Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Feb 9.
Published in final edited form as: Pain. 2020 Nov;161(11):2511–2519. doi: 10.1097/j.pain.0000000000001960

The transition from acute to persistent pain: the identification of distinct trajectories among women presenting to an emergency department

John W Burns 1, Imke Janssen 1, Teresa Lillis 1, Morgan Mulcahy 1, Yanina A Purim-Shem-Tov 1, Stephen Bruehl 2, Helen J Burgess 3, Alexandra Fischer 4, Katie Rim 5, Frances Aranda 6, Linzy Pinkerton 7, Stevan Hobfoll 8
PMCID: PMC10853846  NIHMSID: NIHMS1603282  PMID: 32569094

Abstract

PTSD symptoms and other negative psychosocial factors have been implicated in the transition from acute to persistent pain. Women (N = 375) who presented to an inner-city Emergency Department (ED) with complaints of acute pain were followed for 3 months. They completed a comprehensive battery of questionnaires at an initial visit, and provided ratings of pain intensity at the site of pain presented in the ED during 3 monthly phone calls. Latent class growth analyses were used to detect possible trajectories of change in pain intensity from initial visit to 3 months later. A 3–trajectory solution was found which identified three groups of participants. One group (early recovery; n = 93) had recovered to virtually no pain by the initial visit, whereas a second group (delayed recovery; n = 120) recovered to no pain only after one month. A third group (no recovery; n = 162) still reported elevated pain at 3-months post ED visit. The no recovery group reported significantly greater PTSD symptoms, anger and sleep disturbance, as well as lower social support, at initial visit than both the early recovery and delayed recovery groups. Results suggest that women with high levels of PTSD symptoms, anger, sleep disturbance and low social support who experience an acute pain episode serious enough to prompt an ED visit may maintain elevated pain at this pain site for at least three months. Such an array of factors may place women at increased risk of developing persistent pain following acute pain.

Summary

Women with high levels of PTSD symptoms, anger, poor sleep and low social support who experience acute pain may maintain elevated pain for three months.

Posttraumatic stress disorder (PTSD) and chronic pain conditions are prevalent and are associated with emotional, functional, and financial costs for individuals and society. PTSD is estimated to affect 5.2 million Americans in a given year [43], whereas approximately 100 million Americans are chronic pain sufferers [17, 30, 32]. PTSD and chronic pain are frequently associated, with chronic pain reported in 20–70% of people with PTSD [1, 2, 4, 5, 6, 10, 37, 39, 40, 41, 42, 52, 54]. The comorbidity of PTSD and chronic pain may magnify pain, distress and disability beyond the effects of PTSD or chronic pain alone [22, 23, 24, 49, 58].

Understanding of how comorbid PTSD and chronic pain develops over time is, however, limited. Most studies of this phenomenon are cross-sectional, precluding examination from a longitudinal perspective. Still, a handful of longitudinal studies provide initial insight into this issue. Shaw and colleagues [50] studied first onset of low back pain (LBP). Patients with a pre-LBP lifetime diagnosis of PTSD had a significantly increased risk of transitioning from acute to chronic LBP by 6 months. Jenewein and colleagues [31] found that PTSD was related to the development of chronic pain following motor vehicle accidents (MVA). In a second study, Jenewein and colleagues [32] found that at each of three time points, people with severe PTSD had higher pain intensity. However, these results are limited to people whose pain and PTSD had a common initiating event (an MVA). Such a common source is not present or detectable in the vast majority of people with PTSD and chronic pain [5].

In this study, we used a prospective design to evaluate whether PTSD symptoms play a role in the transition from acute to persistent pain. The sample consisted of women who presented to an inner-city Emergency Department (ED) with an acute pain complaint that was not the result of a traumatic event. Preliminary surveys suggested that women at this particular ED reported high levels of traumatic stress relative to men. Women approached in the ED presented with a new onset of acute pain that was not due to a diagnosed, chronic condition. Pain intensity was assessed in the ED, at an initial interview, and at one-, two-, and three- months post-initial visit. Note that pain assessments at all sessions referred specifically to the pain complaint originally presented at the ED. At the detailed initial visit, we assessed trauma exposure, PTSD symptoms, and psychosocial resilience (e.g., social support) and vulnerability factors (e.g., depressive symptoms).

As others have argued [20], focusing on mean levels of pain intensity may obscure important subgroups within a sample. We used latent class growth analyses to detect possible distinct trajectories of change in pain intensity over 3.5 months. Following work by Downie and colleagues [20, 21], we expected at least two trajectories to emerge: one trajectory including women who recovered from the ED pain complaint relatively quickly, and another including women who did not recover fully from the ED pain complaint. We also expected that participants described by the latter trajectory would be characterized by high levels of PTSD symptoms and other vulnerability factors.

Method

Participants

Participants (N = 375) were women who presented to a Chicago inner-city ED with an acute pain complaint. Only women comprised the sample because the intent of the overall study was to examine co-occurring PTSD symptoms and chronic pain. The impact of this combination remains decidedly understudied among inner-city women, especially African American women, in whom it may be most pronounced [15, 53]. All participants read and signed a written informed consent form. The study protocol was approved by the Institutional Review Board at Rush University Medical Center. Pilot data collected at our ED indicated that abdominal/pelvic, shoulder/neck, chest and low back pain were the most common complaints among women, and thus we confined our sample to women presenting with these pain complaints. Study staff approached women at the ED with information about participating in a study about trauma and pain, and indicated that they would be compensated for their time and travel expenses. For women who expressed interest in participating, study staff collected contact information and then conducted a brief telephone-screening interview within 72-hours of the ED visit to determine eligibility to participate in the study. Participants received a total of $350 for completing an initial detailed session, phone call follow-ups, and another detailed session at 3-months following the initial session.

Inclusion criteria were: (1) female, (2) between 18–40 years old, (3) premenopausal, (4) able to read and write English well enough to provide informed consent and participate in interviews, and (5) presented to the ED with an acute pain complaint of the chest, abdomen/pelvis, neck/shoulder, or back (i.e., not extremity or head pain). Exclusion criteria were: (1) pain intensity or any injury or illness great enough to impair concentration or capacity to understand study instructions or the nature of being in the study, (2) current chronic illness that involved constant or frequent pain, (3) self-reported history of chronic pain on presentation to the ED or documented in the Electronic Medical Record (EMR), (4) appearing intoxicated or under the influence of substances at the ED visit, (5) self-reported or EMR-documented daily opiate use over the prior 3 months, or (6) the presenting ED pain complaint was due to a traumatic circumstance (e.g., a motor vehicle accident [MVA], physical assault, sexual assault, etc.). This latter exclusion criterion was established in order to avoid the confounding effects of the presenting pain complaint and any reported PTSD symptoms being from the same event (e.g., a MVA survivor may have both pain and PTSD from the MVA, not because pain and PTSD impact each other). Moreover, participants who underwent a procedure or surgery as a consequence of their ED visit (pre or post) were not contacted regarding participation in the study. Note that women over 40 years of age were excluded due to a greater likelihood of having a preexisting chronic pain condition. In addition, participant EMRs and notes from their ED encounter were individually reviewed by a co-author (YPST), who is an Emergency Medicine physician, to confirm the location and suspected cause of the acute pain complaint.

