library(mongolstats)
library(dplyr)
library(ggplot2)
# Checked teaching helpers are shipped with the package and can be inspected.
source(system.file("examples", "analysis-helpers.R", package = "mongolstats"), local = TRUE)
source(system.file("examples", "plot-style.R", package = "mongolstats"), local = TRUE)Overview
Mongolia’s National Statistics Office (NSO) publishes vital statistics and disease surveillance data that are useful for research and policy analysis. This guide shows how to find the right tables, check what they measure, and turn counts into rates.
Finding Health Tables
Search by Keyword
nso_search() searches table titles across the whole NSO
catalogue. Some tables are listed in more than one folder, so
distinct() keeps one row per table:
# Mortality
mortality <- nso_search("mortality")
mortality |>
distinct(tbl_id, tbl_eng_nm) |>
head(10)
#> # A tibble: 10 × 2
#> tbl_id tbl_eng_nm
#> <chr> <chr>
#> 1 DT_NSO_2100_014V1 NUMBER OF INFANT MORTALITY, aimags and the Capital and by …
#> 2 DT_NSO_2100_014V2 NUMBER OF INFANT MORTALITY, by soum and discrict, and by y…
#> 3 DT_NSO_2100_014V4 INFANT MORTALITY, by sex, by soum, and by year
#> 4 DT_NSO_2100_014V5 INFANT MORTALITY RATE, per 1000 live births, by sex, by so…
#> 5 DT_NSO_2100_015V1 INFANT MORTALITY RATE, per 1000 live births, aimags and th…
#> 6 DT_NSO_2100_015V2 INFANT MORTALITY, per 1000 live births, by soum, and by ye…
#> 7 DT_NSO_2100_023V1 MATERNAL MORTALITY RATIO, per 100000 live births, by soum,…
#> 8 DT_NSO_2100_030V2 UNDER-FIVE MORTALITY, per 1000 live births, aimags and the…
#> 9 DT_NSO_2100_040V1 NEONATAL MORTALITY, aimags and the Capital and by month
#> 10 DT_NSO_2100_040V2 NEONATAL MORTALITY RATE, per 1000 live births, aimags and …
# Cancer
cancer <- nso_search("cancer")
cancer |> distinct(tbl_id, tbl_eng_nm)
#> # A tibble: 5 × 2
#> tbl_id tbl_eng_nm
#> <chr> <chr>
#> 1 DT_NSO_2100_012V1 NEW CASES OF CANCER, per 10000 population, by type of cance…
#> 2 DT_NSO_2100_013V1 DEATHS OF CANCER, per 10000 population, by type of cancer, …
#> 3 DT_NSO_2100_044V1 NEW CASES OF CANCER, by age group, by year
#> 4 DT_NSO_2100_045V1 NEW CASES AND MORTALITY OF CANCER, per 10000 population, ai…
#> 5 DT_NSO_2100_045V1 NEW CASES AND MORTALITY OF CANCER, per 10000 population, ai…
# Communicable diseases
infectious <- nso_search("communicable")
infectious |> distinct(tbl_id, tbl_eng_nm)
#> # A tibble: 3 × 2
#> tbl_id tbl_eng_nm
#> <chr> <chr>
#> 1 DT_NSO_2100_020V2 INCIDENCE OF COMMUNICABLE DISEASES, by type of diseases, by…
#> 2 DT_NSO_2100_028V1 INCIDENCE OF COMMUNICABLE DISEASES, aimags and the Capital,…
#> 3 DT_NSO_2100_035V1 CASES OF COMMUNICABLE DISEASES, by type of selected disease…Read the title before you fetch: it tells you whether a table holds counts (“NUMBER OF …”, “CASES OF …”) or rates (“per 1000 live births”, “per 10000 population”), and at which level (“aimags and the Capital”, “by soum”).
