There were 46448 patients with pre-existing T2D in the Montefiore health system, of which 13801 (29.71%) tested positive for COVID-19. Of the patients with COVID-19, 8071 patients were hospitalized, of which 6402 returned to our health system, and 5730 patients were not hospitalized, of which 5137 returned. In the non-COVID cohort, 1638 patients were hospitalized for LRTI of which 1373 returned, and 31009 patients were not hospitalized and did not have LRTI of which 21317 returned.
Table 1 summarizes the characteristics of all patients with T2D with and without COVID-19, stratified by hospitalization status, who returned to our health system. Compared to the hospitalized LRTI cohort, patients hospitalized for COVID-19 were younger (64.08 ± 14.99 vs 69.08 ± 13.81 years, p < 0.001), had lower prevalence of pre-existing cardiovascular disease (40.73% vs 46.69%, p = 0.005), asthma (27.12% vs 32.37%, p = 0.004), and COPD (20.67% vs 32.29%, p < 0.001), higher prevalence of pre-existing obesity (45.38% vs 42.71%, p = 0.021), and lower HbA1c levels within 1-year pre-index (7.75 ± 2.19 vs 7.13 ± 1.81, p < 0.001).
Table 1
Demographics and pre-existing comorbidities at index date for hospitalized COVID-19 versus hospitalized LRTI, and non-hospitalized COVID-19 versus non-hospitalized LRTI. [Highlighted] p < 0.05 between COVID + and COVID- patients and hospitalized COVID + and LRTI patients. SD: Standard Deviation.
| Hosp COVID-19 N = 6402 | Hosp LRTI N = 1373 | p value | Non-hosp COVID-19 N = 5137 | Non-Hosp COVID-19/LRTI negative N = 21317 | p value |
Demographics N (%) | | | | | | |
Age, yrs, mean (± SD) | 64.08 (14.99) | 69.08 (13.81) | < 0.001 | 60.03 (14.01) | 61.28 (16.96) | 0.001 |
Male | 2675 (47.14%) | 556 (47.12%) | 0.396 | 1890 (38.09%) | 10695 (50.17%) | < 0.001 |
Hispanic | 2414 (42.54%) | 554 (46.95%) | 0.072 | 2208 (44.50%) | 7671 (35.99%) | < 0.001 |
White | 643 (11.33%) | 123 (10.42%) | 0.240 | 505 (10.18%) | 2617 (12.28%) | 0.003 |
Black | 2124 (37.43%) | 442 (37.46%) | 0.501 | 1722 (34.70%) | 6713 (31.49%) | < 0.001 |
Other | 2480 (43.71%) | 543 (46.02%) | 0.597 | 2167 (43.67%) | 8141 (38.19%) | < 0.001 |
Pre-existing Comorbidities N (%) | | | | | |
Hypertension | 5009 (88.28%) | 1080 (91.53%) | 0.760 | 4052 (81.66%) | 14530 (68.16%) | < 0.001 |
CKD | 2869 (50.56%) | 641 (54.32%) | 0.217 | 1423 (28.68%) | 4239 (19.89%) | < 0.001 |
Cardiovasc Disease | 2311 (40.73%) | 551 (46.69%) | 0.005 | 1064 (21.44%) | 2421 (11.36%) | < 0.001 |
Asthma | 1539 (27.12%) | 382 (32.37%) | 0.004 | 1447 (29.16%) | 2779 (13.04%) | < 0.001 |
COPD | 1173 (20.67%) | 381 (32.29%) | < 0.001 | 590 (11.89%) | 1299 (6.09%) | < 0.001 |
Obesity (or BMI > 30) | 2575 (45.38%) | 504 (42.71%) | 0.017 | 2287 (46.09%) | 5949 (27.91%) | < 0.001 |
1-year pre-index HbA1c, mean(± SD) | 7.75 (2.19) n = 3110 | 7.13 (1.81) n = 534 | < 0.001 | 7.40 (1.81) n = 1004 | 7.98 (2.16) n = 3408 | < 0.001 |
Compared to the non-hospitalized non-COVID cohort, the non-hospitalized COVID-19 cohort was younger (60.03 ± 14.01 vs 61.28 ± 16.96), more likely to be Hispanic (44.50% vs 35.99%, p < 0.001), Black (34.70% vs 31.49%, p < 0.001), and of race Other (43.67% vs 38.19%, p < 0.001), less likely to be male (38.09% vs 50.17%, p < 0.001) and White (10.18% vs 12.28%, p = 0.003), and had higher prevalence of pre-existing hypertension (81.66% vs 68.16%, p < 0.001), CKD (50.56% vs 54.32%, p < 0.001), cardiovascular disease (40.73% vs 46.69%, p < 0.001), asthma (27.12% vs 32.37%, p < 0.001), COPD (11.89% vs 6.09%, p < 0.001), obesity (46.09% vs 27.91%, p < 0.001), and lower HbA1c levels within 1-year pre-index (7.40 ± 1.81 vs 7.98 ± 2.16, p < 0.001).
