Abstract
Objective:
The co-existence of chronic pain conditions with anxiety and/or depression is common in the general population but poorly described during pregnancy. In this study, we sought to describe trends in chronic pain among a sample of delivering people and describe the co-existence of chronic pain with anxiety and/or depression among delivering people.
Methods:
This cross-sectional study used data from Optum’s de-identified Clinformatics® Data Mart Database between 2008 and 2021, for delivering persons with coverage by single employer-based health plan. We computed predicted margins from generalized estimating equations to determine the marginal predicted probability of chronic pain among all delivering and non-delivering persons who identify as women with and without diagnosed anxiety and/or depression.
Results:
Musculoskeletal and pelvic pain occurred most often regardless of delivering status. Delivering persons with anxiety and/or depression had higher marginal predicted probabilities of chronic pain compared to all delivering persons. Between 2008 and 2021, the predicted probabilities ranged from 0.400 to 0.527 and 0.221–0.261, respectively.
Conclusion:
Chronic pain conditions are common in pregnancy and nearly two times higher among individuals with anxiety and/or depression. The frequency of comorbid depression and/or anxiety with pain disorders among delivering persons highlights the importance of proper detection, coordination of care, and safe treatment options for this population.
Keywords: Pregnancy, Mental health, Pain, Anxiety, Depression
1. Introduction
One in five individuals who identify as women suffer from a chronic pain condition, and nearly 9% report that pain limits their life or work activities. [1] Pain interferes with activities of daily living, participation in social relationships, and contributions to work, all of which can substantially and cyclically impact mental health. [2–4] When pregnant individuals experience chronic pain, several unique considerations arise. Pregnancy can exacerbate some types of chronic pain, such as headaches, pelvic pain, or musculoskeletal pain, [5,6] yet pregnancy precludes some effective treatment options.
Chronic pain and behavioral health conditions frequently co-occur; chronic pain sufferers are two to four times more likely to have anxiety, depression or a substance use disorder. [7–10] The relationship between chronic pain and behavioral health becomes particularly concerning during pregnancy. Anxiety or depression affect one in five people during pregnancy or postpartum, [11–13] and behavioral health conditions are a major contributing factor to maternal morbidity, mortality, and pregnancy-associated deaths. [14–17] Unfortunately few studies document the patterns of behavioral health conditions in the context of pain diagnoses and fewer compare delivering and non-delivering women. [18,19]
Given that a biopsychosocial lens can explain the experience of chronic pain, [20] we can anticipate that psychological factors, including mental health diagnoses, might contribute to pain severity and impaired functioning overall and among delivering individuals. Previous, small, or single site studies have documented high rates of anxiety and/or depression among pregnant people with chronic pain conditions. [18,19] However, the prevalence of chronic pain and its overlap with anxiety and/or depression in the broader population of pregnant people remains unknown. Addressing this knowledge gap has crucial implications for efforts aimed at reducing the alarming rates of behavioral health-related maternal morbidity and mortality among delivering people. [21]
Clarifying the co-existence of chronic pain (CP) with anxiety and/or depression (AD) during pregnancy could lead to more comprehensive screening, risk assessment and individualized treatment plans that address both behavioral health and pain conditions present. Accordingly, in this study we sought to 1) describe trends in CP diagnosis among a sample of delivering people and 2) describe the co-existence of CP with AD among delivering people. To accomplish these objectives, we examined trends in CP over time among a large population of all delivering people and delivering people with AD diagnoses. We also examined similar trends in CP in non-delivering people who identify as women to contextualize our findings.
2. Methods
We evaluated the likelihood of CP among delivering aged people enrolled in employer sponsored insurance (ESI) in the U.S. between January 1, 2008, and December 31, 2021. Note that the dataset used in this analysis only includes binary gender classification (male/female); therefore, we used these binary gender classifications to identify “women.”
2.1. Data source
The Maternal Behavioral Health Policy Evaluation (MAPLE) study is a retrospective, observational cohort study of the impact of federal health policy changes on utilization, expenditures, and delivery outcomes among pregnant and postpartum people. The MAPLE study uses administrative claims data from Optum’s de-identified Clinformatics® Data Mart Database. This data consists of de-identified medical and pharmacy commercial insurance claims and includes information on medical diagnoses and procedures, hospitalizations, inpatient and outpatient visits, and patient demographic characteristics. The study site’s Institutional Review Board deemed the study not regulated because it involves analysis of de-identified data only (HUM00164685).
