Two state constraints were significantly associated with the percentage of LTSS expenditures for HCBS (Table III.3) and with the percentage of LTSS users receiving HCBS (data not shown). First, increased demand for HCBS (as measured by the percentage of aged potential Medicaid enrollees in a state) was negatively associated with HCBS balance. The source of this negative association is impossible to identify with certainty. It might indicate that state policymakers fear a "woodwork effect," as described in Chapter I. That is, larger elderly populations may lead to fears of a greater demand response to availability of HCBS. Second, as in 2006, greater availability of personal and home care aides was positively associated with greater balance toward HCBS in LTSS expenditures and use. The consistent relevance of this factor points to the value of further consideration of this relationship and how efforts to improve rates of HCBS use should account for workforce supply factors. One factor -- average winter precipitation -- that was negatively correlated with HCBS spending in 2006 did not have a significant association with HCBS expenditures or use in 2009.17 It may be that states that were most affected by weather-related challenges have made progress in providing HCBS in ways that address these challenges. Or, this difference in results may be caused by the inclusion of different states in this analysis or point to the instability of bivariate associations in a given year. Overall, these results suggest that, as in 2006, we have not identified exogenous factors that can be substantially linked to levels of LTSS system balance across states for the LTSS Medicaid population as a whole.
17 Three factors were significantly associated with the percentage of LTC expenditures for HCBS but were not significantly associated with percentages of LTC users receiving HCBS. First, single-family house price index and total taxable resources were both positively associated with rates of HCBS expenditures. These associations may suggest that communities with greater overall resources are more able to support efforts to provide care in the community. Although we initially expected that single-family home prices would be negatively associated with HCBS use, it is plausible that this indicator is capturing overall wealth and resources in a community, rather than simply the costs required to remain in the community. Neither of these factors, however, was significantly associated with HCBS expenditures in 2006.