Factors influencing COVID-19 Infection in older individuals: History of Alcohol Use Disorder, Major Depressive illness, genetic variation and current use of alcohol

Introduction

The Coronavirus Disease 2019 (COVID-19) pandemic continues to be a major public health problem. Vulnerable populations include older individuals with presumed weakening of the immune response. Identification of factors influencing COVID-19 infection could provide an additional means for protecting such individuals.

Methods

Members of a family study previously interviewed as middle aged individuals were re-contacted and asked to participate in extended phone interview (2-3 hours) covering past and current mental health issues, physical health diagnoses, use of alcohol and drugs, and exposure to anyone with COVID-19. The average follow-up period was 32 years. Detailed medication use was collected to confirm medical diagnoses and to reveal possible protective effects of particular drug classes currently prescribed for the participant by their physician. Serology was available for red cell antigens (ABO, Kell, Duffy, Kidd, Rhesus) and HLA subtypes. Analyses were conducted to contrast COVID-19 + and COVID-19 - individuals for physical and mental health diagnoses, use of alcohol and drugs, and red cell and HLA serology. Additionally, analyses were conducted to contrast these groups with a group reporting known exposure but absence of COVID-19 symptoms or diagnosis by a health professional.

Results

Interviews were completed between September 2020 and November 2021. A total of 42 of the 90 individuals interviewed had been vaccinated at the time of interview. At the time of interview, 11.1% reported having developed COVID-19.

Using quantity per occasion (QPO) and quantity by frequency (QXF) totals in the past month by type of alcohol consumed, we found a significant association between QPO for liquor (p=0.017) and marginal effects for QXF for liquor consumption (p=0.06). Exposed individuals who were COVID-19 negative tended to drink more liquor than those who were positive, an average of about one drink per day. Beer and wine consumption were not statistically significant. A diagnosis of alcohol use disorder at baseline evaluation was not a significant predictor of being COVID positive or negative.

Self-reported current depression or depression in the past only was not a predictor of COVID-19 status based on a single question “Are you depressed currently or only in the past?”. In contrast, completion of a clinical interview designed to elicit depressed mood and concurrent symptoms for determination of the lifetime presence or absence of a depressive episode did reveal a significant effect. Comparison of responses at baseline to follow-up showed those most resilient to developing COVID-19 were those without evidence of a depressive episode by lifetime history at both points in time.

Physical health issues were analyzed for those that were frequently occurring in our sample such as hypertension but not found to be significant. BMI was analyzed and found to be statistically non-significant.

Analysis of HLA variation across the whole sample did not reveal a significant association but among males two variants, A1 and B8, did show significant variation associated with COVID-19+ and COVID-19-status. Analyses of the red cell antigens revealed one significant red cell effect; Kidd genotypic variation was associated with COVID-19 status.

Interpretation

We tentatively conclude that use of specific types of alcohol, namely liquor, is associated with reduced frequency of COVID-19. However, the amount is low, averaging about 1 drink per day. Enlarged samples are needed to confirm these results. The finding that past history of alcohol use disorder does not increase likelihood of developing COVID-19 is important. It should be noted that the 34 individuals diagnosed with AUD at baseline had survived an average of 32 years in order to participate in the current interview suggesting they may be especially resilient to adverse health conditions. The finding that a single question designed to elicit the presence or absence of depressed mood either currently or in the past was not a risk factor for COVID-19 in contrast to report of a clinically significant past history of a depressive episode based on more extensive examination using DSM criteria is important. Results for the KIDD blood group are novel and warrant further investigation.