Parenting,self-regulation,and treatment adherence in pediatric chronic headache: A self-determination theory perspective

Abstract

This study examined parenting factors associated with children’s self-regulation and physician-rated treatment adherence using a self-determination theory framework in pediatric chronic headache. Participants were 58 children and adolescents (aged 10–17 years), who underwent initial and follow-up multidisciplinary evaluation at a headache clinic, and their mothers. Regression analyses showed that higher maternal autonomy support and structure were significantly related to children’s lower treatment-related reactance and higher adherence. Maternal controllingness had associations in the opposite directions. Children’s fear of pain was related to maternal controllingness. Results suggest the importance of parents’ provision of clear expectations and engaging children in treatment problem-solving and decision-making.

Keywords

adherence headache parenting self-determination theory self-regulation

Chronic pediatric headache, including headache and migraine, is the most common chronic pain complaint in pediatric populations and has associations with compromised functioning in school, peer, and family activities (Lipton et al., 2011; Perquin et al., 2000). Children with chronic headache are at risk for physical and psychological symptoms persisting into adulthood (Fearon and Hotopf, 2001). Greater adherence to health behaviors has been found to predict functional improvement over time in children with chronic pain (Simons et al., 2010). Thus, children must take an active role in managing their health, as early engagement in health-promoting behaviors, including medication compliance, proper nutrition, hydration, exercise, and sleep habits, sets the stage for pain management into adulthood (Antonaci et al., 2014).

Given the long-term importance of children with chronic headache engaging in health behaviors, it is imperative to explore factors that may be associated with their participation in headache management. Among such factors are interactions with parents. In the pediatric pain literature, operant behavioral and family systems frameworks have been used to explore social learning factors and family functioning related to children’s pain-related behaviors and disability (Palermo and Chambers, 2005). While these theoretical models have addressed many of the what questions (e.g. parental response styles and family context), there remains an insufficient understanding of why children enact or ignore health behaviors and how parental characteristics may be associated with children’s self-regulation and adherence to treatment recommendations. A motivational perspective is crucial to exploring the processes underlying children’s propensity to engage in behaviors important to treatment and long-term health and the parental environments that may support this development. Therefore, this study applied a self-determination theory (SDT; Deci and Ryan, 1985; Ryan and Deci, 2017) framework to understand factors in the parent–child dyad that may be associated with children’s treatment regulation.

Self-regulation of treatment behaviors

SDT provides a framework for understanding how children with chronic headache may regulate treatment behaviors to manage pain (Ryan and Deci, 2017). SDT proposes that individuals move toward increased self-regulation of their behaviors through the process of internalization, in which individuals move from regulating their behavior due to outside contingencies to greater self, or autonomous regulation. The theory specifies types of motivation along an internalization continuum as well as environmental factors that may facilitate or impede internalization of behaviors. Following the internalization continuum, a child might engage in a treatment behavior, such as taking preventive medication, to avoid punishment or loss of privileges (i.e. external regulation). In introjected regulation, a child may enact a behavior to avoid guilt or to gain approval; for example, a child may take medication to prevent disappointing his or her parents or to avoid feeling guilty for not doing so. In identified regulation, a more autonomous form of regulation, individuals engage in a behavior out of personal value or perceived importance (e.g. taking medication to maintain health). A child demonstrating autonomous regulation engages in behaviors choicefully to fulfill their goals.

When individuals regulate their health behaviors more autonomously, more adaptive outcomes ensue. Research from an SDT perspective has investigated autonomous regulation in chronic health problems such as obesity (e.g. Teixeira et al., 2015) and cardiovascular disease (e.g. Russell and Bray, 2010), with autonomous regulation related to positive physical health outcomes. Despite this evidence, little SDT research has investigated autonomous regulation in chronic pain samples, and no research has been conducted on pediatric chronic pain patients, specifically.

Parenting and children’s self-regulation and reactance

SDT delineates environmental factors that conduce toward internalization by supporting individuals’ innate psychological needs (Ryan and Deci, 2017). Specifically, the environmental dimensions of autonomy support, structure, and involvement are hypothesized to facilitate self-regulation as they help to satisfy individuals’ needs for autonomy (i.e. experience of choice or volition), competence (i.e. being effective in one’s environment), and relatedness (i.e. feeling loved by and connected with others), respectively (Ryan and Deci, 2017). Autonomy support involves parents taking children’s perspectives, allowing choice and input into decision-making, and solving problems jointly (Grolnick and Ryan, 1989). On the other hand, parental controllingness entails parents pressuring children to perform in specific ways and solving problems for them (Grolnick and Ryan, 1989). Structure refers to parents providing clear guidelines, rules, and expectations. Involvement is the extent to which parents are interested in, knowledgeable about, and active in children’s lives.

Relations between parenting dimensions and children’s self-regulation have been examined in home and academic settings but have received less attention in the pediatric literature. Higher levels of parental autonomy support, structure, and involvement have been associated with children’s more autonomous regulation of behaviors, such as schoolwork and chores (Grolnick et al., 2015; Soenens and Vansteenkiste, 2005), while parental controlling behavior has negative correlations with children’s adjustment and functioning in home and school contexts (Barber et al., 2005; Grolnick and Ryan, 1989).

