Trends in the prevalence and economic burden of hypertension and its socioeconomic disparities in rural southwestern China: two repeated cross-sectional studies

Abstract

Objectives

This study aimed to analyse trends in the prevalence and economic burden of hypertension and its socioeconomic disparities in rural southwestern China.

Design

Two repeated cross-sectional studies were conducted to collect interview and health examination data among individuals aged ≥35 years in rural Yunnan Province, China.

Setting

Three rural areas of Yunnan, China.

Participants

We invited 8187 consenting participants in 2010–2011 and 7572 consenting participants in 2020–2021 to undergo interviews and health examinations.

Results

The standardised prevalence of hypertension significantly increased from 26.1% in 2011 to 40.4% in 2021 (p<0.01), and the per capita direct, indirect and disease economic burdens increased from US$1323, US$46 and US$1369 to US$2196, US$49 and US$2244, respectively. In addition to the indirect economic burden, the direct and disease economic burdens increased significantly. The prevalence of hypertension was higher in illiterate population, among participants with low annual household income per capita, and participants with good access to medical services than in their counterparts who had good education, high annual household income per capita and poor access to medical services (all p<0.05). Moreover, the prevalence of hypertension showed a downward trend with improvement in socioeconomic position (SEP) (p<0.05). The per capita direct and disease economic burdens increased most in participants with low SEP, but the per capita indirect economic burden increased most in participants with upper-middle SEP.

Conclusions

The prevalence and economic burden of hypertension have visibly accelerated in rural Yunnan Province over the 10 years studied, and socioeconomic disparities have been found in the prevalence and economic burden of hypertension. These findings highlight that socioeconomic differentials should be tailored to address the timing of effective interventions for hypertension prevention and control and reduce the economic burden of hypertension in rural southwestern China.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The major strength of this research is the first study to analyse temporal trends in the economic burden of hypertension and its socioeconomic disparities in rural southwestern China.

• In both 2011 and 2021, there were high response rates (97.5% and 98.3%, respectively), and the large sample sizes augment the validity of the results in our study.

• We did not collect data on genetics, psychology, exercise, sleep, blood lipids, altitude, climate and traffic conditions, so the full risk factors associated with hypertension were not included.

• Data regarding the intangible economic burden caused by hypertension such as pain, depression, social isolation, increased living costs and a decline in life quality among patients and their family members were not captured and analysed.

Introduction

Globally, hypertension is a major public health concern. It is also one of the most common chronic conditions leading to premature death and a large economic burden.¹ Studies have shown that hypertension is directly associated with the development of fatal diseases, such as stroke, coronary heart disease (CHD), and heart failure.² In 2019, the age-standardised prevalence of hypertension was 32% in women and 34% in men worldwide, similar to the levels of 32% in women and 32% in men in 1990.³ The prevalence of hypertension has decreased in wealthy countries but risen in many low-income and middle-income countries; 82% of people with hypertension worldwide lived in low-income and middle-income countries in 2019. In accordance with the global trend, the burden of hypertension has increased rapidly in recent years in China. The prevalence of hypertension increased from 14.0% in 1991 to 34.1% in 2015, with the development of social economy, urbanisation, ageing of the population, and changes in people’s lifestyles.^{4 5}

In the Philippines, a Southeast Asian country near China, the economic expenditure of hypertension is expected to increase from US$1 billion in 2020 to US$1.9 billion by 2050 at present value, and in 2020, 70% of the total economic burden of hypertension was accounted for direct and indirect cost.⁶ According to a previous study, the per capita economic burden of hypertension for rural residents is US$1157 in Yunnan Province.⁷ However, there is a lack of data regarding recent changes in the economic burden of hypertension.

