Time is health: management of Parkinson’s disease in primary care: a retrospective quantitative study of diagnostic and therapeutic timelines

Abstract

Background

Parkinson’s disease is a neurodegenerative disorder which can be difficult to diagnose due to the non-specificity of the initial symptoms. The objective of our study was to objectify and quantify the delay of the diagnostic process and to identify the factors that influence it.

Methods

An exploratory quantitative, cross-sectional, retrospective, observational study was conducted from 22 March 2023 to 31 March 2024 among general practitioners (GPs) practising in France who were asked to study their patient files concerning the care pathways of their patients with Parkinson’s disease. The data was collected using a standardised, anonymised online questionnaire (Limesurvey^®), the link to which was distributed via email. The questionnaire provided detailed information on the key dates of diagnosis and follow-up, and on the symptoms reported by patients and documented by doctors.

Results

The diagnostic pathway for Parkinson’s disease is substantiated by the observation of protracted delays. On average, patients take more than three years to obtain a formal diagnosis and to receive appropriate treatment. The delays experienced are primarily attributable to prodromal non-visible non-motor symptoms, for which referral delays are tripled in comparison to visible motor symptoms. Challenges in accessing a neurologist and the frequent necessity for unnecessary complementary exams also contribute to the delays.

Conclusion

Our results underscore the pivotal role of GPs in the early detection of Parkinson’s disease, particularly in its prodromal stage marked by non-visible non-motor symptoms. While the diagnostic hypothesis was confirmed in most cases (70%) by neurologists, specific training of GPs in Parkinson’s disease with recognition of non-motor symptoms and treatment initiation would accelerate the process of diagnosis and referral to a neurologist and potentially reduce diagnostic and therapeutic times.

Background

Parkinson’s disease (PD) is the second neurodegenerative disease after Alzheimer’s [1–4]. The population is ageing and the number of people living with Parkinson’s disease (PwP) doubled between 1990 and 2015. In France, at the end of 2015, around 160,000 patients were being treated for PD. Of the 25,000 new cases each year, 17% were under the age of 65. The number of PwP will increase by half by 2030 compared with 2015, with one in every 120 people over 45 affected. These numbers are probably underestimated, as many patients escape diagnosis and therefore go untreated, especially the elderly [2].

Parkinson’s disease is a progressive neurodegenerative disorder of the central nervous system, characterised by damage to the substantia nigra pars compacta. The loss of dopaminergic neurons is a contributing factor to motor disorders, including akinesia, rigidity, walking, speech and balance disorders. Furthermore, patients may present with a wide range of non-motor disorders, including cognitive and behavioural problems, constipation, hyposmia (excluding covid cases), sleep disorders and nightmares [3, 4].

The causes of PD are varied and include environmental, toxic (pesticides, mercury, cadmium, lead) and/or genetic factors (3–5% linked to monogenic and 16–36% of the heritable risk linked to non-monogenic PD genes [4]). Having a family history of PD or tremor, constipation, and not being a smoker, composes a high risk of developing PD, with each condition doubling chances of getting the disease [4].

Parkinson’s disease is characterised by a prodromal phase, estimated to last between five and 20 years. During this phase, symptoms are insidious and highly variable, concerning both pre-clinical motor and non-motor symptoms [4].

Prodromal symptoms such as constipation, anxiety-depression syndrome, mild cognitive impairment and sleep disorders can be overlapping. All of these can be met in general practice, but they are not specific to PD, which makes it difficult for patients and General Practitioners (GPs) to identify a PD origin [5]. Therefore, delays in the diagnosis of PD are common and GPs should be aware of this fact.

Two delays have been identified: 1) Diagnostic delay: the period of uncertainty during which both patients and GPs are unsure about the patient’s condition. 2) Therapeutic delay: the period during which patients are waiting for treatment and care.

According to Plouvier et al., GPs should be alert when patients present with combined multiple prodromal symptoms in a two-year period [6]. In 2015, Postuma et al. defined the Movement Disorder Society (MDS) Clinical Diagnostic Criteria for PD with a view to guiding and systematising the clinical diagnostic process among clinicians [7]. In 2019, the MDS criteria were updated and extended to prodromal symptoms of PD, adding olfactory loss and neurogenic and symptomatic orthostatic hypotension [8]. Postuma et al. also state that voice changes should be considered as a prodromal marker [9]. There are other independent prodromal markers, including changes in global cognitive function and alterations in executive functions. These changes can appear up to six years before clinical PD, when approximately one-fifth of patients with PD present mild cognitive impairment. Consequently, patients with prodromal PD should benefit from early treatment, including non-pharmacological interventions, to prevent or delay cognitive dysfunction in early PD [10, 11].

