Abstract
OBJECTIVE:
Since 2005, after a pilot program, the Massachusetts Child Psychiatry Access Project (MCPAP) has provided point-of-care psychiatry expertise and referral assistance by telephone to primary care providers. We examined its adoption and use and the practice characteristics associated with different adoption timelines and use patterns.
METHODS:
We merged data on calls to MCPAP in 2005 to 2011 with practice data (enrollment year, panel size, regional team assignment). We categorized practices’ days from enrollment to first call (adoption) (0–100, 101–365, >365 days) and quartile of call frequency (use) (annual highest, middle, and lowest quartiles of number of calls per 1000 empanelled patients). We determined associations between adoption and use and practice characteristics using multivariate models.
RESULTS:
Among 285 practices, adoption and use varied: 55% called 0 to 100 days from enrollment and 16% called >365 days from enrollment. Practices in the highest quartile of use made a mean 15.5 calls/year per 1000 patients, whereas the lowest quartile made 0.4 calls/year per 1000 patients. Adoption within 100 days was associated with enrollment during or after 2007 (odds ratio [OR] 4.09, 95% confidence interval [CI] 2.23–7.49) and assignment to the team at the pilot site (OR 4.42, 95% CI 2.16–9.04 for central Massachusetts). Highest-quartile use was associated with team assignment (OR 3.58, 95% CI 1.86–6.87 for central Massachusetts) and panel size (OR 0.10, 95% CI 0.03–0.31 for ≥10 000 vs <2000 patients).
CONCLUSIONS:
Adoption and use of MCPAP varied widely. Timing of enrollment, assignment to the team from the program’s pilot site, and panel size were associated with patterns of adoption and use. Findings may help other programs design effective implementation strategies.
Keywords: psychiatry/psychology, community pediatrics
What’s Known on This Subject:
A program to support pediatric primary care providers in mental health care using point-of-care, telephone-based advice from specialists has been available since 2005 in Massachusetts. Other US states are implementing similar models. Little is known about how providers use this service.
What This Study Adds:
There is wide variability in adoption and use of this program. Patterns are associated with panel size, enrollment timing, and assignment to the program team at the pilot site. Findings will help new programs establish expectations and design implementation interventions.
Caring for children and youth with mental health conditions in primary care is challenging for many reasons, among them a shortage of child mental health specialists,1,2 growing complexity3 and perceived risk4 of treatment regimens, and a fragmented system of mental health services delivery.5 However, the proportion of primary care visits that involve mental health is increasing,6,7 and many patients prefer to receive basic mental health care from their primary care provider (PCP).8 Although most PCPs feel confident as managers of attention deficit hyperactivity disorder (ADHD), they feel unprepared and unsupported when a comorbidity is present with ADHD and with other mental health conditions.9
Innovative programs or service delivery models that aim to increase the capacity of PCPs to deliver mental health care help address this issue. The Massachusetts Child Psychiatry Access Project (MCPAP), established in 2005 after a successful pilot, connects PCPs with mental health specialists to provide advice and coordinate referrals to community-based therapists.10 Practices enroll in MCPAP by meeting with MCPAP staff and signing an agreement acknowledging scope of services; then, PCPs call as needed. Several states have implemented or are considering similar programs, and a national network of programs has recently been formed.11–13
Little is known about how programs such as MCPAP are adopted and used. Sarvet et al10 reported a 6-fold increase in monthly MCPAP encounters between 2005 and 2008. PCPs remained as primary managers of mental health problems for most patients discussed with MCPAP. PCPs indicated that their ability to meet the needs of patients with mental health problems improved after enrolling in MCPAP. Evaluation of a similar program in Washington reported congruent findings.11
These previous studies did not examine patterns of use or investigate reasons for any variability in use. A greater understanding of how practices and PCPs vary in their adoption and use of programs such as MCPAP may help with future program design and implementation strategies. Primary care practices are heterogeneous in the way mental health care is delivered,14,15 and tailoring programs to practice patterns or characteristics may accelerate use, increase perceived value, and more quickly expand capacity of primary care to manage mental health problems.
In this study, we aimed to understand the speed and intensity with which MCPAP was adopted and used among primary care practices. First, we examined to what degree time to adoption and ongoing use of MCPAP varied among practices. Second, we examined call-level and practice-level characteristics to explain variation in adoption and use. Specifically, we asked the following research questions: What was the mean, median, range, and distribution of adoption and frequency of use among practices? Were patterns of adoption and use associated with the reason for the call, the diagnosis discussed, or the outcome of the call? Were patterns of adoption and use associated with practice type, patient volume, number of clinicians, or distance from the practice to the regional MCPAP office? Did adoption and use vary with the regional team to which the practice was assigned? These questions will characterize and inform the complicated process of introducing new models of mental health care in primary care.
