Hospitalization and risk of death due to influenza among vaccinated and unvaccinated patients. A 5-year study in the northeastern Mexican population

ABSTRACT

In Mexico, seasonal influenza epidemics results in substantial mortality and burden to healthcare resources. The country`s health authority provides vaccination to children <5 years old; adults >60 years of age; those aged 5–60 years with risk factors. Inclusion of school-aged children and adults until 59 years of old with no risk factors in the vaccination program would be highly beneficial. A prospective cohort surveillance study was conducted between the influenza seasons of 2014–2015 and 2018–2019 at the Dr. José Eleuterio González University Hospital. The primary outcome was need for hospitalization in vaccinated and unvaccinated patients with ILI or seasonal influenza. Secondary outcomes included outpatient management, admission to the ICU, and mortality during hospitalization among vaccinated and unvaccinated participants. 361patients (37.44%) had a confirmed influenza diagnosis. Being vaccinated made it more probable to be treated as an outpatient (p = .0001). For unvaccinated patients, the risk for hospitalization (OR = 1.70), ICU admission (OR = 8.46) and in-hospital death (OR = 27.17) was higher. Fifty-two patients died due to complications related to seasonal influenza or ILI, and none of them were vaccinated. Most subjects were between 18 and 49 years old. Influenza vaccination significantly reduced hospitalization, need for ICU admission, and in-hospital mortality in a 5-year study from Monterrey, Mexico.

KEYWORDS:

• Influenza
• vaccination
• mortality
• immunization
• Mexico
• middle-income countries

Introduction

Influenza is a highly contagious and rapidly evolving respiratory virus that circulates globally on a seasonal basis around the world. It can cause death in high-risk groups and poses great burden on health systems.^1,2

There is global evidence of the benefits of vaccination against seasonal influenza in studies from various countries and regions, although there are few data from developing and middle-income countries.³ Such is the case of Mexico, where clinical, epidemiological, and mortality data in patients with seasonal influenza are limited, impacting directly in the generation of effective strategies for the prevention and control of this condition.^4–6

In Mexico, seasonal influenza epidemics results in substantial mortality and burden to health-care resources.^5,7 The country's health authority provides vaccination to children <5 years old; adults >60 years of age; those aged 5–60 years with risk factors such as asthma, heart diseases, immunodeficiencies, hemoglobinopathies, asplenia, chronic kidney disease, diabetes, obesity and arthritis; pregnant women and health-care workers.⁸

These target groups are eligible to receive the vaccine free of charge at any public health facility during the influenza season (October to April) and vaccine coverage rates within the groups included in the vaccination program go between 9.9% and 33.9%.⁹

Furthermore, only few reports of seasonal influenza vaccination coverage have been described in Mexican population. One of those included an 8-year cohort of 3,089 patients who died from causes attributable to seasonal influenza and only 2.1% of them were vaccinated.¹⁰

School-aged children and adults until 59 years-old with no risk factors can only access influenza vaccines in the private healthcare system. Inclusion of these groups in the vaccination program provided by health authorities would be highly beneficial since the former has one of the highest incidence rates and risk of transmission to the general population, and the latter exhibits the highest mortality rates attributable to seasonal influenza within an economically active population.^11,12

Therefore, robust evidence is needed to understand the epidemiology and impact of the influenza vaccine on the burden of disease in order to guide necessary adjustments and enhancements to national strategies, which will result in health and economic benefits.

We describe the characteristics, clinical outcomes and vaccination status of a northeastern Mexican group of patients with seasonal influenza or influenza like illness (ILI) through five years of observation.

Methods

Study design and study population

A prospective cohort surveillance study was conducted between the influenza seasons of 2014–2015 and 2018–2019 (five seasons, October through March) at the Dr. José Eleuterio González University Hospital in Mexico. We enrolled patients of all ages that sought medical care for influenza-like illness (ILI) defined as having at least the following signs and symptoms: acute cough, sore throat, and fever (>38°C) for less than 10 days at the time of medical consultation. Patients were evaluated by the emergency department medical team and the infectious diseases physicians who assessed the needed for hospitalization or outpatient treatment.