Browse by Sector
Health and education statistics are grouped together:
# View all sectors
sectors <- nso_sectors()
sectors
#> # A tibble: 8 × 3
#> id type text
#> <chr> <chr> <chr>
#> 1 Economy, environment l Economy, environment
#> 2 Education, health l Education, health
#> 3 Historical data l Historical data
#> 4 Industry, service l Industry, service
#> 5 Labour, business l Labour, business
#> 6 Population, household l Population, household
#> 7 Regional development l Regional development
#> 8 Society, development l Society, development
# Find health-related subsectors
health_sector <- sectors |> filter(grepl("health", text, ignore.case = TRUE))
if (nrow(health_sector) > 0) {
subsectors <- nso_subsectors(health_sector$id[1])
subsectors |> head()
}
#> # A tibble: 6 × 3
#> id type text
#> <chr> <chr> <chr>
#> 1 Births, deaths l Births, deaths
#> 2 Disease l Disease
#> 3 General educational schools l General educational schools
#> 4 General indicators for Education l General indicators for Education
#> 5 Health insurance l Health insurance
#> 6 Main indicators for Health sector l Main indicators for Health sectorCase Study: Cancer Incidence
Exploring the Table
cancer_tbl <- "DT_NSO_2100_012V1" # New cases of cancer per 10,000 population
# Examine available dimensions
meta <- nso_table_meta(cancer_tbl)
meta
#> # A tibble: 2 × 5
#> dim code is_time n_values codes
#> <chr> <chr> <lgl> <int> <list>
#> 1 Type malignant neoplasms Хорт хавдрын төрөл FALSE 7 <tibble [7 × 3]>
#> 2 Year Он TRUE 26 <tibble [26 × 3]>
# Cancer types: codes are not in a meaningful order, so read the labels
cancer_types <- nso_dim_values(cancer_tbl, "Type malignant neoplasms", labels = "en")
cancer_types
#> # A tibble: 7 × 2
#> code label_en
#> <chr> <chr>
#> 1 0 "Total"
#> 2 1 " Liver"
#> 3 2 " Cervix uteri"
#> 4 3 " Stomach"
#> 5 4 " Lung"
#> 6 5 " Oesophagus"
#> 7 6 " Other"
# Year codes are internal ids; the labels hold the years
annual_meta <- nso_dim_values(cancer_tbl, "Year", labels = "en")
range(annual_meta$label_en)
#> [1] "2000" "2025"Incidence by Site
# The 10 most recent years
recent_years <- annual_meta |>
arrange(label_en) |>
tail(10) |>
pull(code)
# Specific sites only: drop the "Total" and "Other" categories. NSO labels
# can carry leading spaces, so trim them before comparing.
site_codes <- cancer_types |>
filter(!trimws(label_en) %in% c("Total", "Other")) |>
pull(code)
cancer_data <- nso_data(
tbl_id = cancer_tbl,
selections = list(
"Type malignant neoplasms" = site_codes,
"Year" = recent_years
),
labels = "en"
) |>
mutate(site = trimws(`Type malignant neoplasms_en`), year = as.integer(Year_en))
# One panel per site, on a shared scale, so overlapping lines do not hide each other
cancer_data |>
ggplot(aes(x = year, y = value)) +
geom_line(color = "#5B6E9A", linewidth = 1) +
geom_point(color = "#5B6E9A", size = 2) +
facet_wrap(~site, ncol = 3) +
scale_x_continuous(breaks = function(x) range(round(x))) +
scale_y_continuous(limits = c(0, NA), breaks = scales::breaks_width(2)) +
labs(
title = "Cancer Incidence in Mongolia by Site",
subtitle = "New cases per 10,000 population (crude rate, rounded by NSO to whole numbers)",
x = NULL,
y = "Incidence (per 10,000)",
caption = "Source: NSO Mongolia, DT_NSO_2100_012V1"
) +
theme(panel.grid.minor = element_blank(), panel.spacing = unit(1, "lines"))Over these years, liver cancer has the highest mean of the published annual rates. This equal-year mean is a descriptive summary, not a pooled incidence rate. Two cautions apply to this table:
- Rounding. NSO rounds these rates to whole numbers per 10,000, so a change of 1 between years can be rounding alone.
- Single-year spikes. An isolated spike is a reason to check source definitions and records. It can reflect ascertainment, screening, classification, data errors or real variation; do not remove it solely because it looks unusual.
The denominator for a sex-specific cancer, such as cervical cancer, must be checked in the source documentation. A rate per total population must not be interpreted as a rate among women. Use compatible sex-specific counts and denominators for that question.
