OBJECTIVE. We assessed the impact of immunization at sick visits on subsequent and overall well-child care.
METHODS. We performed a retrospective cohort study using electronic health record data from 4 urban practices affiliated with an academic medical center. Participants included all children born between September 1, 2003, and July 31, 2004, with a visit at a study practice before 6 weeks of age and ≥1 sick visit (n = 1675). The main outcome measures were (1) attendance at a well-child visit within 60 days after an index sick visit by children due for vaccines and preventive care and (2) the overall number of well-child visits kept by children between 6 weeks and 13 months of age.
RESULTS. Among all demographic and health-related factors considered, immunization receipt at a sick visit was associated most strongly with decreased subsequent well-child care. Among children already delayed (late) for vaccines, 31% returned for well-child care if immunizations were given at eligible sick visits, compared with 47% of those who received no vaccines (risk difference, with adjustment for covariates: −16%). Among those without immunization delay, 42% of children who received vaccines returned for well-child care, compared with 73% of those who received no vaccines (risk difference: −31%). Although 5 well-child visits are recommended, children with no immunizations at sick visits had an adjusted predicted number of 3.8 well-child visits, those with 1 sick visit with immunizations had 3.3 visits, and those with ≥2 sick visits with immunizations had 2.8 visits between 6 weeks and 13 months of age.
CONCLUSIONS. Immunization at sick visits was associated with decreased rates of well-child care, especially among those without previous vaccine delay. This strong association between immunization at sick visits and well-child care should be considered in any plan to restructure pediatric preventive care.
Healthy People 2010 has designated timely immunization as 1 of the 10 leading health indicators for the nation.1 Despite this emphasis, up-to-date rates nationally and in many regions of the United States remain below targets.2 Effective vaccine delivery depends on clinicians capturing available opportunities for immunization in the office.3 Missed opportunities for vaccination, when children come to the practitioner's office but do not receive any or all due vaccines, contribute appreciably to incomplete immunization, especially among those with Medicaid insurance.4
Despite the prioritization of “timeliness” of medical care as one of the quality domains by the Institute of Medicine,5 the receipt of routine pediatric vaccines often is not timely, and failure to immunize children at sick visits contributes to this delay. National data indicate that >1 of 3 children are incompletely vaccinated for ≥6 months during the first 24 months of life and 25% of children have delays for ≥4 distinct vaccines between birth and 24 months.6 Missed opportunities often result from failure to administer all due vaccines simultaneously and rarely are the result of medical contraindications.7 Of particular note, missed opportunities most often occur at sick visits, despite the willingness of parents to accept immunization in this setting.8,9 As a result, strategies to increase immunization at sick visits are a key component of initiatives to improve immunization rates.
Although immunization at sick visits is an effective strategy to increase vaccination rates, many clinicians are concerned that this practice may decrease well-child visits, the primary setting for the delivery of developmental and preventive services for children. At least in part, this concern exists because the well-child visit schedule, which was standardized and endorsed by the Committee on Practice and Ambulatory Medicine of the American Academy of Pediatrics,10 was created around the vaccine schedule. In addition, some families perceive that receiving immunizations is equivalent to receiving well-child care.11 For children at high risk who are prone to missing visits, practitioners consider vaccine receipt an especially important incentive to bring families in for well-child care. Despite these fears, findings from an inner-city population suggested that nearly 75% of families were motivated to attend well-child visits for reasons other than vaccination.12
This study had 2 primary aims. First, for a group of children at urban practices who were due for vaccines and well-child care, we assessed the effect of immunization at sick visits on attendance at the subsequent well-child visit (visit-level analysis). We also examined the overall impact of immunization at these sick visits on the total number of well-child visits for each child between 6 weeks and 13 months of age (patient-level analysis). We hypothesized that children who received immunizations at sick visits would be less likely to attend the subsequent well-child visit and would have fewer well-child visits between 6 weeks and 13 months of age, compared with children who were not vaccinated. Second, we measured whether variability among clinicians providing sick care influenced the probability of children returning for subsequent well-child care. Understanding these associations will be particularly important as our profession increasingly investigates optimal strategies for the delivery of pediatric preventive care and seeks to avoid potential, unintended, negative consequences of system changes.11,13,14
The Children's Hospital of Philadelphia Primary Care Network consists of 4 urban Philadelphia teaching practices, with a combined total of >72000 annual visits. Less than 25% of patients in the practices have private insurance, and >75% are black. Fifty-two pediatricians and 4 nurse practitioners provide care at the 4 sites. These practices also serve as continuity clinics for ∼120 pediatric residents. At study practices, children often see a specific physician for well-child care. At sick visits, patients are scheduled with available clinicians on the basis of the acuity of their complaint and often do not see their regular well-child care provider. Immunization rates in this population closely match the rates for Philadelphia published in the National Immunization Survey.15 To ensure the completeness of vaccine records, it is a standard, routinely followed practice at study sites to check the city's immunization registry and to update information when immunization histories are incomplete.
