predisposes to depression, anxiety, and post-traumatic stress disorder. Thirty-two percent of women Veterans report surviving MST a type abuse that includes both “sexual assault or repeated, threatening sexual harassment”.2,17 Understanding how the diagnosis of CPP impacts narcotic prescribing in the setting of psychiatric disorders and MST is critical to addressing the public health crisis of opioid use. The link between a diagnosis of CPP and the frequency and amount of narcotic prescriptions has not been examined in a longitudinal fashion. Additionally, the risk of chronic opioid therapy (COT) for non-cancer pelvic pain has not been well described. Our objective was to compare COT and narcotic doses prescribed 5 yr prior and following a non-cancer related CPP diagnosis in women veterans. We also anticipated finding distinct patient characteristics that could help predict adverse COT outcomes in women with CPP such a MST based on prior research. We also aimed to look at adverse outcomes related to COT in a population with CPP. We hypothesized that COT use would increase following the diagnosis of CPP. MATERIALS AND METHODS This was a large-scale, retrospective study using clinical records contained in the Veterans’ Health Administration Corporate Data Warehouse which houses all information for Veterans receiving care at a Veteran’s Administration Health Care System since 1997. After obtaining IRB approval (15-H175), with a waiver of HIPAA authorization, data was requested from the Veterans Informatics and Computing Infrastructure (VINCI). Women who declared a single gender, had established care with a primary care physician and who had CPP first diagnosed between 2002 and 2012 by ICD-9 codes comprised the initial cohort. The CPP definition included both non-specific codes as well as specific causes of CPP (such as dysmenorrhea, interstitial cystitis) (see Appendix A). Women were excluded from the study if they had a history of any abdominal or pelvic malignancy. We also excluded women whose CPP seemed related to adnexal torsion, appendicitis or diverticulitis, based on ICD-9 codes. The database also identified all women filling prescriptions at the VA by outpatient VA pharmacy data. To remain in the study cohort, a woman had to be filling prescriptions at the VA for at least a single time point 5 yr before and a single time point 5 yr after the diagnosis of CPP. We used these criteria as we wanted to demonstrate that in some capacity the VA was continuing to provide care for these women. We decided if there were zero prescriptions filled at the VA over 5 yr it was unlikely the VA was providing healthcare for that woman. We choose to look at 10 yr of data per patient both because this was available within the Corporate Data Warehouse and because we thought this would show longer patterns associated between CPP and COT. COT for pain was defined by Veteran’s Health Administration Directive 1005 as “the use of opioids on a daily or intermittent basis for 90 or more calendar days.” In order to identify patients filling narcotic prescriptions we identified all drugs in the VA CN101 drug class (the code for narcotics). We then selected only the oral preparations, excluding transdermal, intranasal and injectables and converted the narcotics into morphine equivalents.18 We excluded non-oral preparations because typically, those preparations treat acute pain or cancer related pain. Narcotic prescriptions and days supply counts were used to identify 90 d of narcotics/calendar year. Age at the time of CPP was calculated. Body mass index (BMI) was calculated as the average BMI over the study period. Because there were multiple responses to the questions of race/ethnicity, MST and smoking over time a specific hierarchy was set for each. Smoking was classified as current, former or never. “Never” smokers were only those patients who always denied smoking. If, at any point, a patient identified as a former smoker but never as a current smoker then she was placed in the former group, and if a patient – identified as a current smoker, they were placed in the current group. Based on self-report patients were broadly grouped into five racial categories (African-American, American Indian, Asian, White, and Unknown) and two ethnicity categories (Hispanic versus non-Hispanic). A patient was considered a member of a minority if she identified herself as African-American, American Indian, or Asian. If a patient ever responded “yes” to a history of MST then the answer was recorded as “yes,” even if subsequent answers were “no” or “declined to answer.” Because these data come from Veterans in the Veteran’s Health Administration data MST (which occurred during military service) preceded the diagnosis of CPP (which was recorded post military service). For this study “declined to answer” and “not asked” responses for MST were grouped together. ICD-9 codes were also used to identify patients’ other chronic non-specific pain conditions such as back pain not otherwise specified, fibromyalgia, headaches, irritable bowel syndrome, and chronic joint pain. If the first date of diagnosis came at any time during the 5 yr prior to the CPP diagnosis date, then the pain condition was considered present at baseline. These other chronic pain conditions were only evaluated at this time point. Adverse outcomes potentially associated with narcotic use included drug abuse, drug overdose, depression, anxiety, and suicide. Adverse events were recorded when emergency room visits or inpatient admissions were associated with these diagnoses. We chose to only include these types of visit encounters as they likely represent the most direct need for immediate intervention rather than the outpatient clinical visits. Emergency services and inpatient admission are also associated with a higher level of care rather than scheduled primary care encounters. Data were analyzed using SAS version 9.4. Cohort demographics were described. Demographic statistics between those who received COT versus those who did not were reported and compared using x2 tests for categorical variables. Student’s t-tests were used for continuous variables, except that Wilcoxon Rank-Sum test was used for non-normally distributed data. The primary outcome was COT, a binary determination for each year of the analysis preceding and following the diagnosis of CPP. Time 0 was identified as the first time of CPP diagnosis. Data were recorded for the 5 yr preceding (−5 to −1) and after (0 to +5). Calculated years before and after were unique and relative to each individual’s CPP diagnosis date. We utilized an interrupted time series approach using a logistic regression method for repeated measures, which were