Introduction

High intensity interval training (HIIT), defined as exercise containing short intervals of vigorous physical activity (PA) with alternate periods of passive or active rest (Gibala et al. 2012), is a well-established training mode in healthy individuals, with a more recent expansion into clinical populations (Cassidy et al. 2017). Previous work has evidenced the effectiveness of HIIT in improving numerous health parameters (Ito 2019), including blood pressure regulation (Edwards et al. 2021, 2022), body composition (Wewege et al. 2017), and other important risk factors of cardiovascular disease (Batacan et al. 2017a). Furthermore, HIIT provides greater improvements in VO2 peak in a 6–8-week intervention when compared to a moderate intensity continuous training (MICT) intervention of the same length (Ito 2019). Importantly, although not to statistical significance, the recent ‘Generation 100’ study demonstrated a 37% and 49% lower all-cause mortality risk following HIIT compared to both the current physical activity guidelines and MICT programmes; respectively (Stensvold et al. 2020).

Currently 1 in 4 adults do not meet the current international PA guidelines (Bull et al. 2020), with lack of time being one of the most frequently reported barriers to exercise (Withall et al. 2011). Indeed, HIIT is a short, time-efficient mode of exercise, increasing its accessibility and potential for increased adherence across clinical and non-clinical populations (Vella et al. 2017). Previous HIIT research has largely focused on physiological adaptations, with primary and secondary disease prevention remaining the predominant interest (Ito 2019). However, as a measure concerned with a person’s perception of self-well-being, the ‘ultimate’ goal of patient care for health practitioners is to improve and maintain Quality of Life (QoL) (Jacobs 2009).

Prior research has shown promise regarding the role of HIIT in improving QoL, particularly in specific clinical groups (Lavín-Pérez et al. 2021; Reed et al. 2022); however, there are no large-scale pooled analyses investigating its effects across varying populations. The use of HIIT to improve QoL may provide an avenue of non-pharmaceutical treatment, which can be individually tailored using a variety of protocols and approaches (Cassidy et al. 2017). Therefore, the aim of this study was to establish the effectiveness of HIIT on physical, mental, and overall QoL in both clinical and non-clinical participant groups compared to a non-intervention control.

Methodology

This review was performed according to the Preferred reporting items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Page et al. 2021).

Search strategy

This work is a sub-study of a larger systematic review and meta-analysis registered to PROSPERO (CRD42022326576). As such, a systematic search of PubMed (MEDLINE), the Cochrane Library and Web of Science was conducted using the key words ‘high-intensity interval training’ and ‘HIIT’ to broadly identify all published HIIT trials of varying outcomes in which QoL data was later extracted from. Studies published before June 2022 were considered. Prior meta-analyses and their respective reference lists were also rigorously screened for any relevant HIIT and QoL research not identified within the broad search.

Screening, eligibility and data extraction

Following conduction of the search, Rayyan was used by two authors (MG and JE) who independently screened all studies for eligibility. All studies were screened by title and abstract for initial relevance. Those studies included following the initial inclusion phase were screened by full text. Subsequently, the QoL, study characteristic and intervention-specific data of all included studies was extracted via Microsoft Excel and any inconsistencies in data collection or confliction regarding study eligibility were discussed by the researchers until a consensus was reached. The opinion of a third researcher (JOD) was provided if necessary.

Studies were considered eligible if they reported the pre and post HIIT intervention changes in questionnaire QoL across any validated domain or scale with a corresponding non-intervention control group. Validated domains included SF-36, SF-12, HRQOL, WHOQOL, WEMBES, MLHFQ, MSqOL-54, KCCQ, EQ-5D, EQ-5D-5L and IBDQ. Participants were required to be ≥18 years of age with no predetermined limitations on health or disease state. The non-intervention control groups of the included papers were required to minimise confounders, with any dietary, counselling or exercise influence resulting in exclusion. Where applicable, studies that included an MICT group adjacent to HIIT and control were included. HIIT was defined as an exercise intervention performed in high intensity intervals that contained active or passive rest periods (Gibala et al. 2012). Exercise intervals were considered high intensity according to the EXPERT tool (Hansen et al. 2017) at intensity metrics falling within the categories of ‘High intensity, vigorous effort’ or ‘Very hard effort’.

Due to variation in QoL instruments and scoring across different studies, data extraction was standardised for consistency. If a specific physical QoL value was not given, then physical functioning/health was used depending on the questionnaire measuring QoL. If a sole mental QoL measure was not available, then a mental/emotional wellbeing value was used. If overall QoL was not available, then a general health value was used.