2.2. Study sample
We generated an analytic sample of people who identify as women enrolled in ESI between 2008 and 2021. We further restricted the study sample to women aged 15 to 44 years with continuous enrollment in only one employer-based health plan for at least one year. Because we do not have individual-level identifiers, individuals that switch plans during the year would not appear to have continuous enrollment; therefore, these individuals are not included in our analysis. We characterized people with and without a delivery during each year of the study period as delivering and non-delivering, respectively. We identified delivery hospitalizations using standardized International Classification of Disease-9th and 10th Revision-Clinical Modification (ICD-9-CM and ICD-10-CM) diagnosis and procedure codes. We used live births as our unit of analysis; therefore, a person could appear in the dataset more than once during the fourteen-year observation period. If a person delivered more than once during the same calendar year, we excluded the second delivery. For the delivery subgroup, we only included people with a delivery that had at least one year of medical service claims prior to the delivery date. We used the pre-delivery period to assess evidence of CP and AD. For the non-delivery subgroup, we restricted to the one year of continuous enrollment, during which we assessed for the presence of CP and AD. Our non-delivering sample included all individuals who identified as women that met age and enrollment criteria but did not have a delivery.
We used a previously validated, claims-based approach to identifying people with CP. [22] We used AD diagnoses from the Healthcare Cost and Utilization Project algorithms for anxiety and depression based on ICD-9-CM and ICD-10-CM codes (Appendix Table 2A). [23,24] We considered people as having AD if codes appeared in either two outpatient encounters or one inpatient stay during the 12 months preceding date of delivery or during the calendar year for the non-delivering women. A similar algorithm was applied to identify diagnoses for CP.
2.3. Analytic methods
We summarized demographic characteristics for all individuals yearly, separated by delivery status, and presence of chronic pain alone and with anxiety and/or depression. We employed Wilcoxon rank sum and Pearson Chi-squared tests to assess demographic characteristic differences between years. To describe the distribution of pain, we tabulated all diagnoses related to pain and rank-ordered the top 25 most prevalent diagnosis codes, which we grouped in six main categories: pelvic pain, headache, musculoskeletal pain, fibromyalgia, widespread pain and other. From there, we calculated the proportions of each pain category among those with CP for both delivering and non-delivering women with and without AD. Since pain categories for an individual are not mutually exclusive, proportions across all pain categories will not equal 100%. We then used Pearson Chi-squared tests to assess proportional differences for each pain category between delivering vs. non-delivering people and delivering vs. non-delivering people with AD. We applied a Bonferroni correction to account for multiple testing. We used generalized estimating equations (GEE), clustered at the individual level, to assess the probability of CP over time for delivering and non-delivering people while adjusting for age, race, household income, and region. We used household income (% FPL) as a proxy for socioeconomic status. We ran separate models for each of the four subgroups: delivering women, delivering women with AD, non-delivering women, and non-delivering women with AD. We then computed predictive margins within each subgroup to determine the marginal predicted probability of CP for each year of the study. We used two-sided statistical tests with an alpha level of 0.05 for all statistical analyses. We performed all claims data management in SAS version 9.4 (SAS Institute) and statistical analyses in Stata version 14.1 (StataCorp) and R version 4.3.0 (R Core Team).
3. Results
The study sample included 9,016,715 people enrolled in 510,310 distinct health plans, with 1,116,937 unique delivering and 7,899,778 non-delivering people from January 1, 2008, and December 31, 2021. Compared to 2008, the delivering population in 2021 was slightly older (mean age of 31.8 years in 2021 vs 30.8 in 2008, p < 0.001), and more likely to live in households at or below 400% of the federal poverty level (47.10% in 2021 vs. 24.59% in 2008, p < 0.001). Individual characteristics of the sample population of delivering people in 2008 and 2021 appear in Table 1. Characteristics of non-delivering people in 2008 and 2021 appear in Appendix Table 1.
Table 1.
Sample characteristics of delivering people with chronic pain (CP), chronic pain with anxiety and/or depression (CP and AD), and overall, 2008 and 2021.