determinations of COT for the 10 yr per veteran (PROC GENMOD in SAS 9.4). This logistic regression considered time as a continuous variable and tested whether the percentage of veterans with COT as a function of time increased after the diagnosis of CPP. This model allowed a discontinuity in COT at year 0 in addition to separate before and after slopes; the presence of this discontinuity tested whether there was extra narcotic usage in the first year following CPP diagnosis. Adverse outcomes: drug abuse, drug overdose, depression, anxiety, and suicide were compared before and after CPP diagnosis. All adverse outcomes covariates were examined for collinearity. Best models were obtained using a backward selection procedure. In these models, each of the adverse outcomes was considered as a complication of COT and was set as the dependent variable with COT as an independent variable. This same method was used to evaluate demographics that increased the likelihood of COT including multivariate modeling with a backward selection procedure. Predictors of COT and adverse outcomes related to COT are reported as odds ratios (OR) with 95% confidence intervals (CI). Significance was set as p < 0.05. RESULTS We identified 596,815 women Veterans who sought care with a VA primary care physician. Of those, 95,860 (16%) were diagnosed with CPP. We excluded 23,718 women as they were not diagnosed between 2002 and 2012 and an additional 11,779 were excluded as they did not fill prescriptions at the VA or seek primary care either during 5 yr before and/or the 5 yr following receipt of their CPP diagnosis. After excluding 10,762 for history of abdominal/pelvic cancers, appendicitis, diverticulitis, or adnexal torsion, there remained 49, 601 women in the study cohort (see Fig. 1). FIGURE 1. View largeDownload slide Diagram of cohort identification. FIGURE 1. View largeDownload slide Diagram of cohort identification. At the time of CPP diagnosis, the average age of women in this cohort was 40.3 (SD ± 11.7) yr, the population was overweight with BMI 29.4 (SD ± 6.0) and most women smoked (51.1%, n = 25,330). Over a third (37.2%, n = 18,465) self- identified as being a minority. Thirty-four percent (n = 16,890) had experienced MST and 30% (n = 15,780) had at least one other chronic pain disorder at the time of CPP diagnosis. The most common co-occurring pain disorders were headaches (15.3%, n = 7,597), back pain (7.9%, n = 3,899), and fibromyalgia (8.2%, n = 4,085). Chronic narcotic (3 or more 30-d prescriptions) use increased significantly (p < 0.001) in the 5 yr preceding the diagnosis of CPP from 6.3% (n = 3124) of women at time −5 to 13.6% (n = 6746) at time 0. In the first year following the diagnosis of CPP, 16.8% (n = 8,333) of women with CPP met the criteria for COT (p < 0.001) and 15% (n = 7440) of women with CPP remained in the COT group for the remaining 5 yr following the diagnosis (see Fig. 2). On average women in the COT group had 250–292 d of opioids/year. FIGURE 2. View largeDownload slide Interrupted time series of percent of patients using COT relative to the time of CPP diagnosis. Percentage COT before CPP diagnosis date (solid blue regression line) increases from 6.3% to 13.6% (p < 0.001) There is an immediate step discontinuity in COT to 16.8% (light blue solid circle) in the year following the CPP diagnosis date (p < 0.001). The regression line after CPP diagnosis (red solid line) remains high (15.8% to 15.2%), significantly higher than the COT usage before CPP diagnosis (p < 0.001). FIGURE 2. View largeDownload slide Interrupted time series of percent of patients using COT relative to the time of CPP diagnosis. Percentage COT before CPP diagnosis date (solid blue regression line) increases from 6.3% to 13.6% (p < 0.001) There is an immediate step discontinuity in COT to 16.8% (light blue solid circle) in the year following the CPP diagnosis date (p < 0.001). The regression line after CPP diagnosis (red solid line) remains high (15.8% to 15.2%), significantly higher than the COT usage before CPP diagnosis (p < 0.001). When comparing women who received chronic narcotics following the diagnosis of CPP versus those who did not receive chronic narcotics, women who received COT were older, more likely to smoke and more frequently diagnosed with other pain conditions such as back pain, headaches and fibromyalgia (all p < 0.001; see Table I). Table I. Demographics for CPP Cohort Comparing Women Veterans Who Received Chronic Opioids Following CPP Diagnosis to Women Veterans Who Did Not Receive Chronic Opioidsa Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  aData presented as N (percentage) unless otherwise noted. Percentages may not add up to 100 due to rounding. Table I. Demographics for CPP Cohort Comparing Women Veterans Who Received Chronic Opioids Following CPP Diagnosis to Women Veterans Who Did Not Receive Chronic Opioidsa Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  aData presented as N (percentage) unless otherwise noted. Percentages may not add up to 100 due to rounding. In the multivariable logistic regression, there were several baseline characteristics which predicted subsequent COT following CPP diagnosis. The strongest predictor was prior COT (OR = 10.0 (95% CI 9.4, 10.6)). Other significant predictors in the model included a positive history of MST, smoking and other chronic pain conditions such as irritable bowel syndrome, fibromyalgia, back pain, and headache. Interestingly, the regression model also identified that as the time of CPP moved forward women were less likely to be prescribed chronic opioids (Table II). Table II. Patient Characteristics in Multivariable Analysis That Increased the Odds of Receiving Chronic Narcotics After the Diagnosis of CPP Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  Table II. Patient Characteristics in Multivariable Analysis That Increased the Odds of Receiving Chronic Narcotics After the Diagnosis of CPP Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  When evaluating adverse outcomes in women with CPP and COT compared with women with CPP without COT, all outcomes evaluated were significantly associated with COT including suicide, drug abuse, drug overdose and admissions for mental health comorbidities (Table III). Table III. Risk of Adverse Outcomes in CPP Patients Using Chronic Narcotics Versus CPP Patients Not Using Chronic Narcotics Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  Table III. Risk of Adverse Outcomes in CPP Patients Using Chronic Narcotics Versus CPP Patients Not Using Chronic Narcotics Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  DISCUSSION We found that women Veterans with CPP have increasing COT in the 5 yr prior to