HBAM Overview Presentation Low Health-Based Allocation Model (HBAM) Overview June 29, 2010 Ontario Ministry IPM/HSIMI of Health and Long-Term Care Health System Information Management and Investment Division
Purpose of this Presentation The aim of this presentation is to provide the audience with basic information about Health Based Allocation Model (HBAM) without getting into a significant level of detail. This includes: 1/ An overview of the reasons for the introduction of HBAM 2/ The guiding principles behind the development of HBAM 3/ A general overview of how the HBAM hospital methodology works a. Service model b. Cost model c. Integration of Service and Cost Models 4/ A general overview of how the HBAM Home Care Methodology works 5/ Examples of HBAM Information and Uses In the coming months, further documentation will be developed with focus on the details of the HBAM methodology and how the results from the methodology can be used. 2
1/ An overview of the reasons for the introduction of HBAM 3
Context With the creation of LHINs, the government signaled its intention to develop and implement a new funding approach that will promote transparency, simplicity, predictability, data quality, cost effectiveness and sector integration The new Health-Based Allocation Model (HBAM). The government created the Local Health Integration Networks (LHINs) to work with local health care providers and community members to determine the health service priorities of their regions. LHINs are based on a principle that community-based care is best planned, coordinated and funded in an integrated manner at the community level, because local people are best able to determine their health service needs and priorities. Authority for funding and managing approximately $20 billion in health care expenditures was transferred to LHINs on April 1, 2007. 4
Impetus for Change Existing funding approaches: Are sector focused and do not support an integrated health care system; Do not complement one another. Thus, it is difficult to get a complete picture of the health care system; Are difficult to understand because of the complexity of the funding methodology; Do not make best use of available health information to determine health needs of the population; Use demographics to determine resource requirements, which did not adequately account for the health status of the population. As a result, sicker populations may have been disadvantaged; and Are disconnected and do not enhance integration of services in a network. 5
2/ The guiding principles behind the development of HBAM 6
HBAM Guiding Principles Prior to developing HBAM, the Ministry and stakeholders established a set of guiding principles: Provide an evidence-based distribution of funding to LHINs within the government s budget limits; Recognize provider characteristics that are commonly accepted to affect the cost of providing care; Maintain patient freedom to choose their health service providers; Ensure stability in funding from year-to-year; Facilitate health sector integration; Encourage quality improvement in health outcomes; and Be simple to understand and communicate. 7
HBAM Development Process With the guiding principles in hand, the ministry embarked on a rigorous process to review and assess existing funding models, approaches and best practices from Ontario and other jurisdictions. The review identified many applicable best practices, but also pointed to the need for a Made in Ontario solution. The ministry therefore set out to develop a new, made-in-ontario funding model. The development process has been collaborative and involved stakeholders from the ministry, LHINs, health service providers and their organizations, several Ontario and U.S. universities, research organizations, and clinicians. The result is the Health Based Allocation Model, which builds on Ontario s data investments and experience in funding models by incorporating the best practices from other jurisdictions and leading edge developments from health services research. 8
3/ A general overview of how the HBAM hospital methodology works 9
What is HBAM? A population health-based funding formula which provides decision makers with actionable information and supports the strategic alignment of funding with management decisions. 10
What is HBAM? HBAM is an enhancement over population and population health-based models as it is based on direct measures of health status, and takes into account both population and provider-based factors. The model estimates each LHIN s and health service provider s (HSP s) share of the system, i.e., available funding The model is made up of two main components: Service model - estimates annual health service requirements, taking into account each Ontario resident s clinical, social, and demographic conditions. Unit cost model - determines unit costs for each health service provider and recognized provider characteristics that justifiably lead to higher unit costs. 11
What Will HBAM Cover? HBAM will cover the following sectors: Developed Hospitals Acute Inpatient & Day Surgery Inpatient Mental Health Complex Continuing Care Inpatient Rehabilitation Emergency To Be Developed Long-Term Mental Health* Mental Health Clinics* Ambulatory Clinics* Community Outpatient* Developed Home Care To Be Developed Long-Term Care Community Mental Health Community Health Centres Community Support Services *Expenses for these hospital care types accounts for ~30% of total hospital expenses. 12
