Analysis of the California In-Home Supportive Services (IHSS) Plus Waiver Demonstration Program. Appendix C. Testing Factors Associated with Provider Selection

07/01/2008

The analyses presented in this Appendix extend the descriptive findings using logistic regression to adjust for recipient differences within a provider group. Separate analyses are conducted by recipient age group to address the following questions:

  • Do the functional limitations, task assistance needs, and chronic health conditions of individuals differ between waiver and non-waiver recipients?

  • Do waiver and non-waiver recipients differ in terms of living arrangement household size, race/ethnicity, and availability of legally responsible relatives?

  • Adjusting for recipient characteristics and living arrangements, are there differences between new and continuing IHSS recipients in provider selection?

The “outcome” of these models is provider type. The recipient data are from the first available IHSS assessment instrument available prior to the start of the observation period. Most of the models, Table C-1, Table C-2 and Table C-3, include all IHSS recipients in the modeled age group, but there are three exceptions. The models predicting Parent providers among those age 3-17 are limited to minor children with a parent in the household. Similarly, the models predicting Spouse providers among adults are limited to persons with a spouse in the household. The Authorized IHSS Service Hours, while it may reflect unmeasured functional status is not included in these models as this measure is determined in part by living arrangement and provider type.

The coefficient of primary interest in these models is the dummy variable representing recipients entering IHSS in 2005 versus those who were continuing from 2004. Among minor children, those with Parent providers are less likely to be newly entering IHSS recipients in 2005, and those with Non-Relative providers are more likely to be new recipients. Non-aged adults reflect a somewhat similar pattern with those having Parent or Spouse providers being less likely to be new recipients. Correspondingly, new recipients had higher odds of having either Other Relatives and Non-Relative providers. This pattern changes somewhat among those 65+. New IHSS recipients have higher odds than continuing recipients of having a Spouse provider, and slightly lower odds of having a Non-Relative provider. There is no difference between recipient participation status in the odds of having Other Relative providers. These findings suggest that between 2004 and 2005 (the starting year for the IHSS Plus Waiver program) there was no consistent pattern in the enrollment of new recipients shifting them toward waiver covered providers (i.e., parents for minor children recipients, spouses for adult recipients). Among children the opposite occurred, with new recipients being less likely than continuing recipients to be among those with Parent providers. Among non-aged adults there was no difference in the odds of new recipients using Spouse providers. Only among those age 65+ did new recipients have greater odds than continuing recipients of having a Spouse provider.

Turning now to the other policy questions, the coefficients in the models provide indicators of the statistically significant differences within provider type relative to race/ethnicity, functional ability, and living arrangement.

The findings from the tables are briefly summarized here:

  • Females show no difference from males in predicting providers among children. However among adults, they are less likely to have Parent or Spouse providers. They are also more likely to have Other Relative providers and less likely to have Non-Relatives.

  • Hispanic recipients across all age strata are substantially more likely than Whites to have Parent or Spouse providers, and less likely to have Non-Relative providers; and among adults to be more likely to have Other Relative providers.

  • Black recipients across children and non-aged adults are less likely than Whites to have Parent providers, more likely to have Other Relative providers, and less likely to have Non-Relative providers. Among the aged they are more likely to have a Spouse provider and somewhat less likely to have Other Relative providers.

  • Asian and the other minority children recipients are more likely than Whites to use Parent and Other Relative providers, and less likely to use Non-Relatives. Among adults they are more likely than Whites to use Spouse and Other Relative providers and less likely to use Parents and Non-Relatives.

  • The odds of having a paid Parent provider increase substantially among minor children who require task assistance from another individual in three or more ADLs, or in three cognitive tasks, or with breathing problems. These same factors reduce the odds of having other providers (with no difference between Relative and Non-Relative providers). Among adult recipients these higher levels of limitations are generally associated with increased odds of Parent and Spouse providers (when they are available). Correspondingly these frailty measures are generally associated with reduced odds of having either Other Relative or Non-Relative providers.

