A number of reports have estimated the frequency of children with chronic illness and disability in primary care settings as being between 3% and 20%, depending on how the severity of the disability is defined. It has been estimated that 4%, or 3.1 million, of America's children under the age of 21 have some activity-limiting condition.1 The leading causes of activity limitation in children aged 21 and under are mental and nervous system disorders, respiratory disease, musculoskeletal and connective tissue diseases, and a variety of other diagnoses. Children with activity-limiting conditions are twice as likely as other children to be hospitalized over the course of a year and spend four times as many days in the hospital as other children. These children make twice as many visits to physicians, and see non-physician providers such as nurse practitioners, psychologists, social workers, and early intervention workers five times more frequently than other children.
In 1980, estimates were made that if healthcare costs for "normal" children were $263 a year, then for children with complex medical needs, the costs would be at least $760 per year. This 4% of children who had activity- limiting conditions accounted for 11% of the healthcare expenditures for children under the age of 21. 2 A study in Massachusetts over a 3-year period indicated that approximately 2200 children under the age of 18 required technology assistance, including but not limited to oxygen, tracheostomy, dialysis, and other ostomy care. The authors concluded that there was a need for longterm, community-based planning for children dependent on medical technology and that planning needed to include discharge planning, community providers, community resources, and accessibility of community-based services.3
Estimates by pediatricians about the frequency of children with special healthcare needs in their own practices are frequently lower than the stated estimates from academic experts. This difference may reflect a bias related to the pediatrician's perception of the number of children who have disabilities who are cared for in primary care practice. The majority of primary care pediatricians view the principle focus of their practice as being related to provision of preventative services and diagnosis, treatment, and follow up of acute illness, with additional services related to provision of immunization services, developmental or behavioral screenings, and provision of laboratory and diagnostic testing.
Opportunities for caring for children with chronic illness in primary care settings have frequently been limited due to perceptions of the difficulty of integrating care for such children into an office setting. The three principle biases against the successful integration of these children are ( 1 ) that the physician will not have a sufficient knowledge base to care for them, (2) that the physician will not have enough time to adequately or appropriately care for them, and (3) that the reimbursement the physician will receive for such care will not be adequate in terms of the time spent or the effort expended. A fourth bias, which has been articulated occasionally, is the possibility that the care of children with chronic illness would expose the practitioner to increased medicolegal liability. Review of closed claims data from medical liability carriers reveals that the ongoing management of chronic illness and disability does not necessarily expose physicians to increased liability. It is clear, however, that inadequate or inappropriate diagnosis of such conditions at the time of presentation does increase the provider's legal exposure.
It is reasonable to examine each of these previously stated biases individually. The concept that a primary care physician does not have an adequate knowledge base to care for a child with chronic illness is particularly perplexing in the context of new practice environments. Residency training in tertiary-care facilities should prepare the pediatrie generalist for practice encounters with the chronically ill. However, exposure is usually related to treatment of exacerbations or complications, not ongoing management. It may also be the perception of the trainee that he ot she will need to defer to the specialist and thus, de facto, "turf1 the patient. Although an individual practitioner may have never seen, during the course of his or her training or practice, a specific illness or disability, there are adequate resources and references to educate him or her about the problem. It is indeed at times intimidating to be ptesented with a patient whose parents may have reviewed the world's literature about their child's disability. However, the willingness of a physician to learn from specialist colleagues, medical literature, the Internet, and the family serves as a wonderful opportunity to form a therapeutic bond between the child and the parent and the individual practice and practitioner.4
The perception of inadequate time available to care for a child with a disability is all too real in primary care. There are frequently too many patients to be seen in too short a period of time. The primary care physician who cares for children with disabilities and chronic illness will frequently arrive early and stay late to meet the demands of his or her patient population. However, the opportunities to learn and the satisfaction to be gained by caring for this population will frequently outweigh the time spent in an absolute sense.