How Does HBAM Work? Step 1. Construct electronic health profile for each Ontario resident Step 2. Assign each person to a refined clinical group By Program Step 3. Assign resource weight to each person based on refined clinical group, age group, socioeconomic status and rural group Step 4. Calculate CD/CSD specific resource weights per person within each clinicalage-gender group and multiply by CD/CSD population forecasts Step 5. For each CD/CSD, allocate share of resource weights forecasts to providers by clinical- age-gender group Step 6. Apply provider HBAM unit cost to estimate total costs Step 7. Sum total costs estimates for providers within each LHIN Step 8. Compile HBAM components to calculate total LHIN HBAM costs Step 9. HBAM costs determine LHIN s share of funding 13
Step 1 Step 2 Step 3 Step 4 Step 5 Step 6 HBAM Components HBAM Population Grouper adjusts for health, demographics, SES, and geography Each Ontario resident assigned to a refined clinical group and an estimate of annual resource requirements for each hospital care type (e.g. acute, emergency, CCC, inpatient rehabilitation, inpatient mental health) Sum estimated resource requirements per person by age and gender group for each geographic region (105 census divisions (CD) and subdivisions (CSD)) Each CD/CSD has estimated total annual resource requirement summarized by age and gender group Apply population growth projections by age and gender group by CD/CSD Regional forecasts of health service requirements based on local burden of disease and provincial average treatment intensities, and incorporate regional variation in population growth and ageing. HBAM Patient Grouper used to estimate each hospital s share of regional resources HBAM estimates each hospital s share of each regions total annual resources based on observed market sharing patterns by age/sex group and type of care (e.g. by partition: medical, surgical, paediatric) HBAM Unit Cost Model estimates reference unit costs, taking provider characteristics into account HBAM estimates unit costs for each hospital care type: acute, emerg, mental health, CCC, and inpatient rehab Step 7 Step 8 Step 9 For each hospital and care type, multiply HBAM services times HBAM unit costs to estimate total HBAM expenses. Sum HBAM estimated total expenses for all hospitals by LHIN. LHIN total hospital expenses determines share of hospital funding 14 14
HBAM Population Grouper: Structure Refined Clinical Groups HBAM s population grouper for the acute module is comprised of 18 Major Clinical Groups and 315 Refined Clinical Groups 15
3a/ HBAM Service Model The HBAM Service model estimates the total relative weight of services provided by facilities by examining the distribution of patients throughout the province. 16
Step 1. Construct Electronic Profile for Each Ontario Resident An Electronic Health Profile is constructed for each person by coalescing all geographic, demographic, socioeconomic, clinical data elements from all the person s hospital and home care encounters within a fiscal year Each person s Electronic Profile can be aggregated by program and care type to profile clinical conditions and service use relevant to each care type: Acute Inpatient and Day Surgery Inpatient Mental Health Inpatient Rehabilitation Emergency Department Complex Continuing Care Home Care The Electronic Health Profile is used to assign people to refined clinical groups in each of HBAM s modules (i.e. hospital care types) 17
Step 1. Construct Electronic Profile for Each Ontario Resident Information captured in the Electronic Health Profile: Patient Demographics Age group, region of residence, SES status, Rural geography Patient Grouper elements Casemix groups (HBAM casemix groups) Diagnostic and Procedural data elements (ICD10-CA/CCI codes) HBAM episode cost weight Population Grouper Elements Episode Refined Clinical Group, Episode Cost Rank Excluded Episodes Non Ontario, Non OHIP 18
Step 2. Assign each Person to a Refined Clinical Group Each person is assigned to a single Refined Clinical Group based on information captured in their electronic health profile People with a single hospital encounter are assigned to an RCG based on their MRDx People with multiple hospital encounters within a fiscal year are assigned to their highest ranked RCG from all their encounters RCG rankings are based on the provincial mean Refined Clinical Group s episode specific HBAM case mix weight (from HBAM s patient grouper) Rankings are specific to paediatric (age 0-17) and adult (age 18+) populations Ranking of 1 indicates the highest mean HBAM weight 19
Refined Clinical Group Ranks Refined Clinical Group Ranks (Sample) Refined Clinical Group Rank 0-17 17+ 097.1 Cardiomyopathy 4 35 103 Pulmonary heart disease 34 115 109.1 Cerebrovascular disease 17 33 120 Hemorrhoids 235 310 131 Respiratory failure; insufficiency; arrest (adult) 1 1 150 Liver disease; alcohol-related 78 158 Chronic renal failure 16 166 226 Fracture of neck of femur (hip) 62 36 232 Sprains and strains 159 235 250 Nausea and vomiting 183 242 An adult with separate hospitalizations for Chronic Renal Failure (rank 166) and Respiratory Failure (rank 1) within a fiscal year, would be assigned to the Respiratory Failure Refined Clinical Group in HBAM s acute population grouper 20
