Table 3.

Bivariate and Multivariate Associations of Food Insecurity and Severe Hypoglycemia: Weighted^* Frequencies and Relative Risks (95% CI) from Weighted Log-Binomial Regression Analyses

Variable Name	Level	Total participants No. (%)	Self-reported severe hypoglycemia, Weighted Row %	Unadjusted (bivariate) Relative Risk (95% CI; p-value)	Adjusted† Relative Risk (95% CI; p-value)
Any food insecurity	No	993 (90.9)	4.1	REF	REF
	Yes	100 (9.2)	15.3	3.76 (1.82–7.76; p = 0.0003)	4.12 (1.71–9.92; p = 0.002)
Economic food insecurity (due to financial limitations)	No	1006 (93.3)	4.3	REF
	Yes	72 (6.7)	16.4	3.80 (1.74–8.32; p = 0.0008)	4.31 (1.33–14.02; p = 0.02)
Physical food insecurity (due to difficulty cooking or shopping for food)	No	1020 (94)	4.1	REF
	Yes	65 (6)	18.6	4.52 (2.07–9.88; p = 0.0002)	4.36 (1.76–10.78; p = 0.002)
Both economic and physical food insecurity‡	No	1064 (96.4)	4.4	REF
	Yes	40 (3.6)	22.6	5.18 (2.16–12.4; p = 0.0002)	5.44 (1.54–19.23; p = 0.009)

^*In 1,122 patients who responded to question about severe hypoglycemia. Weights account for both the age-stratified random sampling design as well as survey non-response

^†Adjusted for age, frailty, HbA1c, Charlson comorbidity score and financial strain

^‡The reference group includes patients who have no or only one type of food insecurity (i.e., either economic or physical) but not both