A total of 622 people were deemed eligible for participation during a telephone screening. Of those, 452 completed initial assessments. Further, 216 people declined to give contact information in the ED, 195 people could not be contacted, and 318 people were phone screened but deemed ineligible. The participants included in this study were those with complete data at the initial visit and three-month follow-up (N = 375). To address possible sampling bias, we compared the 77 participants who completed initial assessments but did not complete three-month assessments to the 375 people who did. Please see Results section.

Measures

PTSD Symptoms.

The total score from the 20-item PTSD Checklist for DSM-5 [(PCL-5); 7] was used to measure PTSD symptoms. Prior to administration of the PCL-5, participants were asked to reflect on their worst or most distressing trauma (i.e., their index trauma) and were then asked to rate the degree to which over the previous month, on a 0 (not at all) to 4 (extremely) scale, they had experienced PTSD symptoms related to re-experiencing, avoidance, hyperarousal, and negative alterations in cognition and mood. Total scores range from 0–80 with higher scores indicating more severe symptoms of PTSD. Scores above 33 are suggestive of a potential PTSD diagnosis. Based on this clinical cut-off score, 44% of the sample could meet criteria for a PTSD diagnosis. The measure has adequate reliability in this sample (α = .96) and in past samples (α = .94); 45].

Trauma History

The total number of Criterion A traumas was measured using the Trauma History Questionnaire (THQ), a self-report measure wherein participants were asked to recall their lifetime trauma history [29]. The questionnaire consists of 24 “yes/no” questions pertaining to crime related events, general disaster and trauma, and physical and sexual experiences such as “Have you ever seen someone seriously injured or killed?” If participants disclosed an experienced trauma, they were asked how many times it happened, a brief description of the trauma/traumas, at what age was the first occurrence of said trauma/traumas, and if applicable, the nature of the relationship with the persons involved.

Resilience Factors.

Social support was assessed with a 10-item scale based on Weiss’s [57] theory of social provisions using a scale from 0–2 (No= 0, 1 = Sometimes, 2 = Yes) with 5 negatively worded items reversed scored. Respondents were asked if they had someone in whom they can confide, with whom they can talk about problems, who will help with chores and responsibilities, to whom they can turn for support, and who makes them feel loved and wanted. Total scores can range from 0–20, with higher scores indicating higher perceived social support. The measure displayed adequate reliability in this sample (α = .81) and in past samples (α = .81); 57].

Optimism was assessed with the total score from The Life Orientation Test – Revised [LOT-R; 14], a 10-item measure designed to assess individual differences in generalized optimism versus generalized pessimism on a 0 (Strongly Disagree) to 4 (Strongly Agree) scale. The scale contains three optimism items, three pessimism items and four “filler” items. Pessimism items are reversed scored and total scores are calculated without “filler” items. Higher total scores on the LOT-R indicate higher Optimism. The measure had adequate reliability in this sample (α = .73) and in past samples (α = .73); 47].

Vulnerability Factors.

Social undermining was assessed with The Social Undermining Scale [28], a 7-item scale that asks respondents to rate, on a scale from 1 (not at all) to 5 (nearly all the time), how often individuals closest to the respondent display negative affect towards the respondent, negatively evaluate or criticize the respondent, and hinder the respondent’s attainment of personal goals. Total scores can range from 7–35, with higher scores indicating more frequent experiences of social undermining. The measure had adequate reliability in this sample (α = .88) and in past samples (α = .89); 28].

Depressive symptoms were assessed with the total raw score from the PROMIS-4 Short-Form Depression scale [44] that asks respondents to rate, on a scale from 1 (not at all) to 5 (very much), how much they felt worthless, helpless, hopeless, and had a depressed mood over the previous 7 days. Total raw scores can range from 4–20, with higher scores indicating higher total of depressive symptoms. The measure displayed adequate reliability in this sample (α = .91) and in past samples (α = .91); 44].

Anger symptoms were evaluated with the total raw score from the PROMIS Anger Short-Form Scale [44] that asks respondents to rate, on a scale of 1 (never) to 5 (always), how often they felt irritated, angry, ready to explode, grouchy, and annoyed over the past 14 days. Total raw scores can range from 5–25, with higher scores indicating increased frequencies of experienced anger. The measure had adequate reliability in the current (α = .90) and in past samples (α = .90); 44].

Anxiety symptoms were assessed with the total raw score from the PROMIS-4 Short-Form Anxiety Scale [44] that asks participants to rate on the scale of 1 (never) to 5 (always), the frequency of experienced feelings of fearfulness, anxiety, being overwhelmed, and uneasiness over the past 14 days. Total raw scores can range from 4–20, with higher scores indicating more severe anxiety symptoms. The measure showed adequate reliability in this sample (α = .89) and in past samples (α = .89); 44].

Sleep disturbance was measured utilizing total raw scores from the PROMIS Sleep Disturbance Short-Form Scale [44]. Respondents were asked to rate on the scale of 1 (not at all) to 5 (very much), if they felt their sleep was refreshing, restless, problematic, or if they experienced trouble falling asleep over the past 7 days. Total raw scores can range from 5–25, with higher scores indicating greater experienced sleep disturbance. The measure had adequate reliability in this sample (α = .87) and in past samples (α = .87); 44].

Pain Intensity.

Pain intensity was measured at each assessment point on an 11-point numeric rating scale [16] of how much pain the participant was experiencing at that moment on a scale of 0 (none at all) to 10 (extreme) in the same area of their body that they had experienced pain on initial presentation to the ED.

Procedure

Research assistants (RAs) had access to the live/active list of patients in the Rush ED via the EPIC Medical Records System. RAs monitored the live/active list of patients in the ED throughout the work day. RAs were able to assess eligibility criteria of each patient on the active patient list through access to medical and surgical history/MD and triage notes/presenting symptoms/diagnosis via EPIC EMR. Patients identified on the list of active patients in the ED as having met inclusion criteria were approached in the ED waiting room or their assigned ED treatment room by the RA.

Eligible participants were asked to complete an initial detailed in-person assessment within 14 days of their ED visit (actual M = 10.48; SD = 5.15), during which they provided informed consent upon receiving a written and verbal description of the study. At this assessment, the interview was conducted and questionnaires completed. A follow-up in-person assessment session was conducted three months later. In addition, phone assessments of pain intensity for the original source of pain reported in the ED were conducted at one- and two-months following the initial detailed assessment.

Statistical Analyses

All analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC). Descriptive statistics were calculated for all demographic and study variables. Trajectory analyses were first conducted to examine patterns of pain intensity from the source of pain originally reported at the ED from initial visit to 3 months later using pain data collected at the initial visit, and one, two, and three-month follow-up assessments. The SAS PROC TRAJ procedure [8] was used to determine models of pain intensity across all time points. While the range of pain intensities ranged from 0 to 10, there were a disproportionate number of zeros, so we estimated zero-inflated Poisson regression models. The TRAJ procedure is a mixture model that estimates a regression model for each discrete group within the sample. PROC TRAJ can include missing observations because it uses all values available from each case to estimate an individual’s timeline. Model complexity and overall fit in PROC TRAJ is determined by the Bayesian information criterion (BIC), which are negative values in which values closer to 0 indicate a better fit.

