Clinical and Health Workforce Implications of Improving Population - - PowerPoint PPT Presentation

Clinical and Health Workforce Implications of Improving Population Health

Annual Research Meeting

June 25, 2017 Tim Dall, tim.dall@ihsmarkit.com Will Iacobucci Ritashree Chakrabarti Frank Chen Terry West

Research Question

• How will improvements in

population health affect aggregate demand for health care services and providers? > Will demand decline because people are healthier thus requiring fewer services in hospitals or other settings? > Will demand increase because people live longer?

2

Population Health Improvement Scenario Modeled

• Modeling assumptions

> Sustained 5% body weight loss for overweight and obese adults > Improved blood pressure, cholesterol, and blood glucose levels for adults with elevated levels – 34.42 mg/dL (CI, 22.04-46.40) reduce total blood cholesterol 1 – 14.5 mm Hg reduction in systolic blood pressure by and 10.7 mm Hg reduction diastolic blood pressure 2 – 1 percentage point annual reduction in hemoglobin A1c until diabetes control reached at 7.5% 3 > Smoking cessation

• Hypothetical scenario covers portion of population health/preventive care goals

1Taylor et al. Statins for the primary prevention of cardiovascular disease. Cochrane Database Syst Rev 2013;1:CD004816. 2Baguet et al. Updated meta-analytical approach to the efficacy of antihypertensive drugs in reducing blood pressure.

Clin Drug Investig. 2007;27(11):735-753.

3Sherifali et al. The effect of oral antidiabetic agents on A1C levels: a systematic review and meta-analysis. Diabetes

• Care. 2010;33(8):1859-1864.

3

Disease Prevention Microsimulation Model (DPMM) Healthcare Demand Microsimulation Model (HDMM) Health Workforce Supply Model (HWSM) Medical Expenditure Model (MEM)

Healthcare Simulation Model

Healthcare Simulation Model

• Four integrated components: demand, supply, disease, expenditures
• Supports work with governments, associations, hospital systems,

health plans

• National, state, and local projections incorporating population health

risk factors

• Microsimulation: individuals are the unit of analysis

4

Publications Using The Model

6

5

The DPMM is designed to simulate disease onset and economic

• utcomes under disease prevention/health promotion scenarios

MODEL CHARACTERISTICS PROJECTION HORIZON PERSPECTIVE OUTCOMES

• Microsimulation model: individual people are modeled
• Markov approach: current characteristics and status used to predict next year’s status
• Monte Carlo simulation: each person is modeled multiple times, and outcomes can differ each time
• Prediction equations come from published clinical trials and observational studies, and original empirical

analysis

• Annual projections until the end of the projection horizon or death (have modeled lifetime,10- and 20-

year projections)

• Can be societal, health-system, employer, individual/household, or payer
• Disease incidence/prevalence, medical expenditures, mortality, economic outcomes (employment,

productivity, earnings, taxes, social security), and quality adjusted life years

The Disease Prevention Microsimulation Model was developed to answer the following questions

• For each person in a given

population and over a specified period of time: > What is the likelihood and timing of disease onset and severity? > How will health affect:

– Health care use? – Medical expenditures? – Employment and productivity? – Quality of life? – Mortality?

• If an intervention changes one
• r more health risk factors,

how will this affect the above questions?

6

Change in disease health states Change in health risk factors Starting health profile Outcomes

• Disease incidence
• Medical costs
• Productivity
• Quality of life
• Mortality

6

7

• Diabetes incidence and prevalence
• Prediabetes incidence and prevalence
• Annual progression rate from prediabetes to diabetes
• Amputation (diabetes-related only)
• Retinopathy (diabetes-related only)

Diabetes & sequelae

• Hypertension
• Ischemic heart disease
• Myocardial infarction
• Stroke
• Congestive heart failure
• Dyslipidemia

Cardiovascular

• Depression
• Alzheimer’s Disease
• Bipolar disorder
• Schizophrenia

Mental & Cognitive

• Gallbladder disease
• Gastroesophageal Reflux Disease (GERD)
• Non-alcoholic fatty liver disease (NAFLD)

Gastroenterology

• Osteoporosis
• Osteoarthritis
• Chronic back pain

Musculoskeletal

• Medical expenditure
• Household/personal income
• Probability of employment
• Social security cost
• Absenteeism
• Life years/death
• QALY

