AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 153:190–202 (2014)

Maize Dependence or Market Integration? Caries Prevalence Among Indigenous Maya Communities With Maize-Based Versus Globalized Economies Elma Maria Vega Lizama1 and Andrea Cucina2* 1 2

Facultad de Odontologıa, Universidad Autonoma de Yucat an, Merida, Yucat an, Mexico 97000 Facultad de Ciencias Antropologicas, Universidad Autonoma de Yucat an, Merida, Yucat an, Mexico 97305 KEY WORDS

caries; diet; modern Maya; Yucat an

ABSTRACT The relationship between diet and oral health is widely known, yet data on dental caries prevalence is lacking for many indigenous groups with traditional or rapidly modernizing diets. This research documents caries prevalence in two Maya communities from northern Yucat an (Mexico) with significantly different levels of market integration, subsistence, and diet: Yalsihon, with a traditional, maize-based subsistence economy, and Dzilam, with access to globalized food markets. Each sample was subdivided by sex into 15–19, 20– 24, and 25–30 years-of-age classes. Caries prevalence was considered separately both when the lesion affected the enamel superficially (grade 11) and when it reached the dentin (grade 21). In both villages, females of all age classes manifest more caries than males. Results show higher prevalence of caries at Dzilam than at Yalsihon,

except for grade 11 caries among 15–19-year-old males and grade 21 caries among 15–19-year-old females. Though differences are not significant, earlier pregnancies among 15–19-year-old females at Yalsihon could be a causative factor. A survey indicated a more balanced diet at Yalsihon despite a heavier intake of maize than at Dzilam. Striking differences were documented in the ingestion of soda and globalized foods; sodas were virtually absent at Yalsihon, while at Dzilam they were ingested daily in great quantities. The decline in oral health at Dzilam is inferred to result from consumption of industrially processed foods and drinks, while a traditional diet leads to less caries despite daily heavy consumption of maize, which must be considered when interpreting caries rates in archaeological samples. Am J Phys Anthropol 153:190–202, 2014. VC 2013 Wiley Periodicals, Inc.

Oral pathologies, and carious lesions in particular, represent a long-standing health problem that affects human beings’ quality of life at both individual and population levels. The etiology of carious lesions is multifactorial and complex, for caries prevalence is related to such extrinsic factors as diet, lifestyle, daily habits, socio-economic level, as well as intrinsic factors such as oral pH, hormonal level or dental morphology (Larsen et al., 1991; Aufderheide and Rodrıguez-Martın, 1998; Hillson, 2008; Lukacs, 2008, 2011a,b; Cucina et al., 2011). Caries became endemic almost everywhere in the world with the adoption of an agricultural subsistence strategy; therefore it is unsurprising that modern populations, whose subsistence economy is based on agriculture, are affected by high levels of such infections (Turner, 1979). Sugars included in the diet enhance the oral bacteria’s acidification process, with an eventual net mineral loss leading to the initiation and progression of dental caries (Hillson, 2008; Takahashi and Nyvad, 2011). In ancient Mesoamerica, analyses of skeletal collections have yielded caries frequencies consistent with those of populations of other parts of the world where cariogenic staple crops were introduced, for high rates of carious lesions have been reported in almost every preHispanic site analyzed, with the frequency of teeth affected ranging from between 8% to more than 30% (Magennis, 1999; Reed, 1999; White, 1999; Whittington, 1999; Cucina and Tiesler, 2003; Seidemann and McKillop, 2008; Cucina et al., 2011). Maize, the main and most important staple crop for subsistence in the region, is generally considered the causative factor for the high

frequencies of oral infectious disease, while protein-rich foods, such as marine resources consumed in coastal areas or animal proteins consumed by high-status sectors of the society, are often cited to account for lower frequencies encountered at specific sites (Cucina and Tiesler, 2003; Marquez and Hern andez, 2007). With few exceptions, women were found to have suffered higher caries rates than men. This finding has commonly been explained as a consequence of consuming a diet richer in carbohydrates, thereby implying a differential access to resources tied to socially constructed gender roles. Lukacs (1996, 2008) reported that women suffered from carious lesions more frequently and more severely than males both in the Old World and in the New World due to the decline in health and living conditions soon after the adoption of a production economy based on agriculture (see also Larsen et al., 1991). Together with changes in daily habits and a gender-based division of

Ó 2013 WILEY PERIODICALS, INC.

Grant sponsor: National Council for Science and Technology (CONACyT, Mexico); Grant number: CB-2005-50091. *Correspondence to: Andrea Cucina, PhD, Facultad de Ciencias Antropol ogicas, Universidad Aut onoma de Yucat an, Km. 1 Carretera M erida-Tizimin, M erida, Yucat an, 97305, M exico. E-mail: [email protected] Received 16 March 2013; accepted 14 October 2013 DOI: 10.1002/ajpa.22418 Published online 15 November 2013 in Wiley Online Library (wileyonlinelibrary.com).

CARIES IN MAIZE-BASED AND GLOBALIZED COMMUNITIES labor that exposed women to a more frequent contact with food during the day (Larsen et al., 1991), women’s reproductive biology and increased fertility contributed further to the decline in women’s oral health (Lukacs, 1996, 2008, 2011b; Lukacs and Largaespada, 2006; Lukacs and Thompson, 2008). As in pre-Hispanic and colonial times, maize still represents the basic subsistence staple for the majority of the modern populations in the Maya area. Maize dependence is significant in the native sector of the society, which consumes it daily in different forms during each meal. Here, only recently has the process of globalization granted access to nonlocal foods at relatively affordable prices. Unfortunately, the consumption of processed foods and refined sugars, that complement or replace the traditional, natural diet, has contributed to a decrease in oral health. In this context, the term “natural” in used to refer to foods and drinks that are prepared using natural ingredients without artificial preservatives and undergoing industrial processing. These observations align with studies on dental decay in modern populations, which tend to give priority to children. Such studies indicate that the daily intake of sugar-rich foods and sweets hampers dental health in children, decreasing the number of caries-free individuals as the number of sweets consumed per day increases (Rodrıguez et al., 1989; Angelillo et al., 1998). In contrast with the large number of publications on oral health on archaeologically derived Mayas, little is known about the effects of carbohydrate consumption in modern Maya populations. This is especially true for groups that still rely on an “original” autochthonous production-economy diet in comparison with the modern subsistence patterns that result from a globalized economy. In this respect, and differently from pre-Hispanic times, the easy and cheap access to modern goods also brings the advantage of providing the population with such technological benefits as fluoride-based toothpaste that helps prevent the development of carious lesions. Likewise, little is known about the potentially higher caries risks for women in a society with changing roles between sexes, and where more women are engaged in activities outside the household. For these reasons, the present study focuses on dental caries in modern samples from two different communities located in the northern part of the State of Yucat an (Mexico) (Fig. 1). The first, Yalsihon, is an isolated village structured according to the indigenous Maya society that still retains a traditional production and subsistence economy. The second is a sample of residents of the small town of Dzilam Gonz alez (Dzilam) who have access to modern resources in terms of both subsistence and technological benefits, and where women within extended nuclear families are, little by little, detaching from the traditional relegation to household activities. This study was designed to test the hypothesis that maize is responsible for high rates of caries, which, in theory, should be higher even than those reported for societies that have access to refined foods but who also have access to dental care and are characterized by higher economic status and educational levels.

