Dual influenza and pneumococcal vaccination was associated with lower short-term risks of all-cause and acute respiratory hospitalizations among the elderly in Shenzhen, China: a retrospective cohort study

Figures & data

Figure 1. Time periods of each observation in the vaccinated and control groups. The matching of potential individuals for the control group and further selection of control group individuals is as following. (1) Vaccinated individuals were matched to all inpatient and outpatient medical occurrences of controls using age at index date/medical occurrence, sex, and district. (2) The index date of each vaccinated individual were assigned to the matched potential controls. (3) Four controls among the matched potential controls were randomly selected for each vaccinated individual. (4) Some individuals in the control group might have multiple appearances because they were matched to more than one vaccinated individual with distinct index dates. To resolve this, a distinct set of control observations were further filtered by keeping the observation of each control with the earliest index date.

Figure 2. Flowcharts of sample selection in the vaccinated and control groups.

Table 1. Demographic and baseline clinical characteristics of the vaccinated and control groups before and after entropy balancing.

	Control group before balancing (N=93,692)			Control group after balancing (weighted N=48,116)			Vaccinated group (N=48,116)
	% or mean	variance	p-value of testing against the vaccinated group	% or mean	variance	p-value of testing against the vaccinated group	% or mean	variance
Age at index	70.0	55.0	<0.001	69.6	46.9	0.997	69.6	46.9
Female	51.52%	0.2498	<0.001	54.71%	0.2478	1.000	54.71%	0.2478
District			<0.001			1.000
Guangming	2.62%	0.0255		4.22%	0.0404		4.22%	0.0404
Nanshan	12.86%	0.1121		13.09%	0.1138		13.09%	0.1138
Pingshan	1.54%	0.0152		1.42%	0.0140		1.42%	0.0140
Dapeng	1.22%	0.0121		1.63%	0.0161		1.63%	0.0161
Bao’an	10.77%	0.0961		10.33%	0.0926		10.33%	0.9262
Yantian	1.69%	0.0166		1.93%	0.0189		1.93%	0.0189
Futian	33.76%	0.2236		29.54%	0.2082		29.54%	0.2082
Luohu	20.95%	0.1656		19.96%	0.1597		19.96%	0.1597
Longhua	3.08%	0.0298		4.13%	0.0396		4.13%	0.0396
Longgang	11.50%	0.1018		13.75%	0.1186		13.75%	0.1186
Any baseline hospitalizations	21.93%	0.1712	<0.001	13.40%	0.1161	1.000	13.40%	0.1161
Any baseline outpatient visits	57.62%	0.2442	<0.001	39.30%	0.2386	1.000	39.31%	0.2386
Baseline inpatient costs	¥5,965	7.56E+08	<0.001	¥2,187	1.15E+08	1.000	¥2,187	1.15E+08
Baseline outpatient costs	¥1,420	2.33E+07	<0.001	¥865	1.09E+07	1.000	¥865	1.09E+07
Respiratory illness in the baseline period (not including chronic respiratory diseases)
Any baseline flu or pneumonia hospitalizations	1.83%	0.0179	<0.001	0.93%	0.0092	1.000	0.93%	0.0092
Any baseline acute upper respiratory infection hospitalizations	0.63%	0.0062	<0.001	0.34%	0.0034	1.000	0.34%	0.0034
Any baseline acute lower respiratory infection hospitalizations other than flu and pneumonia	0.52%	0.0052	<0.001	0.35%	0.0034	1.000	0.35%	0.0034
Any baseline hospitalizations due to other disorders of lung	2.19%	0.0214	<0.001	0.80%	0.0079	1.000	0.80%	0.0079
Any baseline cough-related hospitalizations	0.14%	0.0014	0.192	0.11%	0.0011	1.000	0.11%	0.0011
Any baseline hospitalizations due to abnormalities of breathing	0.38%	0.0038	<0.001	0.19%	0.0019	1.000	0.19%	0.0019
Any baseline hospitalizations due to pain in throat and chest	0.24%	0.0024	0.001	0.13%	0.0013	1.000	0.13%	0.0013