However, most prodromal symptoms, such as constipation, voice changes, cognitive impairment, depression and sleep disorders, are easily identifiable to GPs and can be efficiently managed with symptomatic treatments in primary care, thereby preserving and enhancing the patient’s quality of life. In addition, research indicates that physical and intellectual activities such as learning new content (e.g. music, poetry, cooking) have a positive impact on brain plasticity. These activities appear to be particularly beneficial when carried out at an early stage.

The early administration of dopaminergic treatments and physiotherapy has been shown to improve many symptoms [3, 4].

The diagnosis of PD is based on a clinical examination, which reveals the main symptoms of the disease. This highly specific examination is performed by a doctor who has received training in this area (i.e. the doctor is familiar with the MDS criteria and the MDS-UPDRS-scale [12]). In most cases, this examination is sufficient to establish a diagnosis without the need for further tests [4].

The results of these tests, such as a brain CT scanner or MRI (magnetic resonance imaging), are normal in PD. These are reserved for people with an atypical presentation [4, 6].

Cerebral scintigraphy, also known as DaTscan, is a diagnostic imaging procedure that shows the consequences of dopamine deficiency on the basal ganglia. This test does not specifically confirm the diagnosis of PD and is therefore not routinely performed [4]. Isaacson et al. observed that DaTscan results were abnormal in 58.7% of patients, normal in 37.8%, inconclusive in 3.5%, changed clinical diagnosis in 39.8%, and led to medication therapy changes in 70.1% of cases [13].

According to Heim et al., transcranial sonography allows for high diagnostic accuracy in discriminating early PD from essential tremors. This examination is a promising low-cost, widely available, non-invasive marker diagnostic tool. It is suitable for use in primary care settings [14].

The challenges associated with accessing a neurologist, conducting complementary exams such as brain MRI and DaTscan to confirm or refute the diagnosis of PD, and the subsequent impact on diagnosis and treatment, require urgent attention. Once a patient has been diagnosed with PD, the GP or neurologist will have a strong suspicion or confirmation of this. The next step is to discuss a therapeutic strategy with the patient, whether this is drug or non-drug based. This discussion will be based on the patient’s preference, the doctor’s objective need for symptomatic treatment and the patient’s subjective experience.

Treatment goals are not uniform but vary from person to person, emphasising the need for personalised decision-making and shared management. Currently, there is no therapy that can slow down or stop the progression of PD. Nevertheless, several promising neuroprotection strategies are being tested for their disease-modifying potential, with initial results that are very encouraging. These strategies will require early treatment for greater effectiveness, and therefore diagnosis at the prodromal stage [4]. However, Khan et al. concluded after their scoping review of 1844 studies that there is a lack of studies assessing the accuracy of PD diagnoses when made by GPs [14]. It is imperative that more focused research is conducted in this area, as diagnostic delays and errors may lead to potentially harmful but preventable delays in treatment initiation, resulting in decreased quality of life for PwP [15].

The primary objective of this exploratory pilot study in a real-life primary care setting was to estimate the time taken and the chronology of the various stages in the diagnosis of PD, from the first symptom to the initiation of drug treatment.

The secondary objectives were to compile a list of the various clinical symptoms reported by the patient in a primary care setting.

Methods

A quantitative, cross-sectional, observational study was conducted among practising GPs in France between 22 March 2023 and 31 March 2024.

For this preliminary exploratory study, thirty GPs from the members of the regional GP college (teaching faculty) were invited to participate in an online survey. The survey was administered via the Limesurvey^® platform. A link and a QR code directing respondents to the questionnaire were included in this e-mail. The questionnaire for this study was developed based on the chronological course of the diagnostic and therapeutic care pathway in primary care. The questionnaire has not been published elsewhere and is available to consult in the supplement.