Beyond self-regulation, psychological reactance may be one undesirable correlate of parental controllingness. Psychological reactance (Brehm, 1966) is theorized to occur when an individual perceives their autonomy to be threatened or eliminated, thereby responding in ways to “recapture the freedom(s) affected and preventing the loss of others” (Fogarty, 1997: 1278). Consistent with SDT, an individual whose need for autonomy is compromised may produce defensive behaviors, at times performing the exact opposite behavior of what is requested, to regain a sense of choice and volition (Ryan and Deci, 2017). Research has linked parental controllingness to psychological reactance in adolescent community and clinical samples (e.g. Van Petegem et al., 2015), yet this relation has remained largely unexplored in pediatric chronic illness settings.

Parental autonomy support and structure in the pediatric literature

In the pediatric literature, researchers have explored parenting dimensions related to autonomy support versus controllingness and structure provision, though different terminologies have been used. In a study of children diagnosed with spina bifida, Holmbeck et al. (2002) found that maternal overprotection, defined as intrusiveness (e.g. demanding the child tell the parent everything that happens when away from home), was negatively associated with children’s ratings of decision-making around nonmedical issues (e.g. homework completion) and mothers’ willingness to provide autonomy in the future. In the diabetes literature, higher ratings of parents’ responsiveness (i.e. degree of support, warmth, and love) and lower psychological control (i.e. pressure to comply with rules and standards) were related to better adherence to treatment recommendations over a 1- to 2-week period (Goethals et al., 2017). Specific to physiological health markers, a more authoritative style in parents, characterized by both structure and autonomy support, was positively and directly associated with diabetic adolescents’ adherence (e.g. number of daily blood glucose checks) and indirectly linked to their glycemic control (HbA1c) through adherence and quality of life (Radcliff et al., 2017).

Despite these parallels, differential language and constructs have created confusion regarding relations between parenting and adherence in pediatrics. In diabetes research, the construct of diabetes responsibility has been used (Rubin et al., 1989). This variable is typically coded linearly as diabetes-related management (e.g. determining insulin doses) by child alone (1), joint parent and child (3), and parent alone (5). Declines in this variable (decreased parent responsibility) over time have been associated with deterioration in children’s adherence and poorer metabolic control (e.g. Holmes et al., 2005; Wiebe et al., 2014). This has led some researchers to conclude that “parental control” is positive for adherence. Clearly, parents taking roles in children’s health is important, but with joint responsibility coded in the middle of the scale, it is unclear whether it is beneficial for parents to help organize treatment or to lead and control the treatment. In this study, we differentiate structure (i.e. parents’ organization and informational input) and autonomy support versus control to disentangle various aspects of parental roles and analyze how each is important to children’s self-regulation and adherence.

Predictors of controlling parenting

This study also investigated potential antecedents of controlling parent behavior in the context of children’s chronic headache. Grolnick (2003) conceptualized controlling parenting as influenced by several factors, including pressure from below—child-specific psychological and behavioral difficulties that evoke controlling behavior from the parent and pressure from within—parents’ own emotional and cognitive states generating internal pressure for their children to act in specific ways. This conceptualization depicts parental controllingness as a transactional process between parents and children.

This study examined pressures from below and within by analyzing two important constructs in the pediatric chronic illness literature—parent/child pain catastrophizing and fear of pain—as potential predictors of controlling parenting. Pain catastrophizing refers to a cognitive process characterized by rumination, magnification, and helplessness around pain (Sullivan et al., 1995). In parents, high levels of catastrophic thinking are associated with emotional distress, which, in turn, predicts their attempts to restrict children’s engagement in activities believed to worsen pain (Caes et al., 2011). It may be that parents’ catastrophizing and associated distress may incite pressures from within, fostering controllingness around children’s actions as they attempt to minimize their children’s pain. While parental responses to children’s pain have been examined (Palermo and Chambers, 2005), an understanding of parental catastrophizing in relation to their tendency toward controllingness remains unstudied.

In children, catastrophic thinking involves focus on negative information about their pain experience and functioning (Sullivan et al., 1995). Children’s catastrophizing about their pain has been connected to both positive (e.g. provision of emotional support) and negative (e.g. punishing responses) parental responses (Vervoort et al., 2010). That a child’s catastrophizing over pain may elicit controlling responses from his or her parent as he or she experiences pressure from below is a plausible yet unexplored explanation for controlling parental responses.

Related to, but distinct from, pain catastrophizing is fear of pain, which entails cognitions about the threat of pain and corresponding avoidance behaviors (Simons et al., 2011). A parent who observes a child in pain may experience a heightened awareness of threat and emotional distress, which elicits protection in the form of encouraging the child’s avoidance tendencies; over time, the child may adopt these pain-related fears and corresponding escape behaviors as his or her own. Exploring both parents’ and children’s fear of pain may provide insight into concomitants of parental controllingness in pediatric conditions.

The study also considered children’s level of disability in the examination of parental controllingness and children’s treatment outcomes. Children with chronic pain who struggle to engage in everyday activities, whether due to fear, rumination about their pain, or otherwise, may have difficulty engaging in treatment; this could evoke parental controllingness around treatment as overcompensation for the child’s limitations. Given that parental controlling behavior and lower adherence by the child may be, in part, a function of the child’s level of disability, children’s reported disability was included in the study.