Step 3. Assign Resource Weight to each Person HBAM s Population Grouper assigns each person a resource weight that describes their expected annual resource requirements This resource weight is the mean Actual Annual Resource Use for each Refined Clinical Group & Age Group combination Because all resource weights from all hospitalizations are included in HBAM s annual resource weight: Provincial Sum(HBAM Patient Grouper Weights) = Provincial Sum(HBAM Population Grouper Weights) HBAM Population Grouper weights calibrated with three years of data 21
Adjustments for SES and Rural Geography In Ontario and elsewhere, it has been demonstrated empirically that low socioeconomic status and rural geography can impact requirements for health care HBAM adjusts for SES and rural geography to ensure that estimates of health service use are not biased against residents of low SES or rural regions People receive rural geography adjustment if they are likely to receive care at a small hospital OR live in a census subdivision with a high RIO score RIO is the Ontario Medical Association s Rurality Index for Ontario People receive adjustment for low SES if they live in a neighborhood (dissemination area) in which the average income is in the lowest area-based income quintile Note that these multipliers are required to address bias not already addressed by HBAM s detailed clinical groups Bias associated with SES and rural geography that manifests as variation in prevalence of disease are likely accounted for in unadjusted HBAM 22
Adjustments for SES and Rural Geography Adjustments for low SES and rural geography are specific to each Major Clinical Group because these population characteristics may have differential effects on requirements for care by clinical group Major Clinical Group Non Rural Provider Rural Provider Non Low SES Low SES Non Low SES Low SES RIO < 4 RIO = 4 RIO = 5 RIO < 4 RIO = 4 RIO = 5 RIO < 4 RIO = 4 RIO = 5 RIO < 4 RIO = 4 RIO = 5 01 Infectious/Parasitic Dis 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.10 1.10 1.00 1.00 1.00 02 Neoplasms 1.00 1.00 1.00 1.00 1.03 1.03 1.14 1.26 1.27 1.24 1.36 1.36 03 Blood and Immune 1.00 1.00 1.18 1.07 1.07 1.07 1.10 1.10 1.63 1.13 1.29 1.68 04 Endocrine, Nutrition, Metabolic 1.00 1.00 1.00 1.02 1.06 1.06 1.14 1.32 1.32 1.17 1.23 1.38 06 Nervous System 1.00 1.00 1.00 1.00 1.07 1.07 1.02 1.02 1.21 1.04 1.11 1.31 07 Eye and Adnexa 1.00 1.00 1.08 1.00 1.00 1.00 1.07 1.08 1.08 1.03 1.11 1.11 08 Ear and Mastoid 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.08 1.00 1.00 1.00 09 Circulatory 1.00 1.00 1.00 1.00 1.00 1.00 1.08 1.15 1.16 1.13 1.15 1.15 10 Respiratory 1.00 1.00 1.00 1.00 1.00 1.00 1.05 1.05 1.11 1.09 1.12 1.12 11 Digestive 1.00 1.08 1.08 1.04 1.04 1.10 1.05 1.08 1.18 1.04 1.11 1.17 12 Skin and Subcutaneous Tissue 1.00 1.15 1.15 1.13 1.13 1.13 1.07 1.07 1.22 1.17 1.35 1.35 13 Musculoskeletal and Connective 1.00 1.01 1.11 1.00 1.03 1.03 1.17 1.46 1.46 1.23 1.45 1.50 14 Genitourinary 1.00 1.04 1.08 1.00 1.07 1.11 1.13 1.15 1.15 1.08 1.38 1.38 15 Pregnancy and Childbirth 1.00 1.02 1.02 1.00 1.00 1.00 1.04 1.04 1.08 1.08 1.08 1.08 17 Congenital and Chromosomal 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 19 Injury and Poisoning 1.00 1.08 1.08 1.00 1.07 1.07 1.25 1.31 1.31 1.33 1.38 1.38 20 External Causes 1.00 1.13 1.14 1.03 1.15 1.15 1.15 1.15 1.15 1.27 1.33 1.40 23
Assignment of Resource Weights Expected Population Weight is the mean Actual Annual Resource Use for each Refined Clinical Group & Age Group combination, prior to SES and Rural geography adjustment HBAM Annual Resource Use = [Expected Annual Resource Use] x [SES & Rural Adjustment] ELECTRONIC HEALTH PROFILE Major Clinical Group Refined Clinical Group Actual Annual Resource Use Expected Annual Resource Use (A) SES and Rural Adjustment HBAM Annual Resource Use (B) = A x B 0001, Age Group=4, Rural Provider, Low SES, RIO=4 Geo105=3543000 09 Circulatory 100.1 Acute Myocardial Infarction with other Major Cardiac Conditions 4.63 4.22 1.15 4.85 0002, Age Group=4, Non Rural Provider, Non Low SES, RIO<4 Geo105=3510000 02 Neoplasms 039 Leukemias 5.99 5.72 1.00 5.72 0003, Age Group=4, Rural Provider, Non Low SES, RIO=5 Geo105=3539000 02 Neoplasms 019 Cancer of bronchus; lung 3.54 2.07 1.27 2.62 0004, Age Group=3, Rural Provider, Non Low SES, RIO<4 Geo105=3519049 10 Respiratory 127 Chronic obstructive pulmonary disease and bronchiectasis 2.77 1.86 1.05 1.95 24
Step 4: Sum Expected Resources by Community Age Group Major Clinical Group = 09 Circulatory Geo105 = Toronto Expected Annual Resource Use Medical Paediatric Surgical Mental Health F01 8 F02 0 F03 1 F04 2 13 1 F05 6 14 F06 13 43 F07 17 30 F08 23 52 F09 59 79 F10 88 60 F11 108 172 F12 132 136 F13 165 189 F14 269 207 F15 332 263 F16 471 357 F17 695 342 F18 1,052 354 M01 10 M02 1 M03 1 M04 13 6 0 M05 14 44 M06 13 33 M07 48 42 M08 56 80 M09 97 124 M10 128 254 M11 167 341 M12 273 459 M13 335 535 M14 429 550 M15 491 489 M16 528 424 M17 572 415 M18 550 231 Pregnancy and Childbirth Newborn and Neonate Total estimated resources for each community are derived by summing up each person s estimated resources by their community of residence HBAM uses 105 communities, which are a mix of census divisions and subdivisions Within each community, estimated resources are summed by age/sex group and partition for each Major Clinical Group these tables are called community ratings Example table at right is for the Circulatory MCG in the Toronto region Grand Total 7,146 39 6,320 25
