1. Top of page
  2. Abstract


Little is known about inter-individual variation in psychological stress in modernizing rural populations in developing countries. The objective of this study was to investigate the association between Epstein-Barr virus (EBV) antibody titer, a biomarker of psychological stress, and lifestyle parameters in rural communities in Hainan Island, China.


EBV antibody titer was measured in dried blood spot (DBS) samples collected from 240 adults living in rural communities in Hainan Island. Measures of two major lifestyle changes in rural areas of Hainan Island, i.e., diet and commercial goods possession, were examined by using information on the weekly frequency of pork consumption and monthly mobile phone fees. Both parameters were standardized to create two composite scores: a total affluence index (TAI—the sum of these scores), and an investment tendency index (ITI—the difference between them). Least-squares regression analysis was used to investigate the association between EBV antibody titer and these lifestyle parameters.


A positive association was found between the frequency f pork consumption and EBV antibody titer (P = 0.040), whereas a negative association was found between mobile phone fees and EBV antibody titer (P = 0.028). A negative association was also observed between ITI and EBV antibody titer (P = 0.002) after adjusting for the effect of TAI.


Psychological stress among local residents was linked to the type of lifestyle changes they had experienced, where the adoption of a more market-oriented lifestyle, irrespective of current affluence, was associated with less psychological stress. Am. J. Hum. Biol. 26:36–42, 2014. © 2013 Wiley Periodicals, Inc.

In recent decades, rapid economic growth in developing countries has brought about drastic changes in people's lifestyles. Such changes have been observed at both the individual level [e.g., changes in diet, physical activity and disease profiles (Monda et al., 2007)] and societal level [e.g., changes in infrastructure, land use, transportation planning and urban design (Popkin, 2001)]. The combined effects of this process of change have been termed “modernization.”

It has been suggested that modernization is a psychological stressor (Brown, 1982; Jenner et al., 1987; Pearson et al., 1993). For example, using the urinary concentration of catecholamine as a biomarker of psychological stress, Pearson et al. (1993) showed that Samoan adults who resided in Honolulu, the provincial capital of Hawaii, the United States of America, had higher levels of psychological stress than their counterparts living in rural Western Samoa. Similarly, McDade et al. (2000a) used another biomarker of psychological stress—Epstein-Barr virus (EBV) antibody titer—to highlight how urban children and adolescents had higher psychological stress than their rural peers in Western Samoa.

As yet, however, little attention has been paid to whether there might be inter-individual variation in psychological stress within modernizing rural populations. This may be an important gap in our knowledge especially as some evidence suggests that the modernization process has induced diversification in the lifestyle of individuals (e.g., types of activity and diet), even within rural areas (Adelman and Morris, 1973). Moreover, social comparison theory provides a mechanism for explaining how diversification within what were once homogenous communities might play a role in deteriorating psychological health. It highlights how the perception of an undesirable difference in lifestyle parameters with other community members, especially those whom a person believes to be comparable to him or herself (defined as a reference group), may be detrimental in terms of increasing the psychological stress of those who fare worse in the comparison (Festinger, 1954). Given this, investigating different types of lifestyle in such an environment might shed light on the impact that modernization has on local residents' psychological stress due to the changes it brings about in different forms of behavior (i.e., lifestyle).

This study was conducted among rural communities in Hainan Island, China. People in the rural communities in this location lived generally impoverished lives until the 1990s. However, from the late 1990s onwards rapid economic development occurred there and the communities as a whole became much wealthier. One result of this was that people's lifestyles increasingly diversified. The authors' observations among the local communities on the island suggest that there may have been two typical changes in lifestyle. First, there has been a change in some aspects of dietary behavior. In particular, after the year 2000 the population started to consume meat on a regular basis (Jiang et al., 2006). In traditional society, meat was rarely consumed as it was served exclusively on special occasions in Hainan Island. At that time, the regular consumption of pork was merely an aspiration and symbol of an ideal life for local residents. Second, commercial goods such as the motorbike, television, and mobile phone were introduced into the communities. For example, research by the authors has highlighted how by 2009, many of the households in one of island's rural communities owned televisions and tractors, and some owned mobile phones and motorbikes, whereas both poverty and a lack of electricity prevented residents from owning such durables before the late 1990s (Inoue et al., 2012).

Since the lifestyle changes induced by the process of modernization might not be uniform and might affect people's psychological status differently, even within a single community, the authors examined two major components of lifestyle change: diet and the possession of commercial goods. These were chosen to reflect different aspects of the change and its potential to impact on psychological stress, to better understand the psychological consequences of modernization in the rural areas of Hainan Island.

