Native Americans (NAs) have a higher prevalence of chronic pain than other U.S. minorities and non-Hispanic Whites (NHWs).4,42,100 Unfortunately, there is a lack of research that attempts to understand the mechanisms for this pain disparity. Most studies have focused only on pain prevalence, finding that arthritis,34,57 back pain,4,27 headaches,72 and dental pain49 are among some of the chronic pain conditions that have the highest prevalence in NAs. Given that chronic pain results in tremendous health care utilization, economic burden, disability, and loss of productivity,51 the mechanisms contributing to this pain disparity must be understood.
Previously, we conducted a small pilot study of healthy, pain-free NAs and NHWs and found that NAs were less pain sensitive (eg, higher pain tolerance) and showed reduced central sensitization on a physiological marker of spinal nociception (eg, lower temporal summation of the nociceptive flexion reflex [TS-NFR]).65 These findings suggested that risk factors for chronic pain may be different for NAs than other minorities, because the opposite findings had been noted in African Americans and Hispanics.15,18,20,31,32,61,62,84,103,109,113 We argued that because pain is normally adaptive, chronically dampened pain in NAs (lower tolerances and reduced spinal cord excitability) may lead to difficulties detecting physical harm, protecting oneself from injuries, and learning to avoid future dangers.23,25,96 Moreover, chronic activation of descending inhibition could exhaust the capacity to inhibit pain, perhaps through disruption of the opioid system.8–10,58
Conditioned pain modulation (CPM) is the traditional method to assess central nervous system (CNS) inhibition in the laboratory. Although we did not assess CPM in our pilot study, we did assess emotional modulation of pain and found that NAs were less able to facilitate (disinhibit) pain during negative emotions.64 We hypothesized that this might be due to chronic overactivation of CNS inhibition. Thus, when taken together, chronic pain risk in NAs could be due to problems detecting physical harm (low pain sensitivity) and/or overactivation of inhibitory mechanisms that reduce spinal excitability (but might ultimately deplete inhibitory resources).
The current study was designed to replicate and extend our small pilot study to a larger sample of NAs and NHWs using a wider array of assessment methods. Pain/nociception measurement tasks were chosen to assess differences in pain sensitivity (eg, pain threshold/tolerance), central sensitization (eg, TS-NFR and TS-pain), and CNS pain inhibition (eg, CPM). Thermal detection thresholds (ie, cool/warm thresholds) were also used to evaluate peripheral nociceptive fiber function. Outcomes included both subjective (eg, pain response) and physiological (eg, TS-NFR and CPM of NFR) targets. To rule out potential confounds, care was taken to ensure groups were similar on important background variables or statistically controlled. Moreover, only healthy, pain-free participants were recruited to help ensure that any observed differences in pain processing were not due to disparities in disease severity and/or pain treatment.21,59,97 Based on our pilot data, we hypothesized that relative to NHWs, NAs would exhibit: (1) reduced pain sensitivity, (2) reduced central sensitization, and (3) increased CNS inhibition.
2. Materials and methods
2.1.1. Sample size determination
Power analyses conducted in G*Power (version 22.214.171.124) were based on effect sizes derived from our pilot study of pain processing in NAs,65 as well as effect sizes for race/ethnic differences in CPM from Campbell et al.16 These analyses suggested that 120 per group (N = 240 total) would result in power of 0.80 for most of the outcomes used in the current study (power = 0.80 for effect sizes d ≥0.36) at alpha = 0.05. So, N = 240 was the targeted sample.
2.1.2. Recruitment and inclusion/exclusion criteria
These data were collected from the Oklahoma Study of Native American Pain Risk (OK-SNAP). Healthy, pain-free participants were recruited through newspaper ads, tribal newspapers, fliers, personal communications with NA groups, email announcements, and online platforms (e.g., Facebook). Potentially eligible participants were invited to attend the first laboratory testing day, which began with a thorough screening for inclusion/exclusion criteria. Participants were excluded for: (1) <18 year old, (2) history of cardiovascular, neuroendocrine, musculoskeletal, and neurological disorders, (3) chronic pain or current acute pain problems, (4) body mass index (BMI) ≥35 (due to difficulties recording electromyogram for NFR), (5) use of antidepressants, anxiolytic, analgesic, stimulant, or antihypertensive medication, (6) current psychotic symptoms (assessed by Psychosis Screening Questionnaire7) or substance use problems, and/or (7) an inability to read and speak English. Data collection occurred between March 2014 and October 2018. The study was approved by institutional review boards of The University of Tulsa, Cherokee Nation, and the Indian Health Service Oklahoma City Area Office. Native American status was verified from Certificate of Degree of Indian Blood or tribal membership cards. Participants were given an overview of all procedures and told they could withdraw at any time. All participants provided verbal and written informed consent before enrollment and received a $100 honorarium for the completion of each testing day (or $10/hour for noncompleted days).
2.1.3. Final sample
A total of 391 participants were screened for the study, but 62 did not meet inclusion criteria. Of the 329 who were eligible to participate, 247 completed both testing days, 41 completed only one day, and 39 completed part of one testing day. Two participants' data were lost due to a computer malfunction. Twenty-two participants were non-NA minorities and thus were excluded from the current analyses. Thus, the final sample of participants who could contribute at least partial data was 155 NA (64 males) and 150 NHW (76 males). Table 1 provides characteristics of these participants. Native American participants in the current study represent tribal nations predominately from the southern plains and eastern Oklahoma tribes.
2.2. General overview of procedures/testing
Testing was conducted over a 2-day period, each lasting 4 to 6 hours (Fig. 1). Informed consent and inclusion/exclusion screening were conducted on the first day, followed by a brief semistructured interview about the meaning of pain (data presented elsewhere). On one of the testing days (referred to here as Pain Sensitivity Day), the following tests were administered: cool and warm thresholds, heat pain threshold/tolerance, mechanical pressure pain threshold, stimulus response to heat, temporal summation of heat pain, cold-pressor pain threshold/tolerance, and ischemia pain threshold/tolerance. On the other testing day (referred to here as Pain Modulation Day), the following tests were administered: NFR threshold, Pain30 threshold (if necessary), temporal summation of NFR threshold, 52°C heat pulse series, single electric stimulations, temporal summation of NFR and electric pain, sham CPM (no conditioning stimulus [CS] presented, see description below), CPM (which includes 2-minute 10°C cold water CS), and emotional controls of nociception (data not presented). Order of testing day was counterbalanced, blocking for race and sex. Moreover, tests within each day were partly randomized (Fig. 1). Breaks were provided between tasks to minimize carryover. Several questionnaires were administered on each testing day at the beginning and during breaks to assess background characteristics and control variables.
2.3. Background questionnaires and control variables
These questionnaires were administered to assess eligibility (eg, demographic and health status) and to examine whether groups differed on important background characteristics.
2.3.1. Demographics and health exclusion
A custom-built demographic and health status questionnaire was used to obtain standard background information, as well as information regarding health problems. It was administered immediately after obtaining informed consent. The questionnaire asked about: age, sex, marital status, education level, employment, and income level, as well as potential exclusionary criteria such as cardiovascular problems, neurological problems, chronic pain, and medication use. Weight and height were assessed from a medical scale to calculate BMI.
2.3.2. Blood pressure and resting heart rate
A medical-grade device (Dinamap; Tampa, FL) was used to measure resting systolic blood pressure (BP), diastolic BP, mean arterial pressure (MAP), and heart rate (HR) 3 times at the beginning of each testing session (3 minutes intertest interval) while the participants sat comfortably in the recliner with their arm resting on the arm of the chair. The average of each variable from the first testing day is reported.
2.3.3. Traditional (dispositional) pain catastrophizing
The 13-item Pain Catastrophizing Scale93 was used to assess dispositional catastrophizing at the beginning of testing to assess how each participant generally reacts to painful events. This was done to establish whether groups generally differed in pain catastrophizing. Scores ranged from 0 to 52, with higher scores indicating greater catastrophic thoughts.
2.3.4. Current affect
Positive affect and negative affect before testing were assessed from the Positive and Negative Affect Schedule.106 Each positive and negative affect subscale consists of 10 items that measure positive and negative emotions, with subscales ranging from 10 to 50. Higher scores on each scale indicate greater positive or negative affect.
2.3.5. State anxiety
State anxiety before testing was assessed by the 20-item state anxiety subscale of the State Trait Anxiety Inventory.87 The subscale ranges from 20 to 80, with higher scores indicating greater current anxiety.
2.3.6. Sleep quality
The perceived sleep quality item from the Pittsburgh Sleep Quality Index was used to assess differences in sleep.13 The item ranges from 0 (very good) to 3 (very bad), where higher scores represent poorer sleep.
2.3.7. Perceived stress
The Perceived Stress Scale is a 10-item measure that assesses psychological stress within the past month. Scores range from 0 to 40, with higher scores indicating more perceived stress.22
2.3.8. Psychological distress
The Symptom Checklist-90-Revised (SCL-90-R) was used to assess general psychological functioning.26 The scale consists of 90 items that assess various psychological symptoms (eg, somatization, obsessive-compulsive, depression, phobic anxiety, and paranoia). The Global Severity Index of the SCL-90-R was used to assess overall psychological distress. Total Global Severity Index scores range from 0 to 4, with higher scores indicating greater distress.
