Database.sav

The study protocol was approved by the Institutional Review Board of School of Public Health, Sichuan University. Informed consent was obtained from each participant following a detail explanation about the purpose of the study.

This cross-sectional study was conducted by trained
investigators (medical and public health students and community volunteers) in
Sichuan province, from January to August in 2014.

Measure

A
multistage stratified random sampling method was used. In the first stage,
three cities in Sichuan Province were randomly selected. In the second stage, a
city district and a county were randomly selected from each city. In the third
stage, two communities or townships were randomly selected from each city
district or county. In the fourth stage, we randomly selected 90 residents in
each community or township and asked about their choices of PCFs as the initial
contact for medical care during their latest illness episode.
Residents who were 18 years old and above and had resided in the community/town
for at least 6 months were eligible to participate.

The
investigators took an average of 15 minutes to interview each participant.
Questionnaires were checked by investigators immediately after the survey for
completeness.

A total of 1,080 residents were interviewed (540 urban and 540 rural).
Because the present survey focused on residents’ choices of PCFs for the first
treatment during the latest illness episode, only those who sought medical
treatment were used in the analysis and 165 residents (81 urban and 84 rural)
were excluded for not seeking medical treatment during the latest illness
episode. The final sample included 915 residents (459 urban and 456 rural).

Questionnaire

Respondent
characteristics, provider of initial treatment and principal reason for the
choice were collected.
Respondent characteristics were collected, including socio-demographic
characteristics, walking time (minutes) from home to the nearest PCFs and
self-reported health status. Socio-demographic characteristics included age,
gender, marital status, employment status, education, status of medical
insurance, per capita annual income of household and individual annual income.

The places of care for the initial treatment were categorized into two
basic types: PCFs and higher-tier
hospitals. For residents who chose to receive initial treatment at
medical institutions, we asked them to select principal reason for the choice,
including convenience, whether they felt that the charges were reasonable,
quality of care, trust in doctors, good patient-doctor communication, prior
experience with the doctors, and the medical insurance designation status of
the facility.

Data were entered using the Epidata 3.1 database and were analyzed using
the IBM SPSS version 23.0. We reported means and SDs for continuous variables
and percentages for categorical variables. Pearson’s Chi-square tests or
Fisher’s exact test was used to assess the differences in categorical
variables. We conducted multivariable logistic regression models to identify
the factors associated with rural or urban residents’ decisions about whether
to seek care at PCFs for the initial treatment. The dependent variable was
whether the residents used PCFs (0 for no and 1 for yes). The odds ratio (OR)
was reported along with 95% confidence interval (CI). Results with a p-value of
<0.05 were considered statistically significant.