Methods
Design
We performed a cross-sectional analysis of 29 202 calls to MCPAP from May 2005 to July 2011 merged with practice-level data from 285 primary care practices.
Description of MCPAP
In 2005, MCPAP was established by the Executive Office of Health Services in Massachusetts to enhance patient access to child psychiatry and community mental health resources. MCPAP began as a pilot with 22 practices in central Massachusetts in 200316 and expanded to 6 teams at regional sites across the state. Practices are assigned to regional teams based on geographic proximity. Each team consists of a psychiatrist, a mental health social worker, and a care coordinator. Teams establish formal relationships with practices by contacting the practice to set up an enrollment visit in which the team gives a short presentation to PCPs and answers questions. When a PCP from an enrolled practice sees a patient in the office with a mental health problem and is uncertain about an aspect of care, he or she can call MCPAP. MCPAP can offer advice regarding a patient’s diagnosis, comorbidities, or psychopharmacology and can assist with a referral to a mental health therapist or psychiatrist. If needed, the MCPAP psychiatrist offers a single-visit, in-person evaluation and renders a detailed, timely report to the PCP. MCPAP staff maintain lists of mental health specialists, their availability, and insurance plans they accept and can contact the family directly for referral assistance. Currently, the program is funded by both the Commonwealth of Massachusetts and commercial insurers and receives some additional reimbursement from billable procedures (eg, in-person patient evaluations). MCPAP aimed to enroll every primary care practice serving children, and currently >95% of children in Massachusetts have PCPs in practices enrolled in MCPAP.
Database Development
We merged 2 databases. The first contained calls from PCPs to MCPAP from May 18, 2005, to July 31, 2011. MCPAP documents each phone call with PCPs using a structured electronic form that includes the date, the practice from which the call originated, the reason for calling, the mental health condition discussed, and the outcome of the call. The second database included data collected by MCPAP as part of the enrollment process and yearly communication with practices. At enrollment, MCPAP collected from each practice the number of full-time provider equivalents (FTEs), estimated number of patients aged 0 to 21 years, and practice type (eg, family medicine or pediatrics). The information was typically given by the practice’s director or practice manager. Practices’ enrollment dates and addresses were added from a separate database. For this study, we included practices that had been enrolled in MCPAP for ≥1.5 years. We merged call-level and practice-level databases, linked by practice identifier.
Outcome Variables
We created variables that represented adoption (how soon after enrollment a PCP from the practice called MCPAP) and use (how frequently MCPAP received calls from a practice). We created 4 categories of adoption: (1) practices with a PCP calling before enrolling or on the day of enrollment; (2) practices for which the first call was 1 to 100 days after enrollment; (3) practices for which the first call was 101 to 365 days from enrollment; and (4) practices for which the first call was >365 days after enrollment. We then created a variable (“use”) that described frequency of calling. To standardize frequency to patient panel size, we calculated the mean number of calls per 1000 pediatric-aged patients empanelled in each practice per year. We then categorized frequency of use by quartiles and created 3 categories: highest quartile, middle quartiles, and lowest quartile.
Practice Characteristics
Because adoption of innovative programs and tools can vary with practice characteristics such as specialty, size, and organizational structure,17–20 we created the following variables: practice type (pediatric or family medicine); number of FTEs providing primary care (1–2 or >2); and panel size (<2000, 2000–9999, or ≥10 000 empanelled child-aged patients). We hypothesized that adoption in newly enrolled practices may be faster among those enrolling after the program was well known across the state. Therefore, we categorized practices as enrolling in 2005 or 2006 and 2007 or later. Because practices located closer to MCPAP regional offices might use the program differently, we determined travel time by car from the practice to the MCPAP office using practice and MCPAP office addresses and Google Maps, and categorized this as ≤35 or >35 minutes. In addition, a pilot program similar to MCPAP was delivered to 22 practices in central Massachusetts from 2003 to 2005, which might affect use of the subsequent program. We categorized practices as being assigned to the central Massachusetts site or another site.
Missing Data
Approximately 25% of practices making calls to MCPAP had incomplete data and therefore were not included. Practices with complete data made 87% of all calls to the program, cared for 85% of all patients in practices enrolled in MCPAP, and made a mean of 102 calls to MCPAP over the study period, versus 32 calls for practices with incomplete data. Missing data were primarily due to incomplete information collected from practices at the time of enrollment as well as practices that had called MCPAP but had not ever officially enrolled.