The final concern about inadequate reimbursement is real. Under the old reimbursement system, there was no easy mechanism for demonstrating time spent. However, in 1992, with the redirection of the American Medical Association's Current Procedural Terminology (CPT) system and the expansion of evaluation and management codes, coupled with the adoption of a resource-based relative value scale, there was a real opportunity to expand opportunities for adequate reimbursement. Sadly, the pediatrie community has frequently chosen to either not understand or not embrace the CPT system. Surgeons are intimately familiar with the sections of the CPT manual that are relevant to them. It is surprising to many practitioners that there are more than 10 pages in the CPT manual that relate to repair of a simple laceration, including issues relating to jaggedness, depth of laceration, and materials used. When primary care physicians, however, are surveyed about codes that they use, they show a tendency to use only a few codes and to code at lower levels. Reasons for downcoding include lack of experience with the CPT system, undervaluation of the worth of pediatrie services, and fear of insurance company audit or oversight. Coding visits at higher levels, even if medically justifiable, may trigger review by third-party payers. A practitioner who has large numbers of chronically ill children in his or her practice must embrace the CPT methodology for as long as the modified fee-forservice environment continues to exist. Because a physician with large numbers of chronically ill children in his or her practice will have to code his or her patient encounters at higher levels of documentation, it is necessary for him or her to document the level of service provided to substantiate the need for higher codes. Although any E and M code can be useful for a child with chronic illness and disability, there are some codes that have particular relevance for a chronically ill patient population. See Table 1 for a listing of such codes, clinical descriptors, and relative value units (RVUs).5 These RVUs are, except for geographic variation in practice expense, intended for national use. Fees are generated by multiplying the global RVU by a conversion factor determined by a Healthcare plan. CPT codes are not uniformly reimbursed by all states and insurers, and variations in reimbursement patterns, although at times bewildering, provide insight into how care for the disabled is viewed by third parties,
Relative Value Units (RVUs)1T
Case Coordination for Children With Special Healthcare Needs
DELIVERY OF CARE
Once the child with chronic illness is integrated into a practice setting, it is important that the staff of the physician's office (including secretaries, receptionists, billing clerks, nurses, and nurse practitioners) have some understanding of the particular problems that children with chronic illness present with in a primary care practice. For example, the office staff should be flexible in terms of appointment scheduling to allow a child and family with a particular handicapping condition easier access to the office. It is incumbent upon physicians and office staff to understand the implications of individual disabilities as they relate to the Americans with Disabilities Act. It is also necessary for the office to understand the concept of reasonable accommodations for children with disabilities (Table 2).
Some authors have begun the process of evaluating efficient office and recordkeeping systems to facilitate the care of children with chronic illnesses. Table 3 lists some available materials that are presently being used in one such system.
As medical care reimbursement evolves from a modified fee-for-service environment to a capitated or prospective payment system, practitioners need to be aware of the determinants that allow for calculation of appropriate cap rates for children without chronic illness and then to build on this dataset with tools to modify this rate in the case of chronically ill children.6
It is important to acknowledge that children with chronic illness will be acted on by managed care in general and capitation in particular. An understanding of data analysis and medical information systems is critical to pediatrie practice. It has particular relevance for understanding the fiscal implications in caring for children with chronic illness. Most capitated schema presented to primary care physicians involve a risk adjustment that, at best, is done on an age and gender basis. Non-adjusted plans, especially for younger patients, need to be carefully evaluated by a practice and probably avoided as being fiscally unsound. However, an agegender adjustment that may be perfectly appropriate for a child without a chronic illness may provide woefully inadequate reimbursement for a child who has significant illness or disability. Although a number of managed care plans acknowledge issues related to risk adjustment and risk assessment, this concept is truly in its developmental infancy.7
Risk assessment refers to making an evaluation of the individual diagnoses, comorbidities, and conditions that any child has when entering a practice. Such an assessment will then be used to develop a risk adjuster to modify a capitation that can more fairly compensate a practice or individual physician around a chronic illness. A frequently asked question about such adjustment is what is the "right number" for a child with a chronic illness. The set answer is "there is no right number." The child must be considered in the context of the individual diagnosis and potential comorbidities.8 Table 4 lists the risk adjusters tfiat might, in fact, be employed to more accurately capture the level of illness or disability of an individual patient. Table 5 includes a number of critical definitions.
Some individuals have suggested that an individual diagnosis or group of diagnoses without other risk adjustment could be employed to reflect a more adequate capitation figure. The fallacy of this approach can easily be illustrated by a single clinical diagnosis: spina bifida. As any clinician who cares for a number of children with spina bifida knows, this individual diagnosis can include children who are wheelchairbound, with hydrocephalus requiring ventriculoperitoneal shunting, and who are fecally incontinent and require manual disimpaction and frequent daily urinary catheterization. At the other end of the spectrum are children who are minimally disabled in terms of motor issues, who may have some minimal bowel or bladder dysfunction but who are, in fact, able to actively participate in full school and sports programs. An individual diagnosis coding schema might grossly undervalue or overvalue the needs of an individual patient.
