Patients with ALS Survey (PALS)
Greater Plains Collaborative
Coordination Center:
Department of Neurology,
University Kansas Medical Center
Kansas City, Kansas
913-588-6970
Protocol Version 1.0
Table of Contents
1.0Purpose of Research...... 3
2.0Specific aim...... 3
3.0Prevalence, Background and Significance...... 3
3.1Prevalence:...... 3
3.2Background...... 3
3.3Greater Plains Collaborative (GPC)...... 4
4.0 Study design and rationale 5
4.1Inclusion Criteria:...... 5
4.2 Exclusion criteria...... 5
4.3Procedures...... 5
4.4 Recruitment………………………………………………………………………..5
4.5Adverse events...... 5
5.0Data Management and Case Report Forms...... 6
6.0Biostatistical Approach: Outcome Measures, Data Analysis and Power...... 6
7.0Benefits and Risks of Research...... 6
7.1Known and Anticipated Risks...... 6
7.2Anticipated Benefits...... 6
7.3Privacy...... 6
7.4Informed Consent...... 6
7.5Changes to Protocol...... 6
7.6Maintenance of Records...... 6
7.7Early Termination...... 6
7.8Inclusion of women...... 6
7.9Inclusion of minorities...... 7
8.0Data and safety monitoring plan...... 7
8.1 Data Backup and Storage………………………………………………………7
9.0Reference...... 8
Appendix 1: GPC Survey
Appendix 2: Pioneer Recruitment Registry Online Consent
Appendix 3: Letter to subjects/instruction sheet
1.0Purpose of Research:
Our PCORI-funded Greater Plains Consortium (GPC) network is a powerful new approach for utilizing electronic medical records (EMR) to collect and analyze clinical data to improve patient care and outcomes. Patients with amyotrophic lateral sclerosis (ALS) have a rapidly progressive neurodegenerative disorder and often experience many functional changes between routine clinic visits. The ability to track changes in functional status via surveys completed by the patient at home would help practitioners better assess therapeutic interventions and maximize care. In addition a better understanding of the demographic distribution of patients with ALS across the greater GPC network, and differences in survey response rates in different ALS patient groups, will be important as we design and implement new interventions to improve ALS patient care. We developed a GPC survey which incorporates elements of a standard ALS functional assessment tool, along with specific questions identified as being important in a patient focus group meeting. In order to demonstrate proof of concept for our GPC network, we plan to identify ALS clinic patients using our EMR and send our new GPC survey. Response to this survey will demonstrate the feasibility of our GPC ALS network, and the possibilityofusing apatient-reported survey to follow functional status in ALS patients between clinical visits. This collaboration between the patient and their health care provider team will broaden the tools available for improving clinical care.
2.0Specific aims
Specific Aim #1:
We will demonstrate proof of concept of using the GPC networkto track patient reported outcomes. To achieve this aim, we will send a one-time survey to all ALS Clinic patients at each of the GPC ALS network sites. We will utilize a scheme where patient surveys will be given an unique identifier which can be associated back to the EMR to obtain limited demographic data. However we will set up local ‘honest brokers’, to serve as a firewall between the survey response data and data obtained from the EMR: final analysis for all aims will be performed on de-identified data sets.
Specific Aim #2:
We will determine the clinical diagnosis and basic demographic characteristics of patients in our GPC ALS network, and compare across GPC sites. We will comparereposnders to our survey to non-responders. All data included in analysis will be de-identified from the EMR.
Specific Aim #3
We will determine response rates across different patient options for returning our survey. We will compare response rates between different methods: individual email links to surveys; online RedCap surveys, or mail in/in person response in clinic.
3.0Prevalence, background
3.1Prevalence:
ALS is a rare neuromuscular disorder with an estimated prevalence of 40 to 60 cases per 1,000,000 and an incidence of 0.4 to 1.8 per 100,000 (Annegers, Appel et al. 1991). We estimate the ALS population in the US to be 12,000 to 24,000 with 5,000 new cases diagnosed each year, based on a US population of 320,000,000.
3.2Background
Amyotrophic lateral sclerosis is a progressive disease characterized by degeneration of the voluntary motor system. It has a median age of onset of 55 years with a male predominance of 1.5 to 1 (Norris, Shepherd et al. 1993). There is no racial or geographic predisposition except for an increased incidence on the Marianas Islands of Guam. The diagnosis is based on the El Escorial criteria (Brooks 1994; Brooks, Miller et al. 2000), which define the key clinical and electrophysiologic signs of both upper and lower motor neuron dysfunction. Weakness and atrophy begin either in bulbar or limb muscles and spreads to contiguous myotomes. Respiration is usually affected late in the disease. Patients with older age at onset, bulbar dysfunction, greater clinical disability, and poor respiratory function have the poorest survival (Gubbay, Kahana et al. 1985; Eisen, Schulzer et al. 1993). While 10-15% has a family history of ALS, the majority of cases are sporadic (7). The majority of patients followed in multi-disciplinary ALS clinics will have a clinical diagnosis of ALS. But motor neuron disorders exist on a spectrum from pure upper motor disfucntion (primary lateral sclerosis, PLS), to mixed upper and lower motor neuron dysfunction (ALS), to pure lower motor neuron involvement (progressive muscular atrophy, PMA). In addition some patients followed in the ALS clinic will have symptoms confined to the bulbar region, a condition termed progressibe bulbar palsy (PBP). There is considerable overlap between these disorders, and some controversy whether PMA and PLS may simply be varients of ALS. The prevalence of these other motor neuron disorders are assumed to be low compared to ALS, but the actual prevalence in the ALS clinics are not known.
