PEDIATRICS Vol. 108 No. 4 October 2001, pp. 913-922

Estimating the Cost of Developmental and Behavioral Screening of Preschool Children in General Pediatric Practice

Deborah Dobrez, PhD^*, Anthony Lo Sasso, PhD^*, Jane Holl, MD, MPH^*, , Madeleine Shalowitz, MD, MBA^{*, §}, Scott Leon, BA, and Peter Budetti, MD, JD^*

From the ^* Institute for Health Services Research and Policy Studies, Northwestern University, Evanston;  Children's Memorial Hospital, Chicago; ^§ Evanston Northwestern Healthcare, Evanston; and  Department of Psychology, Northwestern University, Chicago, Illinois.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  Despite increased recognition of the importance of development and growth of young children, formal developmental and behavioral screening often is not included in general pediatric practice. Barriers to the provision of developmental and behavioral screening are considerable; among them are the need for specialized training and uncertain reimbursement. This article develops a model for estimating the cost of providing pediatric developmental and behavioral screening that can be scaled to reflect a pediatric practice's patient population and choice of screening offered.

Methods.  The framework for our scaleable cost model was drawn from work done in estimating the Resource-Based Relative Value Scale (RBRVS). RBRVS provides estimates of the work effort involved in the provision of health care services for individual Current Procedural Terminology codes. The American Academy of Pediatrics has assigned descriptions of pediatric services, including developmental and behavioral screening, to the Current Procedural Terminology codes originally created for adult health care services. The cost of conducting a screen was calculated as a function of the time and staff required and was loaded for practice costs using the RBRVS valuation. The cost of the follow-up consultation was calculated as a function of the time and staff required and the number of relative value units assigned in the RBRVS scale.

Results.  The practice cost of providing developmental and behavioral screening is driven primarily by the time and staff required to conduct and evaluate the screens. Administration costs are lowest for parent-administered developmental screens ($0 if no assistance is required) and highest ($67) for lengthy, pediatric provider-administered screens, such as the Neonatal Behavioral Assessment Scale. The costs of 3 different groups of developmental and behavioral screening are estimated. The estimated per-member per-month cost per 0- to 3-year-old child ranges from $4 to >$7 in our 3 examples.

Conclusions.  Cost remains a significant barrier to greater provision of formal developmental and behavioral screening. Our scaleable cost model may be adjusted for a given practice to account for the overall level of developmental risk. The model also provides an estimate of the time and cost of providing new screening services. This model allows pediatric practices to select the mix of developmental screens most appropriate for their particular patient population at an acceptable cost.  Key words:  developmental screening, RBRVS, CPT codes, cost.

Pediatric primary care providers are well positioned to monitor and foster early childhood development, in part because of regular contact over time with infants and toddlers and their families.¹ Current guidelines for health supervision encourage primary care clinicians to embrace this role, delineating age-specific content for anticipatory guidance, developmental and behavioral surveillance, injury prevention, assessment of the primary caregiver's mental health, family psychosocial support, and advocacy in addition to providing routine medical care.^2,3 The inclusion of these activities in a typical 20-minute office visit has become a major challenge. Other areas of concern include the optimal delivery model,⁴ the cost of additional services within a busy practice, and the opportunity for reimbursement. This article describes a scaleable economic model for calculating the cost of adding specific screening services that are focused on early childhood development to pediatric primary care practice. Our approach is based on the use of the Resource-Based Relative Value Scale (RBRVS), the metric underpinning the calculation of physician reimbursement for health care services under Medicare that was adapted recently for use in pediatrics by the RBRVS Project Advisory Committee of the American Academy of Pediatrics.⁵

Currently, pediatricians are most likely to gather information about a child's development and behavior in the course of routine, informal surveillance, through discussion of parental concerns and observation of children over time. A survey of 41 randomly selected board-certified pediatricians showed that approximately 20% of physicians use formal developmental screening instruments in their office practices, with most relying on history and physical examination to screen for developmental problems.⁶ The use of formal screening in office practices is hampered by the lack of consensus on which instruments are the most suitable for screening general populations, easily integrated into pediatric practice, and cost-effective. Table 1 is a compilation of a wide range of developmental and behavioral screening tools and instruments for children who are 0 to 3 years of age. For each tool or instrument, we provide information about the target age group and administration time and a brief description.

