The distinct longitudinal impact of pain catastrophizing on pain interference among youth living with sickle cell disease and chronic pain.

Eligible participants were children and adolescents with SCD receiving care at one of three comprehensive sickle cell clinics within a southeastern children’s hospital. Eligibility criteria were: (a) 10–18 years old, (b) diagnosed with SCD (any genotype), (c) report chronic pain (i.e., patient endorsed having pain on most days per month, persisting for at least 3 months), and (d) English fluency. Exclusion criteria included: (a) comorbid medical conditions typically associated with pain but unrelated to SCD (e.g., rheumatologic disorders or inflammatory bowel disease); and (b) significant cognitive or developmental limitations, as per their healthcare provider or parent, that would impair completion of self-report measures (e.g., previous overt strokes, receiving chronic transfusion indicated for central nervous system risks or complications).

Of the 134 families approached for participation, 96 patients met eligibility criteria, 14% ( n = 19) declined or were not interested in research participation, and 10% (n = 14) expressed interest in being re-approached for consent at a later time. 63 patient-parent dyads consented to study participation (82% consent rate) and 48 patients (76%) completed 4-month follow-up. There were no sources of selective bias associated with patients who completed or did not complete 4-month follow-up on patient demographic (e.g., age, sex, income) or clinical characteristics (e.g., SCD genotype, disease-modifying treatments). Sample characteristics are described in Table 1 . Participants (Range = 10–18; M age = 13.90; SD = 2.53) primarily identified as female (58.3%) and Black or African American (95.1%). Most had SCD genotype hemoglobin (Hb) HbSS (68.9%) followed by HbSC (14.7%) and HbSβ + thalassemia (8.2%). Participants had missed an average of 35.16 days of school in the past year (SD = 29.93, range = 0–180). Sixty percent of the sample reported a family income of $30,000 or less per year. Descriptive statistics on pain characteristics, internalizing symptoms, pain catastrophizing, and pain interference revealed skewness and kurtosis within the limits of a normal distribution. At baseline, participants reported an average pain intensity of 5.05 ( SD = 2.24) and 77% reported experiencing pain on most days of the month ( M = 14.20, SD = 9.73 pain days per month). On average, participants reported moderate severity of anxiety ( M = 59.40; SD = 8.48) and depressive symptoms ( M = 63.09, SD = 8.16). All participants endorsed pain catastrophizing with the overall mean falling in the high catastrophizing range ( Pielech et al., 2014 ; Range = 5–52, M = 26.16, SD = 11.57). Average pain interference at four-months follow-up fell within the severe range ( M = 66.34; SD = 7.46). Further, 49.2% of participants reported severe levels of pain interference. Correlational associations among child age, pain, and internalizing variables were examined (see Table 2 ). Neither pain frequency (pain days per month) nor average pain intensity correlated with pain interference, pain catastrophizing, anxiety, or depression. There were no differences in pain interference based on SCD genotype severity (HbSS and HbSB0 versus HbSC and HbSB +), F (1,59) = 0.899, p = 0.35 or disease-modifying treatments, such as a history of chronic transfusion therapy, F (1,44) = 0.836, p = 0.37, or hydroxyurea, F (1,44) = 0.287, p = 0.60. There were no differences in child pain catastrophizing, pain interference at baseline or follow-up, anxiety, or depressive symptoms by child sex (all p ’s > 0.60). Child age, SCD type, and pain intensity were retained in subsequent analyses based on existing theoretical and empirical support of the relation between age, disease severity, and pain characteristics on overall quality of life and functioning (e.g., Gil et al., 1993 ). In a linear hierarchical regression analysis, block 1 (child SCD genotype, age, average pain intensity at baseline, and baseline pain interference) was significant and accounted for 17% of the variance in pain interference at 4-month follow-up, F (4,58) = 2.684, p < 0.05. The inclusion of psychosocial functioning (anxiety, depression) in block 2 was significant, F (6,58) = 2.474, p < 0.05, and accounted for an additional 6% of the variance. Finally, the inclusion of pain catastrophizing in block 3 was significant, F(7,58) = 2.832, p < 0.01, and accounted for an additional 6% of the variance, with the entire model accounting for 28% of the variance. In the final model, pain catastrophizing was the only unique predictor of pain interference at 4-month follow-up, above and beyond baseline anxiety, baseline depression, SCD genotype, baseline age, baseline pain frequency, and baseline pain interference, B = 0.323, t (61) = 2.02, p < 0.05 (See Table 3 ). For every one-point increase in pain catastrophizing, the average increase in pain interference at 4-month follow-up was about 0.32.

