Frequently Asked Questions About the NIDDK Central Repositories
Frequently asked questions about the NIDDK Central Repository can be found below:
- About the Repository
- Submitting Samples from NIDDK Grants
- Obtaining Samples for Research
- Sample Confidentiality
- Sample Ownership
- Study-specific Questions
About the Repository
What is the purpose of having a central repository?
A central repository collects samples and data from designated NIDDK studies, simplifies access to samples, and ensures that samples are stored under uniform conditions. This allows for additional studies to be conducted, enhancing the value of the research. Similarly, reposited datasets can be maintained, queried, and reconfigured to facilitate new analyses after a study's original data coordinating center closes.
What are the NIDDK Central Repositories?
The NIDDK Central Repositories are two separate contract-funded components that work together to store data and samples from significant, NIDDK-funded studies.
The first component is the Biorepository that
- gathers, stores, and distributes biological samples from studies
- works with investigators in new and ongoing studies as a real-time storage facility for archival samples
The second component is a Data Repository that
- gathers, stores and distributes incremental or finished datasets from NIDDK-funded studies
- helps active data coordinating centers prepare databases and incremental datasets for archiving and for carrying out restricted queries of the stored databases
- serves as a Data Coordinating Center and website manager for the NIDDK Central Repositories website
Submitting Samples from NIDDK Grants
Eligibility for Submission
Who can submit samples and data?
Only significant NIDDK-funded studies designated by NIDDK will submit samples or data to the central repositories. In general, investigators funded through individual R01s will not be expected to submit samples and data.
Are completed or currently ongoing NIDDK-funded multi-site studies required to submit samples and data to the repositories?
NIH policy is that the unique resources developed through NIH funding are to be shared with the research community, to the extent that sharing of clinical data and materials is practical. Therefore, all current and complete multi-site NIDDK-funded studies should have a plan or should develop a plan for sharing data and samples that complies with current NIH and NIDDK policies (PDF - 28KB). Wherever appropriate, the NIDDK Central Repositories should be used to facilitate sharing.
Cost of Submission
Is there any cost to investigators to store samples or data to the repository?
What will be the proprietary period for stored samples and data?
The proprietary period for each study's materials will be mutually agreed upon by the study's Steering Committee and NIDDK. For multi-site clinical studies, NIDDK policy (PDF - 28KB) defines this proprietary period in terms of specific study milestones.
Submitting Data to the Data Repository
What data should be sent to the Data Repository?
In general, the repository should receive all of the data being collected by the Data Coordinating Center, provided that the subject has consented to having his/her data included in the repository. However, no personal identifiers should be sent to the repository, and data may need to be transformed in some way to ensure that no personal identifiers are reposited.
When should the linkage file be provided?
A linkage file is a list that associates biological samples with the relevant clinical data by matching the IDs that were used for the bio-samples to those that were used for the clinical records for all of the study subjects. The linkage file should be provided by the DCC at the time the clinical data is delivered to the Repository for completed studies.
What interaction will the Data Coordinating Center have with the Data Repository before data are sent?
The repository will work with the Data Coordinating Center during the study to prepare the data set for archiving. Repository personnel will consult with the Data Coordinating Center to develop appropriate documentation that is comprehensive enough to allow investigators not familiar with the data set to use it. In addition, the repository personnel will interact with the Data Coordinating Center to familiarize themselves with the database and its architecture in preparation for archiving. Finally, the Data Repository will work with Data Coordinating Centers of ongoing studies to provide customized views of regularly updated information about sample inventories in the Biorepository.
Should protected health information be sent to the Data Repository?
In general, the repository should only get limited data sets as defined under HIPAA regulations. In some cases, repository personnel may provide advice to Data Coordinating Centers on how to transform data to prevent transmission of protected health information.
Submitting Samples to the Biorepository
Which samples should be sent to the Biorepository and when should those samples be sent?
In general, the repository should be used to store samples that are not being used for the active conduct of the study. Whenever practical, part of each sample should be designated for the repository and shipped to the repository at the time of collection or sample processing. In general, the repository will not serve the day-to-day needs of active studies through extensive sample processing, aliquoting, and shipping aliquots to clinical labs or individual sites. However, the repository can aliquot and redistribute samples. In addition, the stored samples will be made available to the study if needed.
Who will be in charge of the samples at the Biorepository?
During the proprietary period, the study's Steering Committee will control access to the samples, and the Data Coordinating Center will work with the repository and be the primary contact with regard to the samples. However, once the proprietary period ends, the Data Repository will have the data associated with the samples, and will become the primary contact point with the regard to the samples.
Can the Biorepository do aliquoting?
Yes, the repository can aliquot samples.
Can the Biorepository serve as the primary receipt point for samples, processing, aliquoting and shipping samples out to various analysis laboratories?
No. In general, each study should have its own plan for sample processing. This can involve a central receiving laboratory that processes and analyzes samples, and stores the samples needed for the planned analyses. Alternatively, studies may have the samples processed at the collection site.
Can the Biorepository carry out sample processing or routine clinical analyses?
No. The Biorepository is not intended to serve as a clinical laboratory.
