Results

A total of 8 418 163 individuals aged 65 and over were identified as having received at least one influenza vaccine between 1 September 2022, and 31 March 2023, in France. A total of 315 239 nursing home residents vaccinated against influenza were identified, comprising 123 720 HD recipients and 191 519 SD recipients. The 1:1 matched cohorts yielded 76 858 HD recipients (representing 63.4% of the initial number of HD recipients) who were matched with 76 858 SD recipients (Figure 1). The remaining 36.6% of HD recipients could not be matched primarily because insufficient SD recipients were available under the exact matching constraints (sex, age group, region and vaccination week).

Figure 1: Study population selection process. Individuals with HD were matched to individuals with SD using a 1:1 ratio propensity score (PS) with an exact constraint on sex, age groups (65–74, 75–84, 85 and over), geographical region of residence and week of dispensing of the vaccine. The PS included socio-demographic variables, proxies for health behaviours and comorbidities. DOM-TOM: French overseas territories; HD: High-dose; FDep: French deprivation index; IV: Influenza vaccine; NH: Nursing home; SD: Standard-dose.

Individuals’ characteristics were similar after matching (Table 1). The mean age of the subjects was 87.8 (7.7) years in the HD group and 88.0 (7.8) years in the SD group, and 75.3% were women in both groups. Individuals’ characteristics prior to matching are presented in Supplementary Table S4. The most prevalent comorbidities and medical history were cardiovascular diseases (43.9% in HD vs. 43.0% in SD), dementia (41.9% vs. 42.6%) and undernourishment or a history of undernourishment (37.0% vs. 36.6%). The matching performances for the 1:1 ratio are illustrated in Supplementary Figures S1 and S2 and variable ratio matching characteristics and performances in Supplementary Tables S5–S6 and Figure S3.

Characteristics | HD | SD
Number of individuals | 76 858 | 76 858
Age, years |  | 
Mean (SD) | 87.76 (7.7) | 88.02 (7.8)
Median (IQR) | 89.00 (9.0) | 89.00 (9.0)
65–70 years, n (%) | 2 679 (3.5) | 2 507 (3.3)
71–75 years, n (%) | 4 107 (5.3) | 4 176 (5.4)
76–80 years, n (%) | 6 167 (8.0) | 5 925 (7.7)
81–85 years, n (%) | 11 600 (15.1) | 11 945 (15.5)
86–90 years, n (%) | 20 566 (26.8) | 19 177 (25.0)
91–95 years, n (%) | 21 245 (27.6) | 21 285 (27.7)
96–100 years, n (%) | 9 146 (11.9) | 10 230 (13.3)
> 100 years, n (%) | 1 348 (1.8) | 1 613 (2.1)
Women, n (%) | 57 832 (75.3) | 57 832 (75.3)
Reasons for end of follow-up, n (%) |  | 
Admission into medico-social housing (other than NH) | 18 (0.0) | 14 (0.0)
Exit from NH | 1 818 (2.4) | 2 043 (2.7)
Death | 12 333 (16.1) | 12 316 (16.0)
End of follow-up | 62 689 (81.6) | 62 485 (81.3)
Health care seeking behaviours proxy |  | 
All-cause hospitalisation in the past 12 months, mean number per person (SD) | 0.15 (0.0) | 0.13 (0.0)
Influenza vaccination during the previous season, n (%) | 61 839 (80.5) | 66 846 (87.0)
COVID-19 vaccination at index date, n (%) | 73 927 (96.2) | 74 091 (96.4)
Pneumococcal vaccination in the previous 5 years, n (%) | 14 325 (18.6) | 13 231 (17.2)
Comorbidities, n (%) |  | 
Diabetes | 13 139 (17.1) | 13 233 (17.2)
Obesity and/or history of obesity surgery | 5 958 (7.8) | 6 129 (8.0)
Undernourishment or history of undernourishment | 28 427 (37.0) | 28 105 (36.6)
COPD/Asthma | 8 168 (10.6) | 8 257 (10.7)
Dementia | 32 172 (41.9) | 32 768 (42.6)
Cardiovascular diseases | 33 715 (43.9) | 33 036 (43.0)
Immunocompromised subjects | 11 964 (15.6) | 11 441 (14.9)
Chronic liver disease | 1 328 (1.7) | 1 120 (1.5)
Terminal chronic kidney failure | 221 (0.3) | 214 (0.3)
Number of comorbidities, n (%) |  | 
None | 13 392 (17.4) | 13 502 (17.6)
1 | 27 727 (36.1) | 27 690 (36.0)
2 | 20 982 (27.3) | 21 112 (27.5)
3 | 9 728 (12.7) | 9 602 (12.5)
4 | 3 569 (4.6) | 3 502 (4.6)
5 | 1 094 (1.4) | 1 101 (1.4)
6 or more | 366 (0.5) | 349 (0.5)

Influenza-related hospitalisation rates for the main analysis (Primary diagnosis excluding hospitalisations with COVID-19) were 302 (253–360) per 100 000 person-years (PY) in HD recipients and 438 (379–507) per 100 000 PY in SD recipients, which translated into a rVE (95% CI) of 30.8 [11.8–45.7] (P-value = .0029) (Table 2 and number of hospitalisation in Supplementary Table S7).