Data Analysis
We generated descriptive statistics (mean and median values for annual calls per 1000 patients, proportions for other variables) for outcomes and practice and call characteristics. We created graphical representations of the adoption and frequency of use variables and call volume over time. We tested associations between practice and call characteristics and time to adoption (3 categories: adopting either before or during the first 100 days, 101–365 days, and >365 days after enrollment) and frequency of use (3 categories: lowest quartile, middle quartiles, and highest quartile) using Pearson χ2 test. Using multivariate logistic regression models, we then tested associations between practice characteristics and the following 4 specific outcomes: adoption <100 days of enrollment; adoption >365 days after enrollment; highest quartile in frequency of use; and lowest quartile in frequency of use. Covariates practice type (pediatric, family medicine), travel time by car to MCPAP office (≤35, >35 minutes), pediatric patient volume (<2000, 2000–9999, ≥10 000 patients), number of full-time equivalent providers (1–2, >2), MCPAP region (central Massachusetts site, other sites), and year of enrollment (2005 or 2006, 2007 or later) were included in the models. Regression diagnostic tests assessed goodness of fit; we found no undue influence by collinearity of independent variables, interaction terms, or outliers. As a sensitivity analysis, we analyzed the data without the practices assigned to the central Massachusetts MCPAP team, and main findings from the study were largely maintained.
Results:
Characteristics of Practices Participating in MCPAP
Of 285 practices, 82% were pediatric (Table 1). Three-fourths were located ≤35 minutes’ driving time from their regional MCPAP office (range 0–100 minutes). Most practices (67%) enrolled in MCPAP in 2005 or 2006. Practices ranged in size: 28% of practices cared for <2000 pediatric-aged patients, and 23% had ≥10 000 patients. Thirty percent had 1 to 2 FTE PCPs.
TABLE 1.
Characteristics of Practices (n = 285)
| Characteristic | n (%) |
|---|---|
| Practice type | |
| Pediatric | 238 (82) |
| Family medicine | 47 (18) |
| Travel time from practice to MCPAP regional site by car, mina | |
| ≤35 min | 217 (77) |
| >35 min | 68 (23) |
| Pediatric-age patients in practice, n | |
| <2000 | 77 (28) |
| 2000–9999 | 142 (50) |
| ≥10 000 | 66 (23) |
| FTE PCPs in practice, n | |
| ≤2 | 86 (30) |
| >2 | 199 (70) |
| MCPAP region | |
| Central Massachusetts | 56 (20) |
| Boston area or Western Massachusetts | 229 (80) |
| Year of enrollment in MCPAP | |
| 2005 or 2006 | 190 (67) |
| 2007 or later | 95 (33) |
Range 0–100 min, median 21 min.
Adoption and Frequency of Use of MCPAP
The mean time to a practice’s adoption of MCPAP was 178 days (median 78 days). More than half (55%) called within the first 100 days (Table 2). By 1 year after enrollment, 84% of practices had made at least 1 call (Fig 1), and all practices included in this analysis had adopted by the end of the study period. Frequency of use varied: 14% had call volumes of >10 calls/year per 1000 patients, with the majority of practices calling much less frequently (Fig 2). The mean call frequency was 15.5 calls/year per 1000 patients for practices in the highest quartile and 0.4 calls/year per 1000 patients for the lowest quartile, a 38-fold difference (Table 3).
TABLE 2.
Adoption of MCPAP By Practices (n = 285)
| Time Until Adoption, da | n (%) | Proportion of All Calls, % |
|---|---|---|
| 0–1b | 43 (15) | 19.6 |
| 2–100 | 113 (40) | 36.3 |
| 101–365 | 82 (29) | 33.5 |
| >365 | 47 (16) | 10.5 |
Time elapsed between enrollment and a practice’s first call.
First call made before enrollment or on the day of enrollment.
FIGURE 1.
Time from enrollment to practices’ first call to MCPAP.
FIGURE 2.
Variation in frequency of calls by practices to MCPAP.
TABLE 3.
Frequency of use of MCPAP by practices (n = 285)
| Frequency of Usea | Mean | Median | SD | Proportion of All Calls, % |
|---|---|---|---|---|
| All practices | 5.2 | 2.2 | 11.6 | 100 |
| Quartile | ||||
| Highest | 15.5 | 10.3 | 19.7 | 60.0 |
| Middle | 2.4 | 2.2 | 1.2 | 37.6 |
| Lowest | 0.4 | 0.4 | 0.2 | 3.1 |
Annual number of calls per 1000 empanelled patients.
