HHS Public Access

Author manuscript
Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
Author Manuscript

Published in final edited form as:

Trop Med Int Health. 2016 January ; 21(1): 41–51. doi:10.1111/tmi.12622.

Altitude and regional gradients in chronic kidney disease

prevalence in Costa Rica: Data from the Costa Rican Longevity
and Healthy Aging Study
Meera N. Harhay1, Michael O. Harhay2, Fernando Coto-Yglesias3, and Luis Rosero Bixby4,5
1Divisionof Nephrology, Department of Medicine, Drexel University College of Medicine,
Philadelphia, PA, USA
Author Manuscript

2Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, PA,

USA
3Department of Geriatric Medicine, National Geriatrics and Gerontology Hospital, San José,
Costa Rica
4Central American Population Center, University of Costa Rica, San José, Costa Rica
5Department of Demography, University of California, Berkeley, CA, USA

Summary
Objectives—Recent studies in Central America indicate that mortality attributable to chronic
kidney disease (CKD) is rising rapidly. We sought to determine the prevalence and regional
Author Manuscript

variation of CKD and the relationship of biologic and socioeconomic factors to CKD risk in the
older-adult population of Costa Rica.

Methods—We used data from the Costa Rican Longevity and Health Aging Study (CRELES).
The cohort was comprised of 2657 adults born before 1946 in Costa Rica, chosen through a
sampling algorithm to represent the national population of Costa Ricans >60 years of age.
Participants answered questionnaire data and completed laboratory testing. The primary outcome
of this study was CKD, defined as an estimated glomerular filtration rate (eGFR) <60 ml/min/
1.73m2.

Results—The estimated prevalence of CKD for older Costa Ricans was 20% (95% CI 18.5 –
21.9%). In multivariable logistic regression, older age (adjusted odds ratio [aOR] 1.08 per year,
95%CI 1.07–1.10, p<0.001) was independently associated with CKD. For every 200 meters above
Author Manuscript

sea level of residence, subjects’ odds of CKD increased 26% (aOR 1.26 95% CI 1.15–1.38,
p<0.001). There was large regional variation in adjusted CKD prevalence, highest in Limon (40%,
95% CI 30%–50%) and Guanacaste (36%, 95% CI 26–46%) provinces. Regional and altitude
effects remained robust after adjustment for socioeconomic status.

Conclusions—We observed large regional and altitude-related variations in CKD prevalence in

Costa Rica, not explained by the distribution of traditional CKD risk factors. More studies are

Corresponding Author: Meera Nair Harhay, 245 North 15th Street, New College Building, Mail Stop 437, Philadelphia, PA 19102,
USA. Phone +1-215-779-7553, [email protected].
 Harhay et al. Page 2

needed to explore the potential association of geographic and environmental exposures with the
Author Manuscript

risk of CKD.

Keywords
Chronic Kidney Disease; Costa Rica; Epidemiology; Mesoamerican Nephropathy; Tropical
Chronic Disease; Altitude

Introduction
Many developing countries have experienced a surge in chronic disease states, including
chronic kidney disease (CKD).1–3 CKD is a major risk factor for cardiovascular morbidity
and mortality, and among the world’s growing elderly population, CKD is also an
independent risk factor for physical and cognitive dysfunction and poor quality of life.4,5 As
Author Manuscript

adverse outcomes can often be prevented by early identification of CKD, improving

knowledge of population-specific distributions and determinants of CKD is a public health
priority.6,7 At present, the burden and determinants of CKD have been thoroughly
characterized only in high-income countries.8 For example, in the United States (US), there
is an estimated national prevalence of CKD of 11.6%,9 though higher prevalence is observed
in socioeconomically disadvantaged subgroups.10,11 Older age is the strongest predictor of
CKD in US-based studies;12,13 in the US, the prevalence of CKD in individuals over the age
of 60 years is 20%, and it is 35% in those over 70 years.9 However, little is known on the
epidemiology of CKD in low- and middle-income countries, where populations are also
aging but there are fewer resources to handle the sequelae of CKD including high healthcare
costs, disability, morbidity, and mortality. Also, there are likely determinants of CKD that
are distinct in developing countries. For example, in sub-Saharan Africa, higher rates of
communicable disease (e.g., human immunodeficiency virus)14 and genetic variation have
Author Manuscript

been postulated to confer vulnerability to developing CKD.15

In Central America, the rates of renal replacement therapy and transplantation for CKD have
steadily increased in recent years, and some countries have reported alarming increases in
mortality rates attributed to CKD.16,17 Costa Rica is a middle-income country in Central
America that is known for its population’s longevity, social development, universally
available healthcare, and decline in mortality from communicable disease.18 However, a
study of Costa Rican vital statistics revealed that age-standardized mortality rates from CKD
doubled among men and quadrupled among women from 1970 to 2012.19 Based on 2011
census data, the Costa Rican population is comprised of 4.3 million inhabitants, 7% aged 65
years and older, with 24% residing in the three lowland and primarily agricultural provinces
of Guanacaste and Puntarenas on the Pacific coast and Limón on the Caribbean coast
Author Manuscript

(Figure 1). The remaining population resides in four more urbanized central provinces,
namely San Jose, Alajuela, Cartago and Heredia, with the majority of the population living
in the highlands of the Central Valley, including the metropolitan area of San Jose.20

Previous work has highlighted the increasing prevalence of traditional CKD risk factors in
Costa Rica, including an aging population,21 and increased prevalence of hypertension22 and
diabetes.23 A recently described nephropathy called CKDnT (CKD of nontraditional

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 3

causes)24,25 has been described at epidemic levels among young Costa Ricans and other
Author Manuscript

Central Americans without traditional CKD risk factors, and may partially explain the rising
mortality from CKD in the region. Hypotheses for the causative factor of CKDnT include
toxic exposures (e.g., traditional medicines, pesticides),26 geographic factors (e.g., heat and
altitude effects), and harsh working conditions of agricultural workers.14,27–30