Step 4: Applying Population Projections HBAM s projections of resource use are based on growth rates by age/gender group for each region and preserve HBAM s clinical detail 26
Step 4: Applying Population Projections GROWTH MATRIX 1 OF 1,890 Major Clinical Group = 09 Circulatory Geo105 = Toronto Age Group HBAM Annual Resource Use Annual Growth Rate HBAM Services (3-year projection of Expected Annual Resource Use) Medical Paediatric Surgical Medical Paediatric Surgical F01 8 0.16% 8 F02 0-1.96% 0 F03 1-1.01% 1 F04 2 13 1 0.85% 2 13 1 F05 6 14-0.62% 6 14 F06 13 43 0.34% 13 43 F07 17 30 0.64% 17 31 F08 23 52-0.14% 23 52 F09 59 79-3.17% 54 72 F10 88 60 0.68% 90 61 F11 108 172 1.24% 112 179 F12 132 136 1.10% 136 140 F13 165 189 6.01% 196 225 F14 269 207 0.87% 276 212 F15 332 263-0.70% 325 258 F16 471 357-1.23% 454 344 F17 695 342-0.44% 686 337 F18 1,052 354 5.70% 1,242 418 M01 10 0.19% 10 M02 1-1.35% 1 M03 1-1.74% 1 M04 13 6 0 0.96% 13 6 0 M05 14 44 0.08% 14 44 M06 13 33 0.18% 13 33 M07 48 42-0.56% 47 42 M08 56 80-0.10% 55 80 M09 97 124-3.65% 87 111 M10 128 254 1.56% 134 266 M11 167 341 2.16% 178 363 M12 273 459 1.28% 284 477 M13 335 535 5.53% 394 629 M14 429 550 1.50% 449 575 M15 491 489-1.36% 471 470 M16 528 424-0.30% 523 420 M17 572 415 1.36% 596 432 M18 550 231 6.05% 656 276 HBAM services are the 3-year projections of estimated service use for each age/gender group, partition, and major clinical group within each community HBAM services incorporate regional variations in disease burden, SES and rural geography adjustments, and differential population growth and ageing rates HBAM services are based on 3- year projections so that the results are relevant to the year in which they are applied for funding allocation Grand Total 7,146 39 6,320 1.66% 7,547 40 6,604 Growth Rate 5.61% 1.20% 4.49% 27
Step 5: Market Share Community HBAM Services to Providers Each community s HBAM services are allocated to providers based on observed market sharing patterns by age/gender group, partition, and Major Clinical Group Market sharing patterns established using HBAM s patient grouper Each hospital s actual share of each community s HBAM s case mix weights by age/gender group, partition, and MCG determines share of estimated HBAM services HBAM services are allocated to providers based on where patients sought care and on what type and intensity of care was provided Resources follow the patient Market sharing does not rely on patient or provider LHIN, which ensures that LHIN boundaries do not impede patients or affect allocation of resources 28
Step 5: Market Share HBAM Services to Providers MARKET SHARE MATRIX 1 OF 1,890 Major Clinical Category = 09 Circulatory Geo105 = Toronto Total : Toronto Hospital Z : Any Region Hospital Z : Any Region Total : Toronto Hospital Z : Any Region (A) (B) (C = B / A) (D) (E = C x D) Age Actual Annual Resource Use Actual Annual Resource Use Market Share HBAM Services HBAM Services Group Medical Paediatric Surgical Medical Paediatric Surgical Medical Paediatric Surgical Medical Paediatric Surgical Medical Paediatric Surgical F01 7 8 F02 0 0 F03 1 1 F04 1 10 1 1 65% 2 13 1 1 F05 4 12 1 11 22% 91% 6 14 1 13 F06 11 40 5 33 50% 82% 13 43 7 35 F07 17 35 4 18 24% 52% 17 31 4 16 F08 23 52 11 39 49% 74% 23 52 11 39 F09 45 87 16 54 36% 62% 54 72 19 44 F10 69 54 19 10 27% 18% 90 61 24 11 F11 88 177 14 71 15% 40% 112 179 17 72 F12 110 154 35 77 32% 50% 136 140 43 70 F13 117 183 34 69 29% 38% 196 225 57 85 F14 218 240 56 116 26% 48% 276 212 71 102 F15 307 325 76 117 25% 36% 325 258 81 93 F16 440 427 148 184 34% 43% 454 344 153 148 F17 638 442 165 153 26% 34% 686 337 177 116 F18 929 395 282 184 30% 47% 1,242 418 377 195 M01 10 10 M02 1 1 M03 0 1 M04 6 4 1 13 6 0 M05 11 44 1 5 6% 11% 14 44 1 5 M06 9 30 6 18 71% 58% 13 33 9 19 M07 34 37 7 17 21% 47% 47 42 10 20 M08 46 83 16 27 34% 33% 55 80 19 26 M09 67 134 18 54 26% 41% 87 111 23 45 M10 104 282 23 134 22% 48% 134 266 29 127 M11 130 390 22 152 17% 39% 178 363 30 142 M12 218 555 57 235 26% 42% 284 477 74 202 M13 259 584 65 260 25% 45% 394 629 99 280 M14 306 582 90 263 29% 45% 449 575 132 260 M15 447 590 145 277 33% 47% 471 470 154 221 M16 485 518 148 206 30% 40% 523 420 160 167 M17 520 494 173 193 33% 39% 596 432 198 169 M18 492 256 157 111 32% 44% 656 276 209 120 Total 6,152 33 7,203 1,794 3,087 7,547 40 6,604 2,191 2,840 29