Call Volume Over Time and Characteristics of Calls
Call volume overall increased over time to >600 calls per month in 2011 (Fig 3). The most common reason for calling was medication questions (31%) (Table 4), and ADHD was the most common diagnosis discussed (33%). Call characteristics varied among practices with different adoption and use patterns. In call outcome, for example, patients prompting calls from practices in the lowest use quartile were less likely to remain in the care of the PCP and less likely to be referred to the MCPAP care coordinator or a community-based therapist, but more likely to be referred to a community-based psychiatrist.
FIGURE 3.
Number of calls per month to all MCPAP sites by PCP, May 2005 to July 2011.
TABLE 4.
Characteristics of Calls to MCPAP (n = 29 202) and Association With Adoption and Frequency of Use of Practice Making the Call
| Characteristic | All Practices | Time Until Adoption, d | Quartile | ||||
|---|---|---|---|---|---|---|---|
| <100 | 100–365 | >365 | Highest | Middle | Lowest | ||
| Reason for calla | |||||||
| Medication question/evaluation | 31 | 33 | 30 | 28d | 30 | 33 | 33d |
| Resources question | 27 | 28 | 22 | 30d | 28 | 25 | 26d |
| Diagnosis question | 27 | 24 | 27 | 32d | 25 | 28 | 28d |
| Parent guidance | 5 | 6 | 4 | 3d | 5 | 5 | 6 |
| Diagnosisb | |||||||
| ADHD | 33 | 34 | 32 | 32d | 33 | 33 | 26d |
| Anxiety | 24 | 25 | 24 | 24c | 25 | 25 | 26 |
| Depression | 23 | 23 | 24 | 21c | 23 | 24 | 25c |
| Deferred | 12 | 12 | 11 | 13c | 13 | 10 | 10d |
| Other | 11 | 10 | 12 | 12d | 11 | 11 | 9 |
| Oppositional defiant disorder | 10 | 11 | 8 | 9c | 11 | 10 | 8c |
| Outcome | |||||||
| Refer back to PCP | 54 | 53 | 52 | 57d | 53 | 53 | 46d |
| Refer to MCPAP care coordinator (e.g., for identifying resources) | 25 | 28 | 20 | 26d | 27 | 22 | 21d |
| MCPAP psychiatrist 1-time evaluation | 21 | 22 | 22 | 17d | 22 | 22 | 21 |
| Community-based therapist appointment | 17 | 16 | 18 | 19d | 20 | 14 | 12d |
| Refer to community psychiatrist | 6 | 6 | 7 | 6d | 5 | 8 | 12d |
| Emergency department/inpatient | 2 | 2 | 1 | 2d | 1 | 2 | 3c |
Values are expressed as proportion of all calls (%). Categories of call characteristics are not mutually exclusive.
The following reasons, collectively, were <5% of all calls: patient in crisis, follow-up of previous patient, school issues, and second opinion.
The following diagnoses, collectively, were <5% of all calls: Asperger syndrome, conduct disorder, developmental delay, pervasive developmental disorder, and posttraumatic stress disorder.
P < .05 for group comparison.
P < .001 for group comparison.
Association of Practice Characteristics With Adoption and Use of MCPAP
In bivariate (Table 5) and multivariate (Table 6) analyses, adoption within 100 days was positively associated with assignment to the central Massachusetts site (adjusted odds ratio [aOR] 4.42, 95% confidence interval [CI] 2.16–9.04) and enrollment in 2007 or later (OR 4.09, 95% CI 2.23–7.49), and these characteristics were negatively associated with adoption >365 days. Practices with large panels were less likely to have adopted late (>365 days) (aOR 0.21, 95% CI 0.07–0.74, vs <2000 patients), but panel size was not associated with adoption within 100 days.
TABLE 5.