The socio-economic and nutritional contexts Yalsihon is a small, relatively isolated village located in the northern part of the Yucat an peninsula. Its only access to the outside world is a 40-km long narrow

191

Fig. 1. Location of the communities of Yalsih on and Dzilam in Northern Yucat an.

bumpy lane, which limits the movement of population and goods to and from the village. The population numbered 424 individuals at the time of the study. There are few shops, and the residents’ ability to acquire outside products is limited. From an economic perspective, the population as a whole receives a limited income: some 60% of the population receives less than one minimum monthly wage (1,700 Mexican pesos, which corresponds to about 130 USD). The remainder of the community earns between one and two minimum wages. Importantly, only one member of the family usually receives a salary, and on average such families include five individuals (INEGI, 2000). The diets consumed by the residents of Yalsihon are dominated by maize in the form of tortillas, which are prepared with calcium carbonates and ingested during every meal. Fruits and vegetables are grown in the backyards of the villagers. Proteins are consumed in smaller quantities in the form of locally raised chickens, turkeys, and small-sized wild game. Little or no sodas are ingested as logistic and economic constraints limit access to a beverage considered a luxury among village members. Instead, drinks are prepared with local fruits. Abundant and important are citrus fruits, which are commonly picked from trees and eaten between meals. Social roles for women usually involve household chores. Women usually do not attend school after the primary or secondary level, but marry and start having children at a young age (on average 18.3 years, when the first child is born). The population of Yalsihon receives little medical attention and even less dental treatment. For the latter, the villagers depend on clinical facilities located in Panab a, some 20 miles—as the crow flies—from Yalsihon. American Journal of Physical Anthropology

192

E.M.VEGA LIZAMA AND A. CUCINA

TABLE 1. Number of individuals forming the samples by site, sex, and age, frequency of individuals affected by DMFT, average number of DMFT per individual and standard deviations (S.D.) a) caries considered as present from grade 1 and above, b) caries considered as present from grade 2 and above Yalsih on

Males

Females

Males

Females

a

Dzilam

Age class

15–19

20–24

25–30

15–19

20–24

25–30

N total %a Average S.D. N total %a Average S.D. N total %a Average S.D. N total %a Average S.D.

15 80.0 (12) 3.8 3.5 21 90.5 (19) 4.8 3.2 15 40.0 (6) 1.0 1.5 21 57.1 (12) 1.8 2.2

16 87.5 (14) 4.6 3.2 14 85.7 (12) 5.1 5.3 16 50.0 (8) 1.1 1.4 14 71.4 (10) 1.9 2.2

18 88.9 (16) 5.4 2.6 15 100.0 (15) 5.3 2.5 18 83.3 (15) 2.5 1.8 15 86.7 (13) 2.7 1.8

17 64.7 (11) 3.1 3.2 18 83.3 (15) 5.6 4.5 17 58.8 (10) 1.1 1.2 18 72.2 (13) 1.5 1.3

17 94.1 (16) 6.2 3.8 19 100.0 (19) 9.6 5.6 17 94.1 (16) 3.5 2.2 19 100.0 (19) 5.1 3.6

17 100.0 (17) 11.3 3.4 19 100.0 (19) 11.9 4.5 17 88.2 (15) 6.1 4.0 19 94.7 (18) 6.9 4.2

Frequency of affected individuals. The number in parentheses indicates the number of individuals affected.

By contrast, the small town of Dzilam encompasses more than 5,800 inhabitants and is connected to the urban centers of the Yucat an by a modern network of both public and private transportation. The capital of Yucat an, Merida, is reached in 1 h. Economically, the population receives a higher income than the inhabitants of Yalsih on. Population’s income ranges between one and five minimum daily wages. On average, households are small and include four individuals, at least two of which earn a salary. With respect to diet, proteins, fruits and vegetables are purchased in the many shops around town or in the local markets, as is the variety of highly processed, globalized products like sodas and cheaply produced foods, which today are distributed by modern grocery stores. The higher economic level of Dzilam’s inhabitants gives them easy access to processed and refined nonlocal foods. The town has three medical and three dental studios, plus one public healthcare clinic, which allows easy access to medical and dental care. Unlike Yalsihon, female teenagers tend not to get married young and usually do not start having children until after age 20. Instead, Dzilam girls usually attend high school and some even go on to the university (INEGI, 2000).

MATERIALS AND METHODS In both village contexts the analysis focused on individuals between 15 and 30 years of age, divided by sex into three 5-year classes (15–19; 20–24; and 25–30). To be included into one class, the individual had to have already celebrated his/her 15th, 20th, or 25th birthday. Samples consist of 99 individuals at Yalsihon and 107 individuals at Dzilam (Table 1). In the case of Yalsihon, all but one member of the community within the indicated age range were scored. One individual was excluded from this sample for having spent all his life in Cancun, a famous tourist center located along the Riviera Maya (the Maya Riviera). Having abandoned the traditional Maya economy decades ago, this individual was not representative of the traditional subsistence economy. Because of the lack of dental American Journal of Physical Anthropology