The objective of this study was to investigate the relation between changes in two conventional lifestyle markers, i.e., diet and the possession of commercial goods and psychological stress in the rural communities in Hainan Island, China, using EBV antibody titer as a biomarker of psychological stress.


  1. Top of page
  2. Abstract

Research location

Hainan Island is located to the south of Mainland China. It has land area of 33,210 km2 with a subtropical monsoon climate. In 2009, the population consisted of 8.8 million people (Statistical Bureau of Hainan Province, 2010). The largest ethnic group is the Han (comprising 83% of the total population), followed by ethnic minorities such as the Li (16%) and Miao (0.8%). Recent data (for 2009) show that the annual per capita income was 13,751 RMB in urban households and 4,744 in rural households RMB (1 RMB (renminbi) = 0.15 USD in 2010) (Statistical Bureau of Hainan Province, 2010).

Figure 1 shows the location of the regions where this study was undertaken. Four communities are located in Wuzhishan City and three in Baoting Li and Miao Autonomous County. These regions were chosen for this study because they are both in an initial stage of economic development, located inland and share a similar historical and cultural background. Before 2000, the people living in the communities selected from these regions had limited cash income. All households engaged in subsistence lifestyles such as rice cultivation, collecting wild edible plants and catching small animals. Conventional amenities such as electricity, telephone services, and piped tap water supply were unavailable (Umezaki et al., 2009).


Figure 1. Location of the study regions.

Download figure to PowerPoint

It was around 2000 that development projects, which had been initially launched in the coastal areas of the island, started to operate in the more rural mountainous areas. As a result, local residents began to obtain more diverse sources of income from different forms of employment such as cash cropping (e.g., longan, lychee, mango, betel nut, rubber, banana, and tea leaves), migratory work (e.g., shop clerks, construction workers, drivers of automobile or motorbike) as well as the vending of daily foods and various material goods. These changes meant an increased cash income and have led to a growth in meat consumption, and in the purchasing of commercial goods (Jiang et al., 2006; Liang et al., 2003).

However, these changes do not seem to have reduced the level of psychological stress among residents. Based on their fieldwork in a Wuzhishan community in 2001, Umezaki et al. (2009) reported that people evaluated their daily lives positively when they were living in an environment in the initial stage of economic development. In particular, they enjoyed the better production of rice achieved by the introduction of new cultivars and agrochemicals. However, follow-up fieldwork conducted in the same community in 2009 revealed that the scores obtained from the abbreviated version of the World Health Organization's Quality of Life Questionnaire (WHOQOL-BREF) were lower than those previously reported in most of the population around the world. This was especially evident among those without a paid job, possibly as a result of a widening of the gap between what the villagers desired and what they could afford between 2001 and 2009 (Inoue et al., 2012).

Study subjects

This study was conducted as part of a collaborative research project between the Department of Human Ecology, the University of Tokyo and Hainan Provincial Center for Disease Control and Prevention (Hainan CDC). To examine the individual determinants of health transition in rural China 2,272 participants in six different regions of Hainan Island were selected. Biological samples, questionnaire data and anthropometric data were collected by the authors and staff of Hainan CDC, in November to December 2010.

Of the 2,272 participants in the project, 612 resided in the Wuzhishan and Baoting regions. A quota sampling strategy was employed to select the target samples for the measurement of EBV antibody titer (see below). These 612 people were stratified by 10-year age group (20–29 years, 30–39 years, 40–49 years, and 50–59 years), sex (male and female), and region (Wuzhishan and Baoting); 15 people were then randomly selected from each sub-group, to give a total sample size of 240 people for this study.

The study protocol was approved by the University of Tokyo (ethnical approval No. 3235) and Hainan CDC. Written informed consent was obtained from all study participants before the sample and data collection.

Epstein-Barr virus antibody titer: a biomarker of psychological stress

EBV is one of the ubiquitous herpes viruses, with which almost 100% of adults in developing countries and about 90% of adults in developed countries are infected (Henle and Henle, 1973). Once infected, one carries EBV as a lifelong infection in a latent state, which is regulated by cell-mediated immune function (Macsween and Crawford, 2003; Thorley-Lawson and Gross, 2004).

Stress-induced up-regulation of the hypothalamic-pituitary-adrenal axis (HPA axis) increases production of glucocorticoids that suppresses immune function. As a result, latent state EBV reactivates followed by the production of anti-EBV antibodies (Glaser et al., 1991). EBV antibody titer is a useful biomarker in that it does not have a circadian rhythm (McDade et al., 2000a) and has a long half-life of approximately 20 days in serum (Cacioppo et al., 2002).

