 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
* Department of Pediatrics, Boston University School of Medicine/Boston Medical Center, Boston, Massachusetts
Department of Pediatrics, School of Medicine
Department of Health Services, School of Public Health
|| Department of Community Health Sciences, School of Public Health, University of California, Los Angeles, California
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Objective. To describe the proportion of high-quality basic pediatric preventive services, exclusive of immunizations, reported by parents and to identify sociodemographic and health system predictors and health service correlates of provision of these services.
Study Design. The study used cross-sectional data on 2041 children, 4 to 35 months of age, in the 2000 National Survey of Early Childhood Health.
Outcome Measures. The BPS measure assesses the receipt of (1) developmental assessment, (2) injury prevention counseling, (3) screening for parental smoking, (4) guidance on reading to the child, and (5) guidance on 14 other topics (assessed as a composite score). The BPS scale categorizes the receipt of services as excellent, good, fair, or poor.
Results. Most children received excellent (34.9%) or good (31.5%) care, but many received fair (24.9%) or poor (8.7%) care. Sociodemographic and health care factors such as race/ethnicity, insurance, and practice setting were not associated with BPS levels. Higher BPS scores were associated with parental reports of longer well-child visits, more counseling regarding family and community risk factors, lower rates of delayed or missed care, and greater satisfaction.
Conclusions. Two thirds of children receive good or excellent basic preventive care, as determined with this composite, and no disparities according to race/ethnicity, income, or health insurance status of families (which are often found to be associated with health care access) were found. This equitable distribution of high-quality care suggests a high level of clinician professionalism. Duration of visits may be a key factor to improve quality of care. Because of its association with other services, processes, and outcomes of care, the BPS scale may serve as a useful construct for monitoring quality and stimulating efforts to improve national pediatric preventive care.
Key Words: quality of care • pediatric primary care • prevention • disparities
Abbreviations: NSECH, National Survey of Early Childhood Health • BPS, basic preventive services • PHDS, Promoting Healthy Development Survey • AAP, American Academy of Pediatrics
Pediatric primary care is designed to prevent disease, disabilities, and injuries and to promote children's health and well-being.1 Recommendations for the periodicity and content of pediatric well-child visits have been developed by the American Academy of Pediatrics (AAP) and Bright Futures and have increased considerably in scope in the past decade.2,3 National estimates of the quality of pediatric preventive care come largely from the Health Plan Employer Data and Information System, which includes basic indicators such as the receipt of immunizations and the number and timing of well-child visits.4–10 These measures do not assess the full range of recommended preventive services, however.
Other studies assessed a broader array of pediatric preventive services and revealed deficits in some recommended preventive and health promotion services.11–15 A particular challenge to pediatric providers in delivering all recommended preventive care is the large number of services to provide and topics to address. A recent study indicated that it is not feasible, because of the amount of time required, to deliver all preventive services recommended by the US Preventive Services Task Force.16 The implications for practice are particularly compelling when it is considered that US Preventive Services Task Force recommendations for children are actually less comprehensive than those of the AAP or Bright Futures. Although Bright Futures recommends the involvement of other community-based providers in the delivery of preventive care, the expectations for pediatric providers are still considerable. Given these time pressures, there is concern about the extent to which children receive basic preventive services (BPS) beyond immunizations and adherence to a minimal well-child visit schedule.
More comprehensive data on preventive care for young children have become available. The Promoting Healthy Development Survey (PHDS) and PHDS-Plus were developed by the Child and Adolescent Health Measurement Initiative, on the basis of reviews of evidence for each health service item, for assessment of developmental services provided to young children.17 Survey items were selected for inclusion on the basis of the prevalence of the health concern underlying the care and the available evidence on the effectiveness of care for that concern provided in pediatric settings. Until recently, the PHDS had been fielded only among selected health plans and several state Medicaid programs. The 2000 National Survey of Early Childhood Health (NSECH) builds on the PHDS studies by incorporating many of the PHDS measures into a national sample, to provide nationally representative estimates of preventive care.
To date, few studies have evaluated the variability of parental self-reports of preventive services received by young children and their parents. This study used the NSECH to describe the prevalence and correlates of adequacy of BPS provided by clinicians, exclusive of immunizations, for young children. A special focus was on assessing whether the quality of parent-reported provision of preventive care varied according to maternal race, household income, practice setting, or health insurance status or type.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Measures
Measures of services were adapted for NSECH from the PHDS and were originally developed from topics in AAP and Bright Futures guidelines with demonstrated effectiveness. Items included receipt of (1) a developmental assessment,20 (2) injury prevention counseling,21–24 (3) screening for parental smoking,25–30 (4) guidance on reading to children,31–34 and (5) guidance on 14 other topics (expressed as a summary score).