Study Design and Patient Population
We conducted a retrospective cohort study by using data from an electronic health record decision support intervention trial (control period: September 1, 2003, to August 31, 2004; intervention period: September 1, 2004, to August 31, 2005).16 The principal aim of the original study was to use clinical alerts at the point of care to encourage immunization at the earliest possible opportunity at sick or well-child visits. All patients and visits from both the control and intervention periods of that study were eligible for the current study if they met the inclusion criteria.
For the current study, the association of immunization at sick visits with well-child care was studied on both the visit level and the patient level. The visit-level analysis included all sick visits by children <24 months of age who were due or delayed (past due) for ≥1 study vaccine and also were due or delayed for well-child care, according to the American Academy of Pediatrics recommendations for preventive pediatric health care.10 To increase the generalizability of results to community settings, a secondary analysis was conducted by excluding patient visits with trainees. Children who had not received individual doses of vaccines in the recommended age range according to national guidelines and had not caught up before eligible sick visits were defined as immunization delayed.17–19 To prevent the inclusion of multiple sick visits for a single acute illness, which would have violated assumptions of our statistical model regarding independence of the visits, only the first sick visit within a 60-day period that met the aforementioned criteria was included, unless the original sick visit had already been followed by a well-child visit. This approach simplified the problem without introducing bias. For the main patient-level analysis, all children seen at any of the study practices and born between September 1, 2003, and July 31, 2004, were included. To ensure that children were treated from the neonatal period at the study practices, a visit before 6 weeks of age at one of these sites was required. In addition, because the study focused on the effect of immunization practices at sick visits on well-child care attendance, all children in the cohort were required to have ≥1 sick visit before 13 months of age. This approach ensured that the comparison group had the opportunity to be vaccinated at a sick visit.
Vaccines included in this study were diphtheria, tetanus, and pertussis vaccine; Haemophilus influenzae type b vaccine; inactivated poliovirus vaccine; measles, mumps, and rubella vaccine; hepatitis B vaccine; pneumococcal conjugate vaccine; and varicella vaccine.
For the visit-level analysis, the main outcome was attendance at a well-child visit within 60 days after an index sick visit. The 60-day period was chosen because all study sites during the period considered had availability for scheduling well-child visits within that time frame. However, to ensure that results were not biased by the time frame considered, a sensitivity analysis used a well-child visit within 30 or 90 days after the index sick visit as the outcome. At the visit level, we also assessed whether the clinician providing care significantly influenced the likelihood of a patient returning for well-child care.
For the patient-level analysis, the outcome was the overall number of well-child visits kept by children between 6 weeks and 13 months of age, as a function of the number of eligible sick visits before 13 months of age with ≥1 vaccine given. Thirteen months of age was chosen to focus attention on the youngest children, for whom both preventive visits and immunizations are most concentrated. Well-child visits before 6 weeks of age were excluded because of variability in the number of visits typically completed by infants during the neonatal period. In the bivariate analyses, the completion of 5 well-child visits was chosen as the outcome, because this is the number of required visits specified by the American Academy of Pediatrics10 and this was the number of visits recommended uniformly by clinicians at the study sites. In a secondary analysis, however, the completion of 4 well-child visits was considered.