Step 5: Market Share HBAM Services to Providers HBAM Services by Hospital Toronto LHIN Hospitals Major Clinical Category Hospital A Hospital B Hospital C Hospital D Hospital E Hospital F Hospital G Grand Total 01 Infectious/Parasitic Dis 357 707 713 540 770 1,396 697 5,178 02 Neoplasms 2,656 4,244 4,229 2,636 3,541 17,849 8,798 43,953 03 Blood and Immune 581 119 166 138 82 412 176 1,674 04 Endocrine, Nutrition, Metabolic 757 422 1,431 792 624 2,027 1,192 7,245 05 Mental and Behavioural 0 5 0 19 0 25 0 50 06 Nervous System 954 318 934 449 443 1,635 788 5,522 07 Eye and Adnexa 180 293 800 874 458 796 536 3,938 08 Ear and Mastoid 293 9 30 81 53 59 230 755 09 Circulatory 1,055 2,559 13,749 3,463 3,356 20,451 13,594 58,226 10 Respiratory 1,270 1,562 2,678 2,313 2,866 4,681 1,888 17,258 11 Digestive 1,764 2,218 2,826 1,846 2,247 4,380 2,198 17,480 12 Skin and Subcutaneous Tissue 123 147 282 189 256 241 293 1,530 13 Musculoskeletal and Connective 742 1,118 3,354 1,547 1,338 3,707 5,216 17,022 14 Genitourinary 595 981 2,046 1,639 1,682 1,786 2,326 11,056 15 Pregnancy and Childbirth 6,419 2,142 2,872 2,464 86 3,563 17,545 16 Perinatal Period 1,641 3,611 993 954 636 1 2,701 10,535 17 Congenital and Chromosomal 4,791 130 198 63 89 833 133 6,237 19 Injury and Poisoning 2,292 2,436 5,224 2,229 2,268 5,262 8,132 27,842 20 External Causes 455 669 1,254 859 886 2,238 1,247 7,607 Total Toronto LHIN 20,506 27,967 43,049 23,503 24,057 67,863 53,709 260,654 HBAM services are summed by age/gender group, partition, and Major Clinical Group for each hospital Next step is to estimate HBAM reference costs for each hospital 30
3b/ HBAM Hospital Cost Model The HBAM Hospital Cost Model looks at facility specific information to estimate the expected cost per weighted case for each facility. 31
Data Sources HBAM Unit Cost Model 32
What Do we Know about Hospital Unit Cost Variation? Measured case mix Unmeasured Case Mix Standby capacity and specialization Economies of scale and scope Clinical training Research 33
HBAM Unit Cost Model: Features Ensures stability of results across years Is robust to different methodological assumptions Built on reliable, well measured and reported hospital data Covers a large portion of hospital activity Recognizes that cost structures differ across hospitals and care types Integrates seamlessly with HBAM utilization model Allows management tool applications 34
HBAM Unit Cost Model: Cost Modifiers Selected Teaching - Number of student days (acute inpatient activity) - Ratio of student days to patient throughput (complex continuing care and emergency room) Rural geography - Distance from nearest facility with a minimum of 15,000 equivalent weighted cases Economies of Scale - 1/weighted activity Indicator of High Cost Structure Associated with Specialized Services Acute Inpatient: - Acute Specialization Index: weighted portion of a hospital s volume specialized programs Complex continuing care: - Chronic Tertiary Index: proportion of weighted days that is associated with Palliative patients Patients on ventilators 35
HBAM Acute Unit Cost Model Teaching - Number of medical student days in the acute setting Rural geography - Distance from nearest facility with a minimum of 15,000 equivalent weighted cases (JPPC definition) Indicator of High Cost Structure Associated with Specialized Services - Acute Specialization Index =specialized weighted cases/total weighted cases *100 36
HBAM Acute Unit Cost Model Estimated Expenses = Acute Weighted Cases ($4,455 + $141 Specialization Index + $2.7 Distance) + $58 Student Days Estimated Unit Cost Estimated Expenses = /Acute Weighted Cases Units of Service: HBAM weighted cases Predictive Ratio Quartiles: 0.97, 1.06 37
HBAM Mental Health Unit Cost Model Teaching - Number of medical student days in the acute mental health setting 38
HBAM Mental Health Unit Cost Model Estimated Expenses = $4,755 x Acute Mental $58 x Student Days Health Weighted Days Estimated Unit Cost = Estimated Expenses/ Acute Mental Health Weighted Days Units of Service: Acute Mental health HBAM weighted Days Predictive Ratio Quartiles: 0.91, 1.08 39
HBAM CCC Unit Cost Model Teaching - Ratio of student days to patient throughput = Student days in CCC / CCC RUG un-weighted days Economies of Scale - 1/CCC RUG weighted days Indicator of High Cost Structure Associated with Specialized Services Chronic Tertiary Index: - proportion of weighted days that is associated with Palliative patients & patients on ventilators = CCC RUG weighted days (Palliative & ventillation)/ccc RUG weighted days 40
HBAM CCC Unit Cost Model Estimated Expenses = RUG CCC Weighted Days ($401 + $1,095 CCC Teaching Intensity + $428 Percent CCC Tertiary 354,421Scale) Estimated Unit Cost = Estimated Expenses/RUG CCC Weighted Days Units of Service: RUG CCC Weighted Days Predictive Ratio Quartiles: 0.89, 1.16 41
HBAM Emergency Department Unit Cost Model Teaching Ratio of student days to patient throughput = Student days in ER / ER ACW weighted visits 42
HBAM Emergency Department Unit Cost Model Estimated Expenses = ER Weighted Visits ($3,112 $791 ED Teaching Intensity) Estimated Unit Cost = Estimated Expenses/ ER Weighted Visits Units of Service: ER CACS weighted visits Predictive Ratio Quartiles: 0.84, 1.11 43
HBAM Rehabilitation Unit Cost Model Estimated Expenses = NRS Rehabilitation Weighted Cases ($12,429 + 0 Large + $5,684 CCC/Rehab 4,351 AHSC) Estimated Unit Cost = Estimated Expenses/ NRS Rehab Weighted Cases Units of Service: NRS Rehabilitation Weighted Cases Predictive Ratio Quartiles: 0.84, 1.12 44
