Objective. To identify target areas for interventions to improve communication between pediatric generalists (PCPs) and pediatric subspecialists (SPs) in the outpatient care of children with chronic conditions.
Methods. We constructed a 4-page mailed questionnaire probing communication practices, opinions about the role of communication in care, and perceived barriers and facilitators to PCP-SP communication in the care of children with chronic conditions. In the spring of 2001, we surveyed all 495 New England SPs who were members of the American Academy of Pediatrics (AAP) and/or SP societies and a random sample of 495 generalist AAP members in New England. Eligible were those actively providing outpatient care. Most items were rated on a 5-point scale.
Results. Of those eligible, 48% (412/860) completed the questionnaire. Although 98% of respondents agreed that communication was important for good care, reported practices reflected large gaps in this area. Frequent receipt (>60% of the time) of communication about an initial referral was reported by only 28% of SPs. Barriers reported as most important involved inefficiencies in telephone contact, transcription delay, and failure to keep all providers informed when >1 specialist is involved. Important facilitators included letters or phone calls at or before the time of consultation, and clear and specific referral questions from PCPs. PCPs saw communication as more of a problem than did SPs (40% vs 28%), and reported several barriers as more important. Although 86% of respondents had access to e-mail in their practices, <20% used it often.
Conclusions. PCPs and SPs sharing care for children with chronic conditions are troubled by their frequent failure or inability to contact their colleagues by phone and letter. PCPs communicate less frequently than SPs yet perceive more problems with communication. Interventions to promote efficient contact between providers at or before the time of subspecialty visits can lead to improved coordination of care, which in turn may better meet the needs of families.
Providing effective medical care for children with chronic health conditions is frequently a daunting task for child health providers. Comprising ∼15% of children nationwide,1 these children have much higher rates of morbidity, physician visits, and health care expenditures than children in the general population.2,3 Compared with children without a chronic condition, they require more coordination of care, because they may receive care from several different physician subspecialists as well as from providers of other health services (eg, physical therapists and home care nurses). They also may require more frequent reevaluation and more tests, medications, and procedures. All these activities require significant extra time and effort from providers both within visits and outside the visit time.4 When specialists are involved in care, providers and families agree that close involvement of the child’s primary care physician (PCP) is preferred.5–9 This process typically requires frequent communication between the PCP and 1 or more specialty physicians.
Several national groups, notably the Institute of Medicine,10 have discussed the special importance of effective generalist-specialist communication for chronic care.11–13 The American Academy of Pediatrics (AAP) has identified it as 1 of 10 services central to providing a medical home for children with special health care needs.14 Ineffective communication is a problem15 and is a critical target for both research and education.16,17 One recent study of pediatric inpatient care found that poor physician-family communication was the factor that correlated most strongly with parental dissatisfaction.18 To communicate effectively with families, physicians sharing care for a patient must communicate well with each other. Poor communication makes providing care more time-consuming and may result in delayed or missed diagnosis, test duplication, delayed treatment, and potential adverse drug reactions.19 Given the gaps in patient safety uncovered in recent research,20 improving communication among those caring for children with chronic conditions becomes even more important.
Surprisingly little is known about generalist-specialist communication in pediatrics. Studies have been done in internal medicine and general practice, primarily in Canada and Great Britain.21–23 In the United States, 3 studies in the 1980s from general internal medicine and family practice24–26 described rates of communication from specialists to referring physicians ranging from 55% to 80%, and a more recent single-site study mentioned lack of timeliness of information and inadequate referral-letter content as major communication problems between a general internal medicine division and specialty consultants.27 A recent study in pediatrics28 assessed the referral process from the point of view of general pediatricians in a large practice-based research network. Results indicated that specialists communicated their findings to referring generalists 51% of the time and outlined plans for sharing of care in only 31% of cases. Although mention of sharing of care increased referring-physician satisfaction in this study, factors that promote or inhibit this discussion are unknown.