Seasonal influenza diagnosis

All patients who met the ILI definition underwent a nasopharyngeal swab for a rapid influenza antigen test (RAT), and selected patients for a real-time polymerase chain reaction (RT-PCR). RAT was performed with an immunochromatographic assay for the qualitative detection of influenza A and B nucleoprotein antigens in nasopharyngeal swab using the commercial kit Alere BinaxNOW Influenza A&B Card (Abbott Laboratories, IL, United States).

RT-PCR was performed with the method described by the CDC for qualitative detection of influenza virus type A or B in upper respiratory tract clinical specimens using an Eco Real-Time PCR System (Illumina Inc, CA, United States). Previously reported primers were used to detect the matrix (M) gene of influenza A and B virus, and hemagglutinin (HA) gene fragment of influenza A subtype H1N1 or H3N2.^13,14

Patients were classified as having confirmed seasonal influenza if at least one of the two diagnostic tests were positive. If both tests were negative, they were classified as having Influenza-like Illness (ILI). Those who had an inconclusive or invalid RT or RT-PCR, seasonal vaccination of less than 14 days before symptoms onset, or assessed outside the corresponding influenza season, were excluded from the study.

Data collection

The primary outcome was need for hospitalization in vaccinated and unvaccinated patients with ILI or seasonal influenza. Secondary outcomes included outpatient management, admission to the intensive care unit (ICU), and mortality during hospitalization among vaccinated and unvaccinated participants.

Demographic data, high-risk medical conditions, symptoms, the use of antiviral drugs before consultation were collected. All patients were actively asked about their vaccination status for the current season, written proof was solicited when feasible, and electronic registration was assessed from the hospital’s records to confirm vaccination status.

Statistical analysis

A descriptive statistical analysis was performed for seasonal influenza cases and ILI independently, including means for continuous variables and distribution for categorical variables. Multivariate logistic regression analysis was used to estimate the odds ratio (OR) with a 95% confidence interval (CI) to associate non-vaccination with risk of hospitalization, need for intensive care unit (ICU) admission, and mortality. The statistical analysis to contrast the proportions of the demographic and clinical characteristics of our population were performed with the chi-square test or Fisher’s exact test. The MedCalc Software (Version 19.1) and the IBM SPSS Statistics 25 Software were used.

Results

Patients included

From October 1, 2014, through April 30, 2019, 970 patients who met the study criteria were enrolled. Six were excluded from the primary assessment because they had been vaccinated less than 14 days before symptom onset, leaving 964 for the analysis.

All the patients included (ILI and seasonal influenza) were classified according to vaccination status. Patients between 18 and 49 years old (p = .0001) and females (p = .0178) were more frequent in the vaccinated group (Table 1). Also, vaccination was more frequent in the 2017–2018 and 2018–2019 influenza seasons (Table 1).

Table 1. Clinical and demographic characteristics of patients included (ILI and confirmed seasonal influenza) according to the vaccination status.

Characteristic	Vaccinated, n (%)	Unvaccinated, n (%)	p-Value
Total	302 (31.32)	662 (68.67)
Gender
Female	190 (62.91)	361 (54.53)	.0178
Age group (years)
0–8	17 (5.62)	65 (9.81)	.0415
9–17	10 (3.31)	42 (6.34)	.0751
18–49	246 (81.45)	400 (60.42)	.0001
50–64	12 (3.97)	82 (12.38)	.0001
>65	17 (5.62)	68 (10.27)	.0254
No data	0 (0.00)	5 (0.75)	NA
Influenza season
2014–2015	32 (10.59)	76 (11.48)	.7690
2015–2016	19 (6.29)	120 (18.12)	.0001
2016–2017	46 (15.23)	156 (23.56)	.0042
2017–2018	110 (36.42)	154 (23.26)	.0001
2018–2019	95 (31.45)	156 (23.56)	.0120
Antiviral treatment prior assessment
Amantadine	3 (0.99)	16 (2.41)	.2205
Oseltamivir	4 (1.32)	17 (2.56)	.3227
Zanamivir	0 (0.00)	1 (0.15)	NA
High-risk condition^a
Any	4 (8.69)	100 (33.89)	.0001
Asthma/pulmonary	8 (17.37)	117 (39.64)	.0001
Cardiovascular	9 (19.55)	106 (35.92)	.0001
Diabetes	4 (8.69)	82 (27.79)	.0001
Pregnancy	16 (34.78)	57 (19.32)	.0946
Renal	2 (4.34)	24 (8.13)	.0155
Treatment/outcome
Outpatient	256 (84.76)	367 (55.43)
Hospitalized	46 (15.23)	295 (44.56)	.0001
ICU admission	2 (4.34)	82 (27.79)	.0001
In-hospital mortality	0 (0.00)	52 (17.62)	.0001