Obtaining Samples for Research
Available Samples and Data
What types of data/samples are available?
The list of studies available can be found at the Study Search page.
If my only goal is to use a dataset to demonstrate the statistical methodology and I do not plan to draw any scientific conclusions using the dataset, do I need to go through the IRB process?
For any level of access to the data, you must have your institution's IRB approval.
Do you have genotype data available? For which studies?
NIDDK works with dbGaP to make the genotype data generated in NIDDK-sponsored studies available through the NIDDK Repository. Please check the dbGaP Website for the study of interest.
What are renewable and non-renewable samples?
Renewable samples are DNA from cell lines and non-renewable samples are all samples EXCEPT DNA from cell lines.
Do you have a Type 1 Diabetes dataset in the NIDDK Repository? If so, what is the procedure to gain access to it?
We do have several T1D studies in the Repository. The list of studies available can be searched at the Study Search page.
Eligibility for Access
Who can access stored materials?
All qualified investigators will be allowed access to the stored materials at the end of a pre-determined proprietary period.
Can the study group that collected the samples retrieve some of the samples stored in the repositories?
Yes, although the use of archival samples from a large study should be minimized in order to preserve the integrity of the collection for future use. However, studies will always have access to the samples needed to carry out assays that are part of the approved study protocol.
Can non-U.S. investigators apply to use samples and data from the repository?
Yes. In general, all applications will be judged without regard to the geographical location of the applicant, although it is conceivable that there will be certain non-renewable samples collected in specific studies that are restricted to U.S. investigators by the terms of the study or of the consent. At this time, there are no such samples.
Can commercial organizations apply to use samples and data from the repository?
Yes. The data and sample request process is open to all organizations interested in using samples or data for research.
Are there special provisions for samples collected by international consortia?
For samples collected by international consortia, applications for use of samples and data will be considered regardless of the geographical location of the applicant institution or researcher.
Cost for Access
I am an Ancillary Sample Requestor, will I be charged for my sample request?
Yes. All requestors are charged the same prices to receive samples, regardless of the source.
Applying for Access
How do I obtain data from the Repository?
Go to the Data Request page.
Requestors need to provide the following forms:
- NIDDK Data Request Online Form
- IRB Clearance (or waiver)
- Signed copy of the Data Use Certification (DUC) Agreement provided to you by NIDDK Central Repository Staff after submission of the Online Form
If I have several datasets, may I use 1 extension letter to cover all of them?
Yes, you can use the same letter of extension, please list all of the data sets in the letter.
On the Sample and Data Use Certification and Data Use Certification Agreement forms, who should provide the Authorized Signature for my institution?
A Signing Official (SO) from your institution should provide this signature. An SO has institutional authority to legally bind the institution in grants administration matters. The individual fulfilling this role may have any number of titles in the grantee organization. For most institutions, the Signing Official (SO) is located in its Office of Sponsored Research or equivalent.
Who makes the decision about granting access to samples?
Requests for non-renewable samples are submitted in a number of ways. Applicants can submit an X01 grant application for access, but no funds. In addition, the NIDDK often has other Funding Opportunity Announcements (FOAs) specifically aimed at encouraging use of reposited samples. Relevant FOAs are listed at https://repository.niddk.nih.gov and NIDDK funding opportunities.
All applications are reviewed by a peer review panel and judged on numerous criteria, including the qualifications of the researcher and the research environment, the significance and appropriateness of the proposed research, design of the proposed research and ethical considerations including consistency with the terms of the informed consent obtained from the subjects. For renewable samples (principally, DNA from cell lines and whole genome amplified DNA), requests are reviewed administratively by NIDDK staff.
Is there a difference in consideration of requests for renewable (infinite supply) and non-renewable samples?
Yes. Requests for renewable samples, such as DNA from immortalized cell lines or DNA that has been 'whole-genome-amplified', will be approved if the project for which they are being requested is judged to have a reasonable likelihood of achieving its stated aims. Requests for non-renewable samples, such as plasma or urine, will be approved only if judged to have high scientific merit and therefore require a complete scientific justification.
If we were working together with a Study investigator, would this affect our situation and chances of accessing the data?
No. Each request is reviewed independently and all information is considered. If you want to include information that you are working with a Study investigator, the reviewers would consider that in the request. Ultimately, that information alone should not be a deciding factor for access to data.
During the proprietary period, while a study is ongoing, how do investigators gain access to samples?
During the proprietary period, investigators can apply directly to the study group to carry out an ancillary study. Ancillary studies that are approved by the study's Steering Committee for access to samples that are stored in the repositories use the Web-based ancillary study request process at https://repository.niddk.nih.gov to register the ancillary study and request specific samples.
I would like to receive more of the samples that I have requested or additional samples. What process should I follow?
For an ancillary study, contact the DCC of the study group. Otherwise, go to the Comments section of your request and ask for the additional materials, providing a justification.
How do you interpret the following section in Funding Opportunity Announcements: "Applicants must include in this section a statement of how the proposed research fits within the limitations of the subjects'' informed consent. Information about the informed consents for each study can be obtained from the NIDDK Repository."?