| Delivering People | ||||||
|---|---|---|---|---|---|---|
| 2008 | 2021 | |||||
| CP | CP and AD | Overall | CP | CP and AD | Overall | |
| Sample Sizea | 22,594 (20.07%) | 3677 (3.26%) | 112,574 | 13,594 (17.94%) | 5815 (7.67%) | 75,768 |
| Age, years | ||||||
| Mean (SD) | 31.10 (5.60) | 31.60 (5.78) | 30.80 | 32.10 | 32.00 | 31.80 |
| (5.51) | (4.94) | (5.25) | (5.04) | |||
| Age Group, years | ||||||
| 15–19 | 588 (2.60%) | 99 (2.69%) | 3356 (2.98%) | 71 (0.52%) | 60 (1.03%) | 671 (0.89%) |
| 20–24 | 2143 (9.48%) | 312 (8.49%) | 10,899 (9.68%) | 948 (6.97%) | 502 (8.63%) | 6045 (7.98%) |
| 25–29 | 5978 (26.46%) | 884 (24.04%) | 30,659 (27.23%) | 2868 (21.10%) | 1155 (19.86%) | 16,083 (21.23%) |
| 30–34 | 7606 (33.66%) | 1225 (33.32%) | 38,452 (34.16%) | 5427 (39.92%) | 2235 (38.44%) | 29,792 (39.32%) |
| 34–39 | 4727 (20.92%) | 847 (23.04%) | 23,180 (20.59%) | 3413 (25.11%) | 1464 (25.18%) | 18,832 (24.85%) |
| 40–44 | 1552 (6.87%) | 310 (8.43%) | 6028 (5.35%) | 867 (6.38%) | 399 (6.86%) | 4345 (5.73%) |
| Race/ethnicity | ||||||
| Asian | 1178 (5.21%) | 73 (1.99%) | 7066 (6.28%) | 709 (5.22%) | 137 (2.36%) | 4375 (5.77%) |
| Black | 2089 (9.25%) | 259 (7.04%) | 9949 (8.84%) | 1115 (8.20%) | 453 (7.79%) | 6027 (7.95%) |
| Hispanic | 2626 (11.62%) | 321 (8.73%) | 14,066 (12.49%) | 1633 (12.01%) | 499 (8.58%) | 9233 (12.19%) |
| White | 13,601 (60.20%) | 2511 (68.29%) | 65,115 (57.84%) | 8279 (60.90%) | 4055 (69.73%) | 44,874 (59.23%) |
| Unknown/Missing | 3100 (13.72%) | 513 (13.95%) | 16,378 (14.55%) | 1858 (13.67%) | 671 (11.54%) | 11,259 (14.86%) |
| Education | ||||||
| Less than 12th Grade | 93 (0.41%) | <11 | 678 (0.60%) | 35 (0.26%) | <11 | 231 (0.30%) |
| High School Diploma | 4364 (19.31%) | 659 (17.92%) | 20,822 (18.50%) | 2077 (15.28%) | 884 (15.20%) | 11,663 (15.39%) |
| Less than Bachelor’s Degree | 10,811 (47.85%) | 1837 (49.96%) | 52,626 (46.75%) | 6587 (48.46%) | 2939 (50.54%) | 35,945 (47.44%) |
| Bachelor’s Degree Plus | 4739 (20.97%) | 751 (20.42%) | 24,725 (21.96%) | 3254 (23.94%) | 1400 (24.08%) | 17,936 (23.67%) |
| Unknown/Missing | 2587 (11.45%) | 420 (11.42%) | 13,723 (12.19%) | 1641 (12.07%) | 583 (10.03%) | 9993 (13.19%) |
| Number of dependents b | ||||||
| 0 Dependents | 2530 (11.20%) | 408 (11.10%) | 13,467 (11.96%) | 1624 (11.95%) | 571 (9.82%) | 9880 (13.04%) |
| 1 −3 Dependents | 7238 (32.04%) | 1248 (33.94%) | 34,252 (30.43%) | 6241 (45.91%) | 2786 (47.91%) | 33,384 (44.06%) |
| 4+ Dependents | 12,826 (56.77%) | 2021 (54.96%) | 64,855 (57.61%) | 5729 (42.14%) | 2458 (42.27%) | 32,504 (42.90%) |
| Insurance | ||||||
| HMO | 2790 (12.35%) | 484 (13.16%) | 16,434 (14.60%) | 1343 (9.88%) | 575 (9.89%) | 7918 (10.45%) |
| POS | 15,145 (67.03%) | 2446 (66.52%) | 74,364 (66.06%) | 10,366 (76.25%) | 4355 (74.89%) | 57,035 (75.28%) |
| Other | 4659 (20.62%) | 747 (20.32%) | 21,776 (19.34%) | 1885 (13.87%) | 885 (15.22%) | 10,815 (14.27%) |
| Region | ||||||
| Great Lakes/Northern Plains | 5659 (25.05%) | 1080 (29.37%) | 27,261 (24.22%) | 4165 (30.64%) | 2031 (34.93%) | 21,616 (28.53%) |
| Mountain | 2198 (9.73%) | 339 (9.22%) | 10,147 (9.01%) | 1333 (9.81%) | 591 (10.16%) | 8144 (10.75%) |
| Northeast | 2333 (10.33%) | 423 (11.50%) | 11,128 (9.89%) | 1293 (9.51%) | 590 (10.15%) | 7601 (10.03%) |
| Pacific | 2088 (9.24%) | 252 (6.85%) | 13,234 (11.76%) | 1581 (11.63%) | 549 (9.44%) | 8098 (10.69%) |