To test the inclusion of different numbers of trajectories, the estimate of the log Bayes Factor is defined with the following formula: 2loge (B10) ≈ 2(ΔBIC) [8]. The difference (ΔBIC) is calculated by subtracting the BIC value of the simpler model (e.g. the model with a smaller number of trajectories), from the BIC of the more complex model. According to guidelines, values ranging from 0–2 are interpreted as weak evidence for the more complex model, 2 to 6 as moderate evidence, 6–10 as strong evidence and values >10 as very strong evidence. The comparisons are completed in a step-wise manner so that the two-group model is compared to the one-group, and the three-group model is compared to the two-group model, and so on. Trajectory membership for each individual was used as an identifier, and further descriptive analyses were conducted. In addition to BIC values, visual inspection of graphic model curves was used to determine the number of trajectories, similar to cluster and exploratory factor analysis. PROC TRAJ provides fit estimates and probabilities that each woman belongs to each of the modeled trajectory groups. Average probability for members of a trajectory group should be ≥.70 [50]. The models with 3 or 4 groups all fit significantly better than models with 1 or 2 groups. As judged by the BIC criterion, the best model with 4 groups was marginally better than the 3 group model (2ΔBIC) = 6.2). However, the group membership probabilities were higher for the 3 group solution. Detailed examination of results indicated that the 4 group solution split the middle group into 2 equally large groups, each with lower probabilities, which did not appear to be a conceptual advantage.

To assess differences on demographic and questionnaire measures between the resulting trajectory groups, we used analysis of variance for continuous variables and logistic regression for binary variables with subsequent pairwise comparisons. All 375 women provided pain assessments at the initial visit and at 3 months. Only 11 (2.9%) women had missing pain values at intermittent visits (3 at month 1, 7 at month 2, and 1 at both).

Results

Pain Intensity Trajectories

The trajectory analysis rendered three distinct groups of pain intensity courses across the three months. Note that all of the following trajectory groups reported more intense pain at the ED compared to later assessments, but the 3 trajectory groups showed non-significant pain differences at that point (see Table 1). All women in Group 1 (n = 93) reported no pain at the initial detailed assessment visit (within 2 weeks of the ED visit) and throughout the 3 months; hereafter referred to as the “early recovery” group. Group 2 (n = 120) came to the initial appointment with a mild level of pain (M = 2.5, SD= 1.9) and reported significant improvement in their pain (M = 0.3, SD= 0.6) over the course of the 3 months of follow-up, with the biggest improvement occurring between initial visit and 1-month; hereafter referred to as the “delayed recovery” group. Finally, Group 3 (n = 162) reported a moderate level of pain at the initial visit (M = 3.9, SD= 2.5) and showed little to no improvement by the 3-month interview (M = 2.9, SD= 2.3); hereafter referred to as the “no recovery” group.

Table 1.

Trajectory Group Comparisons on Pain and Resilience and Vulnerability Factors

Total 1. Early Recovery 2.Delayed Recovery 3. No Recovery
Outcome (N=375) (N=93) (N=120) (N=162) Eta2 P Overall P(1 vs 2) P(1 vs 3) P(2 vs 3)
ED Pain Intensity 7.2 (2.2) 6.8 (2.4) 7.2 (2.3) 7.4 (2.0) 0.01 0.103
BL Pain Intensity 2.5 (2.5) 0.0 (0.0) 2.5(1.9) 3.9(7.5) 0.38 <.001 < .001 < .001 < .001
M1 Pain Intensity 1.5 (2.2) 0.0 (0.0) 0.4 (0.7) 3.2 (2.4) 0.46 <.001 0.117 <.001 <.001
M2 Pain Intensity 1.5 (2.2) 0.0 (0.0) 0.2 (0.6) 3.2 (2.4) 0.47 <.001 0.303 <.001 <.001
M3 Pain Intensity 1.3 (2.0) 0.0 (0.0) 0.3 (0.6) 2.8 (2.3) 0.42 <.001 0.106 <.001 <.001
PROMIS Sleep 17.0 (6.2) 15.9 (5.9) 16.0 (5.6) 18.3 (6.6) 0.05 0.001 0.938 0.003 0.002
PROMIS Anxiety 9.5 (4.3) 8.0 (3.5) 9.6 (4.2) 10.3 (4.7) 0.04 <.001 0.007 <.001 0.168
PROMIS Anger 13.7 (5.0) 12.0 (4.7) 13.3 (4.9) 15.0 (4.9) 0.04 <.001 0.047 <.001 0.004
PROMIS Depression 7.6 (4.0) 6.1 (2.8) 7.7 (4.2) 8.4 (4.2) 0.06 <.001 0.004 <.001 0.124
Optimism 15.7 (4.1) 16.9 (3.5) 15.6 (4.4) 15.2 (4.2) 0.05 0.004 0.017 0.001 0.399
Past month PTSD 37.0 (15.6) 33.1 (12.5) 34.7 (14.2) 41.0 (17.2) 0.03 <.001 0.425 <.001 0.001
Social undermining 15.0 (5.6) 13.8 (4.9) 14.8 (5.4) 15.9 (5.9) 0.02 0.016 0.210 0.005 0.104
Trauma Total 4.6 (3.0) 4.0 (2.9) 4.6 (3.0) 4.8 (2.9) 0.01 0.106
Social Support 15.3 (4.2) 16.6 (3.6) 15.7 (4.0) 14.4 (4.4) 0.05 <.001 0.106 0.008 <.001
Open in a new tab

Differences between Trajectory Groups and Drop-Outs on Initial Assessments

As indicated above, 77 people who completed initial assessments did not complete 3-month assessments, and so were not included in the trajectory analyses. One concern is that the absence of these people may have affected the composition of the three recovery groups. It may be the case that these people were the most distressed and/or had the highest pain intensity. First we compared the 77 drop-outs to the early recovery group on all initial measures and ED pain intensity. The 77 drop-outs reported significantly less pain intensity at the ED visit than the early recovery group [F(1,171) = 5.64; p < .02; η2 = .03], but significantly more pain intensity at initial detailed assessment visit than the early recovery group [F(1,171) = 65.11; p < .001; η2 = .25]. The drop-outs and early recovery group did not differ significantly on any other measure [F’s < 3.75; p’s > .06; η2’s < .02]. Second, we compared the 77 drop-outs to the no recovery group on all initial measures and ED pain intensity. The 77 drop-outs reported significantly less pain intensity at the ED visit than the no recovery group [F(1,238) = 25.00; p < .001; η2 = .09], and significantly less pain intensity at the initial detailed assessment visit than the no recovery group [F(1,238) = 31.40; p < .001; η2 = .11]. The drop-outs also reported lower anxiety, anger and depressive symptoms than the no recovery group [F’s > 9.57; p’s < .02; η2’s > .04]. The participants who dropped out scored more similarly to the early recovery group, but they reported less pain and distress than the no recovery group. Thus, it is not clear to what extent their inclusion in the trajectory analyses would have altered group characteristics.