  • Managed care membership is in the models to assess the extent to which enrollment in these Medicaid plans might be biased relative to the various provider types. Among minor children there was generally no significant difference in membership among those with each type of provider. The exception was a marginally significant difference with those having Non-Relative providers being less likely to be in managed care. For recipients age 18-64, managed care members were more likely among those with Parent and Spouse providers, and less likely among those with Other Relative and Non-Relative providers. For recipients age 65+, managed care members were more likely among Spouse and Other Relative providers, and less likely among Non-Relatives. Managed care differences may contribute to an under count of chronic health conditions and health service use among those provider types having higher likelihood of enrollment in such plans.

  • Household size, adjusting for race, has no association with the odds of Parent providers for children, but larger households are more likely to have Other Relative and less likely to have Non-Relative providers. Among those 18-64 size is positively related to having a Parent or Other Relative provider, and negatively related to having Non-Relative providers. Recipients age 65 or more in larger households are less likely to have Spouse and Non-Relative providers, and more likely to have Other Relatives.

  • Regardless of recipient age, the presence of a parent or spouse in the household reduces the odds of having non-relatives as paid IHSS providers. Among those 18-64 a present spouse has a modest association with having Other Relative providers.

  • Among minor children the comparison of recipients in counties with IHSS wage levels relative to the average rate in the state (i.e., Los Angeles and Fresno Counties) showed few statistically significant relationships. In counties with hourly wages of $10 or more, the likelihood of a Parent being a paid provider reduced relative to recipients in Los Angeles. No differences were found for the other provider groups. Recipients age 18-64 offer a somewhat similar pattern. Parents in counties with IHSS wages above $9 per hour were less likely to be paid providers, and there was a modest tendency for non-relatives to assume the provider role. The choice of Spouse provider was positive across all wage rate levels, suggesting differences between Los Angeles and Fresno and the other counties in the state that may not necessarily be related to wage rates. Among aged recipients, the prior pattern for Spouse providers holds, accept in the highest wage rate counties, which do not differ from Los Angeles and Fresno. Across all but the highest wage rates, counties show a tendency toward more Other Relative providers and somewhat less likelihood of Non-Relative providers than Los Angeles and Fresno.3

  • County per capita, in the models as an adjustment for cost of living, has an association with provider type, but with a weight near 1.0 across all recipient ages and provider types. These effects reflect the change in likelihood of a provider with each $1,000 increment in county per capita income. There is a modest positive association with Non-Relative providers among all age groups, and no statistically significant association with Spouse providers among adult recipients. Among minor children recipients, higher per capita income is associated with a reduced likelihood of a Parent provider. Among those age 18-64, there is a positive association with per capita income and Parent providers, and for all adult recipients a lower likelihood of Other Relative providers.