the formal diagnosis of CPP. Following CPP diagnosis, COT immediately increases and continues for at least 5 yr. The highest risk factor for ongoing COT following CPP diagnosis was a prior history of COT. Women diagnosed with CPP who also had a history of MST, headaches, back pain, IBS, smoking, and fibromyalgia were at increased risk of developing COT. These other pain conditions likely contribute the COT prescribing however there remains an independent, significant and distinct pattern of COT associated with the diagnosis of CPP. Previous work found that Veterans with COT of greater than 20 mg/d morphine equivalents had increased risk of suicide compared with those that received less than 20 mg/d.3 In our study population, women with CPP who received COT had approximately 2–4 fold increased risk of admissions and ER visits for drug abuse, drug overdose, anxiety, depression, PTSD, and suicide compared with women with CPP not prescribed COT. This study is consistent with general work among Veterans during fiscal years 2004–2012, during which time the prevalence of opioid prescriptions increased from 18.9% to 33.4%, an increase of 76.7%. The groups with the highest prevalence of opioid use were women and young adults (i.e., 18–34 yr old).19 Our work seems to indicate that CPP is associated with COT and in a significant proportion of women, antedates development of COT. The pattern of use before and after CPP diagnosis might indicate that physicians prescribe opiates gradually but then following a formal chronic pain diagnosis become more comfortable with long-term narcotic use (as indicated by the high stable percentage of Veterans prescribed COT after CPP diagnosis for 5 yr). This study has limitations. The population was limited to only women Veterans with 10 yr of data in the CDW database, the data are limited by the accuracy of those who entered it and there may be women treated for CPP that were not coded as having CPP. The conclusions may not be valid for non-Veterans, younger veterans (without 10 yr of data in the CDW) or Veterans with other insurance programs. While we excluded women with abdominal or pelvic cancers there may some women with other cancer types being treated with COT that we included in the CPP cohort. Baseline pain scores as they specifically related to the diagnosis of CPP were also unavailable and this information could help better classify who receives COT. Additionally, there were unmeasured factors related to COT that were not captured in the database including patients’ social situations or medications or other healthcare obtained outside of the VA system. Patients with COT are known to seek multiple providers, visit emergency rooms, and purchase street-drugs in an effort to obtain additional narcotic medications.20 Our data were limited to the narcotics filled at the VA pharmacies. We believe that these issues potentially result in under-representation of the extent of COT in our database. While COT has been looked at in Veteran population it has not been specifically explored in relationship to CPP, a condition that disproportionately affects women Veterans. CPP appears to contribute to the overuse of narcotics in the women Veteran population. A significant strength in this study is that we used prescription fill data (indicating that the patient did pick-up these medications) rather than prescribed data and believe this is important because these opioids were released into the patient’s care Further research could compare prescribed versus filled opioid prescriptions and also investigate actual usage of these prescriptions. Additionally, we were able to evaluate COT over an extended period, which demonstrated the distinct longitudinal association between CPP and COT. This research suggests that COT is a common treatment for CPP in the women Veteran population. It also adds to the growing literature showing harm associated with COT. Given the harms our patients deserve more research on therapeutic alternatives outside of chronic narcotics for the treatment of pelvic pain. The American College of Obstetricians and Gynecologists online patient education site for CPP lists non-narcotic therapies for CPP treatment and provides a useful resource for our patients.21 Interventions that reduce COT prescribing through the education of primary care physicians and mid-level providers should also be explored. CONCLUSION In conclusion, CPP antedates a sustained increase in COT. COT is associated with risks of ER and inpatient visits related to drug abuse and psychiatric illnesses. While this work cannot prove causality between COT and CPP the pattern associated is concerning and suggests that CPP is likely associated with COT. Physicians should be judicious in their prescribing of narcotics for CPP and inform patients of the potential risks. Supplementary Material Supplementary material is available at Military Medicine online. REFERENCES 1 Reuben DB, Alvanzo AA, Ashikaga T, et al.  : National Institutes of Health Pathways to Prevention Workshop: the role of opioids in the treatment of chronic pain. Ann Intern Med  2015; 162( 4): 295– 300. Google Scholar CrossRef Search ADS PubMed  2 Klingensmith K, Tsai J, Mota N, Southwick SM, Pietrzak RH: Military sexual trauma in US Veterans: results from the National Health and Resilience in Veterans Study. J Clin Psychiatry  2014; 75( 10): 1133– 9. Google Scholar CrossRef Search ADS   3 Bohnert AS, Ilgen MA, Galea S, McCarthy JF, Blow FC: Accidental poisoning mortality among patients in the Department of Veterans Affairs Health System. Med Care  2011; 49( 4): 393– 6. Google Scholar CrossRef Search ADS PubMed  4 Ray WA, Chung CP, Murray KT, Hall K, Stein CM: Prescription of long-acting opioids and mortality in patients with chronic noncancer pain. JAMA  2016; 315( 22): 2415– 23. Google Scholar CrossRef Search ADS PubMed  5 White AM, Hingson RW, Pan IJ, Yi HY: Hospitalizations for alcohol and drug overdoses in young adults ages 18---24 in the United States, 1999--2008: results from the Nationwide Inpatient Sample. J Stud Alcohol Drugs  2011; 72( 5): 774– 86. Google Scholar CrossRef Search ADS PubMed  6 Manchikanti L, Helm S 2nd, Fellows B, et al.  : Opioid epidemic in the United States. Pain Physician  2012; 15( 3): ES9– 38. Google Scholar PubMed  7 Paulozzi LJ1, Strickler GK, Kreiner PW, Koris CM: Controlled Substance Prescribing Patterns – prescription behavior surveillance system, eight states, 2013.Centers for Disease Control and Prevention (CDC). MMWR Surveill Summ  2015; 64( 9): 1– 14. Google Scholar CrossRef Search ADS PubMed  8 Seal KH1, Shi Y, Cohen G, et al.  : Association of mental health disorders with prescription opioids and high-risk opioid use in US veterans of Iraq and Afghanistan. JAMA  2012; 307( 9): 940– 7. Google Scholar CrossRef Search ADS PubMed  9 Doggweiler R, Whitmore KE, Meijlink JM, et al.  : A standard for terminology in chronic pelvic pain syndromes: a report from the chronic pelvic pain working group of the international continence society. Neurourol Urodyn  2016; 36: 984– 1008. Google Scholar CrossRef Search ADS PubMed  10 Reiter RC: A profile of women with chronic pelvic pain. Clin Obstet Gynecol  1990; 33( 1): 130. Google Scholar CrossRef Search ADS PubMed  11 Zondervan KT, Yudkin PL, Vessey MP, Dawes MG, Barlow DH, Kennedy SH: Prevalence and incidence of chronic pelvic pain in primary care: evidence from a national general practice database. Br J Obstet Gynaecol  1999; 106: 1149– 55. Google Scholar CrossRef Search ADS PubMed  12 Ahangari A: Prevalence of chronic pelvic pain among women: an updated review. Pain Physician  2014; 17( 2): E141– 7. Google Scholar PubMed  13 Cichowski SB, Rogers RG, Clark EA, Murata E, Murata A, Murata G: Military sexual trauma in american women veterans is associated with chronic pain conditions. Mil Med  2017; 182( 9): e1895– 9. Google Scholar CrossRef Search ADS PubMed  14 Siedentopf F, Weijenborg P, Engman M, et al.  : ISPOG European Consensus Statement – chronic pelvic pain in women (short version). J Psychosom Obstet Gynaecol  2015; 36( 4): 161– 70. Google Scholar CrossRef Search ADS PubMed  15 Steele A: Opioid use and depression in chronic pelvic pain. Obstet Gynecol Clin North Am  2014; 41( 3): 491– 501. Google Scholar CrossRef Search ADS PubMed  16 Latthe P, Mignini L, Gray R, Hills R, Khan K: Factors predisposing women to chronic pelvic pain: systematic review. BMJ  2006; 332( 7544): 749– 55. Google Scholar CrossRef Search ADS PubMed  17 Military Sexual Trauma. Available at https://www.mentalhealth.va.gov/msthome.asp; accessed March 31, 2018. 18 Centers for Medicare and Medicaid Services. Available at https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/Downloads/Opioid-Morphine-EQ-Conversion-Factors-Aug-2017.pdf; accessed May 8, 2018. 19 Mosher HJ, Krebs EE, Carrel M, Kaboli PJ, Vander Weg MW, Lund BC: Trends in prevalent and incident opioid receipt: an observational study in Veterans Health Administration 2004–2012. J Gen Intern Med  2014; 30( 5): 597– 604. Google Scholar CrossRef Search ADS PubMed  20 Hall AJ, Logan JE, Toblin RL, et al.  : Patterns of abuse among unintentional pharmaceutical overdose fatalities. JAMA  2008; 300( 22): 2613– 20. Google Scholar CrossRef Search ADS PubMed  21 American College of Obstetricians and Gynecologists Patient Frequently asked fact sheet. Available at http://www.acog.org/Patients/FAQs/Chronic-Pelvic-Pain; accessed May 8, 2018. Published by Oxford University Press on behalf of Association of Military Surgeons of the United States 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Military Medicine Oxford University Press

A 10-yr Analysis of Chronic Pelvic Pain and Chronic Opioid Therapy in the Women Veteran Population

Loading next page...
 
/lp/ou_press/a-10-yr-analysis-of-chronic-pelvic-pain-and-chronic-opioid-therapy-in-56BhquDQFx
Publisher
Oxford University Press
Copyright
Published by Oxford University Press on behalf of Association of Military Surgeons of the United States 2018.
ISSN
0026-4075
eISSN
1930-613X
D.O.I.
10.1093/milmed/usy114
Publisher site
See Article on Publisher Site

Abstract

Abstract Introduction Chronic pelvic pain (CPP) affects an estimated 30% of women Veterans. Previous research shows high rates of narcotic abuse in the women Veteran population. Narcotics are not recommended for the treatment of CPP. Understanding how CPP impacts narcotic prescribing in the women Veteran population is critical to addressing the public health crisis of opioid abuse. Our objective was to compare chronic opioid therapy (COT) prescribed 5 yr prior to and following CPP diagnosis and to identify predictors of COT as well as adverse events associated with COT. We choose to look at 10 yr of data because we thought this time period would provide unique insight into the longitudinal associations of CPP and COT and was available in the database. Materials and Methods Women with non-cancer CPP were included for analyses from the Veteran’s Affairs Corporate Database Warehouse. COT was defined as 90 d of opiates/calendar year for each of the 5 yr proceeding and following the diagnosis of CPP. Patient characteristics and potential variables influencing COT were collected. We compared baseline demographics between the women who received COT to the women who did not receive COT to find additional demographic predictors of COT in association with CPP. Multivariable analysis identified predictors of COT in this population of women with CPP. We utilized an interrupted time series analysis to understand the impact of the diagnosis of CPP on COT. Results A total of 49,601 women met inclusion criteria with an average age of 40.1 ± 11.5 yr; 37.3% self-characterized as being a racial minority and 24% had a history of military sexual trauma. Chronic use increased significantly (p < 0.001) in the 5 yr preceding the diagnosis of CPP from 6.3% (n = 3124) of women at time −5 to 13.6% (n = 6746) at time 0. In the first year following the diagnosis of CPP, 16.8% (n = 8,333) of women with CPP met the criteria for COT (p < 0.001) and 15% (n = 7440) of women with CPP remained in the COT group for the remaining 5 yr following the diagnosis. On average women in the COT group had 250–292 d of opioids/year. When comparing women who received chronic narcotics following the diagnosis of CPP versus those who did not receive chronic narcotics, women who received COT were older, more likely to smoke and more frequently diagnosed with other pain conditions such as back pain, headaches, and fibromyalgia. (All p < 0.001). In the multivariable model, predictors of COT following CPP diagnosis included prior COT (OR = 10.0 (95% CI 9.4, 10.6), a positive history of military sexual trauma, smoking, and other chronic pain conditions. Conclusions The distinct pattern of prescribing shown in this cohort may mean COT is prescribed for CPP and this prescribing pattern contributes to the adverse events associated with COT. As COT is not recommended for CPP, physicians need more education on the therapies available to help CPP patients. INTRODUCTION In 2014, the National Institutes of Health convened a workshop to address the “silent epidemic” of chronic pain and concomitant opioid use.1 Decreasing the use and abuse of narcotic medications is a national priority adopted by the Veterans Health Administration (VA).1,2 Veterans are twice as likely to die from accidental overdose compared with non-Veterans.3 