The current study

This study sought to (a) examine parenting factors, specifically, children’s reports of parental autonomy supportive behavior versus controlling behavior and structure, associated with children’s reported regulation of their health behaviors and physician-rated adherence to treatment recommendations in pediatric chronic headache using an SDT framework and (b) explore parent- and child-reported factors that may be associated with parents’ controlling behaviors around children’s treatment.

This study tested the following hypotheses:

Higher levels of child-reported parental autonomy supportive behavior and parental structure and lower controllingness will be associated with children’s lower ratings of external and introjected regulation (i.e. controlled regulation), and higher ratings of identified (i.e. autonomous) regulation of their health behaviors.

Higher levels of child-reported parental autonomy supportive behavior and parental structure and lower controllingness around the children’s treatment will be associated with less psychological reactance reported by the child and greater treatment adherence rated by the physician.

Higher ratings of parent- and child-reported fear of pain and pain catastrophizing will be associated with greater parental controllingness, reported by the child.

Method

Participants

Participants were 58 children and adolescents, aged 10–17 (M = 14.3, SD = 2.0) years, and their mothers. Children and adolescents identified as predominantly European American (77.6%) and female (74.1%), representative of patients attending this clinic. The majority of mothers reported completing an undergraduate or graduate degree (72.7%). The children underwent initial and follow-up multidisciplinary evaluation at a tertiary headache clinic in a large, urban, northeast pediatric hospital in the United States. Patients reported a mean pain rating of 5.5 (SD = 2.7) on a 0–10 scale and an average of 16.9 (SD = 11.5) headache days per month at initial evaluation.

Procedure

Institutional Review Board approvals from the hospital and principal investigator’s university were obtained prior to data collection. Initial evaluation prior to study participation included a medical history and physical exam performed by a pediatric neurologist and a structured clinical interview conducted by a clinical psychologist. Individualized recommendations were provided to the families at the initial visit. Medical records were then screened for participant eligibility. Inclusion criteria for participation were as follows: the child/adolescent (a) was age 10–17 years and (b) had an International Classification of Headache Disorders (3rd edn; ICHD-3 beta) diagnosis of either chronic migraine, chronic tension-type headache, or new daily persistent headache by the neurologist. Exclusion criteria included a child’s developmental delay or cognitive impairment that would impact their ability to complete the questionnaires.

Eligible patients who returned for their follow-up appointment approximately 3 months later (M = 114.2 days, SD = 57.5 days) were greeted by the researcher. In total, 98.4 percent of eligible families approached by the researcher consented to participate. During the follow-up appointment, the neurologist evaluated the child’s adherence using a rating form developed specifically for the study. On completion of the appointment, the parent/guardian and child were each given a set of questionnaires and instructed to complete the surveys independently. Both parent consent and child assent were obtained.

Measures

Demographics

Demographic (e.g. child age, gender, and ethnicity; parents’ education) and medical information (e.g. diagnosis) were collected from a review of patients’ records.

Functional disability

The 15-item Functional Disability Inventory (FDI; Claar and Walker, 2006) assesses children’s self-reported difficulty in physical and psychosocial functioning due to their physical health over the previous 2 weeks. Children rated items on a 5-point Likert-type scale from “no trouble” (0) to “impossible” (4), with higher scores indicating greater functional disability. The FDI has good test–retest reliability and validity (e.g. Claar and Walker, 2006) and good scale reliability in this study (α = .89).

Parenting in children with chronic pain

This 36-item questionnaire was adapted for a pediatric chronic pain population from the valid and reliable Parents as a Social Context Questionnaire (Skinner et al., 2005) and was completed by the child/adolescent. To adapt this measure, items were unchanged but “regarding my treatment plan” was added, prompting children to evaluate their mothers’ parenting in relation to their treatment plan, specifically. This instrument assesses children’s perceptions of mothers’ autonomy support (e.g. “My mother allows me to make my own choices regarding my treatment plan”), structure (e.g. “My mother is clear about what she expects me to do regarding my treatment plan”), and controlling behaviors (e.g. “My mother is always telling me what to do regarding my treatment plan”) around the child’s headache treatment. Children rated items from “not true at all” (1) to “very true” (4). The autonomy support and controllingness subscales were kept separate as these may represent distinct constructs (Ng et al., 2014) with unique patterns of relations to child adjustment (Marbell and Grolnick, 2013). The subscales demonstrated good internal reliability: maternal autonomy support (α = .89), maternal structure (α = .75), and maternal controllingness (α = .94).

Self-Regulation Questionnaire for health behaviors

This questionnaire was adapted for a pediatric chronic pain population from the validated Self-Regulation Questionnaire (SRQ; Ryan and Connell, 1989). The SRQ has been modified and adapted with good validity across multiple domains, including health behaviors (e.g. Treatment Self-Regulation Questionnaire; Levesque et al., 2006). This instrument asks children to rate why they engage in specific health behaviors, including staying hydrated, eating healthy, getting enough sleep, and engaging in exercise. The stem of each question is, “When I [drink water] for my headaches/migraines, I do so because . . .” followed by external reasons (e.g. “My parent(s) insist that I do”), introjected reasons (e.g. “I would feel ashamed of myself if I didn’t”), and identified reasons (e.g. “I believe it is important for my future health”). Children rated their reasons for engaging in each of the health behaviors on a 4-point Likert-type scale from “not true at all” (1) to “very true” (4). External, introjected, and identified items were highly correlated across the different health behaviors (r = .7 or above); therefore, items were averaged across health behaviors to create external, introjected, and identified subscales. Due to the high correlation (r = .74) between the external and introjected subscales, external and introjected items were combined into an aggregate controlled regulation summary score. This procedure has been performed in several studies in the SDT literature (e.g. Levesque et al., 2006). Both the controlled (α = .97) and the identified (α = .93) regulation subscales demonstrated good reliability.