Few published studies have focused on estimating the economic burden to providers of providing developmental and behavioral screening services. Previous work used cost estimation approaches that are applicable for activities related to service provision that are directly observable. These approaches include the use of retrospective collection of charges⁷ and activity-based costing⁸ and can provide detailed information regarding the actual costs observed in a single practice. However, several assumptions underlie such estimates. First, they assume that the staff has excess capacity; that is, the staff has available otherwise unused time to provide the developmental service. If, however, the staff is already functioning at full capacity when a new service is introduced, then cost considerations should include the tradeoffs in cost and reimbursement associated with the services the staff no longer can provide. Other assumptions include the current training level of staff, current staff activities, and the development and health of the children in the practice.

Glascoe et al⁹ developed a flexible method for estimating the cost of developmental services. The authors described 4 steps of service provision: tool administration, result interpretation, diagnostic evaluation for positive screens, and treatment. Using a sample of 247 children and average salary and overhead costs, the authors estimated the cost of service provision. Although short-term costs varied significantly among the 4 different screening approaches, the long-term costs (based on assumptions of average expected treatment costs and average expected future cost savings to treatment of detected developmental delays) were similar among the approaches. The benefit of this method is that it includes the long-term cost implications of providing developmental services. However, its reliance on data from a specific sample limits its generalizability to clinical practices that might differ importantly in the characteristics of their patient populations, the proportion of abnormal screens, and the time required for discussion and follow-up.

Our approach to estimating the cost of providing developmental and behavioral screening advances previous work because we base our cost estimates on a widely accepted and applied metric that provides the relative value (based on work effort and expertise required) of the provision of individual services, the RBRVS. The RBRVS is a metric developed by the Health Care Financing Administration for physician payment. Although the scale originally was designed for Medicare reimbursement purposes, private payers frequently use it to determine reimbursement for services to the nonaged population. Recently, the American Academy of Pediatrics worked with the American Medical Association's Current Procedural Terminology (CPT) consensus panel to map the adult-appropriate CPT codes to pediatric services and with the American Medical Association/Specialty Society Relative Value Scale Update Committee to provide to the Health Care Financing Administration relative value units (RVUs) for the pediatric-specific codes.^5,10 In addition, our model further advances previous work in that it is scaleableit can be used to calculate the additional cost expected with the provision of a developmental screening given individual pediatric practice characteristics, such as the combination of developmental screens to be administered and expected levels of development problems among the practice patients.

To build our model, we estimate the costs of conducting individual screens, using the RBRVS. To illustrate our method, we construct 3 examples of combinations of developmental and behavioral screening. Using these 3 examples, we demonstrate how to aggregate individual service costs to estimate the total costs of providing developmental and behavioral screening. We then determine the primary cost drivers, ie, the major determinants of cost, associated with providing developmental and behavioral screening. Finally, we demonstrate the scalability of the model by altering key parameters of the clinic practice setting and patient population.

	METHODS

Top Abstract Methods Results Discussion References

Our scaleable cost model was developed and applied strictly from the provider's perspective. Because we were interested in calculating the economic costs of providing developmental services, we sought to include all costs incurred by the physician practice, including the value of the provider's time, which depends partially on the provider's level of training, practice overhead, and other practice costs. To estimate these practice costs, we used a multistage process, first defining and costing individual services and then aggregating these services to calculate the cost of providing developmental services periodically over the course of the first 3 years of life.

Developmental Service Cost Calculation

We allocated the time and effort associated with service provision for both administration of the screening tool or service and consultation with the parent regarding the results of the screen. We used general estimates of resource cost and tool sensitivity for cost estimation of individual services and combinations of services. We constructed estimates of the cost of providing developmental services using published estimates, input from a panel of child development experts and generalists, and assumptions provided by 2 of the co-authors (J.H., M.S.). Assumptions were necessary to develop credible estimates of the time involved for the administration of a screen, to determine the appropriate staff member to administer the tool (parent, nurse, or MD), and the rate of abnormal results. Because some parents will need assistance in administering screens, because of literacy, language, or comprehension barriers, we allowed for 10% of the parent-administered screens to be administered by a nurse in the office setting. Assignment of dollar values to tool administration and consultation time was achieved using the RBRVS physician fee schedule (modified when appropriate for other nonphysician health care providers). The RBRVS identifies relative work effort for various services defined by CPT codes, using RVUs. The RVUs assigned to each CPT code reflect the work effort generally required by a physician to administer the service to a typical patient, including overhead and other practice costs, such as malpractice insurance. A conversion factor (CF), which translates an RVU into a dollar amount, is established annually for Medicare by congressional mandate. This method of valuing provider time has been used in previous studies of the cost associated with service provision.¹¹

We identified a sample of developmental and behavioral screens for the purpose of illustrating our cost model. We assumed a 5-minute consultation time for normal findings on screens. For abnormal results, a 7- to 15-minute consultation was assumed, depending on the expected complexity of the consultation. Because the analyses were conducted from the provider's perspective, the value of the time and effort required of the parents was not included, although the parents' burden is otherwise an important consideration.

