Lessons Learned in a Physician Referral to Pediatric Telemental Health Services Program

Abstract

Background

This study explores the physician referral and engagement process of a pediatric telemental health program based in a large urban teaching children’s hospital, and identifies the processes, strengths and challenges from the perspectives of Primary Care Physicians (PCPs) and telepsychiatrist consultants.

Methods

A mixed methods approach was used. This included an online survey completed by 43 PCPs in Ontario rural communities who had referred patients to the telemental health program. Qualitative interviews were conducted with 11 child/adolescent telepsychiatrists who provide consultations via teleconferencing.

Results

The majority of PCPs (61%) reported somewhat to moderate satisfaction with referral experiences. Challenges identified by physicians were related to communication and administration issues including: lack of timely follow-up appointments and continuity of care; lengthy referral forms; and recommendations for mental health services not accessible in their communities. Similarly, psychiatrist consultants expressed frustration with the sparse information they received from referring physicians and most significantly, the absence of appropriate service providers/professionals during the consultation to provide collateral information and ensure uptake of recommendations.

Conclusion

Telemental health programs provide a valuable service to PCPs and their child and youth clients that could be significantly enhanced with a different consultation model. Such models of service delivery require protocols to educate PCPs, improve communication and information sharing and establish clear expectations between PCPs and telepsychiatry consultants.

Telemental healthcare (TMH) is rapidly expanding (Hilty, Ferrer, Parish, Johnston, Callahan, Yellowlees, 2013; Gloff, LeNoue, Novins, Myers, 2015) now representing an integral part of mental healthcare service delivery for young people residing in rural and remote regions in North America (Gloff et al, 2015; Boydell, Hodgins, Pignatiello, Teshima, Edwards, Willis, 2014). Factors driving its expansion include: advances in technology; increasing demand for mental health services in the context of a diminishing pool of child and adolescent psychiatrists (Thomas & Holzer, 2006) and growing evidence suggesting that care delivered to young people through TMH is effective (Myers, 2015; Gloff et al., 2015). However, the evidence base related to program development and organization is not keeping up with the rapid growth in these programs (Hilty, Shoemaker, Myers, Snowdy, Yellowlees, Yager, 2016). In this short communication, we focus on lessons learned from the referral process of rural PCPs to TMH programs.

PCPs in rural areas play a critical role in the mental health care of young people; both directly in terms of providing treatment as they are typically the first point of contact (Clatney, MacDonald, Shah, 2008) and indirectly in making appropriate referrals (Steele, Shapiro, Davidson, Floyd, Johnston, Stretch, et al, 2010). However, few receive formal training in child and adolescent psychiatry (Steele et al., 2010; Myers et al, 2015) and report low levels of confidence in managing young people’s mental health problems (Clatney et al., 2006) but high levels in their abilities to make referrals to psychiatrists (Steele, Zayed, Davidson, Stretch, Nadeau, Fleisher, et al., 2012). TMH programs, which situate child and adolescent psychiatrists in consultative roles serving rural health care settings, are well positioned to enhance the knowledge and mental health care delivery of PCPs (Pignatiello, Boydell, Volpe, Braunberger, Minden, 2011). Despite research indicating high satisfaction rates among PCPs and pediatricians (Myers, Valentine, Melzer, 2007), the extent to which PCPs are actively involved in these consultations appears negligible (Boydell, Volpe, Kertes, Greenberg, 2007) or unknown.

The TeleLink Mental Health Program, a pediatric telepsychiatry service based in a large Canadian inner city (described in detail elsewhere, Pignatiello et al., 2011a, 2011b) currently follows a consultation model where the telepsychiatrist provides a formulation and immediate recommendations to the referring provider who maintains ongoing patient care. Referrals from community agencies are “consultee-centred” consultations whereas referrals from PCPs tend to be “client-centred”. For the consultee-centered referral in this model, the telepsychiatrist consults and makes recommendations to the referring provider about the patient, who may or may not be present. Client-centred consultations always involve the presence of the patient and caregiver in the session but the referring provider; in this case, the PCP who receives the recommendation is typically not present.