2.3.9. Health perceptions
The General Health Perceptions subscale of the Medical Outcomes Study 36-item Short Form Health Survey (SF-36)105 was used to assess perceptions of physical health. The Bodily Pain subscale was used to assess the degree of pain and interference due to pain that has occurred over the past 4 weeks (before the first testing day). For both scales, scores are standardized to range from 0 to 100, with higher scores indicating better physical functioning.
2.4. Testing environment
Questionnaires were presented by a computer with dual-monitor capacity. Custom-built LabVIEW software program (National Instruments, Austin, TX) was used to control timing and order of the experimental protocol. One computer monitor was used by the experimenter to monitor experimental timing and physiological signals, whereas the second monitor was used by the participant to complete electronic questionnaires and to make ratings of stimuli. To reduce experimenter influences on testing, all reported experimental outcomes (except pressure pain) were conducted with participants in a sound-attenuated and electrically shielded testing room while experimenters monitored the testing from an adjacent control room using a video camera (with a microphone) connected to a flat-panel monitor. Participants wore a pair of sound-attenuating headphones that allowed them to hear the experimenter and the computer-recorded instructions for each task.
During all tasks, participants were seated in a comfortable reclining chair (Perfect Chair Zero Gravity Recliner, Human Touch, Long Beach, CA) that kept their leg and hip angles constant for NFR testing. All heat stimuli were delivered using a Medoc (Haifa, Israel) Pathway device with a Contact Heat Evoked Potential Stimulator (CHEPS) thermode. The maximum intensity of any heat stimulus was set to 52°C (51°C for heat tolerance).
A circulating water bath (Thermo Fisher Scientific, Pittsburgh, PA) was used to assess cold pain threshold/tolerance and as the painful CS in the CPM task. Participants were asked to submerge their hand and forearm into the water bath and keep their fingers spread apart. The water level was kept constant (6″ deep) across all participants to keep procedures standardized and maintain similar cold exposure to participants' hands/forearms.
To measure responses to mechanical pressure, a Medoc-Wagner computerized algometer was used (Haifa, Israel) and all experimenters were trained to use the software to maintain a constant rate of pressure increase while assessing pain thresholds.
Electric stimuli were delivered by attaching a bipolar electrode (Nicolet; 30 mm inter-electrode distance) filled with a conductive gel (EC60; Grass Technologies, West Warwick, RI) over the retromalleolar surface of the left ankle after the skin had been cleaned with alcohol and abraded with an exfoliating cream (NuPrep; Weaver and Company, Aurora, CO). Stimulations were delivered by an isolated, constant current stimulator (Digitimer DS7A; Hertfordshire, United Kingdom). Each stimulus consisted of a train of five 1-millisecond rectangular wave pulses with a 3-millisecond interpulse interval (250 Hz); however, the train was always experienced as a single stimulation. The maximum stimulation intensity was set at 50 mA to ensure safety.
2.5. Measurement of peripheral fibers
Cool detection thresholds were used to assess Aδ-fiber function,86,114 whereas warm detection thresholds were used to assess C-fiber function.66,85,101,102 For both tasks, the CHEPS thermode was attached to the participants' left volar forearm using a Velcro strap and 5 trials were administered (1 practice and 4 trials averaged). Cool detection started at a baseline of 30°C and decreased at 0.5°C/s until the participants indicated that they perceived a cool sensation by making a button press. Warm detection was similar, except that the temperature started at 30°C and increased at 0.5°C/s until the participants indicated that they perceived a warm sensation. There was a random (25-35 seconds) intertrial interval for both procedures.
2.6. Measures of pain sensitivity
Before pain sensitivity tests, participants were instructed on the difference between pain threshold and pain tolerance. Pain sensitivity was assessed from multiple stimulus modalities representing cutaneous thermal pain (heat pain threshold/tolerance, stimulus–response curve to heat, heat pain45, cold-pressor threshold/tolerance, 52°C heat series, 2-minutes 10°C cold water), deep tissue mechanical (pressure pain thresholds) and ischemic (ischemia pain threshold/tolerance) pain, as well as electric pain threshold/tolerance.
2.6.1. Heat pain threshold/tolerance
To assess heat thresholds and tolerances, the CHEPS thermode was attached to the participants' left volar forearm using a Velcro strap and heat pain threshold/tolerance was each assessed 5 times (1 practice and 4 averaged trials).15 Each trial started from a baseline of 32°C and heated at a rate of 0.5°C/s until the participant terminated the stimulus by pushing a button as soon as the heat became painful (threshold) or intolerable (tolerance). There was a 25 to 35 seconds' randomly determined intertrial interval. In between tolerance trials, the thermode was moved slightly to avoid sensitization before the intertrial interval was started.
2.6.2. Heat stimulus–response curve
To assess responses to nonpainful and painful heat, 11 different heat pulses (40°C-50°C) were delivered to the left volar forearm. Each pulse started at a 35°C baseline, reached the peak temperature in 0.5 seconds, held the peak temperature for 0.5 seconds, and then returned to baseline in 0.5 seconds. After each pulse, participants made a rating using a computerized numerical rating scale (NRS) with the following labels: 0 (no sensation), 10 (warm), 20 (a barely painful sensation), 30 (very weak pain), 40 (weak pain), 50 (moderate pain), 60 (slightly strong pain), 70 (strong pain), 80 (very strong pain), 90 (nearly intolerable pain), and 100 (intolerable pain). After the rating was made, there was a 20 to 30 seconds' random intertrial interval. The order of the 11 temperatures was randomized. After all 11 temperatures were administered, there was a short break, the thermode was moved slightly, and then the 11 randomized pulses were delivered again using the same procedures. The 2 ratings of each temperature were averaged to generate the stimulus–response curve.
2.6.3. Heat Pain45
The 11 temperatures and the pain ratings from the stimulus–response curve described above were used in a regression equation to determine the temperature corresponding to a rating of 45 on the NRS (Heat Pain45). This individually calibrated temperature was used as a measure of pain sensitivity, but also as the suprathreshold stimulus intensity to assess temporal summation of heat pain (TS-Heat Pain) described below.
2.6.4. Cold-pressor pain threshold/tolerance
Participants were asked to submerge their hand and forearm into the circulating water bath that was held at 6 ± 0.1°C.15,60,73 During the task, participants made continuous ratings on a computer-presented visual analogue scale (VAS) ranging from “no pain” to “maximum tolerable pain.” These ratings were converted in real time to scores ranging between 0 and 100. The computer tracked the time following hand/forearm immersion. The time taken to reach a rating of 30 was defined as pain threshold, and the time to reach a rating of 100 was defined as pain tolerance. The maximum cold water exposure was set to 5 minutes, but the participant was not informed of the limit.
2.6.5. Pressure pain threshold
Pressure pain threshold was assessed at 3 body sites (masseter muscle, trapezius muscle, and thumbnail), with site order randomized.70 Pressure stimuli were applied by an experimenter by following the Medoc software to achieve an increase in pressure at a rate of 98 kPa/s. Participants were instructed to push a button as soon as they experienced pain. Up to 5 trials at each site were administered, and pressure pain threshold was defined as the average of the 3 best trials (determined by the accuracy of the experimenter's pressure rate increase).
2.6.6. Ischemia pain threshold/tolerance
A standard forearm tourniquet test was used to measure ischemia pain.36 First, participants used their nondominant hand to conduct hand exercises with a dynamometer (Lafayette Hand Dynamometer; Lafayette Instrument Company, Lafayette, IN) at 50% grip strength for 2 minutes (1×/second). Immediately after, the arm was raised for 15 seconds to allow for desanguination, and then a BP cuff was inflated to 220 mm Hg around the biceps to occlude blood flow to the forearm. Participants were instructed to keep their arm still during the task. During occlusion, participants continuously rated their pain on the same VAS used during the cold-pressor task. Similarly, the computer recorded the time taken to achieve a rating of 30 (pain threshold) and 100 (pain tolerance). Maximum ischemia exposure was 25 minutes, but the participant was not informed of the limit.
2.6.7. Electric pain tolerance
Electric pain tolerance was assessed using a single ascending staircase of stimulations that started at 0 mA and increased in 2-mA steps.75 After each stimulus, the participants rated their pain on the same VAS used during the cold pressor. The stimulation intensity was increased until the participant rated the stimulus as the maximum tolerable pain. The computer recorded the stimulation intensity (in mA) corresponding to a pain rating of 30 (pain threshold) and 100 (pain tolerance).
2.6.8. 52°C heat series
Six blocks of 5 heat pulses (random 8-12 seconds' interstimulus interval) were delivered to the left volar forearm. Each pulse started at a baseline of 32°C and increased to 52°C at a rate of 70°C/s and then returned to baseline at a rate of 40°C/s.39 Each 65-second block of 5 pulses was separated by a short (∼1 minute) break during which the experimenter moved the thermode slightly. The first 5 blocks were used to assess contact heat-evoked potentials from electroencephalogram (data to be presented elsewhere); so, participants had their eyes closed.39 The sixth block was used to assess pain sensitivity in response to the heat; therefore, immediately after each heat pulse, participants rated their pain on the same computerized VAS used during the cold-pressor task. The current study focused only on these pain intensity ratings as a measure of suprathreshold pain sensitivity.