As part of a larger prospective observational longitudinal study, participants completed a battery of measures on pain and psychosocial functioning at a baseline assessment and brief assessment of pain and pain-related impairment at a four-month follow-up visit. The Patient-Reported Outcome Measurement Information System (PROMIS) Pediatric Short Form—Pain Interference is an 8-item patient-report measure that assesses the consequences of pain on aspects of daily life including social, cognitive, emotional, and physical engagement in the past 7 days (e.g., “I had trouble doing schoolwork when I had pain”). Response options range from 1 (never) to 5 (almost always). Domain total scores generate a T-score with a mean of 50 and a standard deviation of 10. Higher scores indicate more pain interference and are categorized as within normal limits (≤ 50), mild (51–55), moderate (56–65), or severe (≥ 66) ( Mann et al., 2020 ; Morgan et al., 2017 ). This measure has shown adequate internal consistency (α = 0.88), adequate test–retest reliability among pediatric samples (α = 0.65), and good convergent validity with the Functional Disability Inventory ( Kashikar-Zuck et al., 2016 ; Varni et al., 2010 , 2014 ). There is also support for scale sensitivity to change over time among children and adolescents with SCD and youth with chronic pain ( Kashikar-Zuck et al., 2016 ; Reeve et al., 2018 ). Within the present study, pain interference items had adequate inter-item reliability (α = 0.88). Participants completed this measure at baseline and at a four-month follow-up visit after their baseline assessment. The Pain Catastrophizing Scale, child version (PCS-C; Crombez et al., 2003 ), is a 13-item self-report measure of overly negative attitudes of pain. Each item begins with “When I am in pain…” and respondents rate items on a 1 (not at all) to 5 (extremely) scale to indicate how strongly they have each thought (e.g., “I feel I can’t go on”). Higher scores indicate more pain catastrophizing and are categorized as low (0–14), moderate (15–25), or high (≥ 26) levels of catastrophizing ( Pielech et al., 2014 ). Within the present study, items showed good inter-item reliability (α = 0.91). This measure has been found to have good reliability and validity among pediatric pain samples ( Crombez et al., 2003 ; Welcom et al., 2013). Participants completed this measure at baseline assessment. PROMIS Pediatric Short Form–Anxiety was used to assess patient-report of anxiety-related symptoms (e.g., “My worries overwhelmed me”) in the past 7 days ( Ader, 2007 ) at baseline assessment. This scale consists of 8 items with response options ranging from 1 (never) to 5 (almost always). Domain total scores generate a T-score with a mean of 50 and a standard deviation of 10. Higher scores indicate worse symptoms of anxiety and are categorized as within normal limits (≤ 50), mild (51–55), moderate (56–65), or severe (≥ 66) ( Mann et al., 2020 ; Morgan et al., 2017 ). Inter-item reliability within the present study was strong (α = 0.92). There is ample support for reliability and validity the PROMIS anxiety domain among families of youth with chronic health conditions including sickle cell disease (e.g., Cox et al., 2020 ). The PROMIS Pediatric Short Form–Depressive Symptoms was used to assess patient-reported depressive symptoms over the past 7 days ( Ader, 2007 ) at baseline assessment. This scale consists of 8 items (e.g., “I was less interested in doing things I usually enjoy”) with response options ranging from 1 (never) to 5 (almost always). Domain total scores generate a T-score with a mean of 50 and a standard deviation of 10. Higher scores indicate worse symptoms of depression and are categorized as within normal limits (≤ 50), mild (51–55), moderate (56–65), or severe (≥ 66) ( Mann et al., 2020 ; Morgan et al., 2017 ). Internal consistency within the present study was strong (α = 0.92). The depression PROMIS domain has shown adequate reliability and validity among youth with sickle cell disease ( Cox et al., 2020 ). Participants rated their pain intensity and pain frequency at baseline assessment. Consistent with assessments among pediatric pain samples, pain intensity was based on adolescent report of their average pain intensity over the past 2 weeks using a numeric rating scale ranging from 0 (no pain) to 10 (worst possible pain) ( Farrar et al., 2001 ; Myrvik et al., 2013 ). Participants reported on pain frequency by reporting total pain days per month (0–31), consistent with pain frequency items developed and based on previous research ( Sil et al., 2016 , 2021 ). Caregivers completed a demographic questionnaire at baseline that assessed adolescent and caregiver age, gender, ethnicity, SCD genotype, household income, and number of school days missed in previous year.