The GP completed the questionnaire based on their own observations, which were documented in their professional software. A tutorial at the beginning of the questionnaire explained how to complete it. This standardised questionnaire was anonymous and consisted of two parts (Part A and Part B). Part A concerned the GP’s socio-demographic data, and Part B the patient’s socio-demographic and medical data, the motor and non-motor symptoms reported by them, as well as additional diagnostic tests. Also included were the date on which the doctor mentioned PD for the first time as a diagnostic hypothesis, the referral letter to the neurologist (already linked to the GP by an existing professional partnership, or chosen by the patient according to the distance of the patient’s home or preexisting trustful relationship), and the feedback letter from the neurologist with confirmed PD diagnosis. Please note that Part B was repeated in the online file so that the doctor could include several patients, up to 10 if possible. To do this, the GP retrospectively consulted each of their patients’ files of their patients with PD, including those who had died, and entered the above-mentioned information in our questionnaire. In the letters from the corresponding neurologists, the doctor noted the following information: the date on which the patient had his first appointment; the date of brain imaging (DaTscan/MRI/FluoroDopa); the date on which antiparkinsonian treatment was started; and the date of the last appointment with the neurologist. Regarding the evaluation of symptoms in the questionnaire, we employed the Parkinson Well Being Map^® due to its comprehensive coverage of both motor and non-motor symptoms [16]. This tool is designed for use in primary care settings. Responses were collected in a spreadsheet format. We began by carrying out a univariate descriptive analysis. Qualitative variables were described in terms of frequency (percentage). All analyses and calculations of average times were carried out using R version 4.4.1 and its console Rstudio^® version 2024.09.0. Missing values were excluded from the delay calculations.

Results

Twenty out of 30 GPs invited (66%) participated in the study. The average age of the doctors was 44.4 (+/−9.1) years, with an average length of service in the profession of 13.8 (+/−9.9) years. Their further socio-demographic characteristics assessed through part A of the questionnaire are presented in Table 1 below.

Table 1.

Characteristics of the 20 GPs who participated into the study

Characteristics	Number of GPs [n]	Percentages [%]
Gender
Female	7	35
Male	13	65
Practice environment
Rural	2	10
Semi-rural	8	40
Urban	10	50
Distance to 1st nearest neurologist
< 5 km	8	40
Between 5 and 10 km	7	35
Between 10 and 20 km	3	15
Between 20 and 50 km	2	10
More than 50 km	0	0
Distance from the 1st PD expert centre
< 5 km	4	20
Between 5 and 10 km	0	0
Between 10 and 20 km	0	0
Between 20 and 50 km	3	15
More than 50 km	13	65
Personal/family history/knowledge of PD
Yes	2	10
No	18	90
Further training on PD
Yes	6	30
No	14	70
Number of People with PD in the patient base
0	1	5
1	4	20
2	2	10
3	4	20
4	3	15
5	1	5
6	0	0
7	1	5
8	1	5
9	0	0
10	3	15
Number of all patients affiliated to the GP
Average	1500
Median	1374
Extreme values	744–2800
Number of people with PD
Average (n)	4
Median (n)	3
Extreme values (n)	0–10
Total (n)	70

The GPs were asked how many patient files they can include to part B of our survey.

The following Fig. 1 shows the flow-chart of the data collection process concerning the patient population included by the 20 GPs.

Fig. 1.

Flow chart

The socio-demographic characteristics of the PwPs studied are presented in Table 2 below. As far as in France professional exposition can be recognized as an origin for PD form farmworkers and welders, we found it important to assess the professional background of the patients.

Table 2.

Socio-demographic characteristics of the population studied (people with parkinson’s disease)

Characteristics (n = 51)	n	%
Gender
Female	18	35
Male	33	65
Profession
Medium-sized farmers (20 to 40 hectares wheat equivalent and similar activities) (12)	9	18
Craftsmen (21)	1	2
Administrative and commercial managers (37)	1	2
Civil service executives (33)	1	2
Retailers and similar (22)	2	4
Company administrative employees (54)	2	4
Civil servants and public service employees (52)	3	6
Commercial employees (55)	2	4
Unskilled industrial workers (67)	6	12
Skilled craft workers (63)	1	2
Skilled industrial workers (62)	4	8
Direct services to individuals (56)	1	2
School teachers (42)	1	2
Professors, scientific professions (34)	1	2
Information, arts and entertainment professions (35)	1	2
Intermediate administrative occupations in the civil service (45)	1	2
Intermediate occupations in health and social work (43)	1	2
Technicians (47)	2	4
Not classifiable	8	16
Other (housewife)	3	6
Family history of Parkinson’s disease
Yes	8	16
No	43	84

The next Table 3 details the range of initial symptoms reported by patients to their GP, classified according to the categories of the Parkinson well-being map [16].