There is a positive association between low income, low educational level and low socioeconomic position (SEP) and the prevalence of non-communicable diseases,⁸ including hypertension, despite the indicators of SEP that are used. Nevertheless, little is known about the relationship between SEP and economic burden of hypertension worldwide, especially in rural areas of southwestern China. Notably, hypertension can reinforce poverty and income inequality by causing an upsurge in direct and indirect expenditures and a loss of productivity.⁹

Previous studies have shown that hypertension prevalence is associated with individual SEP and varies according to ethnic minority groups in Yunnan.^{10 11} However, the above studies lack research on the changes in the economic burden owing to hypertension and its socioeconomic disparities in Yunnan. Due to the large population, the estimation of trends in the prevalence and economic burden of hypertension will be helpful for the distribution of public health resources. Information on trends in the prevalence and economic burden of hypertension, as well as those of socioeconomic disparities, is important for strengthening strategies to prevent and control hypertension.¹² Therefore, our study aimed to reveal trends in the change of the prevalence and economic burden of hypertension and its socioeconomic disparities among adults (aged ≥35 years) in rural Yunnan Province using two cross-sectional surveys conducted in 2011 and 2021.

Methods

Study design and participants

Primary data were derived from two cross-sectional interviews and examination surveys, which were completed in three rural areas of Yunnan, China during 2010–2011 and 2020–2021. Yunnan, an underdeveloped area with a population of 46.9 million in 2021, is located in southwestern China; its gross domestic product (GDP) per capita is much lower than that of eastern, central, and southern China. It is also a region with many ethnicities. A total of 25 ethnic groups are living in the area, with a population of 15.6 million, and 15 ethnic groups are unique to Yunnan Province.

In 2011, a four-stage stratified random sampling method was used to select study participants. First, we divided 129 counties of Yunnan Province into three groups according to GDP per capita: high, medium and low. One county was randomly selected from each of the three groups, for a total of three counties. Second, we established two categories from each of the three counties by GDP per capita: advantaged or disadvantaged. One township was chosen randomly from each of those two categories, for six townships in all. Third, we selected three villages from each township based on the probability proportional to size method, for a total of 18 villages. We obtained a list of residents aged ≥35 years who had lived in each village for ≥5 years from the village committee in each selected village. We then used a simple random sampling method to choose eligible individuals in each selected village. In 2021, we used a consistent four-stage stratified random sampling method to select participants from the same three rural areas.

Sample size

This study was part of a study on common cardiocerebrovascular diseases (hypertension, CHD and stroke). In 2021, the sample size was calculated by using the formula: N=(Z_α ² P(1−P)/δ ²)*deff, where Z_α is 1.96 (for two-sided, 95% CI, α is 0.05), P is 0.01 (according to the Report on National Health Services Survey in China (2013), the prevalence rate of CHD (1%) was the lowest among three cardiocerebrovascular diseases), δ is the permissible error of 0.005 (usually take half of P), the design effect (deff) value was set at 1.5. Based on the formula, the sample size was estimated to be 2282 subjects. Assuming a potential non-response rate of 10%, the sample size was 2500 for each county. Because there were 3 counties, the total of sample size was 7500.

A total of 8400 individuals in 2011 and 7700 individuals in 2021 (both groups aged ≥35 years) were invited to participate in the 2 surveys. Of these, 8187 in 2011 and 7572 in 2021 agreed to participate, yielding an overall response rate of 97.5% and 98.3%, respectively.

Data collection and measurement

In both 2011 and 2021, the two surveys were conducted and used similar pretested and structured questionnaires to collect participants’ data. For each consenting individual, a face-to-face interview was conducted by a qualified interviewer. We collected both individual-level and household-level data from participants. We obtained information about individuals’ demographic characteristics (sex, age, ethnicity, educational level, annual household income and access to medical services), medical expenses, health service utilisation, diagnosis, treatment, and control of hypertension. Data on inpatient hospitalisation expenses and outpatient expenses were gathered from the medical records provided by participants whereas data on direct self-medication costs, non-medical expenses, and work absence due to hypertension were based on participants’ self-reports. Non-medical expenses included costs for transportation, nutrition and accommodation on hospital visits. In our study, the economic burden in 2011 was converted into the value in 2021 based on the consumer price index and subsequently converted to USD from Chinese yuan (RMB) using the average annual exchange rate in 2021 of RMB 6.45 per US$1.00. We documented participants’ measured blood pressure (BP) accompanied by those from the survey.