Any baseline outpatient visits due to respiratory tract infection	19.21%	0.1552	<0.001	17.13%	0.1419	1.000	17.13%	0.1419
Any baseline outpatient visits due to coughing	3.46%	0.0334	<0.001	2.75%	0.0268	1.000	2.75%	0.0268
Comorbid conditions in the baseline period
Any baseline hospitalizations due to lipoprotein metabolism disorders	4.56%	0.0435	<0.001	2.92%	0.0284	1.000	2.92%	0.0284
Any baseline hospitalizations due to coronary heart disease	4.71%	0.0449	<0.001	2.78%	0.0270	1.000	2.78%	0.0270
Any baseline hospitalizations due to hypertension	11.38%	0.1009	<0.001	6.69%	0.0624	1.000	6.69%	0.0624
Any baseline hospitalizations due to chronic lower respiratory diseases	2.77%	0.0269	<0.001	1.60%	0.0158	1.000	1.60%	0.0158
Any baseline hospitalizations due to fatty liver	2.39%	0.0233	<0.001	1.80%	0.0176	1.000	1.80%	0.0176
Any baseline hospitalizations due to type 2 diabetes	5.58%	0.0527	<0.001	3.18%	0.0308	1.000	3.18%	0.0308
Any baseline outpatient visits due to arthritis	4.10%	0.0394	<0.001	3.64%	0.0351	1.000	3.64%	0.0351
Any baseline outpatient visits due to hyperlipidemia or atherosclerosis	9.85%	0.0888	<0.001	6.99%	0.0650	1.000	6.99%	0.0650
Any baseline outpatient visits due to cerebrovascular diseases and sequelae	3.45%	0.0333	<0.001	2.41%	0.0235	1.000	2.41%	0.0235
Any baseline outpatient visits due to hypertension	22.49%	0.1743	<0.001	16.28%	0.1363	1.000	16.28%	0.1363
Any baseline outpatient visits due to COPD	0.65%	0.0064	<0.001	0.42%	0.0042	1.000	0.42%	0.0042
Any baseline outpatient visits due to diabetes	10.71%	0.0956	<0.001	7.55%	0.0698	1.000	7.55%	0.0698
Outcomes in the DiD analytic period
Any hospitalizations in 3rd pre-index month	3.47%	0.0335	<0.001	1.82%	0.0179	1.000	1.82%	0.0179
Any hospitalizations in 2nd pre-index month	3.47%	0.0335	<0.001	1.94%	0.0190	1.000	1.94%	0.0190
Any hospitalizations in 1st pre-index month	3.86%	0.0371	<0.001	1.44%	0.0142	1.000	1.44%	0.0142
Any acute respiratory hospitalization in 3rd pre-index month	0.59%	0.0059	<0.001	0.28%	0.0028	1.000	0.28%	0.0028
Any acute respiratory hospitalization in 2nd pre-index month	0.59%	0.0059	<0.001	0.28%	0.0028	1.000	0.28%	0.0028
Any acute respiratory hospitalization in 1st pre-index month	0.59%	0.0058	<0.001	0.16%	0.0016	1.000	0.16%	0.0016

Figure 3. Percentages of samples that had all-cause and acute respiratory hospitalizations in each month of the difference-in-difference analytic time period.

Table 2. Results of weighted logistic regressions using difference-in-difference specifications.

	All-cause hospitalization	Acute respiratory hospitalization
Post × vaccinated	0.59*** (0.55–0.63)	0.49*** (0.41–0.59)
Post	1.56*** (1.51–1.62)	1.70*** (1.55–1.86)
Vaccinated	1.00 (0.95–1.05)	1.01 (0.89–1.14)
E-value to explain away post × vaccinated [point estimate (upper bound)]^a	2.80 (2.55)	3.51 (2.80)

Results are presented as odds ratios (95% confidence intervals) unless otherwise specified.

^a The value for upper bound is the E-value required to generate an upper bound of a confidence interval that would be greater than 1.00.

*p<0.05.

**p<0.01.

***p<0.001.

Data availability

The data that support the findings of this study are available from the Shenzhen Medical Informatics Center. Restrictions apply to the availability of these data, which were used under a license for studies. Data are available from the authors upon reasonable requests for requestors who have permission from Shenzhen Medical Informatics Center.