This nationwide prospective observational study conducted among symptomatic US adult outpatients with influenza showed that influenza infection was associated with substantial declines in perceived quality of life, work productivity, and nonwork activity levels. HRQoL was impaired over approximately 2 weeks, with lower EQ-VAS scores (through day 14) and lower EQ-UI scores (through day 10). Early declines were clinically meaningful, but by day 14 (EQ-VAS) and day 10 (EQ-UI), effect sizes had diminished to < 0.20, indicating negligible clinical impact. Work productivity and nonwork activity levels were also impaired due to influenza, with adults missing more days of work (absenteeism) than prior to infection over the first week (through day 7). They also experienced higher presenteeism, greater overall work impairment, and more activity impairment, persisting through day 14. Adults at high risk for influenza also experienced decreases in HRQoL, work productivity, and activity levels that were consistent with patterns observed in the overall sample. However, their HRQoL as assessed by EQ-VAS and EQ-UI persisted, remaining below baseline and meeting thresholds for clinical significance slightly longer.

When referring to the companion-symptom burden analysis of this cohort [18], HRQoL and WPAI changes aligned with the symptom trajectory reported there. That analysis found that severe symptoms returned to baseline by day 6, while mild-to-moderate symptoms persisted for up to 4 weeks. This aligns with the finding in the present study that absenteeism returned to baseline by day 14, but presenteeism, work impairment, activity impairment, and HRQoL deficits extended through day 14. These findings reinforce that acute influenza symptoms are a primary driver of short-term declines in quality of life and productivity. Peak impairments in EQ-VAS, EQ-UI, absenteeism, presenteeism, overall work impairment, and activity impairment occurred during the period of highest severe symptom burden, and improvements coincided with return to baseline of severe symptoms. It is likely that lingering less severe symptoms contributed to continued functional limitations and delayed recovery through day 14. These patterns are consistent with the clinical course of influenza, where functional recovery often precedes complete symptom resolution [33, 34].

Prior research on respiratory infections similarly shows that individuals may return to work while still experiencing residual symptoms [33, 34], and that HRQoL impairment during recovery is often driven by these lingering complaints [35, 36]. Notably, in the companion symptom burden analysis, a considerable proportion of participants experienced mild cough, stuffy or runny nose, and fatigue at 2 weeks and beyond [18]. It is likely that these symptoms continued to affect daily functioning and perceived health status despite improvements in acute illness.

For HRQoL, prior to infection, high mean scores on EQ-VAS and EQ-UI, signified that patients perceived themselves to be in good overall health, yet declines were observed postinfection. In the current study EQ-VAS on day 1 was 63.2, whereas prior US research reported EQ-VAS score of 48.5 at initial assessment post influenza infection [37], indicating a less severe decline in HRQoL. This may be partially attributable to different characteristics (fewer with underlying conditions), or a more recent season (2024/25 versus 2011/12–2019/20 seasons).

Also, this study included participants within 4 days of illness onset compared with 3 days in the prior study, potentially capturing individuals at a later stage of illness when symptoms may have already begun to resolve [37]. However, similar declines in HRQoL have been reported in England and Spain [7, 9]. In England, EQ-VAS declined from 84.0 preinfection to 43.0 at the worst day of illness, and EQ-UI from 0.92 to 0.36 [9]. In the current study, EQ-VAS declined from 89.6 at baseline to 63.2 on day 1, and EQ-UI from 0.95 to 0.75. Differences in magnitude likely reflect timing, as this study assessed burden on day 1 of study participation capturing participants’ current health status rather than asking them to recall their worst point since illness onset, whereas the English study captured the worst day.

In Spain, EQ-UI scores declined from 0.93 preinfluenza to 0.50 during infection [7], compared with 0.95 to 0.75 on day 1 in the current study’s sample. Differences are potentially due to season (2024/25 versus 2009 H1N1 pandemic season), age distribution (> 18 years versus > 8 years), and fewer comorbidities (27.5% versus 37.3%). Together, these findings support that influenza is associated with a temporal impact on HRQoL, although the magnitude may vary across populations, seasons, time since infection, and healthcare contexts.