2.6.9. Two-minute 10°C cold water
During the CPM task (described below), participants were asked to place their right hand in a 2-minute circulating cold water bath held at 10°C, after which they rated their cold pain using a computerized NRS (0 “no pain,” 20 “mild pain,” 40 “moderate pain,” 60 “severe pain,” 80 “very severe pain,” and 100 “worst possible pain”). Although its primary use was for CPM, it also serves as a measure of suprathreshold pain sensitivity.
2.7. Measures of central sensitization
2.7.1. Nociceptive flexion reflex threshold
Nociceptive flexion reflex is a polysynaptic, spinal reflex elicited in response to Aδ-fiber activation; so, it is used as a measure of spinal nociception.82 Nociceptive flexion reflex was measured from left biceps femoris electromyogram (EMG) using 2 Ag-AgCl electrodes positioned approximately 10 cm superior to the popliteal fossa. A common ground electrode was placed over the lateral epicondyle of the femur. Sensors were filled with conductive gel (EC60; Grass Technologies). The EMG signal was collected, filtered (10 Hz to 300 Hz), and amplified (×10,000) using a Grass Technologies (West Warwick, RI) Model 15LT amplifier (with AC Module 15A54). The skin was cleaned with alcohol and exfoliated (Nuprep gel; Weaver and Company, Aurora, CO) to achieve impedances <5 kΩ. EMG was sampled and digitized at 1000 Hz using a National Instruments analog-to-digital converter (Austin, TX).
Nociceptive flexion reflex threshold (the stimulus intensity necessary to elicit the reflex) was used as a static measure of spinal nociceptive sensitivity (lower thresholds = greater sensitization). Nociceptive flexion reflex threshold was assessed from 3 ascending–descending staircases of stimulations.74 The first staircase began at 0 mA and increased in 2-mA increments until a reflex was observed (defined as when the averaged, rectified EMG activity of the biceps femoris in the 90-150 ms poststimulus interval was 1.4 SD greater than the averaged, rectified biceps femoris EMG activity during the 60-ms prestimulus baseline interval).74 Once an NFR was obtained, the stimulus intensity decreased in 1-mA steps until a reflex was not observed. The second and third ascending–descending staircases used 1-mA step increments. The interval between electric stimuli varied randomly (8-12 seconds) to minimize predictability and reflex habituation. Nociceptive flexion reflex threshold was defined as the average stimulus intensity (mA) of the 2 peaks and 2 troughs of the last 2 ascending–descending staircases.
In the event that stimulations at NFR threshold were not experienced as painful, the stimulation intensity was increased from NFR threshold in 2-mA steps until a rating of 30 was achieved on the VAS (ie, Pain30; this was only used for stimulation intensity determination for CPM and thus not reported here).
2.7.2. Three-stimulation threshold
Three-stimulation threshold was defined as the stimulus intensity required to evoke an NFR on the last stimulus of a series of 3 electric stimulations. To assess this, a series of 3 electric stimulations were delivered with an interstimulus interval of 0.5 seconds (2.0 Hz). The series started at 0 mA and increased in 2-mA steps until an NFR was evoked by the third stimulus in the series. The stimulus intensity at which the NFR was evoked was used as 3-stimulation threshold. This task is akin to what has been described in the literature as temporal summation of NFR threshold54; however, its definition does not require summation of the reflex across the 3 stimulus series (reflexes could decrease across the 3 stimulations, although this rarely happens based on our experience).
2.7.3. Temporal summation of nociceptive flexion reflex
Temporal summation of NFR assesses the degree of spinal neuron excitability after a series of suprathreshold stimulations,94 and is believed to be a correlate of wind-up in animals.63,110,111 The suprathreshold stimulus intensity used for this task was set at 120% of NFR threshold or 120% of 3-stimulation threshold, whichever was higher (in mA). In this study, TS-NFR was defined as the degree of reflex summation after a series of 3 suprathreshold stimuli. To assess this, 5 trains of 3 suprathreshold stimuli (0.5 seconds ISI) were used to assess temporal summation. After each train of stimuli, participants were instructed to rate the pain intensity corresponding to each of the 3 stimulations, using a set of 3 computer-presented VASs ranging from “no pain sensation” to “the most intense pain sensation imaginable.” After the participant completed the ratings, there was an intertrain interval of 8 to 12 seconds. The baseline EMG in the 60 milliseconds before the third stimulus in the stimulus series was visually inspected in real time by the experimenter for excessive muscle tension or voluntary movement. If the mean rectified EMG exceeded 5 µV, the train was repeated to ensure that EMG activity in the poststimulus interval was not contaminated by muscle tension unrelated to the NFR.94 NFR magnitudes in response to each stimulus in the 3-stimulus train were calculated in d-units by first subtracting the 60-millisecond baseline before the first stimulus in each train from the EMG response 70 to 150 milliseconds after each stimulus in the train. This difference was then divided by the average of the standard deviations of the rectified EMG from these 2 intervals. Note: the poststimulus interval used here differs from the assessment of NFRs in response to single stimuli (ie, 90-150 ms poststimulus) because repeated stimulations with a short (0.5 seconds) interstimulus interval can result in a shorter NFR onset latency.94
2.7.4. Temporal summation of electric pain
Temporal summation of electric pain (TS-electric pain) is believed to be the perceptual correlate of wind-up in animals.90 For this study, TS-Electric Pain was defined in 2 ways. First, it was defined as the increase in pain across the 3-stimulus train described in the TS-NFR section above. However, this approach is limited because participants are asked to make ratings of all 3 stimuli after the series has been delivered because there is not enough time to rate stimuli in the 0.5-second interstimulus interval. This retrospective reporting could result in bias, particularly because the perception of the first 2 stimuli could be influenced by the more painful third stimulus. To overcome this problem, 5 single electric stimuli were delivered using the same suprathreshold intensity (8-12 seconds' interstimulus interval) and participants made pain ratings using the same VAS immediately after each stimulation. Thus, like Farrell and Gibson,33 TS-Electric Pain was also defined as the difference between ratings of these single stimuli and the rating of the third stimulus in the 3-stimulus train. Both definitions were used in analyses.
2.7.5. Temporal summation of heat pain
Temporal summation of heat pain was measured in response to 5 trains of 10 suprathreshold heat stimuli according to published procedures.88,89 The first train was a practice, and the remaining 4 were used in analyses. The stimulus intensity was individually calibrated to each participant (ie, Heat Pain45 described above). Before the TS-Heat Pain task, participants were instructed on the difference between first pain and second pain. During each train, the NRS was continuously displayed and participants were asked to call out a numerical rating after each heat pulse corresponding to their second pain, which was recorded by the experimenter. Heat pulses in the series had a pulse-to-pulse interstimulus interval of 2.5 seconds and each pulse started at a 35°C baseline, reached the peak temperature in 0.5 seconds, held the peak temperature for 0.5 seconds, and then returned to baseline in 0.5 seconds.88,89 However, after 24 participants were ran using these parameters, results suggested that almost no one was showing pain summation; so, the heat parameters were changed based on the recommendations of Kong et al.46 (baseline temperature changed from 35°C to 39°C, ramp up and ramp down speed changed from 0.5 to 0.37 seconds, peak temperature duration changed from 0.5 to 0.75 seconds; and pulse-to-pulse interval stayed at 2.5 seconds). However, this did not improve pain summation; thus, analysis of TS-Heat Pain ignored the difference in these 2 parameter variants after demonstrating no significant difference between the 2 (there were no differences in average pain ratings or summation of pain ratings based on the 2 stimulus parameters).
2.8. CNS inhibition: conditioned pain modulation
The CPM task was used to assess descending inhibition. In this study, CPM involved the assessment of pain and NFR in response to a test stimulus before, during, and after a tonic CS delivered at a distal body site from the test stimulus. In healthy humans, the CS should inhibit pain evoked by the test stimuli.43 The test stimuli were electrical stimulations (intensity = highest of 120% NFR threshold, 120% temporal summation threshold, or 100% electric Pain30) delivered to the left ankle, as well as 52°C heat pulses delivered to the left volar forearm (heat started at baseline of 32°C increased to 52°C at a rate of 70°C/s and then returned to baseline at a rate of 40°C/s). The CS was exposure to a 2-minute long circulating cold-water bath maintained at a temperature of 10 ± 0.1°C.
Conditioned pain modulation consisted of three 2-minute phases: baseline (test stimuli delivered prior to cold water), conditioning (test stimuli delivered while hand/arm is submerged in cold water), and posttest (test stimuli delivered after conditioning). A 2-minute rest occurred between baseline and conditioning, and a 5-minute rest occurred between conditioning and posttest. During conditioning, the participants were instructed to submerge their right hand up to their forearm in the painfully cold water and to keep their hand palm down with fingers spread. Each 2-minute phase consisted of a 20-second wait period, followed by 5 electric test stimuli (random 8-12 seconds' interstimulus interval), and then five 52°C heat pulses (random 4-8 seconds' interstimulus interval). Participants provided pain ratings in response to the electric stimulations verbally using a NRS that was constantly displayed (0 “no pain,” 20 “mild pain,” 40 “moderate pain,” 60 “severe pain,” 80 “very severe pain,” and 100 “worst possible pain”), and an experimenter recorded the ratings. Pain ratings of heat stimuli were made using the same NRS immediately after the last heat pulse (this was done so that making ratings of heat pulses would not interfere with physiological responses; physiological data not presented). After the conditioning phase, participants used the NRS to rate the pain evoked by the cold water.