| Southeast | 10,303 (45.60%) | 1582 (43.02%) | 50,658 (45.00%) | 5210 (38.33%) | 2052 (35.29%) | 30,245 (39.92%) |
| Unknown/Missing | 13 (0.06%) | <11 | 146 (0.13%) | 12 (0.09%) | <11 | 64 (0.08%) |
| Poverty level c | ||||||
| <250% FPL | 2434 (10.77%) | 405 (11.01%) | 11,753 (10.44%) | 3079 (22.65%) | 1381 (23.75%) | 17,414 (22.98%) |
| 250–400% FPL | 3413 (15.11%) | 540 (14.69%) | 15,925 (14.15%) | 3300 (24.28%) | 1435 (24.68%) | 18,273 (24.12%) |
| >400% FPL | 9797 (43.36%) | 1588 (43.19%) | 48,953 (43.49%) | 4898 (36.03%) | 2160 (37.15%) | 26,089 (34.43%) |
| Unknown/Missing | 6950 (30.76%) | 1144 (31.11%) | 35,943 (31.93%) | 2317 (17.04%) | 839 (14.43%) | 13,992 (18.47%) |
Small cells with values <11 redacted.
HMO = Health Maintenance Organization; POS = Point of Service; FPL = Federal Poverty Level.
Data presented as N (% of Overall).
Adults and Children.
Calculated using number of dependents.
Among delivering people with CP, 4.02% had an anxiety diagnosis, 7.30% had a depression diagnosis, and 2.67% had a concomitant AD diagnosis in 2008. In comparison, 16.13% of delivering people with CP had an anxiety diagnosis, 4.05% had a depression diagnosis, and 9.78% had a concomitant AD diagnosis in 2021. Further, among non-delivering people with CP, 5.98% had an anxiety diagnosis, 9.30% had a depression diagnosis, and 3.82% had a concomitant AD diagnosis in 2008. In comparison, 18.73% of non-delivering people with CP had an anxiety diagnosis, 5.5% had a depression diagnosis, and 13.95% had a concomitant AD diagnosis in 2021.
Figure 1 shows the distribution of specific pain conditions among delivering and non-delivering people with CP. Among delivering and non-delivering people, musculoskeletal and pelvic pain were the most common diagnosed chronic pain conditions. Comparing the delivering and non-delivering groups, we found significantly higher proportions of people with chronic pelvic pain (12.45% vs. 7.35%, respectively, p <0.001) and musculoskeletal pain (18.69% vs. 8.79%, respectively, p <0.001). However, we found similar proportions of headache (3.19% vs. 3.29%, p = 1.00), but significantly lower proportions of fibromyalgia (0.18% vs. 0.30%, p < 0.001), and widespread pain (0.08% vs. 0.28%, p < 0.001) underlie CP conditions in delivering people compared to non-delivering people. When comparing delivering and non-delivering people with AD, we identified significantly higher proportions of musculoskeletal (28.38% vs. 17.46%, p < 0.001) and pelvic pain (19.75% vs 14.57%, p < 0.001). We also found significantly lower proportions of headache (9.36% vs. 10.22%, p = 0.001), widespread pain (0.20% vs. 0.82%, p < 0.001), fibromyalgia (0.80% vs. 1.90%, p <0.001).
Fig. 1.
Distribution of specific chronic pain conditions among those with pain diagnoses by delivering status and anxiety/depression diagnosis.
Note, some individuals might have more than one pain condition therefore totals across listed conditions may >100%. An * indicates statistical significance, and NS indicates no statistical significance.
Figure 2 shows the marginal predicted probabilities of CP for delivering people between 2008 and 2021.
Fig. 2.
Trends of chronic pain among delivering people, 2008–2021.
Note, the marginal predicted probabilities of chronic pain above control for age, race, household income, and region.
