An Exploratory Randomised Trial of a Self-Managed Home-Based Exaggerated Spatial Cueing Intervention for Handwriting in Parkinson’s Disease

Abstract

Handwriting impairment is a cardinal symptom of Parkinson’s. However, treatment options are limited. Here we evaluate the utility and estimate effects of a novel low-resource handwriting intervention (Clinicaltrials. gov NCT03369587). Forty-eight people with Parkinsons with self-reported handwriting problems were recruited to an exploratory, assessor-blind two-arm parallel randomized trial to either diverging (n = 24, n = 19 analysed) or parallel (n = 24, n = 20 analysed) groups. Both received a six-week, five times a week, handwriting program: writing a daily diary on lined paper (diverging: 10 mm increasing to 13 mm apart, parallel: 10 mm apart). Outcomes were measures of impairment (cursive ‘el’, single and dual-task), handwriting function (sentence and free writing) and self-reported difficulties. Median diary entries (31, IRQ: 17.5–39) were greater than requested (30) with no differences between groups, p = 0.302. No adverse events were reported. Regardless of group, improvements were found in writing ‘el’ speed (single task: d = −0.90, 95% CI: −1.41: −0.38, p = 0.001; dual task: d = −0.72, 95% CI: −1.24: −0.21, p = 0.09) and amplitude (single task: d = 1.07, 95% CI: 0.49: 1.66, p < 0.001; dual task: d = 0.86, 95% CI: 0.35: 1.37, p = 0.002). Sentence amplitude (d = 0.80, 95% CI: 0.30: 1.29, p = 0.003) and perceived difficulties also improved (OR = −3.6, 95% CI: −12.6: −1.0, p = 0.047). Between-group effects were small (d = 0.11 to 0.48). Large improvements to handwriting, which required less attention, were found after self-directed well-adhered-to practice. Potential additional benefits of exaggerated cueing were small.

1. Introduction

The ability to write is an instrumental activity of daily living [1]. Problems with handwriting affect the majority of people with Parkinson’s [2] and are associated with depression and leaving employment early [3,4]. The few randomised studies evaluating interventions have found that practice can improve handwriting impairment [5,6,7,8]. However, evidence of the extent that this translates to better daily functioning is lacking [8,9]. Moreover, studies are heterogeneous, prominently with a high risk of bias, and evidence to support treatment options for this symptom is limited [10].

The distinctive feature of handwriting impairment in Parkinson’s is micrographia [2] which has two, often co-existing, presentations: constantly small handwriting (consistent micrographia) and progressive reduction in character size during a sentence (progressive micrographia) [11]. Deficits also include reduced writing speed, variability of movement size and increased interruptions [12], with freezing induced when writing speed and amplitude are challenged [13].Although handwriting problems are consistent with general motor symptoms of hypokinesia and bradykinesia, dopaminergic medication only partially improves symptoms [14], with progressive micrograpthia responding less well [11].

Cueing has been found to acutely improve symptoms, with writing speed and amplitude (size) responding to auditory (including verbal) and visual cues (see Nackaerts et al. [12] for a review). There is also evidence that cueing can illicit longer-term benefits. Nackaerts et al. [6] evaluated a visual spatial cueing intervention consisting of writing practice using a coloured line as a target for writing amplitude. They found that after six weeks of intensive training (5 × a week), amplitude improved. Moreover, they also found improvement when assessed whilst performing a concurrent cognitive task (dual task), inferring greater automaticity of the writing task.

Visual cues are believed to facilitate movement through visual–motor pathways bypassing the basal ganglia [15]. Cueing has been extensively used in walking training in Parkinson’s, improving the spatial aspect of gait and freezing [16], and there appears to be some overlap between walking and handwriting impairment [17]. Therefore, insights from gross motor rehabilitation approaches may help to inform fine motor intervention. Emerging evidence in forced exercise, whereby the cadence of the exercise is augmented to assist the participant in achieving a greater rate, suggests that exaggerating performance beyond what an individual would choose can lead to a greater training response [18]. In addition, exercise programs specifically for Parkinson’s often include large amplitude movements. Considering this, we investigated an exaggerated spatial cue handwriting intervention whereby the target cueing lines diverge across the page. To deliver the intervention we used a low-resource home-based approach whereby handwriting practice was delivered via a daily diary completed without supervision. Since we can expect that handwriting practice will improve impairment compared to a none handwriting control [5,6,7,8], we compared the diverging lines intervention to that using parallel lines. We sought to estimate effects of the intervention on impairment and function and, using cognitive motor interference, automatic control.