The chronic care model, as conceived by Bodenheimer et al,29 was designed as a tool to improve chronic illness management for adults in the primary care setting. It focuses on providing patient-centered care for chronic conditions via a team approach, incorporating changes at the level of the provider, the practice, and the delivery system. Within this model, generalist-specialist communication is an important part of the domains of delivery system design (creating practice teams) and decision support (integrating specialist expertise). We conceptualize effective generalist-specialist communication as information transfer between physicians that enables them to work efficiently as a team to provide coordinated care for children and their families. In the referral-consultation process, there are 3 points at which communication is critical: when the generalist and family make the initial decision to refer; when the specialist has evaluated the child and makes a plan for diagnosis and treatment; and when follow-up care by either provider is relevant to the management plan.
We believe that optimizing generalist-specialist communication has the potential to increase patient safety, use health care resources more efficiently, and better meet the needs of families. In a previously published qualitative study,30 we identified a variety of factors that local providers believe may hinder communication as well as some successful practices. The objective of the present study was to determine the priority of these factors as targets for office-based interventions. We reasoned that by using a larger and more representative sample of physicians than the qualitative study and having them describe their communication practices and rate the importance of barriers and facilitators to effective communication, we would gain more insight into addressing this problem. In our previous work, specialists and generalists also described differing communication needs: specialists were more interested in the question asked by the generalist, whereas generalists in turn were more interested in a timely answer to questions and specific guidance about management. Consequently, we hypothesized that generalists and specialists might have different perceptions of communication, requiring different approaches to effect improvement.
Eligible participants were posttraining, actively practicing members of the AAP and/or pediatric specialty societies who practiced in New England and provided outpatient primary care or consultation. To focus on the specialties that typically provide care on an ongoing basis rather than a 1-time consult, we excluded emergency physicians, anesthesiologists, intensivists, neonatologists, geneticists, radiologists, and pathologists.
Initial Generalist and Specialist Samples
In February 2001, a random sample of 550 AAP members not in any specialty section was generated to form the initial “generalist” sample. For the “specialist” sample, a similar list composed of all eligible members from 17 different AAP specialty sections was generated. The list was supplemented by hand-searching the 2000–2001 membership list of the AAP for members whose addresses indicated that they were pediatric specialists and membership directories from 7 pediatric specialty societies. After eliminating duplicates and names known to the authors to be ineligible, 495 physicians remained in each group.
We designed the questionnaire using the results of a physician focus-group study previously described30 to identify barriers and facilitators to effective communication in the shared care of children with chronic conditions. Domains from the study that were addressed in the questionnaire included timing of communication, content, method (including e-mail), system supports/barriers, and provider education. Critical communication points in the care process, specifically when the decision to refer is made, when the consultation has occurred, and when follow-up care is arranged, formed the basis for many of the questions. Because no published surveys on this topic were known to exist, we administered an initial 90-item questionnaire in person to 10 local general and subspecialty pediatricians to gauge content and face validity, ease of understanding, and optimal length. We revised the questionnaire and then mailed it to focus-group participants and 10 other general and subspecialty pediatricians. Comments were solicited from all participants, and questions that were redundant or confusing were eliminated. The final 4-page, 48-item questionnaire asked specifically about the shared care of children with chronic conditions and covered opinions about the role of communication in care, communication practices, barriers and facilitators to communication, and the use of e-mail. To ensure that questions were relevant to both generalists and specialists, the questionnaires for each group differed slightly in wording, and 4 questions were different between questionnaires.
Questions were closed-ended, and most were in a 5-point, Likert-type scale format. Questions about communication practices such as “percentage of time you send a referral letter” had possible responses for each practice of 0% to 100% in 20-point increments. Questions about potential barriers and facilitators to communication had possible responses ranging from “insignificant barrier” to “important barrier” and “improves communication ‘not at all’ to ‘a great deal.’”