The potential utility of EBV antibody titer is suggested by the fact that it has been correlated with the scores from conventional psychological stress questionnaires, e.g., the Profile of Mood Stress short form (Lutgendorf et al., 2001), and the Life Event Scale (Cao et al., 2009). In addition, it has also been associated with life strain and traumatic life events (Esterling et al., 1992), poor quality marriages (Kiecolt-Glaser et al., 1987a), medical students' examinations (Glaser et al., 1987), family stressors (Panter-Brick et al., 2008), and caregiving to patients with Alzheimer's disease (Kiecolt-Glaser et al., 1987b). Given this, in this study the authors used EBV antibody titer as a biomarker of psychological stress.

EBV antibody in dried blood spot

Dried blood spot (DBS) samples were used for the measurement of EBV antibody titer. A tiny amount of whole blood was directly collected on quantitative filter paper (903 Protein Saver Cards, Whatman, Kent, UK). Samples were dried overnight, and then transported to the community hospital, where they were refrigerated at −20°C before shipment to the laboratory in Hainan CDC. Upon arrival at Hainan CDC, samples were stored at −80°C until the analysis.

EBV antibody titer was analyzed by enzyme-linked immunosorbent assay (ELISA) in the laboratory in Hainan CDC. Anti-EBV viral capsid antigen (VCA) IgG was measured using a commercial kit (ETI-VCA-G P001606, DiaSorin, Saluggia, Italy). One disk of dried blood (1/8 inches) was eluted overnight in 250 μL of sample diluent. Then, 100 μL of the diluent was put into microtiter wells coated with anti-EBV VCA antibody on the second day. After incubating at 37°C for 1 h, secondary antibody was added and again incubated at 37°C for 1 h. 100 μL of chromogen/substrate was added to the wells, followed by reactivation. 200 μL of stop solution was then added. Absorbance was read at 450 nm by microplate photometer (Thermo Multiscan FC, Thermo Fisher Scientific, Waltham, MA). Every sample that gave a reading over the highest calibrator was immediately two-fold diluted with stop solution and the absorbance was read again. The protocol for analyzing EBV antibody titer from DBS was developed by the Department of Anthropology in Northwestern University (McDade et al., 2000b).

Samples whose antibody titers were less than 20 ELISA units (EUs) were regarded as EBV-seronegative or “not infected” (McDade et al., 2000b). As a result of this the authors excluded 21 participants from further analysis. McDade et al. (2000b) reported that the measurement of EBV in DBS had good sensitivity, precision, reliability, and a high correlation with matched plasma EBV antibody titer. C-reactive protein (CRP) concentration in DBS was also measured by ELISA to distinguish individuals at an acute stage of systemic inflammation, infection, or tissue injury (Libby and Ridker, 1999; Mortensen, 2001), following the protocol by Brindle et al. (2010). This process was justifiable since immune function is influenced by current infection; higher levels of inflammation may drive EBV reactivation (McDade et al., 2000b). Samples that registered higher than 6.25 mg/l in DBS CRP, equivalent to 10 mg/l in serum CRP (Brindle et al., 2010), were excluded before the samples were stratified.

Identification of modernization parameters

Information was collected from the study participants about their demographic characteristics, educational level, the weekly consumption frequency of eight selected food items that would have represented dietary change across Hainan Island (i.e., major sources of protein and instant noodles) and mobile phone fees paid in the previous one month. Details of the consumption frequencies of the selected food items were obtained by asking the question “how many times a week do you eat each food item?” During the interview, information on the respondents' monthly mobile phone fees was obtained from the mobile operator by means of a text message.

From the information on the weekly consumption frequency of the food items, the frequency of pork consumption was hypothesized to signify dietary change associated with modernization in Wuzhishan and Baoting regions. The use of pork consumption as the marker of potential differences emanating from the modernization process was based on the authors' observation that since the year 2000 pork was the food item that varied most in terms of the frequency of its consumption among the residents. This contrasts markedly to the situation before 2000 when few people consumed it in their daily life.

As distinct from the possession of other consumer goods such as a television or motorbike, a monthly mobile phone fee can be regarded not so much as a marker of the degree of household-level involvement in the cash economy, but as a marker of the degree of individual involvement; the authors thus used mobile phone fees quantitatively as a marker of the degree of involvement in the cash economy. This was done as even within the same household lifestyles became increasingly individualized during the process of economic development. Furthermore, a mobile phone fee is not only a marker of success in the cash economy but also a possible determinant of further diversification in lifestyles, because having a mobile phone might enable people to access information needed for future success.