Figure 2 shows the cumulative incidence for HTN stratified by hospitalization status from 2 months to 3-years post-index date. The cumulative incidence of HTN among the hospitalized COVID-19 cohort was not significantly different from that of the hospitalized LRTI cohort (26.13% vs 15.67%, p = 0.137; HR = 1.14, 95% CI [0.69, 1.47], p = 0.251). The cumulative incidence of HTN among the non-hospitalized COVID cohort was significantly higher than that of the non-COVID cohort (22.32% vs 9.13%, p < 0.001; HR = 1.84, 95% CI [1.34, 2.22], p < 0.001).
Multivariate logistic regressions were performed to examine the effect of COVID-19 relative to other potential risk factors for incident HTN using adjusted odds ratios for all patients without preexisting hypertension stratified by hospitalization status (Table 2). Patients hospitalized for COVID-19 had similar odds of developing HTN during (aOR 1.26, 95% CI [0.70, 2.27], p = 0.441) relative to patients hospitalized for LRTI. By comparison, non-hospitalized patients with COVID-19 had greater odds of developing HTN (aOR 1.99, 95% CI [1.54, 2.57], p < 0.001) relative to patients not hospitalized for COVID or LRTI. Older (aOR 1.02, 95% CI [1.01, 1.02], p < 0.001) and male (aOR 1.39, 95% CI [1.19, 1.64], p < 0.001) patients had greater odds of developing HTN, while obese patients (aOR 0.58, 95% CI [0.39, 0.88], p = 0.009) had lower odds of developing HTN.
Table 2
Odds ratios for newly diagnosed HTN among patients with T2D for (A) hospitalized COVID vs. LRTI and (B) non-hospitalized COVID vs NonCOVID/LRTI during at 3-year follow-up.
| (A) Hosp COVID-19 vs Hosp LRTI | (B) Non-hosp COVID vs. NonCOVID/LRTI |
| OR | 95% CI | p-values | OR | 95% CI | p-values |
COVID positive status | 1.26 | [0.70, 2.27] | 0.441 | 1.99 | [1.54, 2.57] | < 0.001 |
Age (years) | 1.01 | [0.99, 1.02] | 0.097 | 1.02 | [1.01, 1.02] | < 0.001 |
Male | 1.30 | [0.91, 1.85] | 0.152 | 1.39 | [1.19, 1.64] | < 0.001 |
Black vs. White | 1.00 | [0.58, 1.71] | 0.995 | 1.10 | [0.83, 1.43] | 0.452 |
Other vs. White | 0.97 | [0.56, 1.68] | 0.918 | 0.97 | [0.75, 1.25] | 0.804 |
Hispanic vs. Non-Hisp | 0.83 | [0.56, 1.22] | 0.334 | 0.95 | [0.84, 1.09] | 0.473 |
CKD | 1.21 | [0.81, 1.80] | 0.340 | 1.12 | [0.60, 2.10] | 0.719 |
Cardiovascular Disease | 1.27 | [0.84, 1.93] | 0.252 | 1.05 | [0.50, 2.21] | 0.902 |
Asthma | 0.88 | [0.54, 1.40] | 0.586 | 1.07 | [0.68, 1.67] | 0.770 |
COPD | 1.04 | [0.62, 1.74] | 0.876 | 0.87 | [0.36, 2.10] | 0.752 |
Obesity (or BMI > 30) | 1.04 | [0.72, 1.49] | 0.849 | 0.58 | [0.39, 0.88] | 0.009 |
Figure 3 shows the cumulative incidence for CKD stratified by hospitalization status from 2 months to 3-years post-index date. The cumulative incidence of CKD among the hospitalized COVID-19 cohort was not significantly different from that of the hospitalized LRTI cohort (11.27% vs 11.31%, p = 0.786; HR = 0.98, 95% CI [0.87, 1.23], p = 0.404). The cumulative incidence of CKD among the non-hospitalized COVID cohort was significantly higher than that of the non-COVID cohort (6.20% vs 2.03%, p < 0.001; HR = 1.97, 95% CI [1.57, 2.43], p < 0.001).
Multivariate logistic regressions were performed to examine the effect of COVID-19 relative to other potential risk factors for incident CKD using adjusted odds ratios for all patients without preexisting hypertension stratified by hospitalization status (Table 3). Patients hospitalized for COVID-19 had similar odds of developing CKD at follow-up (aOR 0.96, 95% CI [0.79, 1.36], p = 0.131) relative to patients hospitalized for LRTI. By comparison, non-hospitalized patients with COVID-19 had greater odds of developing CKD at follow-up (aOR 2.09, 95% CI [1.69, 2.76], p < 0.001) relative to patients not hospitalized for COVID or LRTI. Older (aOR 1.04, 95% CI [1.03, 1.05], p < 0.001) and male (aOR 1.24, 95% CI [1.52, 2.23], p < 0.001) patients had greater odds of developing CKD, as well has patients with preexisting HTN (aOR 2.89, 95% CI [1.85, 6.01], p < 0.001) and cardiovascular disease (aOR 1.50, 95% CI [1.01, 2.22], p = 0.020).