Nevertheless, physicians continue to prescribe opioids to treat debilitating pain without proven efficacy, and with significant risk.4 The narcotic epidemic’s startling cost was $737 million 2008 U.S. dollars5 and is presently responsible for more deaths than heroin and cocaine combined.6 Recent prescribing data show U.S. physicians are more likely to prescribe opioids to women as compared with men.7 Seal et al8 retrospectively reviewed a cohort of Iraq and Afghanistan veterans with non-cancer related pain and found that both mental health diagnoses and female sex were associated with increased risk of a minimum 20-d opioid prescription. It is unclear why women receive more opioids than men but one chronic pain condition that disproportionately affects women is pelvic pain. Chronic pelvic pain (CPP) is defined as persistent pain of 6 wk to 6 mo duration, which arises from the pelvis and causes dysfunction.9 Ten percent of women referred to a gynecologist come for CPP.10 The estimated prevalence of CPP in the USA is 38 per 1,000 women aged 15-73.11 In studies from around the world, CPP population prevalence varies between 5.7% and 26.6%.12 Women Veterans disproportionately suffer from CPP at an estimated prevalence of 30%.13 Based on expert opinion, narcotics are not recommended as long-term, primary treatment for CPP.14,15 A systematic review evaluated over 60 risk factors in 122 studies for CPP and found that the strongest associations were between CPP and the “presence of pelvic pathology, history of abuse and coexistent psychological morbidity.”16 This association of CPP, sexual abuse and psychological morbidity could explain the increased risk of CPP in women Veterans given the high prevalence of military sexual trauma (MST) which predisposes to depression, anxiety, and post-traumatic stress disorder. Thirty-two percent of women Veterans report surviving MST a type abuse that includes both “sexual assault or repeated, threatening sexual harassment”.2,17 Understanding how the diagnosis of CPP impacts narcotic prescribing in the setting of psychiatric disorders and MST is critical to addressing the public health crisis of opioid use. The link between a diagnosis of CPP and the frequency and amount of narcotic prescriptions has not been examined in a longitudinal fashion. Additionally, the risk of chronic opioid therapy (COT) for non-cancer pelvic pain has not been well described. Our objective was to compare COT and narcotic doses prescribed 5 yr prior and following a non-cancer related CPP diagnosis in women veterans. We also anticipated finding distinct patient characteristics that could help predict adverse COT outcomes in women with CPP such a MST based on prior research. We also aimed to look at adverse outcomes related to COT in a population with CPP. We hypothesized that COT use would increase following the diagnosis of CPP. MATERIALS AND METHODS This was a large-scale, retrospective study using clinical records contained in the Veterans’ Health Administration Corporate Data Warehouse which houses all information for Veterans receiving care at a Veteran’s Administration Health Care System since 1997. After obtaining IRB approval (15-H175), with a waiver of HIPAA authorization, data was requested from the Veterans Informatics and Computing Infrastructure (VINCI). Women who declared a single gender, had established care with a primary care physician and who had CPP first diagnosed between 2002 and 2012 by ICD-9 codes comprised the initial cohort. The CPP definition included both non-specific codes as well as specific causes of CPP (such as dysmenorrhea, interstitial cystitis) (see Appendix A). Women were excluded from the study if they had a history of any abdominal or pelvic malignancy. We also excluded women whose CPP seemed related to adnexal torsion, appendicitis or diverticulitis, based on ICD-9 codes. The database also identified all women filling prescriptions at the VA by outpatient VA pharmacy data. To remain in the study cohort, a woman had to be filling prescriptions at the VA for at least a single time point 5 yr before and a single time point 5 yr after the diagnosis of CPP. We used these criteria as we wanted to demonstrate that in some capacity the VA was continuing to provide care for these women. We decided if there were zero prescriptions filled at the VA over 5 yr it was unlikely the VA was providing healthcare for that woman. We choose to look at 10 yr of data per patient both because this was available within the Corporate Data Warehouse and because we thought this would show longer patterns associated between CPP and COT. COT for pain was defined by Veteran’s Health Administration Directive 1005 as “the use of opioids on a daily or intermittent basis for 90 or more calendar days.” In order to identify patients filling narcotic prescriptions we identified all drugs in the VA CN101 drug class (the code for narcotics). We then selected only the oral preparations, excluding transdermal, intranasal and injectables and converted the narcotics into morphine equivalents.18 We excluded non-oral preparations because typically, those preparations treat acute pain or cancer related pain. Narcotic prescriptions and days supply counts were used to identify 90 d of narcotics/calendar year. Age at the time of CPP was calculated. Body mass index (BMI) was calculated as the average BMI over the study period. Because there were multiple responses to the questions of race/ethnicity, MST and smoking over time a specific hierarchy was set for each. Smoking was classified as current, former or never. “Never” smokers were only those patients who always denied smoking. If, at any point, a patient identified as a former smoker but never as a current smoker then she was placed in the former group, and if a patient – identified as a current smoker, they were placed in the current group. Based on self-report patients were broadly grouped into five racial categories (African-American, American Indian, Asian, White, and Unknown) and two ethnicity categories (Hispanic versus non-Hispanic). A patient was considered a member of a minority if she identified herself as African-American, American Indian, or Asian. If a patient ever responded “yes” to a history of MST then the answer was recorded as “yes,” even if subsequent answers were “no” or “declined to answer.” Because these data come from Veterans in the Veteran’s Health Administration data MST (which occurred during military service) preceded the diagnosis of CPP (which was recorded post military service). For this study “declined to answer” and “not asked” responses for MST were grouped together. ICD-9 codes were also used to identify patients’ other chronic non-specific