To elucidate the associations between psychological stress and monetary expenditure on pork or a mobile phone, the authors calculated two composite scores, a total affluence index (TAI) and an investment tendency index (ITI). After the weekly frequency of pork consumption and monthly mobile phone fees were standardized to a mean of 50 with a standard deviation of 10 and log-transformed (named ST-PK, ST-MB, respectively), TAI was defined as the sum of ST-PK and ST-MB, and assumed to represent individual affluence better than either one of the lifestyle parameters alone. ITI was defined as the score after the subtraction of ST-PK from ST-MB, with higher scores indicating an individual tendency to spend more money on the use of a mobile phone than on pork consumption.

Statistical analysis

Two-way analysis of variance (ANOVA) was used to investigate the association between log-transformed EBV antibody titer and sex and region. Least-squares regression analysis was used to investigate the association between log-transformed EBV antibody titer (the dependent variable) and the weekly frequency of pork consumption (dichotomized at the median) and monthly mobile phone fees (also dichotomized at the median). The association between log-transformed EBV antibody titer and ITI was investigated after adjusting for the effect of TAI. Age, sex, body mass index (BMI), and region were included in each model. BMI was included as an adjustment variable in the models, since EBV antibody titer is influenced by the nutritional status of individuals. The statistical analysis was performed by using the statistical software package R version 2.12.2 (R Development Core Team, 2011).


  1. Top of page
  2. Abstract

Descriptive statistics for age, BMI, prevalence of marriage and EBV antibody titer of the participants by region and sex are presented in Table 1. Results of the two-way ANOVA indicated that age did not differ by sex or region, but BMI was significantly different by region (P < 0.001). Participants in Wuzhishan region had a comparatively higher BMI than those in Baoting. The proportion of married participants was much higher among females than males (P < 0.001). Log-transformed EBV antibody titer ranged from 1.31 to 2.41 log EUs with a mean and standard deviation of 1.79 and 0.25 for male participants, and 1.91 and 0.25 for female participants, respectively. This meant that it differed significantly by sex (P < 0.001). It also differed significantly by region (P = 0.041). More specifically, females had higher EBV antibody titer scores than males, while participants in Baoting region had higher EBV antibody titer scores than participants in Wuzhishan region.

Table 1. Characteristics of the study participants by sex and region
 WuzhishanBaotingTwo-way ANOVA
  1. EU: ELISA unit.

  2. a

    Mean and standard deviation are shown.

  3. b

    P-value of Chi-squared test.

 n = 50n = 58n = 55n = 58 P-value 
Age (year)a39.4 ± 11.739.0 ± 11.840.2 ± 11.439.8 ± 11.40.7970.6180.998
BMI (kg/m2)a21.6 ± 2.522.4 ± 2.520.7 ± 2.920.4 ± 3.00.536<0.0010.152
Marriage (%)<0.001b0.154b 
log-EBV (EU)a1.77 ± 0.241.86 ± 0.251.80 ± 0.261.96 ± 0.24<0.0010.0410.264

Information about the respondents' education level, their weekly consumption frequency of the eight selected food items, and monthly mobile phone fees by region and sex is presented in Table 2. More than 38% of the participants had an education level of less than 9 years (equivalent to graduation from middle school in China), while 6.4% had an education level of more than 9 years. Pork was the most commonly consumed food item; the median value of its consumption frequency was 2.5 times a week, while the other seven items were rarely consumed with the median values of at most once a week. These data support the authors' observation that pork consumption was the only food item that varied among local residents. The median value of the monthly mobile phone fee was 20 RMB.

Table 2. Socioeconomic status and lifestyle characteristics of the participants
  1. RMB: renminbi (The official currency of China).

Education (%)
Less than 9 years48.044.622.237.938.1
Equal to 9 years48.051.863.058.655.5
More than 9 years4.03.614.83.46.4
Consumption of foods (times/week)Median (Inter-quartile range)
Pork3 (1–4)2 (1–3)2 (1–3)3 (2–3)2.5 (1–4)
Beef0 (0–0)0 (0–0)0 (0–0)0 (0–0)0 (0–0)
Chicken1 (0.1–2)0.5 (0–1)1 (0–1)1 (0–2)1 (0–1)
Egg0 (0–0)0 (0–1)0 (0–1)0 (0–1)0 (0–1)
Marine fish1 (0.5–4.8)1 (0–3)1 (0.3–2)2 (1–3)1 (0.5–3)
River fish1 (0–2)1 (0–2)1.5 (0–3)1 (0–2)1 (0–2.8)
Blood of animals0 (0–0.1)0 (0–0.2)0 (0–0)0 (0–0)0 (0–0)
Noodles0 (0–1)0 (0–1)0 (0–0)0 (0–1)0 (0–1)
Mobile phone fee (RMB/month)20 (0–50)20 (10–41.9)30 (0–50)0 (0–30)20 (0–49)

