Individualized Endurance Training Based on Recovery and Training Status in Recreational Runners

Olli-Pekka Nuuttila; Ari Nummela; Elisa Korhonen; Keijo Häkkinen; Heikki Kyröläinen


Med Sci Sports Exerc. 2022;54(10):1690-1701. 

In This Article

Abstract and Introduction


Purpose: Long-term development of endurance performance requires a proper balance between strain and recovery. Because responses and adaptations to training are highly individual, this study examined whether individually adjusted endurance training based on recovery and training status would lead to greater adaptations compared with a predefined program.

Methods: Recreational runners were divided into predefined (PD; n = 14) or individualized (IND; n = 16) training groups. In IND, the training load was decreased, maintained, or increased twice a week based on nocturnal heart rate variability, perceived recovery, and heart rate–running speed index. Both groups performed 3-wk preparatory, 6-wk volume, and 6-wk interval periods. Incremental treadmill tests and 10-km running tests were performed before the preparatory period (T 0) and after the preparatory (T 1), volume (T 2), and interval (T 3) periods. The magnitude of training adaptations was defined based on the coefficient of variation between T 0 and T 1 tests (high >2×, low <0.5×).

Results: Both groups improved (P < 0.01) their maximal treadmill speed and 10-km time from T 1 to T 3. The change in the 10-km time was greater in IND compared with PD (−6.2% ± 2.8% vs −2.9% ± 2.4%, P = 0.002). In addition, IND had more high responders (50% vs 29%) and fewer low responders (0% vs 21%) compared with PD in the change of maximal treadmill speed and 10-km performance (81% vs 23% and 13% vs 23%), respectively.

Conclusions: PD and IND induced positive training adaptations, but the individualized training seemed more beneficial in endurance performance. Moreover, IND increased the likelihood of high response and decreased the occurrence of low response to endurance training.


Successful endurance training requires a proper balance between training load and recovery. Although adequate training stimulus is necessary to induce favorable adaptations, inadequate recovery between training sessions and periods may lead to excessive fatigue, and if the imbalance is extended, even to nonfunctional overreaching or overtraining.[1] It has been observed that acute responses and recovery kinetics to similar training sessions[2–4] as well as adaptations to training periods[5–7] vary between individuals, and processes related to adaptation could be affected by multiple factors not connected with actual training, such as nutrition,[8] sleep,[9] or psychological stress.[10] Therefore, monitoring both training and recovery could help to take individual differences into account and in this way provide useful information for the estimation of proper training load in each case.[11]

The evolution of wearable technology has produced more options for the monitoring of training and recovery, which in turn makes individual training approaches more feasible. Lately, heart rate variability (HRV)–guided training has been utilized in various populations, leading to more beneficial training effects compared with predefined training in untrained,[12] recreationally trained,[13–15] and well-trained[16,17] participants. The assumption in HRV is that because it reflects the cardiac parasympathetic nervous system activity, it would also relate to current readiness to adapt to training stimulus.[18] The basic idea in all studies utilizing the HRV-guided approach has been similar—training intensity has been modified based on changes in daily recorded resting HRV with respect to the individually defined reference range. Furthermore, values below and above the normal range have been regarded as a sign of an abnormal state, and only low-intensity training has been prescribed until HRV has reached the individual reference value.[14,16,17] Interestingly, none of the previous studies have tried to manipulate training volume based on HRV, although it is an important variable in the endurance training prescription.[19]

Despite the fact that HRV-guided training has induced some promising results, a single marker could not establish all aspects critical to recovery. Although HRV mainly reflects cardiac autonomic nervous system activity and cardiovascular homeostasis, aspects such as muscle tissue repair or muscle glycogen repletion may not necessarily be aligned with the parasympathetic reactivation.[18] Indeed, neuromuscular and perceptual recovery has differed from the pattern of HRV in several studies.[3,4,18,20] It can also be argued that training adaptation and HRV (or its responses) may not be as directly associated as sometimes it has been assumed,[21] especially, when taking into consideration the challenging interpretation of HRV after intensified training[22,23] and the possible influence of plasma volume expansion.[18] Therefore, supplementary monitoring methods providing information on perceived fatigue and musculoskeletal strain could help to gain a more comprehensive picture of the recovery status. To the best of our knowledge, only one previous study has considered multiple variables in the training decision scheme by analyzing the rating of perceived exertion, the ability to reach target heart rate (HR), and the HR recovery from a submaximal cycling test.[24] Although it seems obvious that combining both objective and subjective markers would provide the best quality for monitoring, there certainly exists a lack of research on how to implement such an approach in practice.

To investigate the effectiveness of individualized training volume and intensity, the present study compared the individually adjusted training prescription based on nocturnal HRV, perceived recovery, and estimated running performance to the predefined training program in recreationally endurance-trained males and females. We hypothesized that individualized training would induce greater training adaptations in maximal running performance compared with predefined training and decrease the likelihood of low response.