Covariates were chosen to adjust for factors potentially influencing immunization decisions at sick visits or the amount of well-child care received. Covariates in the visit-level analysis were as follows. Demographic factors included patient gender, race (black, white, or other), age (<2, 2 to <6, 6 to <12, 12 to <18, or 18 to < 24 months), and insurance type (nonprivate or private). Health-related factors specific to the index sick visit included the presence of immunization delay, fever, visit complexity (based on Current Procedural Terminology evaluation and management codes; low: 1 or 2 codes; medium: 3 codes; high: 4 or 5 codes), and type of clinician (attending physician, nurse practitioner, or trainee). In addition, appointment no-shows (0, 1 or 2, or >2) in the past 12 months, social work involvement in the past year, and active chronic medical conditions (derived from a previously published list) were considered.20 The conditions included diabetes mellitus, sickle cell anemia, cerebral palsy, seizures, asthma, congenital heart disease, trisomy 21, cancer, immunocompromise, autoimmune disease, major organ disease such as cystic fibrosis, developmental delays, and congenital anomalies. Site and season were also included. To confirm that findings were not biased by the use of immunization alerts in the parent study, we also controlled for whether the visits included in this analysis occurred during the control or intervention period of the underlying study. Patient-level covariates included gender, race, insurance type, chronic medical problems, a history of appointment no- shows, social work involvement, and site of care.
For the visit-level analysis, the number and proportion of sick visits with shots given, as well as each covariate, were described. The relative risk and 95% confidence interval (CI) of having a well-child visit within 60 days were calculated for each covariate. In addition, because analyses revealed effect modification by immunization status at the index sick visit, results in the multivariate analyses were reported separately for visits by children with immunizations due versus delayed.
Marginal standardization by using logistic regression was then used to model the probability of keeping a well-child visit after a sick visit, with adjustment for included covariates.21 Standardization leads to better predictions of what the outcome would be if the children in the sample were all either immunized or not immunized. The standardized risk difference in receipt of subsequent well-child care was then calculated for this sample of children. The 95% CIs for these point estimates were obtained by bootstrapping, with 999 repetitions.
To determine whether variability among clinicians providing sick care influenced the probability of children returning for subsequent well-child care, we used a mixed-effects model, estimated the physician-level random effects, and identified physicians who were outliers on the basis of the size of those random effects. In this analysis, there was 80% power to detect a 10-percentage point difference in the rate of return for subsequent well-child care among as few as 5% of physicians.
For the patient-level analysis, we described the numbers and proportions of children in the cohort with 0, 1, or ≥2 sick visits with immunizations, as well as each of the other covariates. The association between each covariate and the outcome of having completed 5 well-child visits between 6 weeks and 13 months of age was assessed by using χ2 tests. In addition, the relative risk of the outcome was calculated for each covariate. To ensure that results were not biased by children lost to follow-up monitoring, the median, top decile, and top quartile of ages when children were last seen at study practices before 13 months of age were calculated.
With adjustment for covariates, estimates for the number of well-child visits expected for children with 0, 1, or ≥2 sick visits with immunizations were obtained through marginal standardization based on Poisson regression. Bootstrapping with 999 repetitions was again used to generate CIs.
The statistical analysis was performed by using Stata 9 (Stata Corp, College Station, TX). The institutional review board of the Children's Hospital of Philadelphia approved the study.
The patient-level analysis included 1675 patients. The population was nearly equally divided between male and female patients, and <25% of patients had private insurance. Eighty-two percent of the patients were black, 7% were white, and 11% were of other races. Nine percent of the children had a chronic medical problem on their problem list. Patients had high levels of nonadherence to visits. Nearly 85% of children had not shown for ≥1 visit before 13 months of age, and >40% had ≥3 appointment no-shows during this period (Table 1). Indicating that patients were not lost to follow-up monitoring in large numbers during the study period, the median age when children were last seen before 13 months of age was 12.1 months, >75% of the patients were seen after 10 months, and >90% of the patients had a visit after 7 months of age.
Included Visits and Visit-Level Results
A total of 4845 sick visits by children who were due or delayed for both well-child care and immunizations were included (Table 2). Many children (43%) were delayed for vaccines. Fever of >101°F (8.8%) was common at sick visits.
Among all factors in the bivariate analyses (Table 2), those related to immunization receipt and status predicted most strongly whether children would return for subsequent well-child care. Those who received any immunization at a sick visit were 45% less likely to return for well-child care than were those who were not vaccinated (relative risk: 0.55 [95% CI: 0.51 to 0.60]), and those who were already delayed for shots were 37% (relative risk: 0.63 [95% CI: 0.60 to 0.67]) less likely to return than were those who were not delayed.
