Chen et al.

BMC Family Practice (2020) 21:196

https://doi.org/10.1186/s12875-020-01268-w

RESEARCH ARTICLE Open Access

Effect of trust in primary care physicians on

patient satisfaction: a cross-sectional study
among patients with hypertension in rural
China
Wenqin Chen1, Yingchao Feng1, Jiyuan Fang1, Jin Wu1, Xianhong Huang1, Xiaohe Wang1, Jian Wu2 and
Meng Zhang1*

Abstract
Background: In rural areas of China, hypertension is on the rise and it is drawing the Chinese government’s
attention. The health outcomes of hypertension management can be positively impacted by patient satisfaction
with primary care physicians (PCPs), and the influence of patient trust on satisfaction cannot be ignored. This study
aimed to analyze the effect of trust in PCPs on patient satisfaction among patients with hypertension in rural China,
and the influence of patients’ socio-demographic characteristics and hypertension-management-related factors.
Methods: A multi-stage stratified random sampling method was adopted to investigate 2665 patients with
hypertension in rural China. Patient trust and satisfaction were measured using the Chinese version of the Wake
Forest Physician Trust Scale and the European Task Force on Patient Evaluation of General Practice. Multiple linear
regression was used to analyze the factors influencing patient satisfaction, and structural equation modeling was
conducted to clarify the relationships among patient trust and patient satisfaction with PCPs.
Results: Patients’ trust in their PCPs’ benevolence had a positive main effect on all three satisfaction dimensions
(clinical behavior: β = 0.940, p < 0.01; continuity and cooperation: β = 0.910, p < 0.01; and organization of care: β =
0.879, p < 0.01). Patients’ trust in their PCPs’ technical competence had a small negative effect on all three
satisfaction dimensions (clinical behavior: β = − 0.077, p < 0.01; continuity and cooperation: β = − 0.136, p < 0.01;
and organization of care: β = − 0.064, p < 0.01). Patient satisfaction was also associated with region, gender,
insurance status, distance from the nearest medical/health-service institution, and number of visits to PCPs in the
past year.
Conclusions: Patients focused more on physicians’ benevolence than on their technical competence. Hence,
medical humanities and communication skills education should be emphasized for PCPs. Regarding region-based
and health-insurance-based differences, the inequities between eastern, central, and western provinces, as well as
between urban and rural areas, must also be addressed.
Keywords: Patient trust, Patient satisfaction, Hypertension management, Rural area

* Correspondence: [email protected]
1
School of Medicine, Hangzhou Normal University, No. 2318, Yu-hang-tang
Road, Hangzhou, China
Full list of author information is available at the end of the article

© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this article are included in the article's Creative Commons
licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons
licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the
data made available in this article, unless otherwise stated in a credit line to the data.
Chen et al. BMC Family Practice (2020) 21:196 Page 2 of 13

Background Patient satisfaction refers to people’s assessment of the

Hypertension is a common chronic disease and is the health-care services that they receive, and is based on
most important risk factor for cardiovascular and kidney their requirements regarding health, disease, quality of
disease [1]. Worldwide, over one in five adults have high life, etc. [11]. As a patient-reported outcome and a major
blood pressure (of whom 52% have uncontrolled blood component of health-care quality, patient satisfaction
pressure), and there are 9.4 million deaths from hyper- can impact therapeutic outcomes [12]. For patients with
tension complications annually [2]. In 2018, there were hypertension, treatment satisfaction may provide insight
over 200 million people with hypertension in China, and into attitudes toward hypertension treatment [13]. Such
this figure is rising at a rate of 10 million people per year satisfaction is associated with higher adherence to anti-
[3]. The number of adults from rural areas in China pre- hypertensive drugs and improved health-related quality
senting with hypertension has increased rapidly: the of life [2].
prevalence rate was 18.94% from 2004 to 2006, 21.24% Several studies have investigated the factors that influ-
from 2007 to 2009, and 26.68% from 2010 to 2013 [4]. ence patient satisfaction regarding primary health-care
Since 2009, China has considered the health manage- services and have shown that regular visits to a particular
ment of patients with hypertension to be a “national general practitioner (GP), distance from a primary health-
basic public health services project”; a basic public care center, age, gender, socioeconomic status, and health
health service for key populations (e.g., older adults, status are associated with patient satisfaction [14, 15].
women, and children), focusing on key diseases (e.g., Along with these objective factors concerning patients’
chronic or infectious diseases) and meeting residents’ sociodemographic characteristics and health status, pa-
basic health needs. The National Basic Public Health tient satisfaction is also heavily influenced by patients’
Service Project Regulations (2011) stipulate that subjective perceptions and interpersonal relationships
primary-level medical and health institutions should [16]. Notably, patient trust is the foundation of the
provide community health management for hyperten- doctor-patient relationship and leads to perceiving doctors
sion patients. The rural health-service system services as reliable, acting in the patient’s best interests, and pro-
800 million Chinese rural dwellers and directly affects viding support and assistance regarding the patient’s
the health status and service utilization of the rural health problems [17]. Patient trust has been shown to be
population [5]. The health administration’s “Guidelines an important factor in fostering satisfaction [2, 18–20].
for the Management of Hypertension in China” also However, existing China-based research regarding the
propose emphasizing hypertension prevention and man- factors affecting patient satisfaction has mostly focused
agement in rural areas [6]. Long-term adherence to life- on urban areas, with little attention to physicians in rural
style improvement is the cornerstone of associated primary health-care institutions. The few studies in this
treatment approaches, and rational use of antihyperten- latter category have been limited to specific geographic
sive drugs is key to achieving normal blood pressure [6]. areas [21–24]. Although researchers largely acknowledge
Nonadherence to treatment can be due to the patient’s that trust impacts satisfaction, some studies have only
lack of knowledge about the provider’s decision-making performed this analysis from a qualitative viewpoint,
process and low physician credibility [7]. To ensure the while others have used unsuitable or limited question-
effectiveness of this treatment, establishing an enduring naires [25–28]. Univariate and regression analyses have
and harmonious relationship marked by mutual under- been commonly used for analysis [2, 29–32]. Given that
standing between doctors and patients is essential [8]. social science research cannot directly measure trust and
Indeed, trust is the foundation of the doctor-patient rela- satisfaction, measurement error is inevitable [33]. To ad-
tionship, and given that patient satisfaction is an indica- dress this, recent studies have applied structural equa-
tor of health service quality, high levels of trust and tion modeling (SEM) when evaluating patient trust and
patient satisfaction indicate a good relationship between satisfaction [18, 34]. However, few studies have analyzed
patients and service providers [9]. However, studies have the relationship between trust and satisfaction from the
shown that primary-level institutions in the rural health- perspective of refining their internal dimensions. Thus,
service system – in which township health centers or in our study, different dimensions of the two variables
community health-service centers represent primary were scientifically divided according to literature re-
hubs, and village clinics or community health-service search based on the mature scales, and the influence
stations represent the lowest level – have weak service among them was explored by using the structural equa-
capabilities and low resident satisfaction [10]. Import- tion model.
antly, this is not conducive to effective long-term follow- In the present study, the Chinese version of the Wake
up management of rural patients with hypertension and Forest Trust Scale (WFPTS-C) was used to measure
has a negative impact on the prevention and treatment trust among rural-based patients with hypertension. This
of hypertension. scale was developed by Hall and has been used in several
Chen et al. BMC Family Practice (2020) 21:196 Page 3 of 13