Methodological quality of studies

Study quality was measured via the ‘Tool for the assEssment of Study qualiTy and reporting in EXercise’ (TESTEX) (Smart et al. 2015). The TEXTEX scale is a 15-point scoring tool, with 5 points allocated for study quality and 10 points for reporting. Full TESTEX scores for the included studies can be found in the supplementary file (Table S1).

Statistical analysis

Pooled analyses of all studies were performed individually for overall, physical, and mental QoL. Owing to variances in QoL instruments used across the different studies, the standardised mean difference (SMD) was selected as the appropriate outcome measure. SMD effect thresholds were as follows: 0.2-0.5 small effect, 0.5-0.8 medium effect and above 0.8 as a large effect (Cohen 1988). The SMD between the HIIT group and non-intervention control group was measured for each QoL category. Separate secondary analyses were also performed comparing HIIT and MICT. Data was synthesised using Comprehensive Meta-Analysis (Comprehensive Meta-Analysis Version 3, Biostat, Englewood, NJ, USA). Statistical heterogeneity was performed alongside the pooled analysis and reported as the I2 statistic. If the I2 statistic is >40% it was considered significant. Past the I2 threshold, Eggers Regression Test (Egger et al. 1997) was systematically planned to create a funnel plot looking for asymmetry related to potential publication bias. Random effects analysis was conducted when interstudy variability was confirmed through significant heterogeneity. Pooled analysis results were considered significant if the P value was <0.05 and the Z-Value >2.

Results

Search selection

Figure 1 details the PRISMA flowchart. 4033 papers were identified through the systematic search. Following duplicate removal, 2508 papers remained, which then underwent abstract and title screening. 258 papers were full text screened against the inclusion criteria with 9 studies included from the search. Subsequent screening of previous meta-analyses and their respective reference lists identified a further 18 papers, of which 13 were included in the final pool. Ultimately, 22 studies constituting 24 effect sizes were analysed.

Fig. 1
figure 1

PRISMA flowchart of study selection

Full size image

Study characteristics

1322 individuals participated in the included studies. The included studies contain a wide variety of populations, QoL instruments and training characteristics, as seen in Tables 1 and 2. All studies except one (Burn et al. 2021) were randomised trials. 14 studies compared the effects of HIIT on physical QoL, 13 on mental QoL and 17 studies on overall QoL. Five studies additionally compared HIIT vs MICT, constituting the secondary analysis of this work. As demonstrated in the study TESTEX scoring, common limitations include limited activity monitoring in the control groups, no blinding of assessors and participants, and a lack of intention to treat analysis.

Table 1 Study Characteristics
Full size table
Table 2 High-intensity interval training characteristics
Full size table

HIIT and overall, physical and mental QoL

Figures 2, 3 and 4 detail the overall, physical and mental QoL SMD between the HIIT and non-intervention control groups, respectively. There was a statistically significant ‘medium’ improvement in overall QoL in HIIT compared to the control group (SMD: 0.554, CI= 0.210-0.898, p=0.002). There was a statistically significant ‘small’ improvement in physical QoL following HIIT compared to the control group (SMD: 0.405, CI= 0.110- 0.700, p=0.007). Finally, there was also a significant ‘small’ improvement in mental QoL in HIIT compared to the control group (SMD: 0.473, CI= 0.043-0.902, p=0.031).

Fig. 2
figure 2

Forest Plot representing HIIT and Overall QoL, SMD with 95% CI

Full size image
Fig. 3
figure 3

Forest Plot representing HIIT and Physical QoL, SMD with 95% CI

Full size image
Fig. 4
figure 4

Forest Plot representing HIIT and Mental QoL, SMD with 95% CI

Full size image

HIIT Vs MICT on overall QoL

The secondary analysis of 5 studies comparing HIIT against MICT demonstrated no significant difference in improvement between the two modes (SMD= -0.094, CI= -0.506-0.318, p=0.655).

Publication bias and heterogeneity

All analyses demonstrated significant statistical heterogeneity (I2: overall QoL= 70.319%, physical QoL= 83.605%, mental QoL= 78.737% and HIIT vs MICT= 51.067%). Eggers Regression test demonstrated funnel plot asymmetry and therefore evidence of publication bias for the physical QoL domain (p=0.0085, Figure S1). Overall QoL, mental QoL, and HIIT vs MICT showed no evidence of publication bias.

