Study sample

The sample of both focus groups and questionnaires was collected in Liaoning Province, including both developed areas and less developed areas. In 2007, the poverty line in Liaoning Province was defined as 2,000 Yuan, while the average per capita income of the province was 4,773 Yuan. Areas with average income per capita above (below) 2,000 Yuan were regarded as developed (less developed) areas. In our sample 10 villages were randomly selected in each developed and less developed areas respectively and 3,000 people aged 18-72 were involved. Both in-home questionnaire and additional focus groups were conducted between 10th January and 20th August 2008 [16].

In a typical interview, one expert held the interview and two professionals recorded the conversation, expression and mood of the interviewees throughout the process [17]. Informed consent was obtained from the participants before the interview¹.

Data collection and measures

All participants completed a structured questionnaire. The following items were included: socio-demographics(gender, age, education level, marital status), self-rated health status, whether being sick within 2 weeks, and attitudes towards the NCMS. Initially, items of attitudes towards the NCMS consisted of five options: "Very Satisfied" "Satisfied", "Neutral", "Dissatisfied", and "Very Dissatisfied". Our initial descriptive analysis indicated that answers to each extreme were rare. We, therefore, combined "Very Satisfied" and "Satisfied" into "Satisfied". "Dissatisfied" and "very dissatisfied" were integrated into "dissatisfied". In the original questionnaire, occupation was included. We excluded it as all participants were farmers.

Statistical analysis

There were two sections in the questionnaire: (1) merits of the NCMS (7 items); (2) drawbacks of the NCMS (8 items). The respondents could select one or more items based on their own perception and experience. To be consistent in comparison, each item was converted to percentage using the following formula:

(1)

Where A = percentage of a particular item

Ω = number of respondents selected the item;

k = total number of valid respondents

In order to investigate factors affecting overall satisfaction, we conducted nonparametric tests to compare differences in attitudes towards the NCMS among respondents with different socio-demographic attributes, self-rated health status and whether being sick within two weeks. We applied ordinal regression models to evaluate effects of factors on attitudes towards the NCMS [18–20]. In the ordinal regression models, dependent variable Y, the respondents' satisfaction of the NCMS was divided into 3 categories (1 as satisfied, 2 as Neutral, 3 as Dissatisfied). The results of nonparametric analysis were substituted into the equation as follows:

The regression equation is:

(2)

where γj(x) = π₁(x)+π₂(x)+...+π_j(x), θ_j = logk_j

X₁ = Gender (male = 1, female = 0); X₂ = Age (younger than 20, X₂ = 0; 20-40, X₂ = 1; 40-60, X₂ = 2; X₂ = 3, older than 60);

X₃ = marital status (Single = 0, Married = 1, Divorced or widowed = 2);

X₄ = self-rated health status (Excellent = 0, Good = 1, Poor = 2, Awful = 3).

The study was approved by Peking University Health Science Center Institutional Review Board (No: IRB00001052-T1) and had no violation to human subjects, privacy and community honor.

(1) Analysis of three factors

This study found 71.6% of study respondents were satisfied with the NCMS and identified that three factors (gender, age and self-rated health status) were significantly associated with satisfaction.

Gender significantly affects attitudes towards the NCMS. 76.9% of men were satisfied, whereas 67.2% of women were satisfied. Through typical interviews, we understood that men provided the main source of household income, therefore they were engaged in agricultural work as the main labor force in rural areas. However, women were mainly responsible for family daily life. Therefore, the NCMS did not provide immediate benefits to women [21]. Hence the men had a higher rate of recognition and satisfaction with the NCMS.

The study showed that as people got older, satisfaction rate also rose (except the 20-40 age group). This could be explained by the fact that people between 20 and 40 were more economically independent than those in other age groups. The elderly were more likely to be satisfied with the current system since the elderly had the most medical needs, and expected government supports the most, especially when a welfare support system has been available for a long time [22, 23]. Through typical interviews, we understood that as long as the elderly were able to afford the premium and co-payments, they were more likely to be satisfied with the NCMS. It was also not surprising that individuals who thought medical care was expensive tended to be less satisfied with the overall performance of the NCMS. Individuals who were satisfied with the reimbursement procedure of the NCMS were more likely to be satisfied with the overall performance of the NCMS.