HBAM Sample Acute Population Grouper ACUTE & MENTAL HEALTH GROUPER Expected Resource Weight SES Multipliers 01 to 17 18 to 59 60 to 79 80+ SES = Y SES = N SES = Y SES = N Rural = N Rural = N Rural = Y Rural = Y 001 Tuberculosis 1.4229 1.4229 2.0174 2.2581 1.00 1.00 1.00 1.00 002 Septicemia (except in labor) 1.6388 2.7864 2.8937 2.3001 1.00 1.00 1.00 1.00 003 Bacterial infection; unspecified site 1.1341 2.5544 2.6415 2.4078 1.00 1.00 1.00 1.00 004 Mycoses 0.7540 0.7540 1.2744 1.9573 1.00 1.00 1.00 1.00 005 HIV infection 2.6499 2.6499 3.0518 3.0518 1.00 1.00 1.00 1.00 006 Hepatitis 0.2719 0.2719 0.3675 0.3675 1.00 1.00 1.00 1.00 007 Viral infection 0.5295 0.5376 0.7992 1.1111 1.00 1.00 1.00 1.00 008.1 Other infections and immunity disorders 1.0989 1.1941 1.3099 1.3099 1.00 1.00 1.00 1.00 009 Sexually transmitted infections (not HIV or hepatitis) 0.2236 0.1565 0.3112 0.3112 1.00 1.00 1.00 1.00 010 Immunizations and screening for infectious disease 0.1012 0.1012 0.1150 0.1150 1.00 1.00 1.00 1.00 011 Cancer of head and neck 1.8282 1.8282 1.9729 1.7053 1.00 1.00 1.15 1.14 012 Cancer of esophagus 2.3368 2.3368 2.2471 2.1532 1.00 1.00 1.15 1.14 013 Cancer of stomach 2.7378 2.7378 2.8659 2.3802 1.00 1.00 1.15 1.14 014 Cancer of colon 2.1266 2.1266 2.4902 2.6820 1.00 1.00 1.15 1.14 015 Cancer of rectum and anus 2.0310 2.0310 2.4503 2.5934 1.00 1.00 1.15 1.14 016 Cancer of liver and intrahepatic bile duct 2.3252 2.3252 2.3740 2.0456 1.00 1.00 1.15 1.14 017 Cancer of pancreas 3.0064 3.0064 2.9980 2.3489 1.00 1.00 1.15 1.14 018 Cancer of other GI organs; peritoneum 3.0143 3.0143 3.0248 2.8699 1.00 1.00 1.15 1.14 019 Cancer of bronchus; lung 2.0080 2.0080 2.1335 2.0966 1.00 1.00 1.15 1.14 020 Cancer; other respiratory and intrathoracic 2.7004 2.7004 2.9887 2.9887 1.00 1.00 1.15 1.14 021 Cancer of bone and connective tissue 7.5732 1.9925 1.4715 1.2787 1.00 1.00 1.15 1.14 022 Melanomas of skin 0.3384 0.3384 0.4230 0.5391 1.00 1.00 1.15 1.14 023 Other non-epithelial cancer of skin 0.2055 0.2055 0.2253 0.2554 1.00 1.00 1.15 1.14 024 Cancer of breast 0.6465 0.6465 0.6827 0.7198 1.00 1.00 1.15 1.14 025 Cancer of uterus 1.2803 1.2803 1.4400 1.6174 1.00 1.00 1.15 1.14 45
3c/ Integration of Hospital Service and Unit Cost Models With estimated weighted cases and cost per unit, the total expected resource use for facilities and LHINs can be calculated. 46
Integration of Hospital Service and Unit Cost Models Actual expenses: = (Acute WC Actual * Unit Cost Actual ) + (MH Wt Days Actual * Unit Cost Actual ) + (CCC Wt Days Actual * Unit Cost Actual ) + (Rehab Wt Cases Actual * Unit Cost Actual ) + (ER Wt Visits Actual * Unit Cost Actual ) + Non Modeled expenses Actual HBAM expenses: = (Acute WC HBAM * Unit Cost HBAM ) + (MH Wt Days HBAM * Unit Cost HBAM ) + (CCC Wt Days HBAM * Unit Cost HBAM ) + (Rehab Wt Cases HBAM * Unit Cost HBAM ) + (ER Wt Visits HBAM * Unit Cost HBAM ) + Non Modeled expenses Growth Adjusted 47
Integration of Hospital Service and Unit Cost Models (cont d) HBAM Service Based expenses: = (Acute WC HBAM * Unit Cost Actual ) + (MH Wt Days HBAM * Unit Cost Actual ) + (CCC Wt Days HBAM * Unit Cost Actual ) + (Rehab Wt Cases HBAM * Unit Cost Actual ) + (ER Wt Visits HBAM * Unit Cost Actual ) + Non Modeled expenses Growth Adjusted HBAM Cost Based expenses: = (Acute WC Actual * Unit Cost HBAM ) + (MH Wt Days Actual * Unit Cost HBAM ) + (CCC Wt Days Actual * Unit Cost HBAM ) + (Rehab Wt Cases Actual * Unit Cost HBAM ) + (ER Wt Visits Actual * Unit Cost HBAM ) + Non Modeled expenses 48
Integration of Hospital Service and Unit Cost Models Growth Adjustment applied to non Modeled Hospital Activity ELDCAP based on hospital specific growth rate of complex continuing care activity Long term MH based on hospital specific growth rate of long term and forensic mental health Outpatient MH based on hospital specific growth rate of acute inpatient mental health activity Ambulatory Clinics based on hospital specific overall growth rate ER D&T based on hospital specific growth rate of ER activity 49
Example: Calculating Hospital Total HBAM Costs HBAM services and HBAM unit costs for all hospital care types are used to derive overall HBAM expenses for Hospital A HBAM expenses for the un-modeled programs are equal to the provider s actual costs, but increased to adjust for hospital specific growth Hospital A HBAM Service HBAM Unit Cost HBAM Expenses Acute 168,406 3,898 581,899,000 MH 13,500 4,005 54,062,000 CCC 16,250 476 7,735,000 ER 86,750 2,298 199,379,000 Rehab 9,125 11,158 101,815,000 MH Long-Term Inpatient 23,275,000 Ambulatory Clinics - Mental Health 11,190,000 Ambulatory Clinics - Other 290,949,000 Other Community Outpatient 22,380,000 Hospital X Total HBAM Expenses 1,292,684,000 50
Example: Summing Estimated Total Costs for Providers Within Each LHIN Assume LHIN 1 has three hospitals: A, B, and C LHIN 1 HBAM Expenses Hospitals A 1,292,684,000 B 898,455,000 C 345,675,000 Hospitals Total 2,536,814,000 HBAM costs can also be amalgamated at the hospital care type level (e.g., Acute Inpatient and Day Surgery) 51
Example: Compiling HBAM Components to Calculate Total LHIN HBAM Costs HBAM Summary for LHIN 1 HBAM Component HBAM Expenses Hospitals: Acute 1,310,867,000 MH 82,501,000 CCC 147,000,000 ER 78,413,000 Rehab 197,103,000 MH Long-Term Inpatient 62,235,000 Ambulatory 658,695,000 Hospital Total 2,536,814,000 CCAC and CSS 250,000,000 Long-Term Care 200,000,000 Community Mental Health 100,000,000 Community Health Centres 20,000,000 LHIN Total HBAM Expenses 3,106,814,000 52