countries to examine trust in primary care providers, in- dimensions of patient satisfaction (clinical behavior, con-
cluding physicians, comprised of 4 dimensions: fidelity, tinuity and cooperation, organization of care; H1-H3), and
competence, honesty and global [29, 30]. The Chinese patients’ trust in PCPs’ technical competence will also
version has previously been shown to have beneficial have positive effects (H4-H6). The framework for our the-
psychological attributes among patients by Dong and oretical relationships is shown in Fig. 1. Furthermore, we
Bao [35]. And a two-dimensional model (comprising explore the influence of patients’ socio-demographic
“benevolence” and “technical competence”) has been characteristics and hypertension-management-related
verified as a better fit to the data among Chinese pa- factors on satisfaction, aiming to identify the means
tients than Hall’s four-dimensional model or Bachinger’s to promote patient satisfaction and to improve the
one-dimensional model [17, 36, 37]. doctor-patient relationship and the rate of hyperten-
To measure satisfaction, we used the European Satisfac- sion control in rural areas.
tion Survey Scale (EUROPEP), which is a comprehensive
tool representative of international standards that measures Methods
service satisfaction and was developed through a rigorous Study design and population
design process [38]. The EUROPEP does not evaluate a This study comprised a cross-sectional analysis of rural
specific visit or doctor, but rather, patients’ satisfaction with Chinese hypertension patients and analyzed the effect
doctors regarding services provided “over the last 12 that trust in PCPs has on patient satisfaction. Using a multi-
months” [30]. As this scale measures continuity-related as- stage stratified random sampling method, between Febru-
pects (i.e., repeated visits over 12 months), it captures pa- ary 2017 and May 2018 we surveyed 2665 hypertension
tients’ satisfaction with the normative management patients (response rate: 99.6%; 2665/2675) receiving care
requirements for hypertension, meaning it can be applied from rural primary-health-service institutions. Any ques-
to primary-health-care institutions. Generally, EUROPEP tionnaire with a completion rate of 90% was regarded as ef-
scores reflect two dimensions: clinical behavior (items 1 to ficient. The scores of efficient questionnaires were
16) and organizational mechanisms (items 17 to 23) [39, accounted, and missing data were replaced with the me-
40]. Of these, the former items can be divided into “relation dian. We selected three Chinese provinces to obtain repre-
and communication,” “medical care,” and “information and sentative samples from the eastern, central, and western
support,” and the latter items into “continuity and cooper- regions: Zhejiang, Henan, and Shaanxi. Next, the counties
ation” and “organization of care” [41]. of each province were divided into two categories (high and
We hypothesize that among rural-based patients with low) based on the level of economic development, and one
hypertension, trust in primary care physicians (PCPs) sample county was randomly chosen from each category.
will have a positive impact on satisfaction. For the in- Then, three townships from each county were randomly se-
ternal dimensions, we hypothesize that trust in PCPs’ lected as sample townships by classifying each township as
benevolence will have positive impacts on all three economically developed, moderately developed, or less

Fig. 1 Structural framework of the theoretical relationships

Chen et al. BMC Family Practice (2020) 21:196 Page 4 of 13

developed and choosing one from each category. Next, European task force on patient evaluation of general
three sample villages were chosen based on their distance practice
from township hospitals (far, medium, and close). Finally, The EUROPEP is a comprehensive 23-item question-
in every village, using the relevant primary health service in- naire that measures patients’ satisfaction with their gen-
stitution’s hypertension management archives, a random eral medical services [40] and assesses patients’ opinions
sample of 50 hypertension patients was selected for the sur- of their regular GP, based on their experiences over the
vey. Our sample size met the requirement that sampling previous 12 months. Items are scored using a 5-point
(using SEM) should contain at least 20 observations per Likert scale (1 = “poor,” 3 = “acceptable,” 5 = “excellent”).
variable of analysis according to the heuristics, and that the The EUROPEP has been tested in 16 European coun-
number of samples for each unit should be more than 30 tries, and its technical quality has been verified [41]. A
(that is, a large sample size) [42]. Additionally, we increased revised Chinese version was created by Han [45]. We di-
the sample size by approximately 10% to account for un- vided the scale into three dimensions: clinical behavior
foreseeable factors. (15 items), continuity and cooperation (3 items), and
Hangzhou Normal University’s scientific research eth- organization of care (5 items concerning facilities, avail-
ics committee reviewed and approved the study proto- ability, and accessibility). In the present study, the total
col. Prior to the administration of the questionnaires, score for the satisfaction scale ranged from 23 to 96, and
oral informed consent was first obtained from all pa- the median (interquartile range) was 50 (21).
tients considering the age and the education level of pa-
tients with hypertension in rural areas. All participants Reliability and validity of the Chinese version of the wake
satisfied the following inclusion criteria: (a) had received Forest physician trust scale and the European task force
hypertension management for more than 1 year; (b) had on patient evaluation of general practice
a normal intelligence quotient; (c) did not have any brain Responses to all 10 and 23 items of the C-WFPTS and
trauma or brain disease, visual or auditory dysfunction, EUROPEP, respectively, were entered into an explora-
or psychiatric disorder; and (d) could speak or read tory factor analysis model. The principal component ex-
Chinese. traction method was used to extract the components of
each scale. Consequently, the two-component model of
Measures C-WFPTS and the three-component model of EUR-
The questionnaire was distributed by trained inter- OPEP were determined to explain 53.75 and 63.601% of
viewers. All subjects were asked a core set of questions the total variance, respectively. The components were
regarding their socio-demographic characteristics (i.e., consistent with previous researches using the two scales.
age, gender, household register, marital status, level of The two components of C-WFPTS were “Benevolence”
education, per-capita annual household income, and and “Technical competence”, and the three components
health insurance), hypertension management (i.e., num- of EUROPEP were “Clinical behavior” , “Continuity and
ber of visits to PCPs in the past year, blood pressure, cooperation” and “Organization of care”. Factor loadings
and distance from the nearest health-service institution), for the two scales were shown in Tables 1 and 2. The
and self-reported health status. Patients’ trust and satis- Kaiser-Meyer-Olkin (KMO) test for sampling adequacy
faction with their PCPs were measured using the WFPT returned values of 0.833 and 0.973, and the Bartlett’s test
S-C and the EUROPEP, respectively. of sphericity χ2 returned 7135.817 and 42,086.746
(p < 0.001), indicating that the scales contained good
Chinese version of the wake Forest physician trust scale construct validity. Both scales and their dimensions also
Hall et al. [43, 44] verified the reliability and validity of showed favorable internal consistency, ranging from
the WFPTS through a large number of empirical studies. 0.728 to 0.958.
A modified Chinese version was developed by Dong and
Bao [37], comprising 10 items that are scored using a 5- Statistical analysis
point Likert scale, ranging from 1 (“strongly disagree”) In the initial analysis, outlier data and multicollinearity
to 5 (“strongly agree”; scoring is reversed for items 2, 3, were assessed before proceeding. The existence of out-
7, and 8). The Chinese scale’s two-dimensional structure, liers was identified using Cook’s distance [46]: if the ob-
“benevolence” and “technical competence,” has been served Cook’s distance was greater than 0.5, the point
verified in previous research. The overall score is com- was considered an outlier or strong influence point. Our
puted through an unweighted summation of the individ- analysis returned a maximum Cook’s distance of 0.036,
ual item scores, with higher scores reflecting greater indicating no outlier data. Next, multicollinearity was
trust. In the present study, the total score for the trust tested by considering tolerance rate and the variance in-
scale ranged from 10 to 50, and the median (interquar- flation factor (VIF) [47]. The results showed no toler-
tile range) was 24 (4). ance rate below 0.10 or VIF above 10; all tolerance
Chen et al. BMC Family Practice (2020) 21:196 Page 5 of 13