Parents were asked whether the child's doctors or other health care providers ever told them they were carrying out what doctors call a developmental assessment of the child. They were also asked whether the child's doctors or other health care providers ever had him or her pick up small objects, stack blocks, throw a ball, or recognize different colors. An answer of yes to either of these questions was counted as indicating that a developmental assessment had been performed.
For injury prevention, reading, and smoking items, parents were asked whether, in the past 12 months (or since the child's birth, for children <12 months of age), the child's doctors or health care providers had talked about the injury items, the importance of reading to the child, and whether the parent or someone in the house smokes, respectively. Responses were yes or no.
For guidance on the 14 other topics, parents were asked whether, in the past 12 months (or since the child's birth, for children <12 months of age), the child's doctors or health care providers had talked about each topic. If the parents responded no, then they were asked whether a discussion of that topic would have been helpful. Respondents who did not receive the service and said that it would have been helpful were counted as having a missed opportunity.
This study examined the prevalence of BPS and the sociodemographic and health care factors (independent variables) associated with BPS (dependent variable). The study also assessed the relationship of BPS (independent variable) to other well-child care processes and care experiences (dependent variables). Descriptions of the measures are as follows.
BPS Measure
A BPS scale was derived from previously used, validated, reliable measures11,17 and included selected parental reports of services received in the past year (or since birth, for children <1 year of age). Figure 1 summarizes the items that constitute the BPS measure and the scoring procedures for the scale. Scoring for items 1 through 4 was dichotomous (0 points = no, 1 point = yes). The composite measure of 14 guidance topics (item 5) was designed differently, for assessment of the receipt of health supervision in relation to parental needs and preferences. This composite was a count of missed opportunities, which we defined as guidance topics that were not addressed but, according to parental report, would have been helpful. Because missed opportunities were common (
45% of parents reported 
1), a 3-point scoring system was created for this composite, with no missed opportunities being considered the best result (2 points = no missed opportunities, 1 point = 1 missed opportunity, 0 points = 2 missed opportunities). The greater weighting of the missed opportunities component, relative to the other 4 components of the BPS, reflects the larger number of services assessed in this composite item.
|
3 on the BPS scale represented inadequate or "spare" care. This BPS scale has elements of interval-level measurement but, because of our weighting scheme (eg, incorporating a composite measure of 14 topics), the scale is most appropriately used ordinally.
Sociodemographic and Health Care Factors
This study examined the association of family sociodemographic features and health care factors with BPS. Family demographic variables included child age (4–9 months, 10–18 months, or 19–35 months), maternal race/ethnicity and language (non-Hispanic white, black, Hispanic with interview in English, or Hispanic with interview in Spanish), and maternal age (19 years, 20–29 years, 30–39 years, or 40 years). Family socioeconomic status factors included maternal education (less than high school education, high school graduate, or college education or more) and annual family income (<$17 500, $17 501–$35 000, $35 001–$60 000, or $60 000). Health care factors included child health insurance coverage (private, public such as Medicaid or state children's health insurance program, other coverage, or uninsured) and provider setting (private office, health center or public clinic, or hospital-based clinic).
Process of Well-Child Care Measures and Ratings of Health Care Experiences
This study also examined the association of BPS scale results with other health care factors and experiences of care. In particular, the study assessed parental reports of (1) duration of the last well-child medical visit (in minutes), (2) receipt of family-centered care, (3) counseling regarding family and community risks to child health and development, (4) satisfaction, (5) ability to ask all questions, (6) appropriateness of time spent during the visit, and (7) receipt of needed care.
To assess counseling regarding family and community risk factors, parents reported whether, in the past 12 months, the provider had asked them about the following 6 items: use of alcohol/drugs in the household, community violence, parental health, emotional support, support of a spouse/partner in parenting efforts, and difficulty paying for basic needs. The 6 dichotomous responses were combined and transformed to a 100-point scale to reflect the proportion of the 6 items that were asked of parents.
