Engaging patients and parents to improve mental health intervention for youth with rheumatological disease

Eligible patients were ages 14–24 years reporting a diagnosis of JIA, JDM or cSLE, and receipt of specific treatment [including: non-steroidal anti-inflammatory drugs (NSAIDs); glucocorticoids (oral, intravenous, or intra-articular); conventional disease modifying anti-rheumatic drugs (DMARDs); and biologic DMARDs]. Parents of patients ages 8–24 years were also eligible to participate; to facilitate anonymity, they were not required to participate in pairs with their children. Age inclusion criteria were generally aligned with the definition of youth according to the US Department of Health and Human Services Healthy People 2020 Objectives (which includes adolescents ages 10–17 and young adults ages 18–25) [21]. However, we utilized the following modifications: lower age of 14 years for patients to enable consent for their own participation, and facilitate anonymous response; eligibility of parents of patients ages 8 and older, based on parent advisor feedback regarding the importance of characterizing mental health for pre-adolescent patients.

The survey was conducted anonymously to maximally elicit personal information about mental health, while reducing underreporting or misreporting which can occur due to social desirability bias around sensitive topics [22]. To recruit participants living in the United States and Canada, we leveraged the patient organizations in the PARTNERS network (Arthritis Foundation, Lupus Foundation of America, and Cure JM Foundation). Participants accessed the REDCap survey via weblink between August 2017 to August 2018, sent via email listservs, and available on websites and social media sites for the above 3 patient organizations; the weblink was also available via local advertisements at nine participating CARRA clinics. Data was managed using the secure, REDCap web application designed to support data capture for research [23].

The anonymous online survey, available in English and Spanish (using forward and back-translation methods), comprised quantitative and qualitative items to examine experiences with mental health care for patients with JIA, JDM or cSLE. The survey captured the following patient disease characteristics: rheumatological diagnosis, medications, disease duration and activity (active vs inactive), visual analog score (VAS) for disease-related health (0 = very well and 10 = very poor). Active disease was defined across disease groups by an affirmative response to both of the following the questions: “Is your disease currently active (not under good control, with or without medication)?”, and “Has your rheumatologist told you that your disease is currently active/ not active?”. Patient demographics included age, gender, race, ethnicity, household income and education level, insurance, and location [United States census region used by the Centers for Disease Control and Prevention [24], or Canada (regions not included due to small Canadian sample]. Participants also provided mental health history including: clinician-diagnosed mental health disorders (by a rheumatologist, psychiatrist, pediatrician/family care doctor, other doctor, or psychologist), past or current self-diagnosed mental health symptoms, and past or current mental health treatment received (e.g. therapy or counseling, psychotropic medications, or inpatient psychiatric care). Additionally, in response to a hypothetical vignette simulating a situation of emotional distress for the patients (Additional File 1), participants used a Likert scale to rate their: i) level of comfort with potential mental health providers (0 = very uncomfortable, 4 = very comfortable), and ii) level of concern about potential barriers (16 for patients, 17 for parents) to seeking help from a mental health professional (0 = not at all, 4 = extremely), based on known correlates of unmet need for mental health services in children and adolescents [20, 25]. We also collected information on the broad scope of types of mental health resources that participants had used (including general counseling services, peer support groups, camps, online resources, and printed handouts/resources). We asked participants who had used these resources if they found them to be helpful, not helpful, or if they were unsure if the resource was helpful. These participants were then asked to provide a qualitative text entry explaining their response.

Participant characteristics were summarized using frequencies and proportions for categorical variables, and means and standard deviations or medians and interquartile ranges for continuous variables, as appropriate. Ineligible responders and incomplete responses (from unsubmitted surveys) were excluded. We measured the prevalence of self-reported mental health problems for patients with rheumatological disease, stratifying responses by mutually exclusive groups for clinician-diagnosed disorders versus self-diagnosed symptoms, and tabulated reported mental health treatments received by these groups. For participants indicating use and helpfulness of mental resources, qualitative text responses were analyzed using qualitative description to summarize positive or negative experiences with mental health resources, and illustrative text responses were selected. Qualitative description is a low-inference approach that presents the facts from exactly the informants’ point of view. It is the qualitative method of choice when only a description of a phenomenon is desired, and is particularly useful in health care research to focus on the patient experience, for the purpose of needs assessment and intervention development [26‐28].