2. Materials and Methods

2.1. Design

This was an exploratory single-centre assessor-blind two-arm parallel randomised trial of two handwriting interventions. The study was approved by Oxford Brookes University Ethics committee (UREC: 171073), registered on Clinicaltrials.gov (NCT03369587) and reported considering CONSORT (extension to randomised pilot and feasibility trials) [19] and the Template for Intervention Description and Replication (TIDieR) guidelines [20].

After providing written informed consent, participants were allocated the next available study number by the blinded assessor. Participants were randomised (1:1) according to their study number, which was assigned sequentially on entry to the study. The study number was entered into an Excel worksheet which assigned group allocation (block sequence). The worksheet used minimisation to balance groups for whether participants were treated with levodopa (yes:no) and whether they exhibited symptoms of ‘progressive’ micrograpthia or not (>30% reduction in writing amplitude). The list was held by the principle investigator (Oxford Brookes University, UK) who informed the lead investigator at the European Parkinson’s Therapy Centre (EPTC), Italy. Writing dairies consisting of lined paper corresponding to group allocation were then provided to participants in opaque envelopes. Group allocation was concealed from the assessor until the end of the study.

2.2. Setting

Participants were recruited from Brescia, Italy, through the EPTC and Azione Parkinson’s. Assessments were carried out at the EPTC.

2.3. Participants

Inclusion criteria were (1) over 18 years old, (2) diagnosis of idiopathic Parkinson’s disease, (3) a score of ≤3 on the Hoehn and Yahr scale, (4) self-reported problems with handwriting (≥1 Unified Parkinson’s Disease Rating Scale UPDRS II, 2.7), (5) able to communicate in Italian, (6) following a stable medication regimen (no changes in the last two weeks and no anticipated changes for the course of the study). Exclusion criteria: (1) Severe depression or psychosis (history of), (2) reduced cognition that would preclude active involvement or capacity to consent, (3) additional neurological, orthopaedic, or other condition that significantly impairs hand function. Participants’ medication was continued as normal and recorded.

2.4. Intervention

Where: The handwriting training program took place in the participants own home. Who: Handwriting practice was self-directed with remote support available from staff from the EPTC. How: Staff from the EPTC instructed participants on how to complete the handwriting training program during a face-to-face session at the baseline assessment. Thereafter the participant performed the handwriting training program at their home unsupported; however, they could contact the EPTC should they need advice.

What: The program was delivered through handwriting diaries. The diaries also served to monitor adherence and fidelity. The dairies consisted of lined paper with participants asked to write a daily diary (a reflection of their day) aiming for letter size to be consistent with the lines. Participants were instructed to take as long as they needed to complete the daily diary and focus and concentrate on handwriting to achieve the size dictated by the lines. When and how much: Participants were asked to write in the dairy each day, with the aim to do five days a week for 6 weeks. Tailoring and progression: The nature of the handwriting program is that it accounts for individual ability; participants were asked to write as much as they can for each dairy entry, while concentrating on their handwriting. Participants could progress through writing more throughout the intervention.

2.4.1. Parallel Group

The dairy consisted of the writing areas within bold parallel lines 10 mm apart, with a midline (5 mm). Each writing area was separated by a 10 mm filled strip (Appendix A).

2.4.2. Diverging Group

The dairy consisted of the writing areas expanding, left to right. The lines started at 10 mm apart and diverged to 13 mm apart, with a midline (starting at 5 mm)). Each writing area was sperate filled strip of 10 mm at the left margin (Appendix A).

2.5. Assessment

Assessments were carried out at the EPTC by the same assessor who was blind to group allocation, and who was trained and experienced in assessing people with Parkinson’s. The staff received specific training on the handwriting measures used in the study. If a participant had ON and OFF periods, assessments were carried out during the ON state. Participants followed their usual Parkinson’s medication regime.