Socioeconomic

• Pneumonia
• Pulmonary embolism
• Asthma
• COPD

Pulmonary

• Breast
• Cervical
• Colorectal
• Endometrial
• Esophageal
• Gallbladder
• Kidney
• Leukemia
• Liver
• Lung
• Multiple myeloma
• Non-Hodgkin's lym.
• Ovarian
• Pancreatic
• Prostate
• Stomach
• Thyroid

Neoplasms

• Obesity
• CKD/ESRD
• Obstructive sleep apnea

Others

The DPMM projects 50+ clinical and economic outcomes

Constructed initial population file from the 2013-2014 National Health and Nutrition Examination Survey (NHANES)

BMI as a key model driver

8 Endocrine

Diabetes (HbA1c) Prediabetes (HbA1c)

Cardiovascular

LVH Hypertension (SBP, DBP) Dyslipidemia (HDL, Total cholesterol) IHD CHF

Direct Effect Disease States Indirect Effect Disease States

Atrial fibrillation Amputation PVD Renal failure CKD Stroke

Myocardial infarction

Blindness Body weight (BMI)

Respiratory

Pneumonia

Pulmonary embolism

Other

Chronic back pain Osteoarthritis

Gallstones & gallbladder

GERD Major depression NAFLD OSA

Cancers

Breast Cervical Endometrial Esophageal Gallbladder Kidney Leukemia Liver NHL Multiple Myeloma Ovarian Pancreatic Prostate Stomach Thyroid Colorectal

Note: Connecting lines show the items in the model that are linked Abbreviations: BMI=body mass index, CHF=congestive heart failure, CKD=chronic kidney disease, DBP=diastolic blood pressure, GERD= gastroesophageal reflux disease, HbA1c=hemoglobin A1c, HDL=high-density lipoprotein, IHD=ischemic heart disease, LVH=left ventricular hypertrophy, NAFLD=non- alcoholic fatty liver disease, OSA=obstructive sleep apnea, PVD=peripheral vascular disease, SBP=systolic blood pressure.

BMI as a key model driver has direct, secondary, and tertiary impact on many outcomes. For example:

BMI HbA1c Diabetes IHD

Example: Use of Cardiology Services

1 Rate ratios from Poisson

regression analysis using 2009-2013 MEPS/2013 NIS.

2 Odds ratios from logistic

regression analysis using 2009-2013 MEPS. Statistically significant at the 0.05 (*) or 0.01 (**) level.

Health Risk & Behavior Economic & Policy Care Delivery Demographics

Cardiologist Cardiology-related Primary Diagnosis Parameter Office Visits1 Outpatient Visits1 Emergency Visits2 Hospital- ization2 Inpatient Days1 Race- Ethnicity Non-Hispanic White 1.00 1.00 1.00 1.00 1.00 Non-Hispanic Black 0.79** 0.97 1.36** 1.32** 1.14** Non-Hispanic Other 0.90** 0.75** 0.86 0.94 1.10** Hispanic 0.79** 0.68** 0.93 0.84** 1.07** Male 1.13** 1.59** 0.89* 1.11 0.97** Age 18-34 years 0.11** 0.24** 0.66** 0.40** 0.84** 35-44 years 0.22** 0.63** 0.95 0.76** 0.80** 45-64 years 0.50** 0.86** 1.05 1.10 0.86** 65-74 years 0.83** 1.21** 1.11 1.50** 0.93** 75+ years 1.00** 1.00** 1.00** 1.00** 1.00 Smoker 0.73** 0.84** 1.22** 1.11 Diagnosed with Hypertension 1.55** 1.13** 3.86** 2.66** Heart disease 8.50** 10.73** 2.93** 3.84** History of heart attack 1.63** 1.36** 2.36** 2.60** History of stroke 1.08** 1.26** 2.92** 3.04** Diabetes 1.15** 1.34** 1.01 1.19** 1.02** Arthritis 1.10** 1.24** 0.96 0.96 Asthma 1.04* 1.08** 1.00 1.07 History of cancer 1.06** 1.11** 1.01 0.99 Body Weight Normal 1.00** 1.00** 1.00** 1.00** Overweight 1.04** 1.09** 0.87** 0.82** Obese 1.11** 1.18** 1.01 1.02 Insured Has insurance 2.61** 2.09** 0.92 1.09 0.99* In Medicaid 1.36** 1.30** 1.59** 1.71** 1.23** In managed care plan 1.00 1.24** 0.99 0.99 Household Income <$10,000 0.90** 0.97 1.23** 1.19** $10,000 to <$15,000 0.92** 0.91** 1.16* 1.20** $15,000 to < $20,000 0.93** 0.93* 0.82 0.99 $20,000 to < $25,000 0.89** 0.73** 1.15 1.06 $25,000 to < $35,000 0.92** 0.96 1.16* 1.05 $35,000 to < $50,000 0.88** 1.07* 0.91 0.93 $50,000 to < $75,000 0.96* 1.17** 0.93 0.82** $75,000 or higher 1.00 1.00 1.00 1.00 Metro Area 1.31** 1.09** 1.07 0.91 1.03**