Multivariate analyses suggested that, of children with vaccine delay, 31% would return for well-child care within 60 days if shots were given, compared with 47% if no vaccines were given (risk difference: −16% [95% CI: −21% to −11%]). For children without vaccine delay, 42% of those who received vaccines would return for well-child care, compared with 73% of those who received no vaccines (risk difference: −31% [95% CI: −37% to −25%]). This pattern persisted in a secondary analysis that excluded trainees. In this setting, 34% of children with vaccine delay would return if shots were given, compared with 47% if no shots were given (risk difference: −13% [95% CI: −19% to −7%]). For children without delay, 50% would return if shots were given, compared with 74% if no shots were given (risk difference: −24% [95% CI: −31% to −18%]). Although rates of subsequent well-child care increased as the follow-up time for care increased from 30 to 90 days, the overall patterns of results were similar in sensitivity analyses that considered receipt of a well-child visit within 30 or 90 days after the initial sick visit (Table 3).
We next analyzed whether variability among clinicians providing sick care influenced the probability of children returning for subsequent well-child care, with adjustment for covariates. Among children with vaccine delay, we found only random variability; among those without vaccine delay, we identified only 1 clinician whose rate was significantly higher than expected. These results indicated that, controlling for fixed effects, there was no significant variation in outcomes among the >200 clinicians in the study.
At the patient level, receipt of immunizations at sick visits was associated most strongly with whether children received recommended well-child care (Table 1). In the bivariate analyses, children with 1 sick visit with shots were 69% (relative risk: 0.31 [95% CI: 0.23 to 0.41]) and those with multiple sick visits were 89% (relative risk: 0.11 [95% CI: 0.04 to 0.28]) less likely to attend the 5 well-child visits recommended between 6 weeks and 13 months of age, compared with those with no shots at any sick visit. Other factors associated significantly with decreased well-child care included appointment no-shows before 13 months of age, contact with a site-based social worker, and visit site. White children and children of other races (compared with black children) and those with private insurance were more likely to return for subsequent well-child care. Of note, results were similar when the completion of ≥4 well-child visits was chosen as the outcome.
We conducted the multivariate analysis at the patient level to measure the association between immunization at sick visits and the total number of preventive visits by children between 6 weeks and 13 months of age. With adjustment for covariates, children with no sick visits with shots would have 3.8 well-child visits (95% CI: 3.7 to 3.9 well-child visits), those with 1 sick visit with shots would have 3.3 well-child visits (95% CI: 3.2 to 3.4 well-child visits), and those with ≥2 sick visits with shots would have 2.8 well-child visits (95% CI: 2.4 to 3.0 well-child visits) (Table 4). The results indicated that, on the patient level, immunization at sick visits was associated strongly and inversely with the receipt of well-child care.
In terms of covariates in this multivariate analysis, ≥3 no-shows, ≥1 encounter with the on-site social workers, and receiving care at sites 2, 3, or 4 all predicted decreased well-child care (P < .01). Children with chronic medical problems were more likely to have received well-child care. Of note, race and insurance type, which were significant in the bivariate analyses, did not predict receipt of well-child care in the multivariate analysis.
In providing comprehensive, high-quality care, pediatric clinicians ideally vaccinate children in the most timely manner while providing the full range of recommended preventive services.10 Research has historically shown a strong association between vaccine receipt and the receipt of pediatric preventive care,22–26 but those studies did not specifically assess the impact of immunization at sick visits. Our findings from a group of urban teaching practices challenge this paradigm and suggest that efforts to improve vaccination rates by immunizing at sick visits may reduce children's subsequent and overall preventive care.
Our results are similar to previous publications that showed low overall completion rates for recommended well-child care, especially among those without private insurance.27–29 On the visit level, our results demonstrated a strong link between vaccination at a sick visit and decreased likelihood of children returning for well-child care. Although previous work demonstrated that families consider immunization the most important part of well-child care,30 it is remarkable that the effect of vaccination in predicting subsequent well-child care was more important than previous visit adherence, even among those with ≥3 no-shows. Results at the patient level demonstrated that increased immunization at sick visits between 6 weeks and 13 months of age was associated with a dose-responsive decrease in well-child visits for young children.
The contrast in the behavior of those due versus delayed for vaccines at sick visits in returning for subsequent well-child care clarifies this association. Among those delayed for vaccines at sick visits, the impact of immunization at a sick visit on subsequent well-child care was nearly half as large as that for children without delay. For many pediatric clinicians, the 16% decrease (13% with the exclusion of trainees) in subsequent preventive visits for immunization-delayed children vaccinated at sick visits might be acceptable. The association of vaccination at sick visits with a 31% decrease (24% with the exclusion of trainees) in subsequent preventive care among those without delay warrants further investigation. However, given the proven protection of timely immunization against multiple, potentially life-threatening illnesses, clinicians must carefully weigh any decision to withhold vaccination. Ultimately, a cost-effectiveness analysis assessing the trade-off between prompt vaccination and benefits of well-child care, including developmental and growth screening and detection of anemia and plumbism, might prove helpful in clarifying optimal strategies for vaccination for each group.