Demographic and descriptive information, including age, years since diagnosis, current medication use, and UPDRS were ascertained at the baseline assessment. Outcome measurements were performed at 0 weeks (baseline) and six weeks (end of intervention).

2.6. Outcomes

2.6.1. Impairment

Microgapthia was measured using the ‘el’ cursive character sequence task [21]. This consisted of the eight-character text sequence ‘elelelel’ (1 block) written 10 times (total 10 blocks). This sequence was used to derive Amplitude, average of blocks 1, 5, 6 and 10, measured by hand using a ruler; Ratio (%), the progressive reduction in writing amplitude of the 10th block area to the 1st block area of the el sequence, whereby a percentage of less than 100% would represent a reduction in amplitude; and Speed, the time taken to complete the sequence. The ‘el’ sequence was repeated with concurrent cognitive motor interference (dual task), whereby participants were simultaneously asked to count backwards in 3 s from a starting number of either 100, 101, or 102.

2.6.2. Function

Pangram participants wrote the following pangram sentence on blank A4 paper [8]:

‘Chi lascia la via vecchia per la nuova, sa quello che lascia ma non sa quello che trova’.

This sentence was used to derive Amplitude, the average of both ‘lascia’ in the pangram sentence; Ratio (%), the progressive reduction in writing amplitude of 2nd ‘lascia’ to the 1st ‘lascia’; and Speed, the time taken to complete the sentence.

Free writing: participants were given 5 min to write, on lined paper, a reflection on the day. The number of words written and number of legible words were measured. Self-reported problems with handwriting, Item 2.7 on the UPDRS (‘Over the past week, have people usually had trouble reading your handwriting?’ (0: normal: no problems, 4: severe: most or all words cannot be read)), was measured.

2.6.3. Intervention Fidelity

Those who were lost to follow up, discontinued the intervention, and reported adverse events were recorded. In those who did not discontinue the intervention, adherence was determined by the number of dairy entries and total pages completed in the handwriting workbooks. Following completion of the intervention, experience was captured through a process questionnaire.

2.7. Analysis

This was an exploratory study; no pre-existing data was available for the intervention (diverging lines) to estimate possible between-group effects. A study of similar intensity and duration of intervention found a statically significant improvement in microgapthia with a sample size of n = 18. We aimed to recruit a total of 50 people (25 people per group) in order to estimate intervention and between group effects and variability. This sample size was deemed adequate to evaluate intervention fidelity. Descriptive statistics were calculated for participant characteristics and pre-intervention data. Independent samples T-test, independent-samples medians, and the chi-square test were used to assess differences between groups for continuous, ordinal, and nominal data, respectively, at baseline.

All outcome data were analysed based on the intention-to-treat principle (if baseline data was available). The linear mixed models (LMM) procedure of SPSS V. 29 was used to determine the changes in measures as response variables according to assessment (baseline versus 6 week) and intervention (diverging versus parallel), with an autoregressive covariance structure. The model utilised maximum likelihood estimation and included individual intercept as a random effect. Data is reported with Cohens d and 95% confidence intervals for both assessment (pre-post) and between-group estimates of effect (groups*assessment). Self-reported handwriting difficulties (item 2.7 of the UPDRS) were considered ordinal data and analysed longitudinally using the generalized mixed model procedure of SPSS, reported with odds ratios and 95% confidence intervals as estimates of effect. Odds ratios of 1.68, 3.47, and 6.71 and Cohen’s d of 0.2, 0.5 and 0.8 were interpreted as small, medium and large, respectively [22]. Alpha was set at 0.05.

3. Results

3.1. Recruitment and Participant Flow

Between November 2017 and February 2020, 48 people were recruited and randomised; 26 were recruited via the EPTC and 22 via Azione Parkinson’s. All local individuals (i.e., not international clients) attending the EPTC during the study period were screened for eligibility and invited to take part (n = 31). It was not possible to record the number of people informed about the study through Azione Parkinson’s. Participant flow can be found in Figure 1. Pre-intervention assessment was not performed in the last six people randomised due to the COVID-19 pandemic, and a further three people dropped out prior to this assessment. These individuals were not included in the analysis. A protocol violation was revealed at analysis whereby one person in the diverging group entered the study despite scoring zero at baseline on the UPDRS 2.7; impairments to handwriting were evident so it was decided to include their data. Attrition at six weeks was the same in both groups (n = 5 lost to follow up and n = 4 people discontinued). There were no adverse events reported in either group.

