Does nocturnal hypokinesia affect daytime mobility in Parkinson's disease?

How the study defined nocturnal hypokinesia

Nocturnal hypokinesia (NH) is reduced movement at night: staying in one position during sleep, making fewer postural transitions, and having difficulty turning or getting out of bed. The study did not measure bed movements with sensors. It used a single item of the MDS-UPDRS Part II, item 2.9, and counted anyone scoring 1 or more as having NH. That choice matters for reading every result: the exposure is a clinician-scored report of nighttime mobility difficulty, not an inertial-sensor measurement of turning in bed.

The comparison group was built with propensity score matching. From a cross-sectional dataset of 121 people with Parkinson's disease, the authors formed 44 pairs, one participant with NH and one without, matched on age, sex and disease duration. All participants were in the ON medication state, had modified Hoehn and Yahr stage below 4, could walk without aids, and scored at least 18 on the Montreal Cognitive Assessment. The design is cross-sectional, so the paper compares groups at one point in time rather than tracking whether NH precedes a decline in daytime walking.

Supervised tests and five days of free-living steps

The study separated mobility into two measurement settings that should not be treated as interchangeable. Supervised mobility was captured in the laboratory with a lower-back inertial device during three tasks: 20-metre forward walking, an instrumented Timed-Up-and-Go, and a 3-metre backward walking test. Forward walking yielded speed, cadence, height-normalized stride length, and stance and swing phase percentages. The Timed-Up-and-Go yielded total duration plus mean and peak angular velocity of the 180-degree turn. Backward walking yielded speed.

Unsupervised mobility was the average daily step count (AvDS) from a Garmin Vivosmart 4 smartwatch worn for five consecutive days on the wrist of the less affected side. Recording days counted as valid when wear time while awake reached at least 80%, with awake time estimated as 24 hours minus device-indicated sleep time. Participants were told to go about daily activities as usual and received no reminders. The five-day window was chosen because the authors report prior work showing at least four days are needed for reliable daily step estimates in Parkinson's disease and that five days gave acceptable reliability across severity, stage and phenotype.

Turning, backward walking and daily steps diverged

After false discovery rate correction, the two matched groups differed on four mobility outcomes. The NH group took longer on the Timed-Up-and-Go (12.6 versus 11.5 seconds), turned more slowly (mean angular velocity 65.6 versus 74.4 degrees per second; peak angular velocity 128 versus 147 degrees per second), walked backward more slowly (0.52 versus 0.64 metres per second), and accumulated fewer average daily steps (5126 versus 6566 steps per day). All four adjusted p values were between 0.031 and 0.035.

Forward-walking measures did not separate the groups. Speed was 1.14 metres per second with NH versus 1.23 without it, and cadence, normalized stride length, stance and swing percentages were all non-significant. This is a useful contrast rather than a contradiction: in this sample, the differences appeared in turning, backward walking and overall daily activity, not in straight-ahead gait during a short supervised walk. The paper also reports that NH severity, scored on the same MDS-UPDRS item, correlated negatively with forward-walking speed, normalized stride length, Timed-Up-and-Go duration, mean and peak angular velocity, backward-walking speed and average daily steps.

Fear of falling weakens the supervised findings

The authors ran sensitivity analyses because fear of falling is known to affect mobility in Parkinson's disease. They recomputed partial Spearman correlations between the MDS-UPDRS 2.9 score and each mobility variable while controlling for the Falls Efficacy Scale International score, and ran ANCOVA with NH status as the factor and that score as a covariate. After this adjustment, the group differences in supervised mobility lost significance. The paper therefore describes its supervised-mobility findings as preliminary.

The daily step result behaved differently. The association between NH severity and average daily steps remained significant after controlling for fear of falling, which the authors interpret as evidence that the link between NH and reduced real-world walking is at least partly independent of fear of falling. Readers should keep the two outcomes apart: a laboratory difference that did not survive this adjustment is weaker evidence than a free-living association that did, and neither establishes that NH causes lower activity.

A shared axial mechanism remains untested

The authors propose that NH and daytime mobility impairment could share underlying mechanisms, particularly axial bradykinesia and rigidity that respond poorly to dopaminergic therapy. They note that difficulty turning in bed has been attributed mainly to impaired axial rotation, and that reduced nocturnal turning velocity has been associated with bradykinesia, rigidity and postural instability and gait difficulty scores. An alternative explanation they offer is that NH severity contributes to sleep impairment, which in turn affects daytime mobility.

The study does not test either pathway. It is cross-sectional, the NH classification comes from one questionnaire item, and the authors state they did not systematically test the effectiveness of propensity matching, only the absence of significant differences in the matching variables. They also report that the NH group had higher total MDS-UPDRS scores, indicating more severe global symptoms, but did not differ significantly on MDS-UPDRS Part III or the postural instability and gait difficulty subscore, which the authors suggest may reflect matching on disease duration. The mechanistic reading is therefore the authors' interpretation, not a demonstrated causal chain.

Worse symptoms despite higher levodopa doses

Beyond mobility, the NH group showed worse motor and non-motor clinical scores despite a higher levodopa equivalent daily dose, according to the paper's conclusion. The authors read this as support for evaluating NH as a symptom that reflects both nighttime and daytime function, and they argue for interventions targeting both domains. That recommendation follows from the observed pattern of worse symptoms and lower activity, not from any intervention the study delivered.

The practical distinction the data support is narrower than a general call for treatment. The most robust signal in this paper is the severity-to-daily-steps association, which survived adjustment for fear of falling; the supervised turning and backward-walking differences did not. For a clinician or researcher deciding what to prioritize, that asymmetry points to free-living step counts as the outcome most consistently tied to NH in this dataset, while laboratory mobility differences should be treated as provisional until replicated.

Where the conclusion stops

The findings apply to adults with idiopathic Parkinson's disease at modified Hoehn and Yahr stage below 4 who could walk independently without aids and were assessed in the ON state; people with severe cognitive impairment or other conditions affecting mobility were excluded. The NH definition rests on a score of 1 or more on one MDS-UPDRS item, so the results describe reported nighttime mobility difficulty rather than sensor-quantified bed mobility. Step counts came from a commercial watch over five days, a window the authors justify from prior reliability work rather than from a new validation in this sample.

Within those limits, the paper's most decision-relevant comparison is between the two mobility settings. The NH group's lower average daily step count, 5126 versus 6566 steps per day, and the persistence of the severity-step association after adjusting for fear of falling make real-world walking the finding to watch. The supervised turning and backward-walking differences, though statistically significant before adjustment, did not survive it, so they should not yet be used on their own to characterize daytime mobility in people with NH.

Sources used in this answer