Nociceptive flexion reflex magnitudes in response to electric stimuli were used to assess within-subject changes in spinal nociception.75 Nociceptive flexion reflex magnitudes were calculated as a d-score (NFR d = [mean rectified EMG of 90-150 ms poststimulation interval minus mean of rectified EMG from −60 to 0 ms prestimulation interval] divided by the average standard deviation of the rectified EMG from the 2 intervals). Trials with NFR baselines higher than 3.0 μV were excluded from analyses due to excessive muscle tension and/or noise in the recording (3% of trials were excluded).
A “sham” CPM task was also used as a control condition. Sham CPM was identical to the real CPM task with the exception that a 26°C nonpainful water bath was used. The order of CPM and sham CPM was randomized and there was a 15-minute break between the 2 to avoid carryover effects.
2.9. Data analysis
2.9.1. Data screening
Before analyses, variable distributions were screened using boxplots, histograms, and normality statistics. Those that were skewed were log- or square-transformed to reduce positive skew and negative skew, respectively. If a variable was transformed, this is noted in the Results section and/or Tables, and the reverse-transformed mean values in original units are depicted in Figures. Next, outliers were identified using Wilcox's mean absolute deviation-median procedure (using the recommended 2.24 cutoff) and then winsorized by replacing the outlier value with the next nearest nonoutlier value.108 The following variables were winsorized as a result of outliers (and the number of observations winsorized): BMI (n = 3), systolic BP (n = 15), diastolic BP (n = 5), HR (n = 12), MAP (n = 11), pain catastrophizing (n = 10), negative affect (n = 4), positive affect (n = 4), state anxiety (n = 5), perceived stress (n = 7), psychological distress (n = 19), bodily pain (n = 35), general health perception (n = 15), cool detection threshold (n = 25), warm detection threshold (n = 25), heat pain threshold (n = 19), heat pain tolerance (n = 26), stimulus–response curve for heat ratings (n = 428 of 5653 trials; 7.6%), cold-pressor pain threshold (n = 13), cold-pressor pain tolerance (n = 18), thumb pressure threshold (n = 11), trapezius pressure threshold (n = 11), masseter pressure threshold (n = 16), ischemia pain threshold (n = 18), ischemia pain tolerance (n = 8), electric pain threshold (n = 26), 52°C heat series (n = 21), Pain45 heat used for TS-Heat Pain (n = 3), NFR threshold (n = 12), 3-stimulation threshold (n = 20), TS-NFR (104 trials, 2.8%), CPM of NFR (n = 40 of 2412 trials; 1.7%), CPM of electric pain (n = 77 of 2482 trials; 3.1%), and CPM of heat pain (n = 57 of 966 trials; 5.9%). When there were missing observations on control variables (Table 1), they were replaced by the grand mean before analyses of primary outcomes to avoid listwise deletion and loss of participants.
2.9.2. Background characteristics
Independent-samples t-tests were used to examine group differences on background variables that might affect pain processing. If a variable was found to differ, this was considered as a potential control variable in primary analyses. Significance level was set to α < 0.05 (2-tailed).
2.9.3. Primary analyses
Traditional (ie, general linear model) analysis of covariance (ANCOVA) models were used to analyze group differences on primary outcomes that were measured on a single occasion or averaged to a single score (ie, most pain sensitivity measures). Centered control variables were entered as covariates in these models. By contrast, measures of central sensitization (ie, TS-NFR, TS-Pain, and TS-Heat Pain) and pain inhibition (ie, CPM) were analyzed using multilevel models (MLMs) that examined change in pain outcomes (ie, repeated measures). Centered control variables were entered as covariates. For these analyses, level 1 units were the repeated measures, whereas level 2 units were participants. In most cases, the repeated-measures variance–covariance structure was modeled with an autocorrelation matrix (AR1). Significant interactions were followed up with Fisher least significant difference mean comparisons. Significance level was set to α < 0.05 (2-tailed).
3.1. Group differences on background characteristics
Table 1 presents Ns, mean values, SDs, and inferential statistics for participant characteristics by group. Groups were not different on age, dispositional pain catastrophizing, negative affect, positive affect, state anxiety, bodily pain, general health, biological sex, education level, or income. However, consistent with known physical and psychological health disparities,6,107,112 NAs had higher BMI, systolic BP, diastolic BP, MAP, resting HR, perceived stress, psychological distress, and poorer sleep quality. There were also group differences in marital status (NAs less likely to be single, more likely to cohabitate) and employment (NAs more likely to have full time job, less likely to be a student), but these are likely sample-specific.
Given these differences, all primary analyses controlled for BMI, MAP (to control for all cardiovascular differences), sleep quality, perceived stress, and psychological distress. Moreover, given well-established sex differences in pain,1,77 sex was added as a control variable, but its influence will not be discussed to keep the focus on racial/ethnic differences.
3.2. Peripheral fiber function
Cool and warm thresholds were skewed; so, they were squared and log transformed, respectively. As shown in Table 2, there were no significant group differences, suggesting that differences in Aδ- and C-fiber function did not contribute to pain risk in NAs.
3.3. Pain sensitivity
To briefly summarize the pain sensitivity results, the only group differences noted were cold pain threshold and tolerance. Native Americans had increased pain sensitivity (lower threshold/tolerance) on these measures (Table 2 and Fig. 2).
3.3.1. Heat pain threshold/tolerance
Nineteen participants were excluded from these analyses for failing to follow instructions (eg, poor effort and pain tolerance lower than pain threshold). As shown in Table 2 and Figure 2, there were no significant group differences on either variable. Thus, groups did not differ in their heat pain thresholds/tolerances.
3.3.2. Stimulus–response curve to heat
Log-transformed heat ratings (N = 257; NA = 127, NHW = 130) were analyzed using an 11 (Temperature) × 2 (Temperature Block) × 2 (Race) MLM ANCOVA model that covaried the control variables. Although there was a significant Race × Temperature interaction [F(10, 4705) = 2.025, P = 0.027], no group differences were found when the simple effects of race were examined (all Ps > 0.13); so, groups had similar stimulus–response curves to heat (Fig. 3). There was also no main effect of race [F(1, 257) = 0.250, P = 0.618]. Thus, groups did not differ in their sensitivity to heat pulses ranging between 40°C and 50°C.
3.3.3. Heat Pain45
As Table 2 shows, there was no group difference in the Pain45 stimulus intensity. This means that both groups were exposed to similar stimulus intensities during TS-Heat Pain (see results below).
3.3.4. Cold-pressor pain threshold/tolerance
Both variables were right skewed and so, they were log transformed. Three participants were excluded for failing to follow instructions. As shown in Table 2 and Figure 2, NAs had lower cold pain thresholds and tolerances, suggesting hyperalgesia.
3.3.5. Pressure pain thresholds
As shown in Table 2 and Figure 2, there were no group differences in pressure pain thresholds at any of the 3 body sites (masseter, trapezius, and thumb).
3.3.6. Ischemia pain threshold/tolerance
One participant was excluded for failing to follow instructions and both variables were log transformed. As shown in Table 2 and Figure 2, there were no group differences in ischemia pain threshold or tolerance.
3.3.7. Electric pain threshold/tolerance
As shown in Table 2 and Figure 2, there were no group differences in electric pain threshold or tolerance.
3.3.8. 52°C heat series
These data were log transformed and analyzed using a 5 (Stimulus Series) × 2 (Race) MLM ANCOVA covarying the control variables. There was no main effect of Race [F(1,271) < 1, P = 0.945] or a Race × Stimulus Series interaction [F(4,573) = 1.421, P = 0.226]; thus, groups did not differ on this measure. Descriptive statistics for the main effect of race are depicted in Table 2 and Figure 2.
3.3.9. Two-minute 10°C cold water bath (from conditioned pain modulation)
As shown in Table 2 and Figure 2, there were no group differences in pain ratings in response to the 2-minute circulating cold-water bath. Furthermore, this means that the painfulness of the CS used during CPM did not differ between groups.
3.4. Central sensitization
To briefly summarize these findings, groups did not significantly differ on any measure of central sensitization, indicating that pain risk in healthy NAs is not due to differences in central sensitization.
3.4.1. Nociceptive flexion reflex threshold
As shown in Table 2 and Figure 2, groups did not differ in NFR threshold, indicating that spinal sensitization was similar between groups.
3.4.2. Three-stimulation threshold
As shown in Table 2 and Figure 2, groups did not differ in 3-stimulation threshold, indicating that spinal sensitization was similar between groups.
3.4.3. Temporal summation of nociceptive flexion reflex and temporal summation of electric pain
These data were analyzed using a multilevel growth model approach in which the summation slope was modeled using a linear trend (TS Slope). Race, trial number (coding for the 5 different stimulus trains), and the centered control variables were also entered as predictors. Growth model results for TS-NFR and TS-Electric Pain are presented in Table 3 and outcomes are depicted in Figure 4. In brief, both groups showed significant summation of NFR and electric pain (across the 3-stimulus series and single stimulations vs triple stimulations), but groups did not differ in the degree of summation.
Two hundred seventy-one participants (135 NHW and 136 NA) participated in these TS tasks. Twelve participants (6 NHW and 6 NA) were excluded from TS-NFR analyses because their EMG was contaminated by voluntary movement. Thus, 259 participants contributed data. Results for TS-NFR indicated that both groups showed significant summation of NFR as indicated by the significant TS Slope effect, but groups did not differ in their degree of summation (Table 3 and Fig. 4).