facilities at Yalsihon, the senior author, with the help of colleagues and students from the School of Odontology of the Universidad Autonoma de Yucat an, provided free dental consulting at the village itself. The lack of adequate dental facilities prevented more sophisticated dental treatments, such as fillings and extractions; nevertheless, individuals were given consultations, their teeth were cleaned, and in the most serious cases, individuals were recommended to seek dental care at the official facilities. An analogous age segment of the population—as at Yalsihon—was sampled at Dzilam. To be included in the sampling frame, individuals were not those seeking dental attention, which otherwise would have biased the results. More specifically, the volunteers included in the sample were those between 15 and 30 years of age, who were accompanying relatives and friends to the dental facilities, but who were not seeking treatment themselves. In both cases, participants were informed about the purpose of the analysis, that no personal information would be disclosed at any time, and they then signed an agreement (on file at the School of Odontology, Universidad Autonoma de Yucat an). Caries were scored after the individuals had been informed and agreed that their dental conditions (i.e., the amount of caries) were going to be used for scientific, academic purposes (without disclosing personal and private information). The senior author, who is a certified dentist legally qualified to perform oral inspection, scored each individual. Carious lesions were recorded on a scale from 0 to 4, with zero indicating that the tooth did not present any evidence of decay; grade 1: minor demineralization had just started affecting the enamel surface without penetrating deep into the enamel and without reaching the dentin; grade 2: the carious lesion penetrated the enamel and reached the underlying dentin; grade 3: a deep cavity penetrating the dentin and reaching the pulp chamber, and grade 4: a cavity that had destroyed most of the crown. This grading scheme followed that used by Cucina et al. (2011) in order to facilitate direct comparison with pre-Hispanic Maya samples for future

193

CARIES IN MAIZE-BASED AND GLOBALIZED COMMUNITIES

TABLE 2. Results of statistical pair-wise comparisons by sex and age classes within and between sites, based on data reported in Table 1. a) Chi-Squared test (2x2) calculated on absolute values for the frequency of affected individuals (upper part of the table); b) Student t test for the average DMFT per individual (lower part of the table) DMFT including grade 1 caries Yalsihon vs. Dzilam Males Females Males vs. females Yalsihon Dzilam Yalsihon vs. Dzilam Males Females Males vs. females Yalsihon Dzilam

Age class 2

15–19

20–24

DMFT from grade 2 caries 25–30

15–19

20–24 b

25–30

X P X2 P Age class X2 P X2 P

0.92 0.337 0.44 0.506 15–19 0.80 0.370 1.59 0.207

0.44 0.509 2.89 0.089 20–24 0.02 0.886 1.15 0.283

2.00 0.157 –a – 25–30 1.77 0.183 –a –

1.13 0.288 0.96 0.328 15–19 1.03 0.310 0.70 0.404

6.02 0.142 3.79 0.0517c 20–24 1.43 0.232 1.15 0.283

0.17 0.679 0.68 0.410 25–30 0.07 0.790 0.50 0.481

Age class t P d.f. t P d.f. Age class t P d.f. t P d.f.

15–19 20.577 0.56 30 0.63 0.528 37 15–19 20.848 0.40 34 21.83 0.075 33

20–24 21.36 0.18 31 2.32 0.020* 31 20–24 20.36 0.71 28 22.10 0.042* 34

25–30 5.75 0.000* 33 5.04 0.000* 32 25–30 0.12 0.90 31 0.44 0.65 34

15–19 20.24 0.805 30 20.526 0.601 37 15–19 21.25 0.217 34 20.88 0.38 33

20–24 3.71 0.000* 31 2.92 0.006* 31 20–24 21.30 0.202 28 21.62 0.11 34

25–30 3.46 0.001* 33 3.578 0.001* 32 25–30 20.37 0.712 31 20.56 0.57 34

a

In both cases, one of the values equals zero, co Chi-square cannot be calculated. Yates’ Correction applied. c Yates’ Correction applied. Difference would be significant (0.0129) without Yates’ correction. *Asterisk indicates statistically significant difference at p50.05 level. b

comparative analyses. Grade 1 corresponds to D2 degree of cavitation that is limited to the enamel, while grade 2 corresponds to the D3 degree (caries into dentin), according to the World Health Organization criteria (1997) as modified by Burton and Eklund (1999). Caries were considered present in two different ways: from grade 1 and above, and from grade 2 and above. In both cases, missing and filled teeth were scored as positive for carious lesions. The former procedure yields data comparable with the frequencies reported in the literature on modern individuals, which take into consideration grade 1 caries, while the latter yields data with frequencies comparable to those obtained from archaeologically derived samples (Cucina et al., 2011), which consider a caries as present only when it has reached the dentin (grade 2 caries). Third molars were not counted in the younger individuals since they were not fully erupted. Finally, all individuals were interviewed about their daily (or weekly) diet, and related consumption of sodas, proteins and vegetables, and other foodstuffs. Statistical analyses were performed using StatSoft STATISTICA 7.0 software. Pairwise comparison by age class between sexes and village samples for quantitative values (mean number of teeth affected as expressed by DMFT—decayed, missing and filled teeth) were examined using the Student t-test after the Kolmogorov– Smirnov test confirmed normality, with significance level set at P 5 0.05. Pairwise comparisons by age class between sexes and village samples were analyzed for absolute number of DMFT out of the total number of teeth available for each group, using a 2 3 2 Chi-square test. The STATISTICA software automatically provides

results for Chi-square tests with and without Yates’ Correction for Continuity. Yates’ Corrections were applied only for those cases in which probability fell beyond the 0.05 threshold, and when at least one of the expected values was equal or lower than five (Sokal and Rohlf, 1981).

RESULTS Sex differences within villages Table 1 presents the size of each sample by sex and age, the frequency of individuals affected by carious lesions (calculated as DMFT) and the average number of DMFT per individual by sex and age. Table 2 presents the results of the statistical analyses applied to data in Table 1. As expected, caries rates usually increase with age in both village samples and by sex. Males always present lower frequencies of individuals affected by carious lesions than females by age class. The only exception is 20–24-year-old males at Yalsihon who present a frequency of 87.5% (14/16) versus 85.7% (12/14) among their female counterparts when caries grade 11 are considered positive manifestations (v2 5 0.02, 1 df, P 5 0.8856) (Table 2). Similarly, the average number of affected teeth per individual by age class and sex indicates a lower prevalence in males in all age classes with, again, a single exception: in this case, 25–30-year-old males at Yalsihon for caries grade 11 (t 5 0.12, 31 df, P 5 0.902) (Table 2). The frequency of affected teeth by age and sex cohorts is shown in Table 3 and is represented graphically in Figure 2, while Table 4 shows the results from statistical American Journal of Physical Anthropology