Association Between Practice Characteristics and Adoption and Frequency of Use of MCPAP
| Characteristic | Time Until Adoption, d | Frequency of Use, Quartile | ||||||
|---|---|---|---|---|---|---|---|---|
| <100 | 100–365 | >365 | P | Highest | Middle | Lowest | P | |
| Practice type | .064 | .476 | ||||||
| Pediatric | 24 (15) | 10 (12) | 13 (27) | 15 (15) | 21 (15) | 16 (21) | ||
| Family medicine | 132 (85) | 72 (87) | 34 (72) | 56 (85) | 123 (85) | 60 (79) | ||
| Travel time from practice to MCPAP regional site by car, min | .331 | .031 | ||||||
| ≤35 min | 114 (73) | 67 (82) | 35 (77) | 56 (78) | 115 (81) | 46 (65) | ||
| >35 min | 42 (27) | 15 (18) | 11 (23) | 16 (22) | 27 (19) | 25 (35) | ||
| Pediatric-age patients in practice, n | .015 | .004 | ||||||
| <2000 | 39 (25) | 17 (21) | 21 (45) | 27 (38) | 32 (23) | 18 (25) | ||
| 2000–9999 | 81 (52) | 40 (49) | 21 (49) | 39 (54) | 73 (51) | 30 (32) | ||
| ≥10 000 | 36 (23) | 25 (31) | 5 (11) | 6 (8) | 37 (26) | 23 (42) | ||
| FTE PCPs in practice, n | .126 | .894 | ||||||
| ≤2 | 44 (28) | 22 (27) | 20 (43) | 21 (29) | 42 (30) | 23 (32) | ||
| >2 | 112 (71) | 60 (73) | 27 (57) | 51 (70) | 100 (70) | 48 (68) | ||
| MCPAP region | .001 | <.001 | ||||||
| Central Massachusetts | 43 (27) | 9 (11) | 4 (9) | 27 (38) | 24 (17) | 5 (7) | ||
| Boston area or Western Massachusetts | 113 (72) | 73 (89) | 43 (91) | 45 (62) | 118 (83) | 66 (93) | ||
| Year of enrollment in MCPAP | <.001 | .491 | ||||||
| 2005 or 2006 | 88 (56) | 64 (78) | 38 (81) | 51 (69) | 95 (67) | 45 (63) | ||
| 2007 or later | 68 (44) | 18 (22) | 9 (19) | 22 (31) | 47 (33) | 26 (37) | ||
Values are expressed as n (%).
TABLE 6.
Multivariate Logistic Regression: Association of Early and Late Adoption and High and Low Frequency of Use of MCPAP and Practice Characteristics
| Characteristic | Time Until Adoption <100 d | Time Until Adoption >365 d | Highest Quartile Frequency of Use | Lowest Quartile Frequency of Use | ||||
|---|---|---|---|---|---|---|---|---|
| aOR (95% CI) | P | aOR (95% CI) | P | aOR (95% CI) | P | aOR (95% CI) | P | |
| Practice type: pediatricsa | 1.96 (0.89–4.30) | .093 | 0.38 (0.15–0.94) | .036 | 2.13 (0.87–5.23) | .099 | 0.51 (0.23–1.14) | .102 |
| Total travel time by car: >35 minb | 1.48 (0.80–2.75) | .211 | 0.94 (0.41–2.15) | .885 | 1.04 (0.51–2.12) | .909 | 1.95 (1.05–3.65) | .035 |
| Pediatric-age patients in practicec | ||||||||
| 2000–9999 | 1.65 (0.81–3.38) | .167 | 0.42 (0.18–1.00) | .050 | 0.45 (0.21–0.98) | .045 | 1.28 (0.57–2.87) | .545 |
| ≥10 000 | 1.40 (0.58–3.34) | .454 | 0.21 (0.07–0.74) | .015 | 0.10 (0.03–0.31) | <.001 | 2.60 (0.99–6.84) | .053 |
| FTEs in practice: >2d | 1.39 (0.72–2.62) | .328 | 0.72 (0.32–1.57) | .404 | 2.10 (1.03–4.31) | .041 | 0.61 (0.30–1.25) | .178 |
| MCPAP region: central Massachusettse | 4.42 (2.16–9.04) | <.001 | 0.22 (0.07–0.70) | .010 | 3.58 (1.86–6.87) | <.001 | 0.24 (0.09–0.64) | .005 |
| Year enrolled: 2007 or laterf | 4.09 (2.23–7.49) | <.001 | 0.23 (0.09–0.55) | .001 | 0.82 (0.42–1.58) | .551 | 1.00 (0.54–1.86) | .996 |
Reference categories: afamily medicine; b≤35 min; c<2000; d1–2; eBoston area or Western Massachusetts; and f2005 or 2006.
Frequency of use was associated with panel size, assignment to the central Massachusetts MCPAP site, and travel time to the MCPAP regional office. Only 8% of practices with ≥10 000 patients were in the highest quartile of use, compared with 38% of practices with <2000 patients (P = .004). As with early adoption, practices assigned to the central Massachusetts MCPAP office were more likely to be in the highest quartile of use (aOR 3.58, 95% CI 1.86–6.87). Practices with a travel time of >35 minutes to the regional MCPAP site were more likely to be in the lowest quartile of use (aOR 1.95, 95% CI 1.05–3.65).