Table 3.

Initial symptoms observed by the patient

Breakdown of 1 st symptoms observed by the patient (n = 51)	n	(%)
Mobility disorders	45	88
Trembling	32	71
Slowness of movement	24	53
Morning stiffness	18	40
Reduced movement	13	29
Loss of balance	10	22
Feeling of sticky feet	7	16
Mobility disorders	6	13
Mood changes	23	45
Mood changes during the day, feelings of anxiety/anxiety/panic	13	57
Feelings of depression	12	52
Apathy	10	43
Loss of interest	9	39
Memory and attention disorders	12	24
Difficulty concentrating	8	67
Forgetting recent events	7	58
Loss of the thread of conversation	3	25
Difficulty remembering names/numbers/events.	2	17
Sleep disorders	10	20
Frequent waking at night	7	70
Difficulty falling asleep	6	60
Difficulty getting back to sleep	5	50
Incessant need to move your legs	3	30
Early morning awakening	2	20
Sleep disturbed by changes in position	2	20
Painful symptoms	9	18
Cervical/lumbar pain	6	67
Painful stiffness during the day or in the hand	5	56
Painful contractions on waking	2	22
Burning/tingling/squeezing pain	2	22
Digestive symptoms	8	16
Constipation	7	88
Stomachache	1	13
Excess saliva	1	13
Other symptoms	6	12
Taste and smell disorders	2	33
Unexplained fatigue during the day	2	33
Dizziness/head spinning on lifting	1	17
Loss of self-esteem	1	17
Bladder and sexual function disorders	5	10
Altered libido	3	60
Difficulty having sexual intercourse	3	60
Urgent need to urinate	2	40
Gets up at night to urinate	2	40

The next Table 4 presents the motor and non-motor symptoms recorded by the GP in medical records.

Table 4.

Signs noted on clinical examination by the GP

Clinical signs noted on clinical examination by the GP (n = 51)	n	%
Motor signs	50	98
Resting tremor	39	78
Stiffness	32	64
Akinesia	30	60
Slow motion	22	44
Dysarthria	9	18
Freezing/festination	7	14
Postural instability	4	8
Hypomimia	1	2
Left hemiparesis	1	2
Trembling lips	1	2
Non-motor signs	33	65
Anxiety-depression syndrome/apathy	24	73
Constipation	7	21
Weight loss	6	18
Nightmares	5	15
Restless legs syndrome	5	15
Dysexecutive syndrome/dementia	5	15
Urgent care	2	6
Thermal dysregulation	2	6
Orthostatic arterial hypotension	2	6
Nocturia	2	6
Hyposmia	1	3
Psychosis	1	3
Dysesthesia	1	3

The following Table 5 presents the different periods of the diagnosis pathway and their duration in days, starting from the first symptom reported by the patient up to the first prescription of medication by the GP and/or neurologist. The average age at onset of symptoms was 66.3 years (+/−14.6). On average, GPs referred patients to a neurologist after 28 days if they exhibited motor symptoms. However, they took three times as long (93 days) to refer patients in cases of non-motor symptoms.

Table 5.

Timeframes for the different stages in the diagnosis pathway of PD

Time points of the diagnostic process	Average lead times [days]	Median lead times [days]	Standard deviations [days]	Extreme values (Min*/Max) [days]	Missing data (/n = 50)
From the patient’s first symptoms reported to their documentation by the GP	553	30	+/−1.731	(−2554/7171)	1
From the patient’s initial non-motor symptoms reported to their documentation by the GP	1. 879
From the documentation of the first symptoms of all kinds by the GP to the suspicion of PD by the GP	70	0	+/−236	(−30/1.587)	1
From suspicion of PD based on non-motor symptoms by the GP to referral to the neurologist.	93	0	+/−450	(−1461/2081)
From suspicion of PD based on motor symptoms by the GP to referral to the neurologist.	28
From initiation of treatment by the GP to referral to the neurologist	196		+/−553	(0/2.081)
From referral to the neurologist to the first appointment with the neurologist	114	53	+/−503	(−1505/2826)
From the first appointment with the neurologist to further examinations by the neurologist	141	29	+/−350	(−2/1.460)
From complementary examinations to initiation treatment by the neurologist	33	−9	+/−150	(−116/470)	1
From the first appointment with the neurologist to the initiation of treatment by the neurologist	6	0	+/−168	(−729/477)
From the first symptoms of all kinds to the initiation of treatment by the GP	1.324	121	+/−2.267	(3/7.171)
From the first motor symptoms to the initiation of treatment by the GP	133
From the first symptoms of all kinds to the initiation of treatment by the neurologist	456	243	+/−1.345	(−3.947/4.327)
From the first consultation to the last consultation with the neurologist	1.163	323	+/−2.425	(−75/15.061)
From the start of treatment to the last consultation with the neurologist	1.740	881	+/−2.836	(0/15.061)