Psychological reactance

This 16-item questionnaire was adapted from the Hong Psychological Reactance Scale (Hong and Page, 1989) to measure children’s reported psychological reactance in a pediatric chronic pain population. To adapt this measure, items were unchanged but “regarding my treatment plan” was added, prompting children to rate their thoughts and feelings in relation to their treatment plan, specifically (e.g. “I do exactly the opposite of what my parents expect me to do regarding my treatment plan”). Children rated items from “strongly disagree” (1) to “strongly agree” (5). A composite “child reactance” summary score was created, with higher scores indicating more reactance. This measure showed good validity and reliability in previous studies (Hong and Page, 1989) and good internal consistency in this study (α = .91).

Physician-rated treatment adherence

At the follow-up visit, the clinic’s neurologist, who was trained on study procedures, asked patients if they attempted or completed the recommendations provided at their initial evaluation and rated their adherence. The neurologist rated the patient as “fully adherent” (1), “partially adherent” (0.5), or “non-adherent” (0) for applicable categories using a rating form created for this study. A patient received a “fully adherent” rating if they engaged in the recommendation, a “partially adherent” rating if they attempted but discontinued the recommendation (with exceptions made for adverse reactions or lack of access/resources), or a “non-adherent” rating if the patient did not attempt the recommendation. In addition to recommendations specific to each individual patient (e.g. medication, physical therapy, and psychological treatment), every child received a “lifestyle modification” recommendation related to sleep hygiene, nutritional considerations, increased hydration, and/or exercise. Ratings for adherence to sleep, nutrition, hydration, and exercise recommendations were averaged to create a “lifestyle adherence” summary score for each patient, which was used in this study.

Fear of pain

The Fear of Pain Questionnaire (FOPQ) includes 24 items assessing individuals’ perceptions of pain-related fears and avoidance behaviors (Simons et al., 2011). Children and their mothers rated items on a scale from “strongly disagree” (0) to “strongly agree” (4). “Fear of Pain” summary scores were computed for mother and child, with higher scores indicating greater fear of pain. Simons et al. (2011) reported good internal consistency, 1-month stability, and construct validity for the child and parent versions. In this study, there was good reliability for both child (α = .93) and parent (α = .92) reports.

Pain catastrophizing

The Pain Catastrophizing Scale (PCS) is comprised of 13 items assessing negative thinking associated with pain (Sullivan et al., 1995). Children and their mothers rated items on a 5-point Likert-type scale from “not at all true” (0) to “very true” (4). Summary “Pain Catastrophizing” scores were computed for mother and child, with higher scores indicating higher levels of catastrophic thinking. The PCS has shown good concurrent and discriminant validity, internal consistency, and reliability in previous studies (e.g. Osman et al., 1997) and good scale reliability in this study for the child (α = .90) and parent (α = .90).

Results

Preliminary analyses

Table 1 presents descriptive statistics. Relations among demographic variables, child-reported functional disability, and outcome variables were examined prior to evaluating study hypotheses. Due to their relations with parenting and fear of pain/catastrophizing variables, mothers’ education, child age, child gender, and functional disability were controlled for in the regression analyses to analyze variance in outcomes above and beyond demographic and disability-related factors.

Table 1.

Means, standard deviations, and correlations among study variables (N = 58).

	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
1. Child age	14.37	2.00	1.00	.05	−.13	−.11	.26*	.13	−.30*	−.21	−.04	.01	.04	−.02	−.10	−.01	−.28*
2. Child gender	1.73	0.45		1.00	.18	−.01	.42**	.29*	−.44***	−.25⁺	.32*	−.06	.17	−.35**	−.13	−.24⁺	−.19
3. Mother education	5.98	1.06			1.00	−.18	.01	.24⁺	−.17	−.11	.26⁺	−.16	.32*	−.28*	−.20	−.40**	−.14
4. Child report of disability	0.79	0.59				1.00	−.18	−.04	.15	.37**	−.04	.32*	−.53**	.16	.44***	.05	.18
5. Child report of maternal autonomy support	2.53	0.41					1.00	.60***	−.61***	−.34*	.16	−.36**	.37**	−.07	−.32*	−.08	−.27*
6. Child report of maternal structure	2.46	0.73						1.00	−.57***	−.24⁺	.27*	−.52***	.36**	−.07	−.24⁺	−.12	−.25⁺
7. Child report of maternal control	1.84	0.58							1.00	.34**	−.21	.49***	−.34**	.16	.37**	.32*	.40**
8. Child report of controlled regulation	2.43	0.72								1.00	.10	.33**	−.37**	.22	.33**	.26*	.37**
9. Child report of identified regulation	3.65	0.44									1.00	−.08	.09	−.24⁺	−.01	−.24⁺	−.01
10. Child report of reactance	1.68	0.64										1.00	−.26⁺	.14	.36**	.17	.31*
11. Physician-reported adherence	0.73	0.27											1.00	−.17	−.43***	−.17	−.22⁺
12. Parent report of fear of pain	1.25	0.62												1.00	.22	.68***	.29*
13. Child report of fear of pain	1.72	0.77													1.00	.32*	.77***
14. Parent report of pain catastrophizing	1.49	0.73														1.00	.56***
15. Child report of pain catastrophizing	1.69	0.74															1.00

p < .10; *p < .05; **p < .01; ***p < .001.