Discussion
We examined data from calls to a novel program that provides PCPs with point-of-care child psychiatry expertise to determine variation in adoption and use and practice characteristics associated with different patterns. We present several main findings. First, adoption of this program and use of MCPAP varied considerably among practices, including a 38-fold difference in call frequency between practices in the lowest and highest quartiles. Second, several practice characteristics were associated with early or late adoption and high or low use, such the timing of enrollment, panel size, and distance from the MCPAP office. Third, assignment to a particular MCPAP office that had served as the program’s pilot was associated with early adoption and higher use. Fourth, practices with low call frequency were less likely to call about resources and referral to community therapists.
This is one of the first reports describing practices’ adoption and use of this model. We report a large degree of variability in adoption and use among practices that probably cannot be explained by any natural variation in mental health conditions or patient-level preferences for treatment among patients of different practices. Variation in health services not explained by variation in disease prevalence or patient preference deserves examination because it can reflect discretionary utilization by physicians and health care organizations.21 Optimal use of MCPAP and desired range among practices are not known, and whether variability translates to differences in quality or scope of mental health care as experienced by patients could not be concluded. Beliefs and attitudes about a PCP’s role in mental health differ,9,22 as do local mental health resources,2 likely explaining some variation. However, children and youth experience long delays in treatment.23 Therefore, it is likely this model is not used optimally in some practices with very low call volumes.
Previous research helps to further explain this variation. Limited access to mental health care might influence use of MCPAP, and perceived access varies geographically.4,24 Among practices, some exhibit greater eagerness to change; others are more stagnant.25 Such contextual differences have been found to be related to variable success in implementing quality improvement and complex organizational changes such as team-based care.26 Practices have different structural resources for mental health. For example, some have colocated mental health specialists, streamlined referral processes, and comanagement arrangements,27–29 and for such practices MCPAP may be less relevant. Finally, PCPs are likely heterogeneous in reasons for using MCPAP. Early adoption of MCPAP may signify a PCP’s unmet need for assistance with mental health issues. For some providers, MCPAP may represent a tool that replaces a less efficient process. For example, a physician might call MCPAP in lieu of referring to a psychiatrist with a long wait. For other providers, MCPAP might signal an uncomfortable expansion of one’s scope of care (e.g., prescribing antidepressants when one has never done so) and this can be a complex decision.4 Thus, for some, relative advantage over alternatives would likely be a more influential driver, and for others, innate personality differences and evolving practice norms may play a role.30
We examined practice and call characteristics and adoption and use and found several patterns. First, practices enrolling during or after 2007, after MCPAP was more established, were more likely to adopt the program in the first 100 days, consistent with previous work that examined importance of visible use by others and local leaders’ opinions before adopting a new tool.31 Large panel size was associated with more rapid adoption but overall lower use. Large practices likely encounter patients with mental health problems more frequently simply because of exposure and may be able to support a greater internal structure to handle mental health problems,32 and therefore may need MCPAP’s support less often.
Practices assigned to the central Massachusetts MCPAP office were more likely to be early adopters and frequent callers. This was the site of the pilot program, which ran for 18 months in 2003 to 2005, and ∼30% of practices assigned to this office may have participated.16 Whereas MCPAP has a standard method of enrolling practices and carrying out program procedures, interactions with PCPs, particularly in ongoing outreach efforts after enrollment, may have been subtly different from other sites, altering the perceived value and visibility of the central Massachusetts office. Finally, regional differences in PCPs’ ability to access mental health resources may vary such that MCPAP fills a need more for central Massachusetts than for other regions.
There are several limitations to our study. First, we studied the program in Massachusetts, an urban state with a high number of child psychiatrists per capita.1 Although others have programs based on MCPAP, use may vary geographically. Calls about the same patient made on different days would be counted as separate calls in our analysis, possibly inflating some practices’ use. We had incomplete data on many practices, and we were unable to discern which practices participated in the pilot program. Finally, we were unable to examine behaviors of individual PCPs, as this was not captured in the data collection.
Conclusions
Thirty-one other states or regions have programs similar to MCPAP, and this model represents one way for pediatricians to be prepared to care for such patients and obliquely addresses child psychiatry workforce shortages, and a low rate of mental health referral completion.33 With one-time consultation sessions as a key component, this model could also incorporate telepsychiatry to reduce geographic barriers to use.34 With growth in off-label prescribing and polypharmacy,3,35 use of this model may facilitate more appropriate prescribing. We identified considerable variability in adoption and use and identified reasons for variability, displaying opportunities to decrease variability through targeted implementation efforts. Future research should explore what specific efforts will accomplish this.