For BP measurement, participants were asked to fast and to not smoke or engage in strenuous exercise. After each participant rested for a minimum of 5 min in a seated position, three successive BP measurements using standardised mercury sphygmomanometers were taken by trained interviewers in both 2011 and 2021. The BP measurement methods followed the recommendations of the American Heart Association.¹³ In this study, we calculated the mean BP of three readings. To ensure the accuracy of BP results taken with a mercury sphygmomanometer, an oscillometer device was used before each survey day.

Patient and public involvement

This study did not involve patients or members of the public.

Definition

Hypertension was defined as mean systolic BP ≥140 mm Hg, or diastolic BP ≥90 mm Hg, and/or currently on antihypertensive medications, according to recommendations in the 8th Report of the Joint National Committee on the Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 8).¹⁴ In this study, we also regarded people who had been diagnosed with hypertension at a qualified medical institution before the questionnaire as having hypertension.

Ethnicity was determined from among China’s 56 state-recognised ethnicities, with 55 of these designated ethnic minorities having a different culture, history, or language from that of the Han population. In Yunnan, the Han population accounts for 66.9% of the total population, and the proportion of ethnic minorities accounts for 33.1% of the total population. In this study, we classified two groups: Han and ethnic minorities (including Naxi, Buyi, and others). Illiterate individuals were defined as participants aged ≥15 years who cannot read with comprehension or write simple sentences in daily life. The level of education was then divided into illiterate and primary (grades 1–6) or higher. Annual household income per capita was categorised as high and low. High referred to an annual household income per capita>US$1083.00, and low referred to a yearly household income per capita ≤US$1083.00. Access to medical services was classified into two categories: good and poor. Good was defined as walking time to the nearest village hospital ≤30 min, and poor was defined as walking time to the nearest village hospital >30 min. SEP is a composite indicator that can evaluate individual’s work experience and living conditions, and its evaluation effect is superior to a single education or income factor. We determined individual SEP according to education level, annual household income per capita and access to medical services. SEP was divided into four levels: low, middle lower, middle upper and high, according to quartiles.

Analysis of economic burden in hypertension

The economic burden of hypertension includes the direct economic burden, indirect economic burden and intangible economic burden.^{15 16} We used a two-step model method to calculate the direct economic burden of hypertension, including direct medical expenditures and direct non-medical expenditures incurred due to hypertension within 1 year. Direct medical expenditures included outpatient costs, inpatient costs and drug costs. Direct non-medical expenditures included transportation and accommodation expenses for the patient and family members. The indirect economic burden was calculated by multiplying the total workdays lost by both patients and family members by the annual household income per capita per day in 2011 and 2021. Thus, a productivity weight was not used in our study. The intangible economic burden of hypertension was difficult to measure, and only the direct and indirect economic burdens of hypertension were calculated in this study. The calculation formulae were as follows: (in per capita)

Direct economic burden=outpatient costs+inpatient costs+drug costs+transportation and accommodation expenses.

Indirect economic burden=lost workdays * annual household income/365.

Economic burden of disease=direct economic burden+indirect economic burden.

Statistical analysis

We used IBM SPSS V.22.0 (IBM Corp.) to analyse the data with double-entry into EpiData V.3.1 (The EpiData Association, Odense, Region Syddanmark, Denmark) software. Descriptive analysis and the χ² test were used in this study. Categorical variables are expressed in terms of number and percentage. A χ² test was used for comparisons of categorical variables between survey years. The prevalence of hypertension was adjusted for age and sex using a direct standardisation method for the total population of surveyed participants aged ≥35 years in 2011 and 2021. Analysis of variance was used to compare the economic burden of hypertension among different groups. The principal component analysis method was used to construct the individual SEP indicators. When Bartlett’s spherical hypothesis test result was that the spherical hypothesis was rejected (p<0.05), this indicated that the data can be subject to principal component analysis, and the principal component extraction standard was that the characteristic root was >1. Statistical significance was set at a two-tailed p value of<0.05 in this study.