Group Differences on Demographic Characteristics, Analgesic Medication Prescriptions, and Pain Sites

As shown in Table 2, results of analyses comparing the three trajectory groups on demographic characteristics showed that differences across all indexes were nonsignificant. Thus, the groups did not differ significantly on characteristics (e.g., employment) that have been shown in cross-sectional studies to be related to pain intensity [12, 13, 18, 19, 22, 26, 35, 51]. Indeed, to replicate these findings in this sample, we compared demographic groups on pain intensity at initial visit. See Table 3. Omnibus F-tests for the relationships between initial visit pain intensity and most demographic variables including education, income level, and race/ethnicity [F’s > 3.99; p’s < .05; η2’s > .011] were significant in directions previously reported by other investigators. Also in Table 2 are results of analyses comparing the three groups on the frequency with which participants were advised during their ED visit to take over-the-counter or prescription analgesic medication. Group differences were nonsignificant [Χ2 < 1]. Finally, we examined whether groups differed on how many analgesic medications they reported taking at the 3-month follow-up. Again, the groups did not differ significantly [F(2,372) < 1].

Table 2.

Demographic Characteristics and Pain Treatment Prescription Comparisons by Group

Early Recovery Delayed Recovery No Recovery P-value
Characteristic N (%) (N=93) (N=120) (N=162)
Education 0.180
 High school or less 28 (30.1) 31 (25.8) 63 (38.9)
 Some college 43 (46.2) 55 (45.8) 66 (40.7)
 Bachelor or higher 22 (23.7) 34 (28.3) 33 (20.4 )
Employment 0.055
 Full 47 (50.5) 67 (55.8) 69 (42.6)
 Part-time/multiple jobs 23 (24.7) 33 (27.5) 41 (25.3)
 Unemployed 23 (24.7) 20 (16.7) 52 (32.1)
In a relationship 0.654
 No 40 (43.0) 44 (36.7) 68 (42.0)
 Yes 51 (54.8) 75 (62.5) 90 (55.6)
 Missing 2 (2.2) 1 (0.8) 4 (2.5)
Income 0.182
 <$10,000 25 (26.9) 31 (25.8) 58 (35.8)
 $10,000–$40,000 33 (35.5) 50 (41.7) 68 (42.0)
 >$40,000 34 (36.6) 38 (31.7) 35 (21.6)
 Declined 1 (1.1) 1 (0.8) 1 (0.6)
Income meets needs 0.264
 No 41 (44.1) 60 (50.0) 92 (56.8)
 Enough 41 (44.1) 51 (42.5) 60 (37.0)
 More than enough 11 (11.8) 9 (7.5) 10 (6.2)
Race 0.055
 White 13 (14.0) 28 (23.3) 14 (8.6)
 Black 51 (54.8) 52 (43.3) 93 (57.4)
 Latina 25 (26.9) 33 (27.5) 46 (28.4)
 Other 4 (4.3) 7 (5.8) 9 (5.6)
Analgesic Prescribed 0.876
 No 52 (56.0) 69 (57.5) 96 (59.3)
 Yes 41 (44.0) 51 (42.5) 66 (40.7)
Open in a new tab

Table 3.

Pain Intensity Initial Visit Comparisons by Group Demographics

Characteristic N Pain Intensity M (SD) P-value Overall
Education 0.015
 1 High school or less 119 2.8 (2.7)
 2 Some college 164 2.6 (2.6)
 3 Bachelor or higher 89 1.8 (2.5)
Employment 0.119
 1 Full 183 2.3 (2.4)
 2 Part-time/multiple jobs 97 2.4 (2.5)
 3 Unemployed 92 2.9 (2.7)
Income 0.001
 1 <$10,000 114 2.8 (2.7)
 2 $10,000–$40,000 151 2.7 (2.6)
 3 >$40,000 107 1.7 (2.4)
Income meets needs 0.010
 1 Not Enough 193 2.8 (2.6)
 2 Enough 149 2.2 (2.4)
 3 More than enough 30 1.5 (1.9)
Race 0.042
 1 Black 194 2.7 (2.7)
 2 Non-Black 178 2.2 (2.3)
Open in a new tab

We also compared the groups on the frequency of pain sites presented at the ED. As shown in Table 4, the three groups did not differ significantly on how many women presented with acute pain at each of the pain sites we recorded.

Table 4.

Table of Pain Location by Recovery Group

Recovery Group
Pain Location N (%) Overall (N=375) Early (N=93) Delayed (N=120) No (N=162)
Abdomen 199 (53.1) 47 (50.5) 72 (60.0) 80 (49.4)
Back 43 (11.5) 8 (8.6) 13 (10.8) 22 (13.6)
Chest 80 (21.3) 29 (31.2) 22 (18.3) 29 (17.9)
Combination 24 (6.4) 4 (4.3) 7 (5.8) 13 (8.0)
Other (Neck, Shoulder, Pelvis) 29 (7.7) 5 (1.3) 6 (1.6) 18 (4.8)
Total 375 (100.0) 93 (24.8) 120 (32.0) 162 (43.2)
Statistics for Table of Pain Location by Recovery Group
Statistic DF Value P-value
Chi-Square 8 14.1679 0.0775
Likelihood Ratio Chi-Square 8 13.7058 0.0898
Mantel-Haenszel Chi-Square 1 3.0218 0.0822
Phi Coefficient 0.1944
Contingency Coefficient 0.1908
Cramer’s V 0.1374
Sample Size = 375
Open in a new tab

Group Differences on Pain Intensity

As shown in Table 1, results of ANOVAs comparing the three trajectory groups on pain intensity at the five assessment points are provided to further detail the nature of pain changes observed over time in the three trajectory groups. There were no significant differences in pain when participants presented at the ED [F(2,372) = 2.3; p > .10; η2 = 0.01]. Thus, the three groups initially reported similar intensities of the acute pain that prompted them to visit the ED. This was not the case for the detailed initial assessment [F(2,372) = 113.7; p < 0.001; η2 = 0.38], or one- [F(2,368) = 155.2; p < 0.001; η2 = 0.46], two- [F(2,365) = 158.4; p < 0.001; η2 = 0.47], and three-month follow-up visits [F(2,372) = 135.5; p < 0.001; η2 = 0.42], wherein all the omnibus F tests were significant. Pairwise comparisons for initial assessment visit values revealed that the early recovery, delayed recovery, and no recovery groups all differed significantly from each other [t’s(211 or 253 or 280) > 5.8; p’s < 0.001]. By one-month follow-up, however, pairwise comparisons revealed that the early recovery and delayed recovery groups no longer differed significantly [t(208) = 1.6; p = 0.117], whereas both the early recovery group [t(249) = 14.9; p < 0.001] and the delayed recovery group reported significantly lower pain intensity than the no recovery group [t(279) = 14.5; p < 0.001]. This pattern of omnibus F tests and pairwise comparison results were similar for the other two assessments (two- and three-month). Thus, the delayed recovery group recovered by one-month follow-up, whereas the no recovery group continued to report elevated pain out to three-month follow-up that was greater in intensity than the pain intensity reported by the other two groups.