Example: Determine LHIN s Share of Funding Each LHIN s share of funding is determined using the total HBAM expenses from all care settings. Total HBAM Expenses HBAM Share LHIN 1 3,106,814,000 16% LHIN 2 396,498,000 2% LHIN 3 746,578,000 4% LHIN 4 1,198,756,000 6% LHIN 5 2,134,231,000 11% LHIN 6 132,928,000 1% LHIN 7 1,750,874,000 9% LHIN 8 574,689,000 3% LHIN 9 2,170,000,000 11% LHIN 10 772,382,000 4% LHIN 11 2,748,593,000 14% LHIN 12 1,097,465,000 5% LHIN 13 1,150,908,000 6% LHIN 14 2,019,284,000 10% Total 20,000,000,000 100% 53
Example: HBAM Costs Determine LHIN s Share of Funding While the share of LHIN funding is determined at the aggregate level, the estimates can be disaggregated to present actual to HBAM expenses for each program within the LHIN and by care type within each program This information can support LHIN and provider decision making For example: actual and HBAM expenses can be analyzed by: LHIN Program within the LHIN Acute care type within the hospital program Medical partition within the Acute care type of the hospital program Cancer major clinical groups within the hospital medical partition Cancer refined clinical groups within the cancer major clinical group 54
4/ HBAM Home Care Model 55
How Does HBAM s Home Care Model Work? Step 1. Construct electronic health profile for each Ontario resident Step 2. Assign each person to a refined clinical group Step 3. Estimate home care service costs for each person based on their refined clinical group, age group, socioeconomic status, and rural group Step 4. Calculate CD/CSD specific service cost estimates per person within each clinicalage-gender group and multiply by CD/CSD population forecasts Step 5. Allocate forecasted resources to LHINs based on LHIN providing service Step 6. Make adjustments for CCAC non-service costs Step 7. Sum HBAM CCAC expenses for all LHINs Step 8. LHIN HBAM expenses determine LHIN s share of CCAC funding 56
HBAM Home Care Model HBAM s Home Care model aims to derive the best estimate of the cost of each individual s annual home care services A basic model might use only age to estimate costs Age Group 00-09 10-44 45-64 65-74 75-84 85+ Per Capita Cost 38 19 59 180 455 1018 Major improvement in accuracy of predicted costs can be achieved using clinical information available for people who had contact with the hospital system (acute people) Inpatient, day surgery, chemotherapy and dialysis clinics 57
1600 Average Home Care Cost per Person: Acute and Non Acute 1400 1416 1200 Average Cost per Person 1000 800 600 494 830 804 400 295 270 200 0 123 74 77 33 10 22 00-09 10-44 45-64 65-74 75-84 85+ Non Acute Acute 58
Home Care Use by HBAM Home Care Model Component HBAM Home Care Model Component Number and Percent of Ontario Residents Home Care Recipients % of Home Care Recipients Acute and DS 1,439,250 12% 184,901 46% Cancer 152,834 1% 40,720 10% No Hospital Contact 10,550,105 87% 173,062 43% Totals 12,142,189 100% 398,683 100% 59
What Home Care Costs are Estimated? HBAM estimates the direct cost of the following services using the provincial average cost per service These services account for ~70% of total CCAC expenses Service Type Provincial Mean Unit Cost 2006/07 Nursing Visit $70 Nursing Shift $41 Respiratory $185 Nutrition $105 Physiotherapy $94 Occupational Therapy $104 Speech $107 Social Work $129 Personal Support $26 Respite $25 60
HBAM Service Costs for Acute People Each acute person s HBAM home care service cost is the average cost of the individual s assigned group: HBAM Service Cost ijklm i i Total Cost Individuals ijklm jklm where: i indexes individuals; j indexes their clinical group; k indexes their age group; l indexes their SES status; m indexes their rural status HBAM Home Care contains 700 acute clinical groups each with up to 7 age groups 61
HBAM Home Care: Propensity and Cost Chronic Bronchitis Cancer of Bronchus; Lung Total 2005/06 Service Cost $6M $5M Number of Acute People 6,000 5,000 Number of Acute People with Home Care 2,000 2,500 Likelihood of Receiving Home Care 33% 50% Average Home Care Cost per Acute Person with Home Care $3,000 $2,000 Mean Home Care Cost per Acute Person* $1,000* $1,000* * These values are the HBAM estimated service costs. They reflect both the likelihood of receiving home care and the cost of home care conditional on receiving home care. 62
HBAM Estimated Service Costs Parent Group Sub Group 00-01 01-09 10-44 45-64 65-74 75-84 85+ 049.1 Diabetes mellitus wo comps or w minor comps 1 806 806 161 604 2,115 1,884 1,884 1.09 1.00 050.2 Diabetes mellitus with renal complications 1 920 920 1,820 1,820 2,220 2,220 1.09 1.00 050.3 Diabetes mellitus with circ., neuro., or ophth. comps 1 875 2,308 2,420 2,811 2,302 1.09 1.00 050.4 Diabetes mellitus with multiple complications 1 3,772 4,787 4,787 4,787 4,787 1.09 1.00 050.5 Diabetes Mellitus with Complications with Dialysis 1 2,687 2,687 2,491 2,491 2,491 1.09 1.00 100 Acute myocardial infarction 1 0 50 120 347 665 1,048 1.09 1.00 100.1 Acute Myocardial Infarction w oth Maj. Cardiac Conditions 1 156 466 781 1,010 1,284 1.09 1.00 108 Congestive heart failure; nonhypertensive 1 710 710 710 875 1,200 1,489 1,789 1.09 1.00 109.1 