Population Health Improvement

• National outcomes cumulative 2015 to 2030

> 10.2 million fewer people with heart disease > 3.2 million fewer strokes > 3 million fewer heart attacks > Reduced incidence of cancer and other diseases, e.g., – 2.7 million fewer cases of prostate cancer – 460,000 fewer cases of thyroid cancer – But, more cases of ovarian cancer, stomach cancer, Alzheimer, osteoporosis and other conditions associated with an older, living population

• Initially, improved population health means fewer hospitalizations and less

demand for care

• National, per capita utilization declines by 1-2%

10

By 2030, Additional 6.3 M People Living, Mostly Elderly

11

IHS Markit Inc., The Complexities of Physician Supply and Demand 2017 Update: Projections from 2015 to 2030., Exhibit 28

• Higher utilization
• f care associated

with 6.3 million additional people vs slightly lower utilization from improved health

> Over 6 million additional hospital inpatient days > 1.7 million additional ED visits

289.4 247.9 283.1 220 230 240 250 260 270 280 290 300 2015 2020 2025 2030

Millions of Adults Year

Achieving Population Health Goals Census Bureau Projections

Additional People Still Living in 2030 under the Population Health Scenario by Age Cohort

Residence 35 to 44 45 to 64 65 to 74 75+ Total Community 30,600 1,025,200 2,302,900 2,721,200 6,079,900 Residential Care

• 500

5,800 75,400 81,700 Nursing Home

• 1,900

16,700 119,700 138,300 Total 30,600 1,027,600 2,325,400 2,916,300 6,299,900 % Pop Change 0.1% 1.2% 5.9% 8.4%

12

Implications of Achieving Modeled Population Health Goals: % Difference in RN FTE Demand

13

• 2%

0% 2% 4% 6% 8% 10% 2015 2020 2025 2030

Percent Change in Demand versus Status Quo Year

Residential Care Nursing Home Home Health Total RN Demand Inpatient All Other Outpatient Office Emergency

Implications of Achieving Modeled Population Health Goals: Net Difference in Nurse FTE Demand

In 2030, nurse FTE demand would be higher

• RNs: 106,000
• LPNs: 70,000
• APRNs: >7,700

> 7,700 increase under current delivery patterns > Physician demand would increase by 15,500 FTEs; portion of this increase could be provided by APRNs

14

(20,000)

• 20,000

40,000 60,000 80,000 100,000 120,000 2015 2020 2025 2030

Impact on Full Time Equivalent Nurses Year

RNs LPNs APRNs

Implications of Achieving Modeled Population Health Goals: Net Difference in Physician Demand

15

IHS Markit Inc., The Complexities of Physician Supply and Demand 2017 Update: Projections from 2015 to 2030., Exhibit 29

• Intervention

impact in 2030

> +15,500 FTE increase in total national demand > 8% increase in demand for geriatricians > 9% decrease in demand for endocrinologists

• 3,000
• 2,000
• 1,000

1,000 2,000 3,000 4,000 5,000 6,000 2015 2020 2025 2030

Full-Time-Equivalent Physicians

Year Primary Care Other Surgery Medical Specialties Hospitalist

Discussion and Implications

• Achieving these population goals could rely heavily on nurse practitioners,

registered nurses, physician assistants, nutritionists, primary care providers and others thereby increasing demand for these professions accordingly

• Keeping people healthier might not reduce future demand for health care

services and providers > Shift care from the near future to more distant future > Shift care from some medical specialties (e.g., endocrinology) to others (e.g., geriatric medicine) > Demand impact varies by care delivery setting

• This was a hypothetical scenario, true impact would be smaller/larger

depending on goals achieved

• Future work might look at additional areas of prevention

> Screening, early diagnosis and treatment > Intervention/prevention among underserved/disadvantaged populations

16