Predictive factors for chronic pain related to childbirth: A prospective cohort study

Beata Molin: conceptualization; methodology; formal analysis; validation; investigation; writing – original draft; writing – review and editing; project administration. Anna‐Karin Berger: conceptualization; methodology; writing – review and editing. Anna Sand: conceptualization; validation; writing – review and editing. Susanne Georgsson: conceptualization; methodology; validation; supervision; writing – review and editing. All authors have read and approved the final version of the manuscript and contributed to its revision.

The study was approved by the Regional Ethical Review Board in Stockholm on 4 March 2015 (Ref. 2015/236‐31). Participants received both verbal and written information about the study. They were informed that participation was voluntary and were assured that their data would be treated confidentially. The participants provided written informed consent to participate in the study. Due to ethical restrictions and participant confidentiality, the dataset used in this study is not publicly available. Summary data are available from the corresponding author upon reasonable request.

This work was supported by a grant from the Capio Research Foundation (grant no. 2016‐2900, 2017‐2996) and funding from Sophiahemmet University, Stockholm. The funding sources were not involved in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

Of the 1171 women included in the analysis, 89 (7.6%) reported chronic pain related to labour at 8 months postpartum, whereas 1082 (92.4%) reported no such pain. Table  1 presents characteristics obtained from the baseline questionnaire and the medical records, together with breastfeeding status obtained from the eight‐month follow‐up questionnaire, for the total study population and stratified by pain status at 8 months postpartum. In the multivariable logistic regression analysis, pain‐related conditions before pregnancy were independently associated with chronic postpartum pain, including menstrual pain (aOR 2.07, 95% CI 1.24–3.47), dyspareunia (aOR 2.08, 95% CI 1.12–3.89), and chronic pain (aOR 2.11, 95% CI 1.19–3.74). Pain lasting more than 3 months during pregnancy (aOR 2.14, 95% CI 1.24–3.70) as well as moderate or severe acute postpartum pain (NRS 4–10; aOR 2.16, 95% CI 1.16–4.02) were also associated with chronic postpartum pain. Characteristics of the study population, presented for the total sample and stratified by chronic pain status at 8 months postpartum. Note : Pain refers to chronic pain related to labor reported at 8 months postpartum. N denotes the total number of women in each analytical group, whereas values in the table are presented as n (%). Percentages were calculated among women with available data for the respective variable within each group. Variable‐specific denominators vary because of missing data and, for some variables, subgroup‐specific applicability; available and missing n are reported for each variable. Unless otherwise stated, variables were obtained from the baseline questionnaire. Pain and breastfeeding status at 8 months were obtained from the eight‐month follow‐up questionnaire, and obstetric variables were obtained from Obstetrix. Includes diagnoses such as hypertension, diabetes mellitus and asthma. Vacuum extraction. In the crude analysis, women who had a caesarean section had 1.73 times the odds of chronic pain related to childbirth compared to those with vaginal delivery (28/255 vs. 61/916; OR 1.73, 95% CI 1.08–2.77). Results from the crude and multivariable logistic regression analyses are presented in Tables  2 and 3 , respectively. Crude Analysis – odds ratios for chronic pain after childbirth. Note : * p value ≤ 0.2. Abbreviations: CI, confidence interval; OR, odds ratio. Multivariable Analysis – adjusted odds ratios for chronic pain 8 months after childbirth. Note : Bold values indicate statistical significance ( p < 0.05). Abbreviations: aOR, adjusted odds ratio; CI, confidence interval.