ALS Scale - Revised (ALSFRS-R Functional Rating)
Functional rating scales have become the standard primary outcome measure for clinical trials of neurodegenerative diseases including Parkinson’s disease, Huntington’s disease, and Alzheimer’s disease. Functional scales predict disease progression and measure activities of significance to patients. The ALS Functional Rating Scale (ALSFRS) was designed to assess the ability of ALS patients to perform activities of daily living and to detect functional changes during clinical trials(1996). Precedent for using this scale in clinical trials stems from the only positive treatment trial of riluzole in ALS, when a predecessor of the current ALSFRS-R, showed slower decline in treated patients relative to placebo(Bensimon, Lacomblez et al. 1994; Miller, Bouchard et al. 1996). A revised version of the ALSFRS, the ALSFRS-R, encorporates questions about respiratory function into the scale, improving its sensitivity, and has become the current standard outcome measure for ALS studies(Cedarbaum, Stambler et al. 1999). The ALSFRS-R is a quickly administered,by research personnel or study staff, (five minutes) ordinal rating scale that assesses capability and independence in 12 functional activities. These include six bulbar-respiratory functions, three upper extremity functions (writing, cutting food, and dressing), and three gross motor functions (walking, climbing, and turning in bed). Each activity is recorded to the closest approximation from a list of five choices, scored 0-4, with the total score ranging from 48 (normal function) to 0 (no function). The ALSFRS-R has been used extensively in previous clinical trials and validity has been established by correlating ALSFRS-R scores with quantitative strength testing and changes in strength over time(Gordon, Moore et al. 2007; Miller, Bradley et al. 2007; Lauria, Campanella et al. 2009; Moviglia, Moviglia-Brandolino et al. 2012; Sacca, Quarantelli et al. 2012; Beghi, Pupillo et al. 2013; Cudkowicz, van den Berg et al. 2013; Dorst, Cypionka et al. 2013; Shefner, Watson et al. 2013). The ALSFRS-R can also be used to follow disease progression in clinic. Limited studies have assessed patient-reported versions of the ALSFRS-R(Montes, Levy et al. 2006; Kaufmann, Levy et al. 2007). The ability to use a patient-reported tool like the ALSFRS-R to assess patient function between clinic visits from the home would be a powerful new tool for clinicians to monitor the effect of interventions including: medications, orthotic devices, and respiratory or feeding support. In order to make this possible we developed a new patient-reported survey which incorporates many of the key elements of the ALSFRS.
Greater Plains Collaborative Patient Reported Outcome Measures
The ALSFRS- R was sent to approximately twenty ALS patients throughout the Greater Plains Collaborative region and their caregivers. They were asked if the scale was difficult to understand; if there were items they felt should be dropped from the ALSFRS-R or new items that should be included. There were 2 focus group conference calls where the patients gave their opinion of the scale. Patient focus group recommendations were that there needed to be some clarification in meanings of some of the words in the ALSFRS-R, but overall the impression was that this instrument reflected the types of functional limitations they experienced on a daily basis. There were also several items they believed should be added to the ALSFRS-R: a question about pain; a question about emotional liability; and a general non-demoninational question about faith. Our new GPC survey follows the suggestions of the patient focus groups and modifies the language of the ALSFRS-R and adds suggested items.
3.3Greater Plains Collaborative (GPC)
The Greater Plains Collaborative (GPC) is a PCORNet Clinical Data Research Network (CDRN) composed of 10 leading medical centers repurposing the research programs and informatics infrastructures developed through Clinical and Translational Science Award (CTSA) initiatives. Partners are the University of Kansas Medical Center (KUMC), Children’s Mercy Hospital, University of Iowa Healthcare, the University of Wisconsin-Madison, the Medical College of Wisconsin, and Marshfield Clinic, the University of Minnesota Academic Health Center, the University of Nebraska Medical Center, the University of Texas Health Sciences Center at San Antonio and the University of Texas Southwestern Medical Center. The GPC network brings together a diverse population of over 10 million people across 1300 miles covering 7 states with a combined area of 679,159 square miles. Of these, over 6 million have significant data maintained in electronic health records. This population covers the spectrum from primary care networks serving rural and small communities to urban populations with significant African American and Hispanic representation. ALS clinics exist at nine of the ten medical centers in the GPC. The GPC selected amyotrophic lateral sclerosis (ALS) as the rare disease for which we could readily access all patients across our network and is working under the direction of Dr. Barohn to survey the population regarding their willingness to engage in research and provide patient reported outcome measures.