To estimate the cost of both the administration and the consultation associated with each screen, an hourly earnings rate (which includes practice and other overhead costs) was calculated for providers. To do this, we identified the evaluation and management 60-minute individual counseling CPT code, 99404, from the 1999 Coding for Pediatrics Manual. (The individual consultation CPT code was used for estimating the cost of developmental service provision only, and does not constitute any recommendation regarding billing for these services.) This code is intended for use for a wide variety of services, including discussion of laboratory and diagnostic test results available at the time of the encounter. Using the RVUs and CF reported for 1999, the total cost associated with CPT code 99404 is $133.36 (3.84 RVU × $34.73 CF). Thus, $133.36 represents the "fully loaded" average hourly earnings rate for an MD, incorporating provider training, overhead, and other practice costs (eg, malpractice insurance). Overhead includes the cost of photocopying of questionnaires. We ignored the initial purchase cost of screening instruments, which generally is <$200, because when factored over the large number of children in a clinical practice, the cost per child is negligible. The median hourly wage for registered nurses for 1999, $17.43,¹² was loaded for overhead and other practice costs by dividing the wage by the percentage of the RVU that is assumed to be provider wage (44.2%; the remainder is the overhead and other costs). The loaded registered nurse hourly wage therefore was equal to $39.43. Administration and consultation times were multiplied by the appropriate loaded wage to estimate the costs of service provision.

Example of a Combination of Screening Services Cost Calculation

We first identified a selection of individual developmental and behavioral screens, described in Table 1 (descriptions of many of the screens described in Table 1 were drawn from Glascoe¹³). We next combined these services into 3 illustrative combinations of services. The 3 examples are described in Table 2. Administration time (and cost) for each of the screens is independent of whether additional screens are administered at the same visit and therefore can be added. However, total consultation time for several tools administered at the same visit is not likely to be additive. For example, if a child scored in the normal range on 1 screen, then we would assume a 5-minute consultation to communicate those findings. It is not likely that a 15-minute consultation would be used to communicate the findings of 3 normal screens in the same visit. We therefore assume a 5-minute consultation for all normal findings per visit, regardless of the number of screens administered. Abnormal findings on any given screen requires additional time for explanation and future assessment or treatment planning (7-15 minutes). In the case of abnormal results, the issues and implications for future intervention will differ for each abnormal result. Therefore, consultation time for abnormal findings was treated as additive.

The rate of detection of developmental delay or behavioral problems for many screening tools has not been well established. We also recognize that variability in positive detection rates across instruments and variability across population and setting exists. The most serious outcomea disabilityis estimated to exist for 6.5%¹⁴ to 17%¹⁵ of US children. Among a group of preschool children seen in primary care practices, pediatricians estimated that 8.5% had a behavior problem and psychologists estimated the prevalence of behavior problems to be 13%.¹⁶ Thus, in presenting this cost model, we for illustrative purposes assumed an abnormal (positive) detection rate of 10% for each screening tool. In the application of this model, a different percentage can be inserted for each screen to modify the model to fit a particular practice. Also, we did not consider the cost of treatments that result from abnormal findings other than consultation costs. Thus, for example, if an assessment of maternal depression leads to treatment for depression, then we include only the costs associated with performing the assessment and communicating the results of the assessment and not costs for treatment. Similarly, our estimates do not factor in any potential cost offsets that may result from the provision of pediatric developmental screening, because insufficient evidence exists on which to base any estimate of the magnitude of the cost offset.

Furthermore, we calculated the expected (average) cost per child (up to 3 years of age) of providing the 3 example combinations of developmental services. In addition, we calculated the expected steady-state cost per 0- to 3-year-old child per year and per month. The steady-state cost identifies the cost per 0- to 3-year-old child when each child is enrolled in the program. For illustrative purposes, this cost was calculated by assuming that a practice treated equal numbers of children through ages 1, 2, and 3 (increased visit frequency for the younger children is explicitly included in this calculation). The steady-state cost calculations provide a basis for estimating the required capitation increment for the provision of developmental services for children aged 0 to 3 years. The steady-state cost also can be based on the age distribution of children in the practice.