Cerebrovascular disease 1 86 86 86 452 617 705 993 1.09 1.00 109.1 Cerebrovascular disease 2 284 284 284 507 822 1,065 1,285 1.09 1.00 109.2 Stroke 1 3,055 348 348 573 924 1,148 1,459 1.09 1.00 127 Chronic obstructive pulmonary disease and bronchiectasis 1 1,024 1,024 1,024 1,024 1,024 1.09 1.00 127 Chronic obstructive pulmonary disease and bronchiectasis 2 676 676 676 785 1,051 1,338 1,561 1.09 1.00 127 Chronic obstructive pulmonary disease and bronchiectasis 3 1,562 1,562 1,562 1,773 1,773 2,235 2,311 1.09 1.00 127.2 Chronic Bronchitis 1 589 589 589 555 807 1,203 1,579 1.09 1.00 159 Urinary tract infections 1 0 79 65 419 419 750 1,247 1.09 1.00 159 Urinary tract infections 2 70 593 965 2,011 2,011 2,011 2,011 1.09 1.00 251 Gastrostomy And Colostomy Procedures 1 5,576 6,292 3,375 3,139 3,137 2,679 2,259 1.09 1.00 251 Gastrostomy And Colostomy Procedures 2 6,158 6,158 6,158 3,475 3,078 3,078 3,078 1.09 1.00 352 Hip Replacement 1 705 705 631 750 1,069 1,355 1.09 1.00 352 Hip Replacement 2 1,994 1,994 1,994 2,149 2,149 1.09 1.00 354 Knee Replacement 1 648 475 546 753 1,190 1.09 1.00 354 Knee Replacement 2 1,509 1,446 1,446 1,446 1,915 1.09 1.00 Non Acute 1 33 33 10 22 77 270 804 1.0 1.12 1.0 1.52 Adj. for Low SES Adj. for Rural Geog. 63
5/ Examples of HBAM Information and Uses 64
HBAM: A Tool for Managers HBAM Expenses HBAM Volume X HBAM Unit Cost Program Care Type Provider Major Clinical Group Refined Clinical Group Hospitals Home Care Long-Term Care Rehab Acute ER Hospital A Hospital B Hospital C Respiratory Neoplasms Circulatory Acute myocardial infarction.. Pulmonary heart disease Community Mental Health Community Support Services Other Community Services MH CCC Ambulatory... Hospital D Hospital E Hospital F... Nervous System Eye and Adnexa..... Congestive heart failure.. Stroke.. 65
HBAM for Planning and Forecasting Estimated Annual Rates of Growth by LHIN 4.5% 4.0% Annual Rate of Growth by Program 3.5% 3.0% 2.5% 2.0% 1.5% 1.0% 0.5% 0.0% 01 Erie St. Clair 02 South West 03 Waterl oo Wellin gton 04 Hamilt on Niagar a Haldim and Brant 05 Central West 06 Mississ auga Halton 07 Toront o Central 08 Central 09 Central East 10 South East 11 Champ lain 12 North Simcoe Musko ka 13 North- East 14 North- West Homecare 1.7% 1.9% 2.8% 2.1% 3.5% 4.0% 1.6% 3.3% 2.4% 1.7% 2.1% 3.5% 1.4% 0.5% 2.3% Hospitals 1.5% 1.6% 2.2% 1.7% 3.2% 3.4% 1.7% 2.8% 2.2% 1.6% 1.6% 3.0% 1.0% 0.5% 2.0% Overall 66
HBAM for Managing Resources Profile of High Use Population 100% Healthcare Estimated Costs Cumulative Distribution % of total LHIN/Provincial Estimated Cost 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% This chart shows: The proportion of Central LHIN s total hospital and home care weighted services accounted for by each percentage of Central LHIN s population (solid red line) for one fiscal year The proportion of Ontario s total hospital and home care weighted services accounted for by each percentage of Ontario s population (dashed black line) for one fiscal year For example: % of Population Line of Perfect Equality Central (1%=16,050) Ontario (1%=126,870) 1% of Central LHIN s population accounts for 52% of Central LHIN s total hospital and home care services 67
HBAM for Improving Quality High quality evidence that dramatic results can be achieved with focused homebased post-discharge maneuvers for high-risk populations Significant reductions in length of stay, readmission rates, mortality rates Benefits demonstrated for multiple diseases Congestive Heart Failure, Stroke, Acute Coronary Syndrome Some trials have demonstrated significant and long term reductions in readmission for CHF patients treated with home based interventions (HBI) Ontario average readmission rate is 35% higher that rates achieved with HBIs 68
Investigating Hospital Costs A hospital s actual expenses are 8.5% higher than its HBAM expenses Actual HBAM Variance Expenses Expenses Hospital $385,000,000 $355,000,000 8.5% 69
Investigating Hospital Costs By disaggregating of the hospital s actual and HBAM expenses by care type, it is shown that biggest contributor to the hospital s overall variance in expenses is the acute care type Actual HBAM Variance Expenses Expenses Acute 212,934,000 184,038,000 15.7% MH 12,622,000 12,411,000 1.7% ER 31,552,000 30,950,000 1.9% CCC 3,382,000 3,212,000 5.3% Rehab 16,710,000 16,589,000 0.7% Other 107,800,000 107,800,000 0.0% Hospital Total $385,000,000 $355,000,000 8.5% 70
Investigating Hospital Costs Investigating the extent of volume variance and unit cost variance in the hospital s acute care type, the hospital s acute unit cost is 15.2% higher than its HBAM acute unit cost Volumes Unit Cost Actual HBAM Variance Actual HBAM Variance Acute 35,000 34,836 0.5% $6,084 $5,283 15.2% This information can be used to identify the cause of the hospital s high variance in acute unit cost 71
Closing Remarks The HBAM methodology is a made in Ontario approach which builds on Ontario s data investments and experience in funding models by incorporating the best practices from other jurisdictions and leading edge developments from health services research. In the coming months, further documentation will be developed with focus on the details of the HBAM methodology and how the results from the methodology can be used. 72