Disease forecasts are market analytics, aimed at understanding and predicting a product’s market potential. Prediction modelling usually involves a prediction profile (algorithm) from past studies. The model is then “deployed” so that a new individual can get a prediction instantly for whatever the need be, such as an epidemiological endpoint. Predictions can range from variations in disease prevalence to changing diagnostic rates and responses to medications.
Predictive analytics increase the accuracy of diagnosis, help to develop better preventive medicine and provide predictive models that become more accurate over time. Patients may enjoy better outcomes due to such analytics, being better informed, with physicians also taking on the roles of consultants. For hospitals, biopharma companies and insurance providers, accurate disease forecasts will not only save time spent on analyses but also provide detailed disease epidemiology for optimal business strategy planning.
The challenges: the example of multiple sclerosis
Multiple sclerosis (MS), a chronic inflammatory disease of the central nervous system, predominantly affects young women, with an increasing prevalence.1 Forecasting epidemiological data for this disease presents multiple challenges due to the varying population dynamics, diagnostic and treatment challenges observed. These factors are briefly detailed here, together with emerging tools such as shared decision-making for MS treatment.
Gender bias: Relapsing-remitting is the most common MS subtype; up to 85% of MS patients belong to this category (range 48% to 91%), characterized by the acute worsening of neurologic function followed by variable recovery, with a stable period between attacks.2,3 A marked predominance of females is noted among relapsing-remitting MS patients and gender ratios of 3:1 have been reported.4
Genetic risk: MS has a familial recurrence rate of about 20%.5 The disease is most frequent in white Europeans but is also seen in non-white populations, including African-Americans and south Asians. Genome-wide association studies focusing mainly on European populations have identified rs7318477 as the single-nucleotide variant most associated with an increased risk of developing MS in south Asian cases. Disease etiology thus is independent of ethnicity.1
Prevalence: Incidence and prevalence of MS varies geographically. In Europe, Canada and the United States, prevalence rates range from 60 to more than 100 cases per 100,000 inhabitants. Data from African and Asian countries suggest prevalence lower than 40/100,000 for those regions.
Differential diagnosis: MS diagnosis is based on neurological symptoms and signs, together with the spread of central nervous system lesions. MRI alone may not confirm diagnosis, and further supportive information is required from cerebrospinal fluid examination and neurophysiological testing. In the absence of a specific diagnostic test to detect MS, this disease is often ascertained by differential diagnosis.
Silent disease: It is presumed that the disease is far more common than suspected. MS may progress as a silent/asymptomatic disease, in which case the exact number of afflicted persons remains unknown. Typical MS plaques were reported when MRI scans were performed for various complaints in patients not suspected of having MS, and in the siblings of MS patients. Other studies also reported finding unexpected, asymptomatic MS lesions at autopsy (overall prevalence of 100 per 100,000).6
Misdiagnosis: A number of surveys have reported the misdiagnosis of MS. One such report included North American MS specialists who evaluated patients carrying a diagnosis of MS (given by another provider) for longer than a year. The specialists then conducted a neurologic exam and reviewed lab data and stated that the patient did not have MS and had been misdiagnosed. In this survey, up to 95.1% of the specialists had evaluated such a patient in the past; in the past year, about 40% of the specialists estimated that they had seen three to five such patients; more than one-third (34.4%) reported seeing six or more misdiagnosed patients in the last year, including 20 (17.2%) respondents who had seen 10 or more such patients.7
Disease-modifying therapy: Several treatments are now available for relapsing-remitting MS, but effective therapies for progressive MS are still under development. Treatment decisions are challenging due to varying risk-benefit profiles associated with new therapies, and the absence of reliable treatment response biomarkers. Adverse effects are feared when misdiagnosed MS patients are put on disease-modifying therapies (as noted in 26% to 75% of misdiagnosed patients in the Solomon study). 7
Adherence: While 75% of patients are reported to adhere to their prescribed treatment regimens, the rate of patients not missing a single dose differed depending on the treatment route and number of interventions. In a study by the Devonshire group, patients receiving intramuscular (i.m.) interferon (IFN) beta-1a (applied 4 times per 4 weeks) were more treatment adherent than patients receiving all other disease modifying therapies (subcutaneous [s.c.] IFN beta-1a 22 mcg [12 times per 4 weeks]; s.c. IFN beta-1a 44 mcg; IFN beta-1b [14 times per 4 weeks]; glatiramer acetate [GA] [28 times per 4 weeks]).8
Ineffective treatment: MS is usually diagnosed in the third and fourth decades of life and can lead to fixed and/or progressive neurological impairment. The last years have seen the advent of both disease-modifying therapies and new symptomatic therapies (such as pain management).9 However, since neurological disability caused by MS expands during progressive forms of the disease, the 14 anti-inflammatory drugs that have regulatory approval for treatment of the multifocal inflammatory lesions (seen in relapsing-remitting MS) have little or no efficacy against the increasing disability in progressive MS (that show no inflammatory lesion activity). The inability of disease-modifying therapies to improve chronic neurologic deficits has left progressive MS patients still experiencing a variety of neurologic symptoms that often interfere with daily activities and reduce quality of life.10
Shared decision-making: The present scenario focuses on the concepts of personalized therapy and early treatment for relapsing-remitting MS and emerging breakthroughs in progressive MS treatment. A more proactive management strategy, including earlier use of high efficacy disease-modifying therapies and close monitoring of the clinical/radiological response to treatment, is recommended to slow the progression of physical and cognitive impairments in patients with relapsing-remitting MS. Effective therapies are in the pipeline for progressive MS that prevent worsening, reverse damage and restore function.