pain conditions such as back pain not otherwise specified, fibromyalgia, headaches, irritable bowel syndrome, and chronic joint pain. If the first date of diagnosis came at any time during the 5 yr prior to the CPP diagnosis date, then the pain condition was considered present at baseline. These other chronic pain conditions were only evaluated at this time point. Adverse outcomes potentially associated with narcotic use included drug abuse, drug overdose, depression, anxiety, and suicide. Adverse events were recorded when emergency room visits or inpatient admissions were associated with these diagnoses. We chose to only include these types of visit encounters as they likely represent the most direct need for immediate intervention rather than the outpatient clinical visits. Emergency services and inpatient admission are also associated with a higher level of care rather than scheduled primary care encounters. Data were analyzed using SAS version 9.4. Cohort demographics were described. Demographic statistics between those who received COT versus those who did not were reported and compared using x2 tests for categorical variables. Student’s t-tests were used for continuous variables, except that Wilcoxon Rank-Sum test was used for non-normally distributed data. The primary outcome was COT, a binary determination for each year of the analysis preceding and following the diagnosis of CPP. Time 0 was identified as the first time of CPP diagnosis. Data were recorded for the 5 yr preceding (−5 to −1) and after (0 to +5). Calculated years before and after were unique and relative to each individual’s CPP diagnosis date. We utilized an interrupted time series approach using a logistic regression method for repeated measures, which were determinations of COT for the 10 yr per veteran (PROC GENMOD in SAS 9.4). This logistic regression considered time as a continuous variable and tested whether the percentage of veterans with COT as a function of time increased after the diagnosis of CPP. This model allowed a discontinuity in COT at year 0 in addition to separate before and after slopes; the presence of this discontinuity tested whether there was extra narcotic usage in the first year following CPP diagnosis. Adverse outcomes: drug abuse, drug overdose, depression, anxiety, and suicide were compared before and after CPP diagnosis. All adverse outcomes covariates were examined for collinearity. Best models were obtained using a backward selection procedure. In these models, each of the adverse outcomes was considered as a complication of COT and was set as the dependent variable with COT as an independent variable. This same method was used to evaluate demographics that increased the likelihood of COT including multivariate modeling with a backward selection procedure. Predictors of COT and adverse outcomes related to COT are reported as odds ratios (OR) with 95% confidence intervals (CI). Significance was set as p < 0.05. RESULTS We identified 596,815 women Veterans who sought care with a VA primary care physician. Of those, 95,860 (16%) were diagnosed with CPP. We excluded 23,718 women as they were not diagnosed between 2002 and 2012 and an additional 11,779 were excluded as they did not fill prescriptions at the VA or seek primary care either during 5 yr before and/or the 5 yr following receipt of their CPP diagnosis. After excluding 10,762 for history of abdominal/pelvic cancers, appendicitis, diverticulitis, or adnexal torsion, there remained 49, 601 women in the study cohort (see Fig. 1). FIGURE 1. View largeDownload slide Diagram of cohort identification. FIGURE 1. View largeDownload slide Diagram of cohort identification. At the time of CPP diagnosis, the average age of women in this cohort was 40.3 (SD ± 11.7) yr, the population was overweight with BMI 29.4 (SD ± 6.0) and most women smoked (51.1%, n = 25,330). Over a third (37.2%, n = 18,465) self- identified as being a minority. Thirty-four percent (n = 16,890) had experienced MST and 30% (n = 15,780) had at least one other chronic pain disorder at the time of CPP diagnosis. The most common co-occurring pain disorders were headaches (15.3%, n = 7,597), back pain (7.9%, n = 3,899), and fibromyalgia (8.2%, n = 4,085). Chronic narcotic (3 or more 30-d prescriptions) use increased significantly (p < 0.001) in the 5 yr preceding the diagnosis of CPP from 6.3% (n = 3124) of women at time −5 to 13.6% (n = 6746) at time 0. In the first year following the diagnosis of CPP, 16.8% (n = 8,333) of women with CPP met the criteria for COT (p < 0.001) and 15% (n = 7440) of women with CPP remained in the COT group for the remaining 5 yr following the diagnosis (see Fig. 2). On average women in the COT group had 250–292 d of opioids/year. FIGURE 2. View largeDownload slide Interrupted time series of percent of patients using COT relative to the time of CPP diagnosis. Percentage COT before CPP diagnosis date (solid blue regression line) increases from 6.3% to 13.6% (p < 0.001) There is an immediate step discontinuity in COT to 16.8% (light blue solid circle) in the year following the CPP diagnosis date (p < 0.001). The regression line after CPP diagnosis (red solid line) remains high (15.8% to 15.2%), significantly higher than the COT usage before CPP diagnosis (p < 0.001). FIGURE 2. View largeDownload slide Interrupted time series of percent of patients using COT relative to the time of CPP diagnosis. Percentage COT before CPP diagnosis date (solid blue regression line) increases from 6.3% to 13.6% (p < 0.001) There is an immediate step discontinuity in COT to 16.8% (light blue solid circle) in the year following the CPP diagnosis date (p < 0.001). The regression line after CPP diagnosis (red solid line) remains high (15.8% to 15.2%), significantly higher than the COT usage before CPP diagnosis (p < 0.001). When comparing women who received chronic narcotics following the diagnosis of CPP versus those who did not receive chronic narcotics, women who received COT were older, more likely to smoke and more frequently diagnosed with other pain conditions such as back pain, headaches and fibromyalgia (all p < 0.001; see Table I). Table I. Demographics for CPP Cohort Comparing Women Veterans Who Received Chronic Opioids Following CPP Diagnosis to Women Veterans Who Did Not Receive Chronic Opioidsa Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  aData presented as N (percentage) unless otherwise noted. Percentages may not add up to 100 due to rounding. Table I. Demographics for CPP Cohort Comparing Women Veterans Who Received Chronic Opioids Following CPP Diagnosis to Women Veterans Who Did Not Receive Chronic Opioidsa Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  Characteristic  No