Thirty-one participants (15 NHW and 16 NA) were excluded from analyses of ratings from the 3-stimulus train because ratings of stimulation 1 were either at ceiling (VAS ≥ 95) or floor (VAS ≤ 5). Thus, 240 participants contributed data. The main effect of TS Slope indicated that pain summated, but this summation did not differ between groups (Table 3 and Fig. 4).
Twenty-seven participants (12 NHW and 15 NA) were excluded from analysis of TS-Pain defined as the difference between single stimulations and the 3-stimulus train because ratings of the single stimulations were at ceiling (VAS ≥ 95) or floor (VAS ≤ 5). Thus, 244 participants contributed data. The main effect of TS Slope indicated that pain summated, but this summation did not differ between groups (Table 3 and Fig. 4).
3.4.4. Temporal summation of heat pain
Temporal summation of heat pain ratings were analyzed using a 2 (Race) × 10 (TS Slope) × 4 (Trial Number) MLM ANCOVA model that covaried the control variables. Two hundred sixty participants had TS-Heat Pain data available for analysis (NA = 128, NHW = 132). There was a significant TS Slope main effect [F(9, 8083) = 328.52, P < 0.001], suggesting that pain ratings decreased across the 10 pulse heat series (ie, habituated), but that effect was qualified by a significant Race × TS Slope interaction [F(9, 8231) = 3.24, P = 0.001]. Despite this interaction, there were no significant differences between NAs and NHWs on any of the 10 ratings of the heat stimulus series, suggesting both groups habituated to the heat pulses at the same rate (Fig. 5).
3.5. Conditioned modulation of pain and nociceptive flexion reflex
Two approaches were taken to analyzing CPM of NFR, CPM of electric pain, and CPM of heat pain. One approach examined the difference in pain/NFR between the baseline period and the conditioning period of the active CPM task. To do so, 2 (CPM phase: baseline vs conditioning) × 2 (Race) MLM ANCOVAs were used that covaried the control variables. The second approach examined the difference in pain/NFR between the cold-pressor conditioning phase and the sham (nonpainful water) conditioning phase. To do so, 2 (Conditioning Type: Cold vs Sham) × 2 (Race) MLM ANCOVAs were used that covaried the control variables. Results are presented in Table 4 and Figure 6. In brief, both groups showed significant CPM of electric pain and heat pain, but did not differ from one another. Neither group showed significant CPM of NFR.
Two hundred fifty-six participants participated in the CPM tasks, but 2 participants were excluded from CPM analyses for not following instructions. Thus, there were 254 considered for analysis (129 NHW, 125 NA).
3.5.1. Conditioned pain modulation of nociceptive flexion reflex
Three participants (2 NHW and 1 NA) were excluded from CPM of NFR analyses because their EMG was contaminated by voluntary muscle activity and 1 participant's NFR data were lost due to equipment failure. As shown in Table 4 and Figure 6, there were no group differences in CPM of NFR when comparing the baseline period with the conditioning period, or when the sham conditioning period was compared with the conditioning period. Moreover, neither group showed significant inhibition of NFR, although there was a trend for the NHW group to show inhibition when baseline was compared with conditioning (P = 0.05; Fig. 6).
3.5.2. Conditioned pain modulation of electric pain
Four participants (3 NA and 1 NHW) were excluded from analyses of CPM of electric pain because all ratings during baseline were at ceiling (VAS ≥ 95). None were at a floor (VAS ≤ 5). Thus, 250 participants contributed data. As shown in Table 4, when the baseline period was compared with the conditioning period, there was a significant Race × CPM Phase interaction. This interaction was also found in the analysis that compared conditioning with sham conditioning. However, in both cases, simple effect tests did not find a group difference in the inhibition of pain (Ps > 0.05, Fig. 6). Thus, both groups showed similar electric pain inhibition (ie, there were no detectable group differences).
3.5.3. Conditioned pain modulation of heat pain
Nine participants (5 NA and 4 NHW) were excluded from analyses of CPM of heat pain because all ratings during baseline were at floor (VAS ≤ 5). None were at ceiling. Thus, 245 participants contributed data. As shown in Table 4 and Figure 6, both groups showed significant inhibition of heat pain in the analysis that compared baseline with conditioning and the analysis that compared conditioning with sham. However, there were no group differences in this inhibition.
Contrary to hypotheses, there were no group differences on measures of peripheral fibers, central sensitization, CNS inhibition, or pain sensitivity, except that NAs had lower cold thresholds/tolerances (ie, greater cold pain sensitivity). To understand these results, it is important to consider the results from another recent OK-SNAP study that examined subjective responses to pain tasks (McGill Short-Form sensory and affect ratings, pain-related anxiety, and situational pain catastrophizing). Those analyses showed that, compared with NHWs, NAs reported greater pain-related anxiety and situational catastrophizing during pain,76 which we interpreted to mean that NAs have a stronger affective-motivational reaction to pain. The current difference in cold pain is consistent with this interpretation because cold pressor evokes a strong affective response.71 However, it is worth noting that other tasks (eg, ischemia) also evoke notable affective reactions.71 So when taken together, a stronger affective-motivational reaction to pain (ie, greater pain-related anxiety/situational catastrophizing) might place NAs at higher risk for chronic pain and suffering.53,68,92
To further examine this possibility, 2 subsequent studies have used bootstrapped mediation analyses to test whether pain-related anxiety and situational catastrophizing provide indirect pathways linking NAs with pronociceptive processes (eg, NA race→anxiety→reduced CPM). Results indicated pain-related anxiety mediated relationships with lower pain tolerances (electric, cold, ischemic, and heat) and ineffective CPM of NFR (Rhudy et al., under review). Furthermore, situational catastrophizing mediated a relationship with ineffective CPM of NFR (Toledo et al., under review).99 These data suggest that there are racial/ethnic differences in pain processing that stem from NAs' greater affective-motivational reactions to painful input.
It is currently unclear why OK-SNAP results are not consistent with our original pilot study, but there are a few study differences worth considering. First, the pilot study was small and so, it is possible the results were impacted by sampling error. Second, there are some notable sample differences. Average ages were lower in OK-SNAP (NHW = 28.5, NA = 30.8) compared with those in the pilot study (NHW = 33.6, NA = 41.1), particularly for the NA group. Moreover, BP and BMI were slightly lower in OK-SNAP. And finally, because the pilot study had a small sample size, this precluded statistical control of potential confounding variables in the analyses. Given these issues, it will be important to replicate our findings to establish the stability of the conclusions.
4.1. Integration with the literature, implications, and future directions
Accumulating evidence indicates that race/ethnicity can play a role in pain experience and chronic pain prevalence.15,17,28,30,31,38,41,42,44,45,52,59,61,69,72,79,84,104,109 To date, the majority of this work has been conducted with African American and Hispanic populations, but the number of studies examining other racial/ethnic groups is growing. These studies generally find that ethnic/racial groups who are at higher risk for chronic pain have lower pain tolerances and report more pain in response to suprathreshold stimuli.14,44,69 This is believed to be due, in part, to central sensitization (hyperexcitability of spinal cord neurons). Supporting this, studies have found greater temporal summation of pain in response to heat24,38,61,78 and mechanical pressure38,78 in ethnic/racial minorities. Although it is possible that these ethnic/racial differences represent a general bias towards greater pain report within these groups, this cannot account for the fact that Campbell et al. found African Americans had a lower threshold on the physiologically derived NFR,17 indicating greater spinal neuron sensitization. Thus, there is some physiological evidence that pain signals are amplified in African Americans.
A dysfunction of CNS pain inhibition may also contribute to racial/ethnic differences in pain because CPM-related inhibition16,24 and stress-induced analgesia62 are less effective in African Americans. Given these findings, one implication of the current study is that the mechanisms of chronic pain in NAs are not the same as those in other racial/ethnic minorities (especially in African Americans), because NAs do not show enhanced central sensitization or reduced CNS pain inhibition in the absence of anxiety/catastrophizing. When all the OK-SNAP findings are taken together, they imply that interventions to prevent chronic pain that focus on how to dampen affective-motivational reactions to pain may be helpful for NAs. Indeed, cognitive–behavioral techniques that target anxiety and catastrophizing reactions to improve pain tolerance may be particularly effective.3,83,95 Given the group difference in cold pain, these treatments could incorporate graded exposure to cold water. Moreover, it has been reported that some NAs are reluctant to engage in conventional treatments42; therefore, treatments may need to be tailored to include traditional practices (eg, smudging).11,42
This study focused only on between-group differences in pain processing that might contribute to NA pain risk and there are likely several within-group differences that affect progression to chronic pain. For example, there may be important cultural (eg, acculturation) or geographic (eg, reservation-dwelling) variables to consider. Moreover, NAs are more likely to experience traumatic/adverse life events,5,6,12,47,50,55,67,80 discrimination,2 and metabolic problems, such as diabetes.19 Given that these latter variables are associated with pain risk,29,35,56 it is possible that they could also contribute to the higher chronic pain prevalence in NAs. Even if the relationships between these factors and pain are not stronger in (or specific to) NAs (ie, relationships not moderated by race), they could still contribute to NA pain disparities because the base rates of these risk factors are higher in NAs (so, they could mediate the relationship between NA race and pronociceptive outcomes). Additional OK-SNAP studies will address within-group factors.40,48,91,98
As noted, we used stringent inclusion criteria to rule out confounding factors and to help ensure that groups were as similar as possible on background variables. Thus, it is possible this strategy excluded NAs at the highest risk for developing chronic pain, impairing our ability to identify between-group mechanisms that promote pain. Although possible, it did not seem to eliminate the elevated pain risk in NAs. Every 6 months, we contact participants (by phone/mail) to assess for onset of chronic pain (pain lasting >3 months, which does not remit at future assessments). Seventy-three percent have responded to our follow-up screens and 16% of those have developed chronic pain. Preliminary analyses of these follow-up data suggest that NAs are developing chronic pain at 3× the rate of NHWs.81 So, although NAs were healthy and pain-free when they enrolled in OK-SNAP, they are developing pain at a higher rate. Once follow-up data collection is complete, it will be important to determine whether individual differences in pain processing that were measured in this study can help explain what promotes chronic pain in NAs.
