Affiliation concerning lithium use and the incidence of dementia and its subtypes: A retrospective cohort analyze

S1 Appendix. More details on examine techniques and findings.

Fig A: Association of lithium use with the improvement of dementia and its subtypes by Cox proportional dangers types: sensitivity evaluation with a for a longer time (2-year) criterion for determining prescription drugs/comorbidities (see Methods). Altered HRs, 95% CIs, and p-values were extracted from inverse probability weighted Cox regression. Modified for age, sex, marital status, ethnicity, cigarette smoking status, alcoholic beverages ailments, antipsychotic use, despair, mania or BPAD, hypertension, central vascular sickness, diabetic issues mellitus, and hyperlipidemias. Fig B: Affiliation of length of lithium publicity with the progress of dementia and its subtypes by Cox proportional dangers styles: sensitivity evaluation with a more time (2-yr) criterion for determining medications/comorbidities (see Methods). Altered HRs, 95% CIs, and p-values were extracted from inverse probability weighted Cox regression. Adjusted for age, sexual intercourse, marital status, ethnicity, smoking position, alcohol ailments, antipsychotic use, melancholy, mania or BPAD, hypertension, central vascular ailment, diabetic issues mellitus, and hyperlipidemias. NA implies no end result (no corresponding scenarios). Fig C: Association of lithium use with the advancement of dementia and its subtypes by Cox proportional hazards designs: sensitivity investigation demanding at minimum 2 decades of observe-up. Modified HRs, 95% CIs, and p-values ended up extracted from inverse likelihood weighted Cox regression. Altered for age, sex, marital standing, ethnicity, smoking position, liquor ailments, antipsychotic use, despair, mania or BPAD, hypertension, central vascular disorder, diabetic issues mellitus, and hyperlipidemias. Fig D: Affiliation of duration of lithium exposure with the development of dementia and its subtypes by Cox proportional hazards models: sensitivity examination necessitating at least 2 several years of comply with-up. Adjusted HRs, 95% CIs, and p-values have been extracted from inverse likelihood weighted Cox regression. Altered for age, sexual intercourse, marital status, ethnicity, using tobacco position, liquor problems, antipsychotic use, depression, mania or BPAD, hypertension, central vascular illness, diabetes mellitus, and hyperlipidemias. NA implies no end result (no corresponding conditions). Fig E: Association of lithium with the advancement of dementia and its subtypes by Cox proportional hazards designs: sensitivity investigation assuming that all lithium buyers experienced BPAD. Modified HRs, 95% CIs, and p-values have been extracted from inverse probability weighted Cox regression. Altered for age, sex, marital status, ethnicity, smoking cigarettes status, alcoholic beverages diseases, antipsychotic use, depression, mania or BPAD, hypertension, central vascular illness, diabetes mellitus, and hyperlipidemias. Fig F: Association of period of lithium publicity with the improvement of dementia and its subtypes by Cox proportional hazards products: sensitivity assessment assuming that all lithium users had BPAD. Adjusted HRs, 95% CIs, and p-values ended up extracted from inverse probability weighted Cox regression. Modified for age, sex, marital standing, ethnicity, smoking position, alcohol disorders, antipsychotic use, melancholy, mania or BPAD, hypertension, central vascular illness, diabetic issues mellitus, and hyperlipidemias. NA indicates no result (no corresponding circumstances). Fig G: Association of lithium with the enhancement of dementia and its subtypes by Cox proportional hazards versions: sensitivity examination by excluding persons diagnosed with recurrent melancholy. Adjusted HRs, 95% CIs, and p-values were extracted from inverse probability weighted Cox regression. Altered for age, sex, marital standing, ethnicity, using tobacco status, liquor problems, antipsychotic use, depression, mania or BPAD, hypertension, central vascular illness, diabetes mellitus, and hyperlipidemias. Fig H: Affiliation of length of lithium publicity with the development of dementia and its subtypes by Cox proportional hazards styles: sensitivity analysis by excluding individuals diagnosed with recurrent depression. Altered HRs, 95% CIs, and p-values have been extracted from inverse probability weighted Cox regression. Altered for age, sexual intercourse, marital position, ethnicity, smoking cigarettes standing, alcoholic beverages issues, antipsychotic use, despair, mania or BPAD, hypertension, central vascular disease, diabetic issues mellitus, and hyperlipidemias. NA signifies no outcome (no corresponding instances). Fig I: Association of lithium with the improvement of dementia and its subtypes by Cox proportional dangers models: sensitivity investigation by excluding the ethnicity variable. Adjusted HRs, 95% CIs, and p-values had been extracted from inverse chance weighted Cox regression. Altered for age, sexual intercourse, marital standing, smoking cigarettes standing, alcoholic beverages issues, antipsychotic use, depression, mania or BPAD, hypertension, central vascular disease, diabetes mellitus, and hyperlipidemias. Fig J: Affiliation of duration of lithium publicity with the enhancement of dementia and its