Residents Preferences for Pharmacist-Managed Clinic in China: A Discrete Choice Experiment

Figures & data

Table 1 Attributes and Attribute Levels for DCE

Attributes	Attribute Levels
Pharmacist-Managed Clinic Core (Focus)	Evaluation of medication regimen (core4)
	Developing or adjusting medication regimen (core3)
	Patient medication education and counseling (core2)
	Follow-up on medication (core1)
Frequency	Full or half-day clinic on weekdays (times3)
	3–4 times per week (times2)
	1–2 times per week (times1)
Type	Physician-pharmacist joint clinic (type3)
	Comprehensive pharmaceutical clinic (type2)
	Pharmaceutical specialist clinic (type1)
Consultation Room	Dedicated consultation room (place3)
	Shared consultation rooms with other clinics (place2)
	Only a consultation window (place1)
Pharmacists’ knowledge and clinical medication practice competency	Good (ability3)
	Fair (ability2)
	Poor (ability1)
Fee	Free (fee3)
	Fixed fee (Ұ10–12) (fee2)
	Charge according to pharmacist title (refer to doctor registration fee) (fee1)

Table 2 An Example of Discrete Choice Question

	Option 1	Option 2	Option 3
Pharmacist-Managed Clinic Core (Focus)	Developing or adjusting medication regimen	Patient medication education and counseling	Neither.
Frequency	Full or half-day clinic on weekdays	3–4 times per week
Type	Physician-pharmacist joint clinic	Comprehensive pharmacy clinic
Consultation Room	Only a consultation window	Shared consultation rooms with other clinics
Pharmacists’ knowledge and clinical medication practice competency	Good	Poor
Fee	Free	Fixed fee (Ұ10)
Which one would you prefer?

Table 3 Characteristics of Respondents Who Passed the Consistency Test

Characteristics	Percentage
Gender
Male	46.15%
Female	53.85%
Age (Mean, SD)	44.98 (19.77)
Education
Completed or did not complete primary school	13.46%
Completed Middle school	15.38%
Completed High school	27.56%
Bachelor or above	43.59%
Marital status
Unmarried	29.49%
Married	61.54%
Divorced or widowed	8.98%
Occupational status
Employed	48.08%
Retired	24.36%
Unemployed	27.56%
Medical insurance (multiple choice)
Participation in basic endowment insurance for the urban working group	62.18%
Participation in social endowment insurance for urban and rural residents	41.67%
Participation in commercial medical insurance	14.10%
Uninsured	3.85%
Self-assessed health status
Very poor	1.28%
Poor	7.69%
Fair	22.44%
Good	46.79%
Very good	21.79%
Chronic disease
Yes	30.77%
No	69.23%
Monthly income per capita of household (CNY)
<1000	1.28%
1000–3000	11.54%
3000–5000	23.72%
5000–10,000	31.41%
>10,000	25.64%
Have used PMCs
Yes	32.69%
No	67.31%
Degree of knowledge of PMCs
Very well	2.56%
Fairly well	15.38%
Fair	31.41%
Fairly poor	33.97%
Very poor	16.67%

