Caring for previously hospitalized consumers : progress and challenges in mental health system reform : final report to the Joint Legislative Program Evaluation Oversight Committee

              Mental Health System Services Report No. 2008- 12- 04
Page 6 of 23
were requested from MHDDSAS on all consumers who met the criterion of
having had at least one hospitalization of 60 days or less between
January 1 and December 31, 2006. These data included information on
services received and rehospitalizations between January 1 and December
31, 2007. In addition to electronic data, the Program Evaluation Division
conducted a survey of LME administrators on service availability for
previously hospitalized consumers, ability to provide care, and barriers to
serving this population.
The evaluation sample consisted of 22,516 previously hospitalized
consumers. MHDDSAS provided information on 24,353 individuals who
met study criteria. The data were drawn from Medicaid and state provider
claims that had been filed for payment. Upon examination of the dataset,
the Program Evaluation Division identified 206 cases that did not have
substance abuse and/ or mental health diagnoses; these were dropped
from the sample. In addition, differences in data submission requirements
rendered unreliable the 2007 data submitted by Piedmont Behavioral
Healthcare and Smoky Mountain Center LMEs. 9
The resulting evaluation sample of 22,516 previously hospitalized
consumers represented 7% of the total population served by MHDDSAS in
Fiscal Year 2005- 06. All consumers in this sample qualified for Medicaid
and/ or state funds to cover services. 10 Findings are limited to this group
and are not intended to represent the wider population of MHDDSAS
consumers.
The sample included 4,955 children ( or 22% of the sample; defined as
individuals under 21 years of age) and 17,561 adults ( 78%). A small
majority of consumers were male ( 53%), and most were White ( 60%) or
Black ( 35%). Another 5% of the sample consisted of individuals from other
backgrounds ( unknown race, Hispanic, and other groups too small to be
analyzed alone). As compared with the total population of Fiscal Year
2005- 06 mental health consumers, the evaluation sample had fewer
children ( 22% vs. 29%) but was otherwise demographically similar.
The Program Evaluation Division developed three diagnostic categories to
describe the sample for analysis: those with only mental health problems,
those with only substance abuse disorders, and those with dual diagnoses
( i. e., mental health co- occurring with substance abuse). As shown in Exhibit
2, half of the sample had only mental health problems, and half had
substance abuse alone or in conjunction with mental health problems. This
distribution was different when children and adults were examined
separately: over three- quarters ( 78%) of children were diagnosed only
with mental health disorders.
9 Whereas single- stream funding can increase LME efficiency, it also can contribute to unreliable service data because reporting is no
longer tied to payment. MHDDSAS Analyst Adam Holtzman advised the Program Evaluation Division about significant missing 2007
service data from two LMEs: Smoky Mountain ( which has had single- stream funding since 2004) and Piedmont ( operating under a
Medicaid waiver).
10 Medicaid eligibility varies by disability, age, and other factors; see http:// www. dhhs. state. nc. us/ DMA/ medicaid/ who. htm for
detailed information. See http:// www. dhhs. state. nc. us/ MHDDSAS/ iprsmenu/ iprseligibilitymatrix. xls for more information on eligibility
for state- funded services.