We sent 3 mailings at 4-week intervals between February and April 2001 and sent a reminder card 2 weeks after the original mailing and a phone/fax reminder 2 weeks after the second mailing. We included a personalized cover letter with the mailings along with a card explaining eligibility and general versus specialty group membership. If participants indicated on the card that they were members of the other group, then we changed their group status and mailed the appropriate questionnaire. The questionnaire was accompanied by a postage-paid return envelope. We offered participants a prepublication summary of study results for completing the questionnaire. Questionnaires were coded to protect the identity of study participants. The Human Subjects Committee of the University of Massachusetts Medical School and the Institutional Review Board of the Fallon Clinic approved the study.
Data were entered into a Microsoft Access database, with statistical analysis done by using SPSS. Questions about opinions and potential barriers and facilitators were rated for importance according to the proportion of respondents rating them as a 4 or 5 on the 5-point scale; generalist and specialist groups were compared by using χ2 tests. Continuous numeric variables were compared by using the Student t test. In situations where multiple comparisons were done (eg, lists of barriers or facilitators), a level of α = 0.01 was used to determine statistical significance; otherwise, a level of α = 0.05 was used.
Of the 990 questionnaires initially mailed, 130 were undeliverable or the addressees described themselves as ineligible. Seventeen physicians in the original “generalist” sample and 14 in the original “specialist” sample requested to change groups. Questionnaires were completed by 224 (50.6%) of 443 eligible specialists and 188 (45.1%) of 417 eligible generalists for a total response rate of 47.9% (P = .11 between groups). There were no significant differences in response rate by state of residence.
Table 1 describes sample demographics. As anticipated, specialists were more likely to be male, were more likely to work in an urban environment, and had more postresidency experience than generalists. Additionally, 8% of generalists spent some time seeing patients in a subspecialty role, and 11% of specialists saw patients occasionally in a primary care role. When compared with AAP Periodic Surveys of Fellows from 2000–2002 and another survey,31,32 there were no significant differences between our sample and national samples in terms of number of years in practice; the proportions from each state were also similar to those described in national surveys.
Role of Communication in Care
Although 98% of generalists and specialists agreed that communication was important in providing good care, they disagreed about several issues. When answers of “agree” and “agree strongly” were grouped together, 40% of generalists saw communication as problematic, while only 28% of specialists did (P = .015). Overall, 25% of respondents disagreed that communication was a problem in their practice. Table 2 summarizes other selected opinions about communication, including all that were different between groups. Generalists felt more strongly than specialists that timely use of the telephone was helpful in improving patient and provider satisfaction as well as being an efficient way to communicate with other providers.
Both groups agreed that consultation letters should include the reasoning behind a specialist’s treatment plan (93% responded “agree” or “agree strongly”), the expected course of the patient’s condition (91%), and information about how to manage acute problems (89%). They agreed modestly that more education of providers is necessary to improve communication (57%) and had mixed opinions about the importance of e-mail (41%).
For sending and receiving communication, we defined communication as “frequent” if respondents reported that it occurred >60% of the time. Although 60% of generalists reported frequent sending of information about initial referrals, only 28% of specialists reported receipt of generalist communication about a first referral frequently (P < .001). Similarly, although 92% of specialists reported sending communication about an initial consultation frequently, only 70% of generalists reported frequent receipt of specialist communication about an initial consultation (P < .001). The pattern was the same for repeat consults: 19% of generalists sent and 8% of specialists received communication from generalists about repeat referrals frequently, and 87% of specialists sent/62% of generalists received communication from specialists about repeat consultations frequently.
Discussion of sharing of responsibility for care was also uncommon, with only 22% of generalists and 45% of specialists reporting that they frequently mention sharing of care. To explore this topic further, we asked about the importance of discussing responsibility for routine follow-up care, acute situations, feedback to the patient’s family, and negotiating division of responsibility over time. All respondents found the importance of each of these to be high (levels of agreement ranging from 71%–85%).