The marginal predicted probability of CP ranged from 0.400 to 0.527 and 0.221–0.261 among delivering people with AD and all delivering people, respectively. In all years, the marginal predicted probabilities of CP for delivering people with AD were persistently higher compared to all delivering people. For context, the marginal predicted probabilities of CP for non-delivering people were reported in the Appendix Figure. In 2021, the marginal predicted probability of CP for delivering people was higher than for non-delivering people (0.418 (95% CI 0.411–0.427) vs 0.349 (95% CI 0.347–0.351) and 0.256 (95% CI 0.252–0.259) vs 0.152 (95% CI 0.151–0.153), for those with AD and overall, respectively).
4. Discussion
This study demonstrates that CP conditions among a large, national sample of all delivering people, including those with and without AD, remains common. Previous studies have not well characterized patterns of CP conditions among delivering people and its co-existence with AD. Recognizing the significant co-existence of CP and AD among delivering people has crucial relevance for efforts aimed at reducing mental health-related maternal morbidity and mortality. [25] Interventions aimed at reducing suicidality and suicide itself among pregnant people must also account for the high prevalence of co-existing pain conditions.
Recognizing the prevalence and clinical overlap of CP conditions and AD among pregnant people is a crucial component of clinical care. An essential first step involves identifying and treating AD around the time of pregnancy to reduce the alarming rate of maternal morbidity and mortality related to mental health conditions. [11] Currently recommended AD screening tools for use during pregnancy do not routinely screen for CP. [11] Following detection, effective treatment should also include addressing CP conditions, which may require coordinating care with specialized pain treatment programs. Clinicians must also recognize the bidirectional relationship between pain and mental health. Numerous studies, including functional neuroimaging studies, support this relationship suggesting that overlapping brain areas appear altered in CP and mental health disorders. [10]
The strengths of this study included its observation of a very large, national sample. These data allowed us to observe diagnoses and health service use in groups of delivering and non-delivering people over time. However, even given the extreme sample size of our study population, our findings do not generalize to uninsured, inconsistently employed, or Medicaid-covered people because they are not represented in this sample. Optum data may be incomplete in recent years (e.g., 2021) and because it draws from individual enrolled in employer sponsored health insurance, shifts in employment patterns could impact shifts in unidentified population characteristics over time. Furthermore, reliance on diagnostic codes may underestimate both CP and AD. Because we used diagnostic codes and insurance claims data, we were unable to account for parity, which may be associated with an elevated risk of chronic pain. [26] Finally, the timing of observed level change in CP prevalence aligns with the adoption of ICD-10 in 2015–2016. Therefore, while it may appear that annual prevalence of chronic pain conditions in our sample population declined over time, this observed trend is more likely related to changes in billing codes.
This study provides new and clinically relevant information for clinical care. Previous studies on chronic pain and pregnancy focused on the development of chronic pain after delivery, and treatment [25] (e.g., opioid prescribing patterns) during pregnancy, but limited literature addressed the prevalence of chronic pain during pregnancy. Our findings align with one small study of recently postpartum people reporting that 38% of participants had a chronic pain condition prior to pregnancy. [19] Another small, single site study (N = 156) reported that 28.2% of people who were referred to a perinatal moods disorder clinic also had a chronic pain condition. [18] According to a systematic review of studies on interventions to treat pelvic pain during pregnancy, approximately one-fifth experience pelvic pain, which corresponds to findings from this study. [5] Our study also aligns with musculoskeletal pain as a commonly experienced pain condition during pregnancy and with a higher prevalence relative to non-delivering individuals. [6] By examining CP and AD in a large, national sample, our findings indicate that CP remains extremely common during pregnancy, especially among people with AD. Our clinical approach to screening and treatment of AD during pregnancy should recognize the magnitude of this problem.
Mental health problems are increasing, including rates of AD during the perinatal period. These mental health problems bring serious consequences, especially as contributors to maternal morbidity and mortality. Among 421 pregnancy-related deaths occurring between 2008 and 2017 reviewed in 14 states, 11% had causes associated with mental health conditions. [16] During a time when more commonly recognized causes of maternal mortality, like sepsis and hemorrhage, have declined, death by suicide remains high, accounting for about 20% of maternal deaths in the postpartum period. [17] Non-delivering individuals with chronic pain also have an increased risk for suicide; death by suicide occurs at rates as much as two-fold higher than the general population. [10,27] Recognizing the significant co-existence of CP and AD among delivering people has crucial relevance for efforts aimed at reducing mental health-related maternal morbidity and mortality. Specifically, interventions aimed at reducing suicidality and suicide itself among pregnant people must also account for the high prevalence of co-existing pain conditions. Although we do not know whether chronic pain conditions mitigate perinatal suicide risk, future intervention studies should consider this possibility. For example, non-pharmacological transdiagnostic treatments such as acceptance and commitment-based therapy and cognitive behavioral therapy may prove effective in reducing pain and mental health issues based on their benefits shown in other populations. [28]
In this large, national sample about half of delivering people with AD also had a CP condition. To become effective, efforts to address the alarmingly high, and growing rate of mental health-related maternal morbidity and mortality in the U.S. must recognize the co-occurrence of CP and AD. Furthermore, the relatively high proportions of delivering people with CP conditions also highlight the need for safe treatment options and novel care coordination systems for this vulnerable population.