The approach used in this study has several limitations. First, by using a nonexperimental design, we were unable to test causality formally. Nonetheless, the results observed at both the patient and visit levels consistently supported the association between immunization at sick visits and decreased well-child care. Second, we lacked information on the specific reasons why shots were not given and the type of counseling provided to families regarding well-child care when immunizations were given at sick visits. However, by controlling for a wide range of demographic and health-related factors, such as age, chronic health conditions, no-shows, and social work involvement, this study accounted for many influences on clinicians' decisions regarding vaccination at sick visits. It is also possible that results were biased by incomplete visit or immunization records. Because the median age when patients were last seen was >12 months, >75% of patients were seen after 10 months, and incomplete vaccine records were checked against the Philadelphia immunization registry, these biases are unlikely. Furthermore, we were unable to capture specific information on factors such as the influence of clinical and office staff members, as well as office operations at the study practices, on families' likelihood of returning for well-child visits after sick visits. Differences in these areas might have contributed to variation among practices. Finally, although the patient population seen at the 4 centers included in this study is typical of many inner-city teaching practices in distinct settings and the effect of immunization at sick visits remained highly significant when controlling for site, it is possible that the associations observed for this population might not be observed in other locations.
The strong associations seen in this study between vaccination at sick visits and preventive care indicate the need for strategies that promote both timely immunization and well-child care. A simple solution might be effective counseling at sick visits regarding the non–vaccine-related benefits of well-child care. For example, recent work emphasized the role of preventive care early in life as a means to optimize school readiness.31 The lack of variability among physicians suggests that our observations reflect not the behavior of a few individual physicians but rather a generalizable phenomenon relevant across health systems. Given this finding, office-based counseling might work best if paired with a marketing campaign to define and to raise public awareness of these benefits. Furthermore, automated or manual reminder calls for well-child care, which were not made consistently during the study period, or reminders in electronic health records could be used to improve adherence for preventive care, as they have been used to improve immunization rates.16,32–34 Previous work suggested that these initiatives should be implemented in comprehensive, organized, intensive programs.26,35–37 An additional option that would eliminate the need for a return visit would be to conduct a preventive visit in addition to a sick visit when children present for acute care with well-child care and vaccines due. The challenges in such cases would involve scheduling and billing, and flexible approaches would be needed to adopt this strategy on a large scale.
More globally, increasing efforts have centered on better understanding and reforming the entire process for delivering well-child care.11,13,14,38,39 In particular, Bergman et al39 characterized potential models of well-child care that include the delivery of services in homes, churches, and schools, in addition to more traditional settings. Our findings indicate that any new systems of care must account for the strong links for families between immunization and preventive visits, so that both can be delivered with a timely, comprehensive, patient-centered approach.
This work was supported by institutional development funds from the Children's Hospital of Philadelphia.
We thank David Asch, MD, MBA, Anthony A. Luberti, MD, MS, and Lisa M. Biggs, MD, for helpful comments and technical insights. We recognize Asad Moten for background research during the course of this study. We also thank the network of primary care physicians and their patients and families for their contribution to clinical research through the Pediatric Research Consortium (PcRC) at the Children's Hospital of Philadelphia.
- Accepted September 14, 2007.
- Address correspondence to Alexander G. Fiks, MD, MSCE, Pediatric Generalist Research Group, Children's Hospital of Philadelphia, 3535 Market St, Room 1546, Philadelphia, PA 19104. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject
Up-to-date immunization rates nationally and in many regions of the United States remain below targets. Effective vaccine delivery depends on clinicians capturing available opportunities for immunization in the office, especially at sick visits.
What This Study Adds
Our findings suggest that efforts to improve vaccination rates by immunizing children at sick visits may reduce children's subsequent and overall preventive care. This link should be considered in plans to restructure the delivery of pediatric preventive care.