Description of Supportive Tools and Resources for Chronic Condition Management (CCM)
A number of prospective case mix methodologies presently in development may help to more favorably address the area of risk adjustment. A recent study reviewed existing methodologies for capitation adjustments in pediatrie populations. Although such alternative methods improve capitation rates over a strict demographic model, children with chronic illness are still at risk for fiscal discrimination in a competitive healthcare environment. None of the existing risk adjustment approaches were developed specifically for children with special healthcare needs. The NACHRI approach, which is pediatric-specific, has a broader consensus-based approach.9 The Expert Work Group of the Division of Services for Children with Special Healthcare Needs Oversight Group on Pediatrie Risk Adjustment have made the following recommendations:
1. More research is urgently needed to develop risk adjustment models appropriate for children with special healthcare needs.
2. A new comparative analysis of the strengths and limits of the existing capitation risk adjustment approaches for children with chronic conditions should be undertaken using several Medicaid and commercial datasets.
3. A combination of risk adjustment, reinsurance, and carve-out strategies is needed to prevent serious financial difficulty while rewarding highquality care among plans and providers serving high-cost children.
4- Regulatory strategies, contract requirements, and oversight must accompany risk adjustment strategies.
5. Previous examples of underutilization, particularly among children with developmental, behavioral, and emotional problems, must be taken into consideration when developing capitation rates.
Another reimbursement principle that primary care physicians who care for children with chronic illness must be aware of is the concept of carve-outs. This term refers to care for children whose medical problems are so severe that they can never comfortably fit under a capitation schema. Such patients are best cared for under a modified fee-for-service schema with some sort of enhancement around case management and telephone management issues. The one single factor that should at least raise the question of the need for a carve-out is that, irrespective of the quality of care provided, the physician has little or no discretion over expenditures that will be employed to provide adequate care for a patient.10 ("Carving out" conditions from global capitation rates is a useful strategy for protecting high-cost patients, physicians, and healthcare plans in a managed care environment.)
The purpose of this study was to develop a method for identifying individuals with high-cost medical conditions, to determine the percentage of healthcare spending they represented, and to explore policy implications of carving out their care from managed care capitation models. The main outcome measure was to look at enrollees with total annual payments in excess of $25 000 and the contribution of such highcost enrollees in each health care's total plan. The results demonstrated that among children, the carveout list included 32 diagnoses and procedures. A substantial proportion of high-cost children were those born with congenital anomalies, low birth weight, or with respiratory problems. The carve-out was to identify over 50% of high-cost individuals enrolled in a fee-for-service plan or Medicaid program at approximately 20% of high-cost managed care enrollees.
These data confirm that it is possible to identify the majority of high-cost individuals in health plans with the use of a carve-out list. Carving out high-cost patients for managed care arrangements may protect patients, physicians, and managed care organizations.
Certain types of physicians are more likely to attract high-cost patients due to individual interests or expertise. Children were found to account for a smaller proportion of high-cost individuals. Unlike the adult population, where researchers were able to develop a concise list of relatively common conditions, high-cost patients in the pediatrie population had a wider variety of less common problems. Although both groups have high representations of diagnoses such as cancer and organ system failure accounting for high-cost conditions, a significant number of high-cost children are those who are bom prematurely, those born with low birth weight, or those bom with severe congenital anomalies. It should also be noted that a separate subset of highcost patients include those with severe pneumonia, pulmonary insufficiency, respiratory failure, and respiratory distress syndrome. It is clear there is enormous tension between the delivery of optimal care for children and the need to control costs, but we must distinguish the fiduciary responsibility that the physician has in terms of serving the best needs of his or her patients versus the stewardship responsibility of providing care to a large group of patients with a limited number of dollars.
List of Definitions
A difficult group of children are those who have high costs but have a combination of chronic conditions, thus not easily fitting into a category for carve-out purposes. It is clear that carve-out methodologies do have limitations. However, once a carveout methodology is agreed to, it could be implemented on a statewide basis, whereby each health plan could contribute to a "carve-out pool" to provide funds to these plans and providers who have received an inordinate share of the chronically ill.