Table 1. Baseline Characteristic and Pre-intervention Assessment Data.

Demographics	Diverging	Parallel	Difference
	n = 20	n = 22
Age (years)	66.2 ± 7.2	66.7 ± 0.8.7	p = 0.827
Progressive micrograpthia (Y:N)	15:5	16:6	p = 0.867
L-Dopa (Y:N)	14:6	18:4	p = 0.369
Gender (M:F)	11:9	14:8	p = 0.569
Years since diagnosis	9.0 ± 6.3	5.3 ± 4.2	p = 0.034
Hoen Yahr	2 (2–2.5)	2 (1–3)	p = 0.890
el sequence
Speed (s)	93.2 ± 53.8	101.5 ± 77.2	p = 0.708
Amplitude (mm)	5.4 ± 1.7	5.0 ± 1.6	p = 0.467
Ratio (%)	103.2 ± 38.3	92.5 ± 25.3	p = 0.337
Dual Task el sequence
Speed (s)	145.1 ± 106.9	175.2 ± 167.8	p = 0.543
Amplitude (mm)	5.1 ± 2.0	4.8 ± 1.7	p = 0.639
Ratio (%)	109.7 ± 35.6	119.0 ± 36.6	p = 0.463
Handwriting
Pangram speed (s)	75.0 ± 56.2	47.2 ± 15.6	p = 0.057
‘Lascia’ (mm)	5.5 ± 0.15	5.3 ± 1.4	p = 0.330
‘Lascia’ ratio (%)	91.5 ± 17.0	90.1 ± 23.2	p = 0.848
Free writing words	65 ± 39	80 ± 34	p = 0.098
Free writing legible words	50 ± 36	68 ± 32	p = 0.054
UPDRS 2.7	2 (1–4)	2 (1–3)	p = 0.650

Mean ± SD, difference (Parallel vs. Diverging) p = probability value (Independent-samples T-test, Independent-samples medians, chi-square test for continuous, ordinal or nominal data, respectively. Abbreviations: Y:N = Yes:Nom, M:F = Male:Female, UPDRS 2.7 = Movement Disorder Society Unified Parkinson’s Disease Rating Scale question 2.7 (handwriting problems), s = seconds, mm = millimetres.

shows demographic and pre-intervention assessment data. Those in the diverging group had a significantly longer time since diagnosis (p = 0.034). There were no statistically significant differences found between groups in any other baseline outcome data. However, there was a trend that scores for time to complete the pangram and number of words/legible words in the free writing task were better in the parallel group.

3.2. Intervention Fidelity

The median number of days the diary was completed was 31 (IRQ 17.5–39) out of the requested 30 entries (5 × per week × 6 weeks), with no difference between groups (diverging: 25.5 IRQ 17.5–35.5, parallel: 32 IRQ 13.5–40.5, p = 0.302). The average number of diary pages completed over the 6 weeks was 31 (~1 page per day) IRQ 17.5–39)) with no difference between groups (diverging: 33 IRQ 15–41, parallel: 30 IRQ 9–41, p = 0.772).

In total, 25 people completed the process questionnaire (n = 13 parallel, n = 12 diverging), only 2 people specifically reported a negative aspect to the intervention. These were that it was too demanding (parallel) and the daily frequency was too much (this was caveated that they felt the frequency was important (parallel)). However, there were also general perceptions that could be perceived as negative, such as the intervention felt like being at Primary School‘ (n = 3, all in the parallel group), and n = 3 (diverging) finding it difficult (n = 2 of these specifying this was only so in the beginning). Most people (n = 19 (86%), 11 parallel, 8 diverging) perceived benefits to their handwriting. Comments centred around writing better, with individuals also commenting it improved their confidence and one individual reporting that it improved their mood. One person commented that it was good that they could do the intervention at home. In total, 18 people (72%) expressed their intention to continue with handwriting practice after the intervention had finished.

3.3. Outcome

Outcome data are reported in

Table 2. Outcome.