Table 3 shows the results from the least-squares regression analysis that was used to investigate the association between log-transformed EBV antibody titer (dependent variable) and age, sex, BMI, and region. A positive association was observed between age and EBV antibody titer (P = 0.041). An association was also seen between sex and EBV antibody titer (where it was higher in females, P < 0.001).

Table 3. Least-squares regression analysis between log-transformed EBV antibody titer and characteristics of the participants
  1. SE: standard error; BMI: body mass index.

  n = 221 
Age (year)0.0030.0010.041
Sex (ref. female)−0.1280.033<0.001
BMI (kg/m2)−0.0020.0060.756
Region (ref. Baoting)−0.0640.0350.066
  R2 = 0.095 

Table 4 shows the results of the least-squares regression analyses that investigated the associations between log-transformed EBV antibody titer (dependent variable) and the weekly frequency of pork consumption (Model 1), monthly mobile phone fees (Model 2), and the TAI and ITI (Model 3) after adjusting for the effects of age, sex, BMI, and region. A positive relation was observed between the frequency of pork consumption and EBV antibody titer (coefficient = 0.081, P = 0.040), while a negative association was found between monthly mobile phone fees and EBV antibody titer (coefficient = −0.078, P = 0.028). The results from the model investigating the association between EBV antibody titer and TAI and ITI revealed that ITI was negatively associated with EBV antibody titer (coefficient = −0.550, P = 0.002), while TAI had no statistically significant association.

Table 4. Least-squares regression analysis between log-transformed EBV antibody titer and lifestyle parameters and the total affluence index and investment tendency index
  1. SE: standard error; BMI: body mass index; TAI: total affluence index; ITI: investment tendency index.

  2. a

    The weekly frequency of pork consumption and monthly mobile phone fee were correlated with each other (Spearman's rho = 0.22, P = 0.003). To avoid statistical instability, these two variables were entered into separate models.

  3. b

    Pork: weekly frequency of pork consumption; mobile phone: monthly mobile phone fee. Participants were categorized into two groups using the median value.

  4. c

    TAI: the sum of the standardized frequency of pork consumption (a mean of 50 and a median of 10) and standardized mobile phone fees (a mean of 50 and a median of 10).

  5. d

    ITI: the value obtained by subtracting the standardized frequency of pork consumption (a mean of 50 and a median of 10) from the standardized mobile phone fee (a mean of 50 and a median of 10).

Model 1a n = 220 
Pork (ref. less than median)b0.0810.0390.040
Age (year)0.0040.0010.013
Sex (ref. female)−0.1360.034<0.001
BMI (kg/m2)−0.0020.0060.759
Region (ref. Baoting)−0.0710.0350.042
  R2 = 0.113 
Model 2a n = 221 
Mobile (ref. less than median)b−0.0780.0350.028
Age (year)0.0020.0010.166
Sex (ref. female)−0.1170.0330.001
BMI (kg/m2)<0.0010.0060.939
Region (ref. Baoting)−0.0650.0340.060
  R2 = 0.115 
Model 3 n = 220 
Age (year)0.0020.0020.166
Sex (ref. female)−0.1100.0340.001
BMI (kg/m2)0.0010.0060.879
Region (ref. Baoting)−0.0640.0340.062
  R2 = 0.135 


  1. Top of page
  2. Abstract

Summary of findings

Among the rural communities in Hainan Island, China, EBV antibody titer was positively associated with the weekly frequency of pork consumption and negatively associated with monthly mobile phone fees. The tendency to use money on a mobile phone rather than for the consumption of pork was negatively associated with EBV antibody titer. These results suggest that psychological stress was higher among those who reported consuming more pork, and lower among those who were more likely to spend money on a mobile phone.