Acknowledgments
We would like to acknowledge the support of Erin Miller and Sarah Block in the technical support for this manuscript.
Funding
The National Institutes of Health provided funding for this study (R01 MH120124; R01 MD014958).
Appendix A
Table 1A:
Sample characteristics of non-delivering people with chronic pain (CP), chronic pain with anxiety and/or depression (CP and AD), and overall, 2008 and 2021.a
| 2008 | 2021 | |||
|---|---|---|---|---|
| CP | CP and AD | CP | CP and AD | |
| Sample Size | 284,732 | 67,230 | 145,855 | 90,052 |
| Age, years | ||||
| 15 −19 | 27,276 (9.58%) | 5589 (8.31%) | 14,133 (9.69%) | 10,368 (11.51%) |
| 20–24 | 24,802 (8.71%) | 5481 (8.15%) | 16,096 (11.04%) | 12,588 (13.98%) |
| 25–29 | 39,023 (13.71%) | 8826 (13.13%) | 17,055 (11.69%) | 11,148 (12.38%) |
| 30–34 | 51,507 (18.09%) | 12,365 (18.39%) | 25,751 (17.66%) | 15,262 (16.95%) |
| 34–39 | 66,954 (23.51%) | 16,557 (24.63%) | 33,225 (22.78%) | 18,891 (20.98%) |
| 40–44 | 75,170 (26.4%) | 18,412 (27.39%) | 39,595 (27.15%) | 21,795 (24.2%) |
| Race/ethnicity | ||||
| Asian | 10,547 (3.7%) | 1222 (1.82%) | 6094 (4.18%) | 2038 (2.26%) |
| Black | 28,421 (9.98%) | 4927 (7.33%) | 13,824 (9.48%) | 7792 (8.65%) |
| Hispanic | 31,434 (11.04%) | 5405 (8.04%) | 17,521 (12.01%) | 8172 (9.07%) |
| White | 178,128 (62.56%) | 47,141 (70.12%) | 92,337 (63.31%) | 63,232 (70.22%) |
| Unknown/Missing | 36,202 (12.71%) | 8535 (12.7%) | 16,079 (11.02%) | 8818 (9.79%) |
| Education | ||||
| Less than 12th Grade | 1110 (0.39%) | 159 (0.24%) | 413 (0.28%) | 169 (0.19%) |
| High School Diploma | 61,833 (21.72%) | 14,067 (20.92%) | 25,886 (17.75%) | 15,840 (17.59%) |
| Less than Bachelor’s Degree | 138,823 (48.76%) | 33,436 (49.73%) | 74,020 (50.75%) | 46,580 (51.73%) |
| Bachelor’s Degree Plus | 52,897 (18.58%) | 12,469 (18.55%) | 32,172 (22.06%) | 20,374 (22.62%) |
| Unknown/Missing | 30,069 (10.56%) | 7099 (10.56%) | 13,364 (9.16%) | 7089 (7.87%) |
| Number of dependents b | ||||
| 1 −3 dep | 144,531 (50.76%) | 36,067 (53.65%) | 74,968 (51.4%) | 48,513 (53.87%) |
| 4–7 dep | 108,307 (38.04%) | 23,563 (35.05%) | 56,142 (38.49%) | 33,561 (37.27%) |
| >8 dep | 2508 (0.88%) | 648 (0.96%) | 1628 (1.12%) | 1000 (1.11%) |
| Unknown/Missing | 29,386 (10.32%) | 6952 (10.34%) | 13,117 (8.99%) | 6978 (7.75%) |
| Insurance | ||||
| HMO | 37,030 (13.01%) | 9217 (13.71%) | 15,938 (10.93%) | 10,175 (11.3%) |
| POS | 189,738 (66.64%) | 44,335 (65.95%) | 106,848 (73.26%) | 64,000 (71.07%) |
| Other | 57,964 (20.36%) | 13,678 (20.35%) | 23,069 (15.82%) | 15,877 (17.63%) |
