Background Work-life balance (WLB) is the extent to which individual’s multiple life roles and demands carry over between each role. WLB can be divided into work interference with personal life (WIPL) and personal life interference with work (PLIW). This study aimed to investigate longitudinal associations between WIPL, PLIW and work ability outcomes.

Methods In this cohort study, 224 employees in the energy and water sector in Sweden were followed-up over 2 years. Three questions derived from the Work Ability Index were used for measuring work ability outcome: current work ability compared with lifetime best; work ability regarding physical; and mental demands. Logistic regression models were used to analyse longitudinal associations between work ability and WIPL and WIPL respectively, controlling for workplace (company), position at work, experience of leadership quality, demographics, and work ability.

Results Work ability compared to lifetime best were associated with WIPL in the adjusted logistic regression models (odds ratio (OR) 1.77, 95% confidence interval (CI) 1.15–2.73), and PLIW (OR 3.34, 95% CI 1.66–6.74). Work ability regarding physical demands was associated with WIPL (OR 1.60, 95% CI 1.07–2.40). Work ability regarding mental demands was associated with WIPL (OR 1.59, 95% CI 1.03–2.44) and PLIW (OR 2.88, 95% CI 1.31–6.32).

Conclusion In this two-year longitudinal study, lower WIPL predicted good/excellent overall work ability compared with lifetime best, higher work ability regarding physical and mental demands, and lower PLIW predicted good/excellent overall work ability compared with lifetime best and higher work ability regarding and mental demands.