Chronic Opioid Use After CPP Diagnosis N = 35,292  Chronic Opioid Use After CPP Diagnosis N = 14,309  p-Value  Age at time of CPP diagnosis (years ± SD)  39.7 ± 11.8  41.7 ± 11.5  <0.001  Average BMI over 10-yr period (kg/m2 ± SD)  29.2 ± 5.9  30.0 ± 6.3  <0.001  Positive history of MST  10918 (30.9)  5972 (41.7)  <0.001  Ethnicity/race   Non-Hispanic White  17309 (49.1)  8654 (60.5)  <0.001   Hispanic  2125 (6.0)  569 (4.0)     Minority  13876 (39.3)  4589 (32.1)     Unknown/undisclosed  1982 (5.6)  497 (3.47)    Smoking status current  16198 (45.9)  9141 (63.9)  <0.001  Fibromyalgia  2812 (8.0)  2339 (16.4)  <0.001  Generalized abdominal pain  2723 (7.7)  1154 (8.1)  0.19  Irritable bowel syndrome  1867 (5.3)  904 (6.3)  <0.001  Non-specific joint pain  3432 (9.7)  1596 (11.2)  <0.001  Back pain  4810 (13.6)  2976 (20.8)  <0.001  Headaches  11942 (33.8)  5866 (41.0)  <0.001  Anxiety  9065 (25.7)  5159 (36.1)  <0.001  Depression  7768 (22.0)  4763 (33.3)  <0.001  Post-traumatic stress disorder  5119 (14.5)  3053 (21.3)  <0.001  Prior chronic narcotic use  2158 (6.1)  6344 (44.3)  <0.001  aData presented as N (percentage) unless otherwise noted. Percentages may not add up to 100 due to rounding. In the multivariable logistic regression, there were several baseline characteristics which predicted subsequent COT following CPP diagnosis. The strongest predictor was prior COT (OR = 10.0 (95% CI 9.4, 10.6)). Other significant predictors in the model included a positive history of MST, smoking and other chronic pain conditions such as irritable bowel syndrome, fibromyalgia, back pain, and headache. Interestingly, the regression model also identified that as the time of CPP moved forward women were less likely to be prescribed chronic opioids (Table II). Table II. Patient Characteristics in Multivariable Analysis That Increased the Odds of Receiving Chronic Narcotics After the Diagnosis of CPP Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  Table II. Patient Characteristics in Multivariable Analysis That Increased the Odds of Receiving Chronic Narcotics After the Diagnosis of CPP Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  Patient Characteristic  OR (95% CI)  Prior chronic narcotic use  10.0 (9.4, 10.6)  Current smoker  1.8 (1.7, 1.8)  History of MST  1.3 (1.3, 1.4)  Fibromyalgia  1.6 (1.5, 1.7)  IBS  1.2 (1.1, 1.2)  Back pain  1.3 (1.2, 1.4)  Headaches  1.3 (1.2, 1.3)  Progression in time  0.95 (0.95, 0.96)  When evaluating adverse outcomes in women with CPP and COT compared with women with CPP without COT, all outcomes evaluated were significantly associated with COT including suicide, drug abuse, drug overdose and admissions for mental health comorbidities (Table III). Table III. Risk of Adverse Outcomes in CPP Patients Using Chronic Narcotics Versus CPP Patients Not Using Chronic Narcotics Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  Table III. Risk of Adverse Outcomes in CPP Patients Using Chronic Narcotics Versus CPP Patients Not Using Chronic Narcotics Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  Adverse Outcome  OR (95% CI)  Drug abuse ER visits  4.7 (4.0, 5.5)  Drug abuse admissions  3.6 (3.2, 3.9)  Drug overdose ER visits  3.9 (2.8, 5.4)  Drug overdose admissions  4.4 (3.9, 5.1)  Anxiety ER visits  2.4 (2.2, 2.6)  Anxiety admissions  3.6 (3.2, 3.9)  Depression ER visits  2.4 (2.0, 2.8)  Depression admissions  3.2 (3.0, 3.4)  Suicide ER visits  2.6 (2.3, 2.9)  Suicide admissions  2.6 (2.4, 2.8)  Post-traumatic stress disorder ER visits  2.3 (2.1, 2.6)  Post-traumatic stress disorder admissions  3.2 (3.0, 3.5)  DISCUSSION We found that women Veterans with CPP have increasing COT in the 5 yr prior to the formal diagnosis of CPP. Following CPP diagnosis, COT immediately increases and continues for at least 5 yr. The highest risk factor for ongoing COT following CPP diagnosis was a prior history of COT. Women diagnosed with CPP who also had a history of MST, headaches, back pain, IBS, smoking, and fibromyalgia were at increased risk of developing COT. These other pain conditions likely contribute the COT prescribing however there remains an independent, significant and distinct pattern of COT associated with the diagnosis of CPP. Previous work found that Veterans with COT of greater than 20 mg/d morphine equivalents had increased risk of suicide compared with those that received less than 20 mg/d.3 In our study population, women with CPP who received COT had approximately 2–4 fold increased risk of admissions and ER visits for drug abuse, drug overdose, anxiety, depression, PTSD, and suicide compared with women with CPP not prescribed COT. This study is consistent with general work among Veterans during fiscal years 2004–2012, during which time the prevalence of opioid prescriptions increased from 18.9% to 33.4%, an increase of 76.7%. The groups with the highest prevalence of opioid use were women and young adults (i.e., 18–34 yr old).19 Our work seems to indicate that CPP is associated with COT and in a significant proportion of women, antedates development of COT. The pattern of use before and after CPP diagnosis might indicate that physicians prescribe opiates gradually but then following a formal chronic pain diagnosis become more comfortable with long-term narcotic use (as indicated by the high stable percentage of Veterans prescribed COT after CPP diagnosis for 5 yr). This study has limitations. The population was limited to only women Veterans with 10 yr of data in the CDW database, the data are limited by the accuracy of those who entered it and there may be women treated for CPP that were not coded as having CPP. The conclusions may not be valid for non-Veterans, younger veterans (without 10 yr of data in the CDW) or Veterans with other insurance programs. While we excluded women with abdominal or pelvic cancers there may some women with other cancer types being treated with COT that we included in the CPP cohort. Baseline pain scores as they specifically related to the diagnosis of CPP were also unavailable and this information could help better classify who receives COT. Additionally, there were unmeasured factors related to COT that were not captured in the database including patients’ social situations or medications or other healthcare obtained outside of the VA system. Patients with COT are known to seek multiple providers, visit emergency rooms, and purchase street-drugs in an effort to obtain additional narcotic medications.20 Our data were limited to the narcotics filled at the VA pharmacies. We believe that these issues potentially result in under-representation of the extent of COT in our database. While COT has been looked at in Veteran population it has not been