To assess the role of the family as a conduit for information, we asked whether the family was frequently used as the primary avenue of communication between providers and whether respondents were comfortable with families taking that role. Although 39% mentioned that they “agree” or “strongly agree” with the first statement, only 12% agreed with the second.
Barriers and Facilitators to Communication
Of 10 potential barriers queried, the 5 most important for each group are displayed in Table 3. Three of the top 5 had delays in timeliness of communication in common, with 2 concerning telephone difficulties. When generalist and specialist responses were compared, several barriers were rated more highly by generalists. Despite generalists’ perceptions of telephone use as important and efficient, they viewed telephone delays as more of a barrier than specialists did; they also viewed specialists referring to other specialists without involving the generalist and incomplete communication when multiple specialists are involved as more problematic. No individual barriers were seen as more of a problem by specialists.
Table 4 displays the 5 most important potential facilitators for each group of 10 queried. Timeliness again was prominent, with specialists mentioning clear and specific questions from generalists as being very helpful. Generalists and specialists agreed on most facilitators. Both groups viewed a referral letter received before the consultation as extremely helpful. “Curbside consults” were seen as more helpful by generalists, as was the use of a form to pass information between providers (59% vs 35%; P < .001).
Overall, 87% of providers (78% of generalists and 94% of specialists; P < .001) had access to e-mail in their practices, and of those with access, it was “very available” for 62%. Despite this, only 10% of those with access used it “very often” (5 on a 5-point scale), and only 37% with access used it moderately or more often (3, 4, or 5 on a 5-point scale).
Among those with access to e-mail, 61% felt that it is useful. This group did not agree that a lack of confidentiality was a barrier to its use (31% in agreement) or that difficulty in documentation was a barrier (36%). Specialists viewed e-mail as more useful than generalists (68% vs 52%; P < .001), and they were more likely to use it frequently (25% vs 13%; P = .01). Generalists reported more than specialists that not knowing if e-mail is received is a barrier to its use (45% vs 32%; P < .01).
This survey of pediatricians provides information essential to the effort to improve communication between providers in the shared care of children with chronic conditions. It extends the findings of previous qualitative work30 to a larger population and suggests some areas that may be prioritized. It agrees with a previous study28 that mention of sharing of care by specialists in referral letters, while seen by generalists as important, occurs uncommonly. However, the current study points out that specialists also value discussion by generalists of sharing of care, and that continuing this discussion in communication about subsequent visits is also important. Both generalists and specialists provided evidence that shared care is not occurring, although their opinions were mixed about how much of a problem it is for them individually. Generalists admitted that they communicate less frequently than specialists, found inadequate communication to be a greater problem, and perceived many potential barriers as more important. Both groups agreed about ways to improve communication.
There was a large reported gap between the sending and timely receipt of information, especially referrals from generalists to specialists. Specialists have previously described this as a primary reason for wasted time during consultation visits.30 More concerning for the ongoing care of children with chronic conditions is an even lower rate of communication about follow-up visits. Most perceived barriers and facilitators concerned system factors affecting the timeliness of communication, especially by phone, which agrees with a recent study of adults referred within an academic medical center.27 Together with a recent pediatric practice-based study showing phone calls by physicians as the most costly aspect of office-based care coordination for children with special health care needs,4 these data suggest that making phone contact more efficient may reduce costs significantly. Another dimension, unique to children with chronic conditions, is the care that must be taken when multiple specialists are involved. Without clear coordination of tasks, delays in care can occur, and there is potential for medical error as well as duplication and omission of services. This is not a new concept,33,34 although good solutions have yet to be implemented.
Changes informed by these data need to increase communication rates without increasing provider time. Several of these changes might be accomplished by interventions at the system level including automatic sending of clinical information at the time a referral is initiated and keeping a list of providers for children with multiple specialists to whom communication should be sent. Changes may need to be tailored to the needs of different groups of practices, and rather than a universal solution, a quality-improvement model such as that used for asthma and attention-deficit/hyperactivity disorder35 may be needed. The study of provider groups that are known to communicate well, as well as those within integrated delivery systems, may be useful in this regard.