For WPAI, these findings are consistent with previous influenza studies, which show adults experienced greater absence from work and reduced productivity while working [33]. Previous US research analyzed over 1200 employed adults with laboratory-confirmed influenza across four influenza seasons and found that influenza was associated with workplace productivity loss (mean loss of 67–74%) [12]. Nonetheless, slightly higher mean percent changes in work productivity loss were observed on day 1 in this study (Tables 3 and Supplementary Table S7 in the electronic supplementary material): 76.3% in the overall influenza sample, 78.6% in high-risk influenza adults, 78.6% in high-risk COVID-19 adults, and 78.8% in high-risk influenza adults < 65 years (Supplementary Table S7 in the electronic supplementary material). This study extends prior analyses by including four repeated assessments over 1 month, compared with 7–17 days of follow-up in earlier research. This longer follow-up allowed us to estimate recovery timing, showing return to baseline by day 14 or week 4 depending on subgroup. Compared with overall sample, clinically high-risk groups (high-risk subgroup, high-risk subgroup < 65 years, and influenza sample at high-risk for COVID-19) experienced longer meaningful impairment in HRQoL as measured using EQ-UI scores (day 10 versus day 7), indicating slower return to preinfection health state, as reported in the results (Tables 2 and Supplementary Table S6 in the electronic supplementary material). These findings highlight the need for targeted interventions to reduce the humanistic burden in vulnerable populations.

This study has several limitations. All data were self-reported and subject to recall and selection bias with some missingness (survey drop-off). While participant characteristics remained stable through week 4 (Supplementary Table S8 in the electronic supplementary material), findings may not generalize to rural populations, more vulnerable communities, or those seeking care in other healthcare settings. The sample was predominantly White adults from certain US regions who self-selected to get tested at ambulatory care clinic sites in a US national retail pharmacy chain, possibly reflecting greater health consciousness or more severe symptoms. Attrition may have led to underestimation if participants with greater impairment were more likely to drop out (Supplementary Table S9 in the electronic supplementary material), potentially missing data from those experiencing the greatest WPAI and HRQoL impairment. Hence the estimates of recovery may be biased, toward faster resolution. Additionally, requiring English literacy and email access may introduce selection bias by excluding individuals with limited English proficiency and digital access, reducing the sociodemographic representativeness of the sample. Furthermore, compensation may have introduced bias if incentives influenced participants’ motivation or response quality. Notably, because symptom onset could occur up to 4 days before testing, the earliest phase of illness may be missed, potentially underestimating the impact on HRQoL and WPAI. Relatedly, because preinfluenza HRQoL and WPAI were assessed retrospectively at day 1, participants may over- or under- estimate the preillness status due to recall bias and response shift. There may be potential misclassification of influenza status, since the gold standard polymerase chain reaction (PCR) laboratory testing was not used. Influenza virus subtyping was unavailable, limiting assessment of whether differences in symptom severity and related impacts on quality of life, productivity, and activity impairment varied by subtype. Although vaccine-by-time interaction was adjusted for in the models, moderate influenza vaccine effectiveness during the 2024–2025 season [38], potentially influenced by vaccine-strain mismatch, may have reduced the magnitude of differences observed between vaccinated and unvaccinated participants in symptom severity and related quality of life, work productivity and activity impairment. While participants with test-confirmed influenza and COVID-19 coinfection were excluded, coinfections with other respiratory viruses were not assessed and may have influenced symptom severity and related impacts on quality of life, work productivity, and activity impairment.

Despite these limitations, the study has important strengths. Participants in a large nationwide outpatient cohort were enrolled within 48 h of a positive influenza test and prospectively followed, minimizing recall bias. Validated instruments (EQ-5D-5L, WPAI:GH) were used to assess HRQoL and WPAI. Repeated measurements over 4 weeks provided detailed insight into the duration and resolution of influenza-related impairment. By capturing both HRQoL and WPAI impacts longitudinally, it offers new insights into areas that have been previously underexplored and helps fill a critical void in understanding the broader burden of influenza.