Impairment	Pre-Post	Diverging										Parallel										Between Group
		Baseline					6 Weeks					Baseline					6 Weeks
	d, (95% CI), p	EMM ± Standard Error					EMM ± Standard Error					EMM ± Standard Error					EMM ± Standard Error					d (95% CI), p
el sequence
Speed (s)	−0.90 (−1.41: −0.38), 0.001	93.2 ± 14.2					76.6 ± 14.5					101.5 ± 13.8					82 ± 14.2					0.11 (−0.54: 0.75), 0.904
Amplitude (mm)	1.07 (0.49: 1.66), <0.001	5.4 ± 0.4					6.4 ± 0.4					5.0 ± 0.4					5.6 ± 0.4					0.34 (−0.35: 1.02), 0.404
Ratio (%)	0.27 (−0.22: 0.77), 0.279	103.2 ± 10.4					92.32 ± 7.3					92.5 ± 6.6					91.5 ± 7.1					−0.28 (−0.38: 0.95), 0.497
Dual-Task el sequence
speed (s)	−0.72 (−1.24: −0.21), 0.09	145.1 ± 31.3					124.1 ± 32.0					175.2 ± 29.5					135.4 ± 30.1					0.25 (−0.43: 0.92), 0.597
Amplitude (mm)	0.86 (0.35: 1.37), 0.002	5.1 ± 0.5					5.7 ± 0.5					4.8 ± 0.5					5.3 ± 0.5					0.27 (−0.40: 0.93), 0.539
Ratio (%)	0.07 (−0.42: 0.56), 0.779	109.7 ± 7.6					115.4 ± 8.4					119.0 ± 7.4					108.9 ± 8.1					0.26 (−0.41: 0.92), 0.570
Pangram
Speed (s)	0.08 (−0.57: 0.72), 0.812	75.0 ± 9.8					78.1 ± 9.9					47.2 ± 9.7					51.7 ± 9.8					0.48 (−0.00: 0.96), 0.032
‘Lascia’ (mm)	0.80 (0.30: 1.29), 0.003	5.5 ± 0.4					5.8 ± 0.4					5.3 ± 0.4					6.1 ± 0.4					−0.26 (−0.38: 0.91), 0.529
‘Lascia’ ratio (%)	0.28 (−0.20: 0.77), 0.260	91.5 ± 4.9					97.7 ± 5.5					90.1 ± 4.8					92.9 ± 5.2					0.15 (−0.50: 0.81), 0.814
Free writing
Number of words	0.03 (−0.47: 0.53), 0.910	65 ± 28					71 ± 29					80 ± 8					75 ± 8					0.34 (−0.28: 0.97), 0.275
Number of legible words	0.16 (−0.30: 0.63), 0.497	50 ± 8					57 ± 8					68 ± 7					65 ± 8					0.40 (−0.24: 1.03), 0.227
Perceived difficulties	OR, (95% CI), p	0	1	2	3	4	0	1	2	3	4	0	1	2	3	4	0	1	2	3	4	OR, (95% CI), p
% response each level	−3.6 (−12.6: −1.0), 0.047	5	21	32	16	19	13	33	40	13	0	0	45	25	20	10	11	53	24	6	6	0.7 (0.1: 4.1), 0.723

EMM = Estimated Marginal Means (from linear mixed model), pre–post = pre assessment vs. post assessment. d = Cohen’s d, OR = Odds Ratio, 95% CI = 95% confidence interval, p = probability value. s = seconds, mm = millimetres. Perceived difficulties = self-reported problems with handwriting, item 2.7 on the Movement Disorders Society Unified Parkinson’s Rating Scale: ‘Over the past week, have people usually had trouble reading your handwriting?’ (0: normal: no problems, 4: severe: most or all words cannot be read). The grey color is used to differentiate the 0–4 responses for ‘perceived difficulties‘.