A significant problem for both commercially insured patients and for patients covered under state Medicaid systems has been how to safely move children with complex medical needs into managed care scenarios. Although the simplest approach for dealing with this population, from an actuarial point of view, has been to acknowledge that they cost more than children without disabilities, this kind of simplistic system will not allow for effective categorization of such patients. Five states - Missouri, Wisconsin, Ohio, New York, and Massachusetts - have been working with the Boston-based Medicaid Working Group to develop new managed care models for the disabled. Spending for this group has increased at a higher rate than other populations. The philosophy of the Medicaid Working Group is one with which most clinicians would concur. It is their belief that appropriate reimbursement should be tied to the complexity of the conditions of the children cared for. The obvious conundrum is how to develop accurate and effective risk adjusters for patients with chronic illness. Many disabled patients are not necessarily high cost. It is not clear that presently available models are able, at this time, to adequately capture the difficulties of risk-sharing and risk adjustment.
As difficult as it can be at times to care for children with complex illnesses and disabilities, it is sometimes equally difficult to develop mechanisms to successfully transition such children to other providers when their age would mandate that they no longer be part of a pediatrie practice. Families of such children perceive that the care they obtain from their pediatrie providers is frequently more informed and more sensitive than might be offered by physicians who normally care for adults. Although such families may want to receive care from their primary care pediatrician into their mid- and late 20s and ultimately their 30s, this arrangement serves neither the patient nor the pediatrician. Children with complex medical illnesses will develop morbidities and complications in adulthood about which the pediatrician is often illinformed or is ill-equipped to treat. Therefore, a proactive plan needs to evolve to successfully transition such patients.11
First and foremost, planning for transition must begin at least a year in advance of the child's ultimate departure from a practice. Because educational programs for the disabled frequently end by state law by age 22, it would seem to be a logical goal to target the 22nd birthday for transition from a pediatrie practice. The primary care physician should survey family practitioners and internists in his or her community who are willing to accept young adults with complex medical needs. Ideally, if there is a particular area of interest within the medical community around certain diagnostic groups, then such patients could be transferred to physicians who are particularly interested in their medical needs. Not only is it necessary to transfer medical records, but an individual telephone consultation or direct consultation by the pediatrician with the accepting physician is ideal. If the child is technology-dependent or has severe neurologic or motor needs, simultaneous transition to a physiatrist is sometimes helpful. Finally, for the adolescent or young adult patient who has normal cognitive abilities, the empowerment of that patient to begin to assume some of his or her Healthcare needs and responsibilities is critical. Such a process should, in fact, decrease the anxiety and ultimate sense of loss that goes along with the departure of such patients from a pediatrie practice.
HEALTHCARE SYSTEM ISSUES
An often underappreciated area regarding the care of children with chronic illness is the problems they present to the emergency medical system. Children with complex medical needs may be particularly difficult to care for in the context of acute trauma situations. Pediatrie patients account for one third of all emergency room visits and clearly many of these patients will have special healthcare needs. Clinicians have developed a sample emergency dataset, a document that accompanies the child to the emergency room to explain the child's medical diagnoses, a synopsis of the child's past history, ongoing medications, significant baseline physical findings, significant baseline laboratories, and issues around management and contact people. Such a tool can be used on a systems basis to facilitate the care of chronically ill children.12 Repeat admissions may require individual case management strategies to evaluate individual patient needs.
Medical necessity is a term used to decide which services a managed healthcare plan will cover as a benefit. Many managed care organizations have adopted an, adult-oriented approach to medical necessity that does not effectively address issues related to the developmental and behavioral needs of children. In such health plans, they are only required to cover interventions for a medical condition defined as a disease, illness, or injury; implicit in this is that the condition will be of short duration. There are pediatrie conditions, however, which may benefit if addressed early but which can have significant and even crippling consequences if not dealt with. Developmental speech delay is one such example. Services such as pediatrie preventative care, pediatrie speech therapy, prenatal care, obstetrical care, and so on, might be considered discretionary by health plans because they do not treat a medical condition under this medical necessity definition. It is the position of the American Academy of Pediatrics and its Committee on Child Health Financing that medical necessity should be:
1. Appropriate for the age and health status of the individual;
2. Will prevent or ameliorate the effects of the condition, illness, injury, or disorder;
3. Will aid the overall physical and mental growth and development of the individual; and
4. Will assist in achieving or maintaining maximal functional capacity in performing daily activity (Berman S, personal communication, 1997).
It is obvious that children differ from adults. Children may have congenital and acquired disorders that are relatively rare. Clearly, children will require more comprehensive services than required by adults in order to promote optimal physical, intellectual, and emotional development. Although children require such interventional services, outcome studies and evidence-based medicine must be used to determine which interventions are truly effective and which interventions are not.