Results

The basic demographic characteristics of the participants by survey year are shown in table 1. In total, 3960 men (48.4%) and 4227 women (51.6%) participated in 2011, and 3739 men (49.4%) and 3833 women (50.6%) participated in 2021. There were no significant differences in the proportion of men (p>0.05) and people with low annual household income per capita (p>0.05) between the two survey years. However, the rate of illiteracy was reduced from 30.5% in 2011 to 22.8% in 2021 (p<0.01), and the rate of people with good access to medical services increased significantly from 55.5% to 67.9% (p<0.01).

Table 1.

Basic characteristics of the study population by survey year

Characteristics	2011(n=8187)	2021(n=7572)
Sex
Male	3960 (48.4)	3739 (49.4)
Female	4227 (51.6)	3833 (50.6)
Ethnicity
Han	5008 (61.2)	4124 (54.5)*
Naxi	1998 (24.4)	2531 (33.4)
Buyi	542 (6.6)	747 (9.9)
ther	639 (7.8)	170 (2.2)
Age
35–44 years	1851 (22.6)	1256 (16.6)*
45–54 years	2119 (25.9)	1905 (25.2)
55–64 years	2014 (24.6)	1856 (24.5)
65–74 years	1403 (17.1)	1670 (22.1)
≥75 years	800 (9.8)	885 (11.7)
Level of education
Illiterate	2495 (30.5)	1730 (22.8)*
Primary (grade 1–6) or higher	5692 (69.5)	5842 (77.2)
Annual household income per capita (US$)
≤1083	4196 (51.3)	3844 (50.8)
>1083	3991 (48.7)	3728 (49.2)
Accessibility to medical services (min)
≥30	3645 (44.5)	2434 (32.1)*
<30	4542 (55.5)	5138 (67.9)
SEP
Low	2175 (26.6)	1863 (24.6)
Lower middle	1104 (13.5)	1815 (24.0)
Upper middle	3028 (37.0)	2026 (26.8)
High	1880 (23.0)	1868 (24.7)*

*P<0.01.

Table 2 shows the crude rates and age-standardised rates of hypertension in the two survey years and socioeconomic status in rural Yunnan Province. Over the 10 years studied, the age-standardised prevalence of hypertension boosted from 26.1% in 2011 to 40.4% in 2021 (p<0.01), with increasing rates in sex, ethnicity, age, level of education, annual household income per capita, access to medical services and SEP in all subcategories (all p<0.01). The age-standardised prevalence of hypertension in men and women rose from 25.2% and 26.8% to 43.1% and 37.9%, respectively. Among ethnicities, the prevalence of hypertension in Han, Naxi, Buyi and participants of other ethnic minorities increased from 26.7%, 26.3%, 19.9%, and 26.7% to 43.7%, 36.4%, 33.2% and 52.2%, respectively. The prevalence of hypertension increased from 28.5% in 2011 to 46.6% in 2021 in the illiterate population and rose from 27.7% to 41.3% in participants with low annual household income per capita. In the population with poor access to medical services, the prevalence of hypertension increased from 24.7% to 41.3%. Moreover, the prevalence of hypertension in participants with low, lower-middle, upper-middle and high SEP rose from 28.7%, 27.0%, 24.7%, and 24.0% to 42.4%, 42.7%, 41.0% and 35.8%, respectively, over the 10 years studied.

Table 2.