As the population worldwide is aging, there is a need to identify potential risk factors for
CKD in older populations in developing countries like Costa Rica, including geographic,
socioeconomic and educational differences among citizens that drive health behaviors32 and
disparate access to medical care.33 Therefore, using nationally representative data from the
Costa Rican Study of Longevity and Healthy Aging (CRELES), our objectives were to
determine: (1) the prevalence levels of CKD in the older-adult population of Costa Rica, (2)
the relationship of biologic and socioeconomic factors to CKD risk, and (3) the regional
variation in CKD prevalence in Costa Rica.
Author Manuscript

Methods
Data Source and Study Participants
Our data source was the publicly available database of the Costa Rican Study of Longevity
and Healthy Aging (CRELES).34 CRELES is a longitudinal study of a nationally
representative sample of 2827 Costa Rican adults born in or prior to 1946. We used data
from the first two waves of surveys, which took place mostly in 2005 and 2007. The first
survey wave of CRELES collected data on 2827 individuals. The sampling strategy for the
study, which included oversampling of older ages, has been described elsewhere.35
Approximately 5% of participants declined to provide the blood sample in each wave. The
final sample in the first survey wave comprised 2657 individuals (94%) who had laboratory
Author Manuscript

data collected on renal function.

Participants completed in-depth interviews on self-reported health status, functional status,

health care utilization and life-course socioeconomic conditions. All data and specimens in
the study were collected at the participants’ homes, usually during two visits, using handheld
computers. During the first visit, participants answered a 90-minute questionnaire (including
some mobility tests and two blood-pressure measures) and a 10-minute food frequency
questionnaire. The interviewer also checked and recorded all medication prescribed by a
health care professional. During a second visit early the next day, fasting blood samples
were collected by venipuncture: 1 EDTA purple top tube (for 3–4 ml. of whole blood) and 2
serum-separating tubes (SST) with a clot activator (for 10–12 ml. of blood, to obtain 4–6 ml.
of serum). Specimens were analyzed in several laboratories (three laboratories for serum
Author Manuscript

creatinine) according to the methods described elsewhere, which include linear adjustments
to have comparable measurements across laboratories.36 Serum creatinine measurements
were calibrated using the Roche method for Isotope Dilution-Mass Spectometry.37 The
institutional review board of the University of Costa Rica granted human subjects approval
to CRELES and all participants granted their informed consent by means of their signature.

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 4

Definition of CKD
Author Manuscript

We calculated estimated glomerular filtration rate (eGFR) based on the 2009 CKD-EPI
equation.38,39 The CKD-EPI equation utilizes data on participant age, sex, and race, though
race was not relevant in the current analysis. We then categorized eGFR categories per the
2012 Kidney Disease: Improving Global Outcomes (KDIGO) guidelines, as follows: G1)
eGFR≥90 ml/min/1.73m2, G2) eGFR ≥60 and <90 ml/min/1.73m2, G3a) eGFR ≥45 and <60
ml/min/1.73m2, G3b) eGFR ≥30 and <45 ml/min/1.73m2, G4) eGFR< 30 and ≥15 ml/min/
1.73m2, G5) <15 ml/min/1.73m2. As urine albumin data was not available to identify CKD
at higher eGFR levels, only participants with eGFRs in categories G3a, G3b, G4, and G5
were categorized as having CKD.40

Hypothesized Risk Factors

We explored the following putative risk factors for CKD that are unique to Central America
Author Manuscript

and Costa Rican older adults: province of residence19 and altitude of residence above sea
level.30 We also explored the association of socioeconomic status (SES, based on a country-
specific wealth scale) and educational attainment on the risk of CKD. The residential
altitude of the CRELES respondents was an ecological variable representing the mean
altitude of their district of residence, which is the smallest geographic unit in Costa Rica
(i.e., there are 473 districts among the seven Costa Rican provinces).

Known CKD risk factors

We examined the strength of the association between known CKD risk factors to CKD
prevalence in Costa Rica by considering the subjects’ medical history of hypertension and
diabetes. We also adjusted for metrics of poor metabolic control that may lead to CKD
including elevated glycosylated hemoglobin (HbA1C, ≥ 6.5%),41 poorly controlled blood
Author Manuscript

pressure (SBP ≥ 150 mmHg or DBP ≥ 90 mmHg),42 and obesity (defined as having a body
mass index ((BMI) higher than 30 kg/m2). We adjusted for reported medication use for
diabetes, hypertension, or hyperlipidemia. The following variables were also considered
potential confounders in this analysis: sex, age, income, years of education, recent physician
visits, and smoking history.

Statistical analysis
To identify the association of our hypothesized risk factors with CKD among elderly Costa
Ricans, we first explored their univariate associations with CKD. Then, we estimated the
prevalence of CKD by sex, age, altitude and province. Finally, we fit multivariable logistic
regression models for CKD using covariates that were significant (p-value threshold <0.2) at
the univariate level. To identify regions with risk of CKD above that expected by traditional
Author Manuscript

risk factors, we estimated CKD prevalence per region after adjustment for covariates
identified in our multivariable model. Statistical analyses were conducted using STATA MP
version 13.0 (Stata Corporation, 2013). Descriptive statistics included proportions for
categorical variables, and means for continuous variables, with 95% confidence intervals
(CI). All statistical analyses took into account differential sampling weights.

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 5

Sensitivity Analysis
Author Manuscript

As the CKD definition based on KDIGO guidelines includes evidence of consistently low
eGFR in repeat testing after 90 days,40 we performed a sensitivity analysis utilizing repeat
measures of eGFR from the second wave of CRELES, conducted close to two years later.