*Negative extreme values appeared when a patient already has been seen by a neurologist for another reason than PD, before the GP documents symptoms possibly linked to PD. Another situation is when a patient had another GP before the participating GP and already has been diagnosed with PD but the new GP documented new symptoms during PD evolution. We calculated the different delays in days from concrete dates documented in the patient files. We calculated delays from a point X (e.g symptoms noticed by the patient) to a point Y (e.g. symptoms reported by the patient to the GP who documents them in the patient file). This generally supposes a positive number of days as X comes before Y based on our initial hypothesis. All events that happened before X on the timeline inverse the chronology into negative values

In 31% of cases (n = 16), the GP requested further tests, such as MRIs, in 88% (n = 14) of these cases.

In one case, a DaTscan and a blood test were performed (6% of subjects, n = 1).

GPs initiated treatment for 27% (n = 14) of patients. In cases where motor symptoms were present, GPs initiated treatment after 133 days, compared to 1,324 days in cases where symptoms were absent.

Of the drugs prescribed by GPs, Levodopa/Benserazide was the most prescribed (72%; n = 10), followed by Pramipexole LP (14%; n = 2) and Rasagiline (7%; n = 1).

The initial diagnostic hypothesis proposed by the GP was corroborated by the neurologist in 70% (n = 40) of patients. In 85% of cases (n = 34), the neurologist initiated a treatment.

In 37% of cases (n = 19), the neurologist requested further examinations, including 25 imaging tests. The patient has undergone three DaTscan scans, seven brain MRI scans and five PET-fluoro DOPA scans.

The average time between the complementary exams and the initiation of treatment by the neurologist was 33 days (+/−150). From the initial symptoms being noted to the neurologist commencing treatment, a total of 456 days elapsed.

Figure 2 below summarises the average duration of the different periods through the diagnosis pathway from the patient’s 1^st symptoms on up to follow-up by the neurologist.

Fig. 2.

Average time from the first symptoms experienced by the patient to the 1^st treatment by the neurologist after further investigations

Discussion

For the first time, our study examined the diagnostic and therapeutic pathway using a chronological approach based on data from primary care settings to determine diagnostic and treatment delays for PwP. Further research in this field is required to address diagnostic delays and errors, which can result in harmful delays in treatment initiation and negatively impact the quality of life for PwP [15].

The diagnostic process starts with the symptoms reported by patients. Their comprehensive and continuous documentation in the medical record is crucial in identifying early symptoms, which, taken in isolation, would not have aroused suspicion of PD. During the follow-up of patients, when new clinical symptoms appear or existing symptoms increase in frequency or worsen, a retrospective analysis of the medical file can be used to identify clinical trends and developments that are less apparent during one-off consultations. The GP is at the forefront of identifying these combinations and changes in symptoms that may be indicative of PD [6]. It is important to note that a combination of five prodromal non-motor symptoms can be easily and quickly assessed by GPs. These symptoms include constipation, violent nightmares, urgent urination, hyposmia, and anxiety-depressive syndromes [3–6]. Furthermore, sleep disorders are prevalent among PwP, with insomnia, excessive daytime sleepiness, circadian disorders, obstructive sleep apnoea, restless legs syndrome and rapid eye movement sleep behaviour disorder being among the most common [17].

Our findings indicate that the time taken to diagnose and initiate treatment is significantly delayed (tripled compared to motor symptoms) due to a lack of awareness among GPs and patients of often non-visible non-motor symptoms.