As pediatricians contemplate both the end of a century and a millennium, some interesting philosophical questions can be raised about pediatrie practice in the future. The successful pediatrician of the future will have to embrace and care for a number of new morbidities that have not been well-addressed in the primary care office in the past. Increasingly, the "action" will be shifting from the hospital setting, and the primary care physician will be responsible for increasing levels of care on an outpatient and a homehealth basis. Although it is anticipated that pediatricians will continue to be the primary and premiere care providers for infants, children, and adolescents, this is by no means assured. Evolutionary theories suggest that as populations transition during times of environmental stress, traits that have had little or no evolutionary advantage in the past may hold the key to future survival of a species. Although this assumption may seem granthose, it is likely that how pediatricians choose to care for the most vulnerable of their patients will have strong predictive value about the future and survival of pediatrics as a discipline. Pediatricians should commit themselves to adopting the National Agenda for Children with Special Health Needs: Achieving the Goals 2000:
1. Sustain the national consensus for building community-based and family-centered systems of care for children with special healthcare needs.
2. Stimulate professional preparation programs to provide new skills needed by providers in changing delivery systems.
3. Support the development of models that resolve cost and utilization issues for children with special healthcare needs.
4- Provide leadership in the establishment of quality assurance programs for children with special healthcare needs in managed care settings.
5. Identify and support the implementation of models of family participation in managed care settings.
6. Develop strategies for improving data and evaluation activities to monitor the success of the National Agenda for Children with Special Health Care Needs.
7. Integrate managed care programs within the community system of services.13'14
When other healthcare groups, whether family practitioners, general practitioners, nurse practitioners, physician assistants, chiropractors, naturopaths, or others, suggest that they are able to provide care for children, pediatricians must emphatically state that not only are pediatricians best able to orchestrate the care of well children, but they are also able to orchestrate that care for chronically ill children. If pediatricians are able to embrace both populations, our future is bright. If we are unable to do so, our reason for existence will be questioned.
1. Newacheck P, McManusM. Financing health care fat disabled children. Pediatria. 1988;81:385-394.
2. Newacheck PW. Taylor WR. ChikDuod chronic illness: prevalence, sevtiir?. ami impaci. Am J Public Health 1992;82J64-371.
3. Palfry JS .Haynic M. Porter S. et al. Piirvalcnceo of medkal technotogy assistance among children in Mauachmero, 198? and 1990. Pneiic HealA Rtp. 1994:109:226-233.
4. LiptakGS, Revel GM. Community physician's role in case management of children with chronic illnesses. Pediatrics. I989;84:465-470.
5. American Medical Association. Physicians' CKTTCTU Procedural TeraaruJagy Manual. Chicago. Ill: AMA; 1997.
6. Neff JM, Anderson G. Protecting children with chronic illness in a competitive marketplace. JAMA. 1995:274:1866-1869.
7. Fowler E], Anderson GF. Capitation adjustment for pediatrie populations. Pediatrics. 1996;98:10-17.
8. Anderson GF. Stcinberg EP. Powe NR. et al. Setting payment rates for capitated systems; a comparison of various alternatives. Inquiry. 1990:17:225-233.
9. National Association of Children's Hospitals and Related Institutions (NACHRI). Pediatrie Excellence in HîoJiA Car« Sjstími. Alexandria. Va: NACHRI; 1996.
10. Andrews JS, Anderson GF1 Hon C, NeITJM. Pediatrie carve ours. The use of dû· ease specific conditions as risk adjust«* in capitated payment systems. Arch Pedían Adolesc Med. 1997; 1 5 1:236-292.
11. Brewer EJ. McWwrson M, Magrab PR, Hutchins VL. Family-cenrercd. communitybased, coordinated caie for children with special health care needs. Pediatrici. 1989;83:1055- 1060.
12. Sacchetta A, Oerardi M, Barkin R, et al. Emergency data ser fot children with special needs. Ann Emerg Med. 1996:28:324-327.
13. US Department of Health and Human Services, Public Health Service. Heaiihy People 2000; National Heolift Promorion and Disease Prtvenaon Objectives Washington, DC: US Government Printing Office; 1990.
14. National Agenda for Children with Special Health Cate Needs. Summaries of Expert Wort Group Meetings. Rockville, Md: Maternal and Child Health Bureau; 1996.
Relative Value Units (RVUs)1T
Case Coordination for Children With Special Healthcare Needs
Description of Supportive Tools and Resources for Chronic Condition Management (CCM)
List of Definitions