Results
Bivariate Analysis
The first survey wave of CRELES collected data on 2827 individuals, of whom 2657 (94%)
had laboratory data collected on renal function. Of these individuals, 744 survey participants
had evidence of CKD by eGFR criteria (eGFR<60 ml/min/1.73m2).40 The average age of
the study sample was 76 years (standard error 1.3 years); 45% were male and 20% were
diabetic. Subjects with CKD were more likely to be older, female, diabetic, and hypertensive
Author Manuscript

(Table 1). Subjects with CKD also had lower educational attainment, higher systolic blood
pressure, and were more likely to be taking medications for diabetes, hypertension, and
hyperlipidemia. A greater proportion of subjects with CKD had recently seen a physician
and lived at higher altitude. There were also differences in the proportion of subjects with
CKD between provinces. There were no significant (at the p<0.05 level) bivariate
associations between CKD and income or wealth status.

National CKD Prevalence

The estimated prevalence of CKD for older Costa Ricans was 20% (95% CI: 19–22%); 23%
(95% CI: 21–26%) among females and 17% (95% CI: 14–19%) among males. CKD
prevalence increased in a nearly linear fashion with age, from about 15% at age 60 years to
greater than 50% at age 90 years (Figure 2). When further categorized by stage of CKD,40
Author Manuscript

prevalence of stage 3a was 14% (95% CI: 12–16%) among males and 20% (95% CI: 17–
22%) among females, whereas stages 3b-5 CKD were present in 6% (95% CI: 4–7%) of
males and 6% (95% CI: 5–7%) of females. Women exhibited a higher CKD prevalence than
men at all ages, although among individuals over the age of 80 years, the difference between
sexes was no longer statistically significant.

Etiologic Model of CKD

The results of our multivariable modeling approach for CKD among elderly Costa Ricans
are reported in Table 2. In the first model, we examined the association of province and
altitude of residence, adjusted for age and gender, on the risk of CKD. In this model,
subjects of older age (aOR 1.08 per year, 95% CI: 1.07–1.10, p<0.001) and female subjects
(aOR 1.44, 95% CI: 1.13–1.83, p=0.003) were more likely to have CKD. Adjusted for age
Author Manuscript

and gender, living at higher altitudes was independently associated with higher risk of CKD
(aOR 1.28 for every 200 meter increase above sea level, 95% CI: 1.13–1.83, p<0.001). Also,
compared to the capital province of San Jose, residence in Guanacaste (aOR 3.43, 95% CI:
1.84–6.38, p<0.001), Puntarenas (aOR 1.99, 95% CI: 1.14–3.48, p=0.02), or Limon (aOR
4.84, 95% CI: 2.56–9.15, p<0.001) was independently associated with higher risk of CKD.
Adding socioeconomic status (SES) and comorbidity risk factors to the multivariable model
did not substantively change our findings of altitude and regional effects, but did attenuate

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 6

the effect of female gender on CKD risk. In the fully adjusted model, other risk factors for
Author Manuscript

CKD included having poorly controlled blood pressure (aOR 1.60, 95% CI: 1.10–2.32,
p=0.01), taking diabetes medication (aOR 1.73, 95% CI: 1.04–2.87, p=0.04), hypertension
medication (aOR 1.52, 95% CI: 1.06–2.18, p=0.02), and lipid-lowering medication (aOR
1.48, 95% CI: 1.10–1.99, p=0.01). The adjusted CKD prevalence increased monotonically
from 10.4% (95% CI: 6.8–13.9%) for regions below 200 meters to 26.5% (95% CI: 18.2–
34.1%) for regions at 1400–1799 meters above sea level (Figure 3). There were not enough
observations of subjects living at altitudes higher than 1800 meters (N=23) to produce stable
estimates in this model.

Regional Variation in CKD Prevalence

Estimates of CKD prevalence by province differed substantially depending on whether
altitude effects were included (Table 3, Figure 1). Adjusting for age and gender, the lowest
Author Manuscript

CKD prevalence rates were observed in Alajuela and Puntarenas (14% and 15%,
respectively), while the highest prevalence rates were observed in Cartago and Limon (28%
and 25%, respectively). However, after further adjusting for the effect of altitude, San Jose,
Alajuela, and Heredia had the lowest CKD prevalence (16% for all), and Guanacaste and
Limon had the highest (37% and 42%, respectively).

Sensitivity Analysis
Appendix 1 reports the results of a sensitivity analysis incorporating available longitudinal
data on subjects, collected from the second survey wave of CRELES (conducted 2 years
after the first wave). 77% of subjects had repeat eGFR measures in the second survey wave;
the resulting analytical sample for the sensitivity analysis contained 4,740 observations. We
found that the strength of the associations observed between age, region, altitude, and CKD
Author Manuscript

risk remained consistent and statistically significant, although the observed elevated risk of
CKD in Limon and Guanacaste provinces was lower (aOR of 2.72 vs. 4.53 and 2.66 vs.
3.65, respectively). The use of the two waves of survey data resulted in slightly higher
estimates of national CKD prevalence: 26% (95% CI: 24–28%) among all older adults, 23%
(95% CI: 20–25%) among older males and 29% (95% CI: 27–34%) among older females.

