Interview substudy.

The interview substudy was conducted from February to June 2017 and comprised 30–45-minutes, one-to-one interviews among 14 GPs in their respective general practices. Interviews were planned 1–2 weeks in advance in consultation with a doctor’s assistant. The interviews were semi-structured (i.e., open-ended questions followed by probes and transitions) [18, 19], were audio recorded using a voice recorder and were transcribed verbatim by LV. At the start of each interview, GPs were informed about the study’s background and aim. Each GP was also asked to sign an informed consent form. For the interview substudy, 176 GPs were contacted by e-mail and telephone, of whom 14 were included.

Online survey substudy.

Quantitative data were derived from an online survey conducted in May 2017. The completion time of this questionnaire ranged from 10 to 20 minutes. The survey started with an introductory page including the study’s background and aim, as well as a reference to the informed consent form that was attached to e-mails sent to each GP. The online survey consisted of 27 questions based on a question framework. For the online survey substudy, 130 GPs were contacted, of whom 20 responded and 15 filled out the whole questionnaire.

Participants.

GPs who indicated interest in participating in one or both substudies were selected based on purposive sampling [21]. Mandatory sampling criteria included the district in which GPs’ general practices were established, with a maximum of one GP per district for the interview substudy. Since there are 14 districts in the Holland Rijnland region, the sample for the interviews initially has been restricted to n = 14. If two or more GPs from a particular district indicated their wish to participate in the qualitative substudy, the GP who responded first was included. Another sampling criterion was inclusion of experienced GPs who worked for themselves. Therefore, GPs were selected who would have an overview of youth mental health regulations, in order to gain a full picture of how youth mental health care provision is organized within the study region. In both the quantitative and qualitative substudy, the aim was to include GPs who see psychosocial problems among children and youths a minimum of three times per two weeks. However, during the recruitment, GPs often mentioned a lower frequency of once to thrice a month. We therefore decided to change this criterion and to also include GPs who reported to see these problems in a lower frequency. For the quantitative substudy, only GPs who filled out the whole questionnaire were included. Participation was voluntary. GPs who participated in the interviews received two small presents (a compass and a pen with a LUMC logo) in return for their participation. There was no compensation for participating in the online survey.

Clinical vignettes.

In both substudies, three fictional clinical vignettes—A “(suspected) psychiatry,” B “multidimensional problems” and C “safety” (S1 Box)—were used to explore GPs’ everyday CDM. Validation of vignettes was achieved through formulation of the vignettes by a child psychologist (AB) and a general practitioner (PB), based on personal clinical experience regarding referrals of children and youth to a local youth and family center. Also, the vignettes were verbally deemed recognizable with respect to clinical practice during several presentations to GPs and youth mental health workers outside the frame of this study [16, 22].

Question framework.

For both substudies, the research group developed a question framework to explore CDM and to guide GPs through the clinical vignettes (see S2 Box). Unlike the interview study, the online survey included only multiple-choice questions. The question framework included questions on general demographics of the GP (i.e., name, gender, age, years working as a GP, years of working in current general practice), name of district he/she works in, type and frequency of encounters with psychosocial problems in children and youth, frequency of discussing psychosocial problems one on one with a mental health nurse practitioner (MHNP) and self-assessed knowledge of youth mental health care regulations. Questions regarding the clinical vignettes were about the first impression of the problem, thoughts on further diagnostics and management for solving the problem, plan for referral, recognizability of the vignettes with regard to practice, information relevant to achieving a plan for referral (e.g., patients’, parents’ and siblings’ preferences, norms and values regarding treatment) and negative or positive collaboration experiences with other youth care providers (e.g., content, quality and speed of communication). The online survey concluded with three statements on financial cuts to youth mental health care services.

Participants.