Biological determinants of EBV antibody titer

In this study, a positive association was found between EBV antibody titer and age (P = 0.041). There are few previous studies that have reported a positive association between age and EBV antibody titer. Such an association was noted for Samoan adolescents aged 10–20 years (McDade and Worthman, 2004) and among older Americans (mean age of 72) who had higher EBV antibody titer than their younger controls (mean age of 23) (Glacer et al., 1985) while other studies have not shown an age-dependent association with EBV antibody titer (e.g., McDade et al., 2000b; Sorensen et al., 2009). In this study, we suspect older people experience greater psychological stress due to the fact that they find it more difficult to adapt to the lifestyle changes associated with economic development. For example, Inoue et al. (2012) reported that those with a paid job, usually younger, often exchanged potentially beneficial information while drinking beer at Karaoke venues in towns, whereas those without a paid job, usually older, were left behind in the community and shared traditional home-distilled rice wine without such opportunities.

EBV antibody titer was also higher in females than in males in this study. The finding accords with those from previous studies (Panter-Brick et al., 2008; McDade and Worthman, 2004; McDade et al., 2000a, 2000b). Although the potential mechanism(s) underlying this sex difference has yet to be elucidated, there are two possibilities: (1) there is a sex difference in psychological sensitivity to stressors (i.e., lifestyle changes) and (2) females experience more stressors than males in this population. Since the social importance of family collectivism has confined Chinese women, especially young rural women, to a limited sphere of social activity (Zhang et al., 2004), we suspect that female participants in our study are less able to reach out for help, and are thus more vulnerable to psychological stress.

Psychological stress in a rural Hainan context

As described above, monthly mobile phone fees were negatively associated with psychological stress. In these rural Hainan communities, mobile phone usage can be conceptualized as a marker of integration into the outside cash economy; those who owned mobile phones communicated with people outside their home community, exchanging information on paid jobs or looking for drinking (and thus interaction) opportunities in the towns. A recent study argued that there are three advantages to possessing mobile phones. First, they expand private communication within social networks to maintain the existing forms of one's livelihood and mitigate vulnerability. Second, they enable users to enhance their pre-existing social and economic networks; and third, they enable increasing diversification into new activities and enable individuals to maintain or increase their productivity (Duncombe, 2012). All of these possible functions may contribute to lower psychological stress. Lee and Robbins (1998) showed for example, that psychological stress was lowered by intensifying social connectedness and/or by ameliorating an individual's economic situation. In contrast, those who do not own mobile phones might feel more stress due to their relative lack of opportunities. In rural Hainan, being connected to the world outside their community and because of this, possibly having the opportunity to achieve future success, may have alerted local residents to the importance of possessing this item and it might have therefore become the standard marker of what they aspire to.

In contrast to mobile phone usage, pork consumption was positively associated with psychological stress in present day rural Hainan. This result agrees with previous studies suggesting that dietary change is a possible marker of acculturative stress (e.g., Sorensen et al., 2009). However, considering that pork consumption had been the symbol of a better life in traditional rural Hainan society and also became much more commonplace in the course of economic development, other factors might also be important in terms of explaining its association with increased stress. For example, a greater involvement in cash cropping (as indexed by the number of cash crop trees), which was associated with pork consumption in one of the study communities in Baoting region (Spearman's rho = 0.35, P = 0.006, data not shown), may pose a high level of psychological stress on individuals. Among the 50 households that grew cash crops in the community, 33 households did not obtain any monetary profit. In contrast, three households earned very large monetary returns in comparative terms (more than 20,000 RMB in 2011). Cash cropping requires a certain amount of time and a large financial outlay before harvesting can begin (Jiang et al., 2006). It is possible that the profits generated by the three successful households were making it difficult for the other households to abandon cash cropping, since feelings of anxiety about losing money from their investment may have exacerbated the stressful feelings arising from the non-generation of profit. Such psychological stress experienced by those who engage in cash cropping seems to contrast markedly to that experienced by those who earn money by daily employment as construction workers, drivers or venders outside their communities (Jiang et al., 2006). However, as this interpretation is based on the authors' observations from only one of the study communities, further fieldwork is needed to see if these potential explanations are generalizable.

Certainly, from the perspective of social comparison theory, pork consumption does not seem to serve as a sensitive lifestyle marker by which individuals in the rural areas of Hainan Island compare each other. The goods and services that pork consumers envisioned when comparing their status with that of their reference group might have changed in this modernizing population; just eating pork itself is no longer what they aspire to.

This study showed that those individuals who were more likely to spend money on mobile phones than pork (as indexed by the ITI) had less psychological stress after adjusting for current affluence (as indexed by TAI). It is possible that the ITI corresponds to an axis representing the shift in the standard the residents refer to when comparing themselves with others. Given the fact that a lifestyle with the affluent consumption of pork was traditionally sought and a lifestyle with the use of a mobile phone is the one they are currently seeking, it can be proposed that people seem to be subject to less stress as the sphere of their social activities expands in the course of economic development.