TABLE C-1: Predicting Provider Relationships, IHSS Recipients Age 3-17, 2005
Total Recipients
n=16257
Provider Relationship
Parentg
n=11056
Other Relative
n=2090
Non-Relative
n=2589
Predictors   Odds Ratio   95% CI   Odds Ratio   95% CI   Odds Ratio   95% CI
SOURCE: California Department of Social Services, unpublished CMIPS data for 2005.
  1. Reference is White. Race/ethnicity Asian/Other includes in descending order by number in this age group, Vietnamese, Filipino, Chinese, Cambodian, Laotian, American Indian or Alaskan Native, Asian Indian, Korean, Samoan, Hawaiian, Japanese, and all others.
  2. Cognition is defined by: memory, orientation, and judgment. Each scored 1 independent; 2 able to perform, but needs verbal assistance such as reminders, guidance, or encouragement; 5 cannot perform without human assistance. Scores three and four not used. The measure is a dummy variable yes = have three cognitive measures each with a score five.
  3. ADLs refers to activities of daily living (i.e., bathing and grooming; dressing; transferring; bowel, bladder and menstrual; eating). Each task is scored on a four or five point scale: 1 and 2 as per above, 3 Can perform with some human direct physical assistance from the provider, 4 Can perform with a lot of human assistance, 5 cannot perform without human assistance. The measure is a dummy variable yes = have three or more ADLs each with a score of three or more indicating the need for human assistance.
  4. Breathing is scored 1 independent, 5 cannot perform without human assistance, 6 paramedical services needed. The measure is the presence/absence of a breathing item with a score of five or more.
  5. Number of persons in household, including other IHSS recipients, excludes recipients non-IHSS children <age 14.
  6. The omitted category for IHSS wages is $7.50/hr (reflective of Los Angeles and Fresno Counties residence for about 44% of IHSS recipients.
  7. The model includes only recipients who have a parent present, n shown in the column heading is the number of Parent providers, n shown in the Parent Present row is the number of parents available in the sample. “na” is not applicable.
Intercept            
Recipient Characteristicsa
Female Recipient 1.04   0.96-1.12   0.95   0.86-1.05   0.99   0.91-1.08  
Hispanic 1.94 1.75-2.14 0.79 0.70-0.90 0.44 0.40-0.50
Black 0.80 0.71-0.90 1.63 1.42-1.88 0.87 0.77-0.98
Asian/Other 1.28 1.10-1.48 1.21 1.01-1.43 0.57 0.49-0.67
3+ Cognitive Limitationsb 1.22 1.11-1.35 0.83 0.73-0.93 0.86 0.77-0.95
3+ ADL Limitationsc 1.17 1.06-1.30 0.88 0.78-0.99 0.85 0.77-0.95
Respiratory Limitationsd 1.15 1.03-1.29 0.99 0.86-1.14 0.80 0.70-0.91
Managed Care Member 1.08 0.98-1.18 0.93 0.83-1.04 0.90 0.81-0.99
Living Arrangement
Parent Present (n=14623)   na na 0.19 0.17-0.21 0.47 0.42-0.53
Household size (1-5)e 0.98 0.94-1.02 1.10 1.05-1.15 0.93 0.89-0.97
County Measures
IHSS Wage Ratef            
  <$7.50 per hour 1.08 0.96-1.23 0.74 0.63-0.87 1.02 0.89-1.16
  $8 per hour 1.30 1.17-1.45 0.92 0.81-1.05 0.69 0.61-0.79
  $9.50-$9.75 per hour 0.88 0.78-1.01 1.00 0.85-1.18 1.07 0.92-1.23
  >$10 per hour 0.64 0.49-0.83 1.11 0.81-1.52 0.93 0.71-1.23
Per Capita Income (000’s) 0.97 0.97-0.98 1.01 1.00-1.02 1.03 1.02-1.04
New IHSS Recipient 0.69 0.63-0.76 1.03 0.91-1.07 1.43 1.28-1.59
Model Goodness of Fit
-2Log Likelihood   15536.5       11471.0       13493.0    
Maximum Rescaled R2 0.071   0.112   0.078  

 