Up until 2011, referrals initiated by PCPs in the TeleLink Mental Health Program were re-directed through the local children’s mental health agencies that provided case management for young people and their caregivers. However, due to frequent staff turnovers, referrals were delayed and wait times increased. To facilitate a more immediate and effective referral process and encourage PCPs involvement in managing the mental health issues of children, the referral mandate at TeleLink was modified to include PCPs directly in the referral process. TeleLink follows a consultation model (Gloff et al., 2015) where the telepsychiatrist provides a formulation and immediate recommendations to the referring provider who maintains ongoing patient care. When follow-up consultations occur, there is not a guarantee that the patient will see the same psychiatrist as it depends on the reason and availability. However, attempts are made to keep the consulting psychiatrist consistent, whenever possible. The modified direct referral process allows direct referral from PCPs. Referral information required from PCPs includes basic demographic information as well as the reason for referral. Following the consultation, PCPs treating the child/young person receive a telepsychiatry report and are responsible for facilitating further medical/pharmacological interventions. The research objectives of this study were to explore this modified referral process and identifying its strengths and challenges from the perspectives of PCPs and TeleLink psychiatrists.

Methods

A mixed method approach was used. An online survey assessed PCP’s satisfaction (ranked on a Likert-type scale from 1=very poor, 2=poor, 3=fair, 4=good, 5=very good, 6=excellent) with TeleLink services regarding communication methods, referral process, and recommendations for TeleLink. Overall satisfaction with TeleLink referral services were scored from 1=not at all satisfied to 5=extremely satisfied. Despite numerous attempts to recruit PCPs for the survey, including a letter from the Associate Psychiatrist-in-Chief encouraging participation, 43 rural PCPs from a pool of 174 (24.7% response rate) in the TeleLink referral database completed the survey. We drew on the literature that pre-notification emails/letters/phone calls prior to survey release improves physician participation (Dykema, Stevenson, Day et. al., 2011) and had the lead physician researcher send out pre-notification emails 1–2 weeks prior to release of the online survey and fax. All PCPs who completed the survey declined an offer of a follow up telephone interview to further explore their perspectives on the referral process. Brief qualitative interviews were conducted with a convenience sample of 11 (42.3%) of the 26 TeleLink child and adolescent psychiatrist consultants. Participants were asked to describe improvements they would make to the TeleLink PCP referral program. Study approval was received from the Quality Improvement Board at The Hospital for Sick Children.

Survey responses from PCPs were analyzed using simple frequency statistics. Brief qualitative interviews from telepsychiatrists were audiotaped and transcribed verbatim. Transcripts and written responses from the survey and interviews were analyzed using conventional qualitative content analysis (Hsieh, Shannon, 2005). Text data were interpreted through a classification of codes derived from the research objectives and guideline interview questions asked of telepsychiatrists, and repeated patterns were then identified and summarised.

Results

Online Survey (Likert-type and open-ended questions)

PCPs were from northern Ontario, predominantly central northern and northwestern Ontario, including First Nation communities. Survey respondents represented PCPs practicing in rural Ontario, who had made at least one referral to the TeleLink program 2011–2015. One third identified young people with mental health issues as comprising more than 30 percent of their caseload and close to 60 percent reported some discomfort managing them (Table 1). Almost half of PCPs (49%) made between 5 to 30 referrals per year (Table 2). Most frequent reasons for referral were: assistance with management (91%), clarification of diagnosis (88%) and medication queries (74%).

Table 1.

Primary care physicians’ comfort level managing children/adolescents with mental health issues

Response	Percentage	Count
Not at all comfortable	11.6	5
Slightly comfortable	20.9	9
Somewhat comfortable	25.6	11
Moderately comfortable	27.9	12
Extremely comfortable	27.9	12
TOTAL RESPONSES		43

Table 2.

Primary care physician referral frequency to TeleLink per year

Number of referrals/PCP	Percentage	Count
Less than 5	48.8	21
5 to 30	48.8	21
31 to 60	0.0	0
Greater than 60	2.3	1
TOTAL RESPONSES		43

PCPs varied by the number of referrals made per year and referrals were dependent upon multiple factors. For example, a PCP would make multiple referrals depending on the severity of the mental health condition outside of their scope of expertise, limited experience with child mental health or lack of resources in the surrounding community. The PCPs that had more experience with child mental health and had better shared care models interwoven into their practices with community resources referred to TeleLink for consultation regarding severe and complex cases.