Finally, we chose several parameters of the model that are likely to differ across clinical practices. These parametersrate of detection of abnormal findings, time cost, and consultation timewere varied to demonstrate how the model can be scaled to reflect actual patient and provider characteristics, in this case, in an environment of greater service need and provision.

We increased the rate of abnormal findings to 15%, increased consultation time for an abnormal finding to 25 minutes, and increased the total provider cost by 20% to illustrate expected program costs of administering our second example combination of services in an environment of greater developmental needs and more intense service provision. To increase the rate of abnormal findings, we calculated the expected consultation cost for each service with a revised weighting scheme: 0.85 × normal consultation cost + 0.15 × abnormal consultation cost. Increasing the consultation time for an abnormal scheme requires revision of the individual screen cost. Increasing the total provider cost can be conducted by increasing the expected wage, increasing the loading factor to adjust for greater overhead costs, or increasing the overall total costs to adjust for greater overall costs. In our example, we simply multiply the total cost by 20%.

	RESULTS

Top Abstract Methods Results Discussion References

A summary of the estimated costs for each individual developmental service is shown in Table 3. Clearly, parent administration of the screen substantially reduces the total cost of the tool. After accounting for a 10% needed assistance rate in administration, the screens administered by the parent with the shortest expected administration times are the least costly, including the Parents' Evaluations of Developmental Status, the Carey Temperament Scale, and the Pediatric Symptom Checklist. The next least costly tool to administer, administered by a health care provider, was the Bayley Infant Neurodevelopmental Screen ($11.11); administration costs for the remaining tools ranged as high as $66.68 for the Neonatal Behavioral Assessment Scale. As noted previously, we assumed a relatively conservative rate of abnormal outcomes from the screening instruments of 10%; thus, the cost associated with subsequent consultation is variable.

Table 4 displays estimated costs associated with the first example of a combination of developmental screening tools. This example includes 2 screening tools: parent administration of the age-appropriate Ages and Stages Questionnaires, for a recurring expected total cost of $12.86, and administration of the Family Psychosocial Screen annually, bringing the total cost per visit with both screens administered equal to $15.08. The total cost over the course of the first 3 years of life was estimated to be $167.20, $106.66 (64% of the cost) of which is incurred during the first year of life.

Because a managed care provider is likely to be more interested in the steady-state costs per child, we provided the annual incremental cost associated with a combination of developmental screening tools, assuming an equal distribution of children in the practice population at ages 1, 2, and 3 (with greater visit frequency expected for the younger children). In the first example, the per-child per month (PCPM) increment was $4.64 ($167.20  36), spread only over the population of children 0 to 3 years of age. Because we are implicitly assuming 100% compliance, which is unlikely to be realized in practice, the PCPM estimate is likely to represent an upper bound on actual service costs.

Table 5 shows the estimated costs associated with our second example, which features 3 screens administered on an occasional basis. This example includes annual administration of the Family Psychosocial Screen and the Bayley Infant Neurolodevelopmental Screen (administered in the first 2 years only) and biannual administration of the Early Language Milestone Scale. Because the Early Language Milestone Scale and Bayley Infant Neurolodevelopmental Screen are administered by a physician in this model, per-visit costs are higher. The resulting costs were estimated to be $82.54 for each of the first 2 years and $69.88 in the third year, totaling $234.96 per child for the first 3 years of life. At steady state, the expected annual incremental cost was $77.11, with PCPM costs of $6.53.

The third example (Table 6) builds on the second, adding the Behavioral Assessment of Baby's Emotional and Social Style to be administered at 12 and 18 months and the Center for Epidemiologic Studies Depression Scale, administered to the mothers at the 2-week visit. These additional services increase the total costs over the first 3 years of life to $275.02. At steady state, this suggests incremental annual per-child costs of $91.67, or $7.64 per month.