194

E.M.VEGA LIZAMA AND A. CUCINA

TABLE 3. Total number of teeth (including missing teeth), number of decayed, missing and filled teeth, and frequency of DMFT teeth by site, sex and age classes. a) caries considered as present from grade 1 and above (upper part of the table); b) caries considered as present from grade 2 and above (lower part of the table) Yalsih on a

Males

Females

Males

Females

Dzilam

Age class

15–19

20–24

25–30

15–19

20–24

25–30

N total Decayed Missing Filled Total DMFT % N total Decayed Missing Filled Total DMFT %

420 52 0 5 57 13.6 588 85 2 13 100 17.0

448 69 1 3 73 16.3 392 53 2 17 72 18.4

504 81 6 11 98 19.4 420 64 3 13 80 19.1

476 49 0 4 53 11.1 504 98 1 1 100 19.8

476 94 2 10 106 22.3 532 164 14 5 183 34.4

476 170 11 11 192 40.3 532 198 17 11 226 42.5

N totalb Decayed Missing Filled Total DMFT % N total Decayed Missing Filled Total DMFT %

420 10 0 5 15 3.6 588 23 2 13 38 6.5

448 13 1 3 17 3.8 392 8 2 17 27 6.9

504 28 6 11 45 8.9 420 25 3 13 41 9.8

476 15 0 4 19 4.0 504 25 1 1 27 5.4

476 47 2 10 59 12.4 532 78 14 5 97 18.2

476 82 11 11 104 21.8 532 103 17 11 131 24.6

a

Values are rounded to the nearest decimal point. For any comparison with archaeological contexts, which takes into consideration the number of teeth available, the number of missing teeth has to be subtracted from the Total Number of teeth.

b

analyses of data in Table 3. Similar to the results obtained in Table 1, the frequency of teeth affected by caries increases with age, with a higher prevalence in females over males in both samples for every age group, except for the 25–30 year age class at Yalsihon for grade 11 caries, where males slightly exceed females (19.4% vs. 19.1%, difference is not significant, see Table 4). Intra-site comparison by sex yields significant differences for grade 11 caries only at Dzilam, in the 15–19 (v2 5 14.09, 1 df, P 5 0.002) and in the 20–24 age classes (v2 5 18.07, 1 df, P 5 0.000). Instead, grade 21 caries show that females have significantly higher frequencies than males at Yalsihon in the 15–19 age group (v2 5 4.11, 1 df, P 5 0.042) and in the 20–24 age group (v2 5 4.03, 1 df, P 5 0.044), and at Dzilam in the 20–24 age class (v2 5 6.56, 1 df, P 5 0.010) (Table 4).

Sex differences between villages An intersite comparison shows that the frequency of affected individuals at Yalsihon is lower than at Dzilam when examined by sex and age class. Once again, we observed an exception in both sexes in the 15–19 age class for caries grade 11; however, pair-wise comparisons indicate that none of the differences between villagers from the two communities are statistically significant (Table 2). In contrast, when the frequency of affected individuals is considered from grade 2 and above (Table 1) a statistically significant difference occurs for the comparison between 20–24-year-old males (Yates corrected v2 5 6.02, 1 df, P 5 0.0142), while for their female counterparts significance is reached without American Journal of Physical Anthropology

Yates’ correction but is slightly above 0.05 when Yates’ correction is applied (v2 5 6.18, 1 df, P 5 0.0129; Yates corrected v2 5 3.79, 1 df, P 5 0. 0517) (Table 2). When the average number of affected teeth per individual is the basis of comparison and grade 11 lesions are considered (Table 1), the only case when residents of Yalsihon present higher values than their counterparts from Dzilam occurs for males in the 15–19 age class (mean 3.8 vs. 3.1, difference is not significant, Table 2). Pairwise comparisons (Table 2) between individuals of the same sex by age class only show statistically significant differences in the number of affected teeth between males in the 25–30 age class (t 5 25.75, 33 df, P 5 0.000), while for females significant differences occur in the 20–24 age class (t 5 2.32, 31 df, P 5 0.020) as well as for the 25–30 age class (t 5 5.04, 32 df, P 5 0.000). When caries grade 2 and above are the basis of comparison (Table 1), the 15–19-year-old females from Yalsihon present higher average values than their age-matched counterparts at Dzilam (1.8 vs. 1.5 teeth affected per individual). However, this difference is not significant. In contrast, pairwise comparisons between females from the two villages result in significant differences in the 20–24 and 25–30 age classes (t 5 22.9, 31 df, P 5 0.006, t 5 23.58, 32 df, P 5 0.001, respectively). All the other comparisons did not reach the 0.05 significance threshold. A similar pattern of statistical significance is observed among males for the 20–24 (t 5 23.71, 31 df, P 5 0.000) and 25–30 age classes (t 5 23.46, 33 df, P 5 0.001) (Table 2). Intersite comparison shows that frequencies are higher among the residents of Dzilam relative to the residents of Yalsihon, with the exception of the 15–19 males

195

CARIES IN MAIZE-BASED AND GLOBALIZED COMMUNITIES

Fig. 2. Frequency of affected teeth by sex and age groups both for grade 11 and for grade21.

TABLE 4. Results of statistical pair-wise comparisons by sex and age classes within and between sites DMFT including grade 1 caries Yalsihon vs. Dzilam

15–19

20–24

25–30

15–19

20–24

25–30

X P X2 P

1.23 0.267 1.46 0.227

5.27 0.021* 29.03 0.000*

51.28 0.000* 59.09 0.000*

0.11 0.742 0.59 0.441

22.61 0.000* 25.0 0.000*

31.70 0.000* 35.02 0.000*

Age class X2 P X2 P

15–19 2.2 0.138 14.09 0.002*

20–24 0.63 0.428 18.07 0.000*

25–30 0.02 0.879 0.48 0.490

15–19 4.11 0.042*a 1.02 0.312

20–24 4.03 0.044*a 6.56 0.010*

25–30 0.19 0.664 1.08 0.298

2

Males Females Males vs. females Yalsihon Dzilam

Age class

DMFT from grade 2 caries

Chi-Squared test (2 3 2) performed on the absolute values presented in Table 2. Difference does not reach the 0.05 significance threshold if Yates’ Correction is applied.