Following institutional review board approval, patients were screened for eligibility and approached by phone or in-person for participation in an observational longitudinal study on the natural course of chronic pain in youth with SCD. Potentially eligible patients were identified from an existing research database of families who agreed to be contacted for future research and initial review of electronic medical records. All patients were approached and screened during their routine outpatient SCD appointment. Parents or legal guardians of participating patients provided written consent for their child and for themselves to be eligible for inclusion, and youth provided written assent. Patients and parents were provided the option to complete web-based (via REDCap) or paper measures at home, in-clinic while waiting for their medical provider, or during their next scheduled outpatient clinic appointment.

Consistent with the extant literature in non-SCD chronic pain conditions ( Tran et al., 2015 ; Turk & Wilson, 2010 ; Vervoort et al., 2006 ; Wertli et al., 2014 ), this study suggests that baseline pain catastrophizing uniquely impacted pain interference four months later above and beyond that of baseline child age, SCD genotype, baseline pain interference, baseline pain intensity, baseline anxiety, and baseline depressive symptoms. Specifically, for every one-point increase in pain catastrophizing, pain interference at a 4-month follow up increased by 0.32. This study replicates the small body of literature focused on pain catastrophizing in pediatric SCD by demonstrating that all patients endorsed pain catastrophizing with average severity within the high range. Study findings also provide a unique contribution to the field by highlighting that pain catastrophizing, in conjunction with other biological and psychological factors, predicts pain interference at 4-month follow-up among youth with chronic SCD pain. Among youth with chronic SCD pain, pain catastrophizing warrants greater consideration as an important predictor that influences pain management and overall functioning. In line with the fear-avoidance model of chronic pain, catastrophic thinking about a painful experience can contribute to pain-related fear and anxiety ( Asmundson et al., 2012 ). For those with chronic pain, anxiety and pain-related activity avoidance can develop into a vicious, self-perpetuating cycle that contributes to significant impairments in daily functioning. Given the ongoing exposure to SCD pain experiences, youth with chronic SCD may experience catastrophic thinking about pain that contributes to activity avoidance and long-term and persistent functional impairments ( Tran et al., 2015 ; Vervoort et al., 2006 ; Wertli et al., 2014 ). These patterns of high levels of pain catastrophizing emerging as a strong and unique predictor of persistent, long-term impairment beyond the effects of demographic or psychosocial variables are consistent with the extant literature on youth and adults with non-SCD related pain presentations (e.g., chronic fatigue syndrome and endometriosis; Martin et al., 2011 ; Meeus et al., 2012 ), and add to the small but growing body of longitudinal research on the psychosocial predictors of functional outcomes in pediatric SCD ( Brown et al., 2006 ; Powers et al., 2002). Moreover, pain catastrophizing has been identified as higher among patients who identify as Black/African American ( Forsythe et al., 2011 ; Meints et al., 2018 ). According to the Communal Coping Model of pain, pain catastrophizing may be an adaptive coping strategy to solicit more social or emotional support from their environment ( Sullivan et al., 2001 , 2006 ). It is possible that individual and systemic racism and healthcare disparities in pain care influence pain catastrophizing ( Meints et al., 2018 ). For example, some patients may develop and express catastrophic views of their pain in order to be taken seriously or in order to receive more empathetic care from medical providers. Healthcare providers’ increased awareness of pain catastrophizing may help inform their assessment and treatment practices and ensure patient pain experiences are validated to continue optimizing SCD care. There