Crude versus Age-Standardised Incidence
The rates above are crude: new cases divided by the whole population. Cancer risk rises steeply with age, so a crude rate climbs as a population ages even when the risk at each age is flat. Age-standardisation reduces the effect of differences in age composition between the groups used. It applies each year’s age-specific rates to one fixed standard population, here the WHO World Standard Population (2000–2025).
NSO publishes the numerators and denominators needed: new cancer
cases by age group (DT_NSO_2100_044V1) and population by
age group (DT_NSO_0300_003V1).
# New cancer cases by age group and year (counts)
case_years <- nso_dim_values("DT_NSO_2100_044V1", "Year", labels = "en")
case_ages <- nso_dim_values("DT_NSO_2100_044V1", "Age group", labels = "en")
indicator <- nso_dim_values("DT_NSO_2100_044V1", "Indicator", labels = "en")
cancer_case_source <- nso_data(
"DT_NSO_2100_044V1",
selections = list(
Indicator = indicator$code[trimws(indicator$label_en) == "New cases of cancer"],
"Age group" = case_ages$code,
Year = case_years$label_en
),
labels = "en"
)
cancer_case_status <- nso_value_status(cancer_case_source)
cancer_case_raw <- cancer_case_source |>
transmute(year = Year_en, age = trimws(`Age group_en`), cases = value)
cancer_cases <- filter(cancer_case_raw, age != "Total")
# Population by age group, same years, both sexes
pop_years <- nso_dim_values("DT_NSO_0300_003V1", "Year", labels = "en")
pop_ages <- nso_dim_values("DT_NSO_0300_003V1", "Age group", labels = "en")
pop_sex <- nso_dim_values("DT_NSO_0300_003V1", "Sex", labels = "en")
cancer_pop_source <- nso_data(
"DT_NSO_0300_003V1",
selections = list(
Sex = pop_sex$code[trimws(pop_sex$label_en) == "Total"],
"Age group" = pop_ages$code,
Year = pop_years$label_en[pop_years$label_en %in% case_years$label_en]
),
labels = "en"
)
cancer_pop_status <- nso_value_status(cancer_pop_source)
cancer_pop_raw <- cancer_pop_source |>
transmute(year = Year_en, age = trimws(`Age group_en`), population = value)
population <- filter(cancer_pop_raw, age != "Total")
# The two tables use different age groups; collapse both to the cancer
# table's groups (0-4, 5-14, 15-19, ..., 60-64, 65+)
to_cancer_groups <- function(age) {
case_when(
age %in% c("5-9", "10-14") ~ "5-14",
age %in% c("65-69", "70-74", "75-79", "80-84", "85+", "70+") ~ "65+",
.default = age
)
}
# WHO 2000-2025 standard-million representation used by SEER.
# https://seer.cancer.gov/stdpopulations/world.who.html
# These weights sum to 1,000,000 before and after grouping.
who_standard <- tibble::tibble(
age = c(
"0-4", "5-9", "10-14", "15-19", "20-24", "25-29", "30-34", "35-39", "40-44",
"45-49", "50-54", "55-59", "60-64", "65-69", "70-74", "75-79", "80-84", "85+"
),
weight = c(
88569, 86870, 85970, 84670, 82171, 79272, 76073, 71475, 65877,
60379, 53681, 45484, 37187, 29590, 22092, 15195, 9097, 6348
)
) |>
mutate(age = to_cancer_groups(age)) |>
summarise(weight = sum(weight), .by = age)
# Require mutually exclusive source groups before collapsing the population.
stopifnot(
!any(population$age == "70+") ||
!any(population$age %in% c("70-74", "75-79", "80-84", "85+"))
)
analysis_keys(cancer_cases, c("year", "age"))
analysis_keys(population, c("year", "age"))
population <- population |>
mutate(age = to_cancer_groups(age)) |>
summarise(population = analysis_sum(population), .by = c(year, age))