Correlations were conducted to examine associations among the variables. Children’s reports of mothers’ autonomy support were negatively correlated with children’s controlled self-regulation, while maternal controllingness was positively related to children’s controlled regulation. Children’s reports of mothers’ provision of structure were positively associated with identified regulation. Children’s reports of maternal autonomy support and structure were positively related to physician ratings of children’s adherence to lifestyle recommendations and negatively correlated with psychological reactance; children’s reports of mothers’ controllingness showed associations in the opposite direction, consistent with study hypotheses. As hypothesized, children’s reported fear of pain and mothers’ and children’s pain catastrophizing were positively correlated with mothers’ controllingness.

Primary analyses

A series of simultaneous regression analyses were conducted to examine relations between children’s perceptions of mothers’ parenting behaviors and outcome variables. Regressions were conducted separately for the parenting variables of maternal autonomy support, controllingness, and structure to eliminate multicollinearity.

Analyses regressing outcomes onto maternal autonomy support

Regression analyses were first conducted to analyze relations between children’s perceptions of maternal autonomy support and outcome variables (Table 2), controlling for child age, gender, mother education, and functional disability. Functional disability was significantly related to controlled regulation, t(53) = 2.26, β = 0.31, p < .05, such that children who reported more disability from their headaches reported more controlled regulation. Beyond the effects of disability, children’s reports of maternal autonomy support were marginally significantly and negatively associated with children’s controlled regulation of treatment behaviors, t(53) = −1.64, β = −0.26, p < .10. For identified regulation, only child gender was marginally significant, with higher identified regulation reported among girls.

Table 2.

Regressions of study outcomes onto independent variables (N = 58).

	Child age			Child gender			Mother education level			Functional disability			Maternal autonomy support			R ²
	t	β	SE	t	β	SE	t	β	SE	t	β	SE	t	β	SE
Controlled regulation	−0.84	−0.12	0.05	0.04	0.01	0.24	−0.39	−0.06	0.09	2.26	0.31*	0.01	−1.64	−0.26⁺	0.26	0.21
Identified regulation	−0.01	−0.01	0.03	1.78	0.27⁺	0.16	1.60	0.23	0.06	−0.06	−0.01	0.01	0.51	0.08	0.18	0.18
Child reactance	0.46	0.06	0.68	1.22	0.18	3.55	−0.80	−0.11	1.33	1.75	0.23⁺	0.15	−3.11	−0.46***	3.95	0.27
Lifestyle adherence	−0.34	−0.04	0.02	−0.88	−0.11	0.08	2.31	0.27*	0.03	−3.64	−0.43**	0.01	2.78	0.35**	0.09	0.43
	Child age			Child gender			Mother education level			Functional disability			Maternal controllingness			R ²
	T	β	SE	T	β	SE	T	β	SE	T	β	SE	T	β	SE
Controlled regulation	−0.58	−0.08	0.05	0.07	0.01	0.24	−0.08	−0.01	0.09	2.37	0.32*	0.01	1.54	0.24	0.18	0.21
Identified regulation	−0.04	−0.01	0.03	1.82	0.28⁺	0.16	1.50	0.21	0.06	−0.11	−0.02	0.01	−0.38	−0.06	0.12	0.17
Child reactance	1.24	0.16	0.66	1.70	0.23⁺	3.31	−0.09	−0.01	1.24	2.01	0.25*	0.14	4.23	0.60***	2.47	0.36
Lifestyle adherence	−0.50	−0.06	0.02	−0.59	−0.08	0.08	1.75	0.22	0.03	−3.78	−0.45***	0.01	−2.04	−0.28*	0.06	0.39
	Child age			Child gender			Mother education level			Functional disability			Maternal structure			R ²
	T	β	SE	T	β	SE	T	β	SE	T	β	SE	T	β	SE
Controlled regulation	−1.01	−0.14	0.05	−0.43	−0.06	0.02	−0.04	−0.01	0.09	2.54	0.35*	0.01	−0.97	−0.14	0.24	0.18
Identified regulation	−0.05	−0.01	0.03	1.92	0.27⁺	0.15	1.25	0.18	0.06	−0.20	−0.03	0.01	1.71	0.24⁺	0.16	0.20
Child reactance	0.36	0.04	0.60	0.84	0.10	2.96	0.46	0.06	1.22	2.71	0.32**	0.14	−4.75	−0.58***	3.15	0.40
Lifestyle adherence	−0.07	−0.01	0.02	−0.16	−0.02	0.08	1.47	0.18	0.03	−4.14	−0.49***	0.01	2.30	0.28*	0.08	0.40

p < .10; *p < .05; **p < .01; ***p < .001.

For reactance, there was a significant effect of maternal autonomy support, t(53) = −3.11, β = −0.46, p < .001, with higher maternal autonomy support associated with lower psychological reactance. For physician’s ratings of lifestyle adherence, there was a significant positive effect of mother education level, t(53) = 2.31, β = 0.27, p < .05, and a significant negative effect of functional disability, t(53) = −3.64, β = −0.43, p < .01. Finally, children’s reports of higher maternal autonomy support were associated with greater adherence, t(53) = 2.78, β = 0.35, p < .01.