| Region | ||||
| Great Lakes/Northern Plains | 64,299 (22.58%) | 17,837 (26.53%) | 44,609 (30.58%) | 28,791 (31.97%) |
| Mountain | 25,846 (9.08%) | 5732 (8.53%) | 14,617 (10.02%) | 9009 (10.00%) |
| Northeast | 28,460 (10.00%) | 7668 (11.41%) | 12,428 (8.52%) | 8830 (9.81%) |
| Pacific | 26,759 (9.4%) | 4852 (7.22%) | 15,445 (10.59%) | 7466 (8.29%) |
| Southeast | 139,141 (48.87%) | 31,109 (46.27%) | 58,585 (40.17%) | 35,912 (39.88%) |
| Unknown/Missing | 227 (0.08%) | 32 (0.05%) | 171 (0.12%) | 44 (0.05%) |
| Poverty level c | ||||
| <250% FPL | 22,542 (7.92%) | 5497 (8.18%) | 23,448 (16.08%) | 14,383 (15.97%) |
| 250–400% FPL | 44,234 (15.54%) | 10,759 (16.00%) | 44,703 (30.65%) | 29,000 (32.2%) |
| >400% FPL | 135,292 (47.52%) | 30,962 (46.05%) | 57,364 (39.33%) | 35,379 (39.29%) |
| Unknown/Missing | 82,664 (29.03%) | 20,012 (29.77%) | 20,340 (13.95%) | 11,290 (12.54%) |
Data presented as N (%).
Adults and Children.
Calculated using number of dependents HMO = Health Maintenance Organization; POS = Point of Service; FPL = Federal Poverty Level.
Fig. A1.
Trends of chronic pain among non-delivering people, 2008–2021
Note, the marginal predicted probabilities of chronic pain above control for age, race, household income, and region.
Appendix Table 2A:
ICD-9 and ICD-10 codes used to identify anxiety and depression disorders.
| Group | ICD Diagnosis Code |
|---|---|
| Anxiety | 300.xx, 308.xx, 313.xx, 293.xx, F06.xx, F40.xx, F41.xx, F42.xx, F43.xx, F48.xx, R45.xx |
| Depression | 311.xx, 296.xx, 300.xx, F32.xx, F33.xx |
Footnotes
Prior presentations
An early version of findings from this analysis appeared at the North American Society for Psychosocial Obstetrics and Gynecology (NASPOG) Conference in Ann Arbor, MI, April 23, 2022.
CRediT authorship contribution statement
Vanessa K. Dalton: Conceptualization, Methodology, Formal analysis, Writing – original draft. Andrea Pangori: Formal analysis, Methodology, Writing – original draft. Sawsan As-Sanie: Conceptualization, Data curation, Methodology, Investigation, Writing – original draft. Karen M. Tabb: Conceptualization, Formal analysis, Writing – original draft. Stephanie Hall: Supervision, Project administration, Investigation, Writing – original draft. Anca Tilea: Project administration, Investigation, Writing – original draft. Amy Schroeder: Conceptualization, Writing – review & editing. Jennifer Burgess: Data curation, Formal analysis, Methodology. Kara Zivin: Funding acquisition, Conceptualization, Methodology, Formal analysis, Writing – original draft.