Age-standardised prevalence of hypertension by survey year and socioeconomic status in rural Yunnan Province, China

Characteristics	2011			2021
	Number (N)	Prevalence (%)	Standardised prevalence (%)	Number (N)	Prevalence (%)	Standardised prevalence (%)
Sex
Male	953	24.1	25.2	1663	44.5*†	43.1
Female	1105	26.1‡	26.8	1485	38.7†	37.9
Ethnicity
Han	1296	25.9	26.7	1841	44.6†	43.7
Naxi	493	24.7	26.3	961	38.0†	36.4
Buyi	103	19.0	19.9	254	34.0†	33.2
other	166	26.0*	26.7	92	54.1*†	52.2
Age
35–44 years	211	11.4	11.4	234	18.6†	18.6
45–54 years	395	18.6	18.6	644	33.8†	33.8
55–64 years	549	27.3	27.3	857	46.2†	46.2
65–74 years	563	40.1	40.1	903	54.1†	54.1
≥75 years	340	42.5*	42.5	510	57.6*†	57.6
Level of education
Illiterate	715	28.7*	28.5	797	46.1*†	46.6
Primary (grade 1–6) or higher	1343	23.6	25.3	2351	40.2†	38.2
Annual household income per capita (US$)
≤1083	1141	27.2*	27.7	1642	42.7†‡	41.3
>1083	917	23.0	24.3	1506	40.4†	39.2
Accessibility to medical services (min)
≥30	876	24.0	24.7	964	39.6†	38.5
<30	1182	26.0‡	27.0	2184	42.5†‡	41.4
SEP
Low	653	30.0*	28.7	756	40.6†	42.4
Lower-middle	283	25.6	27.0	797	43.9†‡	42.7
Upper-middle	718	23.7	24.7	864	42.6†	41.0
High	404	21.5	24.0	731	39.1†	35.8
Total	2058	25.1	26.1	3148	41.6†	40.4

*P < 0.01 (comparison within the same characteristic groups in the same year).

†P < 0.01 (comparison to in 2011).

‡P < 0.05.

In both 2011 and 2021, the prevalence of hypertension was higher in the illiterate population, among participants with low annual household income per capita, and participants with good access to medical services than in their counterparts who had good education, high annual household income per capita and poor access to medical services (all p<0.05). Among ethnicities, the prevalence of hypertension by ethnicity was in the order of Han, Naxi, and Buyi in both 2011 and 2021 (both p<0.01). In both survey years, the prevalence of hypertension increased with age (both p<0.01). Moreover, it showed a downward trend with improvement in SEP (p<0.01 and p<0.05). In 2011, the prevalence of hypertension was higher in female participants than in male ones (p<0.05), but it was higher in female participants than in male participants (p<0.01) in 2021.

Table 3 presents the economic burden of hypertension by survey year and socioeconomic status in rural Yunnan Province. Over the 10 years from 2011 to 2021, the per capita direct, indirect and disease economic burdens increased from US$1323, US$46 and US$1369 to US$2196, US$49 and US$2244, respectively. In addition to the indirect economic burden, the direct and disease economic burdens increased significantly. Increasing costs of the per capita direct, indirect and disease economic burdens were discovered among the most subgroups. However, the per capita indirect economic burden was decreased in women, ethnic minorities and participants with lower-middle SEP from 2011 to 2021. The per capita direct, indirect, and disease economic burdens increased by US$873, US$2 and US$875, respectively. Moreover, increments of the per capita direct and disease economic burdens were higher in men and Han Chinese than in women and ethnic minorities. Among SEP subgroups, the per capita direct and disease economic burdens increased most in participants with low SEP, but the per capita indirect economic burden increased most in the upper-middle economic status group. The costs of outpatient, inpatient and drug rose significantly from US$115, US$1005 and US$146 in 2011 to US$228, US$1664, and US$249 in 2021, respectively. However, the costs of transportation and accommodation declined from US$57 to US$54. Importantly, direct costs comprised the major constituent of total expenses, and inpatient costs were the biggest driver of the direct economic burden in both 2011 and 2021.

Table 3.