is evidence of long-term benefits from brief (8-min) online trainings among medical providers to improve providers’ perceptions towards patients presenting with SCD pain (see Singh et al., 2016 ). Trainings, such as these, may be an essential next step to reduce negative interactions with patients, which may further reinforce pain catastrophizing over time. Assessment and treatment of psychosocial functioning is an integral part of comprehensive SCD care with a particular emphasis on limiting health-related barriers to education, social relationships, and activity during childhood and adolescents ( NIH, 2002 ). Because pain catastrophizing may lead to less engagement in daily responsibilities over time beyond the effects of anxiety or depressive symptoms for youth with chronic SCD pain, targeted assessment of pain catastrophizing during routine SCD clinic visits may enhance both prevention and intervention services. For example, youth with chronic SCD pain may benefit from targeted services, such as planned activity pacing combined with cognitive restructuring, as patients are likely at higher risk for psychosocial distress and functional impairment due to their chronic pain ( Kazak et al., 2007 ). Given the potential role of racial disparities in pain treatment on pain catastrophizing, it may also be necessary to explore the adaptive nature of pain catastrophizing among youth with chronic SCD pain, such as attending and responding to symptoms to minimize development of serious health complications. Better understanding of maintaining factors can further inform treatment approaches to target pain catastrophizing and interference. Screening and brief interventions for pain catastrophizing may also be essential to reduce persistent functional impairment for youth with chronic SCD pain. The PSC-C has been identified as an easy, feasible measure (11 items) for use within medical settings to assess for pain catastrophizing with clinical reference points to support quick interpretation of findings ( Crombez, 2003 ; Pielech, 2014 ). Youth identified with severe range of pain catastrophizing may benefit from referrals to cognitive-behavioral treatment to target negative thoughts and beliefs associated with pain ( Pielech, 2014 ; Zagustin, 2013 ). Brief cognitive-behavioral interventions may also be helpful for reducing symptoms. For instance, a one-session cognitive-behavioral therapy (CBT) intervention targeting pain catastrophizing among adults with non-SCD chronic pain conditions found large effect sizes in reducing pain catastrophizing at two- and four-weeks post treatment (Cohen’s d = 0.85 and d = 1.15; Darnall et al., 2014 ). Moreover, studies among adults with non-SCD pain have also found that targeting pain catastrophizing initially in treatment has subsequent benefits on perceived pain intensity such that early reductions in pain catastrophizing longitudinally improved pain outcomes but not vice versa ( Burns et al., 2003 ; Campbell et al., 2012 ). While these findings are promising, current research among pain catastrophizing interventions primarily targets adult populations with non-SCD related pain. More research with pain catastrophizing interventions among youth with SCD is needed as it is highly prevalent (100% within the present study) and impacts daily activities. Considering the genetic and life-limiting nature of SCD in which SCD pain may be associated with disease-related medical complications, some engagement in pain catastrophizing may be adaptive to facilitate appropriate medical care. As such, youth with chronic SCD pain may require tailored interventions targeting pain catastrophizing that addresses the unique context of the disease. The interpretation of these study findings should be considered within the context of a few limitations. It is possible that the negative effects of pain catastrophizing persist beyond 4 months and may be reactive to inter-individual variability in disease-related complications. Additional studies are required to examine the association between catastrophizing and pain interference across longer periods of time (e.g., 12 months, 24 months) while considering the stability or change in disease severity. Future studies with a larger sample size will help enhance generalizability of findings and allow consideration of additional biopsychosocial predictors (e.g., medical complications, parent pain catastrophizing) to inform targeted interventions to reduce catastrophic thinking and persistent pain interference.