Influenza hospitalisation rate per 100 000 PY (95%CI) | IRR (95%CI) | rVE (95%CI) | P-value
Main analysis: Primary diagnosis excluding hospitalisations with COVID-19 |  | 
HD | 302 (253–360) | 0.69 [0.54–0.88] | 30.8 [11.8–45.7] | .0029
SD | 438 (379–507) |  |  | 
Sensitivity analysis: Primary or non-primary discharge codes, excluding hospitalisations with COVID-19
HD | 482 (420–554) | 0.73 [0.60–0.89] | 27.2 [11.4–40.3] | .0016
SD | 660 (586–743) |  |  | 
Sensitivity analysis: Primary diagnosis including hospitalisations with COVID-19
HD | 307 (258–366) | 0.69 [0.54–0.88] | 30.8 [12.0–45.6] | .0027
SD | 446 (385–515) |  |  | 
Sensitivity analysis: Primary or non-primary discharge codes, including hospitalisations with COVID-19
HD | 504 (440–578) | 0.73 [0.60–0.89] | 26.9 [11.3–39.8] | .0015
SD | 687 (611–772) |  |  | 

Including hospitalisations with COVID-19 diagnosis induced a rVE (95%CI) of 30.8 [12.0–45.6] (P-value = .0027). When using a broader definition considering influenza coded in non-primary discharge position, a rVE (95%CI) of 27.2 [11.4–40.3] (P-value = .0016) of HD vs SD was observed. The inclusion of hospitalisations with COVID-19 translated into a rVE (95%CI) of 26.9 [11.3–39.8] (P-value = .0015) (Table 2). The results of the variable matching ratio yielded a comparable effect size, which is available in Supplementary Table S10.

No significant difference in relative vaccine effectiveness was observed against pneumonia and pneumonia/influenza hospitalisations (Tables 3 and 4). The incidence of pneumonia and P/I-related hospitalisations was higher in recipients of HD compared to those receiving SD, when analysed in the cohort matched with a variable matching ratio (Supplementary Tables S11 and S12).

Pneumonia hospitalisation rate per 100 000 PY (95%CI) | IRR (95%CI) | rVE (95%CI) | P-value
Main analysis: Primary diagnosis excluding hospitalisations with COVID-19 |  | 
HD | 3 771 (3 588–3 964) | 1.06 [0.98–1.15] | −6.4 [−15.2–1.8] | .1292
SD | 3 550 (3 373–3 737) |  |  | 
Sensitivity analysis: Primary or non-primary discharge codes, excluding hospitalisations with COVID-19
HD | 6 280 (6 042–6 527) | 1.01 [0.95–1.08] | −1.1 [−7.6–5.1] | .7440
SD | 6 207 (5 970–6 452) |  |  | 
Sensitivity analysis: Primary diagnosis including hospitalisations with COVID-19
HD | 3 832 (3 647-4 026) | 1.07 [0.99–1.16] | −6.9 [−15.7–1.3] | .1000
SD | 3 592 (3 413-3 780) |  |  | 
Sensitivity analysis: Primary or non-primary discharge codes, including hospitalisations with COVID-19
HD | 6 867 (6 618-7 123) | 1.01 [0.95–1.07] | −0.8 [−7.1–5.2] | .8019
SD | 6 806 (6 558-7 063) |  |  | 

 | P/I hospitalisation rate per 100 000 PY (95%CI) | IRR (95%CI) | rVE (95%CI) | P-value
Main analysis: Primary diagnosis excluding hospitalisations with COVID-19 |  | 
HD | 4 073 (3 882–4 273) | 1.02 [0.95–1.10] | −2.30 [−10.4–5.2] | .5580
SD | 3 988 (3 800–4 186) |  |  | 
Sensitivity analysis: Primary or non-primary discharge codes, excluding hospitalisations with COVID-19
HD | 6 687 (6 441–6 942) | 0.99 [0.93–1.05] | 1.3 [−4.9–7.1] | .6731
SD | 6 762 (6 515–7 018) |  |  | 
Sensitivity analysis: Primary diagnosis including hospitalisations with COVID-19
HD | 4 139 (3 947–4 340) | 1.03 [0.95–1.11] | −2.7 [−10.8–4.7] | .4845
SD | 4 037 (3 847–4 236) |  |  | 
Sensitivity analysis: Primary or non-primary discharge codes, including hospitalisations with COVID-19
HD | 7 371 (7 113–7 639) | 0.98 [0.93–1.04] | 1.8 [−4.2–7.4] | .5574
SD | 7 492 (7 232–7 762) |  |  | 

The NCO yielded IRR (95%CI) values of 1.06 (0.98–1.15), 1.05 (0.94–1.18) and 1.00 (0.83–1.22) for UTI, cataract surgery and erysipelas hospitalisations, respectively.