In total, 29 GPs (15 in the online survey, 14 in the interviews) were recruited, of whom 12 were female (5 in the online survey, 7 in the interviews). Unpurposely, all GPs who participated in the qualitative substudy did not fill out the questionnaire of the quantitative substudy. Therefore, there was no overlap between substudies with regards to participants. GPs had worked an average of 18.6 years (range 5–38 years) and 18.7 years (range 3–33 years) in the field in the online survey and interview study, respectively. Moreover, they worked 15.7 years (range 2–36 years) and 13.0 years (3–32 years) in their current general practice in the online survey and interview study, respectively. Eighteen GPs worked with a mental health nurse practitioner (MHNP) who provides care for children and youth. Nearly all GPs encountered psychosocial problems at least monthly. According to interviewed GPs, vignettes A and C were most recognizable with respect to clinical practice (Table 1). The online survey did not question recognizability of clinical vignettes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of GPs–online survey (n = 15) and interviews (n = 14), N = 29.

https://doi.org/10.1371/journal.pone.0278314.t001

Preferred approach.

To obtain an overview of the child’s functioning in different life domains which may have been impeded by the problem situation (e.g., disruptive behavior at home, school, leisure), all but one surveyed GPs would ask for the child’s or the adolescent’s opinion, as well as the parents’ views of the problem situation. Less than half of the surveyed GPs would ask for another care provider’s opinion and a youth health care provider’s opinion.

Interviewed GPs mentioned contacting abovementioned persons to obtain more information but added that they would use this information to assess whether they attributed the presented problem(s) to somatic or psychosocial origins. A few interviewed GPs expressed their preference to arrange further physical diagnostic tests and, if needed, treatment by the GP themselves or by a somatic care provider, e.g., pediatric neurologist in vignette A. Presumed by the interviewer that physical cause(s) were unlikely or excluded, eventually all interviewed GPs considered psychosocial origins of the problem(s). After considering somatic and psychosocial origins of the problem, all interviewed GPs weighed whether the problem situation would solve itself without intervention or whether an intervention was necessary.

Quote 1B: “Mother has consulted me on her own, but I also think it would be interesting to hear what Sanne has to say.”

Quote 12A: “I would like to ask the child neurologist what causes his (Dave) symptoms, is it perhaps epilepsy?”

Competence assessment based on interest in the field.

The online survey contained no questions on perceived competence of the GP.

However, the interviews showed GPs’ self-assessed competence to manage or refer the presented problem(s) was related to interest in the field on one hand and knowledge about the field on the other hand. A few interviewed GPs explicitly expressed feeling particularly responsible to take up youth-care-related tasks, having experience recognizing or treating psychosocial problems. Other interviewed GPs did not mention anything about felt responsibility. A majority of GPs mentioned that identifying alarming problem situations and coordinating youth-care-related activities were perceived as their tasks, irrespective of their personal interest. Half of the interviewed GPs who explicitly expressed their interest in the field also participated in the district council to discuss recent political developments. A few GPs had previously participated in youth-care-related research. A small minority of GPs had done additional training in this area.

Quote 2A: “My colleague’s expertise is cutting and chopping, mine’s communication.”

Competence assessment based on knowledge about the field.

The online survey showed some GPs have knowledge of recent developments in regional and national youth care, initiatives for interdisciplinary collaboration, regional referral options, legal regulations and money flows.

Interviewed GPs who claimed to have knowledge of changes in regional youth care provision stated they were kept well informed by their local authority about these changes. Following from the interviews, knowledge about the field also determined whether GPs first assessed (co)morbid mental health problems instead of choosing immediate referral to a child and youth psychiatrist. One interviewed GP called this immediate referral “my common route.” In comparison, other GPs referred based on the specific request for help by the child, its parents or its school. If the GP was constrained by time or not well trained, a MHNP would be consulted to explore this request for help.

Quote 9A: “I know that in this district youth care providers communicate with each other using an information loop.”

Sense of self-limitedness.

Throughout the vignettes, more than half of all surveyed GPs would advise parents to seek help, i.e., they thought further intervention was necessary.