Table 1 Factor analysis with factor loadings for C-WFPTS

Component
1 2
1. My doctor will do whatever it takes to provide me all the care I need. 0.72
2. Sometimes my doctor cares more about what is convenient for him/her than about my medical needs. 0.80
4. My doctor is extremely thorough and careful. 0.70
6. My doctor is totally honest in telling me about all of the different treatment options available for my condition. 0.73
8. My doctor only thinks about what is best for me. 0.66
3. My doctor’s medical skills are not as good as they should be. 0.74
5. I completely trust my doctor’s decisions about which medical treatments are best for me. 0.51
7. Sometimes my doctor does not pay full attention to what I am trying to tell him/her. 0.76
9. I have no worries about putting my life in my doctor’s hands 0.75
10. All in all, I have complete trust in my doctor. 0.56
Variation % 35.01 18.74
Kaiser-Meyer-Olkin measure of sampling adequacy: 0.833
Bartlett’s test of sphericity: χ2: = 7135.817, p < 0.001

Table 2 Factor analysis, with factor loadings, for EUROPEP

Component
1 2 3
1. Making you feel you had time during consultations 0.68
2. Interest in your personal situation 0.77
3. Making it easy for you to tell him or her about your problems 0.75
4. Involving you in decisions about your medical care 0.63
5. Listening to you 0.77
6. Keeping your records and data confidential 0.67
7. Quick relief of your symptoms 0.71
8. Helping you to feel well so that you can perform your normal daily activities 0.74
9. Thoroughness 0.73
10. Physical examination 0.73
11. Offering you services for preventing diseases 0.64
12. Explaining the purpose of tests and treatments 0.68
13. Telling you what you wanted to know about your symptoms and/or illness 0.71
14. Help in dealing with emotional problems related to your health status 0.62
15. Helping you understand the importance of following his or her advice 0.64
16. Knowing what s/he had done or told you during previous contacts 0.83
17. Preparing you for what to expect from a specialist or hospital care 0.80
18. The helpfulness of staff (other than the doctor) 0.51
19. Getting an appointment to suit you 0.63
20. Getting through to the practice on the phone 0.60
21. Introducing you to other doctors in time or arranging a referral to the best hospital, if necessary 0.76
22. Waiting time in the waiting room 0.59
23. Providing quick services for urgent health problems 0.78
Variation % 53.23 5.74 4.64
The Kaiser-Meyer-Olkin measure of sampling adequacy: 0.973
Bartlett’s test of sphericity: χ2: = 42,086.746, p < 0.001
Chen et al. BMC Family Practice (2020) 21:196 Page 6 of 13

values were above 0.77 and all VIFs were below 1.30, in- Univariate analysis of factors associated with patient
dicating no multicollinearity. satisfaction
We used Cronbach’s α values to test the reliability Wilcoxon rank-sum tests revealed that patient satisfac-
of the scale and factor analysis to test structural val- tion was associated with region, gender, insurance status,
idity. Categorical variables were presented through per-capita annual household income, distance from the
frequencies and percentages. The normality of the nearest medical/health-service institution, and number
distribution of the continuous variables was tested of visits to PCPs in the past year. Residents of eastern
using a one-sample Kolmogorov-Smirnov test. Con- and central zones (p < 0.001), males (p < 0.001),
tinuous variables with normal distribution were pre- recipients of medical insurance for urban residents
sented as means ± standard deviations; non-normal (p < 0.001), and those living near a medical/health-ser-
variables were reported as medians (interquartile vice institution (p < 0.001) had better self-reported
range). Comparisons of continuous variables (scores health status, made fewer visits in the past year, and had
for patient trust and satisfaction) were conducted significantly lower satisfaction with their PCPs (Table 4).
using t-tests and one-way analyses of variance
(ANOVA) tests, while Wilcoxon rank-sum tests (the
Multiple linear regression analysis
Mann-Whitney U test and Kruskal-Wallis test) were
The results of our multiple linear regression analysis
used for non-normally distributed values. p values of
(Table 5 shows the assignment of demographic vari-
< 0.05 indicated statistical significance. Multiple linear
ables), with satisfaction score as the dependent variable,
regression analysis was conducted, with patient satis-
trust score as the independent variable, and after con-
faction as the dependent variable and patient trust, as
trolling for other covariates, showed that trust score
well as its two dimensions, as the independent vari-
(β = 0.435, p < 0.01), receiving medical insurance for
able. Model covariates were selected from those that
urban residents (β = 0.133, p < 0.01), and living in the
returned a p-value of less than 0.2 in the univariate
central province (β = 0.149, p < 0.01) were associated
analysis. Next, SEM was conducted to test our hy-
with significantly higher satisfaction scores. In contrast,
potheses. SEM can be used to measure latent vari-
the number of visits in the past year (β = − 0.121,
ables, and it allows the measurable variable and the
p < 0.01) and distance from the nearest medical/health
latent variables to be placed in a common model,
service institution (β = − 0.074, p < 0.01) were associated
which can include multiple dependent variables in
with significantly lower scores. In addition, males had
one measurement, reducing the error of multiple lin-
significantly higher satisfaction scores compared to fe-
ear regression analysis. We also used several fit indi-
males (Table 6). We conducted a second linear regression,
ces, including chi-square ratio (< 3), goodness of fit
taking the two dimensions of trust as independent vari-
index (GFI; > 0.9), adjusted goodness of fit index
ables; this analysis showed that the score for “benevo-
(AGFI; > 0.9), normal fit index (NFI; > 0.9), and root
lence” was associated with significantly increased
mean square error of approximation (RMSEA; < 0.05)
satisfaction (β = 0.532, p < 0.01), while “technical compe-
to evaluate overall model fitness. All analyses were
tence” did not feature in the model (Table 6).
conducted using SPSS 16.0 and AMOS 22.0 (SEM).

