History of coronary heart disease increased the mortality rate of COVID-19 patients: a nested case-control study

Read the published paper in BMJ Open

China has experienced an outbreak of a novel human coronavirus (SARS-CoV-2) since December 2019, which quickly became a worldwide pandemic in early 2020. There is limited evidence on the mortality risk effect of pre-existing comorbidities for coronavirus disease 2019 (COVID-19), which has important implications for early treatment. This study aims to evaluate the risk of pre-existing comorbidities on COVID-19 mortality, and provide clinical suggestions accordingly. Under the nested case-control design, a total of 94 publicly reported deaths in locations outside of Hubei Province, China, between December 18th, 2019 and March 8th, 2020 were included as cases. Each case was matched with up to three controls, based on gender and age ± 1 year old (94 cases and 181 controls).

Quick links:

• Study Design and Rationale
• Data Collection Procedure
• Data Summary
• Weighted Cox Proportional Hazard Model-Univariate
• Weighted Cox Proportional Hazard Model-Multivariate
• KM plot
• Sensitivity Analyses

Study Design and Rationale

This study performed survival analysis under a nested case-control (NCC) design to assess the roles of common comorbidities (cardiocerebrovascular, endocrine and respiratory disease, etc.) in predicting mortality for COVID-19, among patients in mainland China outside of Hubei Province. The study period was from December 18th, 2019, when the first laboratory-confirmed case was announced in China, till March 8th, 2020.

The study cohort was defined as all the publicly reported confirmed COVID-19 patients outside of Hubei Province in mainland China between the study period. During this period, 112 deaths outside of Hubei Province were reported by the National Health Committee of China, and 18 were excluded from the present study due to missingness of important clinical information. A total of 448 publicly reported laboratory-confirmed COVID-19 cases (94 deaths and 354 survivors) were initially collected. The data collection procedure was blinded to patient comorbidity information. All deaths were included as cases, and each case was matched with up to three controls on gender and age ± 1 year old (94 cases and 181 controls).

Data Collection Procedure

We routinely searched for daily news and public health reports on confirmed COVID-19 cases in all areas in mainland China outside of Hubei Province. Patients’ clinical and comorbidity characteristics were recorded and doubly confirmed by national/provincial/municipal health commission websites, the official COVID-19 data reporting websites in China. Follow-up time was defined as the duration from the date of disease onset till the end of observation on March 8th or when the participant died, whichever came first. For each eligible patient, we followed local reports to update their survival status until the end of follow-up time.

As illustrated in Figure 1, the inclusion criterion was publicly reported COVID-19 patients who had complete information on basic demographics (age, gender and region), disease onset date--the first time a patient became symptomatic, and history of comorbidities (include but not limited to hypertension, cardiovascular disease, diabetes and respiratory diseases) were included in the analysis. Asymptomatic patients were not included in this study. In addition, we defined “comorbidity-free patients” as those who were specifically described as “no pre-existing medical condition/comorbidity” on the national/provincial/municipal health commission websites.

In the following three steps, we used the No. 214 patient as an example to introduce the dynamic tracking method we used to identify any missing dates:

Step 1. Conducting an internet search on confirmed cases on baidu.com, the largest search engine in China, using keywords “confirmed COVID-19 cases report” and “pre-existing comorbidities.” A search result pertained to one confirmed case reported on the website of Municipal Health Commission of Binzhou (Shandong Province) on February 17th, described as “the 15th confirmed case: 30-year-old male without pre-existing morbidities, who lives in the neighborhood of Xincun Village. This patient was diagnosed positive on February 16th and is being treated with precaution in Bincheng hospital.” We recorded age, gender, region and comorbidity-free for this patient.

Step 2. We then determined the onset date of this patient based on another announcement on the same website. In this announcement titled “Possible exposure locations and times of the 15th confirmed case,” it says, “the patient was symptomatic on February 14th.”

Step 3. Finally, we confirmed the event status of this patient as discharged on March 3rd, by following the updates on this website.