All data were collected via REDCap ( http://projectredcap.org ), a secure Web application for online surveys and databases, and then imported into the Statistical Package for Social Sciences Version 26 (SPSS-26) for analysis. Data were examined for missingness and baseline characteristics of participants with and without missing data were compared to evaluate potential sources of bias or selective attrition. Upon examination of the data, the proportion of missing data values on variables of interest at baseline and 4-month follow-up ranged from 3 to 6%, which were primarily item-level variables on the Pain Catastrophizing Scale or PROMIS Pain Interference. Based on preliminary analysis of missing values, multiple imputation was selected to handle missing data because it produces more accurate estimates for longitudinal designs with missing observations at baseline and follow-up assessments than standard methods of handling missing data ( Huque et al., 2018 ; Rubin, 1987 ). We generated 5 imputed datasets to achieve a relative efficiency of at least 0.97 with the percentage of missing values. Results when including participants with missing data from imputed datasets were very similar to results from original, observed data (Shaefer, 1997). Descriptive statistics on demographic and child-reported variables were examined to determine whether the data met the underlying assumptions of the proposed analytic procedures (e.g., skewness, kurtosis). Next, bivariate correlational analyses were used to help identify potential sociodemographic and clinical characteristics to be included as covariates. A hierarchical linear regression model was used to examine biopsychosocial predictors of pain interference at four-month follow-up. Baseline biological and clinical characteristics and baseline pain interference were entered into block 1 of the regression model. Block 2 included baseline anxiety and depressive symptoms. Finally, baseline pain catastrophizing was entered into block 3. Change in R-squared was used to evaluate the predictive impact of psychosocial variables on the overall model. A post-hoc power analysis for linear multiple regression indicated power > 0.90 given the medium effect size detected for pain catastrophizing uniquely predicting pain interference at 4-month follow-up.

Pediatric sickle cell disease (SCD) is an inherited red blood cell disorder that affects approximately one in every 365 Black or African American births and one in every 16,300 Hispanic American births in the United States ( CDC, 2020c ). Globally, SCD affects about 112 out of 100,000 live births ( Wastenedge et al., 2018 ). SCD symptoms can include, but are not limited to fatigue, swelling (hand-foot syndrome), anemia, leg ulcers, blood clotting, pain, and infections ( CDC, 2020a ). Moreover, symptoms can cause serious health problems such as damage to body organs, acute chest syndrome, and stroke ( CDC, 2020a ). Pain is a hallmark feature among youth with SCD and can range from acute, intermittent (e.g., vaso-occlusive episodes), to chronic in nature ( Dampier et al., 2017 ; Fingar et al., 2019 ). The severity of pain from vaso-occlusion and disease complications can vary across individuals and genotypes (e.g., higher severity among sickle cell anemia, such as hemoglobin (Hb) SS and HbSβ 0 thalessemia vs HbSC or HbSβ + thalassemia; CDC, 2020b ; Driscoll, 2007 ). Approximately 23% of youth with SCD report chronic pain, defined as pain on most days persisting for at least 6 months ( Dampier et al., 2017 ; Sil et al., 2016 ). Youth with chronic SCD pain are more likely to experience pain-related interruptions in day-to-day activity (i.e., pain interference), which has been related to poorer quality of life, impaired academic performance, less social engagement, worse emotional functioning (i.e., anxiety and depression), and poorer social skills that can all persist into adulthood ( Crosby et al., 2015 ; Lim et al., 2019 ; Valrie et al., 2020 ). Among youth with SCD, pain interference can be exacerbated by pain intensity, fatigue, anxiety, depressive