Just over 50% of PCPs rated the recommendations that they received from the consultation as being good to very good and 7% as being excellent (Table 3). However, 41% rated them as being fair to poor. They expressed frustration when telepsychiatrists made recommendations that could not be met locally, reflecting limited telepsychiatrist knowledge regarding resources available in the PCPs communities.

Table 3.

Primary care physicians’ satisfaction with TeleLink recommendations and consultations

	Recommendations		Consultations
Response	Percentage	Count	Percentage	Count
Excellent	7.0	2	27.5	11
Very Good	22.0	9	25.0	10
Good	29.3	12	25.0	10
Fair	29.3	12	17.5	7
Poor	12.4	5	5.0	2
TOTAL		40		40

Regarding satisfaction with TeleLink consultations, over three quarters (77%) of PCPs rated their satisfaction as being good to excellent (Table 3). However, in response to overall satisfaction with their TeleLink experience, the majority (61%) reported that they were only somewhat to moderately satisfied and 27% indicated they were extremely satisfied (Table 4). The varied levels of satisfaction are reflected in respondent comments. Many remarked that the consultation experience improved their confidence, confirming they were providing appropriate care, making suitable treatment decisions and helping them to manage complex cases. They also expressed considerable frustration with onerous referral forms to complete, lack of continuity with the same telepsychiatrist and the length of time before the next consultation, which required them to complete additional paperwork. To improve their TeleLink experience, PCPs suggested a model of care where rapport and communication could be developed between themselves and the telepsychiatrists. PCPs preferred communicating over the phone, where they could ask questions freely rather than obtaining a faxed document of listed medications and recommendations, rendering it difficult for the physicians to contact the consulting psychiatrists to clarify the recommendations. In this study context, physicians could not easily contact the psychiatrist with concerns and questions regarding the recommendations.

Table 4.

Primary care physicians’ satisfaction with overall TeleLink experience

Response	Percentage	Count
Not at all satisfied	4.9%	2
Slightly satisfied	7.3%	3
Somewhat satisfied	34.1%	14
Moderately satisfied	26.8%	11
Extremely satisfied	26.8%	11
TOTAL		41

Brief Interviews

Close to two-thirds (63%) of the consultant psychiatrists interviewed tended to have extensive experience in child and adolescent psychiatry, practicing for over 15 years. They participated in Telelink consultations from one to four times weekly. Telepsychiatrists were generally positive about their program involvement, however, they expressed frustration with issues around communication, both in terms of receiving patient information and in providing recommendations. They indicated that referral documents from referring PCPs were often difficult to read, disorganized and missing information. PCP referrals that included input from community agencies, however, were reported to be, on the most part, comprehensive and information rich, enabling telepsychiatrists to adequately prepare for the consultation. In some cases, community mental health agencies either directly referred patients or contributed to the PCP referral process and to the consultation.

The presence of a PCP or an informed (often allied health) service provider during the consultation was critical for obtaining patient collateral information and ensuring that prescribed treatment was appropriate and implementable. PCP referrals not involving a local mental health agency were identified as most challenging because, as one participant noted, physicians “rarely show up”. PCPs were viewed as “having the mindset of referring to a specialist”, which meant their involvement halted from time of referral to receipt of the consultant’s report. Without the presence of a service provider (e.g. therapist, case manager, nurse) who was familiar with the young person, relevant context and history was often missing and telepsychiatrists were unable to make immediate and appropriate assessments or recommendations. It appeared that psychiatrists viewed professionals as also providing structure and safety in the consultation process, helping to build rapport and comfort for the patient.