Patients themselves may understate or deny the importance of these non-motor symptoms, which can delay medical consultation and diagnosis.

As stated by Nomoto et al., patients’ inadequate or misinterpreted information about PD may result in delays in accessing healthcare services, leading to diagnostic delays [18].

Non-motor symptoms, which are often isolated and unspecific, can be easily overlooked or attributed to other causes, such as physiological ageing. Should these symptoms appear in combination, it is vital that the GP is alerted and encouraged to explore the diagnostic hypothesis [4, 6, 19, 20].

A recent study has confirmed this diagnostic difficulty based on non-visible (non-motor) symptoms. Indications of constipation and sleep disturbances may emerge 15 years prior to the manifestation of the parkinsonian triad (resting tremor, plastic rigidity, akinesia), which is swiftly identified by medical professionals [20].

Our study provides further confirmation of this hypothesis, showing that the time between the onset of symptoms and the initiation of treatment by a GP (1,324 days) is reduced by 10 days (to 133 days) when motor symptoms are predominant.

On average, the time between the initial suspicion of PD and referral to a neurologist for motor symptoms is reduced by two-thirds (28 days), in comparison with non-motor symptoms. In practice, it is often more straightforward to identify motor symptoms that are visible or reported by relatives.

However, Nomoto et al. have demonstrated a delay in diagnosis when faced with motor symptoms attributed by patients themselves to ageing. In their study, tremor (42%) was the most frequent symptom leading to a doctor, whereas gait disturbance was the least frequent (14%) [18]. In addition to the misinterpretation of symptoms, patients may be in denial about their condition as a defence mechanism. This can help a person to cope with the different stages of illness and treatment. However, it can also lead to a delay in seeking treatment, impaired adherence and reduced self-management, resulting in undesirable outcomes [21]. GPs must be able to recognise denial when it occurs and develop strategies to assist their patients in overcoming it.

Among our GPs, some took more time to observe the evolution of symptoms before suggesting PD. This attitude is understandable and justified in view of the polymorphous nature of possible symptoms, particularly prodromal ones such as the shoulder disorders described by Paggou et al. [22]. According to the latest research, shoulder pain, frozen shoulder and arm swing asymmetry are the most prevalent clinical findings preceding cardinal PD symptoms. Shoulder MRI or US detects common abnormalities in mild or severe PD stages, including supraspinatus tendon thickening or tearing, adhesive capsulitis, acromioclavicular changes, bursa and joint effusion. Fractures resulting from falls or osteoporosis are a secondary shoulder health concern. These orthopaedic conditions are prevalent in GP surgeries, and it is logical for GPs to attribute them primarily to other aetiologies.

However, the time between referral and the first consultation with the neurologist may be influenced by the ease or difficulty of accessing the neurologist.

GPs with specific training in PD are often able to formulate a diagnostic hypothesis swiftly and efficiently refer patients to their trusted neurologist, facilitating prompt appointments and advanced prescriptions for medication. This enables the GPs to titrate the minimum effective dose with the patient, monitor the clinical course, and manage any adverse effects. The development of comprehensive, inter-professional support initiatives has the potential to reduce hospital admissions and repeat consultations. This collaborative approach between professionals is crucial and emphasises the significance of interdisciplinary and interprofessional practice in facilitating the pathway for PwPs [23].

GPs will immediately suspect PD if they see the cardinal motor symptoms of PD, such as akinesia, tremors and rigidity. In 70% of patients referred by GPs to neurologists, this hypothesis was confirmed, underlining the importance of GPs having legitimacy in diagnosing PD at an early stage in GP surgeries. Approximately one-third of patients (n = 13) had undergone complementary exams prescribed by their general practitioner. GPs are requesting brain MRIs to rule out differential diagnoses, such as multiple sclerosis, for patients with non-motor symptoms.

One patient (6%) had a DaTscan prescribed by a PD-experienced GP for a 29-year-old woman with an atypical clinical picture, including gastroparesis in the context of early-onset PD. This minimal prescription of DaTscan is consistent with Isaacson’s findings [13].

However, the process of waiting for a definitive diagnosis and, above all, the introduction of a drug treatment, meant that patients (in our sample) lived in uncertainty for almost 6 months, during which time they continue to experience symptoms daily, such as muscle pain and fatigue, which they frequently report to their GP, who accompanies them closely during this period of uncertainty.

