Background Report for the Request for Public Comment on Initial, Recommended Core Set of Children's Healthcare Quality Measures for Voluntary Use by Medicaid and CHIP Programs
Mental and Behavioral Health
Information about the prevalence of mental and behavioral health problems is difficult to come by.58 Parents of children ages 2-17 reported to the NSCH that 3.8 percent of their children were depressed, 4.5 percent were anxious, and 6.9 percent had behavior or conduct problems.38 For older children, self-reports using standardized assessment tools may be more accurate. Relatively old research shows that one in five children and adolescents aged 9 to 17 years experience symptoms of mental health problems that cause some level of impairment in a given year, and that only 20 percent get treatment.59
- Followup after hospitalization for mental illness (NCQA). This measure assesses whether individuals aged 6 years and older who have had a mental hospitalization and were discharged from the hospitalization had an outpatient visit, intensive outpatient encounter, or partial hospitalization with a mental health practitioner within 7 or 30 days after discharge (two rates are reported). To be eligible for this measure, the member must be continuously enrolled from the day of discharge through 30 days after.
This is a process measure with an evidence grade of B. SNAC voting resulted in a ranking of 14 for this measure. Current measure specifications are available at the health plan and provider levels, and current reporting is at the health plan level. Administrative records provide the data. The measure is reported to NCQA by 55 health plans across 17 States. Participation in reporting may be affected by the number of States using carve-out plans to provide mental health services.
Child and adolescent hospitalizations for mental health conditions typically indicate high levels of severity. In 2006, Medicaid paid for 69,200 community hospital discharges for mental health and substance abuse conditions as a primary diagnosis among children 0-17.60 Another 238,383 discharges were for mental health and substance abuse conditions as a secondary diagnosis. In 1997, 37.6 percent of all child inpatient psychiatric hospitalizations were paid for by Medicaid, equal to 107,602 children.61 The most dominant conditions for which children with Medicaid as expected payer were hospitalized in community hospitals in 2006 were mood disorders (38,000 discharges) and ADHD/disruptive behaviors (14,293 discharges).60 The Child Health Insurance Research Initiative (CHIRI™) reports that mental health issues (e.g., attention and behavior disorders) were second only to asthma as the top health problems of State CHIP-enrolled children with special health care needs, as reported by their families.62
Performance data for this measure are not currently available separately for children. For all ages combined, commercial rates were higher than Medicare and Medicaid rates. For all ages combined, 42.5 percent of Medicaid health plans had followup within 7 days (bottom region, South Atlantic; top region, Northeast). 61.5 percent had followup within 30 days (bottom region South Atlantic, 39.1 percent; top region Northeast, 80.7 percent).26
Validity and reliability of reporting may be affected by variations in the use of carve-outs across plans and States. NCQA has indicated their willingness to change the measure reporting fields so that children can be reported separately from adults.
Health care providers and others are raising concerns about the rising rate of diabetes among children and adolescents. Part of the reason for concerns about overweight and obesity in children is the relationship between those problems and diabetes.
- Annual hemoglobin A1C testing (all children and adolescents diagnosed with diabetes). This State of Alabama Medicaid program measure assesses the extent to which patients ages 5-17 with a diagnosis of diabetes and/or notation of prescribed insulin/oral hypoglycemics/antihyperglycemics for at least 12 months receive at least one hemoglobin A1C test. Receipt is defined by documentation of a face-to-face visit for diabetes care between the physician and the patient that predates the most recent visit by at least 12 months.
This is a process measure with an evidence grade of D. SNAC voting resulted in a rank of 13 for this measure. Data sources are administrative and medical records. Alabama Medicaid is currently using this measure
For people with diabetes, hemoglobin A1c testing is important. This is a well-known health care quality measure for the adult population.63 The prevalence of diabetes is relatively low among young people (0.22 percent among people younger than 20 years of age, or 176,000 people64), although there are a substantial number of young people considered to be pre-diabetic. In addition, there is concern that the current epidemic of obesity and overweight among children and adolescents may result in an explosion of diabetes among children and young adults. Both overweight and diabetes tend to be more prevalent in African Americans and Hispanics. Using the AHRQ Pediatric Quality Indicator (PDI), the rate of hospital admissions for short-term complications of diabetes among 6-17 year-olds increased from 27.1 per 100,000 population in 2000 to 36.1 per 100,000 population in 2005.63 Data are not available by source of insurance, but rates were lower for children who lived in high-income communities.
Performance data for this measure as specified are not available.
A challenge to feasibility of use by States and plans may be low numbers of children and adolescents with diabetes in any one setting. Low numbers can be highly unreliable.
Family Experiences of Care
The CHIPRA legislation explicitly asked for core measures for family experiences of care. Such measures have been in use and constantly updated for children with and without chronic conditions for more than a decade. The SNAC recommended that all children be surveyed using the core survey and all supplements, given their appropriateness for children with and without chronic conditions.