Economic burden of hypertension ( $\overline{x}$ ± s, in USD, in per capita) by survey year and socioeconomic status in rural Yunnan Province, China

Characteristics	Years	Direct economic burden					Indirect economic burden	Economic burden of disease
		Outpatient costs	Inpatient costs	Drug costs	Transportation and accommodation expenses	Total
Sex
Male	2011	105.1±47.3	1247.1±699.4	156.9±72.6	31.9±14.7	1541.0±682.9	50.3±26.5	1591.4±774.3
	2021	227.8±69.8	1975.6±839.8*	235.1±69.8	84.1±40.6†	2522.6±777.1	58.3±25.2	2580.9±777.9
Female	2011	120.8±63.2	822.8±590.0	140.1±73.8	74.5±5.7‡	1158.2±467.0	43.4±14.4	1201.6±555.8
	2021	228.9±75.0	1276.4±538.8	260.6±77.8	20.1±5.5	1786.0±518.6	37.9±13.6	1823.9±520.7
Ethnicity
Han	2011	112.1±59.8	807.7±546.8	131.8±74.3	62.6±36.7	1114.2±570.3	27.4±11.8	1141.7±659.1
	2021	208.9±66.0	1562.3±600.1	246.1±81.0	82.6±40.8*	2099.9±630.5	31.2±9.7	2131.1±634.5
Minority	2011	120.4±67.5	1398.6±605.4‡	172.5±89.8‡	47.0±15.4	1738.4±662.2	78.6±26.0 §	1817.1±624.0
	2021	254.2±80.2	1720.1±663.7	256.0±96.8	38.7±15.7	2269.0±726.5	59.8±23.5	2328.7±677.7
SEP
Low	2011	119.7±64.4	690.9±404.4	132.6±70.8	75.1±9.3	1018.3±531.1	29.3±7.7	1047.6±537.4
	2021	191.5±82.8	1424.2±496.9	228.3±86.8	25.8±6.6	1869.8±786.4	47.6±15.7	1917.4±898.1
Lower-middle	2011	135.9±50.4	1812.1±651.7	137.5±78.4	45.8±11.8	2131.4±816.1	76.8±12.1	2208.2±773.2
	2021	209.9±67.0	2275.4±613.3	246.9±109.5	26.7±8.8	2758.9±1100.4	33.2±11.6	2792.2±945.9
Upper-middle	2011	117.7±67.6	836.6±460.0	135.3±80.5	52.2±13.2	1141.8±536.1	43.9±19.3	1185.7±677.0
	2021	232.9±78.1	1376.4±536.8	250.1±123.8	41.9±16.3	1901.3±1043.6	68.8±23.8	1970.1±1013.5
High	2011	85.2±38.8	1409.1±675.4	204.9±72.4	58.8±17.1	1758.0±721.6	47.9±15.4	1805.9±718.9
	2021	275.9±125.0*	1742.0±1131.8	273.8±87.4	125.0±54.1	2416.6±997.9	48.4±12.4	2465.0±1131.8
Total	2011	115.2±66.2	1004.7±388.9	146.4±67.8	56.8±11.7	1323.0±667.5	46.4±12.4	1369.4±651.2
	2021	228.4±74.3	1664.2±882.5	249.0±139.5	54.0±19.2	2195.6±1217.3	48.8±11.3	2244.4±1080.9

*P<0.05.

†P<0.01 (comparison within the same characteristic groups in 2021).

‡P<0.05.

§P<0.01 (comparison within the same characteristic groups in 2011).

In 2011, expenses for transportation and accommodation were higher in women than in men, and inpatient and drug costs as well as the per capita indirect economic burden were higher in ethnic minorities than in the Han population. However, in 2021, inpatient and transportation and accommodation costs were higher in men than in women, and expenses for transportation and accommodation were higher for Han Chinese than for ethnic minorities. Additionally, outpatient costs were greatest in the group with high SEP among all SEP subcategories in 2021.