Table 3
Odds ratios for newly diagnosed CKD among patients with T2D for (A) hospitalized COVID vs. LRTI and (B) non-hospitalized COVID vs NonCOVID/LRTI at 3-year follow-up.
| (A) Hosp COVID-19 vs Hosp LRTI | (B) Non-hosp COVID vs. NonCOVID/LRTI |
| OR | 95% CI | p-values | OR | 95% CI | p-values |
COVID positive status | 0.96 | [0.79, 1.36] | 0.131 | 2.09 | [1.69, 2.76] | < 0.001 |
Age (years) | 1.02 | [0.99, 1.03] | 0.063 | 1.04 | [1.03, 1.05] | < 0.001 |
Male | 1.07 | [0.83, 1.38] | 0.596 | 1.24 | [1.52, 2.23] | < 0.001 |
Black vs. White | 1.60 | [0.99, 2.53] | 0.051 | 1.19 | [0.87, 1.63] | 0.274 |
Other vs. White | 1.25 | [0.98, 2.51] | 0.064 | 1.03 | [0.76, 1.40] | 0.859 |
Hispanic vs. Non-Hisp | 0.82 | [0.62, 1.09] | 0.170 | 1.12 | [0.96, 1.31] | 0.165 |
CKD | 1.79 | [1.22, 2.61] | 0.002 | 2.89 | [1.85, 6.01] | < 0.001 |
Cardiovascular Disease | 1.71 | [1.31, 2.23] | < 0.001 | 1.50 | [1.01, 2.22] | 0.020 |
Asthma | 0.96 | [0.71, 1.29] | 0.777 | 1.06 | [0.71, 1.59] | 0.758 |
COPD | 0.86 | [0.62, 1.21] | 0.402 | 0.90 | [0.52, 1.59] | 0.727 |
Obesity (or BMI > 30) | 1.15 | [0.89, 1.50] | 0.288 | 0.84 | [0.59, 1.19] | 0.336 |
The new incident HTN and CKD and their ORs for developing new HTN and CKD due COVID-19 status were also computed for three follow-up periods (2–6, 6–12, 12–36 months post index date) (Table 4). There were no significant differences across all time points for hospitalized cohort, but there were significant differences across all time points (except the 2–6 month follow-up for incident HTN, and OR for new CKD) for non-hospitalized cohort. P values became smaller at the later time points.
Table 4
(A) Incidence and (B) ORs due COVID-19 status for developing new HTN and CKD for three different follow-up periods (2–6, 6–12, 12–36 months post index date)
(A) | | Hosp COVID-19 vs Hosp LRTI | Non-hosp COVID vs. non-hosp controls |
| | COVID | LRTI | p | COVID | controls | p |
New HTN | 2–6 months | 44/524 (8.40%) | 8/87 (9.20%) | 0.96 | 29/745 = 3.89% | 125/3833 = 3.75% | 0.445 |
6–12 months | 47/455 (10.33%) | 3/59 (5.08%) | 0.28 | 40/688 = 5.81% | 140/4047 = 3.46% | 0.004 |
12–36 months | 52/407 (12.78%) | 4/43 (9.30%) | 0.67 | 52/534 = 9.74% | 277/5092 = 5.44% | < 0.001 |
New CKD | 2–6 months | 59/2289 (2.58%) | 16/467 (3.43%) | 0.38 | 42/3088 = 1.36% | 57/9801 = 0.58% | < 0.001 |
6–12 months | 58/1987 (2.92%) | 15/352 (3.69%) | 0.24 | 49/2686 = 1.82% | 63/10224 = 0.62% | < 0.001 |
12–36 months | 123/1830 (6.72%) | 19/276 (6.88%) | 1.00 | 56/2031 = 2.76% | 188/12880 = 1.46% | < 0.001 |
(B) | | Hosp COVID-19 vs Hosp LRTI | Non-hosp COVID vs. non-hosp controls |
| | OR | 95% CI | p | OR | 95% CI | p-values |
New HTN | 2–6 months | 0.87 | [0.37, 2.06] | 0.75 | 1.73 | [1.05, 2.86] | 0.032 |
6–12 months | 1.54 | [0.44, 5.44] | 0.49 | 2.07 | [1.29, 3.30] | 0.002 |
12–36 months | 1.66 | [0.53, 5.15] | 0.37 | 2.46 | [1.67, 3.64] | < 0.001 |
New CKD | 2–6 months | 1.03 | [0.89, 1.76] | 0.07 | 1.45 | [0.67, 2.75] | 0.384 |
6–12 months | 0.79 | [0.44, 0.99] | 0.13 | 1.87 | [1.23, 3.30] | < 0.001 |
12–36 months | 0.69 | [0.39, 1.22] | 0.20 | 1.84 | [1.31, 2.26] | < 0.001 |