References
- [1].Zelaya CE, Dahlhamer JM, Lucas JW, Connor EM. Chronic pain and high-impact chronic pain among U.S. adults, 2019. NCHS Data Brief 2020;390:1–8. [PubMed] [Google Scholar]
- [2].National Institutes of Health, National Pain Strategy. Interagency Pain Research Coordinating Committee; 2022. [Google Scholar]
- [3].Smith BH, et al. The impact of chronic pain in the community. Fam Pract 2001;18 (3):292–9. [DOI] [PubMed] [Google Scholar]
- [4].Mills SEE, Nicolson KP, Smith BH. Chronic pain: a review of its epidemiology and associated factors in population-based studies. Br J Anaesth 2019;123(2). e273–e83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Liddle SD, Pennick V. Interventions for preventing and treating low-back and pelvic pain during pregnancy. Cochrane Database Syst Rev 2015;9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Fiat F, et al. The Main changes in pregnancy-therapeutic approach to musculoskeletal pain. Medicina 2022;58:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Miller LR, Cano A. Comorbid chronic pain and depression: who is at risk? J Pain 2009;10(6):619–27. [DOI] [PubMed] [Google Scholar]
- [8].Bair MJ, Robinson RL, Katon W, Kroenke K. Depression and pain comorbidity: a literature review. Arch Intern Med 2003;163(20):2433–45. [DOI] [PubMed] [Google Scholar]
- [9].Munce SEP, Stewart DE. Gender differences in depression and chronic pain conditions in a national epidemiologic survey. Psychosomatics 2007;48(5):394–9. [DOI] [PubMed] [Google Scholar]
- [10].Hooten WM. Chronic pain and mental health disorders: shared neural mechanisms, epidemiology, and treatment. Mayo Clin Proc 2016;91(7):955–70. [DOI] [PubMed] [Google Scholar]
- [11].American College of Obstetricians and Gynecologists. ACOG committee opinion no. 757: screening for perinatal depression. Obstet Gynecol 2018;132(5). e208–e12. [DOI] [PubMed] [Google Scholar]
- [12].Gavin NI, et al. Perinatal depression: a systematic review of prevalence and incidence. Obstet Gynecol 2005;106(5 Part 1):1071–83. [DOI] [PubMed] [Google Scholar]
- [13].Fawcett EJ, Fairbrother N, Cox ML, White IR, Fawcett JM. The prevalence of anxiety disorders during pregnancy and the postpartum period: a multivariate Bayesian Meta-analysis. J Clin Psychiatry 2019;80:4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Glazer KB, Howell EA. A way forward in the maternal mortality crisis: addressing maternal health disparities and mental health. Arch Womens Ment Health 2021;24 (5):823–30. [DOI] [PubMed] [Google Scholar]
- [15].Campbell J, Matoff-Stepp S, Velez ML, Cox HH, Laughon K. Pregnancy-associated deaths from homicide, suicide, and drug overdose: review of research and the intersection with intimate partner violence. J Women’s Health 2021;30(2):236–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [16].Trost SL, et al. Preventing pregnancy-related mental health deaths: insights from 14 US maternal mortality review committees, 2008–17. Health Aff 2021;40(10): 1551–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [17].Chin K, Wendt A, Bennett IM, Bhat A. Suicide and maternal mortality. Curr Psychiatry Rep 2022;24(4):239–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [18].Ray-Griffith SL, Morrison B, Stowe ZN. Chronic pain prevalence and exposures during pregnancy. Pain Res Manag 2019;2019:6985164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [19].Munro A, George RB, Chorney J, Snelgrove-Clarke E, Rosen NO. Prevalence and predictors of chronic pain in pregnancy and postpartum. J Obstet Gynaecol Can 2017;39(9):734–41. [DOI] [PubMed] [Google Scholar]
- [20].Edwards RR, Dworkin RH, Sullivan MD, Turk DC, Wasan AD. The role of psychosocial processes in the development and maintenance of chronic pain. J Pain 2016;17(9, Supplement):T70–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Till SR, As-Sanie S, Schrepf A. Psychology of chronic pelvic pain: prevalence, neurobiological vulnerabilities, and treatment. Clin Obstet Gynecol 2019;62(1): 22–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Schrepf A, et al. ICD-10 codes for the study of chronic overlapping pain conditions in administrative databases. J Pain 2020;21(1–2):59–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Healthcare Cost and Utilization Project. Clinical classifications software (CCS) for ICD-9-CM. Available from: https://www.hcup-us.ahrq.gov/toolssoftware/ccs/ccs.jsp; 2017.
- [24].Agency for Healthcare Research and Quality. Beta Clinical Classifications Software (CCS) for ICD-10-CM/PCS. Healthcare Cost and Utilization Project (HCUP); 2018. Available from: www.hcup-us.ahrq.gov/toolssoftware/ccs10/ccs10.jsp. [Google Scholar]
- [25].Ray-Griffith SL, Wendel MP, Stowe ZN, Magann EF. Chronic pain during pregnancy: a review of the literature. Int J Women’s Health 2018;10:153–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Aggarwal SR, et al. Higher number of live births is associated with left ventricular diastolic dysfunction and adverse cardiac remodelling among US Hispanic/Latina women: results from the echocardiographic study of Latinos. Open Heart 2017;4 (1):e000530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Racine M Chronic pain and suicide risk: a comprehensive review. Prog Neuro-Psychopharmacol Biol Psychiatry 2018;87:269–80. [DOI] [PubMed] [Google Scholar]
- [28].McCracken LM, Yu L, Vowles KE. New generation psychological treatments in chronic pain. BMJ 2022;376:e057212. [DOI] [PubMed] [Google Scholar]