We calculated mean Likert scores for level of comfort with potential mental health providers, and frequency of barriers to seeking mental health treatment. Using multivariate linear regression to control for potential confounders (patient age, gender, disease duration, and patient/parent VAS), we used two separate models to compare patient and parent responses for ratings for comfort level with mental health providers, and frequency of barriers to mental health treatment. We also conducted two stratified regression analyses to compare responses by:1) patient age group (14–17 versus 18–24 years) to determine if the results differed between adolescents and young adults; and 2) presence or absence of reported mental health problem (either clinician- or self-diagnosed), to determine if actual experience with a mental problem impacted the results. Analysis was conducted using Stata^(R) Data Analysis and Statistical Software version 15.0. Alpha ≤0.05 denoted a statistically significant result.

Four hundred forty-seven respondents included 123 patients and 324 parents. Nine respondents (2%) completed the Spanish version of the survey. The most prevalent diagnosis was JIA for both patient (41%) and parent (49%) respondents, followed by JDM (35 and 40%), and cSLE (24 and 11%). The most common treatments received for rheumatological disease were conventional disease-modifying anti-rheumatic drugs (DMARDs) by > 90% of patients. For JIA participants, ever use of methotrexate (80%) and oral glucocorticoids (54%) approximated that for large CARRA JIA Registry cohort (73 and 47%) [29]. Additional disease and demographic characteristics are shown in Table 1.

A history of any clinician-diagnosed or self-diagnosed mental health problem among patients was reported by 92 patients (75%) and 201 parents (62%). Patients most commonly reported problems with anxiety (66%), depression (53%), and adjustment (37%) (Fig. 1), which paralleled those reported by parents (Table 2). History of suicidal thoughts (23%), attention deficit disorders (20%), and eating disorders (18%) were also frequently reported by patients. Adjusted models controlling for patient age, gender, disease duration, and patient/parent VAS showed no significant difference between patients and parents for reported mental health problems.

Of the 92 patients reporting a history of at least one mental health problem, 32 (35%) indicated a clinician-diagnosed disorder only, 37 (40%) indicated self-diagnosed symptoms only, and 17 (18%) indicated both a clinician-diagnosed disorder and self-diagnosed symptoms (6 did not specify). For every type of mental health problem, more than a third of patients reporting the problem had self-diagnosed symptoms rather than a clinician diagnosis, with the exception of post-traumatic stress disorder and attention deficit disorder (Fig. 1). For patients with mental health problems, 46% of patients indicated that their rheumatologist was aware of the problem, compared to 70% of parents (p < 0.0001, Z = -3.87, test of proportions).

Of patients with a reported mental health problem (including clinician and self-diagnosed), receipt of mental health treatment was reported by 44 of 92 (48%) patients and 128 of 201 (65%) parents (Table 2). Psychotherapy or counseling was the most common reported treatment modality, reported by 41% of patients and 57% of parents. About a third of patients reported receiving psychotropic medications, and < 10% of reported receiving inpatient psychiatric care. Of those reporting a clinician-diagnosed disorder, 39 of 49 (80%) patients reported receiving mental health treatment; parents indicated that 90 of 115 (78%) patients had received treatment. Of the 37 patients reporting self-diagnosed symptoms only, 4 (11%) reported receiving mental health treatment; parent report indicated that 27 of 62 (44%) received treatment.

In response to the hypothetical vignette simulating a situation of emotional distress for youth with rheumatological disease, patients were significantly less comfortable seeking help from all potential mental health providers than parents, adjusting for patient age, gender, disease duration, and patient/parent VAS (Fig. 2). While both groups felt most comfortable with rheumatologists (mean Likert score for patients 2.85 vs parents 3.54, p < 0.0001) and primary care providers (PCPs) (patients 2.53 vs parents 3.36, p < 0.0001), they were least comfortable with social workers at the primary care (patients 1.54 vs parents 2.26, p < 0.0001) and rheumatologist office (patients 1.85 vs parents 2.66, p < 0.0001), and school counselors/therapists (patients 1.69 vs parents 2.47, p < 0.0001). There were no significant differences in level of comfort with mental health providers when respondents were stratified by patient age (14–17 vs 18–24). When stratified by reported mental health problem in the patients, those reporting a problem were significantly more comfortable interacting with counselors/therapists in the community than those without (2.72 vs 2.51, p = 0.038), adjusting for other potential confounders.

