Patients’ Preferences Regarding Osteoarthritis Medications: An Adaptive Choice-Based Conjoint Analysis Study

Figures & data

Table 1 Attributes and Levels Used the Pilot and in This ACBC Study

Attribute		Levels
Pilot Study	Current Study	Pilot Study	Current Study
Availability		Prescription drug Over-the-counter drug Internet purchase drug
Way of taking the medication		Cream/Gel Oral
Frequency		Once a day Twice a day 3–4 times a day As needed
How much you would expect to benefit	How much you would expect mobility improvement	Expect 25% benefit Expect 50% benefit Expect 75% benefit	Expect 25% mobility improvement Expect 50% mobility improvement Expect 75% mobility improvement
	How much you would expect pain reduction		Expect 25% pain reduction Expect 50% pain reduction Expect 75% pain reduction
Risk of gastric ulcer		No risk of gastric ulcer Low risk of gastric ulcer Moderate risk of gastric ulcer High risk of gastric ulcer
Risk of addiction		No risk of addiction Low risk of addiction Moderate risk of addiction High risk of addiction
Risk of kidney and liver impairment		No risk of kidney and liver impairment Low risk of kidney and liver impairment Moderate risk of kidney and liver impairment High risk of kidney and liver impairment
Risk of heart attacks and strokes		No risk of heart attacks and strokes Low risk of heart attacks and strokes Moderate risk of heart attacks and strokes High risk of heart attacks and strokes

Figure 1 An example of Build Your Own (BYO) question.

Figure 2 Examples of the screening section questions.

Figure 3 An example of the choice task question.

Table 2 Sawtooth recommended 8 Points to Be Checked Before Fielding an ACBC Study

Question	Answer
Is this an appropriate study for ACBC? Appropriate studies typically involve about five or more attributes.	Yes. 9 attributes included.
Have you used the Test Design capability on the Design tab to have the software generate dummy (robotic) respondents that answer randomly? Have you examined the attribute level frequency report for these dummy test records? Does each non-BYO level occur at a minimum of 2 times (3 times, preferably)?	Yes. Generated and examined dummy respondents with random responses. Each non-BYO question appeared 2 times.
Does each respondent evaluate no more than about 7 levels per attribute?	Yes. Maximum is 4 levels per attribute.
If you are studying a large number of attributes or levels per attribute (such that using constructed lists to discard levels and attributes from consideration within the ACBC survey is necessary), is your sample size sufficient to stabilize the parameters across the full list of attributes and levels? Are you willing to assume that discarded attributes are entirely unimportant to the respondent?	N/A. This study does not include a large number of attributes and levels.
When you take the questionnaire, do the Unacceptable and Must-Have questions properly identify (using correct “at least” or “at most” phrasing) the levels you have consistently selected or rejected? (This confirms you set the correct worst to best or best to worst attribute direction).	Yes. The questionnaire was trialed with extreme scenarios and has defined the Unacceptable & Must-Have questions and selected the levels appropriately.
Take a practice questionnaires yourself, making sure that the computed part-worths reflect the preferences you expressed in the questionnaire	Yes. The questionnaire was trialed with specific scenarios and produced part-worths that match the ranking of the levels in these scenarios.
Have you asked colleagues or a small convenience sample of target respondents to take the survey? Have you debriefed them regarding their experiences and analyzed their data? Were any sections/instructions confusing? Do the computed part-worth utilities reasonably reflect their individual preferences? Confusing? Do the computed part-worth utilities reasonably reflect their individual preferences?	Yes. This was completed in the feasibility studies and was repeated by colleagues prior to fielding.
Have you fielded the study (a “soft launch”) among a few target respondents, examining the same issues as directly above?	Yes. This was completed in the feasibility studies and was repeated by colleagues prior to fielding.

Table 3 Frequency of the Participants’ Age Groups, Gender, Years Suffering from OA, Pain Interfere with Normal Life, and Using Medications for OA Treatment

Age Group	Frequency	Male		Female
Under 30	1	0		1
30–39	3	0		3
40–49	9	2		7
50–59	8	3		5
60–69	8	2		6
70–79	9	4		5
Over 79	5	2		3
Total	43	13		30
Years suffering from OA			Frequency
Less than 5 years			14
5–10 years			12
More than 10 years			16
I do not know			1
Total			43
How much did pain interfere with normal life?			Frequency
Not at all			3
A little bit			9
Moderately			16
Quite a bit			11
Extremely			4
Total			43

Medication	Frequency
Paracetamol	39
Capsaicin	2
Glucosamine	20
Diclofenac	21
Etodolac	3
Ibuprofen	28
Meloxicam	1
Naproxen	12
Codeine or Dihydrocodeine	24
Tramadol	13
Fentanyl	0
Morphine	5
Oxycodone	2
Others	0

Figure 4 The average (relative) importance and Standard Deviation (SD) for all attributes.

Table 4 The Utilities of All Levels

Attribute	Level	Utilities Mean	Utilities Interval	SD
Availability	Prescription drug	39.3		26.9
	Over-the-counter drug	19.0	20.3	23.3
	Internet purchase drug	−58.3	77.3	33.5
Frequency	As needed	8.2		19.6
	Once a day	6.6	1.6	16.9
	Twice a day	−2.4	9	16.9
	3–4 times a day	−12.4	10	19.8
Way of taking the medication	Oral	17.7		21.4
	Cream/Gel	−17.7	35.4	21.4
Pain reduction	Expect 75% pain reduction	22.1		25.0
	Expect 50% pain reduction	1.1	23.2	8.0
	Expect 25% pain reduction	−23.2	22.1	19.2
Mobility improvement	Expect 75% mobility improvement	15.5		16.0
	Expect 50% mobility improvement	0.5	15	7.9
	Expect 25% mobility improvement	−16.0	16.5	14.7
Risk of addiction	No risk of addiction	67.3		25.0
	Low risk of addiction	37.4	29.9	20.8
	Moderate risk of addiction	−6.4	43.8	25.2
	High risk of addiction	−98.2	91.8	34.4
Risk of stomach side effects	No risk of stomach ulcer	62.9		28.9
	Low risk of stomach ulcer	30.5	32.4	17.3
	Moderate risk of stomach ulcer	−24.4	54.9	19.7
	High risk of stomach ulcer	−68.9	44.5	23.4
Risk of kidney and liver side effects	No risk of kidney or liver impairment	91.8		27.3
	Low risk of kidney or liver impairment	36.8	55	17.0
	Moderate risk of kidney or liver impairment	−63.1	99.9	18.2
	High risk of kidney or liver impairment	−65.5	2.4	16.7
Risk of heart attacks and strokes	No risk of heart attacks and strokes	93.5		32.5
	Low risk of heart attacks and strokes	23.1	70.4	28.0
	Moderate risk of heart attacks and strokes	−34.9	58	20.6
	High risk of heart attacks and strokes	−81.6	46.7	22.4

Note: Utilities mean in this table may not add up to ZERO, due to rounding.