Use of electronic communication to improve timely contact between providers, encouraged by the Institute of Medicine,10 is a promising approach. It is both fast and “asynchronous,” meaning that senders and recipients need not both be available simultaneously. Patterns of use and perceptions of e-mail in this study illustrate how this technology has diffused into practice. Although e-mail access was widespread at the time of the survey, it was embraced more by specialists, a finding similar to another recent practice-based study.36 Generalists saw a lack of widespread use as a barrier. When combined with the finding that generalists perceive use of the telephone as more useful than specialists, these data imply that at the time of the survey, use of e-mail between providers was not feasible for everyone, and “lower-technology” options should be considered until e-mail is accepted more universally. Alternatively, interventions might be used to accelerate the acceptance of e-mail use in patient care by generalists. For e-mail to be used more widely, standards for use of e-mail by providers37 must be disseminated, and provider concerns about appropriate and effective use38 must be addressed. Additionally, although this was a concern for a minority of providers on our sample, security concerns must be addressed so that the use of e-mail is facilitated rather than prohibited.
Within the content of communication, respondents acknowledged that explicit discussion of co-management between providers, a key determinant of provider satisfaction,28 is very important but rarely mentioned. Specific questions from generalists to specialists and “anticipatory guidance” for the generalist about the chronic condition were seen as important as well. More education of providers about communication was also seen as helpful by a majority of respondents, which has important implications for residency and fellowship training, where patterns of communication are first learned.
When communication is inadequate between providers, families of children may be relied on for important information. This situation was encountered frequently by 39% of respondents in our study, few of whom felt comfortable with the situation. Although it is essential to include families in constructing treatment plans, it is potentially hazardous for providers to rely on them as the primary conduit for complex information. The generalist’s role as the interpreter of information can be compromised, leaving families confused, and the potential for medical error is introduced if information is transmitted incorrectly. This situation has not yet been explored in the literature, and additional work with families is crucial to determine their optimal role in this process.
Several factors limit the generalizability of this study. Our response rate after 3 mailings (48%) did not capture a majority of eligible subjects, although it was typical of other published physician surveys.39 If providers for whom communication is not an important issue were less likely to respond (response bias), the importance of the topic may have been overstated in our sample. If respondents’ actual practices differed from their own reports (recall bias), the actual differences in reports of sending and receiving communication could be smaller. However, even the perception of a large problem in receiving information is important, agrees with our previous study, and deserves additional investigation. Additionally, nonphysicians and physicians who were not listed in the directories of either the AAP or the pediatric specialty societies in our sample did not participate, so differing views of nonphysicians or physicians not in these organizations cannot be assessed. However, the demographics of the sample of pediatricians reflect accurately those of recent national surveys.31,32 Finally, practice experiences in New England, where access to pediatric specialty care may be easier than in other regions, may not reflect those of other regions.
Implications for Practice
The findings in this study suggest several targets for intervention to improve generalist-specialist communication:
Timely, systematic information transfer from generalists to specialists at the time of referral, and between each provider for follow-up, tailored to the needs and preferences of each practice;
Discussion of co-management, explicit questions from generalists to specialists, and education of generalists by specialists about conditions as standard content for communication;
Improvement of the efficiency of telephone contact through direct lines from physician to physician or protocols designed to address the needs of different practices;
Sharing of information among generalists and all specialists involved in care coordinated through the child’s medical home; and
Ensuring that families, while included fully in communication, are not the sole method of communication between providers.