Footnotes
Dr Van Cleave conceptualized and designed the study, oversaw data analysis and interpretation, and drafted the initial manuscript; Dr Le carried out and helped interpret the initial analyses; Dr Perrin substantially consulted on and contributed to study design, data analysis, and interpretation; Drs Le and Perrin reviewed and revised the manuscript; and all authors approved the final version as submitted.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: All phases of this study were supported by National Institute of Mental Health (NIH) grants K23MH083885 and R25MH080916. Funded by the National Institutes of Health (NIH).
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
References
- 1.Thomas CR, Holzer CE, III. The continuing shortage of child and adolescent psychiatrists. J Am Acad Child Adolesc Psychiatry. 2006;45(9):1023–1031 [DOI] [PubMed] [Google Scholar]
- 2.Kim WJ, American Academy of Child and Adolescent Psychiatry Task Force on Workforce Needs . Child and adolescent psychiatry workforce: a critical shortage and national challenge. Acad Psychiatry. 2003;27(4):277–282 [DOI] [PubMed] [Google Scholar]
- 3.Comer JS, Olfson M, Mojtabai R. National trends in child and adolescent psychotropic polypharmacy in office-based practice, 1996-2007. J Am Acad Child Adolesc Psychiatry. 2010;49(10):1001–1010 [DOI] [PMC free article] [PubMed]
- 4.Richardson LP, Lewis CW, Casey-Goldstein M, McCauley E, Katon W. Pediatric primary care providers and adolescent depression: a qualitative study of barriers to treatment and the effect of the black box warning. J Adolesc Health. 2007;40(5):433–439 [DOI] [PubMed] [Google Scholar]
- 5.Foy JM, American Academy of Pediatrics Task Force on Mental Health . Enhancing pediatric mental health care: report from the American Academy of Pediatrics Task Force on Mental Health. Introduction. Pediatrics. 2010;125(suppl 3):S69–S74 [DOI] [PubMed] [Google Scholar]
- 6.Kelleher KJ, McInerny TK, Gardner WP, Childs GE, Wasserman RC. Increasing identification of psychosocial problems: 1979-1996. Pediatrics. 2000;105(6). Available at: www.pediatrics.org/cgi/content/full/105/6/e1313 [DOI] [PubMed] [Google Scholar]
- 7.Olfson M, Blanco C, Wang S, Laje G, Correll CU. National trends in the mental health care of children, adolescents, and adults by office-based physicians. JAMA Psychiatry. 2014;71(1):81–90 [DOI] [PubMed]
- 8.Davis CC, Claudius M, Palinkas LA, Wong JB, Leslie LK. Putting families in the center: family perspectives on decision making and ADHD and implications for ADHD care. J Atten Disord. 2012;16(8):675–684 [DOI] [PubMed]
- 9.Stein RE, Horwitz SM, Storfer-Isser A, Heneghan A, Olson L, Hoagwood KE. Do pediatricians think they are responsible for identification and management of child mental health problems? Results of the AAP periodic survey. Ambul Pediatr. 2008;8(1):11–17 [DOI] [PubMed] [Google Scholar]
- 10.Sarvet B, Gold J, Bostic JQ, et al. Improving access to mental health care for children: the Massachusetts Child Psychiatry Access Project. Pediatrics 2010;126(6). Available at: www.pediatrics.org/cgi/content/full/126/6/e1191 [DOI] [PubMed]
- 11.Hilt RJ, Romaire MA, McDonell MG, et al. The Partnership Access Line: evaluating a child psychiatry consult program in Washington State. JAMA Pediatr. 2013;167(2):162–168 [DOI] [PubMed]
- 12.Kuehn BM. Pediatrician-psychiatrist partnerships expand access to mental health care. JAMA. 2011;306(14):1531–1533 [DOI] [PubMed]
- 13.The National Network of Child Psychiatry Access Programs. Available at: http://nncpap.org/. Accessed March 4, 2014
- 14.Sheldrick RC, Leslie LK, Rodday AM, Parsons SK, Saunders TS, Wong JB. Variations in physician attitudes regarding adhd and their association with prescribing practices [published online ahead of print November 9, 2012]. J Atten Disord. doi:10.1177/1087054712461689 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chan E, Hopkins MR, Perrin JM, Herrerias C, Homer CJ. Diagnostic practices for attention deficit hyperactivity disorder: a national survey of primary care physicians. Ambul Pediatr. 2005;5(4):201–208 [DOI] [PubMed] [Google Scholar]
- 16.Connor DF, McLaughlin TJ, Jeffers-Terry M, et al. Targeted child psychiatric services: a new model of pediatric primary clinician—child psychiatry collaborative care. Clin Pediatr (Phila). 2006;45(5):423–434 [DOI] [PubMed] [Google Scholar]
