GC650 Lecture 2 - Cost Data and Measuring Cost

0.0(0)
studied byStudied by 0 people
0.0(0)
call with kaiCall with Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/27

encourage image

There's no tags or description

Looks like no tags are added yet.

Last updated 12:42 AM on 2/3/26
Name
Mastery
Learn
Test
Matching
Spaced
Call with Kai

No analytics yet

Send a link to your students to track their progress

28 Terms

1
New cards

Health economic analysis is supported by (5)

  1. Health Services and Outcomes Research

  2. Survey Research

  3. Clinical Research

  4. Epidemiology and Pharmacoepidemiology

  5. Preference Research

2
New cards

Types of Cost (3)

  1. Direct Costs - healthcare/medical, non-medical other

  2. Indirect Costs - productivity, transportation, child care

  3. Intangible - Pain/suffering

3
New cards

Types of Cost Data (2)

  1. Retrospective (Claims, EHRs, chart reviews, surveys)

  2. Prospective (RCTs, surveys, Time and Motion)

4
New cards

Types of Resources Consumed for Cost (5)

  1. Hospital Days (per diem, service/procedural costs)

  2. Staff time and benefits

  3. Office visits

  4. Outpatient Medical Supplies

  5. Outpatient pharmacy

5
New cards

Where to get cost data (3)

  1. Public resources

  2. Previously published data

  3. Generated yourself

6
New cards

Retrospective Data - What cost are we measuring?

Actual cost not documented, most US based studies use reimbursed amount

7
New cards

Data Aggregator Workflow (Komodo and TriNetX) (4)

  1. Data from medical records and clinical trials is aggregated

  2. De-ID Process to anonymize

  3. Data is linked to Encrypted User Tokens, treatments linked to individual “Tokens”

  4. Maps changes in healthcare utilization, patient journey mapping

8
New cards

Major approaches for measuring cost (2)

  1. Total incremental cost

  2. Micro-costing

9
New cards

Total Costs and Total Incremental Costs

Involves calculating all healthcare costs for a

patient with condition or treatment X and

subtracting the costs for patients WITHOUT

condition or treatment X

Requires high quality “match” between cases and

controls

10
New cards

Micro-Costing

All care specifically related to the care for a given disease, requires knowledge about coding, all services, all medications

Can be retrospective with claims, prospective with observation or survey

11
New cards

Sentinel Initiative

FDA resource, has a repository of codes and validated algorithms for many conditions, adverse events and outcomes

12
New cards

How can we account for the difference between Total Incremental Cost and Microcosting?

Total incremental costs includes all healthcare for the patient population of interest, while microcosting typically only includes care for the specific condition of interest

13
New cards

Data from Provider Surveys (3)

  1. Estimate time spent on activities

  2. Record number & type of staff involved

  3. Estimate percent of patients with various characteristics

14
New cards

Data from Patient Surveys (2)

  1. Estimate time spent on and costs of (such as transportation, waiting, childcare)

  2. Estimate out of pocket expenses

15
New cards

National Health and Wellness Survey (NHWS)

Large nationwide survey for health data

16
New cards

Prospective Sources - What is Cost?

Typically collect Resource Utilization, need to apply a dollar amount

17
New cards

Ways to Apply a Dollar Amount to Prospective Utilization Data (3)

  1. Cost reports for hospital data (cost-charge ratio)

  2. Hourly rates for staff time

  3. Medicare reimbursement rates

18
New cards

Prospective data collection methods (3)

  1. RCTs

  2. Time and Motion

  3. Surveys

19
New cards

RCTs Advantages (3)

  1. High internal validity for efficacy

  2. Links measurement of the health outcome with the cost

  3. Accurate

20
New cards

RCTs Disadvantages (5)

  1. Not generalizable

  2. May capture protocol driven costs

  3. Time frame is typically short

  4. Expensive to collect cost data, and not the primary focus

  5. Typically not powered to examine differences in cost between treatment arms

21
New cards

Time and Motion Studies

To determine how much time different staff members spend on different tasks related to a disease/condition or treatment

Convert time to human resource costs

22
New cards

Statistical Analysis of Costs (4)

  1. Ordinary least squares (linear regression) is not appropriate, cost data not normally distributed

  2. Matching cases and controls

  3. Censoring is an issue (patient dropout or death), costs are highest right before death

  4. Advanced techniques exist but are not widely used, require a statistician

23
New cards

Adjusting Costs for Inflation

Most of the time we do not find costs from the time frame we wish to model, so we adjust costs using the Consumer Price Index Inflation Calculator

24
New cards

Measurement Considerations (Covered in Pharmacoepidemiology) (7)

  1. Accuracy of diagnoses

  2. Length of follow-up and ascertaining cause of loss to follow-up

  3. Population identification

  4. Treatment identification

  5. Adherence

  6. Outcome ascertainment

  7. Period of Observation

25
New cards

Incidence Costs

Beginning with new diagnosis or new treatment, often best approach for generating cost inputs for an economic model

26
New cards

Prevalence Costs

Snapshot in time, captures people at different time points in their disease state or treatment process, often most relevant to a health plan/payer

27
New cards

Considerations in Assessing Costs (4)

  1. Time Frame

  2. Clinical Efficiency

  3. Price Changes

  4. Choice of Comparator

28
New cards

Consumer Price Index (CPI)

Measures the average change in prices paid by consumers over time for a “basket” of goods and services, measures inflation and allows for comparability of previously collected price data with today’s prices