Furthermore, self-rated health status played an indispensable role regarding the attitudes towards the NCMS. Compared to those of excellent self-rated health status, respondents with awful self-rated health had lower satisfaction rate. One explanation might be that the NCMS could not completely meet the needs of persons who were in a poor health condition [24]. Through typical interviews, we understood that though the NCMS could, to some extent, ease patients' economic burdens, it was not enough to guarantee patients, especially those with catastrophic diseases, to get suitable and timely care without running into financial difficulty. At the same time, hospital expenses were usually reimbursed after patients had paid the full medical costs. So if a patient failed to afford the cost in the first place, he could not even get the necessary medical treatment, not even to mention the reimbursements from the NCMS. The situation was made even worse when those who could not pay the cost in advance were actually those need the NCMS the most. If a certain proportion of expense could be exempted in advance, we anticipated that satisfaction rate would increase.

(2) Farmers ' perception of the NCMS

When the NCMS was implemented in 2003, most farmers did not know the NCMS and they did not want to participate in the NCMS. Some local government officials for political reasons forced farmers to take part in the NCMS. In recent years, the number of farmers who voluntarily participate in the NCMS has increased, though some people still take a wait-and-see attitude [25].

The reason of the climbing participation rate was partly due to the increased governmental subsidies. During the in-depth interviews, respondents generally recognized the NCMS policy, even though 77.71% of respondents did not know the detailed regulations of the NCMS. (Table 2).

The study showed that the NCRMS provided inadequate protection against major diseases. Small amount of premium and low government subsidies caused narrow coverage and small reimbursement rate. Once farmers suffered from serious disease and could not afford the medical costs, they would most likely to abandon treatment.

An issue worth emphasizing is that service ability of grass-roots hospitals should be strengthened. Due to long-term insufficient government investments, the infrastructure of rural clinics and township health centers were far from adequate. Medical facilities and equipments were simple. Physicians were lack of proper medical trainings. The medical infrastructure could not meet farmers' medical needs. The quality of care was not warranted.

(3) Policy suggestions

Since the start of the NCMS, more farmers have acknowledged the NCMS and are willing to participate in the NCMS. They know that NCMS policies could bring real benefits. The NCMS would most likely to be successful in the long run with the continuous improvement and more benefits available to farmers.

By 2008, the Chinese government modified fundraising structure and raised reimbursement rate in order to expand the coverage of the NCMS. In the new structure, farmers' premium increased from 10 Yuan to 20 Yuan per year. The governmental subsidies raised to 80 Yuan per capita per year.

Due to asymmetric information of medical care, farmers are in a disadvantaged position; medical care is dominated by physicians and medical institutions. Under the NCMS, medical institutions may tend to provide unnecessary treatments to the patients. Therefore, the administration of the NCMS should be strengthened. The reform of healthcare system must be modified accordingly, including increasing health investments, optimizing health resources allocation, improving qualities of medical care, developing drug delivery system and pricing mechanism in rural China. Medical institutions should adapt themselves to meet the demands of the NCMS and improve medical practice to ensure the quality of care, and protect farmers' legitimate rights.

In rural China, insufficient supply of physicians and inactive adoption of high technology are mainly caused by low incomes of medical staffs and rural medical institutions. As a result, medical staffs with extensive medical trainings are unwilling to work there. State funding should be devoted in rural clinics and township (Commune) health centers, and physicians training should be strengthened.

In order to improve the NCMS steadily and soundly, explicit policies and rules of the NCMS should be established as soon as possible. All these can clarify governments' and medical institutions' responsibility, and provide enough protection for insured farmers. The NCMS shall be ruled, implemented and administrated by law.

This study has one limitation. More women than expected are involved in the sample of this study as more women than men were available in rural areas when the survey was conducted.