TABLE C-2: Predicting Provider Relationships, IHSS Recipients Age 18-64, 2005
Total Recipients
n=152,591
Provider Relationship
Parent
n=23,492
Spouseg
n=8,132
Other Relative
n=49,407
Non-Relative
n=70,974
Predictors   Odds Ratio   95% CI   Odds Ratio   95% CI   Odds Ratio   95% CI   Odds Ratio   95% CI
SOURCE: California Department of Social Services, unpublished CMIPS data for 2005.
  1. Reference is White. Race/ethnicity Asian/Other includes in descending order by number in this age group, Vietnamese, Laotian, Chinese, Filipino, Cambodian, American Indian or Alaskan Native, Korean, Asian Indian, Samoan, Japanese, and others.
  2. Cognition is defined by: memory, orientation, and judgment. Each scored 1 independent; 2 able to perform, but needs verbal assistance such as reminders, guidance, or encouragement; 5 cannot perform without human assistance. Scores three and four not used. The measure is a dummy variable yes = have three cognitive measures each with a score five.
  3. ADLs refers to activities of daily living (i.e., bathing and grooming; dressing; transferring; bowel, bladder and menstrual; eating). Each task is scored on a four or five point scale: 1 and 2 as per above, 3 Can perform with some human direct physical assistance from the provider, 4 Can perform with a lot of human assistance, 5 cannot perform without human assistance. The measure is a dummy variable yes = have three or more ADLs each with a score of three or more indicating the need for human assistance.
  4. Breathing is scored 1 independent, 5 cannot perform without human assistance, 6 paramedical services needed. The measure is the presence/absence of a breathing item with a score of five or more.
  5. Number of persons in household, including other IHSS recipients, excludes recipients non-IHSS children <age 14.
  6. The omitted category for IHSS wages is $7.50/hr (reflective of Los Angeles and Fresno Counties residence for about 44% of IHSS recipients.
  7. The model includes only recipients who have a spouse present, n shown in the column heading is the number of Spouse providers, n shown in the Spouse Present row is the number of spouses available in the IHSS recipient sample. “na” is not applicable.
Intercept                
Recipient Characteristicsa
Female Recipient 0.50   0.48-0.51   0.31   0.29-0.33   2.04   1.99-2.09   0.90   0.88-0.92  
Hispanic 1.42 1.37-1.48 2.51 2.33-2.71 1.16 1.12-1.19 0.56 0.55-0.58
Black 0.56 0.53-0.58 1.54 1.38-1.72 1.32 1.28-1.36 0.90 0.88-0.93
Asian/Other 0.86 0.81-0.91 1.57 1.44-1.72 1.79 1.72-1.86 0.51 0.49-0.53
3+ Cognitive Limitationsb 4.10 3.92-4.28 2.21 1.84-2.67 0.54 0.51-0.57 0.45 0.43-0.48
3+ ADL Limitationsc   1.53 1.48-1.58 5.27 4.90-5.68 0.75 0.74-0.77 0.85 0.83-0.87
Respiratory Limitationsd 0.97 0.91-1.03 1.29 1.16-1.44 0.91 0.87-0.95 1.05 1.01-1.10
Managed Care Member 1.22 1.17-1.27 1.41 1.31-1.53 0.89 0.87-0.92 0.94 0.91-0.97
Living Arrangement
Spouse Present (n=25,516) 0.03 0.02-0.03 na na 1.55 1.51-1.60 0.40 0.39-0.41
Household size (1-5)e 1.53 1.51-1.54 0.99 0.97-1.02 1.24 1.23-1.25 0.65 0.64-0.65
County Measures
IHSS Wage Ratef                
  <$7.50 per hour 1.10 1.04-1.15 2.07 1.90-2.26 0.61 0.59-0.63 1.27 1.23-1.31
  $8 per hour 1.39 1.33-1.45 2.06 1.89-2.26 0.80 0.77-0.82 0.89 0.86-0.92
  $9.50-$9.75 per hour 0.93 0.88-0.98 1.29 1.17-1.43 0.94 0.91-0.97 1.04 1.00-1.07
  >$10 per hour 0.78 0.70-0.87 1.59 1.28-1.98 0.90 0.83-0.97 1.19 1.10-1.28
Per Capita Income (000’s) 1.01 1.01-1.01 1.00 0.99-1.01 0.98 0.98-0.98 1.01 1.01-1.02
New IHSS Recipient 0.57 0.55-0.60 0.95 0.88-1.03 1.14 1.11-1.18 1.11 1.08-1.14
Model Goodness of Fit
-2Log Likelihood   103,330.5       25,616.7       181,058.3       185,067.1    
Maximum Rescaled R2 0.288   0.307   0.098   0.207  

 

TABLE C-3: Predicting Provider Relationships, IHSS Recipients 65+ or More, 2005
Total Recipients
n=233,856
Provider Relationship
Spouseg
n=5,130
Other Relative
n=123,496
Non-Relative
n=104,971
Predictors   Odds Ratio   95% CI   Odds Ratio   95% CI   Odds Ratio   95% CI
SOURCE: California Department of Social Services, unpublished CMIPS data for 2005.
  1. Reference is White. Race/ethnicity Asian/Other includes in descending order by number in this age group, Chinese, Filipino, Vietnamese, Korean, Laotian, Cambodian, Asian Indian, American Indian or Alaskan Native, Japanese, Samoan, and all others.
  2. Cognition is defined by: memory, orientation, and judgment. Each scored 1 independent; 2 able to perform, but needs verbal assistance such as reminders, guidance, or encouragement; 5 cannot perform without human assistance. Scores three and four not used. The measure is a dummy variable yes = have three cognitive measures each with a score five.
  3. ADLs refers to activities of daily living (i.e., bathing and grooming; dressing; transferring; bowel, bladder and menstrual; eating). Each task is scored on a four or five point scale: 1 and 2 as per above, 3 Can perform with some human direct physical assistance from the provider, 4 Can perform with a lot of human assistance, 5 cannot perform without human assistance. The measure is a dummy variable yes = have three or more ADLs each with a score of three or more indicating the need for human assistance.
  4. Breathing is scored 1 independent, 5 cannot perform without human assistance, 6 paramedical services needed. The measure is the presence/absence of a breathing item with a score of five or more.
  5. Number of persons in household, including other IHSS recipients, excludes recipient’s non-IHSS children <age 14.
  6. The omitted category for IHSS wages is $7.50/hr (reflective of Los Angeles and Fresno Counties residence for about 44% of IHSS recipients.
  7. The model includes only recipients who have a spouse present, n shown in the column heading is the number of Spouse providers, n shown in the Spouse Present row is the number of spouses available in the IHSS recipient sample. “na” is not applicable.
Intercept            
Recipient Characteristics
Female Recipient 0.23   0.21-0.24   1.38   1.35-1.41   0.85   0.84-0.87  
Hispanica 4.75 4.37-5.17 1.13 1.10-1.16 0.76 0.75-0.78
Blacka 3.64 3.10-4.27 0.94 0.92-0.97 0.98 0.95-1.01
Asian/Othera 2.31 2.12-2.52 1.46 1.42-1.49 0.63 0.62-0.65
3+ Cognitive Limitationsb 1.58 1.34-1.86 0.98 0.93-1.03 0.98 0.93-1.03
3+ ADL Limitationsc 7.79 7.12-8.51 1.03 1.01-1.05 0.84 0.82-0.85
Respiratory Limitationsd 1.48 1.34-1.64 0.93 0.90-0.96 1.02 0.98-1.05
Managed Care Member 1.16 1.05-1.29 1.08 1.05-1.11 0.90 0.87-0.93
Living Arrangement
Spouse Present (n=69,034)   na na 1.10 1.08-1.12 0.68 0.66-0.69
Household size (1-5+)e 0.88 0.86-0.91 1.60 1.59-1.61 0.63 0.62-0.63
County Measures
IHSS Wage Ratef            
  <$7.50 per hour 1.99 1.80-2.20 0.79 0.77-0.81 1.16 1.12-1.19
  $8 per hour 1.86 1.70-2.04 1.23 1.19-1.26 0.75 0.73-0.77
  $9.50-$9.75 per hour 1.32 1.19-1.48 1.17 1.13-1.20 0.82 0.79-0.84
  >$10 per hour 0.88 0.72-1.09 1.06 1.00-1.12 0.94 0.89-1.00
Per Capita Income (000’s) 1.00 1.00-1.01 0.99 0.98-0.99 1.02 1.01-1.02
New IHSS Recipients 1.32 1.22-1.44 1.02 0.99-1.04 0.95 0.93-0.97
Model Goodness of Fit
-2Log Likelihood   29,318.4       296,098.0       289,899.0    
Maximum Rescaled R2 0.242   0.147   0.170  

 

View full report

Preview
Download

"IHSSPlus.pdf" (pdf, 4.9Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®