H567/03 Section B Option 4 - Lewis et al. (2014) and exercise for wellbeing

Background and aim

Parkinson's disease is a neurodegenerative condition associated with motor difficulties, but non-motor experiences such as mood disturbance also affect wellbeing. Earlier dance research often examined movement or balance rather than psychological outcomes. Lewis and colleagues therefore investigated mood over a longer programme and immediately around one class, including older adults with Parkinson's disease and age-matched adults without it. Accepted manuscript, pp3–6, OCR Sport guide, pp13–19.

The study connects physical activity to psychological wellbeing. Biological processes associated with activity, cognitive challenge and mastery, and social contact are possible explanations. The study measured mood, not these proposed mediating processes, so it cannot identify which route produced the change. Nor was it a trial showing a cure for Parkinson's disease or clinical depression.

Design, sample and measures

The design combined an independent-groups comparison, Parkinson's versus non-Parkinson's, with repeated mood measurement over time. Disease status was a naturally occurring group variable, not randomly allocated. Every group received dancing; “control” refers to the non-Parkinson's comparison group, not to absence of exercise. The dependent variables were self-reported mood scores.

The recruited sample was 37 adults aged 50–80, mean age 65.5. Of these, 22 had Parkinson's disease: 12 men and 10 women, rated mild to moderate by trained physiotherapists using Hoehn and Yahr stages I–III. The 15 non-Parkinson's controls comprised seven men and eight women; eight were partners of Parkinson's participants. Recruitment through local advertisements and Parkinson's support groups produced a self-selecting sample.

The Profile of Mood States (POMS) provided the longer-cycle measure. Its 64 items use a 0–4 rating scale and cover tension–anxiety, vigour–activity, depression–dejection, fatigue–inertia, anger–hostility and confusion–bewilderment. Total Mood Disturbance (TMD) combines negative dimensions with vigour subtracted; lower TMD indicates better overall mood. Greater vigour, unlike greater negative-subscale scores, indicates a more positive outcome.

The Brunel Mood Scale (BRUMS) is a 24-item shorter measure using the same mood dimensions, suitable for the immediate before/after comparison. The Mini Mental State Examination (MMSE) was administered at baseline to screen cognitive status. POMS and BRUMS measure reported mood; their depression subscales are not equivalent to a clinical diagnosis.

The dance programme and measurement schedule

After informed consent, participants completed demographics, baseline MMSE and POMS in week 1, with POMS referring to feelings over the past month. A qualified instructor then delivered weekly dance classes in weeks 2–11, ten classes in total. The two groups attended separate but identical classes.

Each class lasted 50 minutes: ten minutes of warm-up, 30 minutes of dancing, a five-minute break and five minutes of cool-down. Activities used rhythmic music and were adapted for age and mobility; participants could stand or sit. The style changed every two weeks across Bollywood, Tango, Cheerleading, Old Time Music Hall and party dancing based on Charleston and Saturday Night Fever.

In week 9, BRUMS assessed how participants felt right now, immediately before and after a class: the short cycle. A few days after the last class, week 12 POMS assessed the longer cycle. The programme lasted ten weeks of classes within a twelve-week measurement schedule; it was not twelve dance sessions.

Two people withdrew during the intervention; five could not attend final testing, and one further control participant was excluded after falling below the MMSE cut-off. Incomplete subscales also affected analyses, so the recruited 37 should not be presented as the complete sample for every result. The paper reports differing analysis sample sizes.

Results and proportionate conclusions

There was no significant TMD group effect or group-by-time interaction for either cycle. The results therefore did not establish a greater TMD benefit for the Parkinson's group. Non-significance also does not prove two populations are identical.

Participants with higher initial depression scores reported a larger fatigue reduction in further analyses, but adjustment for multiple testing weakened the relevant subscale findings. It is misleading to convert this into a demonstrated treatment effect on clinical depression. Baseline Parkinson's and control mood scores did not differ significantly, although the Parkinson's group showed greater disturbance than published norms on some dimensions.