Structural equation modeling

Results On the basis of our factor analysis, “T1 benevolence”
Patients’ demographic characteristics, hypertension and “T2 technical competence” were used as exogen-
management, and self-reported health status ous latent variables, while “S1 clinical behavior,” “S2
This study included a total of 2665 patients. Over half medical service continuity and cooperation,” and “S3
(62.4%) were female; the majority were middle-aged organization of care” were used as endogenous latent
or older adults, with only 33 patients under 45 years variables. The corresponding entries acted as observa-
of age. Most respondents were married (82.5%) and tion variables to construct a structural equation
had lower than senior high school education at survey model. The final structural equation model is
time (95.8%). As we conducted our survey in rural depicted in Fig. 2, and the variables were showed in
areas, most respondents’ insurance was provided Table 7. For the fit indices, χ2/df, GFI, AGFI, NFI,
through the New Rural Cooperative Medical Scheme IFI, and CFI were > 0.9, and the RMSEA was < 0.05,
(NRCMS; 75.6%), a rural dwellers’ medical mutual indicating good model fit (Table 8).
helping system organized, guided, and supported by Results of the SEM indicated that benevolence posi-
local government [48]. Specific data, including data tively influenced clinical behavior (0.940), organization
for additional main characteristics, are presented in of care (0.910), and continuity and cooperation (0.879),
Table 3. while technical competence negatively influenced clinical
Chen et al. BMC Family Practice (2020) 21:196 Page 7 of 13

Table 3 Characteristics of the surveyed patients Table 3 Characteristics of the surveyed patients (Continued)
Characteristic N % Characteristic N %
Region Blood pressure
Eastern province 893 33.5 Controlled 1699 63.8
Central province 885 33.2 Uncontrolled 966 36.2
Western province 887 33.2 No. of visits in the past year
Gender <4 1251 46.9
Male 1002 37.6 ≥4 1414 53.1
Female 1663 62.4
Age
behavior (− 0.077), organization of care (− 0.136), and
continuity and cooperation (− 0.064) (Table 9).
< 45 33 1.2
45–59 483 18.1 Discussion
60–74 1533 57.5 In the present study, we found patients’ trust in their
> 75 616 23.1 PCPs to be the strongest predictor of patient satisfaction.
Marital Status Among our sample population, both trust scores and
Married 2138 80.2
satisfaction scores were relatively low. The median pa-
tient trust score was 2.4 per question, which is lower
Other 527 19.8
than the scores reported by Dong and Bao (3.1) [37],
Level of Education who used the same scale to conduct a trust survey of
Primary or lower 2090 78.4 outpatients at Shanghai Third Grade Hospital. The score
Junior high school 462 17.3 was also lower than scores reported in studies conducted
Senior high school or above 113 4.2 overseas regarding patients’ trust in family physicians
Insurance type
and primary care providers [30, 49]. A possible reason
for this discrepancy is the gap between the service cap-
Medical Insurance for Urban Employees 72 2.7
acity of China’s primary hospitals in rural areas and pa-
Medical Insurance for Urban Residents 191 7.2 tients’ demands regarding diagnosis and treatment,
Basic Medical Insurance for Urban and Rural Residents 636 23.9 hindering the formation of a long-term stable partner-
New Rural Cooperative Medical Scheme 1741 65.3 ship between doctors and patients [50]. In the present
Other 25 0.9 study, the median satisfaction score was 2.2 per ques-
Per-Capita Annual Household Income
tion, far below the scores reported in studies of China’s
urban community health service centers (over 4 per
1 (≤ 1000 yuan) 571 21.4
question, also measured using EUROPEP) [45, 51]; the
2 (1000–2160 yuan) 496 18.6 scores in our study were also lower than those reported
3 (2161–4000 yuan) 549 20.6 in a survey assessing patient satisfaction during GP visits
4 (4001–10,000 yuan) 557 20.9 across nine European countries [52]. A possible reason
5 (> 10,000 yuan) 492 18.5 is that the present study targeted rural-based hyperten-
Distance from the nearest medical and health service institutions
sion patients who, as a result of educational and social
environmental factors, have poor self-care awareness
< 1 km 2075 77.9
and inadequately controlled high blood pressure [53].
1–2.99 km 550 20.6 This may result in an overreliance on doctors and higher
≥ 3 km 40 1.5 numbers of PCP visits, thereby creating an impression
Course of disease that the physician is not sufficiently competent. Add-
≤ 3 years 600 22.5 itionally, although China has formed a relatively compre-
4–10 years 1248 46.8
hensive system for chronic disease prevention and
treatment, there are still many issues in rural areas, such
11–20 years 664 24.9
as the unreasonable allocation of medical and health re-
> 20 years 153 5.7 sources, medical staff’s low enthusiasm, insufficient pol-
Self-reported health status icies for hypertension prevention and treatment, and a
Bad 624 23.4 lack of effective supervision and evaluation mechanisms
Neither good nor bad 1800 67.5 for hypertension control [53]. Together, these problems
Good 241 9.0
have created a situation in which rural-based hyperten-
sion patients are less likely to have high satisfaction with
Chen et al. BMC Family Practice (2020) 21:196 Page 8 of 13