Discussion

Our two cross-sectional studies indicated a high prevalence rate of hypertension, a heavy economic burden of hypertension and a total increase in the hypertension prevalence and economic burden in rural southwestern China over the 10-year study period. The findings also demonstrated that socioeconomic disparities have substantial associations with the actual prevalence and economic burden of hypertension, as well as temporal trends in the hypertension prevalence and economic burden.

In recent studies, the standardised prevalence rate of hypertension (40.4%) among adults aged≥35 years in rural Yunnan Province has been found to be greater than the prevalence rates observed in China (at national level) and other countries.^{5 17–19} It was suggested that there are regional differences in the prevalence of hypertension. Furthermore, in our study, we discovered a significantly increasing tendency in hypertension prevalence over the 10 years studied (from 26.1% in 2011 to 40.4% in 2021). This trend was consistent with the finding of our previous study (from 28.4% in 2009 to 39.5% in 2016),¹¹ but was inconsistent with that of another study,²⁰ and it revealed that the prevalence of hypertension decreased moderately in China after 2010. The reason may be that the research scope differed between studies (Yunnan Province and the whole country). Our study showed that hypertension is a leading and increasing public health problem in the studied rural areas of Yunnan Province. This possibly results from China’s Basic Public Health Services project, which was launched in 2009. The project aim was to improve awareness, diagnosis and treatment and control of chronic diseases (eg, hypertension) by making several policies to integrate a health delivery system based on primary healthcare. Yunnan has experienced excellent economic growth over the past 10 years. This growth has augmented the number of individuals exposed to unhealthy lifestyle factors related to hypertension. The findings imply that more effective interventions are required to prevent and manage hypertension in rural areas, particularly by changing poor lifestyle habits.

Our study indicated that the prevalence of hypertension increased among rural residents in Yunnan Province from 2011 to 2021, along with the per capita economic burden of hypertension, which increased from US$1369 in 2011 to US$2244 in 2021. However, the aforementioned costs were lower than those reported by studies conducted in Canada (US$2341) and the United States (US$3914),²¹ but were higher than in most low-income and middle-income countries.^{22 23} This disparity could be explained by variation in SEP and the asymptomatic nature of hypertension,²⁴ a noteworthy number of residents with undiagnosed hypertension, and differences in healthcare systems and the level of care. Additionally, the cost of living, goods and services varies worldwide, which is also an important reason for incongruities.²⁵ In general, consumer prices are lower in low-income and middle-income countries than in high-income countries. China remains the largest developing country in the world.²⁶ Our study revealed that direct economic expenditures made up the principal proportion of total expenses for hypertension, and inpatient costs overwhelmingly drove these direct costs. Moreover, our study indicated that inpatient expenditures place a heavy monetary burden on rural residents in Yunnan Province, with a high risk of causing or exacerbating poverty. Thus, reducing the costs of hospitalisation, reducing the time of hospitalisation and increasing the reimbursement for inpatient services are urgently needed in policy-making to protect rural residents against poverty owing to illness.

The present findings showed ethnic variance in temporal trends of the prevalence of hypertension in rural Yunnan Province. This finding was consistent with our previous study.¹¹ Over the 10 years studied, the prevalence of hypertension has increased in the Han population more than in Naxi and Buyi ethnic groups. Furthermore, the recent study also showed ethnic difference in temporal trends of the economic burden of hypertension. The per capita direct and disease economic burdens of Han people were higher than those of ethnic minorities. In both 2011 and 2021, Han Chinese had a higher prevalence of hypertension than Naxi and Buyi. A possible reason is that ethnic minorities have healthier lifestyles, for instance, consuming healthier food and engaging in more physical activity.²⁷ In 2011, hospitalisation expenses, pharmacy purchase expenditures and the indirect economic burden were higher in ethnic minority residents than those in Han Chinese, but there were no significant differences by 2021. This possibly results from improved health equity among different ethnic groups with economic and social development of Yunnan, especially the popularisation of medical insurance in ethnic minority areas. In 2021, the costs of transportation and accommodation were higher in the Han population than those in ethnic minorities, which may be related to the better economic conditions of Han Chinese as a whole and the ability to attend better medical institutions for treatment after illness.