To illustrate this pattern of change over time, we report the percentage of participants rating their pain as a 4 or greater in pain intensity (i.e., moderate-severe pain) in each group. For the early recovery group, 86% reported pain intensity of 4 or greater at the ED visit, and none reported pain intensity of 4 or greater at the initial assessment visit and 3-months following the ED visit. For the delayed recovery group, 92.5% reported pain intensity of 4 or greater at the ED visit, 29.2% reported pain intensity of 4 or greater at the initial assessment visit, and none reported pain intensity of 4 or greater 3-months following the ED visit. Finally, for the no recovery group, 95.1% reported pain intensity of 4 or greater at the ED visit, 54.9% reported pain intensity of 4 or greater at the initial visit, and 34.6% reported pain intensity of 4 or greater 3-months following the ED visit. Thus, at 3-months, over a third of the participants in the no recovery group still reported moderate to severe pain, compared to no participants meeting this criterion in either of the other two trajectory groups.

Group Differences in PTSD Symptoms, Resilience and Vulnerability Factors

Table 1 also shows the initial visit characteristics for each of the three trajectory groups as well as ANOVA comparisons across groups for the vulnerability and resilience factors assessed at the detailed initial assessment visit conducted within two weeks of the ED visit. The omnibus F-tests for the PCL, PROMIS anger, anxiety, and depressive symptoms, sleep quality, social undermining and social support were all significant [F’s (2,372) > 3.7; p’s < 0.05; η2’s > 0.02]. Pairwise comparisons revealed that the no recovery group differed significantly from the early recovery group on every measure such that the no recovery group reported greater PTSD symptoms, anger, anxiety, depressive symptoms and social undermining [t’s (253) > 2.84; p’s < 0.005] than the early recovery group. The no recovery group also reported less optimism, sleep quality, and social support [t’s (253) > 2.35; p’s < 0.02] than the early recovery group. Pairwise comparisons further showed that the no recovery group differed significantly from the delayed recovery group on some measures. Namely, the no recovery group reported greater initial PTSD symptoms and anger [t’s (280) > 4.0; p’s < 0.005] than the delayed recovery group. The no recovery group also reported less sleep quality and social support [t’s (280) > 2.35; p’s < 0.03] than the delayed recovery group. It should also be noted that the delayed recovery group reported significantly greater anger, anxiety and depressive symptoms than the early recovery group [t’s (280) > 2.0; p’s < 0.05], but the two groups did not differ significantly on PTSD symptoms.

In sum, the group of women who maintained elevated levels of pain intensity three months after presenting to the ED for acute pain was distinguished from other women, who recovered relatively quickly, by high levels of vulnerability factors, including PTSD symptoms, and low levels of resilience factors at the initial visit.

Unique and Common Sources of Variance

The no recovery group differed significantly from both the early recovery and delayed recovery groups on PTSD symptoms in the past month; results which could implicate this factor as a primary characteristic that distinguishes women who report persistent elevated pain from those who do not. However, this pattern of findings was also evident for anger, sleep quality and social support. These results could indicate that these latter factors are also core characteristics of the group that showed no recovery. To explore this issue, we first generated zero-order correlations among the four measures. All four variables were significantly correlated in the expected directions (r’s ranging from −0.24 to 0.55; p’s < 0.001), with anger showing consistently high correlations with the other measures (absolute value of r’s > 0.43). Results hint that all variables seemingly predictive of pain trajectories in the analyses above may distinguish the three trajectory groups from each other in a mostly overlapping fashion. To investigate this issue further, we computed a new group variable that combined early recovery and delayed recovery groups, and left the no recovery group separate. This coding scheme permitted comparisons of women who recover from acute pain from those who do not. We then conducted a simultaneous regression with PTSD symptoms, sleep quality, anger and social support as predictor variables, and the recoded Group variable as the criterion. Despite showing significant individual correlations with the recoded Group variable in the expected directions (r’s ranging from −0.20 to 0.23; p’s < .001), none of the semi-partial correlations for the predictor variables – reflecting their unique effects -- were significant (sr’s ranging from −0.09 to 0.08). Results suggest that the four intra- and interpersonal vulnerability factors did not individually distinguish women who recovered from acute pain from women who did not. Instead, variance shared by these four factors (PTSD symptoms, anger, sleep quality, and social support) distinguished women who recover in terms of acute pain and those who report persistent pain 3 months after their ED visit.

Discussion

We followed 375 women for three months who initially presented to an ED for abdominal/pelvic, chest, low back or shoulder/neck pain. Latent class growth analyses identified three groups of participants defined by different trajectories of recovery in pain intensity over the three months following their ED visit. One group recovered to an essentially zero pain rating by the time of the initial interview, whereas a second group recovered similarly but only after one month following the initial interview. Finally, a third group failed to recover completely even after three months following their ED visit. Indeed, 34.6% of these latter women reported moderate to severe pain intensity (i.e., 4/10 or greater) at the 3-month assessment, whereas none of the women in the other two groups met this pain intensity criterion at the 3-month assessment.

Differences between groups on initial pain intensity during the ED visit could have accounted for the different trajectories. If the no recovery group reported significantly greater pain than other groups at the ED visit, then the no recovery group’s elevated pain after three months could be attributed to them taking longer to recover from initially greater pain. Findings suggested otherwise. None of the three groups differed significantly on their ratings of pain intensity during their visits to the ED. Differences emerged only after a month had passed, indicating that the three groups were evincing distinct recovery patterns following essentially similar levels of initial acute pain in the ED.

Demographic factors could also have accounted for the different trajectories. Factors such as low education attainment, unemployment, low income, too few resources and race/ethnicity have been associated with higher levels of acute and chronic pain severity [18, 19, 22]. However, the three trajectory groups did not differ significantly on any demographic characteristics. Finally, differences in the frequency of specific pain sites across groups could also have accounted for the different trajectories in recovery from acute pain. Again, however, the three groups did not differ significantly on the frequencies of the various pain sites, suggesting that acute pain location did not play a critical role in promoting the transition from acute to persistent pain.

Our results did suggest that psychosocial factors distinguished the three trajectory groups. The no recovery group differed dramatically from the early recovery group on a wide range of negative affect, sleep and interpersonal factors at initial evaluation. Our hypotheses focused on the presence of PTSD symptoms at the time of an acute pain episode as a key contributor to the development of persistent pain, and results support this supposition in that the no recovery group had the highest level of PTSD symptoms. Thus, women with elevated PTSD symptoms while experiencing acute pain severe enough to prompt a visit to the ED may be at increased risk of subsequently developing persistent pain at the original acute pain site. However, elevated general negative affect, poor sleep quality and low social support also distinguished the no recovery from the early recovery group, and thus these factors may also confer increased risk of developing persistent pain.