Table 4 Univariate analysis of factors associated with patient Table 4 Univariate analysis of factors associated with patient
satisfaction satisfaction (Continued)
Characteristic Classification Median Characteristic Classification Median
(Interquartile range) (Interquartile range)
Region Eastern province 51(18) χ2 (p) 7.611 (0.055)
Central province 50(21) Self-reported Bad 49(20)
health status
Western province 47(24) Neither good nor bad 50(20)
χ2 (p) 28.539 (< 0.001**) Good 48(24)
Gender Male 51(22) χ2 (p) 5.559 (0.062)
Female 49(19) No. of visits in the <4 52(22)
past year
Z(p) −3.480 (0.001**) ≥4 48(21)
Age < 45 52(20) Z(p) −5.238 (< 0.001**)
45–59 50(24) * < 0.05, ** < 0.01

60–75 50(22)
PCPs, regardless of whether the issue is caused by the
> 75 50(18)
physician, other medical staff, or associated policies [28].
χ2 (p) 7.424 (0.060) For the two dimensions of patient trust, the score for
Marital Status Married 50(22) “technical competence” was higher than “benevolence,”
Other 49(20) suggesting that, in comparison to benevolence-
Z(p) −0.138 (0.890) associated aspects, when undergoing hypertension man-
Level of Education Primary or below 50(20)
agement in primary-health-care institutions, rural pa-
tients may pay more attention to physicians’ attitudes
Junior high school 48(24)
and communication skills, and may be more likely to
Senior high school 50(19) feel that physicians are sufficiently clinically competent
and above
to address common medical problems, especially high
χ2 (p) 2.645 (0.266)
blood pressure. The second linear regression, which
Insurance Medical Insurance for 46(22) set the two dimensions of trust as the independent
Urban Employees
variables, showed that the score for “benevolence”
Medical Insurance for 60(16) was significantly associated with an increase in satis-
Urban Residents
faction, while the score for “technical competence”
Basic Medical Insurance 49(18)
for Urban and Rural
had no impact. “Benevolence” represents physicians’
Residents attitudes toward care and their communication com-
New Rural Cooperative 49(23) petence [37]. Our findings indicate that patients’ per-
Medical Scheme ceptions regarding physicians’ levels of considerate
Other 49(24) communication are positively related to patient satis-
χ2 (p) 71.755 (< 0.001**)
faction [54]. A similar result was found in a study
assessing factors that contribute to patients’ satisfac-
Per-Capita Annual 1 53(24)
Household Income tion with family physician consultations: most pa-
Quintile 2 50(23) tients highlighted poor communication as a major
3 47(23) factor that negatively affects the physician-patient re-
4 50(18) lationship, rather than physicians’ professional com-
5 50(18) petency [14].
χ2 (p) 13.490 (0.009**)
The results of our multiple linear regression analysis
and SEM differed slightly. In our SEM, all hypothesized
Distance from the < 1 km 50(20)
nearest health paths were significant. Specifically, “benevolence” had a
service institution 1–3 km 49(22) major positive impact on all three satisfaction dimen-
≥4 km 48(25) sions. An expression of high trust in physicians’ “ben-
χ2 (p) 14.553 (< 0.001**) evolence” indicates that patients believe that physicians
Course of disease ≤ 3 years 48(23) care about them and are willing to devote notable time
4–10 years 50(21)
to their treatment [37]. Although patient satisfaction can
be enhanced once a trusting relationship has been estab-
11–20 years 50(20)
lished, “technical competence” had a small but negative
> 20 years 50(18) impact on all three satisfaction dimensions, directly
Chen et al. BMC Family Practice (2020) 21:196 Page 9 of 13

Table 5 Assignment of demographic variables

Variable Reference group Assignment
Region Western zone Central province = 1; Eastern province = 2
Age < 45 45–59 years = 1; 60–75 years = 2; > 75 years = 3
Insurance Medical Insurance for Urban Employees Medical Insurance for Urban Residents = 1; Basic Medical
Insurance for Urban and Rural Residents = 2; New Rural
Cooperative Medical Scheme = 3; Other = 4
Distance from the nearest health < 1 km 1–3 km = 1; ≥ 4 km = 2
service institutions
Course of disease ≤ 3 years 4–10 years = 1; 10–20 years = 2;
> 20 years = 3
Self-reported health status Bad Neither good nor bad = 1; Good = 2

Table 6 Results of linear regression models examining predictors of hypertensive patients’ satisfaction with PCPs
Variable Unstandardized beta SE Standardized beta t p Confidence
interval
Model 1 (taking total trust score as the independent variable)
Constant 12.749 1.803 – 7.071 < 0.001 (9.213, 16.284)
Trust 1.597 0.065 0.435 24.752 < 0.001 (1.470, 1.724)
No. of visits in the past year − 0.100 0.014 −0.121 −7.031 < 0.001 (−0.127, − 0.072)
Insurance
Medical Insurance for Urban Employees (reference group)
Medical Insurance for Urban Residents 6.882 0.917 0.133 7.502 < 0.001 (5.083, 8.681)
Basic Medical Insurance for Urban and Rural Residents 1.243 0.593 0.040 2.096 0.036 (0.080, 2.406)
Region
Western province
Central province 4.219 0.550 0.149 7.666 < 0.001 (3.140, 5.299)
Distance from the nearest medical and health service institution
< 1 km (reference group)
1–3 km −2.423 0.567 −0.074 −4.278 < 0.001 (−3.534, −1.313)
Gender −1.573 0.467 −0.057 −3.369 0.001 (−2.489, −0.658)
R2 0.239
Model 2 (taking the two dimensions of trust as the independent variable)
Constant 22.930 1.193 – 19.225 < 0.001 (20.591, 25.269)
Benevolence 3.107 0.094 0.532 33.010 < 0.001 (2.922, 3.292)
Insurance
Medical Insurance for Urban Employees (reference group)
Medical Insurance for Urban Residents 6.804 0.856 0.132 7.945 < 0.001 (5.125, 8.483)
Basic Medical Insurance for Urban and Rural Residents 1.923 0.552 0.062 3.481 0.001 (0.840, 3.006)
Region
Western province
Central province 3.550 0.509 0.126 6.969 < 0.001 (2.551, 4.549)
No. of visits in the past year −0.080 0.013 − 0.097 −6.053 < 0.001 (−0.106, − 0.054)
Distance from the nearest medical and health service institutions
< 1 km (reference group)
1–3 km −2.351 0.529 −0.072 −4.443 < 0.001 (−3.389, −1.314)
Gender −1.144 0.437 −0.042 −2.620 0.009 (−2.000, −0.288)
R2 0.336
Chen et al. BMC Family Practice (2020) 21:196 Page 10 of 13