Discussion

This work aimed to measure the effects of HIIT on physical, mental, and overall QoL in both clinical and non-clinical populations. The findings of this meta-analysis demonstrate statistically significant improvements in all domains of QoL across different instruments. As determined by Cohen’s SMD effect thresholds (Cohen 1988), the improvements observed in this work are of a ‘small’ to ‘medium’ effect size. Specifically, physical QoL and mental QoL individually produced small effect sizes, while overall QoL elicited a medium effect size. Furthermore, HIIT appears as effective as MICT in improving overall QoL, offering a more time-efficient exercise option. As the largest-scale analysis to-date, these findings support earlier preliminary evidence in the potential utility of HIIT in improving QoL across different populations.

Several reviews and large-scale trials have supported the effectiveness of HIIT for improving physical, mental and overall domains of QoL in various populations, with particular research interest on its role in clinical groups such as those with cancer (Lavín-Pérez et al. 2021), atrial fibrillation (Reed et al. 2022) and heart transplant recipients (Yu et al. 2022). Furthermore, the capacity for HIIT to produce QoL improvements similar to that of MICT is also well-supported in the broad literature, with several meta-analyses demonstrating no significant differences between HIIT and MICT on QoL in clinical populations (Gomes-Neto et al. 2017, 2018; Lavín-Pérez et al. 2021; Anjos et al. 2022). Given the impairment in QoL in patients with debilitating chronic diseases, these findings may be of clinical importance. This is particularly true considering the poor adoption and adherences rates to traditionally recommended MICT in these populations (Argent et al. 2018), highlighting the potential utility of HIIT as an alternative exercise intervention with similar QoL-enhancing capabilities.

In understanding the mechanistic underpinnings of QoL improvements, it is important to consider the probable confounding and interdependence between all domains of QoL, with improvements in any given single domain likely translating into changes in all domains (Post 2014). Physically, the well-established physiological adaptations frequently seen following HIIT may translate into improvements, particularly in older groups and/or those suffering from debilitating chronic conditions, in the capacity to complete more activities of daily living. In clinical populations such as heart failure, QoL is profoundly impaired (in both preserved and reduced ejection fraction) (Hobbs et al. 2002; Stewart et al. 2010), largely owing to patient symptoms and limited functional capacity. Therefore, the frequently described improvements in cardiometabolic health and peak VO2 (Batacan et al. 2017b) following HIIT may be of particular importance regarding the QoL of such patient groups, as has been demonstrated in previous exercise training studies (Edwards and O’Driscoll 2022).

Regarding mental QoL, these findings support that of a recent large-scale systematic review and meta-analysis (Martland et al. 2022) in clinical and non-clinical populations. This work from Martland et al. (2022) reported significant improvements in mental well-being, depression severity and perceived stress, with suggestions of sleep and psychological distress improvements. Alike the current analysis, Martland et al. (2022) found these improvements to be of a small to medium effect size. Combined with the findings of the present study, HIIT certainly appears an effective strategy to elicit improvements in psychological well-being. However, further research into populations with psychological disorders is warranted to establish the transferability of this data into specific clinical sub-groups.

Limitations

We found significant statistical heterogeneity across all analyses in this work. This is likely attributable to inter-study methodological differences such as the utilised QoL instruments and HIIT protocols, as well as wide population variation in the inclusion of clinical and non-clinical populations. We subsequently performed random-effects analyses in an attempt to account for this, and explored the Eggers regression tests for publication bias. We did indeed find publication bias for the physical QoL domain which should be appropriately considered in the interpretation of these results. Furthermore, some studies (Madssen et al. 2014; Malmo et al. 2016; Ellingsen et al. 2017; Mokhtarzade et al. 2017; Romain et al. 2019; Atan and Karavelioğlu 2020; Burn et al. 2021; Mueller et al. 2021; Ochi et al. 2022; Woodfield et al. 2022) measured QoL as a secondary outcome so may not have been appropriately powered.

Future implications

Only 5 studies included both HIIT and MICT QoL data. As such, future research is needed to assess the efficacy of HIIT compared to the traditionally recommended MICT. Additionally, larger-scale homogenous research is needed in specific populations before these findings can be extrapolated to specified clinical and non-clinical groups. Further research into varying HIIT protocols, with specific comparative data between sprint interval training and aerobic interval training protocols are needed to truly discern optimal HIIT prescription practices.

Conclusion

HIIT produces statistically significant improvements in physical, mental and overall QoL at a small to medium effect size across a range of QoL instruments in clinical and non-clinical populations. Furthermore, HIIT appears as effective as MICT in improving overall QoL, offering a more time-efficient non-pharmacological option. As the largest-scale analysis to-date, these findings support earlier preliminary evidence regarding the potential utility of HIIT in improving QoL across different populations.