In response to items for the hypothetical vignette regarding barriers to accessing mental health services (16 for patients, 17 for parents), patients and parents both rated the following in the top two barriers: concerns that mental health professionals would not understand the rheumatological disease; and that insurance coverage would be inadequate (Fig. 3). Another barrier rated among the top five by both patients and parents included patients not wanting to go/feeling uncomfortable with accessing mental health services. However, there were also statistically significant differences between patient and parent Likert ratings among the top five barriers, after adjusting for potential confounders. Specifically, patients reported more concern about not wanting to take medication for mental health problems (patient 1.63 vs parent 1.27, p = 0.01), whereas parents were more concerned that patients would not want others to find out about mental health treatment (patient 1.14 vs parents 1.38, p = 0.04). There were no significant differences in ratings when stratified by age group or presence of a mental health problem.

69% of patients (n = 85) and 63% of parents (n = 207) reported use of a patient mental health resource (Table 3), of which online informational resources were most common by patient report, and patient counseling services by parent report. Fifty-two (61%) patients and 148 (72%) parents thought the mental health resources were helpful.

Illustrative quotes for qualitative text responses of experiences with patient mental health resources are shown in Table 4. Respondents described positive experiences with counselors who helped patients process emotions and pain, but several indicated negative experiences with counselors who did not understand the impact of the rheumatological disease on mental health. Other comments shed light on the high cost and scheduling/logistic challenges associated with counseling services. Respondents indicated that disease-specific online resources were informative, and those with social media communities were helpful; however, patients sometimes felt uncomfortable sharing very personal information through online social platforms, and found it challenging to be reminded of their struggles with rheumatological disease. Peer support groups were also identified as opportunities to hear from other patients experiencing similar problems, although some commented that it was sometimes difficult to overcome unease about sharing personal details about their lives. Summer camps were seen as beneficial because patients could be around people like them who understood what it was like to have a rheumatological disease. However, some patients reported not being able to engage in summer camps if their disease was too severe or if the camp did not have enough focus on their specific rheumatological disease (Table 4). Print books were also helpful, but noted to be scarce by some respondents.

The strengths of this study include the engagement of patients and parents in the survey design, which was critical to constructing survey items with language conducive to inquiring about the sensitive topic of mental health. Engagement of a patient advocacy group such as PARTNERS was also essential to optimizing the content, design, advertisement and distribution of the survey to reach a broad, national audience of patients and parents through social media and other approaches. Administration of a survey in both English and Spanish also aimed to increase representativeness.

This study also has limitations. First, since the survey responses were self-reported and anonymous, there is possibility of misclassification. We were not able to definitively confirm clinician diagnoses of rheumatological disease; however, we think that our sample represents the intended patient population, as reported medications were similar to expected, with treatment use in JIA participants approximating that in the large CARRA JIA Registry cohort [29]. We were also not able to confirm reported mental health diagnoses, nor ask questions surrounding severity of reported mental health problems, due to lack of a mechanism to identify and appropriately refer participants posing potential harm to themselves or others; however, the prevalence of reported clinician-diagnosed mental health disorders is similar to other studies of pediatric rheumatology patients. Second, generalizability of our results to the larger pediatric rheumatology patient population may be limited by several factors, including: the skewed demographic of the sample (predominately white, educated, and high-income); limitation of the study to JIA, JDM, and cSLE participants in order to leverage the PARTNERS member organizations which are focused on these diseases; and potential enrichment of patients with more disease burden (e.g. JDM and cSLE patients), comorbidities, and mental health issues than the general pediatric rheumatology population. Furthermore, while our varied recruitment approach including clinic advertisements, email listservs, and social media allowed us to reach a wider audience, it made it impossible to calculate response rate and gather information about those who did not complete the survey, also limiting assessment of generalizability of our results. Third, although we adjusted for patient characteristics in our comparisons between patient and parent responses, we could not compare pairs of patient respondents and parents from the same family due to the anonymous survey approach. Fourth, while our stratification of patients into adolescents and young adults is generally relevant for availability of pediatric vs adult medical and mental health services, we may not have captured differences in mental health experiences for the subgroup transitioning to adult care in their early 20s. Last, while we included the perspectives from parents of patients as young as 8 years old, the mental and behavioral health needs of these younger children were not directly captured in this survey. Despite these limitations, our study provides important and much needed insight into the mental health needs of youth with rheumatological disease, from their own perspective and from their parents.