- 17.Taliaferro LA, Hetler J, Edwall G, Wright C, Edwards AR, Borowsky IW. Depression screening and management among adolescents in primary care: factors associated with best practice. Clin Pediatr (Phila). 2013;52(6):557–567 [DOI] [PubMed]
- 18.Patel V, Jamoom E, Hsiao CJ, Furukawa MF, Buntin M. Variation in electronic health record adoption and readiness for meaningful use: 2008-2011. J Gen Intern Med. 2013;28(7):957–964 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Goldberg DG. Primary care in the United States: practice-based innovations and factors that influence adoption. J Health Organ Manag. 2012;26(1):81–97 [DOI] [PubMed] [Google Scholar]
- 20.Menachemi N, Powers TL, Brooks RG. Physician and practice characteristics associated with longitudinal increases in electronic health records adoption. J Healthc Manag. 2011;56(3):183–197, discussion 197–198 [PubMed] [Google Scholar]
- 21.Wennberg JE. Unwarranted variations in healthcare delivery: implications for academic medical centres. BMJ. 2002;325(7370):961–964 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Olson AL, Kemper KJ, Kelleher KJ, Hammond CS, Zuckerman BS, Dietrich AJ. Primary care pediatricians’ roles and perceived responsibilities in the identification and management of maternal depression. Pediatrics. 2002;110(6):1169–1176 [DOI] [PubMed] [Google Scholar]
- 23.Wang PS, Berglund P, Olfson M, Pincus HA, Wells KB, Kessler RC. Failure and delay in initial treatment contact after first onset of mental disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):603–613 [DOI] [PubMed] [Google Scholar]
- 24.Pletcher BA, Rimsza ME, Cull WL, Shipman SA, Shugerman RP, O’Connor KG. Primary care pediatricians’ satisfaction with subspecialty care, perceived supply, and barriers to care. J Pediatr. 2010;156(6):1011–1015 [DOI] [PubMed]
- 25.Goldberg DG, Mick SS, Kuzel AJ, Feng LB, Love LE. Why do some primary care practices engage in practice improvement efforts whereas others do not? Health Serv Res. 2013;48(2 pt 1):398–416 [DOI] [PMC free article] [PubMed]
- 26.Rittenhouse DR, Schmidt L, Wu K, Wiley J. Contrasting trajectories of change in primary care clinics: lessons from New Orleans safety net. Ann Fam Med. 2013;11 suppl 1:S60–S67 [DOI] [PMC free article] [PubMed]
- 27.Foy JM, Earls MF. A process for developing community consensus regarding the diagnosis and management of attention-deficit/hyperactivity disorder. Pediatrics. 2005;115(1). Available at: www.pediatrics.org/cgi/content/full/115/1/e97 [DOI] [PubMed] [Google Scholar]
- 28.Williams J, Shore SE, Foy JM. Co-location of mental health professionals in primary care settings: three North Carolina models. Clin Pediatr (Phila). 2006;45(6):537–543 [DOI] [PubMed] [Google Scholar]
- 29.Guevara JP, Greenbaum PE, Shera D, Bauer L, Schwarz DF. Survey of mental health consultation and referral among primary care pediatricians. Acad Pediatr. 2009;9(2):123–127 [DOI] [PMC free article] [PubMed]
- 30.Rogers EM. Diffusion of Innovations. New York, NY: Free Press; 2003 [Google Scholar]
- 31.Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O. Diffusion of innovations in service organizations: systematic review and recommendations. Milbank Q. 2004;82(4):581–629 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.The Commonwealth Fund. Physician organization in relation to quality and efficiency of care: synthesis of recent literature. Available at: http://www.commonwealthfund.org/usr_doc/Tollen_physician_org_quality_efficiency_1121.pdf. Accessed March 5, 2014
- 33.Rushton J, Bruckman D, Kelleher K. Primary care referral of children with psychosocial problems. Arch Pediatr Adolesc Med. 2002;156(6):592–598 [DOI] [PubMed] [Google Scholar]
- 34.Myers K, Cain S, Work Group on Quality Issues. American Academy of Child and Adolescent Psychiatry Staff . Practice parameter for telepsychiatry with children and adolescents. J Am Acad Child Adolesc Psychiatry. 2008;47(12):1468–1483 [DOI] [PubMed] [Google Scholar]
- 35.Comer JS, Mojtabai R, Olfson M. National trends in the antipsychotic treatment of psychiatric outpatients with anxiety disorders. Am J Psychiatry. 2011;168(10):1057–1065 [DOI] [PubMed] [Google Scholar]