SLIDE 1 Exploring !Temporal !Patterns !in ! Hypertensive !Drug !Therapy
- 1. MSIS, Smith School of Business
- 2. Assistant Prof, Smith School of Business
- 3. Assistant Prof, School of Pharmacy
- 4. Research Scientist, HCIL
- 5. Ph.D student in Computer Science
Sophia Wu1, Margret Bjarnadottir2, Eberechukwu Onukwugha3, Catherine Plaisant4, Sana Malik5
SLIDE 2 INTRODUCTION
Patients’ Adherence to Medication Medication Possession Ratio (MPR)
- Does not adequately capture different adherence patterns of
patients, which vary widely. Great importance as non-adherence can lead to worsening of conditions and health decline.
SLIDE 3
Observe and Summarize Common Patterns in Hypertensive Drug Therapy
SLIDE 4 DATA !DESCRIPTION
Pharmacy claims of 493,022 individuals 5 Drug Classes
Angiotension-Converting Enzyme-Inhibitors (Ace) Angiotension II Receptor Blockers (ARB) Calcium Channel Blockers (CCB) Beta blockers (Beta) Diuretics
SLIDE 5 IDEAL !DRUG !USAGE !PATTERN
30 30 30 30 30 30 30
ACE
30 30 30 30 30
Days
SLIDE 6 IDEAL !DRUG !USAGE !PATTERN
30 30 30 30 30 30 30 30 30 30 30 30 30
ACE Beta
30 30 30 30 30
Days
SLIDE 7 IDEAL !DRUG !USAGE !PATTERN
30 30 30 30 30 30 30 30 30 30 30 30 30
ACE Beta
……
30 30 30 30 30
Days
SLIDE 8
What !does !the !whole !picture !look !like?
1st !PASS
SLIDE 9
Randomly selected 5000 events (180 individuals)
SLIDE 10
What !if !we !narrow !down !a !little !bit?
1st !PASS 2nd !PASS
SLIDE 11
SLIDE 12
SLIDE 13 MATCH WEIR
CCB Ace Beta Diur
SLIDE 14
What !is !a !good !pattern?
1st !PASS 2nd !PASS 3rd !PASS
SLIDE 15 NHLBI (2003), JNC 7 Express
SLIDE 16 NHLBI (2003), JNC 7 Express
SLIDE 17 NHLBI (2003), JNC 7 Express
SLIDE 18
7
Search HF_4289 occurring during the CCB Drug usages&medical records for heart failure patients with ICD9 4289 HF_4289 CCB
SLIDE 19
Why !are !Those !Heart !Failure ! Patients !Given !CCB?
SLIDE 20
CCB Non-dihydropyridines (Good) Dihydropyridines (Bad)
SLIDE 21
CCB Non-dihydropyridines (Good) Dihydropyridines (Bad)
SLIDE 22
Are !there !any !heart !failure !patients !on ! bad !CCB?
1st !PASS 2nd !PASS 3rd !PASS 4th !PASS
SLIDE 23
Bad CCB HF_any
SLIDE 24
28
Search HF_any occurring during the Bad CCB Drug usages&medical records for heart failure patients on Bad CCB
SLIDE 25
MEDICAL !STUDY
Atrial Fibrillation (AF) CCB Non-dihydropyridines (good) Dihydropyridines (bad)
SLIDE 26
Do !those !heart !failure !patients !on !bad ! CCB !have !AF?
1st !PASS 2nd !PASS 3rd !PASS 4th !PASS 5th !PASS
SLIDE 27
Search AF_any not occurring
89
Drug usages&medical records for heart failure patients on Bad CCB
SLIDE 28
37454: Heart failure patients in total 416: Heart failure patients on CCB 89: Heart failure on bad CCB without AF
SLIDE 29
37454: Heart failure patients in total 416: Heart failure patients on CCB 89: Heart failure on bad CCB without AF 0.24% among the heart failure Individuals
SLIDE 30
37454: Heart failure patients in total 416: Heart failure patients on CCB 89: Heart failure on bad CCB without AF 0.24% among the heart failure Individuals
21% among the heart failure on CCB
SLIDE 31
CRITICAL !FINDING
It is not compliant with medical guideline.
SLIDE 32
CRITICAL !FINDING
It is not compliant with medical guideline.
Doctor’s mistake
SLIDE 33
CRITICAL !FINDING
It is not compliant with medical guideline.
Doctor’s mistake
SLIDE 34
What !does !the !pattern !look !like !before !and ! after !1st !heart !failure !inpatient !visit?
1st !PASS 2nd !PASS 3rd !PASS 4th !PASS 5th !PASS 6th !PASS
SLIDE 35
Aggregating HF claims into 5 categories
SLIDE 36 Category Place of Services Code
inpatient 21, 51, 56, 61
all others urgent 20, 23, 41, 42 hospice 34 SNF(skilled nursing facility) 31, 32, 33
SLIDE 37
SLIDE 38 Cleaning multiple records
- f inpatient visit
- n the same day
Aggregating HF claims into 5 categories
SLIDE 39
Search → Add Constraint
SLIDE 40
SLIDE 41 Cleaning multiple records
- f inpatient visit
- n the same day
Aligning by the first HF inpatient visit Aggregating HF claims into 5 categories
SLIDE 42
Drug usages&medical records for heart failure patients on Bad CCB
SLIDE 43 Patients with AF claims: started by taking good CCB and ACE → not long after first heart failure inpatient visit, dropped good CCB but took bad CCB instead. Patients without AF claims: started to take hypertensive drugs after first heart failure inpatient visit → dropped bad CCB Patients without AF claims: started to take hypertensive drugs after first heart failure inpatient visit → continued to take bad CCB for some periods of time
SLIDE 44 CONCLUSION
Patterns are far from ideal
- Doctors may have given wrong prescriptions
(use of bad CCB)
- EventFlow:
- visualization reveals limitations of Medical Possession Ratio
- detect common patterns & specific cases
SLIDE 45 FUTURE !AVENUES
Clustering: discovering underlying patterns of behavior of
patients taking medicine, then analyzing the different clusters in EventFlow
- Statistic Analysis: logistic regression, including Charlson
Comorbidity Index (CCI), to quantify drug impacts on outcomes
- EventFlow:
- Generating hypothesis
(eg. Patient on bad CCB may have higher readmission rate)
- Supporting the statistic analysis findings
SLIDE 46 THANK !YOU
For more information: Email: zhusenru.wu@rhsmith.umd.edu margret@rhsmith.umd.edu Website: http://www.cs.umd.edu/hcil/eventflow May 29, 2014 HCIL Symposium