. Considering handwriting practice irrespective of group, moderate to large effects were found on a number of outcomes. Large effects were found for improvements in amplitude of the el sequence under single (d = 1.07 (95% CI, 0.49: 1.66), p < 0.001) and dual task conditions (d = 0.86 (95% CI, 0.35: 1.37), p = 0.002) and pangram (d = 0.80 (95% CI, 0.30: 1.29), p = 0.003). Moderate to large effects were found for improving el sequence writing speed with significant effects under single task conditions (d = − 0.90 (95% CI, −1.41: –0.38), p = 0.001), and a non-significant moderate effect was observed under dual task conditions (d = −0.72 (95% CI, −1.24: −0.21), p = 0.09). Only trivial to small effects were found for progressive micrographia (ratio) and free writing outcomes. However, a significant moderate effect was found for improvement in perceived handwriting problems (UPDRS 2.7, OR = −3.6 (95% CI, −12.6: −1.0), p = 0.047).

The only significant differences between groups were found in pangram writing speed. However, groups were not well balanced at baseline for this outcome. The largest between group effects were found for writing amplitude outcomes. Direction of effect favoured the diverging lines group for el sequence tasks (single task: d = 0.34 (95% CI, −0.35: 1.02), p = 0.404), dual task: d = 0.27 (95% CI, −0.40: 0.93), p = 0.539) and the parallel group for pangram amplitude (d = −0.26 (95% CI, −0.38: 0.91), p = 0.529). Small effects favoured the diverging group for the free writing task (number of words: d = 0.34 (95% CI, −0.28: 0.97), p = 0.275, legible words: d = 0.40 (95% CI, −0.24: 1.03), p = 0.276).

4. Discussion

We found that a novel intervention requiring minimal resources improved the handwriting symptoms of people with Parkinson’s. The largest effects were found for amplitude outcomes, with improvements observed during dual task conditions indicating greater automatic control. Moderate effects were also found for improvement in amplitude during sentence writing and self-reported difficulties with handwriting.

The visual spatially cued intervention, irrespective of parallel or diverging groups, resulted in a significant improvement in el sequence amplitude and writing speed during single and dual task conditions with only small (non-significant) differences between groups. Improvement during dual task conditions is consistent with previous research and indicates less attentional resource was required during the el task after an intervention to improve handwriting [6,9]. De Vleeschhauwe et al. [9] found strengthened cortico-cortical connectivity with the bilateral parietal cortex left frontal cortex and right cerebellum after handwriting practice in Parkinson’s, with improved handwriting performance particularly related to connectivity with the cerebellum. Although they found dual task handwriting improved, they were cautions to infer whether their findings were indicative of automatic control, as functional reorganisations were associated with dorsal attentional networks [9]. While handwriting is a highly automated fine motor skill [23], continual visual attention is required to support spatial control [24]. A model for the development of automatic control proposes that automaticity occurs as a result of transfer of control from the basal ganglia to cortico-cortical projections from relevant sensory areas to the premotor and/or motor areas that initiate the task [25]. Indeed, in healthy individuals, Waldschmidt et al. [26] found that extended practice of a novel task was associated with a gradual transfer of control from the basal ganglia to the cortex. Importantly, they also reported that pre-automatic performance was depended on the putamen before transfer to the cortical regions. The putamen is central to acquire and run automatic programs and its impairment in Parkinson’s implicated in difficulty to develop automaticity [27]. Considering our findings and those of others that also found improvement in dual task conditions, handwriting practice does appear to lead to brain reorganisation that improves processing and frees attentional resources [6,9], seemingly via attentional cerebellar motor circuits that compensate for predominantly posterior putamen [11] basal ganglia dysfunction [15,16].

Recently, Broder et al. [28] investigated the effects of transcranial direct current stimulation after 20 min of handwriting practice using diverging line cueing, and observed improvement in amplitude after the training session, with some improvements maintained after 24 h. In our study of an exaggerated (diverging line) cueing, between-group effects were small and non-significant. However, it is noteworthy that direction of effect consistently favoured the exaggerated cuing for el single task and dual task amplitude outcomes. When considering the response to intervention, the distinct neural correlates of the different presentations of micrographia may have relevance. Consistent mircograpthia seems predominantly due to basal ganglia dysfunction, whereas disconnection between motor areas and the cerebellum are additionally implicated with progressive micrograthia [11]. The exaggerated cuing (diverging lines) stimulus is the opposite to what is observed during progressive micrograpthia, and we hypothesize that it may stimulate an enhanced training response. Therefore, it is notable that we did not observe improvement in progressive micrographia outcomes. However, progressive micrograpthia was not predominant in our preintervention data. Comparing the response of those primarily exhibiting progressive or consistent mircograpthia was beyond the scope of this study. However, examining whether the manifestations of micrograpthia and handwriting impairment respond better to different cueing interventions may lead to better tailored treatment [10]. It should noted that the divergence in lines totalled 3 mm, as to not make practice incongruent to everyday handwriting. However, this subtle increased stimulus may have also contributed to the minimal differences found between groups, alongside a ceiling effect in those whose handwriting amplitude achieved that of the spatial cues. Broder et al. [28] investigated the acute effect of a 10 mm divergence. Future work could consider the influence of the size of the exaggerated stimulus.