Discussion
The prevalence of CKD is increasing worldwide, likely driven by an aging population.8
Worldwide death rates attributable to CKD have also risen dramatically, with a median
increase of 36.9% from 1990 to 2013.43 In addition to heightened mortality risk, older adults
with CKD are also more likely to suffer disability from physical and cognitive dysfunction.5
However, determinants of CKD are poorly understood in the aging population of developing
Author Manuscript

countries, including those in Central America. In this study, we explored traditional and
novel risk factors for CKD in a nationally representative sample of elderly Costa Ricans. We
have identified that the prevalence of CKD among older Costa Ricans is 20%, similar to the
CKD prevalence previously described among older adults in the US.9 Our study also found
that the prevalence of CKD among older Costa Rican women is higher than among older
Costa Rican men. Previous studies have highlighted that women in Costa Rica have higher
burdens of hypertension, diabetes, and obesity.44 Consistent with these reports, our study

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 7

showed that after adjustment for comorbidities and SES, the association of female gender
Author Manuscript

and CKD was no longer statistically significant. Also, while wealth and education disparities
are also known to drive differential risk for chronic disease states including
CKD,10,16,32,33,45 our study did not find wealth or education gradients with CKD among
elderly Costa Ricans. This may be explained by the fact that while most of the subjects had
low educational attainment, the vast majority also had recent access to medical care. Indeed,
nephrologic clinical services have been freely accessible in Costa Rica since 1968, as part of
the national social security system that ensures that citizens have free access to medical
attention as needed.46 Therefore, the lack of socioeconomic gradients in CKD may represent
a success in Costa Rica’s adoption of widespread health care coverage.18

Our study also highlighted a potential relationship between altitude of residence and risk of
CKD. The association between altitude and CKD risk remained robust after adjustment for
age, comorbidities, region, wealth, education, and access to recent medical care. Living at
Author Manuscript

high altitudes is a known risk factor for high altitude renal syndrome (HARS), a syndrome
resulting from chronic hypoxia that is characterized by polycy-themia, hyperuricemia,
hypertension, and proteinuria.30,47,48 However, this syndrome has generally been described
at altitudes hgher than 2400 meters above sea level, and most Costa Ricans live either at sea
level or in the Central Valley, about 1,200 meters above sea level. It is possible that the
direct renal injury or reduction in renal plasma flow observed in HARS is present to a lesser
degree at the highest altitudes in Costa Rica, and that older adults are particularly susceptible
to these exposures. However, further studies are needed to understand possible factors
related to higher altitude that may explain the patterns observed in this study.

This study also identified Guanacaste and Limon provinces as the regions in Costa Rica with
the highest prevalence rates of CKD. A recent study of Costa Rican death records over the
Author Manuscript

last 40 years noted that Guanacaste province had the most marked rise in age-adjusted CKD
mortality.19 These findings in aggregate warrant future studies to further investigate risk
factors specific to these regions, including factors that may distinguish Guanacaste and
Limon from Puntarenas, the other low-altitude coastal province in Costa Rica. In terms of
putative differences in environmental exposures, residents of the low-altitude provinces
typically encounter warmer weather than the central provinces (Figure 1) and rely
predominantly on agriculture for commerce than residents of the Central Valley. The
agricultural products of these regions are distinct: according to the 2014 agricultural census,
the main crops in Guanacaste are beef, sugar cane, and rice, whereas in Limon, bananas and
other fruits are the predominant export items.31 Guanacaste and Limon have also attracted
the highest proportions of immigrants from other Central American countries, including
Nicaragua, where the epidemic of CKDnT has been extensively documented.29,49,50 While
Author Manuscript

the causes of CKDnT are unknown,50 hypotheses include heat stress caused by year-round
extreme high temperatures and humidity, heavy agricultural work and toxic pesticides.
These factors are thought to predispose certain individuals to recurrent volume depletion,
resulting in activation of both intra- and extra-renal pathways that drive renal injury.52
However, while CKDnT have been characterized as a tubulointerstitial nephropathy with
low-grade proteinuria,24,51 previous reports of kidney biopsies performed among Costa
Ricans with CKD have identified focal and segmental glomeruloscerosis, membranous
nephropathy, mesangial proliferative glomerulonephritis and crescent glomerulonephritis as

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 8

the most common findings.46 Future studies, ideally including data on environmental
Author Manuscript

exposures and kidney biopsies, will be important to elucidate the nature of CKD found
among populations living in different regions of Costa Rica.

Study limitations
Our study has several limitations that must be acknowledged. As some of our data is cross-
sectional, we cannot make inferences on causality. Also, while we attempted to remain as
consistent as possible to the 2012 Clinical Practice Guidelines from the KDIGO working
group recommendations on the diagnosis of CKD,40 we did not have data on proteinuria that
is necessary to characterize early CKD (e.g., stage 2, eGFR between 60 and 90 ml/min).
Therefore, we chose to categorize subjects as having CKD only at later stages (i.e., 3a and
beyond), as proteinuria is not a necessary component for diagnosis at later stages of renal
dysfunction, and the CKD-EPI formula is more precise.40 Additionally, there is an active
Author Manuscript

debate on whether stage 3a CKD represents a pathologic condition or a consequence of

natural aging among the elderly.53 However, in one global study of over two million
participants, Hallan and colleagues found that while older age attenuated the risk of
mortality with declining eGFR, the relative risk of mortality was significantly higher, even
among the oldest participants (age>75 years), when eGFR was 56 ml/min/1.73m2 or
lower.54

The health consequences of CKD for older adults include higher risk of morbidity and
mortality, even at higher levels of eGFR.5,12 Therefore we chose to include mild reductions
in eGFR (corresponding to CKD stage 3a) in our estimates. CRELES investigators did not
perform serial measurements of eGFR during the first wave of surveys. Therefore, as the
classification of CKD requires two eGFR measurements at least 90 days apart,40 we
completed a sensitivity analysis including data from the second survey wave, completed two
Author Manuscript

years after the initial wave. The resulting longitudinal random effects models confirmed the
significant associations we observed between province, altitude, and CKD in Costa Rica.

Conclusion
In Central American and other developing countries, where data are most scarce on the
epidemiology of CKD, studies are urgently needed to explain the rising prevalence of CKD
and precipitous rise in CKD-related deaths. While we did not identify socioeconomic
disparities in CKD risk among elderly Costa Ricans, our findings of potential geographic
and regional determinants of CKD in Costa Rica provide a platform for further
investigation.
Author Manuscript

Acknowledgments
The CRELES study was funded by grants from the Wellcome Trust; the National Heart, Lung, And Blood Institute
and the National Institute of Diabetes and Digestive and Kidney Diseases of the United States National Institutes of
Health. The content is solely the responsibility of the authors and does not necessarily represent the official views
of the National Institutes of Health or Wellcome Trust.