- ↵US Department of Health and Human Services, Office of Disease Prevention and Health Promotion. Healthy People 2010. Vol 15. Washington, DC: US Government Printing Office; 2000
- ↵National Center for Health Statistics. The 2005 National Immunization Survey. Hyattsville, MD: National Center for Health Statistics; 2006
- ↵Szilagyi PG, Rodewald LE, Humiston SG, et al. Missed opportunities for childhood vaccinations in office practices and the effect on vaccination status. Pediatrics.1993;91 (1):1– 7
- ↵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
- ↵Lieu TA, Black SB, Sorel ME, Ray P, Shinefield HR. Would better adherence to guidelines improve childhood immunization rates? Pediatrics.1996;98 (6):1062– 1068
- ↵McConnochie KM, Roghmann KJ. Immunization opportunities missed among urban poor children. Pediatrics.1992;89 (6):1019– 1026
- ↵Udovic SL, Lieu TA, Black SB, Ray PM, Ray GT, Shinefield HR. Parent reports on willingness to accept childhood immunizations during urgent care visits. Pediatrics.1998;102 (4). Available at: www.pediatrics.org/cgi/content/full/102/4/e47
- ↵American Academy of Pediatrics, Committee on Practice and Ambulatory Medicine. Recommendations for preventive pediatric health care. Pediatrics.2000;105 (3):645– 646
- ↵Schor EL. Rethinking well-child care. Pediatrics.2004;114 (1):210– 216
- ↵Coker T, Casalino LP, Alexander GC, Lantos J. Should our well-child care system be redesigned? A national survey of pediatricians. Pediatrics.2006;118 (5):1852– 1857
- ↵Moyer VA, Butler M. Gaps in the evidence for well-child care: a challenge to our profession. Pediatrics.2004;114 (6):1511– 1521
- ↵Fiks AG, Alessandrini EA, Luberti AA, Ostapenko S, Zhang X, Silber JH. Identifying factors predicting immunization delay for children followed in an urban primary care network using an electronic health record. Pediatrics.2006;118 (6). Available at: www.pediatrics.org/cgi/content/full/118/6/e1680
- ↵Fiks AG, Grundmeier RW, Biggs LM, Localio AR, Alessandrini EA. Impact of clinical alerts within an electronic health record on routine childhood immunization in an urban pediatric population. Pediatrics.2007;120 (4):707– 714
- ↵Centers for Disease Control and Prevention. Recommended immunization schedules for persons aged 0–18 years: United States, 2007. MMWR Morb Mortal Wkly Rep.2005;53 (51C–52):Q1– Q3
- ↵Korn EL, Graubard BI. Analysis of Health Surveys. New York, NY: Wiley; 1999
- Bordley WC, Margolis PA, Lannon CM. The delivery of immunizations and other preventive services in private practices. Pediatrics.1996;97 (4):467– 473
- Fairbrother G, Friedman S, DuMont KA, Lobach KS. Markers for primary care: missed opportunities to immunize and screen for lead and tuberculosis by private physicians serving large numbers of inner-city Medicaid-eligible children. Pediatrics.1996;97 (6):785– 790
- ↵Bordley WC, Margolis PA, Stuart J, Lannon C, Keyes L. Improving preventive service delivery through office systems. Pediatrics.2001;108 (3). Available at: www.pediatrics.org/cgi/content/full/108/3/e41
- ↵Byrd RS, Hoekelman RA, Auinger P. Adherence to AAP guidelines for well-child care under managed care. Pediatrics.1999;104 (3):536– 540
- ↵Schor EL, Abrams M, Shea K. Medicaid: health promotion and disease prevention for school readiness. Health Aff (Millwood).2007;26 (2):420– 429
- ↵Lieu TA, Capra AM, Makol J, Black SB, Shinefield HR. Effectiveness and cost-effectiveness of letters, automated telephone messages, or both for underimmunized children in a health maintenance organization. Pediatrics.1998;101 (4). Available at: www.pediatrics.org/cgi/content/full/101/4/e3
- ↵Schuster MA, Wood DL, Duan N, Mazel RM, Sherbourne CD, Halfon N. Utilization of well-child care services for African-American infants in a low-income community: results of a randomized, controlled case management/home visitation intervention. Pediatrics.1998;101 (6):999– 1005
- ↵Kuo AA, Inkelas M, Lotstein DS, Samson KM, Schor EL, Halfon N. Rethinking well-child care in the United States: an international comparison. Pediatrics.2006;118 (4):1692– 1702
- ↵Bergman D, Plsek, P, Saunders M. A High-Performing System for Well-Child Care: A Vision for the Future. New York, NY: Commonwealth Fund; 2006
- Copyright © 2008 by the American Academy of Pediatrics