The delayed recovery group also differed from the early recovery group on anxiety, anger and depressive symptoms, but not on PTSD symptoms. Those prone to experience a delayed recovery from acute pain may be characterized by elevated negative affect, but not necessarily by the broad array of problematic factors that characterized the no recovery group. The delayed recovery group also differed from the no recovery group. The latter group was distinguished from the former by even greater anger and sleep dysfunction, low social support and elevated PTSD symptoms. Each of these factors has been shown to be associated with acute and chronic pain in cross-sectional, observational, laboratory, and daily diary studies [11, 27, 36, 38, 45, 48, 55, 58]. To our knowledge, however, these are the first published findings suggesting that an array of anger, sleep, social support and PTSD factors partly define a group of women who appear to be at increased risk of developing persistent pain following an episode of diffuse acute pain.

Results of other analyses suggest that PTSD symptoms, anger, sleep quality and social support may constitute a common substrate of elevated risk. None of these factors predicted a significant portion of unique variance in group membership (i.e., early and delayed recovery groups vs no recovery group). Instead, PTSD symptoms, anger, sleep quality and social support predicted group membership in common. Women suffering from PTSD symptoms, who were also angry, experiencing sleep difficulties and were without adequate support from friends and family did not fully recover from an acute pain episode, whereas people suffering only from high negative affect were more likely to fully recover. It is the range of problems traversing both intra-personal and inter-personal functioning that conferred an elevated risk for maintaining persistent pain out to three months post-ED visit.

Our sample was predominantly African American and reported low-SES, and thus our results speak at least indirectly to racial/ethnic health disparities. First, the high level of PTSD symptoms reported by the sample as a whole underscores the physical and psychological risks posed by living in an inner-city environment. Recall that 44% of the sample may have met criteria for a diagnosis of PTSD based on the clinical cut-off score of the PCL-5. On its own, this pronounced rate of PTSD suggests the operation of social generators of a marked prevalence of mental health problems that are disparate across racial and SES lines. Second, given the role PTSD symptoms may play in fostering the maintenance of persistent pain – per our results – more people with PTSD symptoms in a certain geographical area may translate into more people developing persistent pain. The no recovery group comprised 43.2% of the inner-city and low-SES sample, which appears to be a high rate of transition from acute to persistent pain. We can envision a vicious cycle wherein high rates of PTSD, that are inadequately addressed by mental health services available in low-SES areas, contribute to the development of persistent pain following an episode of acute pain; a phenomenon that is then accompanied by inadequate physical health care and pain management after the initial ED visit.

Some weaknesses of this work should be delineated. First, the composition of the three groups, particularly the emergence of the no recovery group, may be a unique feature of an exclusively female, low SES, predominantly African American sample. The three groups describing distinct trajectories of recovery from acute pain may not exist in a similar fashion in other populations. Again, the high prevalence of PTSD symptoms reported by this sample may have excessive influences not only on group composition in general, but on the large proportion of women in the no recovery group. Second, we did not have information on more specific pain characteristics (e.g., neuropathic, nociceptive) or pain origin. The type and origin of the acute pain presented in the ED may have affected the transition from acute to persistent pain, and thus may have influenced the composition of the recovery groups. Third, initial visit questionnaire responses may have been affected by acute pain intensity. This may especially be the case for vulnerability factors such as negative affect and PTSD symptoms. It is possible that the early recovery group differed from the no recovery group on these vulnerability factors because they had virtually no acute pain at the initial detailed assessment visit. However, the no recovery group was also distinguished by low social support, elevated social undermining and low optimism, which are factors less susceptible to episodes of acute pain.

Clinical implications of our findings revolve around at least two issues. The first issue relates to the possibility that the emergence of persistent pain following an acute pain episode may be predicted by a number of psychosocial difficulties, including elevated PTSD symptoms. “Downstream” interventions in ED settings are needed to address high rates of co-occurring mental health and pain management needs. By three-months following a visit to the ED, patient complaints of persistent pain may signal that mental health and pain management issues have become intertwined, and perhaps would be best treated together as a co-morbid syndrome. Second, our findings suggest that the profile of psychosocial difficulties displayed by the no recovery group may have antedated the pain complaint that prompted their visit to the ED, and its subsequent persistence. “Upstream” assessment of psychosocial factors in the ED could reveal, for individual patients, a profile of difficulties that increase the likelihood of not recovering fully from acute pain. Prophylactic interventions aimed at ameliorating these underlying difficulties – such as elevated PTSD symptoms – could reduce the likelihood of the transition from acute to persistent pain.

In sum, our results suggest that women with high levels of PTSD symptoms who experience an acute pain episode serious enough to prompt an ED visit may maintain persistently elevated pain at this pain site for at least three months. However, elevated PTSD symptoms were not alone in characterizing this group. High levels of anger, poor sleep quality and low social support were equal partners in this cluster. Because the sample was derived from a low-SES inner city catchment area, we could infer that the predictors of persistent pain we identified are part and parcel of a particular environment. Crime, noise, crowding, and inadequate resources to meet basic needs could all contribute to a person developing the profile we identified. Although we may propose individually-oriented upstream and downstream interventions to help people with the psychosocial difficulties described by the no recovery group profile, the SES factors that may partly generate and perpetuate these difficulties will be much harder to combat.

Acknowledgements

This work was supported by NIH grant number R01DA039522.

Footnotes

Dr. Burgess serves on the scientific advisory boards for Natrol, LLC and MovingMindz, Pty, Ltd. All other authors have no conflicts of interest to report.