Regression analyses for maternal controllingness

Regression analyses were next conducted for maternal controllingness (Table 2), controlling for the same variables as the above analyses. Children’s reports of maternal controllingness were not significantly related to either controlled or identified regulation. However, there was a significant effect of maternal controlling behavior on children’s reports of reactance, t(53) = 4.23, β = 0.60, p < .001, such that higher controllingness was associated with more reactance. Furthermore, higher reports of maternal controllingness were related to lower physician ratings of lifestyle adherence, t(53) = −2.04, β = −0.28, p < .05.

Regression analyses for maternal structure

Regression analyses were also conducted to examine relations for children’s perceptions of maternal structure (Table 2), controlling for the same variables as the above analyses. For identified regulation, there was a marginally significant effect of children’s reports of maternal structure, t(53) = 1.71, β = 0.24, p < .10, with higher structure provision associated with more identified regulation of treatment behaviors. Similar to children’s reports of autonomy support, higher reports of maternal structure were associated with lower reactance, t(53) = −4.75, β = −0.58, p < .001, and higher physician ratings of lifestyle adherence, t(53) = 2.30, β = 0.28, p < .05.

Regression analyses of maternal controllingness onto child and parent fear of pain and pain catastrophizing

Simultaneous regression analyses were also performed to evaluate correlates of maternal controllingness, with a focus on fear of pain and pain catastrophizing reported by both mothers and children (Table 3). Because mothers and their children reported on the same constructs and parent–child dyads are assumed to have mutual influences (i.e. nonindependence) on each other’s reports, it was important to account for the dyadic nature of these data. Failing to account for the nonindependence of the data may result in biased significance testing (Kenny et al., 2006). Following the procedure described by Kenny et al. (2006), a Pearson product-moment correlation coefficient was computed between the mother and child reports of fear of pain and pain catastrophizing, and each respective correlation was included in the regression analyses. After accounting for the data nonindependence, results showed that children’s reports of fear of pain were significantly and positively related to mothers’ controlling behaviors, t(52) = 2.34, β = 0.28, p < .05. Contrary to study hypotheses, there was no significant effect of mothers’ report of fearing their children’s pain on children’s perceptions of their controllingness. Furthermore, neither child nor parent report of pain catastro-phizing was related to mothers’ controlling behavior.

Table 3.

Regressions of child report of maternal controllingness onto child and parent reports of Fear of Pain (FoP) and Pain Catastrophizing (PC) (N = 58).

	t	β	SE
FoP
Child age	−2.08	−0.24*	0.02
Child gender	−3.86	−0.45**	0.11
Mother education	0.20	0.02	0.05
Functional disability	0.41	0.05	0.01
Child report of FoP	2.34	0.28*	0.01
Parent report of FoP	−0.70	−0.08	0.01
r_{childFoP, parentFoP}	1.87	0.27⁺	37.4
R²			0.33
PC
Child age	−1.77	−0.21⁺	0.02
Child gender	−3.78	−0.46**	0.11
Mother education	0.29	0.04	0.05
Functional disability	1.06	0.13	0.01
Child report of PC	1.25	0.18	0.01
Parent report of PC	−0.15	−0.02	0.01
r_{childPC, parentPC}	3.47	0.56**	15.3
R²			0.29

p < .10; *p < .05; **p < .01; ***p < .001.

Discussion

This study explored the (a) associations between maternal autonomy support, controllingness, and structure and children’s self-regulation, treatment adherence, and treatment-related reactance and (b) possible correlates of maternal controlling behavior in pediatric chronic headache. This study is novel in its use of an SDT motivational framework to examine parenting in relation to children’s treatment behaviors in pediatric chronic headache. This theoretical perspective adds a more nuanced understanding of how parenting styles around children’s chronic pain treatment relate to children’s regulation and enactment of treatment and addresses conflicting findings with regard to parenting constructs in the pediatric literature. In addition, this study used multiple reporters to examine various aspects and correlates of children’s chronic pain management. With regard to the findings, the more the children perceived their mothers as autonomy supportive and as providing structure around their treatment behaviors, the lower their psychological reactance and the higher their adherence to physician-recommended lifestyle modifications; associations were found in the opposite direction for maternal controllingness. Furthermore, children who reported more fear of pain reported higher levels of maternal controllingness. These results and their implications are discussed further below.

First, the parenting dimensions of autonomy support, controllingness, and structure were expected to be related to children’s self-regulation. There was only partial support for this hypothesis. Correlational analyses showed that maternal autonomy support was negatively associated and controllingness was positively associated with controlled regulation, suggesting that children who perceive their mothers as offering input and choices around their treatment are less likely to use external contingencies, such as threat, reward, or fear of disappointment, to regulate treatment behaviors. Children viewing their mothers as controlling, however, are more reliant on these forms of external regulation. As children’s use of controlled regulation in this study was related to their lower treatment adherence, these findings may have implications for management of chronic pain. With regard to identified regulation, there were no significant relations between maternal autonomy support or controllingness and children’s identified regulation. Interestingly, provision of structure was associated with children’s identified (but not controlled) regulation, suggesting that parenting characterized by predictability and clarity of expectations might link to children’s internalization of important health behaviors.