4.2. Strengths and limitations
The Oklahoma Study of Native American Pain Risk is the largest, most comprehensive study of NA pain processing, and is one of the largest studies of racial/ethnic differences in pain mechanisms to date.44 It also used numerous pain stimulus modalities, assessed subjective and physiological outcomes, and addressed multiple aspects of the pain system (peripheral, spinal, inhibitory, and perceptual). Tasks and testing days were pseudorandomly ordered or counterbalanced to minimize order effects. Potential confounds were either experimentally or statistically controlled. Moreover, most testing was administered by computer and experimenters observed from an adjacent room to minimize bias and/or experimenter effects on participant responses. Despite these strengths, a few limitations should be mentioned.
As noted, only healthy, pain-free individuals were recruited to determine whether preexisting differences in pain processing are present that contribute to chronic pain risk in NAs. Although this helps ensure that observed group differences are not due to disease status and/or access to health care, we do not know whether our findings will generalize to NAs with chronic pain or to older adults (given that the sample was relatively young). Second, although we varied the order of tasks and scheduled mandatory breaks, it is possible that some carryover occurred, which might have obscured some racial/ethnic differences. Third, our NA participants were recruited mostly from northeastern Oklahoma where the majority of NAs are not reservation-dwelling. It is not clear as to whether our results will generalize to NAs from other geographical regions, particularly those living on reservations where access to health care may be more limited. Fourth, although we chose tasks intended to comprehensively assess the pain system, there may be other tasks more sensitive to racial/ethnic differences in pain processing (eg, topical capsaicin).37,104 Fifth, because this was one of the first studies to examine pain risk mechanisms in NAs, we did not adjust for familywise type I error rate to ensure that potentially important differences were not missed. However, this may explain why only 2 significant differences (cold pain) were found among the large number of comparisons. And finally, this study was cross-sectional and so, we cannot guarantee that responses to pain tasks were stable estimates of each participant's pain processing and not influenced by recent events in their lives.
This study found that there were no differences between NAs and NHWs on measures of peripheral fiber function, spinal (central) sensitization, CNS pain inhibition, and most measures of pain threshold/tolerance. Surprisingly, the only difference was that NAs had lower cold pain thresholds/tolerances than NHWs, indicating that they were more pain sensitive. Together, this suggests that pain risk in NAs may be different than other U.S. minority groups who show greater CNS excitability, reduced pain inhibition, and generally enhanced pain sensitivity. Future studies are needed to identify other factors that contribute to their pain risk.
Conflict of interest statement
The authors have no conflicts of interest to declare.
Aspects of this research have been presented at the International Association for the Study of Pain's 2018 World Congress on Pain and at the 2019 American Pain Society conference.
The authors thank Heather B. Coleman, Kathryn A. Thompson, Jessica M. Fisher, Samuel P. Herbig, Ky’Lee B. Barnoski, and Lucinda Chee for their help with data collection. This research was supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health under Award Number R01MD007807. Edward Lannon, Shreela Palit, and Yvette Güereca were supported by a National Science Foundation Graduate Research Fellowship Program. The content is solely the responsibility of the authors and does not necessarily reflect the views of the National Institutes of Health, National Science Foundation, Indian Health Service, or the Cherokee Nation.
. Akinci MK, Johnston GAR. Sex differences in acute swim stress-induced changes in the binding of MK-801 to the NMDA subclass of glutamate receptors in mouse forebrain. J Neurochem 1993;61:2290.
. American Psychological Association. Stress in America: The impact of discrimination. Stress in America™ Survey. APA, 2016.
. Bailey KM, Carleton RN, Vlaeyen JW, Asmundson GJ. Treatments addressing pain-related fear and anxiety in patients with chronic musculoskeletal pain: a preliminary review. Cogn Behav Ther 2010;39:46–63.
. Barnes PM, Adams PF, Powell-Griner E. Health characteristics of the American Indian or Alaska Native adult population: United States, 2004-2008. In: USDoHaH Services editor. Vol. 20. Hyattesville: National Center for Health Statistics, 2010.
. Beals J, Manson SM, Whitesell NR, Spicer P, Novins DK, Mitchell CM. Prevalence of DSM-IV disorders and attendant help-seeking in 2 American Indian reservation populations. Arch Gen Psychiatry 2005;62:99–108.
. Beals J, Novins DK, Whitesell NR, Spicer P, Mitchell CM, Manson SM; American Indian Service Utilization PE, Risk, Team PFP. Prevalence of mental disorders and utilization of mental health services in two American Indian reservation populations: mental health disparities in a national context. Am J Psychiatry 2005;162:1723–32.
. Bebbington P, Nayani T. The psychosis screening questionnaire. Int J Methods Psychatric Res 1995;5:11–19.
. Bruehl S, Chung OY, Burns JW, Diedrich L. Trait anger expressiveness and pain-induced beta-endorphin release: support for the opioid dysfunction hypothesis. PAIN 2007;130:208–15.
. Bruehl S, Chung OY, Chont M. Chronic pain-related changes in endogenous opioid analgesia: a case report. PAIN 2010;148:167–71.
. Bruehl S, McCubbin JA, Harden RN. Theoretical review: altered pain regulatory systems in chronic pain. Neurosci Biobehav Rev 1999;23:877–90.
. Buchwald D, Beals J, Manson SM. Use of traditional health practices among Native Americans
in a primary care setting. Med Care 2000;38:1191–9.
. Buchwald D, Goldberg J, Noonan C, Beals J, Manson S. Relationship between post-traumatic stress disorder and pain in two American Indian tribes. Pain Med 2005;6:72–9.
. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index (PSQI): a new instrument for psychiatric research and practice. Psychiatry Res 1989;28:193–213.
. Campbell CM, Edwards RR. Ethnic differences
in pain and pain management. Pain Manag 2012;2:219–30.
. Campbell CM, Edwards RR, Fillingim RB. Ethnic differences
in responses to multiple experimental pain stimuli. PAIN 2005;113:20–6.
. Campbell CM, France CR, Robinson ME, Logan HL, Geffken GR, Fillingim RB. Ethnic differences
in diffuse noxious inhibitory controls. J Pain 2008;9:759–66.
. Campbell CM, France CR, Robinson ME, Logan HL, Geffken GR, Fillingim RB. Ethnic differences
in the nociceptive flexion reflex
(NFR). PAIN 2008;134:91–6.
. Campbell TS, Hughes JW, Girdler SS, Maixner W, Sherwood A. Relationship of ethnicity, gender, and ambulatory blood pressure to pain sensitivity: effects of individualized pain rating scales. J Pain 2004;5:183–91.
. Centers for Disease Control and Prevention. National Diabetes Statistics Report: Estimates of Diabetes and Its Burden in the United States. Atlanta: US Department of Health and Human Services, 2014.
. Chapman WP, Jones CM. Variations in cutaneous and visceral pain sensitivity in normal subjects. J Clin Invest 1944;23:81–91.
. Cleeland CS, Gonin R, Baez L, Loehrer P, Pandya KJ. Pain and treatment of pain in minority patients with cancer. The eastern cooperative oncology group minority outpatient pain study. Ann Intern Med 1997;127:813–6.
. Cohen S, Kamarck T, Mermelstein R. A Global Measure of Perceived Stress. Journal of Health and Social Behavior 1983;24:385–96.
. Cox JJ, Reimann F, Nicholas AK, Thornton G, Roberts E, Springell K, Karbani G, Jafri H, Mannan J, Raashid Y, Al-Gazali L, Hamamy H, Valente EM, Gorman S, Williams R, McHale DP, Wood JN, Gribble FM, Woods CG. An SCN9A channelopathy causes congenital inability to experience pain. Nature 2006;444:894–8.
. Cruz-Almeida Y, Sibille KT, Goodin BR, Petrov ME, Bartley EJ, Riley JL, King CD, Glover TL, Sotolongo A, Herbert MS. Racial and ethnic differences
in older adults with knee osteoarthritis. Arthritis Rheumatol 2014;66:1800–10.
. Dearborn GVN. A case of congenital general pure analgesia. J Nerv Ment Dis 1932;75:612–15.
. Derogatis LR. SCL-90-R: symptom checklist-90-R: administration, scoring & procedures manual. Minneapolis: National Computer Systems, Inc., 1994.
