Table 4.

Effectiveness outcomes, mortality and hospitalisation, after seasonal influenza vaccination in COPD patients

Reference (Country)	Study design	Influenza season	n	Subgroup analysis	Effectiveness outcome	Comment
					Mortality
Schembri et al., 2009 [42] (UK)	Database study	1988–2006a	40,741		RR (95% CI) all-cause 0.59 (0.57–0.61)	Mortality rates were higher in years when the influenza vaccine did not include all strains circulating during that season (RR 1.19, 95% CI 1.13–1.25).
					RR (95% CI) death associated with respiratory event 0.63 (0.58–0.68)
					RR (95% CI) with respiratory event recorded as cause of death 0.63 (0.55–0.77)
Vila-Córcoles et al., 2008 [41] (Spain)	Prospective cohort study				HR (95% CI) all-cause	Mild-moderate influenza activity during the study. Mixed circulation of influenza A and B, with generally good matches with vaccine strains
		2002	1,298		0.48 (0.22–1.04)
		2003	1,233		0.79 (0.37–1.60)
		2004	1,149		0.95 (0.48–2.03)
		2005	1,050		0.87 (0.43–1.77)
		All seasons	-		0.76 (0.52–1.06)
Wang et al., [12] (Taiwan)	Retrospective population-based cohort study	2001	102,698 elderly		RR (95% CI) with COPD recorded as cause of death	Good match between epidemic strains and vaccine strains [55]
				High-riskb	0.45 (0.32–0.63)
				Low-risk	0.47 (0.26–0.83)
					Hospitalisation
Chen et al., 2013c [45] (Taiwan)	Retrospective, Database study	2000–2007a		Gender	HR (95% CI) due to heart failure	Good match between epidemic strains and vaccine strains except for 2001–02 (B mismatch), 2003–04 (A/H3N2 mismatch) [55]
			11,749	Female	0.48 (0.33–0.68)
			13,860	Male	0.42 (0.32–0.57)
				Age groups	HR (95% CI) due to heart failure
			13,218	≤44 years	3.96 (0.50–31.11)
			4,669	45–54 years	2.67 (0.95–7.50)
			3,455	55–64 years	0.65 (0.38–1.10)
			2,854	65–74 years	0.37 (0.26–0.52)
			1,413	≥75 years	0.38 (0.26–0.55)
			25,609	All subjects	0.44 (0.35–0.55)
Sung et al., 2014c [46] (Taiwan)	Retrospective, Database study	2000–2007a	7,722	≥ 55 years	HR (95% CI) due to acute coronary syndrome	As above
				Influenza season	0.45 (0.35–0.57)
				Non-influenza season	0.48 (0.37–0.62)
				All seasons	0.46 (0.39–0.55)
Menon et al., 2008 [43] (India, New Delhi)	Self-controlled case series	2004–2006		COPD severity	RR (p-value) post-vaccination year compared to pre-vaccination year Hospitalisation	Poorly matched seasons in 2005 and 2006 for influenza A strains (data for Kolkata) [56].
			32	Mild	0.33 (0.31)
			17	Moderate	0.5 (0.41)
			38	Severe	0.14 (0.15)
			87	Total	0.28 (0.02)
					ARI
			32	Mild	0.4 (0.26)
			17	Moderate	0.4 (0.21)
			38	Severe	0.25 (0.02)
			87	Total	0.33 (0.005)
Montserrat-Capdevila et al., 2014 [44] (Spain)	Retrospective cohort study	2011–2012		COPD severity	OR (95% CI) due to COPD exacerbations	Moderately severe influenza season. Moderate-to good matches for predominant circulating A/H1N1 and A/H3N2 viruses. Poor match for type B [14]
			1,099	Mild	0.083 (0.042–0.163)
			108	Moderate	0.133 (0.021–0.844)
			62	Severe	0.305 (0.024–3.813)
			54	Very severe	0.067 (0.009–0.505)
			1,323	Total	0.092 (0.052–0.165)

CI confidence interval, COPD chronic obstructive disease, HR hazard ratio, n number of subjects, OR odds ratio, RR relative risk, UK United Kingdom

Disease severity by Menon et al., 2008: mild: FEV₁ > 70% predicted; moderate: FEV₁ = 50–69% predicted; severe: FEV₁ < 50% predicted. Disease severity by Montserrat-Capdevila et al., 2014: mild: FEV₁ > 80% predicted; moderate: FEV₁ = 50–80% predicted; severe: FEV₁30–50% predicted; very severe: FEV₁ < 30% predicated

^aData from different influenza seasons were not separately analysed

^bHigh risk defined as recent hospital admission or chronic disease

^cIn the study of Sung et al., 2014 a subpopulation of Chen et al., 2014 is used