symptoms, and maladaptive coping ( Alberts et al., 2020 ; Barakat et al., 2007 ; Donohoe & Smith, 2019 ; Miller et al., 2020 ; Reader et al., 2019 ). Drawing upon the broader pediatric pain literature, the fear avoidance-model of pain behavior describes overlapping factors that can impact pain and impairment including fear of pain, anxiety, rumination, anxiety-sensitivity, and catastrophic thinking about pain ( Asmundson et al., 2012 ). Within this model and aligning with previous literature, pain catastrophizing refers to self-reported rumination, magnification, helplessness, or an exaggerated view associated with pain or anticipated pain ( Crombez et al., 2003 ; Sulivan et al., 1995 ; Turk & Wilson, 2010 ; Quartana, 2009 ). Pain catastrophizing is among one of the strongest predictors of pain outcomes (e.g., functional disability, pain severity, somatic symptoms, effectiveness of pain interventions) in both adult and pediatric chronic pain samples ( Tran et al., 2015 ; Vervoort et al., 2006 ; Wertli et al., 2014 ). Moreover, among the broader population of children and adolescents with chronic pain, pain catastrophizing has been identified as having a unique impact on functional outcomes above and beyond psychosocial and pain-related factors, such as anxiety and depression ( Holroyd et al., 2007 ; Jastrowski Mano et al., 2012 ; Tran et al., 2015 ). Pain catastrophizing has been associated with poorer recovery from pain, more sick leave, poorer daily functioning, and greater pain interference among children and adults with chronic fatigue and pain syndromes, highlighting the persistent and long-term negative impact on overall functioning ( Martin et al., 2011 ; Meeus et al., 2012 ; Noel et al., 2015 ; Wertli et al., 2014 ; Westman et al., 2011 ). Recently, the potential influence of pain catastrophizing has been examined among youth with SCD in cross-sectional investigations. Consistent with other chronic pain conditions, pain catastrophizing was related to higher pain intensity, more functional impairment, and reduced health-related quality of life among youth with SCD ( Bakshi et al., 2018 ; Sil et al., 2016 ; Stone et al., 2020 ). Yet, the unique influences of pain catastrophizing beyond psychosocial factors on pain-related functioning, as well as the potential long-term impact of pain catastrophizing on pain impairment among youth with SCD remain largely unknown. The role of pain catastrophizing among youth with SCD may be especially unique within the context of known racial/ethnic disparities that are evident in pain assessment and treatment ( Bonham, 2001 ; Green et al., 2003 ; Hoffman et al., 2016 ), that may not be well-reflected in existing research among the general chronic pain population that includes predominantly White individuals ( CDC, 2020c ; Dahlhamer et al., 2018 ; Murray et al., 2020 ). Higher levels of pain catastrophizing have been identified among African Americans in comparison to White Americans, which, in alignment with the Communal Coping Model of pain, could be related to a history of health care disparities (e.g., pain minimizing; see Sullivan et al., 2006 ) ( Forsythe et al., 2011 ; Meints et al., 2018 ). Given the robust evidence of the negative impact of pain catastrophizing on immediate and long-term health outcomes among pediatric chronic pain populations, examining the role of pain catastrophizing within the context of pediatric SCD may help guide targeted psychosocial screening, prevention, and intervention services to reduce the negative impact of pain on the lives of youth with SCD. The purpose of the present study was to examine the unique, predictive impact of pain catastrophizing on pain interference among youth with chronic SCD pain. It was expected that greater pain catastrophizing would predict greater pain interference at four-month follow-up above and beyond baseline biological (e.g., SCD type, pain intensity, pain interference) and baseline psychosocial (anxiety and depression) factors in children and adolescents with SCD and chronic pain.