. Deyo RA, Mirza SK, Martin BI. Back pain prevalence and visit rates: estimates from U.S. national surveys, 2002. Spine 2006;31:2724–7.
. Edwards CL, Fillingim RB, Keefe F. Race, ethnicity and pain. PAIN 2001;94:133–7.
. Edwards RR. The association of perceived discrimination with low back pain. J Behav Med 2008;31:379–89.
. Edwards RR, Doleys DM, Fillingim RB, Lowery D. Ethnic differences
in pain tolerance: clinical implications in a chronic pain population. Psychosom Med 2001;63:316–23.
. Edwards RR, Fillingim RB. Ethnic differences
in thermal pain responses. Psychosom Med 1999;61:346–54.
. Edwards RR, Moric M, Husfeldt B, Buvanendran A, Ivankovich O. Ethnic similarities and differences in the chronic pain experience: a comparison of African American, Hispanic, and white patients. Pain Med 2005;6:88–98.
. Farrell M, Gibson S. Age interacts with stimulus frequency in the temporal summation of pain. Pain Med 2007;8:514–20.
. Ferucci ED, Templin DW, Lanier AP. Rheumatoid arthritis in American Indians and Alaska Natives: a review of the literature. Semin Arthritis Rheum 2005;34:662–7.
. Fillingim RB, Edwards RR. Is self-reported childhood abuse history associated with pain perception among healthy young women and men? Clin J Pain 2005;21:387–97.
. Fillingim RB, Maixner W, Girdler SS, Light KC, Sheps DS, Mason GA. Ischemic but not thermal pain sensitivity varies across the menstrual cycle. Psychosom Med 1997;559:512.
. Gazerani P, Arendt-Nielsen L. The impact of ethnic differences
in response to capsaicin-induced trigeminal sensitization. PAIN 2005;117:223–9.
. Goodin BR, Bulls HW, Herbert MS, Schmidt J, King CD, Glover TL, Sotolongo A, Sibille KT, Cruz-Almeida Y, Staud R. Temporal summation of pain as a prospective predictor of clinical pain severity in adults aged 45 years and above with knee osteoarthritis: ethnic differences
. Psychosom Med 2014;76:302.
. Granovsky Y, Anand P, Nakae A, Nascimento O, Smith B, Sprecher E, Valls-Solé J. Normative data for Aδ contact heat evoked potentials in adult population: a multicenter study. PAIN 2016;157:1156–63.
. Güereca Y, Kuhn B, Lannon E, Palit S, Sturycz C, Payne M, Hellman N, Toledo T, Huber F, Demuth M, Palit S, Shadlow JO, Rhudy JL. The relationship between discrimination and pain tolerance and its potential mediation by stress: results from the Oklahoma Study of Native American Pain Risk (OK-SNAP). J Pain 2019;20:S40–1.
. Hastie BA, Riley JL III, Fillingim RB. Ethnic differences
in pain coping: factor structure of the coping strategies questionnaire and coping strategies questionnaire-revised. J Pain 2004;5:304–16.
. Jimenez N, Garroutte E, Kundu A, Morales L, Buchwald D. A review of the experience, epidemiology, and management of pain among American Indian, Alaska Native, and Aboriginal Canadian peoples. J Pain 2011;12:511–22.
. Kennedy DL, Kemp HI, Ridout D, Yarnitsky D, Rice AS. Reliability of conditioned pain modulation: a systematic review. PAIN 2016;157:2410–19.
. Kim HJ, Yang GS, Greenspan JD, Downton KD, Griffith KA, Renn CL, Johantgen M, Dorsey SG. Racial and ethnic differences
in experimental pain sensitivity: systematic review and meta-analysis. PAIN 2017;158:194–211.
. Komiyama O, Kawara M, De Laat A. Ethnic differences
regarding tactile and pain thresholds in the trigeminal region. J Pain 2007;8:363–9.
. Kong JT, Johnson KA, Balise RR, Mackey S. Test-retest reliability of thermal temporal summation using an individualized protocol. J Pain 2013;14:79–88.
. Koss MP, Yuan NP, Dightman D, Prince RJ, Polacca M, Sanderson B, Goldman D. Adverse childhood exposures and alcohol dependence among seven Native American tribes. Am J Prev Med 2003;25:238–44.
. Kuhn B, Lannon E, Payne M, Palit S, Sturycz C, Thompson K, Güereca Y, Hellman N, Shadlow J, Rhudy J. Quantitative sensory testing
(QST) and diabetes risk: are abnormalities in small diameter afferents present in healthy persons at high risk for diabetes? J Pain 2017;18:S83.
. Leake J, Jozzy S, Uswak G. Severe dental caries, impacts and determinants among children 2-6 years of age in Inuvik Region, Northwest Territories, Canada. J Can Dent Assoc 2008;74:519.
. Libby AM, Orton HD, Novins DK, Beals J, Manson SM. Childhood physical and sexual abuse and subsequent depressive and anxiety disorders for two American Indian tribes. Psychol Med 2005;35:329–40.
. Loeser JD, Butler SH, Chapman CR, Turk DC. Bonica's management of pain. Philadelphia: Lippincott Williams & Wilkins, 2000.
. Lu Q, Zeltzer L, Tsao J. Multiethnic differences in responses to laboratory pain stimuli among children. Health Psychol 2013;32:905–14.
. Lumley MA, Cohen JL, Borszcz GS, Cano A, Radcliffe AM, Porter LS, Schubiner H, Keefe FJ. Pain and emotion: a biopsychosocial review of recent research. J Clin Psychol 2011;67:942–68.
. Manresa JAB, Neziri AY, Curatolo M, Arendt-Nielsen L, Andersen OK. Test–retest reliability of the nociceptive withdrawal reflex and electrical pain thresholds after single and repeated stimulation in patients with chronic low back pain. Eur J Appl Physiol 2011;111:83–92.
. Manson SM, Beals J, Klein SA, Croy CD, Team AS. Social epidemiology of trauma among 2 American Indian reservation populations. Am J Public Health 2005;95:851–9.
. Mäntyselkä P, Miettola J, Niskanen L, Kumpusalo E. Chronic pain, impaired glucose tolerance and diabetes: a community-based study. PAIN 2008;137:34–40.
. Mauldin J, Cameron HD, Jeanotte D, Solomon G, Jarvis JN. Chronic arthritis in children and adolescents in two Indian health service user populations. BMC Musculoskelet Disord 2004;5:30.
. Mayer DJ, Hayes RL. Stimulation-produced analgesia: development of tolerance and cross-tolerance to morphine. Science 1975;188:941–3.
. McCracken LM, Matthews AK, Tang TS, Cuba SL. A comparison of blacks and whites seeking treatment for chronic pain. Clin J Pain 2001;17:249–55.
. Meagher MW, Arnau RC, Rhudy JL. Pain and emotion: effects of affective picture modulation. Psychosom Med 2001;63:79–90.
. Mechlin B, Heymen S, Edwards CL, Girdler SS. Ethnic differences
in cardiovascular-somatosensory interactions and in the central processing of noxious stimuli. Psychophysiology 2011;48:762–73.
. Mechlin MB, Maixner W, Light KC, Fisher JM, Girdler SS. African Americans show alterations in endogenous pain regulatory mechanisms and reduced pain tolerance to experimental pain procedures. Psychosom Med 2005;67:948.
. Mendell LM, Wall PD. Responses OF single dorsal cord cells to peripheral cutaneous unmyelinated fibres. Nature 1965;206:97–9.
. Palit S, Kerr KL, Kuhn BL, DelVentura JL, Terry EL, Bartley EJ, Shadlow JO, Rhudy JL. Examining emotional modulation of pain and spinal nociception in Native Americans
: a preliminary investigation. Int J Psychophysiol 2013;90:272–81.
. Palit S, Kerr KL, Kuhn BL, Terry EL, DelVentura JL, Bartley EJ, Shadlow JO, Rhudy JL. Exploring pain processing differences in Native Americans
. Health Psychol 2013;32:1127–36.
. Pittenger GL, Mehrabyan A, Simmons K, Amandarice C, Dublin C, Barlow P, Vinik AI. Small fiber neuropathy is associated with the metabolic syndrome. Metab Syndr Relat Disord 2005;3:113–21.
. Pole N, Gone JP, Kulkarni M. Posttraumatic stress disorder among ethnoracial minorities in the United States. Clin Psychol Sci Pract 2008;15:35–61.
. Quartana PJ, Campbell CM, Edwards RR. Pain catastrophizing: a critical review. Expert Rev Neurother 2009;9:745–58.
. Rahim-Williams B, Riley JL III, Williams AK, Fillingim RB. A quantitative review of ethnic group differences in experimental pain response: do biology, psychology, and culture matter? Pain Med 2012;13:522–40.
. Rahim-Williams FB, Riley JL III, Herrera D, Campbell CM, Hastie BA, Fillingim RB. Ethnic identity predicts experimental pain sensitivity in African Americans and Hispanics. PAIN 2007;129:177–84.
. Rainville P, Feine JS, Bushnell MC, Duncan GH. A psychophysical comparison of sensory and affective responses to four modalities of experimental pain. Somatosens Mot Res 1992;9:265–77.
. Rhee H. Prevalence and predictors of headaches in US adolescents. Headache J Head Face Pain 2000;40:528–38.
