Regional differences and determinants of self-rated health in a lower middle income rural Society of China

The study was approved by the Anhui Medical University Research Ethics Committee. Each participant received an explanation of the significance of the study and a form that outlined the protection of personal information from trained data collectors. All participants provided written informed consent.

Anhui province is located in the east of China. The province boasted 61.436 million permanent residents in 2015 (ranking 8th among 34). The Gross Domestic Product (GDP) of Anhui was 5808 dollars per capita in 2015, ranking 25th among 34. The Percentage of healthcare to consumer expenditures in rural areas was 8.99% [18]. The study included Anhui rural residents over the age of 15 years old as research subjects. Under a multi-stage stratified random sampling method, the survey randomly selected 3 of the 16 municipalities, representing three geographical areas (southern, central and northern regions). In each 3 regions, 2 counties were selected by economic level (wealthier and poorer). Three villages were randomly selected from each county. A total of 18 villages in 3 regions were selected. All the permanent and available residents from farming households were invited to participate in the survey. In this study, “permanent residents” were defined as those who spent more than 6 months out of the year at the registered place of residence. From June to August 2015, the survey was conducted in the subjects’ homes. The questions covered demographic and socio-economic information on individuals and households, socio-economic status, physical health, and healthy lifestyle habits.

A total of 3870 valid questionnaires were collected from the 3 regions. Additional information on the geography and economic profiles of the regions was collected from local statistics bureaus (Table 1).

The central region was close to the provincial capital, and the per capita. The GDP per capita was $2986 in 2015. The southern region showed the highest economic level (per capita GDP, $13001) and the highest disposable income ($2835 per capita). Residents of the northern plain lived closer to each other (652 persons per km²), the lowest economic level and the fewest number of medical practitioners (1.07/1000 residents). However, the utilization rate of city and township hospital beds were highest in the northern region (86.66% and 80.70%, respectively). The disparity of utilization of township hospitals beds among the 3 regions was larger than that of city hospitals (Table 1).

As shown in Table 2, the demographic composition of the 3 regions indicated no statistically significant difference in terms of marital status (P > 0.05), but all of the demographic social traits were statistically distinct. There were more residents who were female (63.24%), over 65 years of age (29.50%), without education (52.02%), subsistence farmers (83.49%), and elderly living alone (50.74%) in the northern region than elsewhere. Most of the sampled households used by NRCM coverage, especially in the northern region (98.81%). Northern residents demonstrated the highest social support (47.98%), chronic illnesses (57.90%), injury or illness within the past two weeks (60.00%), hospitalizations within the past year (77.50%), outdoor exercise (25.46%), and abstinence from alcohol (86.21%). However, few northern residents smoked (16.39%).

In general, there were statistically significant differences with regard to the association between SRH and demographic traits, with the exception of gender, medical insurance and active exercise (Table 3). The SRH levels of male and female did not prove significantly different between the central and northern regions.

Among all participants, the best levels of SRH among respondents aged 15 to 44 years were among participants who had higher levels of education, and were not subsistence farmers. Those who were married had higher SRH. Greater income also was associated with higher SRH. A total of 65.09% of the residents in the poorest income quartile in the northern region reported the poorest SRH. Elderly who lived alone reported the poorest.

Among people in the northern region, those without medical insurance reported the poorest SRH (46.23%). However, medical insurance varied in the different regions. The residents who were covered by NRCMs exhibited the poorest SRH in the southern region (24.68%) and northern region (48.00%). However, the SRH of respondent with different medical insurance plans in the central region showed no statistical difference among them. Regardless of the region where the respondents resided, the entire sample with better social support had higher SRH scores. However, the trend was not consistent across the three regions; the highest SRH were found in quartile 4 in the southern and northern regions, and in quartile 3 in the central region.

The participants with chronic illness, illness in the past two weeks, or a hospitalization within the past year, had poorer SRH than other participants. More than half reported the poorest SRH among respondents with chronic disease, had made a clinic visit and had been hospitalized in the northern region (64.60%, 63.88$ and 65.84%, respectively). However, the highest rate of poor SRH was among respondents with poor physical health in the southern and central regions.

In the central region, the population who actively exercised had higher SRH than other participants (33.55%). In the southern and northern regions, the percentage of participants who exercised showed no statistical difference compared to the other respondents. However, most respondent who smoked or drank reported better SRH. These traits were more clearly represented in the southern and central regions; in the northern region, the highest rate of poor SRH was found among responders who did not smoke or drank.

The odds ratios of SRH in a series of models are presented in Table 4. In the first, districts were included, followed by a regression analysis of individual demographic characteristics, family factors, physical health and healthy lifestyle habits.

Model 1 served to demonstrate that the place of residence was a significant determinant of poor SRH. The respondents who lived in the northern region were more likely to have poor SRH.

