Local

• rr_epilepsy_age_ge20,base_prob_3m_epi_death: Based on Subsarahn africa study, defined as local
• base_prob_3m_seiz_stat_freq_infreq, base_prob_3m_seiz_stat_infreq_freq, base_prob_3m_seiz_stat_none_freq, base_prob_3m_seiz_stat_none_infreq, base_prob_3m_seiz_stat_infreq_none: All these rates have limited data, so rate has been assumed. Given high uncertainty of these, I have deemed them to be local
• base_prob_3m_seiz_stat_freq_infreq,base_prob_3m_seiz_stat_none_infreq,prob_start_anti_epilep_when_seizures_detected_in_generic_first_appt: Discrepancy in write-up value and code.

Undetermined

• unavailable_medicine_retry_months, unavailable_appt_retry_months : For retry variables (eg medicine or appt not available, so patient goes back after 1mo) currently have this set to undetermined with value design decision. However, I see that this can also be linked to a local context

Hard coded parameters

• Status of epi seizures defined as (0 = never epilepsy, 1 = previous seizures none now, 2 = infrequent seizures, 3 = frequent seizures). I kept these "numbers" hard coded in the code. In the future, may enhance clarity of code to replace numbering with text description.

Files

• File IHME-GBD_2019_DATA-4a82f00e-1.csv is used only in the analysis script. Left as-is for now
• parameter file structure can be updated to fit the style of ResourceFile_Epilepsy, ...