Fig. 2 Structural equation model of trust in physicians and patient satisfaction

contradicting our original hypothesis. This appears un- may not obviously improve patient-perceived health-
usual but can be explained by the “customer-perceived service quality. Moreover, as a basic public health ser-
quality of service theory” (similar to patient satisfaction, vice, hypertension management is not very demanding
which can be understood as “patient-perceived quality of for doctors in terms of technical competence; instead,
health service”). This theory holds that service quality there is a greater need for doctors to improve the service
comprises two parts: technical quality (the result of the process and to make their patients feel their concern
service) and functional quality (the service process) [55]. and that their communication is enjoyable [56]. In our
Technical quality is a “hygiene factor” in regard to ser- study, although physicians’ technical competence may be
vice quality, which means that high technical quality similar, it was not the patients’ main concern. Instead,
patients frequently made requests concerning the service
Table 7 The variables of the structural equation model process (or functional quality) provided by physicians,
Variable such as physician-based care, communication, and refer-
T1 Benevolence
ral arrangements. Additionally, our investigation focused
on physicians’ clinical behavior (including the patient-
a1-a5 The items of “Benevolence” (eg. My doctor will do whatever
it takes to provide me all the care I need.) doctor relationship and communication, medical care,
and information and support), organization of care, and
T2 Technical competence
continuity and cooperation, which mostly relate to the
b1-b5 The items of “Technical competence” (eg. My doctor’s
medical skills are not as good as they should be.)
hypertension treatment service process. Thus, based on
our results, patients are more likely to express their dis-
S1 Clinical behavior
satisfaction with these dimensions when they have rela-
c1-c15 The items of “Clinical behavior” (eg. Making you feel tively high trust in their PCPs.
you had time during consultations.)
When examining the influence of socio-demographic
S2 Continuity and cooperation
and other variables, we found that patients who lived in
d1-d3 The items of “Continuity and cooperation” (eg. Knowing the central province tended to have higher satisfaction
what s/he had done or told you during previous contacts.)
than patients who lived in the eastern and western prov-
S3 Organization of care inces. A study conducted in eight cities located in rural
e1-e5 The items of “Organization of care” (eg. Getting an areas of China yielded similar results [53]. This region-
appointment to suit you.)
based difference in satisfaction is potentially caused by
Chen et al. BMC Family Practice (2020) 21:196 Page 11 of 13

Table 8 Fit indices of final model

Fit indices GFI AGFI CFI NFI IFI χ2/df RMSEA
Reference value scale > 0.9 > 0.9 > 0.9 > 0.9 > 0.9 <5 < 0.08
Fitted value 0.929 0.915 0.942 0.933 0.942 7.322 0.049
a
GFI Goodness of fit index; AGFI adjusted goodness of fit index; CFI comparative fit index; NFI normed fit index; IFI incremental fit index; RMSEA root mean square
error of approximation

differing health-service conditions and resident needs. across the literature [14, 32]; however, our study
For instance, patients in the eastern province may showed that men had greater satisfaction with their
have had access to better health services, but their family physicians. No other factors, including age,
hypertension-related knowledge, particularly regarding education, marital status, or self-reported status, had
risks and management, may have caused them to an influence in this regard, indicating that patient sat-
make more requests to visit their physicians. In the isfaction is a universal phenomenon across these
western province, financial limitations meant that the variables.
services patients received did not reach the standards This study has several limitations. In particular, we did
of those of the eastern province; thus, patient satisfac- not measure the impact of physician-related characteris-
tion was lower. Notably, patients with more visits in tics and the regression models had a relatively small
the past year reported lower satisfaction levels. For goodness of fit index (R2), indicating a limited ability to
hypertension patients, graded follow-up management explain the variations in the dependent variables, but
is implemented based on blood pressure levels; more still within the acceptable range. However, we did not
visits mean blood pressure is not being adequately aim to predict patient satisfaction ratings; instead, we
controlled or indicate the presence of a more serious primarily aimed to identify the influence of patient trust
condition [6]. Such patients need medication, and and its dimensions on these ratings. Despite these short-
their quality of life is more likely to be affected, comings, by using SEM we have extended the current
thereby increasing the economic burden of the disease literature concerning relationships between trust and
[55]. Patients who lived near health institutions were satisfaction among Chinese rural patients with hyperten-
more satisfied owing to convenience and medical in- sion, clarifying the influencing mechanism of the in-
surance also impacted patients’ satisfaction. For in- ternal dimension.
stance, patients insured with the Medical Insurance
for Urban Residents (MIUR) and Basic Medical Insur- Conclusion
ance for Urban and Rural Residents were more satis- This study analyzed the effect of patient trust with their
fied. Compared with the NRCMS, MIUR reimburses PCPS on satisfaction among rural-based patients with
for a wider range of drugs and has a relatively more hypertension. Results indicated that during hypertension
convenient referral process [53, 57]. Basic Medical In- management, patients focus more on physicians’ benevo-
surance for Urban and Rural Residents is integrated lence than on their technical competence. Thus, medical
into MIUR and NRCMS and features, within its humanities education should be emphasized for PCPs to
scope, medical institutions and drugs that can be re- improve the services they provide, as well as their service
imbursed under both types of insurance, thereby attitude. Concurrently, we found that physicians’ commu-
benefiting rural residents [58]. Thus, integrating med- nication skills played an essential role in improving pa-
ical insurance for urban and rural residents can im- tients’ satisfaction. However, low overall trust and
prove patient satisfaction. Regarding gender, a satisfaction among patients can negatively influence pa-
consistent conclusion has not yet been established tients’ self-management and doctors’ enthusiasm, exerting

Table 9 Results of structural equation modeling

Path Unstandardized Standardized t Hypothesis
regression weights regression weights (Y/N)
Benevolence→ Clinical behavior 1.607 0.940 25.493** Y
Benevolence→ Continuity and cooperation 1.642 0.910 25.041** Y
Benevolence→ Organization of care 1.716 0.879 26.800** Y
Technical competence→ Clinical behavior −0.169 −0.077 −4.116** N
Technical competence→ Continuity and cooperation −0.317 −0.136 −6.255** N
Technical competence→ Organization of care −0.161 −0.064 −3.001** N
* < 0.05, ** < 0.01
Chen et al. BMC Family Practice (2020) 21:196 Page 12 of 13