Parents reported overall satisfaction with the last well-child visit on a 0–10 scale, with 10 being the highest possible rating. This rating was converted to a 100-point scale for similarity of presentation to other measures in the study. Parents also reported whether they were able to ask all of their questions, whether the provider spent enough time with them, and whether there was any time when their child needed care for a problem and did not receive it or the child received care for a problem but received the care later than the parents would have liked. Family-centered care was assessed with 4 questions. Parents were asked how often in the past 12 months the provider (1) took time to understand the specific needs of the child, (2) respected the parent as the expert on the child, (3) asked how the parent was feeling as a parent or guardian, and (4) understood the parent and family and how they preferred to raise the child. A 100-point scale was created from the 4 items on the basis of a point system for the 4 response options (always = 100, usually = 67, sometimes = 33, never = 0); higher scores reflect more family-centered care. This measure was used previously in the PHDS.17
Analyses
This study presents the prevalence, predictors, and correlates of BPS. The survey design involved stratified sampling; therefore, adjustments were made to SEs with survey procedures in Stata version 7.0 (Stata Corp, College Station, TX). For all preventive services, the small number of responses of "don't know" (generally 2–5 responses per service) were coded as "did not receive the service," because we assumed that not knowing whether a service was received likely indicated that the service was not received or was not effectively conveyed. Sensitivity analysis showed no impact on the results of treating "don't know" responses as missing data. Individuals with missing information on BPS were excluded from the analyses, which reduced the analytic sample size to 2041. Some of the analyses used smaller sample sizes because of missing information on the predictors and correlates.
The Pearson 
2 test was used to test the significance of differences in BPS categories according to sociodemographic and health care factors. The relationship of BPS scores to the processes and experiences with care measures was tested with either the Pearson 2 test or pairwise t test. These comparisons were also conducted with percentages and means that were adjusted for several potential confounding factors (such as race/ethnicity, insurance coverage, maternal education, and language).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
|
|
To ensure that we did not overlook important differences because of the relatively small number of children who received poor care, we combined children with poor or fair care into a single category and those with excellent or good care into another category and then we reanalyzed the data. This combined variable did not alter any of the relationships presented in Table 3. We also assessed the impact of assigning the missed opportunities component a slightly greater weight in the scale, because of the large number of items included in this domain and to distinguish among levels of care in this domain (because 75% of parents had 1 missed opportunity). Equalizing the weighting between missed opportunities and the other items in the scale resulted in an overall distribution shift toward poorer care, but no changes in the relationships reported in this study were noted (data not shown). Finally, for Table 4, we repeated the analyses with means and percentages that were adjusted for confounding factors (race/ethnicity, insurance coverage, maternal education, and interview language), and none of the relationships changed significantly (data not shown).
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The results of the study must be understood in the context of certain limitations. First, the purpose of the BPS scale was to evaluate a clinically relevant composite measure of core preventive services; therefore, the measure included a limited number of components, which cannot fully represent the broad range of recommended developmental and counseling services that constitute ideal pediatric care. Similarly, the injury prevention measure has 2 components for the 4- to 9-month and 19- to 35-month age groups but only 1 component for the 10- to 18-month age group. Whether a combination of other measures of related importance would yield similar results is not known. Studies that used the same dataset with different measures of preventive care showed selective disparities associated with socioeconomic status and structural factors. For example, non-Hispanic white children received more counseling than did nonwhite children,11 parents of nonwhite children reported higher rates of missed opportunities,37 and discussion of some family and community risk factors was greater among nonwhite parents.38 With the combination of multiple components of preventive care in a single composite measure, this variation largely disappears with the BPS scale.
The usefulness and validity of the BPS scale as a single composite measure is supported by the consistency and magnitude of the relationship of the BPS scale results with other aspects of care, particularly process measures such as visit length. This suggests that the services studied are good indicators of the quality of preventive care, given that all BPS items are recommended for parents irrespective of child and family factors. Also, when children do not receive BPS, the services are not simply replaced by other preventive services.
Second, the NSECH measured preventive care only in the first 3 years of life. Whether the prevalence and correlates of receiving BPS are similar for older children is not known. Third, all BPS measures were based on parental reports, without validity checks of observable behavior, and might not reflect accurately the preventive services actually delivered or whether problems were prevented effectively. It is possible that parental responses to some questions were biased by the parents liking their child's clinician, in which case they might have been more likely to report receiving certain services. It can be argued that, if parents do not recall a discussion, then the guidance provided was not conveyed effectively. We do not know whether services were actually provided by the clinician or whether they were provided by nonphysician support staff members or through written materials. Also, missed opportunities may reflect both clinician practices and parental preferences and reporting, and this factor is not a measure of clinician behavior alone. Fourth, although the items on the BPS scale are considered best practice and are based on AAP guidelines, the evidence for the effectiveness of these items remains sparse, and we cannot be sure that providing these services prevented or changed outcomes. It is also important to note that BPS items and components were not assigned equal weights.