a

when grade 11 caries are considered (13.6% vs. 11.1%, difference is not significant), and females in the same age cohort when caries of grade 2 and above are the basis of comparison (6.5% vs. 5.4%) (Tables 3 and 4). Once again, this difference is not significant. With regard to males, the higher frequencies at Dzilam are significant at the 0.05 level for the 20–24 (v2 5 5.27, 1 df, P 5 0.021) and 25–30 age classes (v2 5 51.28, 1 df, P 5 0.000) for grade 11 caries, as well as for grade 2 caries (20–24: v2 5 22.61, 1 df, P 5 0.000; 25–30: v2 5 31.70, 1 df, P 5 0.000). Identical results are obtained for females in the 20–24 age class (grade 1 caries: v2 5 29.03, 1 df, P 5 0.000; grade 21 caries: v2 5 25.0, 1 df, P 5 0.000) and in the 25–30 age class (grade 1 caries: v2 5 59.09, 1 df, P 5 0.000; grade 21 caries: v2 5 35.02, 1 df, P 5 0.000) (Table 4).

Dietary differences between villages All participants were asked to complete a dietary questionnaire. All the participants at Dzilam responded,

while at Yalsihon, for reasons beyond our control, only 31 of the 99 individuals responded (31.3%). The results revealed a striking dietary difference between residents of the two villages. Vegetables are eaten daily by all members of Yalsihon’s population (Table 5), compared to less than a third of the interviewed sample at Dzilam. While there were no statistically significant differences between sexes in both villages, the difference in daily intake of vegetables is statistically significant (v2 5 38.8, 2 df, P 5 0.000). In contrast, protein consumption in the form of animal flesh is never ingested on a daily basis among the interviewed sample at Yalsihon, while about two-thirds (70.6% among males and 65.4% among females) of those interviewed at Dzilam consume meat daily. As was the case for vegetable consumption, there were no significant differences in meat consumption between the sexes in both Dzilam and Yalsihon, but the difference between residents of the villages is highly significant (v2 5 52.5, 3 df, P 5 0.000) (Table 6). Access to refined sugars and sodas reflects highly different consumption habits between inhabitants of the American Journal of Physical Anthropology

196

E.M.VEGA LIZAMA AND A. CUCINA TABLE 5. Frequency of ingestion of vegetables per individual on a daily/weekly base Daily

Dzilam Males

Females

Males

Females

One to two times a week

Age class

n

%

n

%

n

%

15–19 20–24 25–30 Total 15–19 20–24 25–30 Total

5 8 5 18 6 9 6 21 39

29.4 47.1 29.4 35.3 33.3 47.4 31.6 37.5 36.4

6 6 9 21 6 2 9 17 38

35.3 35.3 52.9 41.2 33.3 10.5 47.4 30.4 35.5

6 3 3 12 6 8 4 18 30

35.3 17.7 17.7 23.5 33.3 42.1 21.1 32.1 28.0

Age class 15–19 20–24 25–30 Total 15–19 20–24 25–30 Total

n 7 3 3 13 12 1 5 18 31

% 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

n

%

n

%

Total Yalsihon

Every other day

Total

Never n

%

n

%

TABLE 6. Frequency of meat consumed per individual on a daily/weekly base Daily Dzilam Males

Females

Males

Females

One to two times a week

Never

Age class

n

%

n

%

n

%

n

%

15–19 20–24 25–30 Total 15–19 20–24 25–30 Total

9 14 13 36 10 8 16 34 70

52.9 82.3 76.4 70.6 55.5 42.1 84.2 60.7 65.4

4 3 1 8 3 6 1 10 18

23.5 17.6 5.8 15.7 16.6 31.5 5.2 17.9 16.8

4 – 3 7 4 4 2 10 17

23.5 – 17.6 13.7 22.2 21.0 10.5 17.9 15.9

1 1 – 2 2

5.56 5.26 – 3.6 1.9

Age class 15–19 20–24 25–30 Total 15–19 20–24 25–30 Total

n

%

n 5 3 3 11 9 1 4 14 25

% 71.4 100.0 100.0 84.6 75.0 100.0 80.0 77.8 80.6

n 2 – – 2 3 – 1 4 6

% 28.5 – – 15.4 25.0 – 20.0 22.2 19.4

n

%

Total Yalsihon

Every other day

Total

two villages (Table 7). More than half of the individuals at Yalsihon do not ingest sweetened foods (61.3%), while the rest only do so once or twice a week. By contrast, at Dzilam the pattern of consumption is more heterogeneous and the difference between residents of the two villages in highly significant (v2 5 68.7, 3 df, P 5 0.000). From a quantitative perspective, those who eat sugary foods at Yalsihon do not exceed 100 g per day. By contrast, less than half (46.7%) of those interviewed at Dzilam consumed moderate quantities of such foods (2 l

Yalsih on Males

Females

15–19 20–24 25–30 Total 15–19 20–24 25–30 Total

Total Amount Males

Females Total

15–19 20–24 25–30 Total 15–19 20–24 25–30 Total

250–500 ml 2 100.0 – – – – 2 100.0 4 57.1 – – 5 100.0 9 69.2 11 73.3

500–1000 ml

3 1 – 4 4

1997; Angelillo et al., 1998; Emerich and AdamowiczKlepalska, 2007; Chankanka et al., 2011; Chatterjee and Bandyopadhyay, 2012). As a result, and differing strongly from bioarchaeological studies, the academic literature on extant populations is currently short of information on the prevalence of carious lesions among adults.

Impacts of globally available processed foods Literature on carious lesions from the advent of agriculture onward identifies carbohydrate consumption as the causative factor behind the dramatic increase in caries prevalence. While this is true, in comparison with the preagricultural Paleolithic diet (Kaidonis et al., 2012), high frequencies of caries in modern or archaeological populations are not and cannot be simplistically associated with social status or carbohydrate intake (Cucina et al., 2011). In a subsistence economy based on maize, such as those practiced by archaeological and modern communities in American Journal of Physical Anthropology