Finally, we increased the rate of abnormal findings to 15%, increased consultation time for an abnormal finding to 25 minutes, and increased the total provider cost by 20% to illustrate expected program costs of administering our hypothetical moderate-intensity combination of services in an environment of greater developmental needs and more intense service provision. Increasing the consultation time for abnormal findings resulted in a range of expected consultation time for each of the 3 screens of $11.11 to $55.57 ($133.36 × 0.42). Increasing the rate of abnormal findings results in an average expected consultation time of $17.78 for each individual screen ($11.11 × 0.85 + $55.57 × 0.15) and an average expected consultation time of $34.45 when all 3 screens are administered. Replacing these consultation times into Table 5, the total costs of service provision can be calculated. Total costs per 0- to 3-year-old child are $306.20, with first and second year costs equal to $108.55 and third year costs equal to $89.10. The PCPM costs are equal to $8.51. Total costs might be expected to be higher in an environment with wages and overhead costs that exceed those factored into the RBRVS scale. When increased by 20%, the total costs per 0- to 3-year-old child and PCPM costs are $367.44 and $10.21. In sum, the 3 changes together result in a 56% increase in the PCPM amount relative to our baseline assumptions.

	DISCUSSION

Top Abstract Methods Results Discussion References

Our scaleable cost model provides a framework for estimating the costs of providing pediatric developmental services that can be modified to fit the characteristics of a particular clinical practice and patient population. However, there are several limitations to this framework. The perspective chosen for these analyses is that of the provider: provider time and effort are included, whereas the time costs of the parent are not. Inclusion of parental time would increase the cost of some services. We excluded consideration of service provision beyond the initial consultation to evaluate and treat further the developmental delays detected by these screens. In addition, false-positive and -negative test results are likely to occur, with implications for increased cost in terms of unnecessary service provision or future service provision required as a result of the failure to treat developmental delays early (see Glascoe et al⁹ for further discussion). Conversely, 3 additional model assumptions potentially overstate the cost of service provision. First, any future cost savings resulting from improved development or more appropriate use of health services as a result of the service provision is ignored. Second, although evidence suggests that some pediatric practices offer formal developmental screening,⁶ the cost model assumes that the addition of the formal developmental screening does not substitute for any currently provided service. For the practices that do offer formal developmental screening, substitution effects may result in a lower estimated provision cost. Third, the model assumes 100% compliance by both the parent and the provider. Therefore, our model may represent an upper bound on the costs that providers will face in actual practice. Finally, because few published data are available regarding the rate of abnormal screen results in a general pediatric population, we assumed a 10% detection rate for screens. This assumption fails to capture the variation in the detection rate of the individual tools, which potentially could significantly impact the relative cost of the tools. Additional research is needed to establish detection rates for many of the above-mentioned tools.

The provision of developmental screening involves significant commitment from the provider in terms of time and effort. Estimated costs ranged from as low as $11 to as high as $82 per screen, not including any additional follow-up consultation and treatment resulting from detection of developmental delay. The effort and training required to conduct an assessment are primary determinants of cost; parental administration is substantially less expensive to the provider than administration in the office by a pediatric provider. In most models, these screening costs are incurred primarily in the first year of life. In the first example, with screening provided at each well-child visit, >60% of the total costs were incurred during the first year of life.

Calculation of the steady-state costs for various combinations of developmental screening allowed us to evaluate the costs for the provider when the patient population is evenly distributed between ages 0 and 3 years. Although total costs per child in even the least costly example are not negligible, provision of these screens in the steady state costs only $4 per 0- to 3-year-old child per month. If distributed over the entire population of children who are younger than 18 years in a practice, then the incremental costs would be considerably lower. When the potential for future cost savings through the prevention or early detection of delayed development is taken in to consideration, the actual societal economic cost may be substantially less. Of course, more intensive screening increases the steady-state cost. In our most costly example, the steady-state cost may be as high as $7.64 per 0- to 3-year-old child per month.

We demonstrated further the scalability of the model by estimating costs in different screening provision environments. Other model parameters, including rate of assistance required with parental administration of screens, compliance rates, individual who administers the screen, and expected administration time, also may be altered in a similar manner.

Our cost model is based primarily on developmental screening administered by a professional pediatric health care provider. However, there are many alternative approaches. Use of trained nonphysician providers to administer developmental screens can result in lower costs both by reducing the high cost use of physician time and in achieving economies of scale by focusing solely on the developmental screening. For example, the Healthy Steps for Young Children Program, funded by The Commonwealth Fund and local philanthropies across the country, uses a Healthy Steps Specialista dedicated developmental specialist within each pediatric practiceto conduct developmental screens and interventions.