Summary Data

	Survivor (n=181)	Death (n=94)	P-value	Overall (n=275)
Age (years)
Mean (SD)	64.2 (14.7)	70.7 (13.3)	<0.001	66.4 (14.5)
Median [Min, Max]	67.0 [24.0, 90.0]	72.5 [25.0, 94.0]		68.0 [24.0, 94.0]
Sex
Female	64 (35.4%)	38 (40.4%)	0.488	102 (37.1%)
Male	117 (64.6%)	56 (59.6%)		173 (62.9%)
Early non-intervention period
After 01/11/2020	138 (76.2%)	67 (71.3%)	0.453	205 (74.5%)
Before 01/10/2020	43 (23.8%)	27 (28.7%)		70 (25.5%)
History of surgery
No	175 (96.7%)	90 (95.7%)	0.956	265 (96.4%)
Yes	6 (3.3%)	4 (4.3%)		10 (3.6%)
Hypertension
No	114 (63.0%)	52 (55.3%)	0.27	166 (60.4%)
Yes	67 (37.0%)	42 (44.7%)		109 (39.6%)
Coronary heart disease
No	166 (91.7%)	69 (73.4%)	<0.001	235 (85.5%)
Yes	15 (8.3%)	25 (26.6%)		40 (14.5%)
Chronic_Bronchitis
No	169 (93.4%)	87 (92.6%)	0.998	256 (93.1%)
Yes	12 (6.6%)	7 (7.4%)		19 (6.9%)
COPD
No	176 (97.2%)	87 (92.6%)	0.136	263 (95.6%)
Yes	5 (2.8%)	7 (7.4%)		12 (4.4%)
Diabetes
No	135 (74.6%)	68 (72.3%)	0.797	203 (73.8%)
Yes	46 (25.4%)	26 (27.7%)		72 (26.2%)
Cerebral_Infarction
No	173 (95.6%)	83 (88.3%)	0.0446	256 (93.1%)
Yes	8 (4.4%)	11 (11.7%)		19 (6.9%)
Cardiac_Failure
No	169 (93.4%)	84 (89.4%)	0.353	253 (92.0%)
Yes	12 (6.6%)	10 (10.6%)		22 (8.0%)
Renal_Failure
No	175 (96.7%)	88 (93.6%)	0.384	263 (95.6%)
Yes	6 (3.3%)	6 (6.4%)		12 (4.4%)
Hepatic_Failure
No	181 (100%)	91 (96.8%)	0.0711	272 (98.9%)
Yes	0 (0%)	3 (3.2%)		3 (1.1%)
Comorbidity Score
Mean (SD)	1.02 (1.11)	1.60 (2.17)	<0.001	1.22 (1.21)
Median [Min, Max]	1.00 [0.00, 5.0]	1.50 [0.00, 5.0]		1.00 [0.00, 5.0]

Additional Fisher’s exact test & Mann–Whitney U test

## 
##  Fisher's Exact Test for Count Data
## 
## data:  covid_sub$History_of_Surgery and covid_sub$Death
## p-value = 0.7394
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##  0.2618976 5.6202653
## sample estimates:
## odds ratio 
##   1.294993

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  Age by Death
## W = 6255.5, p-value = 0.0003183
## alternative hypothesis: true location shift is not equal to 0

## 
##  Wilcoxon rank sum test with continuity correction
## 
## data:  ComorbidityScore by Death
## W = 5940.5, p-value = 2.296e-05
## alternative hypothesis: true location shift is not equal to 0

Median followup time:

## Call: survfit(formula = Surv(time = Followup_Days, event = event_follow) ~ 
##     1, data = covid_sub)
## 
##       n  events  median 0.95LCL 0.95UCL 
##     275     181      40      38      42

Univariate Cox PH Model

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Age	0.0486394	1.049842	0.0086594	0.0100961	4.817644	1.5e-06

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Male	-0.2732098	0.7609332	0.2103793	0.2329391	-1.172881	0.2408437

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Early_Infection	0.1115256	1.117982	0.2307952	0.2498649	0.4463437	0.655349

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
History_of_Surgery	0.5353023	1.707964	0.5111814	0.5562612	0.9623218	0.335888

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Hypertension	0.3119277	1.366056	0.2075527	0.2339723	1.333182	0.1824719

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
CHD	1.433072	4.191554	0.2346989	0.2710545	5.287023	1e-07

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Chronic_Bronchitis	0.0508731	1.052189	0.3930306	0.4487422	0.1133681	0.9097387