The interviews revealed that this assessment of whether help or further diagnosis was needed was based on GPs’ gut feeling. Interviewed GPs answered that this gut feeling was made up of a combination of factors, including: a sense of something’s wrong, a problem situation not fitting GPs’ expectations of a “normal” biopsychosocial development and expected deterioration if the GP refrained from any intervention, i.e., “watchful waiting.”

Quote 7C: “Well, I think this behavior is normal for her age.”

Assessment of psychiatric (co)morbidity.

The online survey contained no specific questions on the assessment of psychiatric (co)morbidity.

The interviews, however, showed that GPs assessed whether or not to consult a child and youth psychiatrist based on anamnesis, specific request for help, behavioral observations and previous clinical findings such as family history. However, due to long waiting lists, a few interviewed GPs chose to consult the youth and family team or a self-employed care provider instead, mainly guided by existing collaboration agreements with these youth care providers.

Quote 4C: “It is somewhat unclear here, but Melany could be a troubled teenager with ADD or ADHD. If I would want further psychiatric diagnostic evaluation, I would refer to [name of a local mental health institution], instead of the youth and family center.”

Assessment of complexity.

The survey showed that in all three vignettes, GPs frequently thought about contacting a local youth and family team or asking their MHNP for further diagnosis and management. For vignette A, some surveyed GPs were also thinking about contacting a specialized mental health professional.

In the interviews, a majority of GPs reported that prior to selecting an intervention for further diagnosis or management, they assessed complexity of the vignettes. GPs spoke of a “complex” problem if the situation combined both multidimensionality and severity. A problem situation was called “multidimensional” if multiple individuals in a family were involved and multiple life domains were impeded. Furthermore, the degree to which a dimension was impeded was described as “severity.” In noncomplex problem situations, most interviewed GPs would ask their mental health nurse practitioner (MHNP) or a self-employed child psychologist to provide short-term treatment or would consult other youth care providers. In more complex cases, GPs would contact a local youth and family team because these teams were considered to offer rapid social support to multiple individuals in a family at the same time. Furthermore, specialized mental health care services would have been chosen in more complex cases. If a consulted care provider would assess the problem(s) to be more or less complex than assessed by the GP, all interviewed GPs were prepared to refer to another youth care provider.

Quote 2B: “I don’t think I have to refer every teenager who lives in a stressful home situation.”

Quote 3B: “The self-employed child psychologist I usually refer to also pays attention to the child’s social system.”

Existing collaboration agreements.

The online survey contained no specific questions on the importance of existing collaboration agreements in GPs’ CDM.

However, the interviews showed that, in cases where GPs thought psychiatric (co)morbidity seemed less likely, they inventoried which youth care provider they already had collaboration agreements with on, for example, consultations, referrals and interdisciplinary meetings. The most frequently mentioned youth care providers to be contacted were MHNPs, self-employed care providers and local youth and family teams.

Quote 5A: “Every six weeks, I speak to our MHNP about patients she has seen.”

Quote 14A: “The youth and family team of our district advises us on waiting lists.”

Previous collaboration experiences.

Surveyed GPs were asked to react to several statements regarding their previous collaboration experiences with youth care providers. Their answers showed a few GPs opining that youth and family teams are suitable for managing social-system-related problems and most GPs opining that budget cuts in youth care services have led to deteriorated quality of care.

The interviews provided more detail with regards to these answers. Almost all interviewed GPs thought recent budget cuts in youth care provision increased the possibility of having negative collaboration experiences rather than positive ones. Negative experiences included poor quality of written or verbal feedback after referral, previous referrals having a negative impact on patient-doctor relationships due to unsafe exchange of private information, patient-perceived stigma after referral, personal unfamiliarity with care provider(s) and low perceived expertise. Negative collaboration experiences with a care provider—local youth and family teams in particular—resulted in referring to another care provider, even if the former provider might have provided more suitable help. Positive experiences made it more likely for GPs to refer to a self-employed care provider they preferred.

Quote 13A: “After a lot of conversations (with mother of the child, among others), the person from the youth and family team concluded it was a difficult situation. That’s a very thorough conclusion after six months. *laughs cynically*”