Our scaleable cost model achieves 2 goals. First, we estimated the cost of individual developmental screening to aid physicians, practices, and health plans in their construction of a combination of developmental screening to offer infants and toddlers. Second, we developed and illustrated a flexible framework that will allow providers to estimate the total cost of providing their choice of screens to their pediatric population. This model should be of assistance to providers as they evaluate the extent to which they can integrate the provision of developmental screening in their practice.

Two key barriers remain to the provision of broad developmental screening in the general pediatric practice. There is little evidence regarding the actual long-term effectiveness of these screens in terms of their ability to improve development for the child and of their potential for long-term cost savings. Pediatric providers, already feeling pressured by short well-child care visits, are unlikely to integrate additional services without adequate reimbursement. Strategies to ensure such reimbursement for these services are not in place. Additional efforts are needed to estimate accurately the cost of individual developmental screening, and appropriate billing codes must be created for adequate reimbursement.

	ACKNOWLEDGMENTS

This research was supported by a grant from The Commonwealth Fund. The views presented here are those of the authors and not necessarily those of The Commonwealth Fund or its directors, officers, or staff.

We gratefully acknowledge the significant contribution of Barbara Langner, RN, PhD, in the development of this article.

	FOOTNOTES

Received for publication Nov 21, 2000; accepted Mar 19, 2001.

Reprint requests to (D.D.) Institute for Health Services Research and Policy Studies, Northwestern University Wieboldt Hall, 339 E Chicago Ave, Chicago, IL 60611. E-mail: d-dobrez{at}northwestern.edu.

	ABBREVIATIONS

RBRVS, Resource-Based Relative Value Scale; RVU, relative value units; CPT, Current Procedural Terminology; CF, conversion factor; PCPM, per child per month.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Dworkin P Detection of behavioral, developmental and psychosocial problems in pediatric primary care practice. Curr Opin Pediatr 1993; 5:531-536 [Medline]
2. American Academy of Pediatrics. Guidelines for Health Supervision III; 1997. Available: http://www.aap.org/publications/bookstore
3. Green M, ed. Bright Futures. Guidelines for Health Supervision of Infants, Children, and Adolescents. Arlington, VA: National Center for Education in Maternal and Child Health; 1997
4. Zuckerman B, Parker S Preventive pediatricsnew models of providing needed services. Pediatrics 1995; 95:758-762 [Abstract/Free Full Text]
5. American Academy of Pediatrics Issues in the application of the Resource-Based Relative Value Scale System to pediatrics: a subject review. Pediatrics 1998; 102:996-998 [Abstract/Free Full Text]
6. Dobos AJ, Dworkin P, Bernstein B Pediatricians' approaches to developmental problems: has the gap been narrowed? J Dev Behav Pediatr 1994; 15:34-38 [CrossRef][Medline]
7. Newacheck PW. The costs of caring for chronically ill children. Bus Health. 1987;18-24
8. Chan Y-CL Improving hospital cost accounting with activity-based costing. Health Care Manage Rev 1993; 18:71-77 [Medline]
9. Glascoe F, Foster E, Wolraich M An economic analysis of developmental detection methods. Pediatrics 1997; 99:830-837 [Abstract/Free Full Text]
10. AAP RBRVS Project Advisory Committee. Coding for Pediatrics. Elk Grove Village, IL: American Academy of Pediatrics; 1998
11. Berlin M, Faber B. Financial applications using the cost per RBRVS methodology. Med Group Manage J. 1996;28-34
12. Bureau of Labor Statistics. Occupational Outlook Handbook. Washington, DC: US Bureau of Labor Statistics; 1998-1999
13. Glascoe F Using parents' concern to detect and address developmental and behavioral problems. J Soc Pediatr Nurs 1999; 4:24-35 [Medline]
14. Glascoe F, Shapiro H. Developmental and Behavioral Screening. Available: http://www.dbpeds.org/articles/dbtesting/
15. Yeargin-Allsopp M, Murphy CC, Oakley GP, Sikes RK A multiple-source method for studying the prevalence of developmental disabilities in children: the Metropolitan Atlanta Developmental Disabilities Study. Pediatrics 1992; 89:624-630 [Abstract/Free Full Text]
16. Lavigne J, Binns H, Christoffel K, Behavioral and emotional problems among preschool children in pediatric primary care: prevalence and pediatricians recognition. Pediatric Practice Research Group. Pediatrics 1993; 91:649-655 [Abstract/Free Full Text]