When controlling for functional disability, as well as demographic variables, in regression analyses, however, the above relations between parenting variables and self-regulation were attenuated. Children’s reports of maternal autonomy support were marginally significantly negatively associated with children’s controlled regulation, and maternal structure was marginally positively related to their identified regulation. There were no significant relations between maternal controllingness and either controlled or identified regulation in the regressions.

These findings suggest that part of the relation between parenting and children’s self-regulation may be explained by children’s disability. Children who perceived more activity limitations due to pain, such as socializing with friends or completing homework, reported being more externally regulated for their treatment behaviors. Children who are more disabled by their headaches may find it difficult to engage in treatment behaviors, or alternatively, children who rely more on contingencies to regulate their treatment may come to believe they cannot engage in various activities. In any case, taking into account levels of disability is an important factor in children’s treatment regulation (and in following recommendations, as disability was also associated with adherence) and should be considered in future studies.

Hypothesis 2, which addressed relations between parenting and children’s reactance and adherence, received full support. Higher levels of parental autonomy support, lower levels of parental controllingness, and higher levels of structure around the children’s treatment predicted children’s lower psychological reactance and higher adherence to lifestyle modification recommendations. This suggests that children with chronic headache who perceive their mothers as encouraging participation in decisions and choice, as well as providing clear and consistent expectations and guidelines, are less likely to respond to parent behaviors by deliberately ignoring or rebelling against their treatment plans and more likely to be rated by their physician as adherent to making lifestyle changes. An alternative explanation is that mothers become more autonomy supportive and less controlling when their children are more adherent and less reactant. It is likely that the relations uncovered represent a bidirectional process between parents and children. The child-to-parent interpretation seems less likely for parental structure, however, in that parents are likely to provide more structure when children are less rather than more adherent.

The positive effects of structure and negative effects of controllingness suggest the importance of clarity around constructs such as “parental control” and “parental responsibility” in the pediatric literature. Consistent with diabetes literature (e.g. Wiebe et al., 2014), our results suggest that parental roles in children’s treatment are crucial, yet structure and not controllingness may be most facilitative. Future studies, either using SDT or other parenting theories, would be helpful to increase our understanding of the specific behaviors that should be recommended by treatment teams.

Regarding possible antecedents of maternal controllingness, Hypothesis 3 was partially supported. Children’s report of fear of pain predicted maternal controlling behavior; however, neither mother nor child report of pain catastrophizing predicted controllingness. These findings suggest that children’s expression of fearfulness and avoidance of activities believed to worsen pain may evoke more intensified responses from their mothers. While such efforts on behalf of the mother are likely well intentioned and aimed at improving their children’s condition, they may backfire in terms of children’s adjustment. This relation could also be interpreted in the opposite direction—that maternal controllingness might elicit more fear of pain. In either case, the results suggest that attention to children’s attitudes around pain may be helpful for both parents and treatment teams.

Limitations and future research

Results of this study must be interpreted in the context of its limitations. First, the sample size was relatively small and would have benefited from increased statistical power to detect effects. Second, the study utilized questionnaire data. While self-report data may be subjected to reporter bias, this study incorporated multiple informants, including parent, child, and physician, to reduce such bias and balance differing perspectives. Third, the correlational nature of the study precluded causal interpretations and bidirectional relations between variables must be considered. Fourth, the sample lacked gender, ethnic, and racial diversity. Finally, this study included data from a single time point and is therefore unable to explore relations over time.

Despite its limitations, findings from this study suggest the importance of helping parents to actively engage their children in problem-solving and decision-making around treatment, particularly when troubleshooting areas such as lifestyle modifications. Involving children as active agents in the management of their chronic illness may foster their enactment of health-promoting behaviors that may transition into adulthood. Future directions for research include an examination of factors beyond fear of pain and pain catastrophizing that may predict how and why mothers may become controlling around their children’s treatment. In addition, a longitudinal study examining associations among parenting, children’s self-regulation, and treatment adherence using multiple reporters for each of these constructs could help to uncover parent–child dynamics around treatment behaviors over time.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Alessandra Caruso

References

Antonaci

Voiticovschi-Iosob

Di Stefano

, et al. (2014) The evolution of headache from childhood to adulthood: A review of the literature. The Journal of Headache and Pain 15(1): 15.

Barber

Stolz

Olsen

, et al. (2005) Parental support, psychological control, and behavioral control: Assessing relevance across time, culture, and method. Monographs of the Society for Research in Child Development 70: 1–137.

Brehm

(1966) A Theory of Psychological Reactance. New York: Academic Press.

Caes

Vervoort

Eccleston

, et al. (2011) Parental catastrophising about child’s pain and its relationship with activity restriction: The mediating role of parental distress. Pain 152(1): 212–222.

Claar

Walker

(2006) Functional assessment of pediatric pain patients: Psychometric properties of the functional disability inventory. Pain 121(1–2): 77–84.

Deci

Ryan

(1985) Intrinsic Motivation and Self-Determination in Human Behavior. New York: Plenum Press.

Fearon

Hotopf

(2001) Relation between headache in childhood and physical and psychiatric symptoms in adulthood: National birth cohort study. British Medical Journal 322(7295): 1145.

Fogarty

(1997) Reactance theory and patient noncompliance. Social Science & Medicine 45(8): 1277–1288.