Discussion

PCPs and telepsychiatrists shared what might be considered mirrored versions of frustration in communication and information sharing. PCPs viewed referral documents as lengthy and cumbersome to complete, whereas telepsychiatrists found PCP referrals to be brief and difficult to follow. The lack of follow-up and continuity by the telepsychiatrist compromised PCP’s confidence in providing appropriate treatment. For telepsychiatrists, the absence of the PCP or the community agency practitioner during consultation sessions meant potentially missing relevant patient information thereby compromising assessment and treatment recommendations.

The lack of PCP’s involvement in the consultation is a significant issue not extensively reported on, suggesting the need to refine existing adult models of TMH service delivery to better serve children and youth (Gloff et al., 2015; Fortney, Pyne, Turner, Farris, Normoyle, Avery, 2015). Currently within telepsychiatry, direct care and consultation models are most commonly used with young people and families (Gloff et al., 2015). Psychiatrists have suggested that consultation models may encourage PCP participation because telepsychiatrists are typically not responsible for continuing patient care (Volpe, Boydell, Pignatiello, 2013) as compared to collaborative care (a more intensive model entailing on-going care shared among the PCP and the consultant) (Hilty et al., 2013). However, PCP non-participation in the consultation sessions is problematic, particularly in pediatric populations where the mental health needs of children and young people are often complex, requiring family involvement and a comprehensive array of services and providers (e.g. behavioural, education supports) (Hilty et al., 2016). Innovative hybrid consultation models that preserve the PCP-patient relationship and allow the telepsychiatrist to remain in a consulting role by, for example, involving a local service provider(s) who is knowledgeable about the child and can bridge communication gaps, need to be considered (Yellowlees, Chan, Burke, Parish, 2015). This type of model would address PCP’s preference for a shared model of care as the traditional models of shared care between PCPs-telepsychiatrists are not sustainable because they are held during the day when PCPs are unable to attend. Thus, it is unreasonable to expect that PCPs could attend such sessions. Further, placing telepsychiatrists in consultative roles, in contrast to the traditional siloed approach whereby a mental health specialist assumes responsibility for patients’ behavioural care, is suggested to have greater population level impact in the mental health sector (Fortney et al., 2015). In addition, when local case managers are involved, they bring with them a detailed knowledge base regarding local community resources.

The challenges regarding PCP referrals to pediatric telemental health as reported by PCPs and telepsychiatrists might be mitigated by: instilling new requirements such as multiple consultative meetings with the same consultant; clarifying expectations for both PCPs and telepsychiatrists; encouraging dialogue between telepsychiatry specialists and rural PCPs that go beyond a faxed recommendation; developing resource guides for local mental health services, and implementing education/training regarding the referral and consultation processes. Relevant models can be drawn upon to provide guidance regarding these improvement strategies, for example with respect to training (Gadomsky, Wissow, Plainkas et al., 2014) and resource guides (https://www.seattlechildrens.org/healthcare-professionals/access-services/partnership-access-line/). Project ECHO (Project Extension for Community Healthcare Outcomes), a free educational program funded by the Ontario Ministry of Health and Long-Term Care for PCPs, represents an innovative, technology-enabled collaborative learning program. The goal is to share knowledge about child and youth mental health amongst PCPs and with tertiary care specialists, offering CME/CPD for PCPs. Evaluation of the model indicated that high-participant retention is achievable, and provided preliminary evidence for increased knowledge and self-efficacy (Sockalingham et al., 2018). These findings suggest that this intervention may improve mental health management in primary care.

Limitations

The response rate is somewhat lower than usual in such study approaches; however, low response rates of physicians in particular have been previously reported (Abdulaziz et al., 2015) - with the field of psychiatry as particularly low at 27% (Cunningham et al., 2015). The possibility of non-representativeness is also of concern due to the convenience sampling. It is also possible that the unique features of TeleLink, such as its large pool of psychiatrists, might limit generalizability ot other jurisdictions using tele-mental health. Private practice office settings with its “various gatekeepers” poses an additional barrier to PCP participation.

Conclusion

As this study suggests, an effective telemedicine model requires accommodating the needs of both rural medical sites and the telepsychiatry sites that serve them (Hilty, Yellowlees, Cobb, Bourgeois, Neufeld, Nesbitt, 2006). Integral to addressing these needs are models and protocols to ensure on-going communication and information sharing that begins with the referral process.