In this large cohort from a Nordic healthcare setting with universal and largely free maternity care, we identified several predictive factors associated with chronic pain 8 months postpartum. Overall, the results suggest that women with pain‐related conditions before or during pregnancy, as well as those experiencing moderate or severe acute postpartum pain, are at increased risk of developing chronic postpartum pain. Our findings support previous research suggesting that pre‐existing pain may increase vulnerability to chronic postpartum pain. 2 , 6 , 7 , 20 , 21 In our study, several pain‐related conditions present before pregnancy remained independently associated with chronic postpartum pain in the multivariable analysis. In addition, prolonged pain during pregnancy was independently associated with the outcome. Rather than representing isolated clinical entities, menstrual pain, dyspareunia, and chronic pain may reflect a common underlying pain vulnerability. One possible explanation for this increased vulnerability involves altered pain processing. Alterations in endogenous pain modulation and central sensitization have been implicated in the development and persistence of chronic pain states. 35 , 36 , 37 , 38 Central sensitization refers to increased responsiveness of the central nervous system, leading to amplified pain signals even in the absence of substantial peripheral input. 35 Over time, neuroplastic changes driving central sensitization may shift from being reversible to becoming more permanent and resistant to treatment. 18 , 35 , 38 Within this framework, women entering pregnancy with pre‐existing pain conditions may be particularly vulnerable, as the additional nociceptive input from pregnancy and childbirth has the potential to further reinforce and perpetuate these processes. In addition, pain lasting more than 3 months during pregnancy may contribute to sensitization before childbirth, thereby increasing susceptibility to chronic postpartum pain. 2 , 6 , 35 From a clinical perspective, the identified pain conditions may therefore be viewed as markers of a sensitized pain system and as indicators of a subgroup of women who warrant closer antenatal attention. 2 , 6 , 35 In this context, relevant antenatal strategies may include assessment of a woman's pain history, pain assessment, and individualized planning of peripartum care. Such an approach may facilitate early identification of women at increased risk of chronic postpartum pain and support more individualized care. However, whether such strategies reduce the incidence of chronic postpartum pain remains to be determined in future intervention studies. Moderate and severe acute postpartum pain also emerged as a strong predictor of chronic pain 8 months postpartum. These results are in line with studies demonstrating that high‐intensity acute pain after obstetric trauma or surgery increases the risk of chronic pain. 6 , 7 , 21 , 39 In addition, acute postoperative or post‐traumatic pain is recognized as one of the strongest biological risk factors for the transition from acute to chronic pain. 18 , 38 , 40 Several mechanisms may explain why women with moderate or severe acute postpartum pain are at increased risk of chronic postpartum pain. Higher intensity of acute postpartum pain may represent a marker of an underlying individual pain vulnerability, predisposing some women to both more intense acute pain and chronic postpartum pain. Alternatively, acute postpartum pain itself may contribute to peripheral and central sensitization, thereby facilitating the transition to chronic pain. 2 , 6 , 35 Our findings therefore reinforce the importance of systematic assessment and appropriate management of acute postpartum pain. Although prevention of obstetric injury remains an important objective in maternity care, equal attention should be directed towards timely recognition and treatment of acute postpartum pain, particularly among women with pre‐existing pain conditions. Although caesarean delivery was associated with chronic postpartum pain in the crude analysis, the association was no longer statistically significant after adjustment. This finding is consistent with previous studies reporting no long‐term differences in maternal pain outcomes between caesarean and vaginal delivery. 2 It has been suggested that the greater tissue injury associated with caesarean delivery results in more severe acute postoperative pain, and that pain intensity rather than mode of delivery may be a more important determinant of chronic postpartum pain. 6 Consistent with this interpretation, moderate and severe acute postpartum pain remained associated with chronic postpartum pain in our study, reinforcing the importance of timely recognition and effective management of acute postpartum pain. A major strength of this study is the large prospective multicentre cohort, which enabled evaluation of multiple potential predictive factors for chronic postpartum pain. Another strength is that the outcome was defined based on self‐reported pain onset in relation to childbirth, rather than on pain presence at follow‐up alone, thereby reducing the risk that pre‐existing pain conditions were misclassified as childbirth‐related chronic pain. Pain intensity was assessed using the validated Numeric Rating Scale (NRS). The questionnaires were developed by an interprofessional research team, based on systematic evaluation of patient‐reported outcomes, and pilot tested among postpartum women to ensure clarity and relevance. Questionnaire development could have been further strengthened through active involvement of pregnant women in the design process. 41 Such involvement may enhance the clarity, relevance, and comprehensibility of survey items and support more accurate interpretation of responses. Several limitations should be considered. Some exposure categories contained relatively few observations. To improve the robustness of the multivariable analyses, variables with very sparse distributions were excluded from the final model because they produced unstable odds ratio estimates with wide confidence intervals. Consequently, potentially relevant associations for these variables may have been missed. Although analyses stratified by mode of delivery could have provided additional clinical information, sparse data in several exposure categories resulted in unstable estimates and limited precision. Consequently, we focused the final analyses on the overall cohort to obtain more robust and reliable effect estimates. Future studies with larger cohorts are needed to determine whether risk factors differ according to mode of delivery. Recruitment procedures were not systematically documented, preventing calculation of an exact participation rate and introducing the possibility of selection bias. It is possible that women who experienced particularly complicated deliveries, regardless of mode of birth, were less likely to participate or to complete follow‐up, which may have influenced the representativeness of the cohort. The follow‐up rate among baseline participants was high (77.7%), which may have reduced the risk of attrition bias, although selective loss to follow‐up cannot be ruled out. Certain demographic groups were underrepresented, including younger women and foreign‐born women, 42 partly due to the Swedish language requirement. This limits generalizability and may have resulted in underestimation of chronic pain prevalence. Finally, data were collected in two Swedish regions, predominantly in the Stockholm region, which may limit the generalizability of the findings.