Strength and limitations of this study

A strength of this study is that the authors created a unique analytical framework on the basis of their fieldwork. Specifically, the authors evaluated lifestyle changes from two different dimensions and examined the heterogeneous association between the parameters of modernization and psychological stress. This contrasts with earlier studies that have commonly assumed that the process of modernization is comparatively straightforward and which therefore adopted parameters that were essentially similar in nature (McDade et al., 2000a, 2001; Bindon et al., 1997).

There are also several limitations to this study. First, psychological stress was measured by using only EBV antibody titer. Second, seasonal fluctuation in EBV antibody titer and participants' life histories were not considered. For example, a cohort effect might have existed, influencing the association between age and EBV antibody titer. Third, this study employed quota sampling, focusing on the relationships between the variables; the authors cannot extrapolate the findings to the whole population. Further in-depth ethnographic studies are now needed to investigate what factors determine individuals' adoption of modernization-related lifestyle change and how this affects their well-being.


  1. Top of page
  2. Abstract

The authors express their great appreciation to all participants from Wuzhishan and Baoting regions. The authors also thank Dr. Andrew Stickley of the University of Tokyo for comprehensive and constructive advice on the article.


  1. Top of page
  2. Abstract
  • Adelman I, Morris CT. 1973. Economic growth and social equity in developing countries. Stanford, CA: Stanford University Press.
  • Bindon J, Knight A, Dressler W, Crews D. 1997. Social context and psychosocial influences on blood pressure among American Samoans. Am J Phys Anthropol 103:718.
  • Brindle E, Fujita M, Shofer J, O'Connor KA. 2010. Serum, plasma, and dried blood spot high-sensitivity C-reactive protein enzyme immunoassay for population research. J Immunol Methods 362:112120.
  • Brown DE. 1982. Physiological stress and culture change in a group of Filipino-Americans: a preliminary investigation. Ann Hum Biol 9:553563.
  • Cacioppo JT, Kiecolt-Glaser JK, Malarkey WB, Laskowski BF, Rozlog LA, Poehlmann KM, Burleson MH, Glaser R. 2002. Autonomic and glucocorticoid associations with the steady-state expression of latent Epstein–Barr virus. Horm Behav 42:3241.
  • Cao Y, Zhang Y, Chang DF, Wang G, Zhang X. 2009. Psychosocial and immunological factors in neurasthenia. Psychosomatics 50:2429.
  • Duncombe R. 2012. Understanding mobile phone impact on livelihoods in developing countries: a new research framework. Centre for development policy and management, SED. Development informatics. Working Paper Series. Paper No. 48.
  • Esterling BA, Antoni MH, Schneiderman N, Carver CS, LaPerriere A, Ironson G, Klimas NG, Fletcher MA. 1992. Psychosocial modulation of antibody to Epstein-Barr viral capsid antigen and human herpesvirus type-6 in HIV-1-infected and at-risk gay men. Psychosom Med 54:354371.
  • Festinger L. 1954. A theory of social comparison processes. Human Relations 7:117140.
  • Glaser R, Strain EC, Tarr KL, Holliday JE, Donnerberg RL, Kiecolt-Glaser JK. 1985. Changes in Epstein-Barr virus antibody titers associated with aging.179:352355.
  • Glaser R, Pearson G, Jones J, Hillhouse J, Kennedy S, Mao H, Kiecolt-Glaser J. 1991. Stress-related activation of Epstein-Barr-virus. Brain Behav Immun 5:219232.
  • Glaser R, Rice J, Sheridan J, Fertel R, Stout J, Speicher C, Pinsky D, Kotur M, Post A, Et A. 1987. Stress-related immune suppression health implications. Brain Behav Immunity 1:720.
  • Henle W, Henle G. 1973. Epstein-Barr virus and infectious mononucleosis. N Engl J Med 288:263264.
  • Hyman HH. 1942. The psychology of status. Archives of Psychology (Columbia University) 269:94.
  • Inoue Y, Umezaki M, Watanabe C. 2012. Emergence of income inequality and its impact on subjective quality of life in an ethnic minority community in Hainan Island, China. Anthropol Sci 120:5160.