. Rhudy JL, Dubbert PM, Parker JD, Burke RS, Williams AE. Affective modulation of pain in substance dependent veterans. Pain Med 2006;7:483–500.
. Rhudy JL, France CR. Defining the nociceptive flexion reflex
(NFR) threshold in human participants: a comparison of different scoring criteria. PAIN 2007;128:244–53.
. Rhudy JL, France CR, Bartley EJ, McCabe KM, Williams AE. Psychophysiological responses to pain: further validation of the nociceptive flexion reflex
(NFR) as a measure of nociception using multilevel modeling. Psychophysiology 2009;46:939–48.
. Rhudy JL, Lannon EW, Kuhn BL, Palit S, Payne MF, Sturycz CA, Hellman N, Güereca YM, Toledo TA, Coleman HB, Thompson KA, Fisher JM, Herbig SP, Barnoski KB, Chee L, Shadlow JO. Sensory, affective, and catastrophizing reactions to multiple stimulus modalities: results from the Oklahoma Study of Native American Pain Risk. J Pain 2019;20:965–79.
. Rhudy JL, Williams AE. Gender differences in pain: do emotions play a role? Gend Med 2005;2:208–26.
. Riley JL, Cruz-Almeida Y, Glover TL, King CD, Goodin BR, Sibille KT, Bartley EJ, Herbert MS, Sotolongo A, Fessler BJ, Redden DT, Staud R, Bradley LA, Fillingim RB. Age and race effects on pain sensitivity and modulation among middle-aged and older adults. J Pain 2014;15:272–82.
. Riley JL III, Wade JB, Myers CD, Sheffield D, Papas RK, Price DD. Racial/ethnic differences
in the experience of chronic pain. PAIN 2002;100:291.
. Robin RW, Chester B, Rasmussen JK, Jaranson JM, Goldman D. Prevalence, characteristics, and impact of childhood sexual abuse in a Southwestern American Indian tribe. Child Abuse Neglect 1997;21:769–87.
. Ross E, Huber F, Kuhn B, Lannon E, Sturycz C, Payne M, Hellman N, Toledo T, Güereca Y, Palit S, Demuth M, Shadlow JO, Rhudy JL. Assessing chronic pain onset in native Americans
: follow-up results from the Oklahoma Study of Native American Pain Risk (OK-SNAP). J Pain 2019;20:S33.
. Sandrini G, Serrao M, Rossi P, Romaniello A, Cruccu G, Willer JC. The lower limb flexion reflex in humans. Prog Neurobiol 2005;77:353–95.
. Schütze R, Rees C, Smith A, Slater H, Campbell JM, O'Sullivan P. How can we best reduce pain catastrophizing in adults with chronic noncancer pain? A systematic review and meta-analysis. J Pain 2018;19:233–56.
. Sheffield D, Biles PL, Orom H, Maixner W, Sheps DS. Race and sex differences in cutaneous pain perception. Psychosom Med 2000;62:517.
. Shun CT, Chang YC, Wu HP, Hsieh SC, Lin WM, Lin YH, Tai TY, Hsieh ST. Skin denervation in type 2 diabetes: correlations with diabetic duration and functional impairments. Brain 2004;127:1593–605.
. Sorensen L, Molyneaux L, Yue DK. The level of small nerve fiber dysfunction does not predict pain in diabetic neuropathy: a study using quantitative sensory testing
. Clin J Pain 2006;22:261–5.
. Spielberger CD. State-trait anxiety inventory. Palo-Alto: Consulting Psychologists Press, 1983.
. Staud R, Craggs JG, Robinson ME, Perlstein WM, Price DD. Brain activity related to temporal summation of C-fiber evoked pain. PAIN 2007;129:130–42.
. Staud R, Price DD, Fillingim RB. Advanced continuous-contact heat pulse design for efficient temporal summation of second pain (windup). J Pain 2006;7:575–82.
. Staud R, Vierck CJ, Robinson ME, Price DD. Effects of the N-methyl-D-aspartate receptor antagonist dextromethorphan on temporal summation of pain are similar in fibromyalgia patients and normal control subjects. J Pain 2005;6:323–32.
. Sturycz CA, Hellman N, Payne MF, Kuhn BL, Hahn B, Lannon EW, Palit S, Güereca YM, Toledo TA, Shadlow JO, Rhudy JL. Race/ethnicity does not moderate the relationship between adverse life experiences and temporal summation of the nociceptive flexion reflex
and pain: results from the Oklahoma Study of Native American Pain Risk. J Pain 2019;20:941–955.
. Sullivan MJ, Thorn B, Haythornthwaite JA, Keefe F, Martin M, Bradley LA, Lefebvre JC. Theoretical perspectives on the relation between catastrophizing and pain. Clin J Pain 2001;17:52–64.
. Sullivan MJL, Bishop SR, Pivik J. The pain catastrophizing scale: development and validation. Psychol Assess 1995;7:524–32.
. Terry EL, France CR, Bartley EJ, Delventura JL, Kerr KL, Vincent AL, Rhudy JL. Standardizing procedures to study sensitization of human spinal nociceptive processes: comparing parameters for temporal summation of the nociceptive flexion reflex
(TS-NFR). Int J Psychophysiol 2011;81:263–74.
. Terry EL, Thompson KA, Rhudy JL. Experimental reduction of pain catastrophizing reduces pain perception and temporal summation of the nociceptive flexion reflex
(NFR). Proceedings of the International Association for the Study of Pain, 15th World Congress on Pain, 2014. p. PF331.
. Thrush DC. Congenital insensitivity to pain. A clinical, genetic and neurophysiological study of four children from the same family. Brain 1973;96:369–86.
. Todd KH. Pain assessment and ethnicity. Ann Emerg Med 1996;27:421–3.
. Toledo T, Kuhn B, Lannon E, Hellman N, Sturycz C, Palit S, Guereca Y, Payne M, Shadlow J, Rhudy J. Is risk for diabetes associated with disrupted descending modulation of pain and spinal nociception? J Pain 2017;18:S48–9.
. Toledo T, Lannon E, Kuhn B, Hellman N, Sturycz C, Palit S, Guereca Y, Payne M, Shadlow J, Rhudy J. State catastrophizing is associated with facilitation of spinal nociception during conditioned pain modulation (CPM). J Pain 2018;19:S15.
. USDHHS. Summary of health statistics for U.S. Adults: National health interview survey, 2009. In: HaH Services editor. Hyattsville: DHHS, 2010.
. Vlckova-Moravcova E, Bednarik J, Belobradkova J, Sommer C. Small-fibre involvement in diabetic patients with neuropathic foot pain. Diabet Med 2008;25:692–9.
. Vlčková-Moravcová E, Bednařík J, Dušek L, Toyka KV, Sommer C. Diagnostic validity of epidermal nerve fiber densities in painful sensory neuropathies. Muscle Nerve 2008;37:50–60.
. Walsh NE, Schoenfeld L, Ramamurthy S, Hoffman J. Normative model for cold pressor test. Am J Phys Med Rehabil 1989;68:6–11.
. Wang H, Papoiu ADP, Coghill RC, Patel T, Wang N, Yosipovitch G. Ethnic differences
in pain, itch and thermal detection in response to topical capsaicin: African Americans display a notably limited hyperalgesia and neurogenic inflammation. Br J Dermatol 2010;162:1023–9.
. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey manual and interpretation guide. Boston: The Health Institute New England Medical Center, 1993.
. Watson D, Clark LA, Tellegen A. Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 1988;54:1063–70.
. Welty TK, Lee ET, Yeh J, Cowan LD, Go O, Fabsitz RR, Le NA, Oopik AJ, Robbins DC, Howard BV. Cardiovascular disease risk factors among American Indians. The Strong Heart Study. Am J Epidemiol 1995;142:269–87.
. Wilcox RR. Understanding and applying basic statistical methods using R. Hoboken, NJ: John Wiley & Sons, 2016.
. Woodrow KM, Friedman GD, Siegelaub AB, Collen MF. Pain tolerance: differences according to age, sex and race. Psychosom Med 1972;34:548–56.
. You HJ, Dahl Morch C, Chen J, Arendt-Nielsen L. Simultaneous recordings of wind-up of paired spinal dorsal horn nociceptive neuron and nociceptive flexion reflex
in rats. Brain Res 2003;960:235–45.
. You HJ, Mørch CD, Arendt-Nielsen L. Electrophysiological characterization of facilitated spinal withdrawal reflex to repetitive electrical stimuli and its modulation by central glutamate receptor in spinal anesthetized rats. Brain Res 2004;1009:110–9.
. Zamora-Kapoor A, Sinclair K, Nelson L, Lee H, Buchwald D. Obesity risk factors in American Indians and Alaska Natives: a systematic review. Public Health 2019;174:85–96.
. Zatzick DF, Dimsdale JE. Cultural variations in response to painful stimuli. Psychosom Med 1990;52:544.
. Zhou L, Kitch DW, Evans SR, Hauer P, Raman S, Ebenezer GJ, Gerschenson M, Marra CM, Valcour V, Diaz-Arrastia R, Goodkin K, Millar L, Shriver S, Asmuth DM, Clifford DB, Simpson DM, McArthur JC. Correlates of epidermal nerve fiber densities in HIV-associated distal sensory polyneuropathy. Neurology 2007;68:2113–9.