Reddy et al. posit that the period of uncertainty could be reduced in the future by using artificial intelligence devices capable of identifying early combinations of symptoms suggestive of PD [24]. Furthermore, the MDS has developed a web-based prodromal PD risk calculator to estimate the probability of prodromal PD for individuals, which is available under license on their website [8].

Regarding the timing of treatment initiation, Bloem [4] and Lyons [25] advocate a swift commencement, even in cases of neurodegenerative disease, given the often-protracted period before the onset of functional disability.

As Bloem et al. also point out, there is no reason to delay symptomatic treatment for people developing disability due to PD. While Levodopa is the most common medication used as first-line therapy, optimal management should start at diagnosis using a multidisciplinary team approach, including the growing repertoire of non-pharmacological interventions such as speech- and physiotherapy, as well as psychological and social support [4].

According to Lyons et al., patients diagnosed with PD who receive treatment demonstrate improved outcomes in comparison to those who do not receive treatment, with those who receive treatment at an earlier stage exhibiting superior long-term results. Therefore, earlier initiation of a personalised treatment plan is crucial and must consider specific symptoms, severity, and lifestyle. In general, MAO-B inhibitors are suitable for initial treatment of mild symptoms, with the addition of a dopamine agonist in younger patients or levodopa in older patients, according to symptom evolution [25].

GPs can monitor the evolution of symptoms under dopaminergic treatment before deciding, together with the patient, whether to refer the patient to a neurologist to adapt the therapeutic strategy. GPs will also have a pivotal role in promoting change and maintenance of health behaviours and a lifestyle likely to slow the progression of PD and prevent its complications [26].

In elderly people with co-morbidities, the decision to initiate treatment for PD must be taken carefully, weighing up the potential benefits for quality of life against the risks of side effects. This has also had a consequential impact on the time taken to initiate first treatment. A cross-sectional study of the knowledge and practices of GPs practising in GP surgeries regarding the management of PwP showed that they were less comfortable initiating treatment for PD than geriatricians. Conversely, the initiation of dopamine replacement therapy increased with the GP’s years of experience [27].

Research indicates that the time lag between the onset of symptoms and the initiation of treatment by a GP can be reduced by additional training on PD, as evidenced by studies conducted in Canada [28–30]. In our study, six general practitioners were specifically trained in PD diagnosis and treatment, which saved patients a considerable amount of time (treatment was initiated three times earlier than when initiated by the neurologist), even when non-motor symptoms predominated.

The first author developed and delivered a two-day in-person training course for GPs in France, which has been recognised by the National Agency of Continuous Professional Development. To date, 50 GPs have been trained in early diagnosis of PD, including prodromal stages, and in the use of the MDS diagnostic criteria and MDS-UPDRS scale [12] and the initiation of symptomatically and PD-specific treatment [4]. It is vital for GPs to receive training and to take their rightful place alongside their patients and their neurologist colleagues, as well as all the other stakeholders involved in the care pathway for PwPs. This will optimise their early diagnostic and therapeutic pathway (drug and non-drug strategies), with the aim of preserving their autonomy and quality of life. In collaboration with the patients, they will oversee the care pathway, considering their personal preferences and support requirements. Further studies could assess diagnosis and therapeutic delays among these 50 GPs before and after the training course to determine if delays are reduced and more patients are diagnosed and treated. The use of self-administered patient questionnaires in primary care settings has the potential to facilitate the early detection of prodromal stages in patients over the age of 60, as outlined by Baldin [31]. In their study, constipation (27.8%) and hyposmia (19.9%) were more prevalent with age, but not pRBD (4.3%) [31].

Transcranial sonography could be promoted in primary care settings as a non-invasive, easily accessible complementary examination. This could be used to discriminate PD from essential tremor if necessary [14]. In conclusion, the diagnosis of PD remains challenging [32]. However, GPs can be trained by their peers to diagnose PD in its prodromal and early stages. This will allow them to contribute to the early diagnosis and decision-making process regarding shared drug or non-drug therapy in a team setting [23].

Strengths and limitations

This study is distinguished by its innovative use of a chronological approach, offering a comprehensive and longitudinal perspective on the diagnostic and therapeutic process initiated by GPs. GPs serve as the primary interface for patients experiencing motor and/or non-motor symptoms indicative of PD.