a harmful influence on the effective management of Received: 24 April 2020 Accepted: 10 September 2020
hypertension. Thus, it is important for patients and physi-
cians to establish a long-term, stable partnership.
Our findings regarding other factors, such as region- References
and health-insurance-based differences, indicate that the 1. Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global
burden of hypertension: analysis of worldwide data. Lancet. 2005;365(9455):
existing inequities between eastern, central, and western
217–23.
provinces as well as those between urban and rural areas 2. Jneid S, Jabbour H, Hajj A, Sarkis A, Licha H, Hallit S, et al. Quality of life and
must be addressed. Thus, future formulations of policies its association with treatment satisfaction, adherence to medication, and
should fully consider regional characteristics and inte- Trust in Physician among Patients with Hypertension: a cross-sectional
designed study. J Cardiovasc Pharmacol Ther. 2018;23(6):532–42.
grate medical insurance for urban and rural residents. 3. Chinese Medical Doctor Association. China's hypertension prevention and
To meet residents’ needs, services in eastern provinces control status blue book 2018. Beijing: People's Medical Publishing House;
should focus on improving efficiency, while services in 2019.
4. Chen X, Li L, Zhou T, Li Z. Prevalence of hypertension in rural areas of
western provinces need to improve in other areas, such China: a meta-analysis of published studies. PLoS One. 2014;9(12):e115462.
as health facilities and the workforce. 5. Han Y. Present situation and countermeasure of China's rural medical and
health service system. J Changchun Univ Chin Med. 2015;31(1):202–203+
Abbreviations 213.
EUROPEP: European Task Force on Patient Evaluation of General Practice; 6. Revision Committee of China's Guidelines for the Management of
GP: General Practitioner; KMO: Kaiser-Meyer-Olkin test; MIUR: Medical Hypertension. China’s Guidelines for the Management of Hypertension
Insurance for Urban Residents; NRCMS: New Rural Cooperative Medical (2014 Revision). Chin J Hypertens. 2015;23(1):24–43+15.
Scheme; PCP: Primary Care Physician; SEM: Structural Equation Model; 7. Cobos B, Haskard-Zolnierek K, Howard K. White coat hypertension:
VIF: Variance Inflation Factor; WFPTS-C: Chinese version of the Wake Forest improving the patient-health care practitioner relationship. Psychol Res
Physician Trust Scale Behav Manag. 2015;8:133–41.
8. Liu X, Liu C. Doctor-patient communication in medical situations and
enlightenment of knowledge transfer. Chin Gen Pract. 2014;17(34):4146–50.
Acknowledgements
9. Gopichandran V, Chetlapalli SK. Factors influencing trust in doctors: a
The authors would like to thank the School of Medicine of Hangzhou
community segmentation strategy for quality improvement in healthcare.
Normal University and Zhengzhou University for supporting this research.
BMJ Open. 2013;3(12):e004115.
The authors also thank the staff of the primary health service institutions and
10. Chai S. Research on status and countermeasures of interpersonal continuity
the research staff for their assistance with the surveys.
of health Care in Rural Areas [master's thesis], Huazhong University of
Science & Technology; 2016.
Authors’ contributions 11. Pascoe GC. Patient satisfaction in primary health care: a literature review
MZ and WC conceived the paper. YF, JF, and JW (MS) were responsible for and analysis. Eval Program Plann. 1983;6(3–4):185–210.
data collection. WC conducted the data analyses and was primarily 12. Platonova EA, Kennedy KN, Shewchuk RM. Understanding patient
responsible for writing the paper. MZ, XH, XW, and JW (PhD) helped with satisfaction, trust, and loyalty to primary care physicians. Med Care Res Rev.
revisions of the manuscript. All authors read and approved the final 2008;65(6):696–712.
manuscript. 13. Zheng J, Wu Q, Liu H, Zheng T, Tan X, Chen M. Study on the influencing
factors of the satisfaction of patients with hypertension to the village clinics
Funding under the background of rural basic public health services. Med Soc. 2019;
This study was supported by Grant 71403075 from the National Natural 32(9):61–3.
Science Foundation of China and Grant 19BGL250 from the National Social 14. Bawakid K, Rashid O, Mandoura N, Shah H, Ahmad W, Ibrahim A. Patients'
Science Fund of China for the research, authorship, and publication of this satisfaction regarding family physician's consultation in primary healthcare
article. centers of Ministry of Health. J Fam Med Prim Care. 2017;6(4):819–23.
15. Zheng Y, Hua S, Dai J. Satisfaction of community patients with chronic
Availability of data and materials diseases to treatment and their utilization of medical services. Chin Gen
The datasets used during the current study are available from the Pract. 2012;15(19):2153–7.
corresponding author on reasonable request. 16. Yu Z. Research on status and countermeasures of interpersonal continuity
of health Care in Rural Areas [master's thesis], Shanxi Normal University;
2017.
Ethics approval and consent to participate
17. Hall MA, Dugan E, Zheng B, Mishra AK. Trust in physicians and medical
Hangzhou Normal University’s scientific research ethics committee reviewed
institutions: what is it, can it be measured, and does it matter? Milbank Q.
and approved the study protocol (reference number: 2015–032). Considering
2001;79(4):613–39.
the age and the education level of patients in our study, verbal-only in-
18. Chang CS, Chen SY, Lan YT. Service quality, trust, and patient satisfaction in
formed consent, which had been approved by the ethics committee, was
interpersonal-based medical service encounters. BMC Health Serv Res. 2013;
obtained from all patients. All procedures were performed in accordance
13(1):22. https://doi.org/10.1186/1472-6963-13-22.
with the Declaration of Helsinki (2008) and the ethical standards of the ethics
19. Barton JL, Trupin L, Tonner C, Imboden J, Katz P, Schillinger D, et al. English
committee of Hangzhou Normal University.
language proficiency, health literacy, and trust in physician are associated
with shared decision making in rheumatoid arthritis. J Rheumatol. 2014;
Consent for publication 41(7):1290–7.
Not applicable. 20. Liu X, Chen J. Investigation on nurse — patient communication and degree
of trust and satisfaction of inpatients in military comprehensive general
Competing interests hospitals. Hosp Adm J Chin PLA. 2016;23(7):613–6.
The authors declare that they have no competing interests. 21. Chen Q, Fan J, Xi M, Chen Z, Rong Y, Zhou J, et al. Investigation of the
influence factors of inpatients’ satisfaction degree in a grade a
Author details Comprehensive Hospital in Beijing. Chin Hospital Manag. 2015;35(2):56–8.
1
School of Medicine, Hangzhou Normal University, No. 2318, Yu-hang-tang 22. Cheng Q, Liu K, You L, Zhang L, Zhu X, Zheng J, et al. Survey on inpatient
Road, Hangzhou, China. 2College of Public Health, Zhengzhou University, satisfaction with healthcare and its influencing factors in level 2 and level 3
Zhengzhou, China. hospitals in China. Chin Nurs Manag. 2016;16(11):1482–8.
Chen et al. BMC Family Practice (2020) 21:196 Page 13 of 13