subtypes by Cox proportional hazards versions: sensitivity assessment by excluding the ethnicity variable. Modified HRs, 95% CIs, and p-values were extracted from inverse likelihood weighted Cox regression. Modified for age, sexual intercourse, marital status, smoking cigarettes status, alcohol ailments, antipsychotic use, despair, mania or BPAD, hypertension, central vascular condition, diabetic issues mellitus, and hyperlipidemias. Fig K: Association of lithium with the advancement of dementia and its subtypes by Cox proportional hazards models: sensitivity examination by thinking about competing outcomes from loss of life. Altered HRs, 95% CIs, and p-values were extracted from inverse probability weighted Cox regression. Modified for age, intercourse, marital status, ethnicity, using tobacco status, alcoholic beverages conditions, antipsychotic use, despair, mania or BPAD, hypertension, central vascular illness, diabetic issues mellitus, and hyperlipidemias. Fig L: Association of length of lithium exposure with the progress of dementia and its subtypes by Cox proportional hazards models: sensitivity examination by thinking of competing consequences from death. Adjusted HRs, 95% CIs, and p-values were extracted from inverse likelihood weighted Cox regression. Modified for age, sex, marital standing, ethnicity, smoking status, alcohol conditions, antipsychotic use, despair, mania or BPAD, hypertension, central vascular illness, diabetic issues mellitus, and hyperlipidemias. NA implies no outcome (no corresponding situations). Fig M: Affiliation of lithium with the development of dementia and its subtypes by Cox proportional hazards designs: sensitivity evaluation by only controlling for confounders identified at baseline. Altered HRs, 95% CIs, and p-values had been extracted from inverse chance weighted Cox regression. Altered for age, sexual intercourse, marital position, ethnicity, smoking standing, alcoholic beverages conditions, antipsychotic use, melancholy, mania or BPAD, hypertension, central vascular sickness, diabetes mellitus, and hyperlipidemias. Fig N: Association of period of lithium publicity with the progress of dementia and its subtypes by Cox proportional dangers models: sensitivity analysis by only controlling for confounders regarded at baseline. Adjusted HRs, 95% CIs, and p-values were being extracted from inverse chance weighted Cox regression. Altered for age, sexual intercourse, marital standing, ethnicity, using tobacco position, alcoholic beverages ailments, antipsychotic use, melancholy, mania or BPAD, hypertension, central vascular illness, diabetes mellitus, and hyperlipidemias. NA indicates no result (no corresponding instances). Fig O: Association of lithium with the progress of dementia and its subtypes by Cox proportional hazards styles: sensitivity examination by treating publicity to lithium as a time-various variable. Altered HRs, 95% CIs, and p-values were being extracted from inverse probability weighted Cox regression. Adjusted for age, sexual intercourse, marital status, ethnicity, using tobacco standing, alcoholic beverages ailments, antipsychotic use, melancholy, mania or BPAD, hypertension, central vascular illness, diabetic issues mellitus, and hyperlipidemias. Fig P: Affiliation of length of lithium exposure with the enhancement of dementia and its subtypes by Cox proportional hazards designs: sensitivity analysis by dealing with the duration of exposure to lithium as a time-different variable. Adjusted HRs, 95% CIs, and p-values have been extracted from inverse likelihood weighted Cox regression. Adjusted for age, sexual intercourse, marital standing, ethnicity, cigarette smoking position, liquor problems, antipsychotic use, depression, mania or BPAD, hypertension, central vascular condition, diabetic issues mellitus, and hyperlipidemias. Fig Q: Association of length of lithium exposure with the advancement of dementia and its subtypes by Cox proportional dangers types: sensitivity assessment by pooling intermediate lithium exposure durations. Modified HRs, 95% CIs, and p-values ended up extracted from inverse probability weighted Cox regression. Modified for age, sexual intercourse, marital standing, ethnicity, cigarette smoking position, alcoholic beverages disorders, antipsychotic use, melancholy, mania or BPAD, hypertension, central vascular disease, diabetes mellitus, and hyperlipidemias. Fig R: Affiliation of duration of lithium publicity with the advancement of dementia and its subtypes by Cox proportional dangers designs: sensitivity evaluation by incorporating an conversation term involving lithium publicity and publicity duration (in addition to the major result of lithium exposure). HRs are for exposure length (the conversation phrase) consequently, HRs < 1 indicate a progressive reduction in the risk of dementia for each additional year of lithium exposure (over and above the fact of lithium exposure). Unadjusted/adjusted HRs, 95% CIs, and p-values were extracted from inverse probability weighted Cox regression. Adjusted for age, sex, marital status, ethnicity, smoking status, alcohol disorders, antipsychotic use, depression, mania or BPAD, hypertension, central vascular disease, diabetes mellitus, and hyperlipidemias. BPAD, bipolar affective disorder/mania CI, confidence interval HR, hazard ratio.

https://doi.org/10.1371/journal.pmed.1003941.s002