^aPresence of a high-risk health condition is defined as the presence of ≥1 medical record-documented on the Tenth Revision of the International Statistical Classification of Diseases and Related Health Problems (ICD–10) high-risk code from October 1, 2014, to enrollment, as defined by the Advisory Committee on Immunization Practices (ACIP) guidance for conditions that increase risk for complications from seasonal influenza. The high-risk condition column met more than 100% because some patients had more than one high-risk health condition.

ILI: influenza-like illness; NA: not applicable.

Statistically significant p values are remarked in bold letters.

Patients aged 0–8 years (p = .0415), 50–65 years (p = .0001), >65 years (p = .0254), with asthma or another pulmonary condition, cardiovascular condition, diabetes and renal condition (p = .0001 each) were more frequent in the unvaccinated group. Furthermore, hospitalization, ICU admission, and in-hospital mortality were more frequent in unvaccinated patients (p = .0001 for each) (Table 1).

Influenza virus detected

A total of 361 patients (37.44%) had a confirmed seasonal influenza diagnosis by the RT. Among those tested by the RT-PCR, in 154 out of 292 influenza was detected, and 107 had both RT and RT-PCR positive tests. A positive RT-PCR was more frequently observed in unvaccinated patients for subtype AH3N2 (p = .0003) and AH1N1 (p = .044) (Table 2). The risk for having a confirmed seasonal influenza case was more than four-times higher in the group of unvaccinated patients (363 vs 152, OR = 4.76, 95% CI = 3.62–6.25, p < .0001).

Table 2. Total of confirmed seasonal influenza cases.

	Vaccinated (n = 302)	Unvaccinated (n = 662)	p-Value
Rapid antigen test positive*	109/302	252/662	.606
RT-PCR positive**	43/93	111/199	.162
Influenza A	9	23	.977
Subtype A H1N1	0	24	.0003
Subtype A H3N2	23	38	.044
Influenza B	11	26	.943
Rapid antigen test and RT-PCR positive	34	73	.764

*All patients were evaluated by the rapid antigen test.

**Selected patients were evaluated by the RT PCR.

RT-PCR: real-time polymerase chain reaction.

Statistically significant p values are remarked in bold letters.

Risk factors detected

In the univariate analysis, being vaccinated made it more probable to be treated as an outpatient (p = 0001). For unvaccinated patients, the risk for hospitalization (OR = 1.70, 95% CI = 1.16–2.49, p = .0059), ICU admission (OR = 8.46, 95% CI 2.01–35.74) and in-hospital death (OR = 27.17, 95% CI 1.65–446.62, p = .0208) was higher compared to vaccinated patients (Table 3). Multivariate analysis showed that hospitalization, ICU admission, and in-hospital death (p = .0001 each) were independent factors (Table 3).

Table 3. Univariate and multivariate analysis of hospitalization, ICU-admission, and in-hospital mortality risk among vaccinated and unvaccinated patients.

Treatment/Outcome	Vaccinated, n (%)	Unvaccinated, n (%)	Univariate analysis			Multivariate analysis
			OR	95% CI	p-Value	OR	95% CI	p-Value
Outpatient	256 (84.76)	367 (55.43)	0.22	0.16–0.32	0.0001	0.15	0.16–0.30	.0001
Inpatient	46 (15.23)	295 (44.56)	1.70	1.16–2.49	0.0059	1.49	3.13–6.30	.0001
Inpatient in the ICU	2 (4.34)	82 (27.79)	8.46	2.01–35.74	0.0036	2.03	3.70–15.96	.0001
In-hospital mortality	0 (0.00)	52 (17.62)	27.17	1.65–446.62	0.0208	18.73	0.00–0.00	.0001

CI: confidence interval; ICU: intensive care unit; OR: odds ratio.