References

  • [1].Akhtar E, Ballew AT, Orr WN, Mayorga A, & Khan TW. The prevalence of post-traumatic stress disorder symptoms in chronic pain patients in a tertiary care setting: a cross-sectional study. Psychosomatics 2019; 60: 255–262. [DOI] [PubMed] [Google Scholar]
  • [2].Amir M, Kaplan Z, Neumann L, Sharabani R, Shani N, & Buskila D. Posttraumatic Stress Disorder, Tenderness and Fibromyalgia. Journal of Psychosomatic Research 1997; 42: 607–613. [DOI] [PubMed] [Google Scholar]
  • [3].Ann NM & Nagin DS. Group-based trajectory modeling: An overview 2014; 65: 205–210. [DOI] [PubMed] [Google Scholar]
  • [4].Arnold LM, Hudson JI, Keck PE, Auchenbach MB, Javaras KN, & Hess EV. Comorbidity of Fibromyalgia and Psychiatric Disorders. Journal of Clinical Psychiatry 2006; 67:1219–1225. [DOI] [PubMed] [Google Scholar]
  • [5].Beck JG & Clapp JD. A different kind of co-morbidity: understanding posttraumatic stress disorder and chronic pain. Psychological Trauma 2011; 3: 101–108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Bilevicius E, Sommer JL, Asmundson GJ, & El-Gabalawy R. Posttraumatic Stress Disorder and Chronic Pain are Associated with Opioid Use Disorder: results from a 2012–2013 American nationally representative survey. Drug and Alcohol Dependence 2018; 188: 119–125. [DOI] [PubMed] [Google Scholar]
  • [7].Blevins CA, Weathers FW, Davis MT, Witte TK, & Domino JL. The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation. Journal of Traumatic Stress 2015; 28: 489–498. [DOI] [PubMed] [Google Scholar]
  • [8].Bobby L, Jones B, Daniel S, Nagin D, & Roeder K. SAS Procedure based on Mixture Models for Estimating Developmental Trajectories. Sociological Methods & Research 2001; 29: 374–393. [Google Scholar]
  • [9].Bosco MA, Gallinati JL, & Clark ME. Conceptualizing and Treating Comorbid Chronic Pain and PTSD. Pain Research & Treatment 2013; 2013:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Burns JW, Bruehl S, Chung OY, Magid E, Chont M, Goodlad JK, Gilliam W, Matsuura J, & Somar K. Endogenous Opioids May Buffer Effects of Anger Arousal on Sensitivity to Subsequent Pain. Pain 2009; 146: 276–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Burns JW, Gerhart JI, Bruehl S, Peterson KM, Smith DA, Porter LS, Schuster E, Kinner E, Buvanendran A, Fras AM, & Keefe FJ. Anger Arousal and Behavioral Anger Regulation in Everyday Life among Patients with Chronic Low Back Pain: Relationships to patient pain and function. Health Psychology 2015; 34: 547–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Burns JW, Quartana PJ, Gilliam W, Matsuura J, Nappi C, & Wolfe B. Suppression of Anger and Subsequent Pain Intensity and Behavior Among Chronic Low Back Pain Patients: The role of symptom-specific physiological reactivity. Journal of Behavioral Medicine 2012; 35:103–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Bruehl S, Chung OY, Burns JW, & Biridepalli S. The Association between Anger Expression and Chronic Pain Intensity: Evidence for partial mediation by endogenous opioid dysfunction. Pain 2003; 106: 317–24. [DOI] [PubMed] [Google Scholar]
  • [14].Carver CS, Scheier MF, & Segerstrom SC. Optimism. Clinical Psychology Review 2010; 30: 879–889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Catalano SC. Intimate Partner Violence in the United States. Washington, D.C.: U.S. Department of Justice, Bureau of Justice Statistics; 2006. [Google Scholar]
  • [16].Childs JD, Piva SR, Fritz JM. Responsiveness of the Numeric Pain Rating Scale in Patients with Low Back Pain. Spine 2005; 30: 1331–4. [DOI] [PubMed] [Google Scholar]
  • [17].Darrell J, Gaskin, & Patrick R. The Economic Costs of Pain in the United States. The Journal of Pain 2012; 13: 715–724. [DOI] [PubMed] [Google Scholar]
  • [18].Day MA, & Thorn BE. The Relationship of Demographic and Psychosocial Variables to Pain-related Outcomes in a Rural Chronic Pain Population. Pain 2010; 15:467–474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Dorner TE, Muckenhuber J, Stronegger WJ, Ràsky É, Gustorff B, & Freidl W. The impact of socio-economic status on pain and the perception of disability due to pain. European journal of pain 2011; 15: 103–109. [DOI] [PubMed] [Google Scholar]
  • [20].Downie AS, Hancock MJ, Rzewuska M, Williams CM, Lin CW, & Maher CG. Trajectories of Acute Low Back Pain: A latent class growth analysis. Pain 2016; 157: 225–34. [DOI] [PubMed] [Google Scholar]
  • [21].Dunn KM, Jordan K, & Croft PR. Characterizing the Course of Low Back Pain: A latent class analysis. American Journal of Epidemiology 2006; 163:754–761. [DOI] [PubMed] [Google Scholar]
  • [22].Feldman CH, Dong Y, Katz JN, Donnell-Fink LA, & Losina E. Association between Socioeconomic Status and Pain, Function and Pain Catastrophizing at Presentation for Total Knee Arthroplasty. BMC Musculoskeletal Disorders 2015; 16:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Gerhart JI, Burns JW, Post KM, Smith DA, Porter LS, Burgess HJ, Schuster E, Buvanendran A, Fras AM, & Keefe FJ. Relationships between Sleep Quality and Pain-Related Factors for People with Chronic Low Back Pain: Tests of reciprocal and time of day effects. Annals of Behavioral Medicine 2017; 51: 365–375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Gibson CA. Review of Posttraumatic Stress Disorder and Chronic Pain: The path to integrated care. Journal of Rehabilitation Research & Development 2012; 49:753–776. [DOI] [PubMed] [Google Scholar]
  • [25].Glaesmer H, Rief W, Martin A, Mewes R, Brähler E, Zenger M, & Hinz A. Psychometric Properties and Population-based Norms of the Life Orientation Test Revised (LOT-R). British Journal of Health Psychology 2012; 17: 432–445. [DOI] [PubMed] [Google Scholar]
  • [26].Green CR, Anderson KO, Baker TA, Campbell LC, Decker S, Fillingim RB, Kalauukalani DA, Lasch KE, Myers C, Tait RC, Todd KH, & Vallerand AH. The Unequal Burden of Pain: Confronting Racial and Ethnic Disparities in Pain. Pain Medecine 2003; 4:277–294. [DOI] [PubMed] [Google Scholar]
  • [27].Holtzman S, Newth S, & Delongis A. The Role of Social Support in Coping with Daily Pain among Patients with Rheumatoid Arthritis. Journal of Health Psychology 2004; 9: 677–695. [DOI] [PubMed] [Google Scholar]
  • [28].Hooley JM & Teasdale JD. Predictors of Relapse in Unipolar Depressives: Expressed emotion, marital distress, and perceived criticism. Journal of Abnormal Psychology 1989; 98: 229. [DOI] [PubMed] [Google Scholar]
  • [29].Hooper L, Stockton P, Krupnick J, & Green B. Development, use, and psychometric properties of the Trauma History Questionnaire. Journal of Loss and Trauma 2011; 16: 258–283. [Google Scholar]
  • [30].Institute of Medicine of the National Academies. Relieving Pain in America: A Blueprint for Transforming Prevention, Care, Education, and Research. National Academy of Sciences; 2011. [PubMed] [Google Scholar]
  • [31].Jenewein J, Hanspeter M, Lutz W, Büchi S, Kraemer B, & Schnyder U. Development of Chronic Pain Following Severe Accidental Injury. Results of a 3-year follow-up study. Journal of Psychosomatic Research 2009; 66: 119–26. [DOI] [PubMed] [Google Scholar]