  • Jenner DA, Harrison GA, Prior IAM, Leonetti DL, Fujimoto WJ, Kabuto M. 1987. Inter-population comparisons of catecholamine excretion. Ann Hum Biol 14:19.
  • Jiang HW, Umezaki M, Ohtsuka R. 2006. Inter-household variation in adoption of cash cropping and its effects on labor and dietary patterns: a study in a Li hamlet in Hainan Island, China. Anthropol Sci 114:165173.
  • Kiecolt-Glaser JK, Fisher LD, Ogrocki P, Stout JC, Speicher CE, Glaser R. 1987a. Marital quality, marital disruption, and immune function. Psychosom Med 49:1334.
  • Kiecolt-Glaser JK, Glaser R, Shuttleworth EC, Dyer CS, Ogrocki P, Speicher CE. 1987b. Chronic stress and immunity in family caregivers of Alzheimer's disease victims. Psychosom Med 49:523535.
  • Lee RM, Robbins SB. 1998. The relationship between social connectedness and anxiety, self-esteem, and social identity. Journal of Counseling Psychology 45:338345.
  • Liang JY, Umezaki M, Ohtsuka R. 2003. Advantageous and disadvantageous impacts of tourism development on the living of Li ethnic minority villagers in Hainan Island, China. J Hum Ergol (Tokyo) 32:17.
  • Libby P, Ridker PM. 1999. Novel inflammatory markers of coronary risk: theory versus practice. Circulation 100:11481150.
  • Lutgendorf SK, Reimer TT, Harvey JH, Marks G, Hong SY, Hillis SL, Lubaroff DM. 2001. Effects of housing relocation on immunocompetence and psychosocial functioning in older adults. J Gerontol A Biol Sci Med Sci 56:M97M105.
  • Macsween KF, Crawford DH. 2003. Epstein-Barr virus—recent advances. The Lancet Infectious Diseases 3:131140.
  • McDade TW, Stallings JF, Worthman CM. 2000a. Culture change and stress in Western Samoan youth: methodological issues in the cross-cultural study of stress and immune function. Am J Hum Biol 12:792802.
  • McDade TW, Stallings JF, Angold A, Costello EJ, Burleson M, Cacioppo JT, Glaser R, Worthman CM. 2000b. Epstein-Barr virus antibodies in whole blood spots: a minimally invasive method for assessing an aspect of cell-mediated immunity. Psychosom Med 62:560567.
  • McDade T. 2001. Lifestyle incongruity, social integration, and immune function in Samoan adolescents. Soc Sci Med 53:13511362.
  • McDade T, Worthman C. 2004. Socialization ambiguity in Samoan adolescents: a model for human development and stress in the context of culture change. J Res Adolesc 14:4972.
  • Monda KL, Gordon-Larsen P, Stevens J, Popkin BM. 2007. China's transition: the effect of rapid urbanization on adult occupational physical activity. Soc Sci Med 64:858870.
  • Mortensen RF. 2001. C-reactive protein, inflammation, and innate immunity. Immunol Res 24:163176.
  • Panter-Brick C, Eggerman M, Mojadidi A, McDade TW. 2008. Social stressors, mental health, and physiological stress in an urban elite of young Afghans in Kabul. Am J Hum Biol 20:627641.
  • Pearson JD, James GD, Brown DE. 1993. Stress and changing lifestyles in the pacific: physiological stress responses of Samoans in rural and urban settings. Am J Hum Biol 5:
  • Popkin BM. 2001. Nutrition in transition: the changing global nutrition challenge. Asia Pac J Clin Nutr 10 Suppl:S13S18.
  • R Development Core Team, 2011. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Available at: Accessed date: February 25, 2011
  • Sorensen MV, Snodgrass JJ, Leonard WR, McDade TW, Tarskaya LA, Ivanov KI, Krivoshapkin VG, Alekseev VP. 2009. Lifestyle incongruity, stress and immune function in indigenous Siberians: the health impacts of rapid social and economic change. Am J Phys Anthropol 138:6269.
  • Statistical Bureau of Hainan Province, 2010. Hainan statistical yearbook. China: Statistics Press.
  • Thorley-Lawson D, Gross A. 2004. Persistence of the Epstein–Barr virus and the origins of associated lymphomas. N Engl J Med 350:13281337.
  • Umezaki M, Jiang H. 2009. Changing adaptive strategies of two li ethnic minority villages in a Mountainous region of Hainan Island, China (<Special Issue> land use changes in the uplands of Southeast Asia: proximate and distant causes). 47:348362.
  • WHO. 1998. WHOQOL User Manual. Available at:
  • Zhang J, Conwell Y, Zhou L, Jiang C. 2004. Culture, risk factors and suicide in rural China: a psychological autopsy case control study. Acta Psychiatr Scand 110: 430437.