1–2 liters

5 3 3 11 5 – – 5 16

71.4 100.0 100.0 84.6 41.7 – – 27.8 51.6 >2 liters

42.9 100.0 – 30.8 26.7

the Yucat an peninsula, status should confer access to a more varied and protein-rich diet. Indeed, Cucina and Tiesler (2003) reported such findings for the pre-Hispanic Maya royals in Calakmul, even though a subsequent study found that better economic and living conditions at the Classic Maya port of Xcambo (AD 500–750) led to a significant increase in caries (Cucina et al., 2011). The results of the present study of modern populations shows that, despite the better access to resources, higher economic level, and greater access to medical and dental care at Dzilam, carious lesions are of significantly higher prevalence than among members of the traditional Maya community of Yalsihon, who rely heavily upon tortillas made of maize as a dietary staple. This difference has to be considered, not only in terms of carbohydrate consumption, but also within the context of a modern, globalized diet that is high in refined sugars. Despite the high carbohydrate intake at Yalsihon (tortillas are ingested during every meal), dietary intake lacks refined sugars and is rich in nutrients and vitamins. Fruit

199

CARIES IN MAIZE-BASED AND GLOBALIZED COMMUNITIES TABLE 9. Frequency and amount of natural drinks ingested per individual on a daily/weekly base Daily Daily intake

Every other day

Nevera

One to two times a week

Age class

n

%

n

%

n

%

n

%

15–19 20–24 25–30 Total 15–19 20–24 25–30 Total F

2 7 3 12 3 5 4 12 24

11.8 41.2 17.6 23.6 16.7 26.3 21.1 21.4 22.4

2 6 7 15 2 3 8 13 28

11.8 35.3 41.2 29.4 11.1 15.8 42.1 23.2 26.2

2 1 2 5 2 2 3 7 12

11.8 5.9 11.8 9.8 11.1 10.5 15.8 12.5 11.2

11 3 5 19 11 9 4 24 43

64.6 17.6 29.4 37.2 61.1 47.4 21.1 42.9 40.2

250–500 ml 1 16.7 2 14.3 2 16.7 5 15.6 1 14.3 3 30.0 1 6.7 5 15.6 10 15.6

3 11 9 23 3 6 13 22 45

500–1000 ml 50.0 78.6 75.0 71.9 42.9 60.0 86.7 68.8 70.3

2 1 1 4 3 1 1 5 9

Dzilam Males

Females Total Amount Males

Females

15–19 20–24 25–30 Total 15–19 20–24 25–30 Total F

Total

1–2 l

>2 l 33.3 7.1 8.3 12.5 42.9 10.0 6.7 15.6 14.1

Yalsih on Males

Females

15–19 20–24 25–30 Total M 15–19 20–24 25–30 Total F

Total Amount Males

Females Total a

7 3 3 13 12 1 5 18 31

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

250–500 ml 15–19 20–24 25–30 Total M 15–19 20–24 25–30 Total F

3 3 3 9 12 1 5 18 27

500–1000 ml 42.9 100.0 100.0 100.0 100.0 100.0 100.0 87.1

1–2 l

>2 l

4 – – 4

57.1 – –

4

12.9

The table shows only the results of those who ingest natural drinks, and does not include water.

seems to represent an important dietary component: the citric acid it contains helps clean the mouth, irrigates it by stimulating salivation, and protects the oral environment due to its high amount of vitamin C (Nikawa et al., 2008). On the other hand, modern processed foods are highly cariogenic. The escalation in caries prevalence has been reported in association with a variety of factors. Unfortunately, as noted by Marshall et al. (2005), many variables related to the relationship between diet, sugars and oral carious lesions are not clearly specified, which leads to different interpretations. While sugary drinks and snacks are associated with carious lesions (Ismail et al., 1984; Marshall et al., 2003, 2005; Hillson, 2008), authorities disagree over the impact of differences in timing of ingestion of such foods. Ismail et al. (1984) found no difference in the risk of developing carious lesions when considering whether sugary drinks were ingested during or between meals. By contrast, Marshall et al. (2005) found that the ingestion of soft drinks between meals

exposes children to higher risks of acquiring dental caries than when such drinks were consumed as part of regular meals. These researchers also found that the number of times in which sugar-added foods are consumed during the day was related to carious lesions, for when sweetened bread was ingested only once a day it was not associated with an increase in caries, but when the same food was consumed four times a day, there was an increase in the caries rate (Marshall et al., 2005). Quality, quantity, and timing of ingestion of foods are essential factors in the carious lesion prevalence. Protein intake is a daily habit at Dzilam, while it only happens every other day (or less) at Yalsihon. The opposite pattern characterizes the intake of fruits and vegetables. As regards timing of ingestion, snacking involves the intake of foods between meals; Chankanka et al. (2011) found that snacking on industrially processed food (whose access in their sample is related to socioeconomic status) is positively associated with the rise in prevalence of dental caries in school-age children. The same can be claimed American Journal of Physical Anthropology

200

E.M.VEGA LIZAMA AND A. CUCINA

for Dzilam, where individuals tend to ingest snacks and sodas throughout the day. Such consumption patterns stand in sharp contrast to those observed among the residents of Yalsihon, where snacks in between meals are often limited to a tangerine or an orange directly picked from the trees, with very little or no processed foods consumed. Nikawa et al. (2008) concluded that adding citric acid to chewing gum reduces the periodontal pathogen load and increases the salivary flow and concentration, which eventually results in a benefit for oral health. It is likely that the ingestion of citric fruits in between or after meals may represent a mechanical way of cleaning teeth, by providing fibrous materials that brushes the dental and gingival surfaces, coupled with the chemical component that stimulates saliva (Nikawa et al., 2008; PortillaRobertson et al., 2010). Walker et al. (1981) attributed the reduction of caries in European children to the addition of fibrous foods to the diet, instead of the reduction of sweetened foods.