# Both sources must contain the same years and every target age group.
# An inner join must not silently select the analysis population.
cancer_asr <- analysis_asr(cancer_cases, population, who_standard) |>
mutate(year = as.integer(year)) |>
arrange(year)
stopifnot(all(cancer_asr$standard_weight == 1e6))
# Reconcile sums against any supplied, compatible all-age total.
case_totals <- filter(cancer_case_raw, age == "Total") |> select(year, total = cases)
pop_totals <- filter(cancer_pop_raw, age == "Total") |> select(year, total = population)
if (nrow(case_totals)) {
check <- analysis_pair(summarise(cancer_cases, calculated = analysis_sum(cases), .by = year), case_totals, "year")
stopifnot(all(is.finite(check$total)), all(check$calculated == check$total))
}
if (nrow(pop_totals)) {
check <- analysis_pair(summarise(population, calculated = analysis_sum(population), .by = year), pop_totals, "year")
stopifnot(all(is.finite(check$total)), all(check$calculated == check$total))
}
cancer_asr |>
select(year, crude, standardised) |>
knitr::kable(
col.names = c("Year", "Crude incidence", "Age-standardised incidence"), digits = c(0, 1, 1),
caption = "Cancer incidence per 10,000 people; WHO World standard"
)| Year | Crude incidence | Age-standardised incidence |
|---|---|---|
| 2010 | 16.2 | 26.2 |
| 2011 | 16.2 | 26.4 |
| 2012 | 15.8 | 25.5 |
| 2013 | 17.9 | 28.6 |
| 2014 | 17.7 | 27.0 |
| 2015 | 18.2 | 29.4 |
| 2016 | 18.4 | 29.2 |
| 2017 | 18.3 | 27.9 |
| 2018 | 17.9 | 27.1 |
| 2019 | 18.3 | 26.4 |
| 2020 | 19.1 | 26.7 |
| 2021 | 17.5 | 23.8 |
| 2022 | 19.9 | 26.0 |
| 2023 | 20.6 | 26.0 |
| 2024 | 23.0 | 28.0 |
| 2025 | 22.3 | 26.1 |
cancer_asr |>
tidyr::pivot_longer(c(crude, standardised), names_to = "measure", values_to = "rate") |>
mutate(measure = if_else(measure == "crude", "Crude", "Age-standardised (WHO World)")) |>
ggplot(aes(x = year, y = rate, color = measure)) +
geom_line(linewidth = 1.2) +
geom_point(size = 2) +
scale_color_manual(values = c("Crude" = "#167D8D", "Age-standardised (WHO World)" = "#C2623D"), name = NULL) +
scale_y_continuous(limits = c(0, NA)) +
labs(
title = "All-Site Cancer Incidence: Crude and Age-Standardised",
subtitle = "New cases per 10,000 population",
x = NULL,
y = "Incidence (per 10,000)",
caption = "Source: NSO Mongolia, DT_NSO_2100_044V1 and DT_NSO_0300_003V1"
) +
theme(legend.position = "top", panel.grid.minor = element_blank())Between 2010 and 2025, the crude rate changed by +38.0% and the standardised rate by -0.5%. A smaller change after standardisation is consistent with ageing contributing to the crude change. It does not establish unchanged risk in every age group or quantify how much of the change is attributable to ageing. These are descriptive estimates, not a formal trend test. Broad groups, especially 65+, retain some differences in age composition.
Biostatistical note: Age-standardised rates are for comparison only: between years, between countries, or between groups with different age structures. They depend on the standard population chosen. Report the crude rate, or the number of cases, when the question is about the actual burden on health services. The 2021 dip coincides with the pandemic; these counts do not separate changes in disease occurrence from diagnosis and reporting. Confirm the population table’s reference date and resident/total-population coverage against the cancer source before publication.
Finding Numerators and Denominators
Many NSO rate tables have matching count tables. With counts you can pool years, combine areas, and attach confidence intervals, which a published rate does not allow. For infant mortality:
-
DT_NSO_2100_014V1: infant deaths (count), by aimag and month -
DT_NSO_2100_018V5: live births (count), by aimag and month -
DT_NSO_2100_015V2: infant mortality rate (per 1,000 live births), by soum and year
Before you rely on rates you build yourself, check that they reproduce the published ones:
# Preserve source status and require complete observed national years.
national_deaths <- analysis_fetch_counts("DT_NSO_2100_014V1", "0")
national_births <- analysis_fetch_counts("DT_NSO_2100_018V5", "0")
valid_years <- analysis_common_years(national_deaths$annual, national_births$annual, require_values = TRUE)
stopifnot(length(valid_years) > 0)
infant_deaths <- filter(national_deaths$annual, year %in% valid_years)
live_births <- filter(national_births$annual, year %in% valid_years)
# Published annual rate. This id names two tables in the catalogue, so
# mongolstats warns and uses the first match ("by soum, and by year"),
# which includes the national total.
imr_years <- nso_dim_values("DT_NSO_2100_015V2", "Year", labels = "en")
imr_published <- nso_data(
"DT_NSO_2100_015V2",
selections = list(Region = "0", Year = intersect(imr_years$label_en, valid_years)),
labels = "en"
) |>
select(year = Year_en, published = value)
published_years <- imr_published$year
imr_check <- analysis_pair(
select(filter(infant_deaths, year %in% published_years), year, deaths = count),
select(filter(live_births, year %in% published_years), year, births = count), "year"
) |>
analysis_pair(imr_published, by = "year") |>
mutate(
calculated = deaths / births * 1000,
difference = calculated - published,
# Whole-number published rounding permits up to half a unit.
agrees_with_rounding = is.finite(difference) & abs(difference) <= 0.5 + 1e-8
) |>
arrange(year)
stopifnot(all(imr_check$births > 0))
imr_check |>
transmute(
year, calculated, published, difference,
agreement = if_else(agrees_with_rounding, "Yes", "Needs review")
) |>
knitr::kable(
col.names = c("Year", "Calculated IMR", "Published IMR", "Difference", "Agrees within rounding"),
digits = c(0, 1, 1, 1, 0), caption = "Infant deaths per 1,000 live births: reconciliation with the published rate"
)| Year | Calculated IMR | Published IMR | Difference | Agrees within rounding |
|---|---|---|---|---|
| 2016 | 16.8 | 17 | -0.2 | Yes |
| 2017 | 13.6 | 14 | -0.4 | Yes |
| 2018 | 13.4 | 13 | 0.4 | Yes |
| 2019 | 13.3 | 13 | 0.3 | Yes |
| 2020 | 11.5 | 12 | -0.5 | Yes |
| 2021 | 11.6 | 12 | -0.4 | Yes |
| 2022 | 12.4 | 12 | 0.4 | Yes |
| 2023 | 12.3 | 12 | 0.3 | Yes |
| 2024 | 12.2 | 12 | 0.2 | Yes |
| 2025 | 13.6 | 14 | -0.4 | Yes |
The table explicitly shows whether each comparable year agrees within half a unit, the tolerance for a whole-number rounded rate. 0 year(s) need review under this check. Differences outside that tolerance cannot be explained by whole-number rounding alone. Agreement is a reconciliation check, not proof that registration and population coverage are complete. The Getting Started and Mapping guides retain these limitations.
Case Study: Tuberculosis Notifications
The monthly communicable disease table counts reported (notified) cases. Annual summation assumes these are monthly increments, not cumulative year-to-date values; confirm that reporting convention with the source before using the estimates. To compare years, divide by the population at risk. Here we use the national mid-year population.