Model 2 indicated that gender, age and education were all associated with SRH. Female respondents were 1.184 times more likely than males to have poor SRH. The likelihood of poor SRH levels increased with age. Education demonstrated a protective effect; participants with higher level of education were less likely to have poor SRH.

As shown in the Model 3, the residents with higher incomes were less likely to have poor SRH than those who were impoverished. Participants in the wealthiest income quartile reported low SRH less than (0.447 times) those in the poorest quartile. However, marriage status, family composition and employment demonstrated no significant impact on SRH levels.

In Model 4, social support was included. The respondents who reported strong social support were 0.608 times more likely to have a poor SRH. Medical insurance was not included in the model, although the family composition was. The elderly living alone were more likely (1.118 times) to report poor SRH than the others.

Model 5 included the variables related to physical health: chronic diseases, illnesses, clinic visits and hospitalizations. These physical health variables were all significantly associated with SRH.

All of the covariates were analyzed in Model 6. The smoking population showed a lower likelihood (0.745 times) to report poor SRH. Furthermore, the people who drank more than twice a week were 0.627 times less likely to have poor SRH than people who did not drink. Respondents who actively exercised did not show lower SRH levels. Furthermore, gender was proved irrelevant in this model.

Logistic regression analysis revealed that region, age, education, household income, social support, physical health, healthy lifestyle habits were all associated with SRH.

The study’s findings factors of importance among regions with wide disparities, even within the same province. More importantly, the results suggested associations between SRH and a variety of factors, women, elderly, respondents with lower level of education, who lived alone, lower-income individuals, people with low levels of social support, and those with poor physical health all fared worse on SRH [10, 14, 19‐21]. Additionally, smoking and consuming small amounts of alcohol were associated with good SRH.

Region of residence was found to affect SRH scores in all of the models. Previous studies have demonstrated that place of residence proved relevant to SRH scores [8, 22]. The findings form this study were consistent with those form previous studies, that suggested that household better economic conditions positively impacted SRH [15]. In this study, the northern region’s lower economic levels and decreased access to medical resources affected the quality of health services provided that, in turn, affected its residents’ reported health. Policymakers should pay attention to the northern residents’ relatively poor subjective health scores. Economic and public policies that address inequities in the quality of health care services provided may improve residents’ health [23].

This research demonstrated that marital status had no significant effect on SRH. Previous research also has shown that there were that no significant correlations between SRH and housing quality, or marital status [14]. The fact that the majority of the participants in this study were married may have contributed to this finding. High levels of social support corresponded to higher SRH scores. The measure of SRH permitted an individual to independently determine which factors contribute to their health. Rich social relationship networks could have contributed to positive perceptions about health, low levels of social support could be associated with low SRH [10, 24]. Perceptions about one’s own health could be enhanced through the creation of positive community environments and harmonious family atmospheres. The participants who experienced hypertension, diabetes, cancer or had been hospitalized were more likely to report poor health [17, 25]. The perception of a person’s overall health was affected by the appearance of chronic conditions or serious illnesses. Therefore, subjective perception also depended on objective health [5, 26].

This study evidenced no differences in SRH among residents with different types of medical insurance. This may have been because residents in these regions enjoyed coverage primarily through the NRCMs, with less access to other types of insurance. Since 2009, Anhui Province has gradually implemented a merger between NRCMs and urban resident medical insurance. The new insurance model will be referred to as “urban and rural resident medical insurance.” In some regions—particularly in the south—rural businesses have purchased private insurance or urban employee medical insurance for some part-time rural resident workers. The fact that some workers enjoyed double coverage could partially explain why medical insurance did not seem to impact SRH scores. This also demonstrated that access to additional types of insurance coverage (such as commercial medical insurance) and implementation of a more comprehensive medical insurance package could serve to improve rural residents’ perception of their health [22].

The study sustained findings form previous research that better SRH was reported among people who smoked cigarettes and consumed alcohol [27]. However, this finding contradicted an earlier study that associated smoking and drinking with lower SRH scores [28] and another that found that the elderly with healthier lifestyles believed themselves to be healthier than those with less healthy lifestyles [29]. The participants in this study showed high incidence of unhealthy lifestyles; a little over one-fifth of them (20.39%) smoked. Interestingly, most of the smokers and drinkers were male and in good physical health. This could help to explain why smoking and drinking was associated with positive SRH. The possibility exists that the outcome resulted from the populations’ desire to believe that they were healthy despite their smoking and drinking habits. Furthermore, there could also be a cultural explanation for the association between smoking and drinking and good health. Both activities typically serve as a means to socialize. However, this does not in any way contradict the extensive research that has confirmed the harm caused by smoking or excessive drinking.

This was cross-sectional study, no cause-effect relationship between each social factors and SRH could be determined.