This study provides an impetus to improve and enhance the quality of preventive pediatric care. Although 66% of young children received good or excellent preventive care, as indicated by the BPS measure, 34% of young children received poor or fair care. On the basis of parental reports, we know that these children were also more likely to miss or delay needed medical care. It is not known whether poor or fair care has actual effects on health, specifically related to smoke exposure, injury prevention, identification of developmental disabilities and referral for early intervention services, or other problems that could have been successfully addressed by the pediatrician eliciting parents' concerns or needs. Although discussion of a topic does not necessarily indicate effective delivery of guidance or actual prevention or alteration of outcomes, the epidemiologic features of the problems and risks discussed suggest nontrivial potential effects on young children's health and well-being. Fair or poor care appears to be a general problem, affecting nearly one third of a cross-section of all young children in the United States. Identification of process of care factors associated with less than adequate preventive care has important implications. This analysis suggests that visit length is a potential area to address to improve the quality of preventive pediatric care for all children. Parents whose children receive BPS also report that the provider takes time more frequently to understand the child's unique needs. We do not know the predictors of visit length or whether visit length was under the control of the clinicians or imbedded in the logistics or system of care in which the clinicians were practicing. Inadequacy of primary care reimbursement and the scope of current pediatric preventive care recommendations have been identified as barriers to providers covering all recommended counseling topics. Recent approval by the Center of Medicare and Medicaid Services of a relative value for Current Procedural Terminology code 96110 (developmental screening, limited) of $13.81 should help improve delivery of this preventive service in primary care. There are also some efficiency measures that pediatric providers can implement to improve coverage of other preventive developmental services. Questionnaires, written materials, and videotapes could help providers achieve more in the limited time they have to counsel parents of young children.37–40
Other innovative initiatives are underway to improve the quality of preventive care for young children, including efforts to enhance staffing in pediatric practices. For example, Healthy Steps is an effort that incorporates child development specialists with dedicated time and specific expertise into pediatric practices, to provide preventive and developmental services and to address parental concerns and needs more effectively. This effort significantly improves the quality of preventive pediatric care for families. Parents who received care in Healthy Steps offices were 20 times more likely than other parents to receive 4 developmental services during a visit. More importantly, Healthy Steps promotes more positive parenting, including less severe discipline, more timely receipt of immunizations, and increased parental sensitivity to their children (eg, less likely to have children sleep on their stomachs).41,42 Because these programs require additional resources, other strategies17 may be needed to enhance efficiency and target parental needs while ensuring that BPS items are consistently addressed.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
We thank Edward Schor, MD, for his helpful comments on an earlier draft.
|
FOOTNOTES |
|---|
Reprint requests to (B.Z.) Boston Medical Center, 771 Albany St, Dowling 3509 South, Boston, MA 02118. E-mail: barry.zuckerman{at}bmc.org
No conflict of interest declared.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  I. Sandler and T. F. Boat Implications of Parental Bereavement and Other Family Adversities for Preventive and Health Promotion Pediatric Services Arch Pediatr Adolesc Med, May 1, 2008; 162(5): 487 - 488. [Full Text] [PDF]
|
|
|
|
 |
|
 R. Mangione-Smith, A. H. DeCristofaro, C. M. Setodji, J. Keesey, D. J. Klein, J. L. Adams, M. A. Schuster, and E. A. McGlynn The Quality of Ambulatory Care Delivered to Children in the United States N. Engl. J. Med., October 11, 2007; 357(15): 1515 - 1523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Houtrow, S. E. Kim, A. Y. Chen, and P. W. Newacheck Preventive Health Care for Children With and Without Special Health Care Needs Pediatrics, April 1, 2007; 119(4): e821 - e828. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. L. Schor, M. Abrams, and K. Shea Medicaid: Health Promotion And Disease Prevention For School Readiness Health Aff., March 1, 2007; 26(2): 420 - 429. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. J. Hambidge, C. B. Emsermann, S. Federico, and J. F. Steiner Disparities in Pediatric Preventive Care in the United States, 1993-2002 Arch Pediatr Adolesc Med, January 1, 2007; 161(1): 30 - 36. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Selden Compliance With Well-Child Visit Recommendations: Evidence From the Medical Expenditure Panel Survey, 2000-2002 Pediatrics, December 1, 2006; 118(6): e1766 - e1778. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Serwint, K. A. Thoma, S. M. Dabrow, L. E. Hunt, M. S. Barratt, T. R. Shope, P. M. Darden, and for the CORNET Investigators Comparing Patients Seen in Pediatric Resident Continuity Clinics and National Ambulatory Medical Care Survey Practices: A Study From the Continuity Research Network Pediatrics, September 1, 2006; 118(3): e849 - e858. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Needlman Thoughts on health supervision: learning-focused primary care. Pediatrics, June 1, 2006; 117(6): e1233 - e1236. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