specifically explored in relationship to CPP, a condition that disproportionately affects women Veterans. CPP appears to contribute to the overuse of narcotics in the women Veteran population. A significant strength in this study is that we used prescription fill data (indicating that the patient did pick-up these medications) rather than prescribed data and believe this is important because these opioids were released into the patient’s care Further research could compare prescribed versus filled opioid prescriptions and also investigate actual usage of these prescriptions. Additionally, we were able to evaluate COT over an extended period, which demonstrated the distinct longitudinal association between CPP and COT. This research suggests that COT is a common treatment for CPP in the women Veteran population. It also adds to the growing literature showing harm associated with COT. Given the harms our patients deserve more research on therapeutic alternatives outside of chronic narcotics for the treatment of pelvic pain. The American College of Obstetricians and Gynecologists online patient education site for CPP lists non-narcotic therapies for CPP treatment and provides a useful resource for our patients.21 Interventions that reduce COT prescribing through the education of primary care physicians and mid-level providers should also be explored. CONCLUSION In conclusion, CPP antedates a sustained increase in COT. COT is associated with risks of ER and inpatient visits related to drug abuse and psychiatric illnesses. While this work cannot prove causality between COT and CPP the pattern associated is concerning and suggests that CPP is likely associated with COT. Physicians should be judicious in their prescribing of narcotics for CPP and inform patients of the potential risks. Supplementary Material Supplementary material is available at Military Medicine online. REFERENCES 1 Reuben DB, Alvanzo AA, Ashikaga T, et al.  : National Institutes of Health Pathways to Prevention Workshop: the role of opioids in the treatment of chronic pain. Ann Intern Med  2015; 162( 4): 295– 300. Google Scholar CrossRef Search ADS PubMed  2 Klingensmith K, Tsai J, Mota N, Southwick SM, Pietrzak RH: Military sexual trauma in US Veterans: results from the National Health and Resilience in Veterans Study. J Clin Psychiatry  2014; 75( 10): 1133– 9. Google Scholar CrossRef Search ADS   3 Bohnert AS, Ilgen MA, Galea S, McCarthy JF, Blow FC: Accidental poisoning mortality among patients in the Department of Veterans Affairs Health System. Med Care  2011; 49( 4): 393– 6. Google Scholar CrossRef Search ADS PubMed  4 Ray WA, Chung CP, Murray KT, Hall K, Stein CM: Prescription of long-acting opioids and mortality in patients with chronic noncancer pain. JAMA  2016; 315( 22): 2415– 23. Google Scholar CrossRef Search ADS PubMed  5 White AM, Hingson RW, Pan IJ, Yi HY: Hospitalizations for alcohol and drug overdoses in young adults ages 18---24 in the United States, 1999--2008: results from the Nationwide Inpatient Sample. J Stud Alcohol Drugs  2011; 72( 5): 774– 86. Google Scholar CrossRef Search ADS PubMed  6 Manchikanti L, Helm S 2nd, Fellows B, et al.  : Opioid epidemic in the United States. Pain Physician  2012; 15( 3): ES9– 38. Google Scholar PubMed  7 Paulozzi LJ1, Strickler GK, Kreiner PW, Koris CM: Controlled Substance Prescribing Patterns – prescription behavior surveillance system, eight states, 2013.Centers for Disease Control and Prevention (CDC). MMWR Surveill Summ  2015; 64( 9): 1– 14. Google Scholar CrossRef Search ADS PubMed  8 Seal KH1, Shi Y, Cohen G, et al.  : Association of mental health disorders with prescription opioids and high-risk opioid use in US veterans of Iraq and Afghanistan. JAMA  2012; 307( 9): 940– 7. Google Scholar CrossRef Search ADS PubMed  9 Doggweiler R, Whitmore KE, Meijlink JM, et al.  : A standard for terminology in chronic pelvic pain syndromes: a report from the chronic pelvic pain working group of the international continence society. Neurourol Urodyn  2016; 36: 984– 1008. Google Scholar CrossRef Search ADS PubMed  10 Reiter RC: A profile of women with chronic pelvic pain. Clin Obstet Gynecol  1990; 33( 1): 130. Google Scholar CrossRef Search ADS PubMed  11 Zondervan KT, Yudkin PL, Vessey MP, Dawes MG, Barlow DH, Kennedy SH: Prevalence and incidence of chronic pelvic pain in primary care: evidence from a national general practice database. Br J Obstet Gynaecol  1999; 106: 1149– 55. Google Scholar CrossRef Search ADS PubMed  12 Ahangari A: Prevalence of chronic pelvic pain among women: an updated review. Pain Physician  2014; 17( 2): E141– 7. Google Scholar PubMed  13 Cichowski SB, Rogers RG, Clark EA, Murata E, Murata A, Murata G: Military sexual trauma in american women veterans is associated with chronic pain conditions. Mil Med  2017; 182( 9): e1895– 9. Google Scholar CrossRef Search ADS PubMed  14 Siedentopf F, Weijenborg P, Engman M, et al.  : ISPOG European Consensus Statement – chronic pelvic pain in women (short version). J Psychosom Obstet Gynaecol  2015; 36( 4): 161– 70. Google Scholar CrossRef Search ADS PubMed  15 Steele A: Opioid use and depression in chronic pelvic pain. Obstet Gynecol Clin North Am  2014; 41( 3): 491– 501. Google Scholar CrossRef Search ADS PubMed  16 Latthe P, Mignini L, Gray R, Hills R, Khan K: Factors predisposing women to chronic pelvic pain: systematic review. BMJ  2006; 332( 7544): 749– 55. Google Scholar CrossRef Search ADS PubMed  17 Military Sexual Trauma. Available at https://www.mentalhealth.va.gov/msthome.asp; accessed March 31, 2018. 18 Centers for Medicare and Medicaid Services. Available at https://www.cms.gov/Medicare/Prescription-Drug-Coverage/PrescriptionDrugCovContra/Downloads/Opioid-Morphine-EQ-Conversion-Factors-Aug-2017.pdf; accessed May 8, 2018. 19 Mosher HJ, Krebs EE, Carrel M, Kaboli PJ, Vander Weg MW, Lund BC: Trends in prevalent and incident opioid receipt: an observational study in Veterans Health Administration 2004–2012. J Gen Intern Med  2014; 30( 5): 597– 604. Google Scholar CrossRef Search ADS PubMed  20 Hall AJ, Logan JE, Toblin RL, et al.  : Patterns of abuse among unintentional pharmaceutical overdose fatalities. JAMA  2008; 300( 22): 2613– 20. Google Scholar CrossRef Search ADS PubMed  21 American College of Obstetricians and Gynecologists Patient Frequently asked fact sheet. Available at http://www.acog.org/Patients/FAQs/Chronic-Pelvic-Pain; accessed May 8, 2018. Published by Oxford University Press on behalf of Association of Military Surgeons of the United States 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US.

Journal

Military MedicineOxford University Press

Published: May 18, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off