Statistically significant p values are remarked in bold letters.

Fifty-two patients died during hospitalization due to complications related to confirmed seasonal influenza or ILI, and none of them were vaccinated (Table 4). Most of these were males (67.30%, 43.13%) between 18 and 49 years old, and half of them died during the influenza season 2016–2017.

Table 4. Characteristics of patients who died during the hospitalization due to seasonal influenza or ILI and their complications. All patients were unvaccinated.

Characteristic	n (%)
Total	52 (100.00)
Male gender	35 (67.30)
Age group (years)
0–8	1 (1.96)
9–17	1 (1.96)
18–49	22 (43.13)
50–64	16 (31.37)
>65	11 (21.56)
Influenza season
2014–2015	2 (3.92)
2015–2016	4 (7.84)
2016–2017	26 (50.98)
2017–2018	4 (7.84)
2018–2019	15 (29.41)
High-risk condition^a
Any	20 (39.21)
Diabetes	17 (33.33)
Obesity	16 (31.37)
Cardiovascular	16 (31.37)
Asthma/pulmonary	9 (17.64)
Seasonal influenza cases
Rapid antigen test	4 (7.84)
AH1N1	12 (23.52)
AH3N2	2 (3.92)
Virus A	5 (9.80)
Total	23 (45.08)
Total ILI	28 (54.90)

^aPresence of a high-risk health condition is defined as the presence of ≥1 medical record-documented on the ICD-10 high-risk code from October 1, 2014, to enrollment, as defined by the ACIP guidance for conditions that increase risk for complications from influenza. The high-risk condition column met more than 100% because some patients had more than one high-risk health condition. CI: confidence interval; ILI: influenza-like illness; OR: odds ratio.

Discussion

We describe the characteristics and vaccination status of a Mexican group of patients with seasonal influenza or ILI through five years of observation and found that unvaccinated patients had a risk more than 25 times higher of in-hospital death. Our data reaffirms the statement that vaccination against influenza is a public health strategy that aids in reducing influenza-related mortality.

Furthermore, there is evidence that seasonal influenza vaccination may prevent serious outcomes in hospitalized patients, specifically pneumonia, admission to ICU, and in-hospital mortality.^15–17 None of the patients who died in our study was immunized, similar to another analysis in Mexican population where only 2.1% of patients who died from causes attributable to influenza were vaccinated.¹⁰

Moreover, in our study, vaccination resulted effective regardless of gender or underlying high-risk conditions in almost all age groups, preventing severe cases of seasonal influenza or ILI, with only two patients in the vaccinated group needing ICU admission.

Additionally, other case-control studies among subjects with comorbidities have reported hospitalization reduction around 50% in vaccinated patients regardless the influenza season and more than 40% in high-risk patients.^15,18–20 Our study showed 29% less admissions in the vaccinated group representing and important target for reassessing immunization strategies.

Also, immunization of young adults may have potential economic and epidemiological benefits; a study showed an inversed relationship between vaccination coverage in non-elderly adults (18–64 years) and influenza-related illness in the elderly.²¹ In our analysis, more than 40% of fatal cases were between 18 and 49 years old. This age group not only represents the highest proportion of the economically active population but is the one that has less influenza vaccination coverage in the country.⁹

We acknowledge some limitations to our analysis, such as the fact that not all the patients were tested by both methods; thus, we may have missed confirmation of seasonal influenza cases; although all patients had at least a rapid influenza antigen test done, and all were treated with oseltamivir if the infectious disease team deemed it necessary even if the RT was negative.

Finally, influenza vaccination significantly reduced hospitalization, the need for ICU admission, and in-hospital mortality in this 5-year study in Monterrey, Mexico.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors reported there is no funding associated with the work featured in this article.