Before starting a treatment, neurologists are now carefully assessing the state of the disease, prognostic factors and comorbidities, the patient’s response to earlier treatments, and if the patient would accept treatment-related risks. The early detection of suboptimum responders (following accurate clinical monitoring) is allowing clinicians to redesign treatment strategies whenever necessary.
In this context, shared decision-making is a valuable tool in the clinical care of MS. Patients are acquiring a better understanding of both their disease and the benefits/risks of treatments. Patients with MS who are well informed about their disease have significantly higher medication adherence than those who are not.
Accurate forecasting for MS
Predicting future trends for MS poses the critical question:
How can we forecast diagnosis/treatment rates for the coming years, given that any predictions made now could change in the early next years as a result of the intensive research ongoing to develop better diagnostic tools and treatment options?
The answer to this question lies with customized disease forecasts. With current advances in magnetic resonance imaging (MRI) and serological/genetic testing, more accurate numbers and assumptions are expected in these domains. Predictive analytics are now assimilating these evolving developments thereby creating customized disease forecasts for the coming 15 to 20 years.
While baseline forecasts are provided for each data point (e.g., prevalence rate, segmented prevalence rate, diagnosis rate, treatment rate), new assumptions/input variables can be entered for any time point, which can relate to an increase/decrease in the data point (prevalence/diagnosis/treatment), beginning of change (month/year) and transition time (in quarters). The customized forecast data thus provide the cumulative effect of the separate input variables, and present the impact of each assumption/data. Comparative analyses of the baseline prevalence, diagnosed and treated populations (when no assumptions were added) to the corresponding populations when customized data were entered, provide a working report created for accurate disease forecasting.
With emerging improved research criteria and reliable diagnostic and/or treatment rates providing new trends, such novel disease data need to be integrated into the forecast reports, cumulatively creating more realistic forecasting. Since every disease is unique, epidemiology-based forecasting tools should to be designed individually, with each disease forecast pertaining to a particular disease and built in such a way, so as to showcase the most relevant data offered.
Clarivate Analytics now provides Disease Forecasts for MS and numerous other diseases, where new input variables can be integrated for reliable market analytics. Learn more about Disease Forecasts, customizable reports for serious disease forecasting with full source traceability.
These forecasting reports are built on and linked to Clarivate’s Incidence and Prevalence Database, with a 15-year market forecast for a single disease in G7 (U.S., Canada, France, Germany, Italy, Japan, UK) and BRIC (Brazil, Russia, India, China) countries.
- Pandit L et al; “European Multiple Sclerosis Risk Variants in the South Asian Population.” Multiple Sclerosis; V.22; No.12; 10/16; p1536; DOI: 10.1177/1352458515624270.
- Cristiano E et al; “The Epidemiology of Multiple Sclerosis in Latin America and the Caribbean: A Systematic Review.” Multiple Sclerosis; V.19; 6/13; p844; DOI: 10.1177/1352458512462918.
- Applebee A; “The Clinical Overlap of Multiple Sclerosis and Headache.” Headache; V.52; 10/12; p111; DOI:10.1111/j.1526-4610.2012.02243.x.
- Kalincik T et al; “Sex as a Determinant of Relapse Incidence and Progressive Course of Multiple Sclerosis.” Brain; V.136; 12/13; p3609; DOI:10.1093/brain/awt281.
- Compston A et al; “Multiple Sclerosis.” The Lancet; V.372; 10/25/08; p1502; DOI: 10.1016/S0140-6736(08)61620-7.
- Poser CM et al; “The Accuracy of Prevalence Rates of Multiple Sclerosis: A Critical Review.” Neuroepidemiology; V.29; 2007; p150.
- Solomon AJ et al; “Undiagnosing Multiple Sclerosis: The Challenge of Misdiagnosis in MS.” Neurology; V.78; 6/12/12; p1986; DOI:10.1212/WNL.0b013e318259e1b2.
- Devonshire V et al; “The Global Adherence Project (GAP): A Multicenter Observational Study on Adherence to Disease-Modifying Therapies in Patients with Relapsing-Remitting Multiple Sclerosis.” European Journal of Neurology; V.18; 2011; p69; DOI:10.1111/j.1468-1331.2010.03110.x.
- Hadjimichael O et al; “Persistent Pain and Uncomfortable Sensations in Persons with Multiple Sclerosis.” Pain; V.127; 2007; p35).
- Goldman MD et al; “Multiple Sclerosis: Treating Symptoms, and Other General Medical Issues.” Cleveland Clinic Journal of Medicine; V.73; No.2; 2/06; p177.