Impacts of sex and gender Very often, the words sex and gender are used synonymously. However, it is important in this context to stress that sex is intended as the biological, intrinsic and genetically determined factor, while gender identifies the extrinsic and culturally constructed factor. The fact that females are marked by higher caries prevalence than males is not surprising (Larsen, 1983; Burt and Ismail, 1986; Larsen et al., 1991; Lukacs, 2008, 2011a,b). At Yalsihon, women tend to engage in regular household activities that expose them to a more frequent contact with food (Larsen et al., 1991). By contrast, at Dzilam, household activities are more irregular and are not as highly gender-specific. In other words, women who live at Dzilam are more often engaged in activities outside the household, like working and studying, than their counterparts at Yalsihon. Nevertheless, the difference between the two contexts indicates that oral health for females in the traditional Maya social structure is much better than in its globalized equivalent, except for the 15–19 age class, where females at Yalsihon slightly exceed their female counterparts from Dzilam. Given the difference in women’s consumption of cariogenic versus cariostatic foods at Yalsihon and Dzilam, with a more balanced, natural and healthy diet in the former, a lower prevalence of caries would be anticipated among 15–19 year-old women at Yalsihon. However, caries are multifactorial in their etiology, so this difference in caries prevalence (when caries grade 21 are considered) between the women of Yalsihon and Dzilam may also be related to fertility. As noted by Lukacs (2008, 2011a,b; Lukacs and Largaespada, 2006; Lukacs and Thompson, 2008) fertility and female hormonal factors are of paramount importance in the increased prevalence of caries. Alongside the progesterone-induced part of the monthly cycle, which drops the woman’s defenses, hormonal changes during pregnancies result in modifications of the salivary pH and flow that seem to have detrimental effects on oral health. In agreement with this concept, the fact that 80% of females at Yalsihon have their first child between 15 and 19 years of age, while the same percentage of women at Dzilam have their first child in the 20–24 years-of-age class, may explain why it is only in the 15–19 year age class that caries prevalence is higher American Journal of Physical Anthropology

among Yalsihon females than their counterparts from Dzilam, even if this difference is minimal. Women at Yalsihon continue to have children after their first one; despite that, their frequency of carious lesions does not rise much in the following age classes. At this stage, after age 20, the pregnancy factor seems to have equalized between communities, for all the women (or at least the majority of them) already have children; yet, caries frequency at Dzilam is three times higher than at Yalsihon in the 20–24 and 25–30 years age classes. After age 20, when hormonal differences can be ruled out as causative or contributing factors, it is likely that consumption of refined foods is the primary causative factor behind the observed higher caries prevalence in the modernized community, relative to the traditional one, in particular that related to consumption of sugary foods or natural drinks. As for the case of 15–19 year-old women, it must be stressed that 15–19 year-old males from Yalsihon, for whom hormonal factors are not an issue in the prevalence of caries, also present caries from grade 11 at a slightly higher prevalence relative to their male counterparts from Dzilam, although in both cases differences are not significant. Therefore, while we still consider pregnancies to be a contributing factor in the higher prevalence of caries in the younger female class at Yalsihon, we cannot rule out the possibility that in both sexes such differences may simply be a consequence of stochastic variation that results from limited sample size.

Potential additional factors At present, we lack information on the amount of fluoride in the water. Water is obtained on a daily base from wells at Yalsihon. By contrast, residents of Dzilam consume water distributed by the City Hall, where no fluoridation program is in effect, and/or they purchase drinking water in the form of filtered, purified water. In a similar fashion, leaving aside protein intake, the survey did not take into consideration the ingestion of other foods with cariostatic properties, such as cheese (Krobicka et al., 1987). However, in this specific case, cheese is not consumed regularly at Yalsihon; therefore, it is unlikely that it contributed to the reduced levels of oral lesions recorded at the village. Finally, the effect of oral hygiene has not been considered directly in this work because, despite its apparent equivalence in the two communities, we are not certain the responses we obtained in the surveys truly reflected the real oral hygiene habits of the population interviewed. Differently from the answers given on quality and quantity of ingested foods, which confirmed what the senior author witnessed personally while attending the participants, in the case of brushing teeth, we suspect the answers were guided by the wish to avoid reproach in the eyes of the senior dentist author, who conducted the survey.

CONCLUSIONS This study suggests that consumption of balanced natural resources is of great benefit for oral and general health. By contrast, a high intake of processed foods does not bring particular health benefits, especially when its consumption is uncontrolled. With regard to sex, particularly females, the picture is complex. Daily habits linked to household activities and pregnancies certainly played a role in the onset of caries, as

CARIES IN MAIZE-BASED AND GLOBALIZED COMMUNITIES witnessed by the differences between sexes in both villages. On the other hand, the striking differences between the villagers of Yalsihon and Dzilam can clearly be attributed to their different dietary patterns. The picture that emerges from this study is that the entwined interaction between extrinsic and intrinsic factors, in which diet is an important component, is responsible for such differences. These results are also helpful for the interpretation of oral health in archaeological collections. Highly refined sugary foods were not common in pre-industrial societies, and maize (and other forms of carbohydrates), in a strongly maize-based subsistence economy, was responsible for the deterioration of oral health. Yet, from a bioarchaeological perspective, inferring status and diet only from the frequency of carious lesions leads to simplistic and biased interpretations. Although it may not be the norm, the case of Xcambo (Cucina et al., 2011) shows that improved social and economic conditions can sometimes lead to a more sedentary lifestyle and to contact with foods that may not have been regularly accessible by the “normal” people.

ACKNOWLEDGMENTS The authors are grateful to John R. Lukacs for the enriching discussion on the topic and for his many worthy comments and advice on the manuscript. They are thankful to Vera Tiesler and Catherine Harrison for proofreading the final version. They are equally indebted to the Editor in Chief, the Associate Editor and two anonymous reviewers for their valuable comments, suggestions and their kind editorial help, which allowed us to clarify and improve the original manuscript. A preliminary version of this article was given as a podium presentation at the 2012 Annual Meeting of the American Association of Physical Anthropology.

LITERATURE CITED Angelillo IF, Anfosso R, Nobile CGA, Pavia M. 1998. Prevalence of dental caries in schoolchildren in Italy. Eur J Epidemiol 14:351–357. Aufderheide AC, Rodrıguez-Martın C. 1998. Dental caries: In: Aufderheide AC, Rodrıguez-Martın C, editors. The Cambridge encyclopedia of human paleopathology. Cambridge: Cambridge University Press. p 402–404. Burt BA, Eklund SA. 1999. Dentistry, dental practice and the community. Philadelphia: WB Saunders Co. Burt BA, Ismail AI. 1986. Diet, nutrition and food cariogenicity. J Dent Res 65:1475–1484. Chankanka O, Marshall TA, Levy SM, Cavanaugh JE, Warren JJ, Broffitt B, Kolker JL. 2011. Mixed dentition cavitated caries incidence and dietary intake frequencies. Pediatr Dent 33:233–240. Chatterjee M, Bandyopadhyay AR. 2012. A study on nutritional status and dental caries in permanent teeth among schoolgoing girls of Bengalee population, India. Adv Anthropol 2: 112–116. Cucina A, Perera Cantillo C, Sierra Sosa T. 2011. Carious lesions and maize consumption among the Prehispanic Mayas: an analysis of a coastal community in northern Yucat an. Am J Phys Anthropol 154:560–567. Cucina A, Tiesler V. 2003. Dental caries and antemortem tooth loss in the Northern Peten area, Mexico: a biocultural perspective on social status differences among the classic Maya. Am J Phys Anthropol 122:1–10. Edelstein BL. 1997. Commentary on childhood caries. New approach needed to reduce caries in children. Pub Health Rep 112:330–331.

