haematological profile of malaria patients

project structure

• hemo_db.R -> import and cleaning
• hemo_bygroup.R -> cross-sectional analysis using baseline data
• hemo_byvisit.R -> longitudinal analysis using folow-ups!

método: analisis longitudinal

1. definir variable dependiente

• Var. dependiente = cada una de las mediciones hematológicas
• Y: gaussian distribution identity link function

2. definir variables independientes

• Var. independiente [time-invariant] = infeccion (especie de plasmodium, dicotomica)
• Var. independiente [confusores] = edad (continua), sexo (dicotómica)

3. objetivo:

• test if, in average, the change of Y in time is related with a time-invariant exposure
• this translates as including an interaction term between Y and the time-invariant exposure (Vittinghoff, 2nd edition, p271)

4. correlation structure

• the repeated measurements were taken through time
• only three time measurements per pacient were included
• the measurement interval times are not equally spaced
• autoregressive process 1 (ar1) because the correlation between times differ

5. clustered data

• each pacient had 3 visits each with one measurement for all the haematological variables
• dataset is in a long (tidy) format

---

historial

20180103

• recepción de las bases

• estructura:

  • Controles (n=308) del año 2010
  • Pf positivos (n=34) años 2011 y 2012
  • Pv positivos (n=93) año 2011

20181201

• importar y limpiar base de datos

20190506

• union de base de datos y ejecución de análisis descriptivos

20200204

• corrección por autocorrelación debido a mediciones repetidas

---

citar

@software{andree_valle_campos_2020_4014205,
  author       = {Andree Valle Campos},
  title        = {avallecam/hemogr: First release},
  month        = sep,
  year         = 2020,
  publisher    = {Zenodo},
  version      = {v0.1},
  doi          = {10.5281/zenodo.4014205},
  url          = {https://doi.org/10.5281/zenodo.4014205}
}