Wan et al. used a similar approach among 131 newly diagnosed PwP in China [33]. A full record of the patient’s demographic and clinical characteristics, as well as their detailed clinical history, was obtained from the patient files. The diagnostic latency period lasted approximately 15 months and was found to be closely related to the severity of motor symptoms, anxiety and depression, as well as the number of non-motor symptoms. Patients with limb rigidity, more non-motor symptoms, more severe motor symptoms and initial consultations with non-PD experts were all contributing factors to a longer diagnostic delay for PD [33].

In our study sample, 65% of subjects were male (n = 33), which is in line with national [34] and international data which report a 1.5-fold higher risk in men than in women [4].

The study revealed a higher-than-expected proportion of farmers (18%, n = 9) and unskilled workers (12%, 6%) among the PwP. This over-representation can be attributed to the geographical area of recruitment, the North of France, which is characterised by a large proportion of agricultural land. These results align with the existing literature, which highlights the role of occupational exposure to chemicals and pesticides as potential risk factors for PD [35–38]. The prevalence of a family history of PD in our cohort (16%) was also consistent with national data, which report a prevalence ranging from 10 to 15% [35] and international data [4].

We have identified potential declaration and selection bias.

The standardised questionnaire, although completed by healthcare professionals, remains declarative. All GPs are responsible for ensuring that their patients’ medical records are properly kept and authentic. It is reasonable to assume that GPs selected complete files rather than incomplete ones. It is also possible that some GPs who initiated PD symptomatic treatments for their patients without prior neurologist approval may not have selected these files if they were uncertain about the correct treatment.

A limited number of GPs completed their questionnaires in full, and the data available is subject to limitations arising from the time taken to complete the questionnaire. Each patient required completion of a new questionnaire, which took approximately 15 min. Some GPs chose not to continue with the questionnaire.

The initial plan was to include 70 cases, but in the end only 51 files were collected by the 20 GP investigators. It should be noted that this study was based on unpaid participation.

It is also conceivable that some GP investigators opted against including data in their files due to inadequate information. Nevertheless, they offer a unique perspective on the diagnostic delay and the various symptoms, which has not been explored before.

Whilst the size of our patient sample may be considered small, it was purposively selected as it ignored the number and profile of PwP reported by the investigating GPs, who, in turn, composed a convenient sample.

Conclusions

The study’s findings align with our initial hypothesis, confirming that the diagnostic pathway for PD results in significant delays from the onset of symptoms to diagnosis and the initiation of drug treatment. It has been established that these delays are linked to a lack of recognition of prodromal, non-specific, non-motor symptoms of PD by both patients and GPs. On average, GPs referred patients to a neurologist after 28 days if they were experiencing motor symptoms. However, if patients did not have motor symptoms, they were referred on average after 93 days.

Furthermore, the observation periods of clinical signs by the doctor, the difficulties in accessing neurologists and additional examinations to confirm or refute the diagnosis of PD can lead to a diagnostic delay of up to three years. This in turn can delay the introduction of drug-based and non-drug therapies.

The hypothesis of PD put forward by the GPs was confirmed by neurologists in most cases, which serves to substantiate the efficacy of first approaching a GP.

Peer-to-peer training for GPs has the potential to streamline care pathways and reduce delays, especially in the prodromal and early stages of the disease during the pre-diagnosis period, which can be a source of uncertainty for patients.

Abbreviations

PD

Parkinson’s disease

GP

General practitioner

PwP

Person with Parkinson’s Disease

Authors’ contributions

SB had the original idea for the study design and wrote the first draft. XL recruited the GPs and collected, analysed and interpreted the patient data regarding the different timelines/delays between the steps of the diagnostic and therapeutic health course. MC performed the statistical analysis. MB performed the language editing, NM added critical expert view to the discussion. CC and TR reviewed the article. All authors read and approved the final manuscript.”

Funding

The study was not funded.

Data availability

Data are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

In accordance with the declaration of Helsinki, on 19 April 2023 this study received ethical approval from the Data Protection Officer of the institutional review board (IRB) of the University of Lille. According to the current French legislation LAW no. 2012 − 300 of 5 March 2012 on research involving the human person (LOI Jardé), the need for ethics approval was waived by the IRB.

For this study design the need for consent to participate was waived by the Institutional Review Board of the University of Lille.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.