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 9

References
Author Manuscript

1. Meguid El Nahas A, Bello AK. Chronic kidney disease: the global challenge. Lancet. 2005;
365(9456):331–340. [PubMed: 15664230]
2. Correa-Rotter R, Naicker S, Katz IJ, et al. Demographic and epidemiologic transition in the
developing world: role of albuminuria in the early diagnosis and prevention of renal and
cardiovascular disease. Kidney international. Supplement. 2004; (92):S32–37. [PubMed: 15485413]
3. Eckardt KU, Coresh J, Devuyst O, et al. Evolving importance of kidney disease: from subspecialty
to global health burden. Lancet. 2013; 382(9887):158–169. [PubMed: 23727165]
4. Murray AM. Cognitive impairment in the aging dialysis and chronic kidney disease populations: an
occult burden. Adv Chronic Kidney Dis. 2008; 15(2):123–132. [PubMed: 18334236]
5. Maw TT, Fried L. Chronic kidney disease in the elderly. Clin Geriatr Med. 2013; 29(3):611–624.
[PubMed: 23849011]
6. Bello AK, Nwankwo E, El Nahas AM. Prevention of chronic kidney disease: a global challenge.
Kidney international. Supplement. 2005; (98):S11–17. [PubMed: 16108964]
7. Nwankwo E, Bello AK, El Nahas AM. Chronic kidney disease: stemming the global tide. American
Author Manuscript

journal of kidney diseases : the official journal of the National Kidney Foundation. 2005; 45(1):
201–208. [PubMed: 15696462]
8. Jha V, Garcia-Garcia G, Iseki K, et al. Chronic kidney disease: global dimension and perspectives.
Lancet. 2013; 382(9888):260–272. [PubMed: 23727169]
9. Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidney disease in the United States.
Jama. 2007; 298(17):2038–2047. [PubMed: 17986697]
10. Vart P, Gansevoort RT, Coresh J, Reijneveld SA, Bultmann U. Socioeconomic measures and CKD
in the United States and The Netherlands. Clinical journal of the American Society of
Nephrology : CJASN. 2013; 8(10):1685–1693. [PubMed: 23813554]
11. Garcia-Garcia G, Jha V. Chronic kidney disease in disadvantaged populations. Pediatr Nephrol.
2015; 30(2):183–187. [PubMed: 25395360]
12. Prakash S, O’Hare AM. Interaction of aging and chronic kidney disease. Seminars in nephrology.
2009; 29(5):497–503. [PubMed: 19751895]
13. Anderson S, Halter JB, Hazzard WR, et al. Prediction, progression, and outcomes of chronic
Author Manuscript

kidney disease in older adults. Journal of the American Society of Nephrology : JASN. 2009;
20(6):1199–1209. [PubMed: 19470680]
14. Stanifer JW, Jing B, Tolan S, et al. The epidemiology of chronic kidney disease in sub-Saharan
Africa: a systematic review and meta-analysis. The Lancet. Global health. 2014; 2(3):e174–181.
[PubMed: 25102850]
15. Ojo A. Addressing the global burden of chronic kidney disease through clinical and translational
research. Transactions of the American Clinical and Climatological Association. 2014; 125:229–
243. discussion 243–226. [PubMed: 25125737]
16. Cusumano AM, Garcia-Garcia G, Gonzalez-Bedat MC, et al. Latin American Dialysis and
Transplant Registry: 2008 prevalence and incidence of end-stage renal disease and correlation with
socioeconomic indexes. Kidney international supplements. 2013; 3(2):153–156. [PubMed:
25018980]
17. Rosa-Diez G, Gonzalez-Bedat M, Pecoits-Filho R, et al. Renal replacement therapy in Latin
American end-stage renal disease. Clinical kidney journal. 2014; 7(4):431–436. [PubMed:
25349696]
Author Manuscript

18. Rosero-Bixby L. Socioeconomic development, health interventions and mortality decline in Costa
Rica. Scandinavian journal of social medicine. Supplementum. 1991; 46:33–42. [PubMed:
1805367]
19. Wesseling C, van Wendel de Joode B, Crowe J, Rittner R, Jakobsson K. 0204 Mesoamerican
nephropathy in Costa Rica: Geographical distribution and time trends of chronic kidney disease
mortality between 1970 and 2012. Occupational and environmental medicine. 2014; 71(Suppl
1):A27.
20. Augelli JP. Costa Rica’s Frontier Legacy. Georgraphical Review. 1987; 77(1)

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 10

21. Fernández-Rojas, X.; Robles-Soto, A. I Informe estado de situación de la persona adulta mayor en
Costa Rica. San José, Costa Rica: Universidad de Costa Rica; 2008.
Author Manuscript

22. Mendez-Chacon E, Santamaria-Ulloa C, Rosero-Bixby L. Factors associated with hypertension

prevalence, unawareness and treatment among Costa Rican elderly. BMC public health. 2008;
8:275. [PubMed: 18681969]
23. Barcelo A, Rajpathak S. Incidence and prevalence of diabetes mellitus in the Americas. Revista
panamericana de salud publica = Pan American journal of public health. 2001; 10(5):300–308.
[PubMed: 11774801]
24. Correa-Rotter R, Wesseling C, Johnson RJ. CKD of unknown origin in Central America: the case
for a Mesoamerican nephropathy. American journal of kidney diseases : the official journal of the
National Kidney Foundation. 2014; 63(3):506–520. [PubMed: 24412050]
25. Wesseling C, Crowe J, Hogstedt C, Jakobsson K, Lucas R, Wegman DH. The epidemic of chronic
kidney disease of unknown etiology in Mesoamerica: a call for interdisciplinary research and
action. American journal of public health. 2013; 103(11):1927–1930. [PubMed: 24028232]
26. Ordunez P, Silva LC. Pesticides and the epidemic of CKD in Central America. American journal
of kidney diseases : the official journal of the National Kidney Foundation. 2014; 64(3):477.
Author Manuscript