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
COPD	0.9590383	2.609186	0.393382	0.3909177	2.4533	0.0141552

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Diabetes	0.1349872	1.144522	0.2306795	0.2611842	0.5168275	0.6052766

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
ComorbidityScore	0.4018354	1.494565	0.0772606	0.0816546	4.921158	9e-07

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Cerebral_Infarction	1.049599	2.856504	0.3219388	0.3600373	2.91525	0.003554

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Cardiac_Failure	0.6132745	1.846468	0.3346376	0.3758117	1.631866	0.1027076

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Renal_Failure	0.8310933	2.295827	0.4233677	0.4904419	1.69458	0.0901551

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Hepatic_Failure	2.11074	8.254343	0.5915688	0.288125	7.325777	0

Multivariate Cox PH Model

Model 1:

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Age	0.0422091	1.043113	0.0092716	0.0106404	3.9668523	0.0000728
Male	0.0478739	1.049038	0.2145176	0.2295828	0.2085257	0.8348185
Early_Infection	0.2460885	1.279013	0.2320980	0.2455243	1.0022980	0.3161997
ComorbidityScore	0.2673628	1.306514	0.0816799	0.0822388	3.2510526	0.0011498

Model 2:

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Age	0.0425327	1.043450	0.0090123	0.0104228	4.0807517	0.0000449
Male	0.0852805	1.089023	0.2134930	0.2275641	0.3747538	0.7078435
Early_Infection	0.1779544	1.194771	0.2318960	0.2490904	0.7144171	0.4749693
CHD	1.0732430	2.924849	0.2428424	0.2652509	4.0461433	0.0000521

Model 3:

	coef	exp(coef)	se(coef)	robust se	z	Pr(>|z|)
Age	0.0372684	1.0379716	0.0092484	0.0107136	3.4786102	0.0005040
Male	-0.0027273	0.9972764	0.2186409	0.2396018	-0.0113828	0.9909181
Early_Infection	0.1904524	1.2097968	0.2338193	0.2500363	0.7616992	0.4462396
CHD	1.1012042	3.0077858	0.2423555	0.2575034	4.2764651	0.0000190
Cerebral_Infarction	0.6419621	1.9002055	0.3341922	0.3592212	1.7870938	0.0739223
COPD	0.6166034	1.8526247	0.4154428	0.3734279	1.6511980	0.0986982
Renal_Failure	0.7046864	2.0232120	0.4360997	0.4684805	1.5041956	0.1325310

KM Plot

Sensitivity Analyses

Sensitivity analysis was performed using multivariate logistic regression to provide estimated odds ratio (ORs), which includes the same covariates as the multivariate weighted Cox model. The results were similar.

	Estimate	Std. Error	z value	Pr(>|z|)
(Intercept)	-2.9553523	0.8053584	-3.6696111	0.0002429
Age	0.0289566	0.0106839	2.7103099	0.0067220
Male	0.0466922	0.2820745	0.1655316	0.8685256
Early_Infection	0.4753632	0.3048857	1.5591523	0.1189603
CHD	1.1833551	0.3711950	3.1879609	0.0014328

	Estimate	Std. Error	z value	Pr(>|z|)
(Intercept)	-2.5960546	0.8248978	-3.1471228	0.0016489
Age	0.0220286	0.0111294	1.9793162	0.0477804
Male	-0.0416592	0.2890719	-0.1441136	0.8854107
Early_Infection	0.5207312	0.3105032	1.6770559	0.0935316
CHD	1.2364024	0.3786224	3.2655289	0.0010926
Cerebral_Infarction	0.9023069	0.5167924	1.7459757	0.0808152
COPD	0.9351284	0.6386495	1.4642279	0.1431317
Renal_Failure	0.6642553	0.6410489	1.0362007	0.3001085

	Estimate	Std. Error	z value	Pr(>|z|)
(Intercept)	-3.0054603	0.8072070	-3.7232834	0.0001966
Age	0.0271152	0.0109868	2.4679816	0.0135877
Male	-0.0188726	0.2797069	-0.0674726	0.9462054
Early_Infection	0.4863549	0.3054186	1.5924208	0.1112902
ComorbidityScore	0.3104978	0.1148960	2.7024239	0.0068836