Our findings suggest that women with pain‐related conditions before or during pregnancy, together with those experiencing higher intensity of acute postpartum pain, are at increased risk of chronic postpartum pain. Importantly, several of these factors can be identified before childbirth, providing an opportunity to recognize women at increased risk. While pre‐existing pain conditions may help identify vulnerable women, acute postpartum pain represents a clinically important factor that warrants timely assessment and management. Future studies with larger cohorts are needed to confirm these findings and determine how this knowledge can best be translated into clinical practice.

Although childbirth is a natural process, it is commonly associated with pain and tissue injury. Pregnancy itself is linked to a high prevalence of pain conditions, particularly low back and pelvic girdle pain. 1 , 2 During vaginal delivery, women may sustain injuries such as perineal lacerations of varying degrees or levator ani muscle trauma, 2 , 3 , 4 while caesarean section also involves considerable tissue damage. 5 Chronic postpartum pain has been reported in 0.3–56% of women, 2 , 5 , 6 , 7 , 8 , 9 with wide variation explained by differences in design, populations, and the lack of distinction between pre‐existing and new‐onset pain. 5 , 6 , 7 In our previous study, the prevalence of pain 8 months postpartum was 7.6%, 9 supported by a prospective design, large sample size, high response rate, and by excluding pre‐existing pain from the outcome definition. To align with international definitions of chronic pain, 10 , 11 we defined pain with onset in relation to labor and persisting for 8 months as chronic . This approach ensures consistency with IMMPACT 10 and IASP 11 recommendations and strengthens comparability with the broader chronic pain field. Postpartum recovery, both in terms of physical and psychological aspects, is essential for maternal health and quality of life. Pain that persists after labor can have a profound impact on daily functioning, maternal–infant bonding, and overall well‐being. 12 , 13 , 14 Despite its prevalence and considerable impact, chronic pain remains an understudied and under‐discussed topic in maternal healthcare. According to recent research, women are often not informed about risks of developing long‐lasting pain conditions, and pain is often insufficiently recognized, assessed, or managed within maternal healthcare. 15 , 16 Lack of information and support may exacerbate suffering and contribute to the development and persistence of pain. 17 Given that pain can become more complex in its pathophysiology and more difficult to treat over time, 18 , 19 early identification, prevention, and treatment are essential. Identifying predictive factors for pain persisting after childbirth is important for targeted management. Previous studies focusing on pain after a caesarean section have identified risk factors such as pre‐existing pain 20 , 21 and severe acute postoperative pain. 21 , 22 Postpartum vaginal and perineal pain have been associated with operative delivery and the extent of perineal trauma. 6 , 7 , 23 , 24 , 25 Risk factors for chronic back pain include pre‐existing back pain and back pain during pregnancy, 26 , 27 obesity, and pregnancy‐related weight gain. 27 , 28 In addition, psychosocial factors, including depression, anxiety, pain catastrophizing, and limited social support, play an important role in the persistence of pain. 29 However, much of the existing research is limited by small sample sizes, leaving important knowledge gaps in this field. The aim of the present study was to identify predictive factors for chronic pain at 8 months postpartum with onset in relation to labor.