- HEDIS CAHPS® 4.0H including supplements for children with chronic conditions and Medicaid Plans. CAHPS® is a family of surveys of experiences of care, an aspect of patient-centeredness. Parents or other responsible adults report about experiences of care during visits in which they accompany their children. Currently, health plans reporting to NCQA must collect and report data for the CAHPS® Health Plan Survey 4.0H, Child Version, with or without Children with Chronic Conditions items. If an organization does not want to collect and report Children with Chronic Conditions, it uses a version of the questionnaire that does not contain the CCC questions (this is consistent with what AHRQ refers to as the 'core questionnaire'). The questionnaires are the same for Medicaid and commercial plans, but the vendor changes the question wording look-back from 12 months (commercial) to 6 months for Medicaid.
This is an outcome measure with an evidence grade of B. SNAC voting resulted in a ranking of 8 for this combination of measures. Current reporting is at the health plan level. The core instrument is currently reported to NCQA by 38 health plans. HEDIS CAHPS® for children with chronic conditions is currently reported to NCQA by 28 health plans.
Patient-centered (and for children, family-centered) care is a key domain of health care quality. Some leaders in quality have referred to patient-centeredness as the "true north" of health care delivery.17 As recommended by the SNAC, all children enrolled in Medicaid and CHIP should be assessed for patient-centeredness of services using all components of CAHPS® (e.g., core instrument, plus the CAHPS® for Children with Chronic Conditions and the Medicaid 4.0 CAHPS®). The sense of the group was that many of the items previously deemed relevant only to children with chronic conditions (e.g., coordination of care, health plan responsiveness) are equally applicable across the child population. They also noted that the health plan survey items would be relevant to all covered children.
Performance data from CAHPS® are not available from NCQA, probably because of the relatively low number of health plans reporting to NCQA. Other States may use the measure, but not report it to NCQA. However, data are available from other sources that use identical or similar items and composite.65 For example, the National CAHPS® Benchmarking Database reports data on responses to HEDIS measures. For example, 53 percent of 19,019 respondents responded to the item "Child's personal doctor seemed informed and up-to-date about care from other doctors or providers" by checking "always." Only 22 percent nationally responded "never or sometimes" to this item.66
Feasibility is recognized as a major challenge for this measure. Surveys can be resource-intensive.
Availability of Services
CHIPRA directed the Secretary to include measures of the availability of services, either in the initial, recommended core set or as a topic in the pediatric quality measures program. Availability of services is an important topic, given the many reports of lack of providers and some providers' unwillingness to serve Medicaid patients.67-69 We considered measures of realized access to care (e.g., utilization of primary care practitioners) to be incomplete measures of availability because the reasons for lack of utilization could go well beyond lack of availability (e.g., parents don't perceive a need for the service). However, realized access is a piece of availability and a way to measure access under capitated primary care arrangements.
Access to Services
- Access to primary care practitioners, by age and total. For children aged 12-24 months, and 25 months-6 years, this NCQA measure assesses the receipt of one or more visits with a primary care provider during the measurement year. For children aged 7-11 and 12-19 years, the measure assesses the receipt of one or more visits with a primary care provider during the measurement year or the year prior to the measurement year. Continuous enrollment for the younger two age groups is required for the measurement year. For the two older age groups, continuous enrollment is required for both sequential measurement years.
This is a proxy measure for availability of services. There is no evidence grade because it is an access measure rather than a traditional quality measure. Access may be considered a health care quality measure from a systems perspective. Specifications are available at the health plan and provider levels, and reporting is at the plan level. The measure is currently reported to NCQA by between 120 and 129 health plans across 30-32 states, depending on age group. The data source is administrative records.
This measure can be considered important from two perspectives. From a payer perspective, it is good to know whether enrolled children are getting any care. Access to primary care practitioners is essential for children to have well child visits and the opportunity for social and developmental screening, two measures also included in the recommended set. Performance on this measure is generally good. The younger the child, the better the performance is. The national mean for Medicaid-enrolled children aged 12-24 months in reporting plans was 93.4 percent; the 10th and 90th percentiles were 87.7 percent and 98.4 percent, respectively. For children 25 months-6 years of age, the national mean was 84.3 percent; the 10th and 90th percentiles were 74.2 percent and 91.9 percent, respectively. For children 7-11 years of age, the national mean was 85.86 percent; the 10th and 90th percentiles were 75.5 percent and 94.1 percent, respectively. For children aged 12-19 years, the national mean was 82.66 percent; the 10th and 90th percentiles were 70. 6 percent and 91.8 percent, respectively.
The measure is feasible to use because it comes from administrative data. The measure cannot be considered a true availability of services measure because it deals with only primary care, and because the reasons for non-use of primary care services may vary, from lack of availability, to lack of time, to lack of perceived need, and other factors.