Abstract Introduction Intensive care unit-acquired weakness (ICU-AW) is a significant problem. There is currently widespread variability in the methods used for manual muscle testing and handgrip dynamometry (HGD) to diagnose ICU-AW. This study was conducted in two parts. The aims of this study were: to determine the inter-rater reliability and agreement of manual muscle strength testing using both isometric and through-range techniques using the Medical Research Council sum score and a new four-point scale, and to examine the validity of HGD and determine a cutoff score for the diagnosis of ICU-AW for the new four-point scale. Methods Part one involved evaluation of muscle strength by two physical therapists in 29 patients ventilated >48 hours. Manual strength testing was performed by both physical therapists using two techniques: isometric and through range; and two scoring systems: traditional six-point Medical Research Council scale and a new collapsed four-point scale. Part two involved assessment of handgrip strength conducted on 60 patients. A cutoff score for ICU-AW was identified for the new four-point scoring system. Results The incidence of ICU-AW was 42% (n = 25/60) in this study (based on HGD). In part one the highest reliability and agreement was observed for the isometric technique using the four-point scale (intraclass correlation coefficient = 0.90: kappa = 0.72 respectively). Differences existed between isometric and through-range scores (mean difference = 1.76 points, P = 0.005). In part two, HGD had a sensitivity of 0.88 and specificity of 0.80 for diagnosing ICU-AW. A cutoff score of 24 out of 36 points was identified for the four-point scale. Conclusions The isometric technique is recommended with reporting on a collapsed four-point scale. Because HGD is easy to perform and sensitive, we recommend a new two-tier approach to diagnosing ICU-AW that first tests handgrip strength with follow-up strength assessment using the isometric technique for muscle strength testing if handgrip strength falls below cutoff scores. Whilst our results for the four-point scale are encouraging, further research is required to confirm the findings of this study and determine the validity of the four-point scoring system and cutoff score developed of less than 24 out of 36 before recommending adoption into clinical practice.