Retrospective databases

Question 1

TYPES OF HEALTHCARE DATABASES

National/regional registries
National health survey data
Insurance claims records
Electronic medical records
Wearables and apps

Question 2

____________: a systematic collection of standardized data about a group of individuals

Answer 2

Registries

Question 3

__________ contain clinical data about individuals who have a specific condition as well as demographic information

Answer 3

Patient registries

Question 4

Data can be collected prospectively for specific purpose or used retrospectively to study additional research questions

Question 5

REGISTRIES
Used to evaluate care in ___________ setting
___________outcomes are captured
Do not require comparator/______
Open-label
No random allocation

Answer 5

in real world

Long term

no placebo

Question 6

REGISTRIES: EXAMPLES

The Surveillance, Epidemiology, and End
Results (SEER) Program of the National Cancer Institute (NCI) -collects and publishes cancer incidence and survival data covering ~ 30 percent of the US population

NIH Global Rare Diseases Patient Registry
Data Repository

Question 7

WHICH QUESTIONS CAN BE ANSWERED USING
REGISTRIES?

Studies of heterogeneous patient populations, since unlike randomized trials, registries generally have much _____________ criteria and fewer exclusion criteria.

Follow-up for delayed or long-term benefits or harm, since registries can extend over much _________ periods than most clinical trials (because of their generally lower costs to run and lesser burden on participants).

Surveillance for _________events or of rare diseases.

Studies of health care access and barriers to care.

Answer 7

broader, inclusion

longer

rare

Question 8

NATIONAL HEALTH SURVEYS

National Ambulatory Medical Care Survey
(NAMCS)

National Health and Nutrition Examination
Survey (NHANES)

National Hospital Ambulatory Care Survey
(NHAMS)

Medical Expenditure Panel Survey (MEPS)

Healthcare Cost & Utilization Project (HCUP)

Question 9

NATIONAL HEALTH SURVEYS
Advantages:
____________ information from patients and
providers

____________- representation ( US noninstitutionalized population)

Include HRQoL data (SF-12, EQ5D)

Answer 9

detailed

nationwide

Question 10

NATIONAL HEALTH SURVEYS
Disadvantages

Some variables of interest may be missing
(____________, deductibles)

Varying degree of accuracy depending on
services provided

Some collect event-specific versus patient
specific data (e.g., hospitalization as event)

________ time

Answer 10

co-pays

Lag

Question 11

INSURANCE CLAIMS RECORDS
Used for ______________ purposes

Records from healthcare organizations and
providers:
✓ Physicians, nurse practitioners
✓ Hospitals, outpatient clinics, laboratories
✓ Pharmacies

Answer 11

reimbursement

Question 12

INSURANCE CLAIMS RECORDS
Advantages:
__________ number of patients
______________

Most are reliable (adjudicated and paid
claims)

Standard variables collected

Less lag time

Answer 12

Large

Inexpensive

Question 13

INSURANCE CLAIMS RECORDS
Disadvantages:

Created for administrative use NOT for
pharmacoeconomic/pharmacoepidemiologic
research

__________ data not included

Do not capture some healthcare utilization
(_______medications, medications paid out-of pocket, use of non-plan services)

Eligibility changes restrict long term follow-up

Answer 13

HRQoL

OTC

Question 14

ELECTRONIC HEALTH/MEDICAL RECORDS
(EHR/EMR)

Advantages
More detailed than other databases

More complete picture of care (_______ values,
patient information documented by prescriber)

Answer 14

lab

Question 15

ELECTRONIC HEALTH/MEDICAL RECORDS
(EHR/EMR)
Disadvantages:
___________of implementation
Providers may be unwilling to adopt fully
(‘typing’ vs facing the patient)
Data retrieval is not __________
Formulary restrictions (agent prescribed by
physician may not match the actual
medication patient filled)

Answer 15

High cost

straightforward

Question 16

DATABASES VS RANDOMIZED CONTROLLED
TRIALS (RCTS)
Advantages
_________________
Large samples
Inexpensive
____________ criteria & reanalyze – sensitivity analyses

Answer 16

Databases increases generalizability

Change

Question 17

DATABASES VS RCTS
Disadvantages:
_______________
Incomplete data
____________ (ICD9 and CPT)
Duplicate records
____________ data
Out-of-range data
Changes in coding (ICD10)

Answer 17

Selection bias

miscoding

Missing

Question 18

WHEN CRITIQUING STUDIES USING
RETROSPECTIVE DATABASES

1.Clear explanation of the ___________(types of patients covered Medicaid, Medicare, VA or private insurer)

2. Clear explanation of patient _________ (e.g.,
   study of adherence to oral hypoglycemics- at least two ICD9 for diabetes , at least two Rx fills)
3. Patient eligibility for __________(were patients included only if eligible for the entire study period?)
4. _____________ may often provide additional
   insights (all healthcare costs vs diabetes-related costs only)

Answer 18

database

selection criteria

coverage

sensitivity analyses

Question 19

DIAGRAM/FLOW CHART/TABLE EXPLAINING
CRITERIA FOR STUDY INCLUSION

Example- Study of patients with Medicare Part D prescription drug plan newly initiating DPP-4 inhibitors:

Question 20

YOU ARE INTERESTED TO STUDY OUTCOMES &
TOTAL H/C COSTS ASSOCIATED WITH THE USE OF
ENTRESTO™ AMONG PATIENTS WITH HEART
FAILURE. WHICH DATABASE MAY BE MOST
APPROPRIATE?

Answer 20

Insurance claims

Question 21

THIS STUDY USED INSURANCE CLAIMS DATA AND EVALUATED TREATMENT ADHERENCE AMONG PATIENTS WITH HF NEWLY STARTING ENTRESTO. THE AUTHORS STATED THAT THEY INCLUDED ALL PATIENTS, NOT ONLY THOSE CONTINUOUSLY ELIGIBLE FOR INSURANCE DURING THE STUDY PERIOD. THE ADHERENCE FINDINGS ARE VALID.
A. True
B. False

Question 22

_____________________- answer the
question “Can the medication work? ” – high
internal validity

Answer 22

Randomized controlled trials

Question 23

____________- answer the question “Does it work
in real-world? ”- high external validity

Answer 23

Databases

Question 24

Databases are highly flexible to answer a
variety of different clinical question