We conducted an exploratory analysis of data collected between April and December 2015 in a prospective, multicentre cohort study at six hospitals in the Stockholm region (population 2.24 million) and at Västerås Central Hospital in Västmanland County (population 260 000). This study was designed and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for observational studies. Participants were recruited through convenience sampling at the participating study sites. Eligible participants were women who had given birth at one of these hospitals and provided written informed consent. Exclusion criteria were inability to read or understand Swedish and stillbirth. Data were collected using two self‐administered questionnaires: a baseline questionnaire administered 24–36 h postpartum (Supplemental Information Appendix  S1 ) and an eight‐month follow‐up questionnaire (Supplemental Information Appendix  S2 ), together with obstetric data obtained from the Obstetrix medical record system. The questionnaires, developed by the research team (pain researchers, midwives, and an obstetrician) following a systematic literature review and evaluation of existing patient‐reported outcomes (PROs), were used. Pilot testing through face‐to‐face validation with 15 women led to revisions to two questions on socio‐economic status and pain localization. The baseline questionnaire was distributed by midwives on the maternity ward within 24–36 h postpartum, covering socio‐demographics (age, marital status, educational level and country of birth), body mass index (BMI), somatic diseases (high blood pressure, diabetes and asthma) as well as anxiety or depression before and during pregnancy. Pain‐related variables included menstrual pain, dyspareunia, endometriosis, and chronic pain lasting more than 3 months before pregnancy, as well as the presence and duration of pain during pregnancy. The questionnaire also assessed subjective childbirth experience as well as catastrophizing and acute postpartum pain (within the first 2 days after childbirth). As recruitment was carried out by clinical staff during routine care, the total number of eligible women who were approached and informed about the study was not systematically recorded. The principal investigator visited all sites to brief staff and ensure consistent distribution procedures. Completed forms were collected by the researcher. Of 1580 returned questionnaires, 73 (4.6%) were excluded for incomplete data, leaving 1507 valid responses. The eight‐month follow‐up questionnaire was mailed to the participants with a prepaid return envelope, followed by two reminders to non‐respondents. It assessed whether women had experienced constant or recurrent pain during the past week (yes/no). Those reporting pain were asked to specify pain location, onset, frequency, and intensity using a structured questionnaire. Pain location was reported for predefined anatomical regions including the pelvis, lower back, coccyx, abdomen, vaginal and perineal areas, anal region, lower extremities, and, in women with caesarean delivery, the surgical scar. For each pain location, women indicated when the pain began (before pregnancy, during pregnancy, in relation to labor, or weeks to months after childbirth). The questionnaire also assessed breastfeeding status. In total, 1171 (77.7%) out of 1507 women completed the eight‐month follow‐up questionnaire and were included in the current study population. Obstetric data, such as mode of delivery, vacuum extraction, third‐ or fourth‐degree perineal tears, and type of CS, were obtained from the patient record system, Obstetrix. Data on perineal tears of grades 1 or 2 were unavailable. The primary outcome was chronic postpartum pain, assessed in the eight‐month follow‐up questionnaire. Pain was classified as childbirth‐related if the woman reported that it had started in relation to labor and was still present at 8 months postpartum. This definition did not exclude women with pre‐existing pain conditions; such conditions, including endometriosis, dyspareunia, and chronic pain, were therefore analyzed as potential risk factors. For the estimation of pain intensity, a numerical rating scale, NRS, ranging from 0 to 10, was used, with 0 being “no pain” and 10 being the “worst pain imaginable”. 30 Pain intensity was transformed into severity stages of “mild” (NRS 0–3), “moderate” (NRS 4–7), and “severe” (NRS 8–10). For the analyses, moderate and severe pain were combined into one category (NRS 4–10) and compared with mild pain (NRS 0–3). Childbirth experience was assessed 24–36 h postpartum using a self‐reported numeric rating scale (NRS) ranging from 0 (“very negative”) to 10 (“very positive”). Ratings were categorized as “negative” (0–3), “moderate” (4–7), and “positive” (8–10). This measure captures the woman's subjective emotional and psychological appraisal of the birth experience rather than objective obstetric complications or clinical severity. 31 Pain catastrophizing was assessed using the 13‐item Pain Catastrophizing Scale (PCS; score range 0–52), with scores >30 indicating clinically significant catastrophizing. 32 Anxiety and depression were assessed in the baseline questionnaire using self‐reported single items. Women were asked whether they had experienced anxiety and/or depression before pregnancy or during pregnancy (yes/no). Descriptive statistics were presented for the total sample and separately for women with and without childbirth‐related chronic postpartum pain. Categorical variables are shown as counts and percentages. Missing data were handled using listwise deletion within each analysis, which resulted in varying sample sizes across analyses. Logistic regression was used for crude models for the total cohort. The following independent variables were entered into crude logistic regression models: sociodemographic variables (age, educational level and country of birth), health variables before pregnancy (BMI, chronic pain conditions, menstrual pain, dyspareunia, somatic disease, anxiety/depression), variables during pregnancy (somatic disease and pain), variables during and after childbirth: mode of delivery, parity, subjective childbirth experience, intensity of pain within the first 2 days after childbirth and breastfeeding at 8 months after childbirth. Variables with very sparse distributions, such as marital status, endometriosis before pregnancy, depression/anxiety during pregnancy, and pain catastrophizing, were excluded as sparse cells can lead to unstable odds ratio estimates and wide confidence intervals. Variables with p   ≤  0.2 in the crude analyses were entered simultaneously into the multivariable model. The model used the logit link, and p  < 0.05 was considered statistically significant. Multicollinearity was assessed using variance inflation factors (VIF) and tolerance statistics. All VIF values were within acceptable limits (≤5) and tolerance values exceeded recommended thresholds (>0.18), indicating no substantial multicollinearity. 33 , 34

Appendix S1. Baseline questionnaire administered 24–36 h after childbirth. Appendix S2. Follow‐up questionnaire administered 8 months postpartum.