Dr. Waitman (GPC principal investigator) is working with the informatics site leads to tailor existing electronic medical record systems, data repositories based upon the i2b2 software developed through the National Center for Biomedical Computing at Partners Healthcare System, data capture systems based on REDCap (Research Electronic Data Capture) developed by Vanderbilt University, and governance processes to support comparative effectiveness research in alignment with PCORNet objectives. The GPC complements considerable investments in electronic health records by our healthcare systems with existing NIH-funded technology (e.g., i2b2, REDCap) to provide a cost-effective common data model that promotes data transparency and interoperability. This includes:
- Collecting Patient Reported Outcome Measures (PROM) standardized measures deployed using either EMR patient portals or data collection instruments for existing registry and research management systems such as REDCap.
- Configuring comparative effectiveness trial components directly in the EMR (preferred) or integrate existing data capture and trial management systems.
Tailoring existing methods (a lightweight i2b2 plug-in) so that each site’s honest broker can extract limited data sets composed of EMR and PROM and securely transfer them to the GPC data store to support the conduct of the comparative effectiveness research (CER) trial monitoring and analysis.
4.0Study Design and Rationale
The study is a prospectively collectedone-time survey (Appendix 1) that will be sent to all ALS Clinic patients at the nine ALS sites within the GPC. We anticipate that each site follows between 30-230 (depending on location) patients in the ALS Clinic. Each survey will contain a cover letter explaining the survey, and why they are being asked to participate, and instructions for comlpeting the survey.
4.1Inclusion Criteria:
Any patient seen in one of the 9 particpating GPC network ALS Clinics within the last two years.
4.2 Exclusion criteria
Any patient not willing to participate.
4.3 Procedures:
Each participant will receive a letter telling them about the upcoming survey and asking for their participation approximately 1 month prior to the survey being distributed.
A cover letter describing the survey with instructions for completing the survey will be sent to all qualifying patients seen at the nine GPC ALS Clinics. The survey will be sent by email (depending on site requirements), through my chart (if the patient has signed up), or by mail. Patients will have the following options for returning the survey: 1) for patients with a current email address or they have registered through mychart, they can follow a link embedded in their email to their individual survey; 2) patients with internet access can go to a secure RedCap website to enter their unique survey identifierthat will take them to their individual survey; and 3) patients can elect to return the paper survey by mail, or return it in person at their next clinic visit. In addition patients can complete the surveys on-line or in person at their clinic visit. Two weeks after the surveys have been distributed local coordinators from each site will call their own patients to find out if their have received their survey, and answer any questions about the purpose of the survey, or how to complete the survey.
4.4Recruitment
Each of the nine GPC - ALS network sites will provide a list of patients with the ICD-9 code of 335.20, 335.24, or 335.29seen in the ALS clinics within the last two years to the local PI or study coordinator. The PI or study coordinator will confirm if these patients are followed in their clinics, not deceased and otherwise rectify the list to be inclusive. The information technology officer (ITO) will create a final list comprising the clinic ALS patient lists, and possible patients identified in the EMR. After this list is rectified, a letter will be sent to all study participants.
4.5Adverse events
This is a survey and we do not anticipate any adverse events from this study.Participation is voluntary, and patients can chose not to participate.
5.0Data management and case report forms
We will utilize a schematic for data flow where each site will have an honest broker (the local ITO)who can extract limited data sets composed of EMR and GPC survey response data and securely transfer them to the GPC data store to support the conduct of the comparative effectiveness research (CER) outlined in this protocol. At each site the lists of qualifying ALS clinic patients will each be given an unique identifier generated for the GPC survey: each survey will have a unique identifier that will link to that patients EMR. The local ITO will serve as honest broker, forming a firewall between the data collection and analysis. Survey responses will be entered either electronically or manually into the local database managed by the local ITO, and maintained in a password secure REDCAP database on the local University secure IT network. We will allow 3 months from distribution of the surveys for all survey responses to be completed. When we are ready for analysis, the sponsor (University of Kansas Medical Center) will approach the local ITO to retrieve the GPC survey data and selected EMR information for the entire cohort which will be transmitted in a de-identified data set. The following EMR data will be included for analysis: age, gender, race, ethnicity, living situation, occupation, and years of education.
6.0Biostatistical Approach: Outcome Measures, Data Analysis and Power
The goal of the GPC network survey is to provide proof of concept that we can utilize the network to reach out to patients, and the patients will repsond: we will measure the response rate to the survey as the primary study outcome metric and consider the study successful if 50% of patients return thecompleted surveys. Basic descriptive statistics will be used to describe the overall ALS clinic cohort: including basic demographics (age, gender, race, ethnicity, living situation, occupation, years of education) and functional burden as identified by survey response (clinical diagnosis, and functional status). Investigations of participation response rates by site, demographics, GPC survey, and by method of response (email versus mail) will be performed to determine which patients are more likely to return a survey. We will perform simple comparison of binomial proportions (participation/no participation) by these covariates with contingency table analysis along with measures of association (e.g., odds ratios for 2 x 2 comparisons). Analyses will be extended to the use of unconditional logistic regression analysis to look at the joint effects of multiple covariates. Adjusted odds ratios (and 95% confidence intervals) will be use to summary the magnitude of association between a covariate and participation.