Table 4 CLOGIT and MXL Results

Attributes and Levels	Mixed Logit Model (β±SE)	Conditional Logit Mode (β±SE)
Asc	1.37±0.40**	0.57±0.22*
Pharmacist-Managed Clinic Core (Focus) (ref=core1)
Core2	0.11±0.19	0.03±0.14
Core3	−0.01±0.15	−0.02±0.09
Core4	0.11±0.18	0.11±0.12
Type (ref=type1)
Type2	0.17±0.12	0.21±0.08*
Type3	0.36±0.12**	0.34±0.08***
Consultation Room (ref=place1)
Place2	0.06±0.12	−0.04±0.08
Place3	0.65±0.12***	0.30±0.07***
Frequency (ref=times1)
Times2	0.06±0.12	−0.03±0.07
Times3	0.11±0.12	0.08±0.07
Pharmacists’ knowledge and clinical medication practice competency (ref=ability1)
Ability2	0.67±0.12***	0.37±0.08***
Ability3	1.59±0.20***	1.01±0.13***
Fee (ref=fee1)
Fee2	0.44±0.12***	0.27±0.08**
Fee3	1.15±0.21***	0.67±0.14***
Standard deviations
Asc	2.63±0.30
Pharmacist-Managed Clinic Core (Focus) (ref=core1)
Core2	0.75±0.28**
Core3	0.05±0.30
Core4	0.50±0.29
Type (ref=type1)
Type2	0.002±0.39
Type3	0.51±0.18**
Consultation Room (ref=place1)
Place2	0.21±0.28
Place3	0.48±0.17**
Frequency (ref=times1)
Times2	0.21±0.23
Times3	0.51±0.16**
Pharmacists’ knowledge and clinical medication practice competency (ref=ability1)
Ability2	0.48±0.19
Ability3	1.91±0.21***
Fee (ref=fee1)
Fee2	0.81±0.15***
Fee3	1.80±0.21***
Number of observations	5616	5616
chi-squared	389.29	373.13
Log likelihood	−1394.3724	−1589.0186
AIC	2844.745	3206.037
BIC	3030.479	3298.905

Notes: *p < 0.05, **p < 0.01, ***p < 0.001.

Table 5 Hierarchical Importance of Attributes Based on Partial Log-Likelihood Analysis

Attributes and Levels	Log Likelihood	Partial Effect-change in Log Likelihood	Relative Effect-%sum Of Change in Log Likelihood
None	−1394.37
ability3	−1561.85	−167.48	46.33%
fee3	−1499.16	−104.79	28.99%
ability2	−1418.62	−24.24	6.71%
place3	−1415.49	−21.12	5.84%
fee2	−1412.52	−18.15	5.02%
type3	−1404.68	−10.31	2.85%
place2	−1397.83	−3.45	0.96%
times2	−1397.44	−3.07	0.85%
core2	−1397.42	−3.05	0.84%
times3	−1396.77	−2.39	0.66%
core3	−1396.35	−1.97	0.55%
core4	−1395.66	−1.29	0.36%
type2	−1394.55	−0.17	0.05%

Table 6 Latent Class Model Results with Only the Main Effects

Variable	Class1 Experiential Type		Class2 Content Type		Class3 Economic Type
	β	p	β	p	β	p
ASC	2.420	0.000	2.388	0.224	−2.391	0.000
core2	−0.248	0.212	2.013	0.031	−0.268	0.343
core3	−0.285	0.073	1.422	0.015	−0.100	0.698
core4	−0.167	0.385	1.093	0.179	−0.228	0.442
type2	0.180	0.131	−0.264	0.525	0.272	0.247
type3	0.313	0.004	0.370	0.233	0.542	0.018
place2	0.097	0.432	0.019	0.960	−0.224	0.310
place3	0.359	0.002	0.769	0.022	0.339	0.166
times2	0.160	0.183	−0.456	0.212	−0.106	0.663
times3	0.212	0.056	−0.619	0.148	0.170	0.415
ability2	0.287	0.017	0.574	0.101	1.517	0.000
ability3	0.428	0.000	0.944	0.005	3.482	0.000
fee2	−0.100	0.396	1.785	0.000	0.495	0.019
fee3	−0.281	0.057	3.368	0.000	2.236	0.000
chronic	−0.570	0.197	−0.284	0.612	--	--
experience	0.575	0.187	−0.269	0.636	--	--
cons	0.493	0.100	−0.101	0.771	--	--
Class Share	0.484		0.292		0.224

Table 7 Goodness-of-Fit of the LC Model

Measure	AIC	BIC
Latent class model	2854.789	3173.191

Abbreviations

PMC, Pharmacist-Managed Clinic; DCEs, Discrete choice experiments; CLOGIT, Conditional logit model; MXL, mixed logit model; AIC, Akaike information criterion; BIC, Bayesian information criterion; LCA, Latent Class Analysis.

Data Sharing Statement

The data that support the findings of this study are not openly available due to reasons of privacy and are available from the corresponding author upon reasonable request.