In this study, participants who were less educated, had lower income and had good access to medical services were more prone to having hypertension than their counterparts who were more educated and had higher income or poor access to medical services. The findings are consistent with those of previous studies.^{28 29} These results can be partly explained by the fact that less educated and had lower income participants had a higher prevalence of obesity, central obesity than their more educated and had higher income counterparts.¹¹ And overweight, obesity and central obesity are well-documented, established primary risk factors for hypertension.^{7 30} Participants who had good access to medical services usually have more opportunities for professional BP measurement and management than in those with poor access, thereby increasing the detection rate of hypertension. Moreover, the prevalence of hypertension increased in all subcategories, indicating that it is also necessary to strengthen the prevention and management of hypertension, innovate prevention and control measures, and enhance the effect of prevention and control.

In our study, the SEP of men was higher than that of women in both 2011 and 2021. Encouragingly, compared with 2011, SEP has improved in women. In 2011, there was a negative correlation between SEP and the prevalence of hypertension. Although there was also a negative correlation overall in 2021, the highest prevalence occurred in the lower-middle SEP group. In total, the prevalence of hypertension showed a downward trend with increased SEP, which is consistent with a previous study.²⁹ This possibly results from individuals with lower SEP having relatively poor lifestyle habits, such as smoking, drinking alcohol, engaging in few sports activities and poor food quality.³¹ Thus, it is urgent to guide and encourage people with lower SEP to adopt a healthy lifestyle to reduce the risks of hypertension.

In both 2011 and 2021, the per capita economic burden of hypertension was heaviest in participants with lower-middle SEP, which may be related to the government’s attention toward residents’ health and the promotion of hypertension management, as well as increased opportunities and willingness of this population to participate in the diagnosis and treatment of hypertension. Moreover, the outpatient expenses in participants with high SEP were the highest among SEP subgroups in 2021, suggesting that residents with good socioeconomic conditions can visit an outpatient clinic more frequently after illness and adopt better preventive measures. In addition, poor access to medical services is the main barrier to accessing medical resources. Therefore, improving access conditions by adding and strengthening clinics’ level of medical services in rural areas will help to reduce the economic burden of hypertension.

Some limitations in this study should be taken into account. First, we only collected data in two cross-sectional surveys, and the change trend in each year across the decade studied could not be analysed. Furthermore, causal relationships cannot be determined in our study. Second, the 10-year study period is a comparatively short time in which to analyse trends in the prevalence and economic burden of hypertension. Analysis over a longer time range can enhance the robustness of our findings. Third, we did not collect data on genetics, psychology, exercise, sleep, blood lipids, altitude, climate and traffic conditions, which may be vital factors influencing hypertension. Hence, not all risk factors associated with hypertension were included in the analysis. Fourth, we did not collect and analyse data regarding the intangible economic burden caused by hypertension such as pain, depression, social isolation, increased living costs and a decline in life quality among patients and their family members. Therefore, this study may have underestimated the true burden of hypertension.

Conclusions

Both the prevalence and economic burden of hypertension are enormous in rural southwestern China. The hypertension prevalence and economic burden increased significantly from 2011 to 2021, with clear socioeconomic disparities. Because hypertension is also an important risk factor for other cardiocerebrovascular diseases (such as CHD and stroke), strengthening the management and control of hypertension is key to the prevention and treatment of cardiocerebrovascular diseases for individuals and policymakers. Socioeconomic differentials should also be considered when developing effective interventions to promote the prevention and management of hypertension and reduce the economic burden of hypertension in rural southwestern China.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by the Ethics Committee of Kunming Medical University (approval Number: KMMU2020MEC031). Participants gave informed consent to participate in the study before taking part.