201

Emerich K, Adamowicz-Klepalska B. 2007. Dental caries among 7-year-old children in northern Poland. Pub Health Rep 122: 552–558. Hillson S. 2008. The current state of dental decay. In: Irish JD, Nelson GC, editors. Technique and application in dental anthropology. Cambridge: Cambridge University Press. p 111–135. INEGI (Instituto Nacional de Estadıstica y Geografıa, M exico) Survey. 2000. XII Censo general de poblaci on y vivienda 2000. Available at: www.inegi.org.mx. Ismail AI, Burt BA, Eklund SA. 1984. The cariogenicity of soft drinks in the United States. J Am Dent Assoc 109:241–245. Kaidonis J, Ranjitkar S, Townsend G. 2012. Mineral maintenance of dental structures in caries and erosive tooth wear: an holistic model. In: Townsend G, Kanazawa E, Takayama H, editors. New directions in dental anthropology: paradigms, methodologies and outcomes. Adelaide: University of Adelaide Press. p 115–122. Krobicka A, Bowen WH, Pearson S, Young DA. 1987. The effects of cheese snacks on caries in desalivated rats. J Dent Res 66:1116–1119. Larsen CS. 1983. Behavioral implications of temporal change in cariogenesis. J Archaeol Sci 10:1–8. Larsen CS, Shavit R, Griffin MC. 1991. Dental caries evidence for dietary change: an archaeological context. In: Kelley MA, Larsen CS, editors. Advances in dental anthropology. New York: Wiley-Liss. p 179–202. Lukacs JR.1996. Sex differences in dental caries rates with the origin of agriculture in South Asia. Curr Anthropol 37: 147–153. Lukacs JR. 2008. Fertility and agriculture accentuate sex differences in dental caries rates. Curr Anthropol 49:901–914. Lukacs JR. 2011a. Gender differences in oral health in South Asia: metadata imply multifactorial biological and cultural causes. Am J Hum Biol 23:398–411. Lukacs JR. 2011b. Sex differences in dental caries experience: clinical evidence, complex etiology. Clin Oral Invest 15: 649–656. Lukacs JR, Largaespada L. 2006. Explaining sex differences in dental caries rates: saliva, hormones and “life history” etiologies. Am J Hum Biol 18:540–555. Lukacs JR, Thompson LM. 2008. Dental caries prevalence by sex in prehistory: magnitude and meaning. In: Irish JD, Nelson GC, editors. Technique and application in dental anthropology. Cambridge: Cambridge University Press. p 136–177. Magennis AN. 1999. Dietary change of the Lowland Maya site of Kichpanha, Belize. Part II: paleopathology. In: White CD, editor. Reconstructing ancient Maya diet. Salt Lake City: University of Utah Press. p 133–150. M arquez Morfın L, Hern andez PO. 2007. Alimentaci on y salud en algunos pobladores de Jaina, Campeche durante el Cl asico. arquez Morfın L, editors. La poblaci on In: Hern andez PO, M prehispanica de Jaina. Estudio osteobiografico de 106 esqueletos. M exico City: ENAH/INAH. p 111–154. Marshall TA, Broffitt B, Eichenberger-Gilmore JM, Warren JJ, Cunningham MA, Levy SM. 2005. The role of meal, snacks and daily total food and beverages exposure on caries experience in young children. J Pub Health Dent 65:166–173. Marshall TA, Levy SM, Broffitt B, Warren JJ, EichenbergerGilmore JM, Burns TL, Stumbo PJ. 2003. Dental caries and beverage consumption in young children. Pediatrics 112: e184–e191. Nikawa H, Igarashi S, Takasu O, Tataka H, Harano F, Shinohara S, Makihira S, Takemoto T, Murayama T, Satoda T, Amano H, Kurihara H. 2008. Chewing gum containing citric acid reduces burden of periodontal pathogens. Open Food Sci J 2:29–37. Portilla-Robertson J, Pinz on Tofi~ no ME, Huerta Leyva ER, Obreg on Parlange A. 2010. Conceptos actuales e investigaciones futuras en el tratamiento de la caries dental y control de placa bacteriana. Rev Odontol Mexicana 14:218–225.

American Journal of Physical Anthropology

202

E.M.VEGA LIZAMA AND A. CUCINA

Reed DM. 1999. Cuisine from Hun-Nal-Ye. Part III: bone chemistry. In: White CD, editor. Reconstructing ancient Maya diet. Salt Lake City: University of Utah Press. p 183–196. Rodriguez-Contreras Pelayo R, Delgado-Rodriguez M, GalvezVargas R. 1989. Prevalence of dental caries in schoolchildren of the province of Granada (Spain). Eur J Epidemiol 5: 193–198. Seidemann RM, McKillop H. 2008. Dental indicators of diet and health for the Postclassic coastal Maya on Wild Cane Cay, Belize. Anc Mesoam 18:1–11. Sokal RR, Rohlf FJ. 1981. Biometry: the principles and practice of statistics in biological research. New York: W.H. Freeman. Takahashi N, Nyvad B. 2011. The role of bacteria in the caries process: ecological perspectives. J Dent Res 90:294–303.

American Journal of Physical Anthropology

Turner CG II. 1979. Dental anthropological implications of agriculture among the Jomon people of central Japan. Am J Phys Anthropol 51:619–636. Walker ARP, Dison E, Duvenhage A, Walker BF, Friedlander I, Aucamp V. 1981. Dental caries in South African black and white school pupils in relation to sugar intake and snacks habits. Commun Dent Oral Epidemiol 9:37–43. White CD. 1999. Reconstructing ancient Maya diet. Salt Lake City: University of Utah Press. Whittington SL. 1999. Caries and antemortem tooth loss at Cop an. Implications for commoner diet. In: White CD, editor. Reconstructing ancient Maya diet. Salt Lake City: University of Utah Press. p 133–150. World Health Organization. 1997. Oral health surveys. Basic methods, 4th ed. Geneva: World Health Organization.