[PubMed: 25150854]
27. Crowe J, Wesseling C, Solano BR, et al. Heat exposure in sugarcane harvesters in Costa Rica.
American journal of industrial medicine. 2013; 56(10):1157–1164. [PubMed: 23775893]
28. Peraza S, Wesseling C, Aragon A, et al. Decreased kidney function among agricultural workers in
El Salvador. American journal of kidney diseases : the official journal of the National Kidney
Foundation. 2012; 59(4):531–540. [PubMed: 22300650]
29. Almaguer M, Herrera R, Orantes CM. Chronic kidney disease of unknown etiology in agricultural
communities. MEDICC review. 2014; 16(2):9–15. [PubMed: 24878644]
30. Hurtado A, Escudero E, Pando J, Sharma S, Johnson RJ. Cardiovascular and renal effects of
chronic exposure to high altitude. Nephrology, dialysis, transplantation : official publication of the
European Dialysis and Transplant Association - European Renal Association. 2012; 27(Suppl
4):iv11–16.
31. (INEC) INdEyC. VI Censo Nacional Agropecuario, Resultados Generales. San Jose, Costa Rica:
INEC Publications; 2015.
Author Manuscript

32. Rosero-Bixby L, Dow WH. Surprising SES Gradients in mortality, health, and biomarkers in a
Latin American population of adults. The journals of gerontology. Series B, Psychological
sciences and social sciences. 2009; 64(1):105–117.
33. Brenes-Camacho G, Rosero-Bixby L. Differentials by socioeconomic status and institutional
characteristics in preventive service utilization by older persons in Costa Rica. Journal of aging
and health. 2009; 21(5):730–758. [PubMed: 19584413]
34. CRELES: Costa Rican Longevity and Healthy Aging Study. (Costa Rica Estudio de Longevidad y
Envejecimiento Saludable) ICPSR26681-v2 2010–07–21 2013. 2005. http://doi.org/10.3886/
ICPSR26681.v210.3886/ICPSR26681.v2
35. Brenes-Camacho G, Rosero-Bixby L. Metabolic control in a nationally representative diabetic
elderly sample in Costa Rica: patients at community health centers vs. patients at other health care
settings. BMC international health and human rights. 2008; 8:5. [PubMed: 18447930]
36. Mendez-Chacon E, Rosero-Bixby L, Fernandez Rojas X, Barrantes Jimenez K. Comparación de
los resultados de pruebas de laboratorio seleccionadas de un estudio poblacional de adultos
mayores de Costa Rica. Población y Salud en Mesoamérica (Revista electrónica). 2007; 5(1)
Author Manuscript

37. Peake M, Whiting M. Measurement of serum creatinine--current status and future goals. The
Clinical biochemist. Reviews / Australian Association of Clinical Biochemists. 2006; 27(4):173–
184. [PubMed: 17581641]
38. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate.
Annals of internal medicine. 2009; 150(9):604–612. [PubMed: 19414839]
39. Gifford FJ, Methven S, Boag DE, Spalding EM, Macgregor MS. Chronic kidney disease
prevalence and secular trends in a UK population: the impact of MDRD and CKD-EPI formulae.
QJM : monthly journal of the Association of Physicians. 2011; 104(12):1045–1053. [PubMed:
21821654]

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 11

40. KCW Group. KDIGO 2012 clinical practice guideline for the evaluation and management of
chronic kidney disease. Kidney international. Supplement. 2013; 3:1–150.
Author Manuscript

41. Sacks DB, Arnold M, Bakris GL, et al. Position statement executive summary: guidelines and
recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus.
Diabetes care. 2011; 34(6):1419–1423. [PubMed: 21617111]
42. James PA, Oparil S, Carter BL, et al. 2014 evidence-based guideline for the management of high
blood pressure in adults: report from the panel members appointed to the Eighth Joint National
Committee (JNC 8). Jama. 2014; 311(5):507–520. [PubMed: 24352797]
43. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240
causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013.
Lancet. 2014
44. Rehkopf DH, Dow WH, Rosero-Bixby L. Differences in the association of cardiovascular risk
factors with education: a comparison of Costa Rica (CRELES) and the USA (NHANES). Journal
of epidemiology and community health. 2010; 64(9):821–828. [PubMed: 19822554]
45. So BH, Methven S, Hair MD, Jardine AG, MacGregor MS. Socio-economic status influences
chronic kidney disease prevalence in primary care: a community-based cross-sectional analysis.
Author Manuscript

Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant
Association - European Renal Association. 2015
46. Cerdas M. Chronic kidney disease in Costa Rica. Kidney international. Supplement. 2005;
(97):S31–33. [PubMed: 16014096]
47. Arestegui AH, Fuquay R, Sirota J, et al. High altitude renal syndrome (HARS). Journal of the
American Society of Nephrology : JASN. 2011; 22(11):1963–1968. [PubMed: 21566053]
48. Luks AM, Johnson RJ, Swenson ER. Chronic kidney disease at high altitude. Journal of the
American Society of Nephrology : JASN. 2008; 19(12):2262–2271. [PubMed: 18842990]
49. O’Donnell JK, Tobey M, Weiner DE, et al. Prevalence of and risk factors for chronic kidney
disease in rural Nicaragua. Nephrology, dialysis, transplantation : official publication of the
European Dialysis and Transplant Association - European Renal Association. 2011; 26(9):2798–
2805.
50. Torres C, Aragon A, Gonzalez M, et al. Decreased kidney function of unknown cause in
Nicaragua: a community-based survey. American journal of kidney diseases : the official journal
of the National Kidney Foundation. 2010; 55(3):485–496. [PubMed: 20116154]
Author Manuscript

