Abstract
Purpose :
Patients with glaucoma often are non-adherant with eye pressure lowering medications. We utilized decomposition analysis via a conjoint matrix to identify barriers to medication adherance.
Methods :
Four factors were selected: odds of needing a second drop, copay, dosing, and side effects. Each factor had four levels: 0%, 25%, 50%, and 75% for odds; $10, $20, $30, and $40 for copay; twice a day, evening, morning, and every other day for dosing; and blurring, stinging, redness, and none for side effects. Four different 16x16 conjoint matrices were generated to avoid random selection bias. Each package contained one level of each factor. Fifty-nine subjects from Ochsner Health Center in cosmpolitan New Orleans, LA and forty-one subjects from Chabert Medical Center, providing care to a rural, underserved population in Houma, LA, participated in this study. Subjects were asked to rank their best choice from 1 through the 16 options. All scores were averaged by level. Utility was assessed per factor level by the equation (worst score - raw score for a factor level)/rank range. Similar calculations were performed within the individual levels. Percent weighting was calculated by subtracting the sum of the raw scores from each factor level, subtracting that number from the summed scores for the worst possible drop, and dividing by the sum rank range.
Results :
Subjects from each medical center demonstrated a similar linear response to increased copay. Utility fell off faster for Chabert subjects than for Ochsner subjects related to odds of needing a second drop. Fifty percent of the Chabert subject's choice was based on odds of a second drop, whereas 27.7% of the Ochsner subject's choice was based on a second drop. Side effects impacted the Ochsner patient's decision by 29.0% compared to the Chabert subject's choice of 4.2%
Conclusions :
Chabert Medical Center subjects are more influenced by cost than Ochsner subjects. However, the Ochsner Subject pool are more concerned about the glaucoma drop side effect protocol rather than the cost sensitive Chabert subjects. Our results demonstrate that decomposition analysis can be a valuable tool in identifying the preferences of patients in specific populations and may guide the physician in selecting glaucoma medication for their patients. Future research will study use of this technique in other medical conditions..
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.