Goethals

Oris

Soenens

, et al. (2017) Parenting and treatment adherence in type 1 diabetes throughout adolescence and emerging adulthood. Journal of Pediatric Psychology 42(9): 922–932.

10.

Grolnick

(2003) The Psychology of Parental Control: How Well-Meant Parenting Backfires. Mahwah, NJ: Lawrence Erlbaum.

11.

Grolnick

Ryan

(1989) Parent styles associated with children’s self-regulation and competence in school. Journal of Educational Psychology 81(2): 143–154.

12.

Grolnick

Raftery-Helmer

Flamm

, et al. (2015) Parental provision of academic structure and the transition to middle school. Journal of Research on Adolescence 25(4): 668–684.

13.

Holmbeck

Johnson

Wills

, et al. (2002) Observed and perceived parental overprotection in relation to psychosocial adjustment in preadolescents with a physical disability: The mediational role of behavioural autonomy. Journal of Consulting and Clinical Psychology 70(1): 96–110.

14.

Holmes

Chen

Streisand

, et al. (2005) Predictors of youth diabetes care behaviours and metabolic control: A structural equation modeling approach. Journal of Paediatric Psychology 31(8): 770–784.

15.

Hong

Page

(1989) A Psychological Reactance Scale: Development, factor structure and reliability. Psychological Reports 64(3): 1323–1326.

16.

Kenny

Kashy

Cook

(2006) Dyadic Data Analysis. New York: Guilford Press.

17.

Levesque

Williams

Elliot

, et al. (2006) Validating the theoretical structure of the Treat-ment Self-Regulation Questionnaire (TSRQ) across three different health behaviors. Health Education Research 22(5): 691–702.

18.

Lipton

Manack

Ricci

, et al. (2011) Prevalence and burden of chronic migraine in adolescents: Results of the chronic daily headache in adolescents study (C-dAS). Headache: The Journal of Head and Face Pain 51(5): 693–706.

19.

Marbell

Grolnick

(2013) Correlates of parental control and autonomy support in an interdependent culture: A look at Ghana. Motivation and Emotion 37: 79–92.

20.

Ntoumanis

Thogersen-Ntoumani

(2014) Autonomy support and control in weight management: What important others do and say matters. British Journal of Health Psychology 19: 540–552.

21.

Osman

Barrios

Kopper

, et al. (1997) Factor structure, reliability, and validity of the Pain Catastrophising Scale. Journal of Behavioral Medicine 20(6): 589–605.

22.

Palermo

Chambers

(2005) Parent and family factors in pediatric chronic pain and disability: An integrative approach. Pain 119(1–3): 1–4.

23.

Perquin

Hazebroek-Kampschreur

Hunfeld

, et al. (2000) Pain in children and adolescents: A common experience. Pain 87(1): 51–58.

24.

Radcliff

Weaver

Chen

, et al. (2017) The role of authoritative parenting in adolescent type 1 diabetes management. Journal of Pediatric Psychology 43(2): 185–194.

25.

Rubin

Young-Hyman

Peyrot

(1989) Parent-child responsibility and conflict in diabetes care. Diabetes 38(Suppl. 2): 7A.

26.

Russell

Bray

(2010) Promoting self-determined motivation for exercise in cardiac rehabilitation: The role of autonomy support. Rehabilitation Psychology 55: 74–80.

27.

Ryan

Connell

(1989) Perceived locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and Social Psychology 57(5): 749–761.

28.

Ryan

Deci

(2017) Self-Determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness. New York: Guilford Publications.

29.

Simons

Logan

Chastain

, et al. (2010) Engagement in multidisciplinary interventions for pediatric chronic pain: Parental expectations, barriers, and child outcomes. The Clinical Journal of Pain 26(4): 291–299.

30.

Simons

Sieberg

Carpino

, et al. (2011) The Fear of Pain Questionnaire (FOPQ): Assessment of pain-related fear among children and adolescents with chronic pain. The Journal of Pain 12(6): 677–686.

31.

Skinner

Johnson

Snyder

(2005) Six dimensions of parenting: A motivational model. Parenting: Science and Practice 5(2): 175–235.

32.

Soenens

Vansteenkiste

(2005) Antecedents and outcomes of self-determination in 3 life domains: The role of parents’ and teachers’ autonomy support. Journal of Youth and Adolescence 34(6): 589–604.

33.

Sullivan

Bishop

Pivik

(1995) The pain catastrophising scale: Development and validation. Psychological Assessment 7(4): 524–532.

34.

Teixeira

Carraça

Marques

, et al. (2015) Successful behavior change in obesity interventions in adults: A systematic review of self-regulation mediators. BMC Medicine 13(1): 84.

35.

Van Petegem

Soenens

Vansteenkiste

, et al. (2015) Rebels with a cause? Adolescent defiance from the perspective of reactance theory and self-determination theory. Child Development 86(3): 903–918.

36.

Vervoort

Eccleston

Goubert

, et al. (2010) Children’s catastrophic thinking about their pain predicts pain and disability 6 months later. European Journal of Pain 14(1): 90–96.

37.

Wiebe

Chow

Palmer

, et al. (2014) Developmental processes associated with longitudinal declines in parental responsibility and adherence to type 1 diabetes management across adolescence. Journal of Pediatric Psychology 39(5): 532–541.