51. Lopez-Marin L, Chavez Y, Garcia XA, et al. Histopathology of chronic kidney disease of unknown
etiology in Salvadoran agricultural communities. MEDICC review. 2014; 16(2):49–54. [PubMed:
24878649]
52. Johnson AC, Zager RA. Renal cortical pyruvate as a potentially critical mediator of acute kidney
injury. Nephron. Clinical practice. 2014; 127(1–4):129–132. [PubMed: 25343836]
53. Glassock R, Delanaye P, El Nahas M. An Age-Calibrated Classification of Chronic Kidney
Disease. Jama. 2015; 314(6):559–560. [PubMed: 26023760]
54. Hallan SI, Matsushita K, Sang Y, et al. Age and association of kidney measures with mortality and
end-stage renal disease. Jama. 2012; 308(22):2349–2360. [PubMed: 23111824]
Author Manuscript

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 12
Author Manuscript
Author Manuscript

Figure 1.
Map of Costa Rica with regional prevalence, annual heat exposure, and altitude
Author Manuscript
Author Manuscript

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 13
Author Manuscript
Author Manuscript

Figure 2.
Prevalence of CKD by age and sex (smoothed with locally weighted polynomials)
Author Manuscript
Author Manuscript

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 14
Author Manuscript
Author Manuscript

Figure 3.
Probability of CKD by altitude of residence, adjusted for age, gender, comorbidities, and
Province of residence
Author Manuscript
Author Manuscript

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Harhay et al. Page 15

Table 1

Characteristics of elderly Costa-Ricans by CKD status in Wave 1

Author Manuscript

No CKD CKD (stage 3a-5) p-value

n=1913 n=744

Mean age, years 69.2 [68.8–69.6] 74.6 [73.7–75.4] < 0.001

Percent male (n=1,211) 49.6 [46.9,52.3] 39.4 [34.9,44.1] < 0.001
Percent with reported diabetes (n=517) 20.1 [18.1,22.4] 25.4 [21.5,29.9] 0.02
Percent with reported hypertension (n=1,307) 45.1 [42.4,47.8] 63.1 [58.5,67.5] < 0.0001
Mean years of education 5.3 [5.0–5.5] 4.9 [4.5–5.3] 0.17
Education Category, % 0.02
Less than primary school (n=1,862) 50.1 [47.4,52.8] 56.9 [52.0,61.7]
Primary school and beyond (n=795) 49.9 [47.2,52.6] 43.1 [38.3,48.0]
Mean income (monthly thousands of Colones) 307 [247–367] 261 [203–320] 0.34
Author Manuscript

Percent visited physician in last 12 months (n=2,380) 86.8 [84.7,88.6] 95.2 [92.7,96.9] < 0.0001
Mean wealth proxy / scale of goods 8.2 [8.1–8.3] 8.2 [8.1–8.4] 0.66
Mean altitude of home above sea level, m 795 [767–824] 931 [882–979] < 0.001
Percent rural residence (n=981) 35.3 [32.8,37.9] 32 [27.9,36.3] 0.20
Province of Residence, % < 0.001
San Jose (n=928) 35.7 [33.1,38.3] 36.3 [31.9,40.9]
Alajuela (n=425) 16.6 [14.7,18.7] 10.3 [8.1,13.1]
Cartago (n=413) 13.9 [12.2,15.9] 23 [19.2,27.4]
Heredia (n=188) 8.3 [6.8,10.0] 7.7 [5.3,11.1]
Guanacaste (n=275) 9.5 [8.0,11.2] 8.8 [6.6,11.7]
Puntarenas (n=254) 9.8 [8.3,11.5] 6.3 [4.7,8.5]
Limon (n=174) 6.3 [5.1,7.8] 7.5 [5.6,10.1]
Author Manuscript

Percent current smoker (n=211) 11.2 [9.5,13.2] 5.1 [3.3,7.7] < 0.001
Mean HbA1C, % 5.7 [5.7–5.8] 5.9 [5.8–6.0] 0.008
Percent with HbA1C ≥ 6.5% (n=350) 13.3 [11.6,15.3] 17 [13.6,21.0] 0.07
Mean systolic blood pressure, mmHg 143.4 [142.2–144.6] 147.0 [144.7–149.3] 0.006
Mean diastolic blood pressure, mmHg 83.9 [83.3–84.5] 83.5 [82.2–84.8] 0.49
Percent with measured hypertension (>150/90) (n=1212) 42.2 [39.5,44.9] 50.5 [45.8,55.1] 0.002
Percent BMI > 30 kg/m2 (n=593) 27.4 [24.9,29.9] 26.4 [22.3,30.9] 0.70
Percent taking medication for:
Diabetes (n=424) 16.3 [14.4,18.4] 23.3 [19.4,27.7] 0.002
Hypertension (n=1,210) 39.5 [36.9,42.2] 61.2 [56.5,65.6] < 0.0001
Hyperlipidemia (n=498) 18.8 [16.8,21.0] 30.7 [26.3,35.4] < 0.001

CKD defined as eGFR<60 ml/min/1.73m2

Author Manuscript

Continuous variables are reported as means, and categorical variables as percentages, with 95% confidence intervals in bracket.

The reported sample sizes (n) for categorical variables are the number of survey participants within each category.

