understanding temporal patterns
play

Understanding Temporal Patterns in Hypertensive Drug Therapy 1 - PowerPoint PPT Presentation

Understanding Temporal Patterns in Hypertensive Drug Therapy 1 Margret Bjarnadottir, 2 Sana Malik , 2 Catherine Plaisant, 3 Eberechukwu Onukwugha 1 Smith School of Business, University of Maryland, College Park 2 Department of Computer Science,


  1. Understanding Temporal Patterns 
 in Hypertensive Drug Therapy 1 Margret Bjarnadottir, 2 Sana Malik , 2 Catherine Plaisant, 3 Eberechukwu Onukwugha 1 Smith School of Business, University of Maryland, College Park 2 Department of Computer Science, University of Maryland, College Park 3 School of Pharmacy, University of Maryland, Baltimore May 28, 2015 — HCIL 32 nd Annual Symposium

  2. ๏ Worsened conditions ๏ Adverse outcomes ๏ Increased risk of death 2

  3. Hospitalization costs due to medication non-adherence are estimated as high as $13 billion annually. Sullivan et al., “Noncompliance with medication regimens and subsequent hospitalizations: a literature analysis and cost of hospitalization estimate,” Journal of Research in Pharmaceutical Economics, Vol. 2, No. 2. (1990), pp. 19-33. 3

  4. Medication Possession Ratio (MPR) days supplied ∑ length of observatio n period 4

  5. Medication Possession Ratio (MPR) days supplied ∑ length of observatio n period time Study ¡end Study ¡start 4

  6. Medication Possession Ratio (MPR) days supplied ∑ = ¡100% length of observatio n period time Study ¡end Study ¡start 4

  7. Medication Possession Ratio (MPR) days supplied ∑ = ¡83% length of observatio n period time Study ¡end Study ¡start 4

  8. Medication Possession Ratio (MPR) days supplied ∑ = ¡75% length of observatio n period time Study ¡end Study ¡start 5

  9. Hypertensive Drug Therapy Angiotension-Converting Enzyme-Inhibitors ACE ARB Angiotension II Receptor Blockers CCB Calcium Channel Blockers Beta Beta Blockers Diur Diuretics 6

  10. Hypertensive Drug Therapy time

  11. Hypertensive Drug Therapy time

  12. Hypertensive Drug Therapy time

  13. Hypertensive Drug Therapy time

  14. Hypertensive Drug Therapy time

  15. Hypertensive Drug Therapy time

  16. Hypertensive Drug Therapy time

  17. Hypertensive Drug Therapy time

  18. Hypertensive Drug Therapy time

  19. Hypertensive Drug Therapy time

  20. Hypertensive Drug Therapy time

  21. Hypertensive Drug Therapy time

  22. Hypertensive Drug Therapy time

  23. Hypertensive Drug Therapy time

  24. Hypertensive Drug Therapy time

  25. Hypertensive Drug Therapy time

  26. Hypertensive Drug Therapy time

  27. Hypertensive Drug Therapy time

  28. Hypertensive Drug Therapy time

  29. Hypertensive Drug Therapy time

  30. Hypertensive Drug Therapy time

  31. Hypertensive Drug Therapy time

  32. Hypertensive Drug Therapy time

  33. Hypertensive Drug Therapy time

  34. Can we use visualization tools to more accurately understand adherence patterns in hypertensive drug therapy?

  35. Research Questions ๏ Can we identify good vs. bad patterns? ๏ Can we understand patient behavior? 11

  36. Data ๏ Commercial prescription claims ๏ 900,000 individuals (16 million claims) ๏ 5 Drug Classes Angiotension-Converting Enzyme-Inhibitors ACE Angiotension II Receptor Blockers ARB Calcium Channel Blockers CCB Beta Blockers Beta Diuretics Diur 12

  37. Data Simplification ๏ Partitioning ๏ Temporal Windowing ๏ Interval Merging 13

  38. Data Simplification ๏ Partitioning ๏ Temporal Windowing ๏ Interval Merging Dataset 14

  39. Data Simplification — Single ๏ Partitioning } 2 Drugs ๏ Temporal Windowing } ๏ Interval Merging 3 Drugs } 4 Drugs — All 5 14

  40. Data Simplification ๏ Partitioning ๏ Temporal Windowing ๏ Interval Merging time Start ¡of ¡record 15

  41. Data Simplification ๏ Partitioning ๏ Temporal Windowing ๏ Interval Merging time Start ¡of ¡record 2 ¡years 15

  42. Data Simplification ๏ Partitioning ๏ Temporal Windowing ๏ Interval Merging time 16

  43. Data Simplification ๏ Partitioning ๏ Temporal Windowing ๏ Interval Merging time 16

  44. How do allowable gap and overlap assumptions affect adherence analysis?

  45. Sensitivity Analysis: Allowable Gap 0 days 7 days 15 days 30 days 18

  46. Sensitivity Analysis: Allowable Gap 0 days 7 days 15 days 30 days 18

  47. Sensitivity Analysis: Allowable Gap 0 days 7 days 15 days 30 days 18

  48. Sensitivity Analysis: Allowable Gap 0 days 7 days 15 days 30 days 18

  49. Sensitivity Analysis: Allowable Overlap 0 days 7 days 15 days 30 days 19

  50. Sensitivity Analysis: Allowable Overlap 0 days 7 days 15 days 30 days 19

  51. Sensitivity Analysis: Allowable Overlap 0 days 7 days 15 days 30 days 19

  52. Sensitivity Analysis: Allowable Overlap 0 days 7 days 15 days 30 days 19

  53. ACE ARB Single Drugs CCB Beta Diur 20

  54. ACE ARB Single Drugs CCB Beta Diur 20

  55. ACE 2 Drugs Beta Both 21

  56. ACE 2 Drugs Beta Both 21

  57. ACE 2 Drugs Beta Both false overlap 22

  58. ACE 2 Drugs Beta Both 23

  59. ACE 2 Drugs Beta Both 24

  60. All 5 Drugs 2,200 patients 25

  61. All 5 Drugs 2,200 patients 26

  62. All 5 Drugs 2,200 patients 26

  63. All 5 Drugs 2,200 patients 26

  64. All 5 Drugs 2,200 patients 26

  65. All 5 Drugs 2,200 patients 27

  66. All 5 Drugs 2,200 patients 27

  67. ACE ARB Starting on Diuretic CCB Beta 150,000 patients Diur 28

  68. ACE ARB Starting on Diuretic CCB Beta 150,000 patients Diur 28

  69. ACE ARB Starting on Diuretic CCB Beta 150,000 patients Diur 28

  70. Summary ๏ Visualization facilitated rapid understanding of complex patient behavior ๏ Interactions allow systematic analysis of episode modeling 29

  71. Understanding Temporal Patterns 
 in Hypertensive Drug Therapy Sana Malik maliks@cs.umd.edu www.cs.umd.edu/hcil/eventflow Thanks to Sophia Wu. Support from the University of Maryland Center for Health-related Informatics and Bioimaging (CHIB). 30

Recommend


More recommend