Although PCPs and specialists desire to take an active role in sharing care for their patients with chronic conditions, they frequently do not communicate with each other. Generalists, who report that they communicate less frequently than specialists, also feel that communication is more of a problem. This may create gaps in the care of children with chronic conditions, and poses an important problem for successful implementation of the chronic care model in the medical home. Reported barriers to communication involve inefficient contact between providers and failure to maintain contact when multiple providers are involved. Interventions that promote efficient contact between providers at or before the time of subspecialty visits can lead to more effective communication and coordination of care, which in turn may better meet the needs of families. Measuring the impact of improving generalist-specialist communication on both the health care process and the well-being of children and their families is a priority.
This work was supported in part by grants from the Meyers Primary Care Institute and the Robert Wood Johnson Foundation Generalist Physician Faculty Scholars Program (42206).
We thank Dr. Eileen Ouellette for her advice and support as AAP District I Chair and Drs. Richard “Mort” Wasserman, Jerry Gurwitz, Kurt Stange, and Evan Charney for their thoughtful advice and review of the manuscript.
- ↵Antonelli R, Peer M. Providing a medical home: the cost of care coordination services in a community-based, general pediatric practice [abstract]. Pediatric Academic Societies’ Annual Meeting; May 7, 2002
- ↵Miller MR, Forrest CB, Kan JS. Parental preferences for primary and specialty care collaboration in the management of teenagers with congenital heart disease. Pediatrics.2000;106 :264– 269
- Forrest CB, Glade GB, Starfield B, Baker AE, Kang M, Reid RJ. Gatekeeping and referral of children and adolescents to specialty care. Pediatrics.1999;104 :28– 34
- ↵Young PC, Shyr Y, Schork A. The role of the primary care physician in the care of children with serious heart disease. Pediatrics.1994;94 :284– 290
- ↵Institute of Medicine, Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press; 2001
- Serwint JW, Seidel HM. Communication: the key to effective consultation. Contemp Pediatr.1999;16 :138– 152
- ↵American Academy of Pediatrics, Medical Home Initiatives for Children with Special Needs Advisory Committee. The medical home. Pediatrics.2002;110 :184– 186
- ↵American Academy of Pediatrics, Committee on Children with Disabilities. Care coordination: integrating heath and related systems of care for children with special needs. Pediatrics.1999;104 :978– 981
- ↵The Future of Pediatric Education II Task Force. The Future of Pediatric Education II: organizing pediatric education to meet the needs of infants, children, adolescents, and young adults of the 21st century. Pediatrics.2000;105(suppl) :163– 212
- Newton J, Eccles M, Hutchinson A. Communication between general practitioners and consultants: what should their letters contain? BMJ.1992;304 :821– 824
- ↵Cornwall PL. Communication between general practitioners and child psychiatrists. BMJ.1993;306 :692– 693
- ↵American Academy of Pediatrics. Personal/practice characteristics of pediatricians. Available at: www.aap.org/research/periodicsurvey/practicecharacter2002color.htm. Accessed February 13, 2003
- ↵Stoddard JJ, Cull WL, Jewett EAB, Brotherton SE, Mulvey HJ, Alden ER. Providing pediatric subspecialty care: a workforce analysis. Pediatrics.2000;106 :1325– 1333
- ↵Palfrey JS, Levy JC, Gilbert KL. Use of primary care facilities by patients attending specialty clinics. Pediatrics.1980;65 :567– 572
- ↵Kanthor H, Pless B, Satterwhite B, Myers G. Areas of responsibility in the health care of multiply handicapped children. Pediatrics.1974;54 :779– 785
- ↵Kleiner KD, Akers R, Burke BL, Werner EJ. Parent and physician attitudes regarding electronic communication in pediatric practices. Pediatrics.2002;109 :740– 744
- ↵Kane B, Sands DZ. Guidelines for the clinical use of electronic mail with patients. J Am Med Inform Assoc.1998;5 :104– 111
- ↵Bauchner H, Adams W, Burstin H. “You’ve got mail”: issues in communicating with patients and their families by e-mail. Pediatrics.2002;109 :954– 956
- Copyright © 2003 by the American Academy of Pediatrics