---
title: "Live Birth Data from SINASC with healthbR"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Live Birth Data from SINASC with healthbR}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  eval = FALSE
)
```

## Overview

The **SINASC (Sistema de Informacoes sobre Nascidos Vivos)** is Brazil's national live birth information system, managed by the Ministry of Health through DATASUS. It records individual birth certificates (*Declaracao de Nascido Vivo*) with maternal, delivery, and newborn characteristics.

| Feature | Details |
|---------|---------|
| Coverage | Per state (UF), all 27 states |
| Years | 1996--2024 |
| Unit | One row per live birth certificate |
| Format | .dbc files from DATASUS FTP |

## Getting started

```{r setup}
library(healthbR)
library(dplyr)
```

### Check available years

```{r}
sinasc_years()

# include preliminary data
sinasc_years(status = "all")
```

### Module information

```{r}
sinasc_info()
```

## Downloading data

### Basic download

```{r}
births <- sinasc_data(year = 2022, uf = "AC")
```

### Multiple states and years

```{r}
births <- sinasc_data(year = 2020:2022, uf = c("SP", "RJ"))
```

### Filter by congenital anomaly

Use CID-10 code prefixes to filter births with congenital anomalies:

```{r}
# Down syndrome (Q90)
down <- sinasc_data(year = 2022, uf = "SP", anomaly = "Q90")

# All congenital anomalies (Chapter XVII)
anomalies <- sinasc_data(year = 2022, uf = "SP", anomaly = "Q")
```

### Select variables

```{r}
births <- sinasc_data(
  year = 2022,
  uf = "SP",
  vars = c("DTNASC", "SEXO", "PESO", "IDADEMAE", "GESTACAO",
           "PARTO", "CONSULTAS", "CODMUNRES")
)
```

## Key variables

| Variable | Description |
|----------|-------------|
| DTNASC | Birth date |
| SEXO | Sex (1=Male, 2=Female, 0=Unknown) |
| PESO | Birth weight (grams) |
| IDADEMAE | Mother's age (years) |
| GESTACAO | Gestational age (weeks, categorized) |
| PARTO | Delivery type (1=Vaginal, 2=Cesarean) |
| CONSULTAS | Prenatal consultations (categorized) |
| CODANOMAL | Congenital anomaly (CID-10 code) |
| CODMUNRES | Municipality of mother's residence (IBGE 6 digits) |
| ESCMAE | Mother's education level |
| RACACOR | Newborn's race/color |
| APGAR1, APGAR5 | Apgar score at 1 and 5 minutes |

### Data dictionary

```{r}
sinasc_dictionary()
sinasc_dictionary("PARTO")
sinasc_dictionary("GESTACAO")
```

### Explore variables

```{r}
sinasc_variables()
sinasc_variables(search = "mae")
sinasc_variables(search = "peso")
```

## Example: Low birth weight by state

```{r}
births <- sinasc_data(year = 2022, uf = c("SP", "RJ", "MG", "BA", "RS"))

lbw <- births |>
  filter(!is.na(PESO), PESO != "0") |>
  mutate(
    weight = as.numeric(PESO),
    low_weight = weight < 2500
  ) |>
  group_by(uf_source) |>
  summarise(
    total = n(),
    low_weight_n = sum(low_weight),
    low_weight_pct = low_weight_n / total * 100
  )
```

## Example: Cesarean rates over time

```{r}
births <- sinasc_data(year = 2018:2022, uf = "SP",
                      vars = c("PARTO", "CODMUNRES"))

cesarean <- births |>
  filter(PARTO %in% c("1", "2")) |>
  group_by(year) |>
  summarise(
    vaginal = sum(PARTO == "1"),
    cesarean = sum(PARTO == "2"),
    cesarean_rate = cesarean / (vaginal + cesarean) * 100
  )
```

## Example: Teen pregnancy

```{r}
births <- sinasc_data(year = 2022, uf = "SP")

teen <- births |>
  filter(!is.na(IDADEMAE)) |>
  mutate(
    mother_age = as.integer(IDADEMAE),
    teen_mother = mother_age < 20
  ) |>
  summarise(
    total = n(),
    teen_n = sum(teen_mother, na.rm = TRUE),
    teen_pct = teen_n / total * 100
  )
```

## Smart type parsing

```{r}
# parsed types (default)
births <- sinasc_data(year = 2022, uf = "AC")
class(births$DTNASC)  # Date
class(births$PESO)    # integer

# all character
births_raw <- sinasc_data(year = 2022, uf = "AC", parse = FALSE)
```

## Cache and lazy evaluation

```{r}
sinasc_cache_status()
sinasc_clear_cache()

# lazy query
lazy <- sinasc_data(year = 2022, uf = "SP", lazy = TRUE)
lazy |>
  filter(PARTO == "2") |>
  collect()
```

## Further reading

- SINASC on DATASUS (`datasus.saude.gov.br`)
- [SIM vignette](sim-mortality.html) for mortality data
- [Census vignette](censo-denominadores.html) for population denominators