The extent to which improvements observed at impairment level (i.e., el sequence) transfer to performance in daily life writing is equivocal [6,28]. Nackaerts et al. [6] used the Screening of Handwriting Difficulties test, which involves five minutes of free writing, and found increased writing size after 6 weeks of amplitude training, but did not report improvement in total score or letters written. Our results are broadly consistent with these finding, as we found improvements in amplitude when writing ‘actual’ words during the pangram task, but found no improvement in the free writing task where the outcome was more associated with writing speed than amplitude. Indeed, spatial cueing interventions do not target writing speed. Interestingly, improvement in temporal aspects of handwriting have been associated with non-handwriting-specific training [10]. However, perhaps of note is the small (non-significant) difference between group effects found for the diverging group, who improved in eligible words written after intervention during the free writing task.

We found that perceived handwriting ability improved. Particularly conspicuous in the data was the 19% of the diverging group who had rated handwriting problems as severe at baseline compared to none after the intervention. We previously found a less intensive but longer intervention (2 × a week for 6 months) led to improvement in perceived handwriting ability sustained for six months post intervention [8]. Indeed, evidence supports that, to an extent, improvements in handwriting are retained following intervention [6,8]. Ongoing practice may be required for long-term retention of benefits. Therefore, it is also encouraging that most people in the current study expressed their intention to continue with handwriting practice after the intervention period. Home-based dairy or workbook handwriting interventions have been consistently well adhered to in Parkinson’s [6,7,8]. The pragmatic nature and low cost of the intervention allows individuals to continue at low cost with no supervision. Whether they do so is unknown; however, perceived benefit and willingness to continue are elements that would support long-term engagement. Whilst the pragmatic nature of the intervention has these benefits, it may contribute to variability in training response, as people chose when and how to engage with the daily handwriting practice. Future studies should consider co-developing guides and materials to support long-term participation. Furthermore, the use of digital technologies has several benefits of pen and paper, providing a means to incorporate temporal cues and other forms of visual cues, remote monitoring, assessment, and feedback. Indeed, Gardoni et al. [29] recently used a commercially available tablet and stylus to analyse a writing speed, amplitude and repetitive cursive loop task.

There are several factors that should be considered when interpreting our findings. Firstly, the relatively small sample size limits generalizability and likely contributed to the imbalance in groups at baseline. The diverging line group had on average a longer time since diagnosis and tended to have a slower handwriting speed (pangram), and write fewer words and fewer legible words at baseline (free writing). It is also important to note that practical reasons meant we were unable to assess the eligibility of all those attending Azione Parkinson’s and thus determine the extent of selection bias may be present in those recruited through this route. Whilst we recognise these limitations, it is encouraging that we observed the effects in those with severe difficulties. The foremost consideration is that the study was suspended and later stopped due to the COVID-19 pandemic. This resulted in being two participants short of recruitment target, and the originally planned 12-week assessment point was not completed. Therefore, further research is required to assess sustained benefit, which should also consider the impact on daily lives.

5. Conclusions

Our results show that handwriting practice, at home through simply writing a daily diary, improves micograpthia and may translate to improved handwriting function requiring less conscious effort. Whilst we found limited evidence that exaggerated cueing produced greater positive effects, future research might consider its application in tailored approaches for those primarily exhibiting progressive micrograpthia. There is emerging evidence that writing practice can contribute to improved general dexterity and quality of life [30,31]. This seems of particular importance to individuals earlier in the condition and may help to support them to maintain employment [4]. Given the potential for benefit and no reported and minimal conceivable risks, handwriting practice may be simply supported through spatial cued dairy writing.