  • [32].Jenewein J, Lutz W, Moergeli H, Creutzig J, & Schnyder U. Mutual Influence of Posttraumatic Stress Disorder Symptoms and Chronic Pain Among Injured Accident Survivors: A Longitudinal Study. Journal of Traumatic Stress 2009; 22: 540–8. [DOI] [PubMed] [Google Scholar]
  • [33].Kroenke K, Zhangsheng Y, Jingwei W, Jacob K, Patrick OM. Operating Characteristics of PROMIS Four-Item Depression and Anxiety Scales in Primary Care Patients with Chronic Pain . Pain Medicine (2014); 15: 1892–1901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].Lowe SR, Galea S, Uddin M, & Koenen KC. Trajectories of posttraumatic stress among urban residents. American Journal of Community Psychology 2014; 53: 159–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Lykkegaard SR, Bjarke HV, Cardel T, & Andersen TE. The Role of Posttraumatic Stress Symptoms on Chronic Pain Outcomes in Chronic Pain Patients Referred to Rehabilitation. Journal of Pain Research 2018; 11: 527–536. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].McDermott KA, Smith HL, Matheny NL, & Cougle JR. Pain and Multiple Facets of Anger and Hostility in a Sample Seeking Treatment for Problematic Anger. Psychiatry Research 2017; 253: 311–317. [DOI] [PubMed] [Google Scholar]
  • [37].McKernan LC, Johnson BN, Crofford LJ, Lumley MA, Bruehl S, & Cheavens JS. Posttraumatic Stress Symptoms Mediate the Effects of Trauma Exposure on Clinical Indicators of Central Sensitization in Patients With Chronic Pain. The Clinical journal of pain 2019; 35: 385–393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [38].McKillop AB, Carroll LJ, Jones CA, & Battie MC. The Relation of Social Support and Depression in Patients with Chronic Low Back Pain. Disability and Rehabilitation 2017; 39: 1482–1488. [DOI] [PubMed] [Google Scholar]
  • [39].McLean SA, Soward AC, Ballina LE, Rossi C, Rotolo S, Wheeler R, Foley KA, Batts J, Casto T, Colette R, Holbrook D, Goodman E, Rauch SAM, & Liberzon I. Acute Severe Pain is a Common Consequence of Sexual Assault. Journal Pain 2012; 13:736–741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Meltzer-Brody S, & Leserman J. Psychiatric Comorbidity In Women with Chronic Pelvic Pain. CNS Spectrums 2011; 16:29–35. [DOI] [PubMed] [Google Scholar]
  • [41].Moeller-Bertram T, Keltner J, & Strigo IA. Pain and posttraumatic stress disorder - review of clinical and experimental evidence. Neuropharmacology 2012; 62: 586–597. [DOI] [PubMed] [Google Scholar]
  • [42].Morasco BJ, Lovejoy TI, Lu M, Turk DC, Lewis L, & Dobscha SK. The Relationship between PTSD and Chronic pain: Mediating role of coping strategies and depression. Pain 2013; 2013; 154: 609–616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [43].National Institute of Mental Health. Post-Traumatic Stress Disorder (PTSD) 2017. Available at www.nimh.nih.gov.
  • [44].Pilkonis PA, Choi SW, Reise SP, Stover AM, Riley WT, Cella D. Item Banks for Measuring Emotional Distress from the Patient-Reported Outcomes Measurement Information System (PROMIS®): Depression, anxiety, and anger. Assessment 2011; 18: 263–283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [45].Roberts AL, Gilman SE, Breslau J, Breslau N, & Koenen KC. Race/ethnic differences in exposure to traumatic events, development of post-traumatic stress disorder, and treatment-seeking for post-traumatic stress disorder in the United States. Psychological Medicine 2010; 41: 71–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [46].Schalet BD, Pilkonis PA, Yu L, Dodds N, Johnston KL, Yount S, & Cella D. Clinical Validity of PROMIS Depression, Anxiety, and Anger across Diverse Clinical Samples. Journal of Clinical Epidemiology 2016; 73: 119–127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [47].Scheier MF, Carver CS, & Bridges MW. Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A re-evaluation of the Life Orientation Test. Journal of Personality and Social Psychology 1994; 67: 1063–1078. [DOI] [PubMed] [Google Scholar]
  • [48].Schumm JA, Briggs-Phillips M, & Hobfoll SE. Cumulative Interpersonal Traumas and Social Support as Risk and Resiliency Factors in Predicting PTSD and Depression among Inner-City Women. Journal of Traumatic Stress 2006; 19: 825–836. [DOI] [PubMed] [Google Scholar]
  • [49].Schur EA, Afari N, Furberg H, Olarte M, Goldberg J, Sullivan PF, & Buchwald D . Feeling Bad in More Ways than One: Comorbidity patterns of medically unexplained and psychiatric conditions. Journal of General Internal Medicine 2007; 22:818–821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [50].Shaw WS, Means-Christensen AJ, Slater MA, Webster JS, Patterson TL, Grant I, Garfin SR, Wahlgren DR, Patel S, Atkinson JH. Psychiatric Disorders and Risk of Transition to Chronicity in Men with First Onset Low Back Pain. Pain Medicine 2010; 11: 1391–1400 [DOI] [PubMed] [Google Scholar]
  • [51].Sibrava NJ, Bjornsson AS, Pérez Benítez ACI, Moitra E, Weisberg RB, Keller MB. Posttraumatic stress disorder in African American and Latinx adults: Clinical course and the role of racial and ethnic discrimination. American Psychologist 2019; 74:101–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [52].Siqveland J, Hussain A, Lindstrøm JC, Ruud T, & Hauff E. Prevalence of Posttraumatic Stress Disorder in Persons with Chronic Pain: A Meta-analysis. Front Psychiatry. 2017; 8:164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [53].Stampfel CC, Chapman DA, Alvarez AE. Intimate partner violence and posttraumatic stress disorder among high-risk women: does pregnancy matter? Violence Against Wom. 2010;16(4):426–443. [DOI] [PubMed] [Google Scholar]
  • [54].Sterling M, Hendrikz J, & Kenardy J. Similar factors predict disability and posttraumatic stress disorder trajectories after whiplash injury. Pain 2011; 152: 1272–1278. [DOI] [PubMed] [Google Scholar]
  • [55].Toledo TA, Hellman N, Lannon EW, Sturycz CA, Kuhn BL, Payne MF, Palit S, Güereca YM, Shadlow JO, & Rhudy JL. Anger Inhibition and Pain Modulation. Annals of Behavioral Medicine 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [56].Weathers FW, Litz BT, Keane TM, Palmieri PA, Marx BP, & Schnurr PP. The PTSD Checklist for DSM-5 (PCL-5) 2013. Scale available from the National Center for PTSD at www.ptsd.va.gov. [Google Scholar]
  • [57].Weiss RS. The Provisions of Social Relationships. Doing Unto Others 1974:17–26. [Google Scholar]
  • [58].Weiss NH, Schick MR, Contractor AA, & Dixon-Gordon KL. Posttraumatic Stress Disorder and Substance Use: Identifying the underlying role of difficulties regulating positive emotions. Addictive Behaviors 2019; 96: 119–126. [DOI] [PubMed] [Google Scholar]
  • [59].Yu L, Buysse DJ, Germain A, Moul DE, Stover A, Dodds NE, & Pilkonis PA. Development of Short Forms from the PROMIS™ Sleep Disturbance and Sleep-Related Impairment Item Banks. Behavioral Sleep Medicine 2011; 10: 6–24. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES