Annual IDD mortality report, CY 2023

Annual IDD Mortality Report

CY2023

GEORGIA DEPARTMENT of BEHAVIORAL HEALTH and DEVELOPMENTAL DISABILITIES Office of Performance Analysis and Quality Improvement November 2024

TABLE OF CONTENTS
Executive Summary ............................................................................................................................. 2 Major Findings .............................................................................................................................................. 2 Purpose and Scope of this Report ........................................................................................................ 4 Causes of Death Among the Intellectual and Developmental Disability Waiver Population .................. 5 IDD Mortality During CY2023 ............................................................................................................... 8 Age and Mortality ......................................................................................................................................... 8 Health Risk and Mortality ............................................................................................................................. 9 The Central Importance of Age and Health Risk ......................................................................................... 10 HRST Rating Items and Mortality ................................................................................................................ 12 Individual Profile Based on Top Four Rating Items ........................................ Error! Bookmark not defined. Community Mortality Review Committee related Deficient Practice Analysis .................................... 12 Background ................................................................................................................................................. 12 Classification of Deaths ............................................................................................................................... 13 Statewide Analysis of Number and Type of Deficient Practices ................................................................. 14 Key Findings....................................................................................................................................... 19 Appendix A: Method for Mortality Review and Analysis.................................................................... 21 Crude Mortality Rate .................................................................................................................................. 21 Analysis and Measures................................................................................................................................ 22 Appendix B: NOW/COMP Population Demographics ......................................................................... 24 Characteristics of the Intellectual and Developmental Disability Waiver Population................................ 24 Appendix C: DBHDD Sampling Procedure ........................................................................................... 26 Appendix D: HRST Domains ............................................................................................................... 28 Appendix E: HRST Expanded Score Descriptors .................................................................................. 29 Appendix F: Additional Statistical Information ................................................................................... 38
1|Page

EXECUTIVE SUMMARY
This report includes data and information concerning adults who died during calendar year 2023 (CY2023) while receiving intellectual and developmental disability (IDD) Medicaid waiver services authorized by the Georgia Department of Behavioral Health and Developmental Disabilities ("DBHDD") and rendered by its contracted providers.
An analysis of individual deaths and trends in mortality is a component of health and safety oversight and is part of DBHDD's quality management and improvement system. This is the tenth annual mortality report released by DBHDD. The purpose of this report is to provide CY2023 information about what DBHDD has learned about deaths, to identify trends or patterns in mortality, and to identify indicators that may assist DBHDD in the prevention and treatment of certain illnesses or conditions that may lead to deaths or other disorders or diseases in the future. This report does not issue recommendations, as these will emanate from later processes when DBHDD has had the opportunity to consider findings and observations reported within this document.
MAJOR FINDINGS
In CY2023, DBHDD served 13,916 adults with intellectual and developmental disabilities in waiver services. A total of 237 deaths occurred in CY2023, resulting in a crude mortality rate of 17.03 deaths per 1,000 individuals, or 1.7%.1, 2
As in previous years, several of the 10 leading causes of death for general populations of the United States or Georgia were also leading causes of death in the IDD population. Several variables were analyzed to determine their association with mortality in CY2023. These included age, gender, health risk, residential setting, race, region, marital status, service intensity and the 22 rating items of the Health Risk Screening Tool (HRST). Major analytical findings from CY2023 mirror those from CY2021 and CY2022: increasing health risk and increasing age were most strongly associated with mortality, while gender, residential setting, race, region, marital status, service intensity, and other variables were not significantly related to mortality.
DBHDD's Community Mortality Review Committee (CMRC) uses a standardized, systematic process to conduct mortality reviews to identify opportunities to reduce morbidity, mortality, and identify opportunities to improve the quality of services. CMRC data review identified low-, moderate-, high-, and critical-risk provider deficient practices.
1 The mortality rate used in this report is a crude mortality rate, which is an unadjusted mortality rate. The mortality rate is a measure of how many people out of every thousand served by DBHDD died within the calendar year. It is determined by multiplying the number of people who died during the year by 1,000, then dividing by the total number of individuals served in the NOW/COMP waiver program during the same year. The crude mortality rate can be useful when comparing deaths across populations of varying sizes. For the purposes of the remainder of this report, crude mortality rate will be referred to as "mortality rate." 2 Standard recommended by the U.S. Centers for Disease Control and Prevention, National Vital Statistics Report, Age Standardization of Death Rates: Implementation of the Year 2000 Standard, Vol. 47, No. 3, 1998.
2|Page

In CY2023, the most common provider deficiency category was "Medication and Healthcare Management." This category accounted for 53.7% of the deficiencies cited, with the deficiencies primarily being rated as high risk or critical risk. These deficiencies include issues such as proxy caregiving, healthcare plans and risk mitigation, medication administration, nursing oversight, responding to a change in condition, and bowel tracking, monitoring, and intervention.
3|Page

PURPOSE AND SCOPE OF THIS REPORT
This is the tenth annual report on mortality, mortality trends, and related information pertaining to individuals on NOW and COMP waivers. The report focuses on an analysis of mortality data and findings from DBHDD's mortality review process for calendar year 2023. Reports are produced annually and cover the prior calendar year of January 1 through December 31. A description of the method and the analysis conducted in the report can be found in Appendix A. Several considerations are provided for reading and interpreting the findings from this report. Although DBHDD considered the inclusion of other states' findings, given the differences in waiver programs, obligations of the various state agencies, and other state-specific issues, it is difficult to compare mortality rates or draw conclusions between states. Therefore, this report will only present findings for individuals on Georgia NOW and COMP waivers.
4|Page

CAUSES OF DEATH AMONG THE INTELLECTUAL AND DEVELOPMENTAL DISABILITY WAIVER POPULATION
The State of Georgia has a mixed coroner/medical examiner system, making the gathering of information concerning causes and manners of death more difficult than if there were a single statewide system. The state has no uniform method for death reporting (i.e., categorizing the causes of death), and information provided on death certificates varies. Due to this lack of uniformity, it is difficult to aggregate causes of death, and the reliability is somewhat questionable because many death certificates are not completed by medical professionals. Currently, the causes of death are identified by DBHDD through one of the following means: the autopsy report, if an autopsy was conducted; the death certificate issued by the Georgia Department of Public Health's Division of Vital Statistics (if available); the medical examiner or coroner's report (if available); or as reported by law enforcement, the physician, or the individual's family.
DBHDD's process presents an aggregate of all underlying causes of death listed on the death certificate following the methods outlined by the Centers for Disease Control and Prevention (CDC).3 Using CDC direction to create a comprehensive examination of the issues and concerns leading to death in the intellectual and developmental disability population, all underlying causes of death listed on the available death certificates were combined and weighted equally. Modes of death were excluded if present. As stated in the CDC's "Instructions for Classifying the Underlying Cause of Death, 2017" (2017, p. 2):
A death often results from the combined effect of two or more conditions. These conditions may be completely unrelated, arising independently of each other or they may be causally related to each other, that is, one cause may lead to another which in turn leads to a third cause, etc.
This method helps to encompass comorbid conditions that could be missed when assigning a singular cause of death.
A summary of the causes of death, as recorded within death certificates follows (Table 1).

3 (2017). Retrieved from https://www.cdc.gov/nchs/data/dvs/2a_2017.pdf. Accessed January 10, 2020.

5|Page

Rank 1 2 3 4 5 6 7 8 9
10

Table 1: Leading Causes of Death4

U.S. (CY2023) provisional5 Heart Diseases (28.3%)
Malignant Neoplasms (25.4%)
Unintentional Injuries (9.2%)
Cerebrovascular Diseases (6.7%)
Respiratory Diseases (6.0%)
Alzheimer's Disease (4.7%)
Diabetes Mellitus (3.9%)
Renal Disease (2.3%)
Chronic Liver Disease (2.2%)
COVID (2.1%)

Georgia (CY2023)6
Heart Diseases (29.6%)
Malignant Neoplasms (19.4%)
Nervous System Diseases (10.4%)
Respiratory Diseases (9.2%)
Unintentional Injuries (8.9%)
Endocrine, Nutritional, and Metabolic (5.2%)
Infectious and Parasitic Diseases (4.2%)
Digestive System Diseases (3.9%)
Reproductive and Urinary System Diseases (2.9%)
Mental and Behavioral Disorders (2.6%)

DBHDD (CY2023) Heart Diseases (19.3%)
Respiratory Diseases (16.0%) Disability (8.4%)
Malignant Neoplasms (7.3%) Sepsis (7.1%)
Endocrine, Nutritional, and Metabolic (5.7%)
Cerebrovascular Diseases (5.4%)
Aspiration Pneumonia (5.2%)
Pneumonia (4.9%)
Unintentional Injuries (3.0%)

4 Data shown for the U.S. and Georgia include all ages, while the data shown for DBHDD's IDD population are limited to adults only. The information presented above is provided for descriptive purposes only. Due to the lack of consistency in categorizing the causes of death and expertise of those completing the death certificates, readers are strongly cautioned against drawing conclusions based on this information. To use this information to make conclusions or recommendations regarding system or practice changes, it is necessary to conduct further exploration into available information about individual cases or groups of cases. It is important to understand and consider information, such as the underlying causes of death, the circumstances of the death, the medical care provided prior to the death, co-morbid conditions, and potentially important early detection, screening, and preventive care practices.
5 https//wonder.cdc.gov.
6 Data for Georgia mortality is from the Georgia Department of Public Health (https://oasis.state.ga.us/oasis/webquery/qryMortality.aspx). Georgia now includes COVID in its infectious and parasitic diseases category. It is no longer a standalone category.
6|Page

As in previous years, several of the 10 leading causes of death for general populations of the United States or Georgia were also found to be leading causes of death in the IDD population. Common causes of death for general and IDD populations included the following six:
Heart diseases Respiratory diseases Unintentional Injuries Endocrine, Nutritional, and Metabolic diseases Cerebrovascular diseases Cancer Four of the 10 leading causes of IDD deaths in CY2023 were not common to the general population: Sepsis Disability Pneumonia Aspiration pneumonia That "disability" is listed as a leading cause of death is peculiar, as "disability" typically is not considered to be a fatal condition or cause of death, though it often is included as a cause of death on death certificates. It is important to note the prevalence of disability being listed as a cause of death on death certificates. This likely is an artifact of using causes of death from death certificates, complicated by the limitations of Georgia's mixed coroner/medical examiner system.
7|Page

IDD MORTALITY DURING CY2023
This section contains information on deaths reported to DBHDD among the IDD waiver population during CY2023. Appendix A describes the method used to collect and analyze information and data contained in this section. As has been noted in previous versions of this report, eligibility and enrollment criteria are not consistent across states, and generalizations and comparisons may lead to insupportable conclusions. Considering these caveats, this report will consider only DBHDD's data. During CY2023, there were 237 deaths reported for the 13,916 waiver population served. This is a mortality rate of 17.03 deaths per thousand, or 1.7%, which is explored in further detail in the pages that follow.
AGE AND MORTALITY
The average ages of death in CY2021 and CY2022 were 54.60 and 52.50 respectively. The average age of death in CY2023 was 54.34. As in CY2021 and CY2022, mortality rates increased with increasing age as displayed in Figure 1. For additional statistical context, please see Appendix F, Table 4. One can see from the graphic that mortality rates increased with age across the entire age range, and the mortality rates began increasing more dramatically after ages 55-64.
Figure 1: Mortality Rates by Age Category, CY2020-CY2023

8|Page

This report's findings were supported by other research7 which found that mortality rates tend to increase with increasing age, such that younger groups had lower mortality rates, and significant increases in mortality rates were typically found to begin at 45-54 and increased dramatically with increasing age. For the U.S. population, mortality rates also increase more rapidly with increasing years after about 55 years of age.

HEALTH RISK AND MORTALITY

The Health Risk Screening Tool (HRST) 8 is a standardized mechanism designed to evaluate an individual's susceptibility to potential health risks. It aids in the early detection of health deterioration and guides healthcare providers in determining the need for further assessments, evaluations, services, or modifications to the individual's care plan.

The HRST consists of 22 rating items categorized into five health domains: Functional Status, Behavior, Physiological, Safety, and Frequency of Services. Each item is assigned a weighted score, with most items scoring from zero to four, except for Requirements for Licensed Intervention (Item Q), which is scored either zero or four.

The HRST is specifically designed to identify and quantify health and behavior risks. The scored risk dimensions and other detailed information can be found in Appendices D and E of this report. After scoring each rating item, the HRST generates Health Care Levels (HCLs) that correspond to the individual's overall risk. These levels range from Low Risk (HCL 1) to the most severe risk category, Highest Risk (HCL 6).

Table 1: HRST Health Care Levels

HCL

Description

1

Low Risk

2

Low Risk

3

Moderate Risk

4

High Moderate Risk

5

High Risk

6

Highest Risk

Consistent with previous years, there was a statistical association between HCL and mortality rate in CY 2023. Individuals with lower HCLs (1-3) had a group mortality rate (7.9 deaths per 1,000) that was below the population mortality rate in CY 2023 (17.03 deaths per 1,000). Conversely, individuals with higher HCLs (4-6) had a group mortality rate (40.9 deaths per 1,000) that significantly exceeded the overall population mortality rate. Results from previous years have consistently indicated that a two-point increase in HCL is associated with a significant increase in

7 National Vital Statistics Report, Vol. 68 No. 9, June 24, CY2019, p. 8.
https://www.cdc.gov/nchs/data/nvsr/nvsr68/nvsr68_09-508.pdf, accessed March 13, 2020.
8 IntellectAbility. (n.d.). Understanding the HRST and SIS-A. Retrieved from https://support.replacingrisk.com/portal/en/kb/articles/understanding-the-hrst-and-sis-a

9|Page

mortality. This underscores the importance of the HRST in predicting health outcomes and guiding healthcare interventions. Figure 2 presents a graphic illustration of the association between HCL and mortality.
Figure 2: Crude Mortality Rates by HCL, CY2020-CY20239

Additional data on crude mortality rates by HCL, including statistical significance can be found in Tables 5 and 6 of Appendix F.
THE CENTRAL IMPORTANCE OF AGE AND HEALTH RISK
Health risk and age are important factors that need to be considered when investigating mortality. Within the IDD population, high-level risk tends to be present across all age categories, as well as varying degrees of lower-health risks across all age categories. The relationship between health risk and age is not uniform. HCLs are distributed similarly within each age group. Correlations between age (both as continuous and ordinal variables) indicate the association between HCL and age is weak (Pearson's r = 0.047, p < .00001). Although significant, the strength of the

9 The horizontal gray line indicates the crude mortality rate (17.03 per 1,000) for the overall IDD population.

10 | P a g e

association between age and health risk is small, which indicates that, for this population, health risk and age are not necessarily meaningfully associated. Therefore, one would also expect that if health risk and age were related to mortality, these variables would have independent (not interactive) effects.
Data analysis to this point has examined variables as they individually relate to mortality. However, it also is important to consider all variables of interest at once to determine the individual effect of each variable on the occurrence of death, while controlling for the influence of other variables. Analyses considered if and how age, gender, region, waiver type (NOW vs. COMP), current living situation, intensity of residential setting, and health risk (using HCL) were associated with mortality to determine which variables may be of key importance. Such associations were examined using logistic regression10.
While some areas of the differing variables demonstrated more significance than others, all nonsignificant variables were removed from the final model, leaving only age and HCL. Gender, region, and intensity of residential setting were not significantly related to mortality in CY2023. These logistic regression results have remained consistent over time.
This analysis presented findings and observations based on a statistical analysis of all adults with a primary IDD diagnosis who received services funded by NOW/COMP waivers during CY2023. Statistical analyses are useful for identifying variables and trends that are associated with mortality, which provide information for improvement of service quality. It is worth noting that, among the CY2023 IDD population, death was a relatively rare outcome. Large increases in odds (such as with the upper values of HCL and age) do not necessarily mean that individuals with these attributes were in great danger of death; it only means that people in those groups were more likely than others to experience death. It is also worth noting that statistical association does not indicate causation.
10Several advantages of using logistic regression exist. First, logistic regression allows one to determine the association of a variable without the influence of other variables. For example, logistic regression analysis about age pertains only to the effects of age and mortality without the effect of other variables. In this way, each variable is risk-adjusted so that the effects of other variables do not affect it. Another advantage is that logistic regression can be used to determine the importance of each variable and can be easily interpreted using odds ratios. An odds ratio is a measure of association between a variable and an outcome occurring. The odds ratio represents the odds of death occurring given a particular event or condition compared to the odds of death occurring in the absence of that variable.
11 | P a g e

HRST RATING ITEMS AND MORTALITY
Next, the analysis focused on identifying the top four HRST rating items out of a total of 22. These were selected based on their significant impact on the likelihood of mortality. A binary logistic regression11 was used to quantify the relationship and identify the HRST Rating Items that were more likely to influence mortality among the IDD population. To learn more about the binary logistic regression and its outputs, see Table 7 in Appendix F.
The results suggest that hospital admissions, clinical issues, toileting, and use of psychotropic medication are considered the top 4 HRST rating items in terms of their significant impact on mortality risk. The odds ratio results suggest that for each unit increase in the hospital admission score, the probability of mortality increases by 36%. Similarly, for each unit increase in the clinical issues score, the probability of mortality increases by 29%. For each unit increase in the toileting score, the probability of mortality increases by 24%. Lastly, for each unit increase in the psychotropic medication score, the probability of mortality increases by 14%.
The entire model explains 15% of the variations in mortality using these 22 HRST Rating Items. This suggests that while these items are important, other factors not included in this model may also play a significant role in determining mortality rates.
COMMUNITY MORTALITY REVIEW COMMITTEE RELATED DEFICIENT PRACTICE ANALYSIS
BACKGROUND
DBHDD investigates expected deaths, unexpected deaths, suicides, deaths of enrolled individuals, and homicides of individuals receiving services by or through DBHDD community providers. DBHDD's Community Mortality Review Committee (CMRC) uses a standard process to conduct interdisciplinary reviews of investigated deaths. The purpose of the mortality review is to identify opportunities to reduce morbidity or mortality and evaluate and provide information that may improve the quality of services. The overall goals of the mortality review are to provide insight into the way the DBHDD system works; share lessons and learn from an
11 Logistic regression was chosen for this analysis due to several key advantages. To begin with, it allows an examination of the association of each HRST rating item with mortality, independent of other variables. For instance, the effect of `hospital admissions' on mortality is analyzed without the influence of other rating items. This ensures that each variable is risk-adjusted, and its effect is not confounded by others. Another advantage of logistic regression is its ability to identify the importance of each rating item. It quantifies the impact of each variable on the likelihood of mortality, which is crucial in our study to identify the top four HRST rating items. Finally, logistic regression results can be interpreted using odds ratios, a measure of association between a rating item and mortality. The odds ratio represents the odds of mortality given a particular rating item score compared to the odds of mortality in the absence of that score. This interpretability is crucial in this study as it allows us to quantification of the risk associated with each rating item.
12 | P a g e

individual's death; discover if the same or similar situations may affect others served; assist in prevention or mitigation of future harm; and improve overall quality of care.
At a minimum, DBHDD requires providers to correct deficient practices that have the potential for causing harm, which include moderate-, high-, and critical-risk deficient practices. Deficiencies are tracked in DBHDD's incidents and investigations application, Image. Among other things, this database maintains information about deficient practices, entities cited, and categorization of the deficiencies (e.g., low, moderate, high, or critical risk). More information about the deficiency risk determinations can be found in DBHDD policy 13-101, Corrective Action Plan Management12.
CLASSIFICATION OF DEATHS
Community providers report the deaths of individuals receiving services by or through community providers. Each death is classified as one of the following types:
Expected Death: Cause of death is attributed to a terminal diagnosis or diagnosed disease process identified more than 30 days before the date of death, where the reasonable expectation or outcome is death and the individual was receiving residential or community living support services, in the company of or onsite at a community provider or discharged/transferred from a community provider within 30 days of the death.
Unexpected Death: Cause of death is not attributed to a terminal diagnosis or diagnosed disease process where the reasonable expectation or outcome is death and the individual was receiving residential or community living support services, in the company of or onsite at a community provider or discharged/transferred from a community provider within 30 days of the death.
Suicide: Self-inflicted death of an individual and the individual was enrolled in community services or discharged/transferred from a community provider within 30 days of the death.
Homicide: Injury inflicted on an individual resulting in death and the individual was enrolled in community services or discharged/transferred from a community provider within 30 days of the death.
Death of an Enrolled Individual: Death of any individual enrolled in services, excluding any other death incident type.

12 https://gadbhdd.policystat.com/policy/12697461/latest

13 | P a g e

A health and safety risk review is conducted to determine if a referral will be made for investigation as outlined in DBHDD policy 04-118, Investigating Deaths and Other Incidents in Community Services13. In CY2023, there were 97 investigations of deaths of waiver individuals.
Figure 3 Classification of Investigated Deaths CY2023
STATEWIDE ANALYSIS OF NUMBER AND TYPE OF DEFICIENT P R A C T I C E S 14
The analysis of deficiencies presented below is based on provider deficiencies entered into the Image system that were related to deaths reviewed by CMRC. In CY2023, 645 deficiencies were entered into Image, along with an assigned risk level and category.
Figure 4 CY2023 Deficiencies by Risk Level
13 https://gadbhdd.policystat.com/policy/13421590/latest 14 Due to small sample sizes, statistical analysis is not advisable. The reader is cautioned from generalizing findings and observations from the analyses above and below to the DBHDD intellectual and developmental disability population.
14 | P a g e

Figure 5 CY2023 Deficiencies by Category
15 | P a g e

In CY2023, the most common provider deficiencies by category and risk level were "Medication and Healthcare Management." This category accounted for 53.7% of the deficiencies cited. Closer examination of the types of deficiencies in this category, by risk level, shows the following:
16 | P a g e

Figure 6 CY2023 Medication and Healthcare Management Deficiencies by Sub-Category
The top 3 deficiency types within Medication and Healthcare Management are Proxy Caregiver, Medication Management, and Healthcare Plan and Risk Mitigation. These deficiencies include the following issues: Proxy Caregiver
Lack of documentation of staff knowledge and skills to perform health maintenance activities
Not having a written informed consent or plan of care Medication Management
Not having prescriptions available for review Not documenting medication administration in accordance with standards of care Delayed medication procurement Not administering medications as ordered Healthcare Plan and Risk Mitigation
17 | P a g e

Not having a healthcare plan or risk mitigation document when required Not implementing the healthcare plan or risk mitigation document as instructed Lack of documentation of staff training It is also noted that these deficiencies are primarily high or critical risk which is due to the nature of the deficiency. Deficiencies that are healthcare related are more likely to negatively impact an individual's health, and therefore, are risk rated higher.
18 | P a g e

KEY FINDINGS
Below is a summary of the key findings identified in the CY2023 Mortality Report: The CY2023 DBHDD NOW/COMP waiver mortality rate was 17.03 deaths per 1,000 individuals, which is lower than the 2 previous calendar years. Increasing age was significantly associated with mortality. Increasing health risk was significantly associated with mortality. Mortality increased markedly for individuals in the 55-64 and 65-74 age groups. Increased risk of mortality due to increasing age is also found in the general U.S. and Georgia populations. The top four HRST rating items - hospital admissions, clinical issues, toileting, and psychotropic medication - are significant predictors of mortality among the IDD population. Each unit increase in these scores correspondingly increased the probability of mortality: 36% for hospital admissions, 29% for clinical issues, 24% for toileting, and 14% for psychotropic medication. These four items, among the total 22 HRST Rating Items, accounted for 15% of the variations in mortality as indicated by a Pseudo R2 value of 0.15. The analysis of mortality rates by risk level reveals a strong correlation between the level of risk and the crude mortality rate. This correlation is evident in the significant increase in mortality rates as we move from low to high risk levels. Furthermore, the proportion tests provide statistical evidence of this correlation, confirming significant differences in mortality rates across all risk levels.
Six of the 10 leading causes of death for general populations of the United States or Georgia were also found to be leading causes of death in the IDD population. Common causes of death for general and IDD populations included the following six:
Heart diseases Respiratory diseases Unintentional Injuries Endocrine, Nutritional, and Metabolic diseases Cerebrovascular diseases Cancer Four of the 10 leading causes of IDD deaths in CY2023 were not common to the general population: Sepsis Disability Pneumonia Aspiration pneumonia
In CY2023, the most common provider deficiencies by category and risk level were "Medication and Healthcare Management." This category accounted for 53.7% of the deficiencies cited, with the deficiencies primarily being rated as high risk or critical risk. These deficiencies include issues
19 | P a g e

such as proxy caregiving, healthcare plans and risk mitigation, medication administration, nursing oversight, responding to a change in condition, and bowel tracking, monitoring, and intervention.
20 | P a g e

APPENDIX A: METHOD FOR MORTALITY REVIEW AND ANALYSIS
This mortality report analyzes information on individuals and deaths reported to DBHDD that meet the following criteria:
At least 18 years of age during the calendar year of review Primary diagnosis of an intellectual or developmental disability Medicaid waiver recipient (NOW or COMP)
This report does not include data for individuals under the age of 18. Deaths for children and adolescents are analyzed on a case-by-case basis and not included in these statistical analyses due to potential differences between children and adults and the small sample size of children.
Individuals who moved between the NOW/COMP waiver during CY2023 were categorized into the waiver in which they were last enrolled.
The data used to calculate mortality rates per 1,000 people by age group and type of residence were supplied by IDD Connects and Image. IDD Connects data also included identifying, demographic, and payer information, as well as residential setting. Health risk information was extracted from HRST and IDD Connects. Death and incident data were extracted from Image.
For these analyses, the following information was included: Region (IDD Connects) Medicaid number (IDD Connects) Date of birth (IDD Connects) Date of death (Image and IDD Connects) Residential setting (IDD Connects) Cause of death (if known) (death certificates) Whether death was referred for investigation (Image) Whether a mortality review was completed (CMRC) Health risk scores (HCLs from Health Status Risk Screening Tool and IDD Connects) Rating Items (HRST assessment tool) Tracking of deficient practices and corrective action plans (Image)
Due to the large number of statistical comparisons, the statistical significance level was set at = 0.01. Setting = 0.01 as the significance level is to compensate for finding significance due to increased chances afforded by multiple comparisons.
CRUDE MORTALITY RATE
The crude mortality rate is a measure of how many people out of every thousand served by DBHDD died within the calendar year. It is determined by multiplying the number of people who
21 | P a g e

died during the year times one thousand and dividing this by the total number of people served in the NOW/COMP waiver program during the same year. The crude mortality rate can be useful when comparing deaths across populations of varying sizes. Caution should be used when comparing mortality rates across unlike methods and populations.
In the analysis of crude mortality rates, proportion z-test was employed to compare mortality rates across different age categories (Table 4), health care levels (Table 5) and risk levels (Table 8).
Deaths were included, regardless of death category, for all population-eligible adults who died in CY2023.
ANALYSIS AND MEASURES
Analysis was conducted using R Studio, an integrated development environment (IDE) tailored for the R programming language. A variety of statistical techniques were employed to comprehensively assess the relationships between variables and mortality, including significance testing through the Chi-Square test. After conducting the chi-square test with an alpha level set at 0.01, it was determined that gender, region and marital status variables were not significantly associated with mortality, as their p-value exceeded alpha = 0.01. Consequently, these variables were removed from further analysis. The remaining variables were then examined using measures of association such as Cramer's V, and assessment of multicollinearity using variance inflation factor (VIF).
To enhance the interpretability of coefficients and odds ratios, variables underwent appropriate transformations as needed. The variables utilized in logistic regression were categorized as follows:
Death (outcome): Factor with levels, No Death (0) and Death (1) Age: Continuous Numeric (ranging from 18 to 93; Categorical (18-24, 25-34, 35-44, 45-
54, 55-64, 65-74, 75-84, 85+ Health Risk (HRST Health Care Level [HCL]): Continuous Numeric (ranging from 1-6; Table
5); Ordered Factor (HCL 1, HCL 2, HCL 3, HCL 4, HCL 5, HCL 6 Intensity of Residential Setting: Factor with levels, Lower Intensity (0) (independent
apartment/home; live with family/relative/caretaker/friend/other; other); Higher Intensity (1) (personal care home; community living arrangement; host home) Rating Items: Continuous Numeric (ranging from 0 to 4, with the exception of Item Q (Requirements for Licensed Intervention), which can only be rated as either 0 or 4); Rating Items variables are Eating, Ambulation, Transfer, Toileting, Clinical issues, Behaviors, Selfabuse, Aggression, physical, Emergency drugs, psychotropic medications, Physiological, Gastrointestinal conditions, Seizures, Anticonvulsant medication, Skin breakdown, Bowel function, Nutrition, Requirements for licensed interventions, Safety, Injuries, Falls, Frequency of services, Professional health services, Emergency visits, Hospital admissions
22 | P a g e

All variables were entered into regression models individually, and the remaining variables were examined for significant association with death. Variables that were indicated as not being significantly associated with death were removed, and the model was recomputed. Those variables that were indicated as significantly associated with death were retained in the model. This process continued until only significantly associated variables with death remained. Finally, the model was examined for meaningful relationships and interpretation.
23 | P a g e

APPENDIX B: NOW/COMP POPULATION DEMOGRAPHICS
CHARACTERISTICS OF THE INTELLECTUAL AND DEVELOPMENTAL DISABILITY WAIVER POPULATION
Below is a brief demographic description of the CY2023 IDD waiver population: The total number of unduplicated IDD individuals with active NOW/COMP waivers in CY2023 was 13,916. These individuals were aged 18-94, with a mean age of 42.30. Of these, 60.4 percent were male, and 39.6 percent were female. Region 3 (25.4%) was the most populous region, followed by Region 1 (23.3%), Region 2 (16.9%), Region 6 (13.2%), Region 5 (11.2%), and Region 4 (9.9%). Most of the population had COMP waivers (68.8%) as opposed to NOW waivers (31.2%).
More information about the characteristics of the population can be found on the following page (Table 3).
24 | P a g e

Table 3: Characteristics of the Adult IDD Waiver Population, CY2021-CY202315

Characteristic

2021

n

%

2022

n

%

2023

n

%

Age

18-24

1,186

8.6

1,087

7.8

1,167

8.4

25-34

3,968 28.7 3,976 28.7 3,953 28.4

35-44

3,215 23.2 3,301 23.8 3,424 24.6

45-54

2,270 16.4 2,274 16.4 2,229

16

55-64

1,969 14.2 1,940 14.0 1,877 13.5

65-74

960

6.9

985

7.1

997

7.2

75-84

244

1.8

264

1.9

246

1.8

85+

29

0.2

24

0.2

23

0.2

Gender

Male

8,293 59.9 8,338 60.2 8,405 60.4

Female

5,544 40.1 5,510 39.8 5,506 39.6

Unknown

4

0.0

3

0.0

5

0.0

Region

Region 1

3,231 23.3 3,193 23.1 3,238 23.3

Region 2

2,344 16.9 2,335 16.9 2,349 16.9

Region 3

3,417 24.7 3,478 25.1 3,539 25.4

Region 4

1,363

9.8

1,358

9.8

1,380

9.9

Region 5

1,637 11.8 1,631 11.8 1,564 11.2

Region 6

1,822 13.2 1,816 13.1 1,832 13.2

Region 99/Unknown

27

0.2

40

0.3

14

0.1

Waiver Type

NOW

4,680 33.8 4,448 32.1 4,339 31.2

COMP

9,161 66.2 9,403 67.9 9,577 68.8

Residential Setting

Lower Intensity

9,419 68.1 9,363 67.6 9,500 68.3

Higher Intensity

4,422 31.9 4,488 32.4 4,416 31.7

Race

American Indian/Alaskan Native

3

0.0

5

0.0

4

0.0

Asian

97

0.7

103

0.7

124

0.9

Black/African American

6,004 43.4 6,021 43.5 5,994 43.1

Multiracial

90

0.7

97

0.7

102

0.7

Native Hawaiian/Other Pacific Islander

12

0.1

13

0.1

12

0.1

Other Single Race

290

2.1

291

2.1

301

2.2

Unknown/Refused

502

3.6

566

4.1

708

5.1

White/Caucasian

6,843 49.4 6,755 48.8 6,671 47.9

Total

13,841 100 13,851 100 13,916 100

15 Shown for each characteristic are totals and percentages. Total percentages may not total to 100% because of rounding.
25 | P a g e

APPENDIX C: DBHDD SAMPLING PROCEDURE
DBHDD carefully considers information and data to analyze to answer analytical questions. High quality, valid information and data are the basis of useful, practical, and valid research findings and conclusions. Ideally, analysis occurs from data on an entire population, and DBHDD strives to accomplish this when feasible; this produces maximum validity. However, when data on the entire population are not available or feasible, then DBHDD carefully considers how the analytic data sample is built, as the sampling procedure has great impact on the quality, validity, and generalizability of research findings.
DBHDD's sampling procedure proceeds in the following manner:
First, when available, DBHDD utilizes data on the full population under study (e.g., all individuals who received services within a given period such as calendar or fiscal year).
Second, if some individuals within the full population have missing data for variables being used for analysis, DBHDD considers widely-accepted procedures to address missing data. For example, individuals with missing data typically are excluded from analysis using listwise deletion,16 resulting in a subset of the full population. DBHDD may consider other theoretically-sound methods and procedures to understand or address missing data.17
Third, in some cases, DBHDD utilizes some form of random sampling18 (e.g., a random subset of providers or events that occurred). For this approach to be valid, one must be able to define the entire population from which it is being drawn, and each unit (e.g., individual, situation, etc.) must have an equal chance of being included in the sample. This method is unbiased, and the resulting sample is representative of the full population under study.
Fourth, DBHDD also occasionally makes use of purposive sampling, a non-probability sampling method. This method is typically reserved for specific instances (e.g., identifying when a situation occurred, selecting specific cases, identifying specific errors, etc.). Purposive sampling is a selective, non-probabilistic method, and purposive sampling is not representative of the full population under study; therefore, findings or results based on purposive sampling are not generalizable to the full population, rather only to the cases from which data were sampled.
16 Listwise deletion is a method for handling missing data, whereby an entire record is excluded from analysis if any single value is missing. 17 Sensitivity analyses are conducted to evaluate the pattern of missing data, wherein missing data are determined to be either missing completely at random (MCAR) or missing at random (MAR). Data are determined to be MCAR when the probability of missing data on a variable is unrelated to any other measured variable and is unrelated to the variable with missing values itself. Data are determined to be MAR when the missingness can be explained by variables that do not contain missing values. 18 The leading component of simple random sampling is that every case (e.g., individuals or providers) has the same probability of being selected for inclusion in analysis.
26 | P a g e

DBHDD considers sample sizes carefully and analytically to create empirical samples large enough to have sufficient statistical power to detect associations or differences and allow valid inferences to be drawn from and generalized about the population being studied. When the entire population is not used in the analyses, DBHDD relies upon practical application of scientific, statistical, and theory-based techniques and procedures to yield inferences about the population based on a sample smaller than the population that increases the chances that the sample has sufficient size and power to identify and draw valid conclusions from the data and generalize to the larger system.
27 | P a g e

APPENDIX D: HRST DOMAINS

Risk Dimension Functional status
Behaviors
Physiological Safety
Frequency of services

Item Letter (A-V) A B C D E F G H I J K L M N O P Q R S T U V

Item Topic Eating
Ambulation Transfer Toileting
Clinical issues affecting daily life Self-abuse
Aggression towards others and property Use of physical restraints Use of emergency drugs
Use of psychotropic medications Gastrointestinal conditions Seizures Anticonvulsant medication Skin breakdown Bowel function Nutrition
Requirements for licensed interventions Injuries Falls
Professional health services Emergency department visits
Hospital admissions

28 | P a g e

APPENDIX E: HRST EXPANDED SCORE DESCRIPTORS

Score 0 1 2 3
4

Functional Status - Eating (Item A) Expanded Explanation
Eats independently: May require simple adaptive equipment (hand splint, special eating equipment) but is able to eat without assistance/supervision. Individuals needing help only to cut food into regular, bite-sized pieces still rate a 0. Those who require altered food/fluid textures require a higher score. Requires INTERMITTENT physical assistance and/or verbal prompts to eat: May need occasional physical help due to physical limitation or occasional verbal prompts due to issues with attentiveness or behavior.
Requires CONSTANT verbal and/or physical assistance to complete a meal: Has difficulty attending to task or may have motor limitations which require constant physical and/or verbal assistance. No issues with safety or swallowing.
Requires constant assistance or other mealtime intervention to eat SAFELY OR has a feeding tube but maintains some level of oral intake: May have difficulty coordinating breathing/swallowing while eating, dangerous behaviors or other conditions which impair their ability to eat safely. Unable to obtain adequate calories and fluids without assistance. Interventions are required (specific positioning support, eating devices, presentation techniques and/or modifications in food/fluid consistency). May have enteral (feeding) tube but maintains some level of oral eating.
Receives ALL nutrition/hydration via other than oral routes (gastrostomy, jejunostomy or nasogastric tube, or total parenteral nutrition-TPN): Unable to swallow safely OR has other issues requiring other than oral feeding procedures. Individuals who receive food by mouth against physician orders still qualify for a score of 4.

Score 0 1 2 3
4

Functional Status - Ambulation (Item B) Expanded Explanation
Ambulates independently in ALL settings: May use a walker or other means of support but does so independently in all settings without problems of safety.
Walks with minimal supervision: Requires the support of another person in close proximity in one or more settings. The primary issue is safety during ambulation.
Predictably dependent on wheelchair for at least some mobility needs: May or may not have the ability to walk in some settings. Individuals are able to use their upper body strength for repositioning AND have the ability to independently maintain trunk alignment. Able to recognize the need to change positions on a consistent basis. Requires mechanical assistance to maintain upright, seated position in wheelchair. Needs assistance to change position or shift weight: Unable to walk. Able to be placed in an upright sitting position but cannot maintain a seated posture without outside mechanical support (specialized positioning equipment, adaptive wheelchair, etc.) or assistance. Needs assistance to reposition OR may not recognize need to reposition on a consistent basis. May need assistance to propel wheelchair.
Disability prevents sitting in an upright position: UNABLE to flex the hips to at least 45o OR unable to approach reasonable alignment of the head, shoulders, and pelvis. Due to degree of musculoskeletal deficits or deformity has limited positioning options.

29 | P a g e

Score 0 1 2
3
4

Functional Status - Transfer (Item C) Expanded Explanation
Transfers independently in ALL settings: May require verbal prompts, but no physical assistance. Needs someone to supervise the transfer for safety: May need minor hands-on assistance, but able to bear their own weight and transfer safely in all settings. Needs physical assistance of 1 person to transfer or change position: Individual is able to participate in transfers with the assistance of one other person managing a portion of their weight OR is completely dependent for lifting assistance but weighs less than 50 pounds. Needs physical assistance of 2 people to transfer or change position: Individual is able to participate in transfers with the assistance of two other persons managing a portion of their weight OR is completely dependent for lifting assistance and weighs between 50 and 75 pounds. Needs lifting equipment or specialized procedures to safely transfer OR has a history of a fracture caused by a transfer procedure: Requires specialized lifting equipment due to inability to participate in transfers. Includes individuals who weigh more than 75 pounds and are completely dependent for transfers, whether or not they actually use lifting equipment. May need range of specially designed positions due to severe spasticity, history of bone fragility, potential for injury due to size, or due to degree of physical deformity OR has had a history of a fracture caused by a transfer procedure at some time in their life. Note: The influence of this item on the HCL extends beyond 12 months, because it relates to "history of".

Score 0 1 2
3
4

Functional Status - Toileting (Item D) Expanded Explanation
Independently accomplishes ALL toileting tasks: No assistance required or appreciated. Minimal supervision or adaptation required: May require reminders or some verbal and physical assistance to maintain hygiene or manage clothing adjustments. May require adaptations to restroom facilities (grab bars or built-up commode seat) Beyond this, minimal assistance is necessary. Continent of bladder and bowel, but constant attention is needed: Requires physical assistance to complete hygiene tasks (wiping, hand washing) and clothing repositioning. May have occasional accidents but NOT routine, predictable incontinence. Incontinent of bowel or bladder: Individual is predictably incontinent of bowel or bladder in one or more settings (nighttime, work or school settings or engages in willful incontinence.) May require scheduled toileting or use incontinence briefs. Includes infants, for whom incontinence is age appropriate. ANY use of catheterization procedures or colostomy for elimination within the past 12 months: Urinary catheterization for ANY reason or elimination via colostomy, urostomy or ileostomy within the past year.

30 | P a g e

Score 0 1 2 3 4

Functional Status - Clinical Issues (Item E)
Expanded Explanation
None, or person does not participate due to personal preference or guardian objections. No clinical restrictions: No ADLs changed or missed within the past year due to illness, behaviors, or necessary medical appointments (Full or partial day). Less than 2 days (full or partial) in a month on average due to clinical issues: Able to participate in usual activities of daily living, but participation may occasionally be interrupted by illness, behavioral or mental health issues, or may have physician appointments to monitor a diagnosed condition or receive treatment. 2 to 4 days (full or partial) in a month on average due to clinical issues: Able to participate in usual activities of daily living, but participation may be interrupted by illness, behavioral or mental health issues, or may have physician appointments to monitor a diagnosed condition or receive treatment. 5 to 10 days (full or partial) in a month on average due to clinical issues: Able to participate in usual activities of daily living, but due to chronic unstable or progressively worsening health or behavioral issues, there is a significant impact on usual activities. May be due to physician appointments to monitor a diagnosed condition or receive treatment. More than 10 days (full or partial) in a month on average or normal daily activities are completely disrupted due to intensity of clinical issues: Due to chronic, unstable, or progressively worsening health or behavioral issues participation in usual activities is severely impaired. May be ill or have physician appointments to monitor condition or receive treatment OR may be completely unable to participate in usual activities due to intensity of clinical issues.

Score 0 1 2
3
4

Behavior - Self Abuse (Item F) Expanded Explanation
No self-abuse within the past year. Minimal self-abuse, no additional consequences: Behaviors that are considered self-abusive have been identified but have not required first aid or other intervention within the past year. Self-abuse needing additional observation LESS than 2 times a month: Demonstrates behaviors that cause minor self-injury which may require treatment or other intervention but averaging to less than two interventions per month over the past year.
Self-abuse needing medical/nursing attention or other intervention 2 OR MORE times per month: Demonstrates behaviors that cause minor self-injury, which may require treatment or other intervention, but averaging two or more interventions per month over the past year. Self-injury interferes with the ability to engage in structured activities, requires increased staffing or causes extensive physical harm: May be due to an existing behavioral pattern or the result of a single, isolated incident.

Score 0 1
2
3
4

Behavior - Aggression (Item G) Expanded Explanation
No aggression within the past 12 months.
LESS than 5 incidents per month of minor aggression (verbal or physical) WITHOUT injury to others or property damage within the past 12 months. 5 OR MORE incidents per month of aggression (verbal or physical) WITHOUT injury to others or property damage within the past 12 months. LESS than 5 episodes of aggression per month WITH minor injuries to others (injuries not needing medical TREATMENT) or property damage within the past 12 months. Episodes of aggression have required increased staffing ratios, restrictive interventions OR caused serious physical harm within the past 12 months,

31 | P a g e

Score 0 1 2
3
4

Behavior - Physical Restraint (Item H) Expanded Explanation
Has NOT been physically restrained in the past 12 months. Has been physically restrained less than once per month on average in past 12 months: May include restraints used to facilitate some type of urgent medical procedure or care that without using restraint would have been impossible OR an acute behavioral event that required an immediate response. Has been physically restrained more than once per month on average in past 12 months: Restraint use would require a physician's approval. Less restrictive options would have been explored and ruled out. Use of physical restraint procedures or devices MORE than 5 times per month on average but LESS than 12 hours per day: Generally behavioral issues (hitting, biting, head-banging, etc.) that cause injury to self and/or others. May wear protective devices, including helmets to protect from injuries due to anticipated falls. Individual sustained and injury requiring medical TREATMENT as the result of application of physical restraint procedures/devices OR use of some sort of device 12 or more hours per day: Generally, has significant behavioral issues (severe and continuous tissue damage, significant aggression, causing injuries). Includes use of helmets to protect from injuries due to anticipated falls or confinement of individual to a restricted space such as a prison cell.

Score
0
1 2 3 4

Behavior - Chemical Restraints (Item I) Expanded Explanation
Has NOT received additional medications to control mood, mental status, or behavior in the past 12 months: May have behavior issues but coping skills and behavioral intervention are sufficient to help the individual calm down without the necessity of drug/medication administration. Received pre-sedation before any medical or dental appointment in the past twelve months: Anxiety/pain threshold has resulted in use of drugs prior to medical or dental procedure.
Has received medications to control mood, mental status, or behavior 1 time in last 12 months. Has received medications to control mood, mental status, or behavior 2-3 times in last 12 months. Has needed medications to control mood, mental status, or behavior 4 or more times in last 12 months.

Score 0 1 2 3
4

Behavior - Psychotropic Meds (Item J) Expanded Explanation
Has NOT received medication to control behavior or a psychiatric disorder within the past year. Receives 1 medication not associated with or known to cause tardive dyskinesia (TD) to control behavior or psychiatric disorder. Medication dosage has NOT CHANGED within the past year. Receives 2 medications not associated with or known to cause tardive dyskinesia (TD) to control behavior or psychiatric disorder. Medication dosage has NOT CHANGED within the past year: May or may not be taking a traditional psychotropic drug, but is taking medication (e.g., Benadryl, Inderal, Tegretol) for identified behavior or psychiatric diagnosis.
Receives 3 or more behavioral or psychiatric medications not associated with or known to cause tardive dyskinesia (TD) OR psychotropic medication type or dosage has been changed in the past year: On 3 or more medications to control behavior or psychiatric disorder OR receives ANY medication to control behavior or psychiatric disorder with at least one change in type or dosage in past year. Individuals on a drug tapering program will remain a 3 for one year after the medication is discontinued. Has received one or more medications associated with or known to cause Tardive Dyskinesia within the past year: Includes medications such as metoclopramide (Reglan), even when they are not used for psychiatric purposes.

32 | P a g e

Score 0 1 2
3 4

Physiology - Gastrointestinal (Item K)
Expanded Explanation None: No GI concerns within the past 12 months AND no history of GI bleed. Occasional (2 or less) episodes of GI symptoms per month in the absence of acute illness: Health is very stable. Only has an occasional episode of GI symptoms (2 or less per month). GI distress occurs with no apparent explanation. 3-6 episodes of GI symptoms per month: Occasional episodes of GI symptoms occurring 3 - 6 times per month. A documented pattern of incidents may be developing. These episodes are more likely to be associated with a disorder of the stomach or GI tract instead of an acute illness like the flu. This includes individuals who take over the counter medications for upset stomach, heartburn, or other GI symptoms. MORE than 6 episodes of GI symptoms per month, OR coughing within 1-3 hours after meals or during the night, OR hand-mouthing or PICA behaviors, OR has a history of GI bleeding OR has a current diagnosis of gastroesophageal reflux (GER) Note: The influence of this item on the HCL extends beyond 12 months, because it relates to "history of". GI condition requiring hospital admission in past 12 months OR receives more than one medication for GER: Conditions requiring hospital admission include GI bleeding, ulcerative conditions, vomiting, persistent dehydration, aspiration pneumonia, intestinal infections, bariatric surgery, gallbladder or pancreatic surgery, bowel impaction, obstruction or ileus, parasites, etc. OR individual regularly takes more than one medication (including over-the-counter medications) to control GER.

Score 0 1 2 3 4

Physiology - Seizures (Item L) Expanded Explanation
No seizure in lifetime OR more than 5 years since last seizure: Individual has never had seizures OR has a known seizure history but has not had a seizure in more than 5 years. May or may not be taking antiepileptic medication. More than 2 but less than 5 years since last seizure: Has a history of seizure activity but has been seizure-free for at least the last 2 years. May or may not be taking antiepileptic medication. Less than 1 seizure per month which DOES NOT interfere with functional activity: Seizure activity occurs less than one time per month AND does not affect the person's ability to engage in functional activities for longer than 30 minutes. Seizure activity that DOES interfere with functional activities: Seizures of any type which occur more than once a month OR seizure activity of ANY frequency that interferes with functional activities for longer than 30 minutes.
Has required hospital admission for seizures in past the 12 months: Any classification of seizure requiring a hospital ADMISSION (not just an ER visit) to treat seizure complications, diagnose or evaluate a seizure disorder or for surgery to treat a seizure disorder.

Score 0 1 2
3
4

Physiology - Anticonvulsant (Item M) Expanded Explanation
None: Has not taken antiepileptic medication within the past year. Use of SINGLE antiepileptic medication: Dosage or medication type has NOT CHANGED within the past year. Use of 2 antiepileptic medications: Dosage or medication type(s) have NOT CHANGED within the past year. Use of 3 or more antiepileptic medications OR any change in antiepileptic medication type or dosage in past 12 months OR receives valproic acid derivatives (Depakene or Depakote, etc.) in combination with any other antiepileptic medication OR receiving felbamate (Felbatol): Individuals on a drug tapering program will remain a 3 for one year after the medication is discontinued. ER visit OR hospitalization due to antiepileptic drug toxicity in past 12 months.

33 | P a g e

Score 0 1 2
3
4

Physiology - Skin Breakdown (Item N)
Expanded Explanation
No current or potential skin problems within the past year: No issues with skin integrity in the past 12 months AND no known conditions associated with increased skin vulnerability.
Red or dusky discolorations or other minor disorders of skin: Skin may be reddened or have signs of poor circulation. This may also include individuals with typical presentations of psoriasis, acne, eczema, severe dryness, or other skin issues. Individuals with diabetes mellitus or other issues associated with skin vulnerability require a higher score (3 or greater). Either currently has or has had significant disruptions of skin integrity within last 12 months OR has a history of pressure sores: Includes ANY significant wound, including surgical wounds, in individuals who do not have a known condition associated with skin vulnerability AND individuals who have had pressure sores, even if they resolved more than 12 months ago. Note: The influence of this item on the HCL extends beyond 12 months, because it relates to "history of".
Within the past 12 months a significant break in skin has developed which required MORE than 3 months to heal OR has a condition directly associated with skin vulnerability: Examples include spina bifida, spinal cord injury, nutritional compromise, low serum albumin, diabetes mellitus, continuous incontinence, self-injurious behaviors involving skin damage. Individual may NOT have had any actual issues with skin integrity in the past year.
The skin condition required recurrent medical treatment or hospitalization in past 12 months: Individuals have required hospitalization or surgery for a skin problem (invasive skin cancer, graft surgery for wounds or burns, etc.) OR have required visits to a wound care clinic, infectious disease or other specialist for a severe or potentially life-threatening skin issue.

Score 0 1 2 3
4

Physiology - Bowel Function (Item O) Expanded Explanation
No bowel elimination problems within the past year AND no history of hospitalizations for bowel obstruction or ileus Note: The influence of this item on the HCL extends beyond 12 months, because it relates to "history of".
Bowel elimination is easy to manage with diet: Receives a diet modification and/or increased fluids to assist with proper elimination.
Bowel elimination is easy to manage with diet and routine supplements: Has slight problems with constipation requiring intermittent or routine stool softener or fiber supplement.
Receives at least one medication that affects bowel motility OR regularly receives more than one supplement or medication of ANY type to treat diarrhea or constipation: Has recurrent problem with constipation or experiences episodes of intermittent diarrhea. May require suppositories, enemas or manual assessment for impaction.
Any hospitalization in past 12 months required to treat an impaction, bowel obstruction or ileus OR history of ANY hospitalizations for bowel obstruction or ileus Note: The influence of this item on the HCL extends beyond 12 months, because it relates to "history of".

34 | P a g e

Score 0 1 2 3
4

Physiology - Nutrition (Item P) Expanded Explanation
Within ideal body weight range and able to maintain weight: Requires no diet modifications, prescribed nutritional supplements or other intervention to maintain health. Individual may voluntarily take vitamins or other nutritional supplements without physician prescription or recommendation. Is slightly above or below ideal body weight range. May require extra calories or some dietary restrictions: Health is generally stable, though weight is not within ideal range (not more than 10% above or below the far ends of the ideal body weight range.) May require additional calories through supplemental products or snacks, OR may require dietary restrictions (single servings at mealtime, low fat and low-calorie foods, restricted sweets, etc.). Is well managed on a prescribed diet: Within desired weight range, but has a diet prescription for health maintenance or health concerns which have been under control for the past 12 months (low sodium, low cholesterol, etc.) This includes individuals receiving tube feeding formula who are otherwise nutritionally stable and well maintained. Has demonstrated weight instability in the past OR has an identified nutritional risk which required nutrition status monitoring within past 12 months: May have displayed unstable nutritional status episodes or trends in past 12 months which have produced health issues requiring intervention to maintain health OR is being monitored for one or more of the following:
Inability to reach or maintain desired body weight.
Unplanned changes/trends in body weight (up or down). A chronic medical condition which affects nutritional status (diabetes mellitus, anemia, low serum albumin, renal or hepatic disease, GI disorder, impaction, pressure ulcer, etc.).
Medical conditions that require monitoring and control of fluid intake levels.
Difficulty consuming adequate intake, poor appetite, or frequent meal refusals.
Food allergies or intolerance which limits intake of major food groups. Nutritional status unstable within the past 12 months: High risk with an unstable nutritional status. Required intensive nutritional intervention to address any of the following conditions: Unplanned weight loss >10% of usual weight in past 12 months.
Morbid obesity (body weight 100 pounds greater than, or twice the desired weight range or BMI >35). Hospitalization and/or treatment in the past 12 months for recurrent aspiration pneumonia, choking episodes, GI bleeding, unresolved diarrhea, vomiting, or unresolved wounds caused by pressure, diabetes, circulatory disorders, etc. Inability to consume an adequate diet due to chewing or swallowing disorder (for individuals receiving only oral intake). Gastrostomy or jejunostomy tube placement OR complications with existing enteral tube in the last 12 months.

35 | P a g e

Physiology - Requirements for Licensed Intervention (Item Q)

Expanded Scoring Descriptors

Treatments -- Includes interventions or procedures which MAY be performed independently or by unlicensed

family/staff but, by their nature, are inherently high-risk. Also includes treatments which may not, under ANY circumstances, be delegated to non-licensed personnel. Scoring is intended to be consistent from

setting to setting, regardless of policies dictating professional practice delegation. In many cases a Q-score

qualifies the person to receive 24-hour nursing services, although not all individuals require such a restrictive setting. Item is scored either 0 or 4 regardless of how many qualifying issues apply.

1 Tracheotomy that requires suction.

2 Ventilator dependent.

3

Nebulizer treatments one or more times daily: Receives medications such as Ventolin or Theophylline, by oxygen mist nebulizer at least once per day.

4

Deep suction: Requires deep suction, which means entering a suction catheter 6" or more into or below the voice box either via tracheotomy, oral or nasal routes.

5 Requires complex medication calculations for insulin given via insulin pump or injection.

Has an unstable condition that requires ongoing (usually daily or more frequent) assessment and 6 treatment by a licensed health care professional. Including but not limited to:

Medication therapy requiring intramuscular or intravenous injections using a PICC line or port, once or

more times daily.

Daily or more frequent catheterization, requiring sterile technique.

Physician ordered treatments that CANNOT be delegated to a non-licensed person such as

chemotherapy or renal dialysis.

Sterile dressing/wound treatments routinely performed only in clinical settings or by licensed

practitioners.

Individuals in acute and/or end stages of cardiac, liver, lung, or kidney disease.

End-stage terminal illness (cancer, AIDS) or persons with end-stage progressive neurological disorders

(Sanfilippo Syndrome, Multiple Sclerosis, Huntington's chorea).

7

1:1 staffing for behavioral issues: Requires 1:1 staffing 16 or more hours EACH day due to behavioral issues.

Score 0 1 2
3 4

Safety - Injuries (Item R) Expanded Explanation
No injury within the past year OR minor bruises/abrasions requiring only simple first aid: Small cuts or scratches that do not require attention beyond cleansing and simple bandaging or minor bruises, sprains or strains that do not require immobilization.
Bruises or cuts 1 or 2 times in the past year requiring first aid or nursing intervention within the past year: Injuries of any type requiring minor first aid or nursing attention (but NOT physician treatment). Bruises or cuts requiring first aid or nursing intervention occurring 3 or more times within the past year: Injuries of any type requiring first aid or nursing intervention (but NOT physician treatment) occurring 3 or more times within the past year.
Injury requiring medical TREATMENT in the past year: Sustained an injury that required treatment by a physician or in an emergency room (sutures, casting a fracture, etc.) within the past year. Injuries receiving physician evaluation as a precaution but NOT requiring treatment should receive a lower score.
Major injuries requiring hospital admission within the past year: Has documented evidence of fracture or other major trauma which required hospital admission within the past year.

36 | P a g e

Score 0 1
2
3 4

Safety - Falls (Item S) Expanded Explanation
No falls within the past year. 1 - 3 falls within the past year. 4 - 6 falls within the past year OR wears a helmet to protect from injuries due to anticipated falls from events such as seizures or narcolepsy. More than 6 falls in the past year. Any fall that resulted in a fracture or hospital admission due to injuries in the past year.

Score 0
1 2 3 4

Frequency of Services - Professional Healthcare Services (Item T) Expanded Explanation
No visits other than routine screening or health maintenance visits within the past year: Visits to licensed health care providers that did NOT identify or manage a diagnosed condition. These visits are normally only to primary health care providers and NOT to specialists. Required 2 visits per quarter on an average over the past year to health care provider(s): Visits to ANY health care providers intended to identify or manage a diagnosed condition.
Required 1-2 visits per month on average to health provider(s) OR required daily nursing services greater than 14 days continuously in past 12 months. Required 3 visits per month on average to health care providers within the past year. Required 3 visits per month to health care providers PLUS unscheduled appointments within the past year: In addition to 3 or more visits per month, unplanned visits to health care providers were required to treat acute health incidents within the past year.

Score 0 1 2 3 4

Frequency of Services - Emergency Room Visits (Item U) Expanded Explanation
No emergency room visits within the past year. Emergency room visit due to physician absence or non-emergency situation within the past year. One emergency room visit in last year for acute illness or injury. Two or more emergency room visit for acute illness or injury in the past year. Any emergency room visit in the past year that resulted in hospital admission.

Score 0 1
2 3 4

Frequency of Services - Hospital Admissions (Item V) Expanded Explanation
No hospital admissions within the past year. Hospital admission in the past year for scheduled surgery or procedure: Normally for conditions that are not deemed urgent where there is an elapsed period (days to weeks) between diagnosis and admission, including routine childbirth. Hospital admissions for acute illness or injury within the past year: Often occurs from an emergency room or physician's office with little or no elapsed time between diagnosis of the condition and hospital admission. Includes admissions to psychiatric facilities or ICFs. 2 or more hospital admissions for acute illness or injury in the past year. Admission to ICU during a hospitalization in past year: Initial hospitalization may have been for an acute illness or injury, but ICU admission may also occur as the result of scheduled or elective procedures.

37 | P a g e

APPENDIX F: ADDITIONAL STATISTICAL INFORMATION

Table 4: Mortality Rates Among the Adult IDD Waiver Population by Age Category, CY2023

Age Category

Population

Deaths (#)

Deaths (%)

Crude Mortality
Rate

Significance

18-24

1,167

10

4.2%

8.6

--

25-34

3,953

28

11.8%

7.1

NS

35-44

3,424

34

14.3%

9.9

NS

45-54 55-64 65-74

2,229 1,877 997

35

14.8%

15.7

NS

54

22.8%

28.8

|z|= 2.8644; p = .00209

53

22.4%

53.2

|z|= 3.2873; p = .00051

75-84

246

20

8.4%

81.3

NS

85+

23

3

1.3%

130.4

NS

Total

13,916

237

100.00%

17.03

--

Table 5: Description with Statistical Analysis

HCL

Description

HRST Risk Scores Statistics, mean (SD)

1

Low Risk

3.4 (1.9)

2

Low Risk

7.5 (2.0)

3

Moderate Risk

10.7 (2.0)

4

High Moderate Risk

13.2 (2.3)

5

High Risk

14.8 (2.5)

6

Highest Risk

15.9 (2.6)

Table 5 provides a summary of the HRST risk scores corresponding to each Health Care Level (HCL). It is evident that there is a consistent increase in the average HRST risk scores as we move from lower to higher HCLs.

38 | P a g e

HCL
1 2 3 4 5 6 Total

Table 6: Mortality Rates by HCL, CY202319

Population Deaths (#) Deaths (%)

Crude Mortality Rate

Significance

3,315

11

4,071

29

2,671

39

1,609

36

1,099

54

1,151

68

4.6% 12.2% 16.5% 15.2% 22.8% 28.7%

3.3

--

7.1

NS

14.6

|z| = 3.0054; p = .00133

22.4

NS

49.1

|z| = 3.8151; p = .00007

59.1

NS

13,916

237

100.00%

17.03

--

Table 7: Odds Ratio for Logistic Regression Model of Mortality on 22 rating items; CY2023

Rating Items (Predictors)

Odds Ratio

99% CI

p-value

Ambulation

1.38

0.99, 1.91

0.013

Hosp Admissions

1.36

1.13, 1.65

<0.001

Clinical Issues

1.29

1.06, 1.55

<0.001

Toileting

1.24

1.01, 1.52

0.006

Emer Room Visits

1.15

0.97, 1.36

0.035

Psychotropic Meds

1.14

1.02, 1.28

0.002

Eating

1.10

0.92, 1.32

0.2

Beh Support Chemical

1.09

0.94, 1.25

0.14

Falls

1.06

0.88, 1.27

0.4

High Risk Treatments

1.05

0.93, 1.17

0.3

Nutrition

1.04

0.91, 1.20

0.4

Injuries

1.04

0.87, 1.23

0.5

Bowel Function

1.03

0.89, 1.19

0.6

Seizures

0.99

0.81, 1.19

0.8

Antiepileptic

0.98

0.80, 1.19

0.8

Transfer

0.97

0.75, 1.25

0.7

Self-Abuse

0.95

0.79, 1.14

0.5

Skin Integrity

0.94

0.79, 1.11

0.3

Gastrointestinal

0.92

0.81, 1.05

0.10

Prof Healthcare SVCS

0.87

0.71, 1.06

0.072

Beh Support Physical

0.83

0.63, 1.06

0.068

Aggression

0.81

0.68, 0.97

0.003

Pseudo R2: 0.15

19 "--"indicates that a statistical test was not conducted. "NS" indicates non-significance.

39 | P a g e

Table 7 presents the Odds Ratio for the Logistic Regression Model of Mortality, which is based on 22 rating items for CY2023. The odds ratio results suggest that for each unit increase in the hospital admission score, the probability of mortality increases by 36% [OR = 1.36, 99% CI (1.13, 1.65), p < 0.001]. Similarly, for each unit increase in the clinical issues score, the probability of mortality increases by 29% [OR = 1.29, 99% CI (1.06, 1.55), p < 0.001]. For each unit increase in the toileting score, the probability of mortality increases by 24% [OR = 1.24, 99% CI (1.01, 1.52), p < 0.01]. Lastly, for each unit increase in the psychotropic medication score, the probability of mortality increases by 14% [OR = 1.14, 99% CI (1.02, 1.28), p < 0.01].

The Pseudo R2 value is 0.15, indicating that the entire model explains 15% of the variations in mortality using these 22 HRST Rating Items. based on on the significance level of Alpha = 0.01.

Table 8: Mortality Rates by Risk Level based on Top 4 HRST Rating Items; CY202320

Deaths Death

Significance

Risk Levels Population

CMR

(#)

(%)

Total

Low

Medium

Low

898

2

0.84 2.23 |z| = 3.4127; p = 0.00032

--

--

Medium

11,597

141

59.49 12.16 |z| = 3.2075; p = 0.00067 |z|=2.6955; p=0.00351

--

High

1,421

94

39.66 66.15 |z| = 12.1377; p = 0

|z| = 7.5273; p = 0

|z| = 14.4286; p = 0

Total

13,916

237 100.00 17.03

--

--

--

Table 8 presents a comparative analysis of the crude mortality rates for CY2023. The overall crude mortality rate (CMR = 17.03) is significantly higher than the rates observed at low (CMR = 2.23, p < 0.001) and medium (CMR = 12.16, p < 0.001) risk levels. However, the mortality rate at the high risk level (CMR = 66.15, p < 0.001) surpasses the overall crude mortality rate. In essence, the crude mortality rate escalates by approximately 3.88 times at high risk level compared to the overall rate, while it diminishes by about 7.64 times at low risk level.
Subsequently, proportion tests were conducted to compare the crude mortality across different risk levels. The results revealed a statistically significant difference between the total crude mortality and the crude mortality at low, medium, and high risk levels at Alpha = 0.01 [|z| = 3.4127, p = 0.00032], [|z| = 3.2075, p = 0.00067] and [|z| = 12.1377, p = 0.00000] respectively.
Moreover, a statistically significant difference was found between the crude mortality at low risk level and those at medium and high risk levels at Alpha = 0.01 [|z| = 2.6955, p = 0.00351] and [|z| = 7.5273, p = 0.00000]. Lastly, a significant difference was also observed between the crude mortality at medium risk level and at high risk level at Alpha = 0.01 [|z| = 14.4286, p = 0.00000].

20 "--"indicates that a statistical test was not conducted. The term `CMR' is employed as an abbreviation for Crude Mortality Rate, which is calculated per 1000 individuals.
40 | P a g e