tb_tbl <- "DT_NSO_2100_035V1" # Cases of communicable diseases, by type and by month
# Find the code for tuberculosis
indicators <- nso_dim_values(tb_tbl, "Indicators", labels = "en")
tb_code <- indicators |>
filter(trimws(label_en) == "Tuberculosis") |>
pull(code)
stopifnot(length(tb_code) == 1L)
tb_months <- nso_table_periods(tb_tbl)$label
tb_raw <- nso_data(tb_tbl, list(Indicators = tb_code, Month = tb_months), labels = "en")
tb_status <- nso_value_status(tb_raw)
tb_annual <- analysis_annual_counts(mutate(tb_raw, Region = "0"), "0") |>
filter(usable) |>
transmute(year, cases = count, n_months)
stopifnot(nrow(tb_annual) > 0L)
# Mid-year population (denominator)
population <- nso_data(
"DT_NSO_0300_002V1",
selections = list(Region = "0", Year = tb_annual$year),
labels = "en"
) |>
select(year = Year_en, population = value)
tb_rate <- analysis_pair(tb_annual, population, "year")
stopifnot(all(is.finite(tb_rate$population)), all(tb_rate$population > 0))
tb_rate <- tb_rate |>
mutate(
year = as.integer(year),
rate = cases / population * 1e5,
# Exact Poisson 95% confidence interval
lower = qchisq(0.025, 2 * cases) / 2 / population * 1e5,
upper = qchisq(0.975, 2 * (cases + 1)) / 2 / population * 1e5
) |>
arrange(year)
tb_rate |>
transmute(year, cases, population, rate, interval = sprintf("%.1f–%.1f", lower, upper)) |>
knitr::kable(
col.names = c("Year", "Reported cases", "Population", "Notifications per 100,000", "95% CI"),
digits = c(0, 0, 0, 1, 0), format.args = list(big.mark = ",")
)| Year | Reported cases | Population | Notifications per 100,000 | 95% CI |
|---|---|---|---|---|
| 2,016 | 4,045 | 3,088,856 | 131.0 | 126.9–135.1 |
| 2,017 | 3,779 | 3,148,917 | 120.0 | 116.2–123.9 |
| 2,018 | 3,498 | 3,208,189 | 109.0 | 105.4–112.7 |
| 2,019 | 3,624 | 3,267,673 | 110.9 | 107.3–114.6 |
| 2,020 | 3,339 | 3,332,062 | 100.2 | 96.8–103.7 |
| 2,021 | 2,331 | 3,393,455 | 68.7 | 65.9–71.5 |
| 2,022 | 2,355 | 3,443,458 | 68.4 | 65.7–71.2 |
| 2,023 | 2,357 | 3,490,859 | 67.5 | 64.8–70.3 |
| 2,024 | 2,232 | 3,534,503 | 63.1 | 60.6–65.8 |
| 2,025 | 2,104 | 3,572,835 | 58.9 | 56.4–61.5 |
tb_rate |>
ggplot(aes(x = year, y = rate)) +
geom_ribbon(aes(ymin = lower, ymax = upper), fill = "#C2623D", alpha = 0.15) +
geom_line(color = "#C2623D", linewidth = 1) +
geom_point(color = "#C2623D", size = 2) +
scale_x_continuous(breaks = tb_rate$year) +
scale_y_continuous(limits = c(0, NA), expand = expansion(mult = c(0, 0.1))) + # start at 0
labs(
title = "Tuberculosis Notifications in Mongolia",
subtitle = "Reported cases per 100,000 population, with 95% confidence intervals",
x = NULL,
y = "Notifications per 100,000",
caption = "Source: NSO Mongolia, DT_NSO_2100_035V1 and DT_NSO_0300_002V1"
) +
theme(panel.grid.minor = element_blank())The notification rate fell from 131 per 100,000 in 2016 to 59 in 2025.
Biostatistical note: A notification rate counts cases that were diagnosed and reported, not all new cases (incidence). It falls when case finding falls, as well as when transmission falls. The intervals describe count variation conditional on the notification process and population denominator; they do not include missed diagnoses or reporting error. The sharp drop in 2021 coincides with the COVID-19 pandemic, which disrupted tuberculosis services in many countries. These data cannot separate a true decline from missed diagnoses. For incidence, see the World Health Organization’s country estimates, which adjust for under-reporting.
Tips for Epidemiological Research
-
Check time coverage: Use
nso_table_periods()ornso_dim_values()to see which periods a table covers. Monthly tables usually include a partial current year; keep complete years for annual comparisons. - Check what a table measures: The title says whether values are counts or rates and gives the rate’s denominator.
- Prefer counts for analysis: Build rates from numerator and denominator counts, then check them against any published rate.
- Use the right denominator: Live births for infant and maternal mortality; mid-year population for incidence and crude death rates.
- Watch for ambiguous ids: A few table ids name more than one table. mongolstats warns and tells you which one it used.
-
Geographic detail varies: Many health tables are by
aimag; some are by soum (“by soum” in the title). Check the
Regiondimension withnso_dim_values().
Next Steps
- Mapping Health Outcomes: See the Mapping Guide for spatial epidemiology
- Reference Documentation: Explore all available functions in the Reference