Abbreviations: CKD=chronic kidney disease

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 2

Etiologic model of CKD in Costa Rica: regional and traditional risk factors

Model 1 Model 1 + SES & comorbidities

Harhay et al.

aOR 95% CI p-value aOR 95% CI p-value

Age (per 1-year increase) 1.08 [1.07,1.10] <0.001 1.08 [1.07,1.10] <0.001
Female 1.44 [1.13,1.83] 0.003 1.20 [0.93,1.56] 0.16
Altitude (per 200m increase from sea level) 1.28 [1.17,1.40] <0.001 1.26 [1.15,1.38] <0.001

Province of Residence*
Alajuela 1.01 [0.69,1.49] 0.95 1.02 [0.69,1.51] 0.93
Cartago 1.37 [0.97,1.94] 0.07 1.37 [0.96,1.95] 0.09
Heredia 0.93 [0.55,1.58] 0.79 0.84 [0.51,1.41] 0.51
Guanacaste 3.43 [1.84,6.38] <0.001 3.65 [1.91,6.97] <0.001
Puntarenas 1.99 [1.14,3.48] 0.02 2.14 [1.19,3.83] 0.01
Limon 4.84 [2.56,9.15] <0.001 4.53 [2.35,8.76] <0.001
Education (per 1-year increase) 1.00 [0.97,1.03] 0.84
Recent MD Visit 1.64 [0.94,2.87] 0.08
Current smoker 0.65 [0.36,1.16] 0.15
Diagnosed with DM 0.63 [0.39,1.03] 0.07
Diagnosed with HTN 1.25 [0.87,1.80] 0.22
HbA1C % (per 1-% increase) 1.11 [0.95,1.29] 0.18
HbA1C% > 6.5 0.89 [0.47,1.71] 0.74
Systolic Blood Pressure (per 1-mmHg increase) 0.99 [0.99,1.00] 0.08
BP >150/90 1.60 [1.10,2.32] 0.01

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
BMI> 30 kg/m2 0.93 [0.69,1.25] 0.64
Prescribed Medication for:
DM 1.73 [1.04,2.87] 0.04
HTN 1.52 [1.06,2.18] 0.02
Hyperlipidemia 1.48 [1.10,1.99] 0.01

Abbreviations: aOR=adjusted Odds Ratio, CI=confidence interval, CKD=chronic kidney disease, SES=socioeconomic status, m=meter, MD=medical doctor, DM=diabetes mellitus, HTN=hypertension,
HbA1C=glycosylated hemoglobin, mmHg=millimeter mercury, BP=blood pressure, BMI=body mass index
*
Reference: San Jose
Page 16
 Harhay et al. Page 17

Table 3

Adjusted CKD prevalence by province

Author Manuscript

Age and gender adjusted Age, gender & altitude adjusted

Province Prevalence (%) 95% CI Prevalence (%) 95% CI

San Jose 20 (17–22) 16 (14–19)
Alajuela 14 (11–17) 16 (12–19)
Cartago 28 (23–33) 23 (18–28)
Heredia 20 (13–28) 16 (10–22)
Guanacaste 19 (14–25) 37 (25–48)
Puntarenas 15 (11–19) 25 (17–33)
Limon 25 (18–31) 42 (31–54)

Abbreviations: CI=confidence interval, CKD=chronic kidney disease

Author Manuscript
Author Manuscript
Author Manuscript

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Appendix 1

Longitudinal etiologic model of CKD in Costa Rica.

Model 1 Model 1 + SES & comorbidities

Harhay et al.

aOR 95% CI p-value aOR 95% CI p-value

Age 1.13 [1.11,1.15] <0.001 1.13 [1.11,1.15] <0.001
Female 1.55 [1.23,1.97] <0.001 1.23 [0.97,1.57] 0.09
Altitude (per 200 m) above sea level 1.25 [1.14,1.36] <0.001 1.23 [1.13,1.34] <0.001

Province of Residence*
Alajuela 0.92 [0.61,1.37] 0.67 0.95 [0.64,1.43] 0.82
Cartago 1.22 [0.86,1.72] 0.27 1.29 [0.91,1.83] 0.15
Heredia 1.49 [0.97,2.30] 0.07 1.31 [0.85,2.01] 0.22
Guanacaste 2.33 [1.27,4.28] 0.01 2.66 [1.45,4.89] 0.00
Puntarenas 1.22 [0.68,2.19] 0.50 1.38 [0.77,2.48] 0.28
Limon 2.95 [1.53,5.68] <0.001 2.72 [1.41,5.25] <0.001
Years of Education 1.02 [0.99,1.05] 0.24
No Insurance 0.91 [0.67,1.24] 0.57
Recent MD Visit 1.07 [0.72,1.61] 0.73
Current smoker 0.6 [0.38,0.95] 0.03
Diagnosed with DM 0.71 [0.44,1.16] 0.18
Diagnosed with HTN 1.8 [1.31,2.48] <0.001
HbA1C % 1.1 [0.96,1.26] 0.17
HbA1C % > 6.5 1.88 [1.24,2.86] 0.00
Systolic BP (per mmHg rise) 0.99 [0.98,1.00] 0.01

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
BP > 150/90 1.3 [0.96,1.76] 0.09
BMI > 30 kg/m2 0.97 [0.75,1.25] 0.81
DM 1.29 [0.78,2.14] 0.32
HTN 1.6 [1.18,2.16] <0.001
Hyperlipidemia 1.53 [1.20,1.96] <0.001

N = 2681 individuals and 4740 observations.

*
Reference: San Jose
Page 18
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript
Abbreviations: aOR = adjusted Odds Ratio, CI=confidence interval, m=meter, SES=socioeconomic status, MD=medical doctor, DM=diabetes mellitus, HTN=hypertension, HbA1C=glycosylated
hemoglobin, mmHg=millimeter mercury, BP=blood pressure, BMI=body mass index
Harhay et al.

Trop Med Int Health. Author manuscript; available in PMC 2017 January 01.
Page 19