23. Zhou W, Wang Y, Wan Q, Gong W, Zhang Y, Hou L, et al. Path analysis of 48. Yip W, Hsiao W. China's health care reform: a tentative assessment. China
perceived service quality, patient satisfaction and patient trust on patient Econ Rev. 2009;20(4):613–9.
behavioral intentions. Chin J Health Stat. 2018;35(2):272–5. 49. Chu-Weininger MYL, Balkrishnan R. Consumer satisfaction with primary care
24. Liu S. Construction and empirical analysis of patient satisfaction index provider choice and associated trust. BMC Health Serv Res. 2006;6(1):139.
model in large general hospitals, Jilin University; 2013. 50. Wu H, Hao J, Yang J. Study on the patient trust assessment and influencing
25. Qian Y, Wang X, Chen Y, Zhang S, Yu C, Gu Y. Studying on the progress of factors in different level medical institution. Chin Hosp. 2018;22(2):26–9.
research in patients' satisfaction on medical services and its problems. Chin 51. Tan C, Zhu H. Reliability and validity analysis of European satisfaction survey
Health Serv Manag. 2015;32(2):105–7. scale (EUROPEP) and its application in a community in Shanghai. Chin
26. Sun F, Fu Y, Dong E. Study on the factors influencing the life satisfaction of Commun Doctors. 2017;33(13):147–8.
outpatients in public hospitals. Chin Health Qual Manag. 2016;23(3):60–62+ 52. Wensing M, Vedsted P, Kersnik J, Peersman W, Klingenberg A, Hearnshaw H,
66. et al. Patient satisfaction with availability of general practice: an
27. Zhang X. Study on the health service and professionals, satisfaction and the international comparison. Int J Qual Health Care. 2002;14(2):111–8.
influencing factors of residents in Shandong province [master's thesis], 53. Ding P. Analysis on the current situation and effectiveness of hypertension
Shandong University; 2016. control in rural China [master's thesis], Chinese Center for Disease Control
28. Zhang M, Tang G. Problems and countermeasures of prevention and and Prevention; 2013.
control of hypertension in rural areas. Chin Rural Health Serv Adm. 2018; 54. Hesse C, Rauscher EA. The relationships between doctor-patient affectionate
38(1):30–2. communication and patient perceptions and outcomes. Health Commun.
29. Van Den Assem B, Dulewicz V. Patient satisfaction and GP trustworthiness, 2019;34(8):881–91.
practice orientation and performance: implications for selection, training 55. Qi X. Association study of psychosocial characteristics, compliance and
and revalidation. J Health Organ Manag. 2014;28(4):532–47. blood pressure control among hypertensive patients [master's thesis],
30. Baker R, Mainous Iii AG, Gray DP, Love MM. Exploration of the relationship Fudan University; 2013.
between continuity, trust in regular doctors and patient satisfaction with 56. Hu S. Study on the applicability of SERVQUAL scale in evaluating the
consultations with family doctors. Scand J Prim Health Care. 2003;21(1):27– community health service quality in China [master's thesis], Huazhong
32. University of Science and Technology; 2018.
31. Martin KD, Roter DL, Beach MC, Carson KA, Cooper LA. Physician 57. Li Y. Research on the basic health Insurance for Urban and Rural Residents
communication behaviors and trust among black and white patients with in Shanxi Province [master's thesis], Shanxi University of Finance &
hypertension. Med Care. 2013;51(2):151–7. Economics; 2016.
32. Shan L, Li Y, Ding D, Wu Q, Liu C, Jiao M, et al. Patient satisfaction with 58. Xia Y, Jing X, Duan Q. Status quo, problems and recommendations on the
hospital inpatient care: effects of trust, medical insurance and perceived convergence of urban-rural basic medical insurance system in China. Chin J
quality of care. PLoS One. 2016;11(10):e0164366. Health Policy. 2010;3(1):43–8.
33. Wu M. Structural equation model: the operation and application of AMOS:
Chongqing University press; 2010. Publisher’s Note
34. Benkert R, Peters RM, Clark R, Keves-Foster K. Effects of perceived racism, Springer Nature remains neutral with regard to jurisdictional claims in
cultural mistrust and trust in providers on satisfaction with care. J Natl Med published maps and institutional affiliations.
Assoc. 2006;98(9):1532–40.
35. Dong EH, Liang Y, Liu W, Du X, Bao Y, Du Z, et al. Construction and
validation of a preliminary Chinese version of the wake Forest physician
trust scale. Med Sci Monit. 2014;20:1142–50.
36. Bachinger SM, Kolk AM, Smets EMA. Patients' trust in their physician-
psychometric properties of the Dutch version of the "wake Forest physician
trust scale.". Patient Educ Couns. 2009;76(1):126–31.
37. Dong E, Bao Y. Reliability and validity of the Chinese version of wake Forest
physician trust scale. Chin Ment Health J. 2012;26(3):171–5.
38. Grol R, Wensing M, Mainz J, Ferreira P, Hearnshaw H, Hjortdahl P, et al.
Patients' priorities with respect to general practice care: an international
comparison. Fam Pract. 1999;16(1):4–11.
39. Bjertnaes OA, Lyngstad I, Malterud K, Garratt A. The Norwegian EUROPEP
questionnaire for patient evaluation of general practice: data quality,
reliability and construct validity. Fam Pract. 2011;28(3):342–9.
40. Teng F, Guan L, Wang Y, Xuan L. Evaluation of community health service
satisfaction by method based on the EUROPEP-questionnaire. Chin Gen
Pract. 2011;14(25):2844–7.
41. Evans RG, Edwards A, Evans S, Elwyn B, Elwyn G. Assessing the practising
physician using patient surveys: a systematic review of instruments and
feedback methods. Fam Pract. 2007;24(2):117–27.
42. Bartlett J, Kotrlik J, Higgins C. Organizational research: determining
appropriate sample size in survey research appropriate sample size in
survey research. ITLPJ. 2001;19(1):43–50.
43. Hall MA, Camacho F, Lawlor JS, Depuy V, Sugarman J, Weinfurt K. Measuring
trust in medical researchers. Med Care. 2006;44(11):1048–53.
44. Hall MA, Zheng B, Dugan E, Camacho F, Kidd KE, Mishra A, et al. Measuring
patients' trust in their primary care providers. Med Care Res Rev. 2002;59(3):
293–318.
45. Han G. Development and application of extended EUROPEP scale – a
survey of patients' satisfaction with the service of community general
practitioners [master's thesis], Shanxi Medical University; 2014.
46. Sen Roy S, Guria S. Diagnostics in logistic regression models. J Korean Stat
Soc. 2008;37(2):89–94.
47. Field A. Discovering statistics using IBM SPSS statistics: sage publications ltd;
2013.