In general, samples and data stored at the NIDDK Repository are consented for additional research; however, some consents are conditional and/or may require further review by the Repository staff prior to approved release. All requests for samples and data are carefully reviewed against the consent forms by the Repository and NIDDK staff members to ensure adherence.
Under what conditions are samples distributed?
In order to receive samples or data, the institution will have to sign a Usage Agreement with NIDDK (https://repository.niddk.nih.gov/pages/agreement_forms/). The agreement will include certification that the project has IRB approval, that the investigators will only conduct the research consistent with the subjects' informed consent, and that researchers will not attempt to identify any individuals. In addition, the Agreement will forbid any redistribution of the materials by the investigator or institution.
How do I obtain samples?
- To request non-renewable samples, all requestors must apply via PAR-17-123 or PAR-17-270. In order to obtain the required information about sample availability, please fill out the online form and repository staff will reply to your request.
- To request renewable samples (specifically, DNA from cell lines or Whole Genome Amplified DNA), please fill out the online form and repository staff will reply to your request.
- Ancillary studies - please fill out the online form and repository staff will reply to your request.
Is there a cost to obtain samples?
Ancillary requestors and those not affiliated with the original study will be invoiced samples according to the prices outlined on our costs page. This page also contains the policies for submitting payment, including a requirement that all requests over $5000 or being shipped internationally pre-pay prior to the repository beginning preparation of the shipment.
Once my request for samples is approved, before pulling the samples, will I be contacted regarding the method of sample retrieval and shipment method?
If necessary, special processing requests may be taken into consideration.
Once my request for samples is approved, how long will it take for me to receive the samples?
Depending upon availability, a sample request can take anywhere from 3 days to 2 months to fulfill. It may take longer if the samples are from a legacy study and have not yet been delivered to the Repository. The clock begins once the fully completed request is approved by NIDDK. Sample turnaround time depends on many factors, including:
- Does the request require relabeling of vials?
- How many samples are being requested?
- Do they need to be aliquoted?
- Are the specimens stored in one freezer or scattered in several freezers?
- Are specimens being pulled from -80C freezers or liquid nitrogen freezers?
- Where will the samples be shipped?
- Does the request require payment by the recipient before the request may be processed?
I am unable to open a data set I received from the Repository, what do I do next?
Please review the following questions. Send the answers via email to email@example.com
- Which study's data are you having problems with?
- Did you receive this data set from the Repository via Central Repository?
- Are you having problems with any other data sets?
- Are you working with Windows PC or another operating system?
- What version of SAS are you using?
- Do you know the nature of the problem? Example: Cannot open because SAS format is missing; incompatible SAS version; or another specific error message.
- Are you trying to work with a different software tool such as R, SPSS, etc.?
Is IRB clearance required also for internal analysis in an organization, or is IRB clearance required only if the data analysis will be published?
For any level of access to the data you must have your institutions IRB clearance or waiver. If you do not have an IRB, you must use an external IRB. This IRB step is required whether or not you publish your findings. Since the data we offer is provided in Limited Datasets and the data has been de-identified and there is no direct human-subject interaction, sometimes your IRB will consider a waiver so you may want to approach it from that standpoint.
I will have co-investigators from other institutions working with me on the dataset I am requesting. Does NIDDK need IRB approval from each institution separately, or do you just need the IRB approval from the Principal Requestor's institution?
All co-investigators who will receive the data needs IRB approval to work with it. Similarly, each institution needs its own signed Agreement form.
My IRB requires confirmation that all Repository data is deidentified. Can the Repository sign a Confidentiality Agreement so I can move forward with my request for NIDDK samples and data?
NIDDK policy is that documents aside from the repository's own agreement will not be signed by NIDDK officials. Additionally, the repository is providing limited datasets rather than completely de-identified data. One of the repository's tasks, as defined by NIDDK, is to develop systems to help ensure protection of the human subjects who contributed research samples and/or data by maintaining the confidentiality of these subjects and by maintaining a searchable inventory of informed consents and NIDDK Approval Letters from sample contributors.
Datasets have been received without personal identifiers and thus are unlinked to the subjects. We do not accept any information, nor use any information collected, that would permit us as the keeper of a record system to establish the identity of a given subject or query the databases and retrieve information for an individual subject whose identity is known from some other source.
Traditional personal identifiers that may be utilized to establish the identity of individual subjects (e.g., surname, address, social security number, etc.) are never accepted by us when we receive materials from a study. Instead, we work with study's Data Coordinating Center and sample repositories to develop an identifier code for each subject that will link the clinical and descriptive data to stored biosamples. All data received by us is untraceable to the original sources.
Producing Manuscripts using NIDDK Samples
If I produce a manuscript using NIDDK data and/or samples, am I required to acknowledge this?
Researchers producing manuscripts using NIDDK data and/or samples are required to use the general NIDDK Repository acknowledgement. Study-specific acknowledgements are required with some of the data/sample sources.
What provisions should be included about the repository in the informed consent for subjects?
NIDDK has developed model language for use in informed consents that describes the repository and explains what will happen to the samples and data that are collected. The most important point is that the repository will not accept any personal identifiers on samples or in the datasets. Your NIDDK Project Officer can provide you with a copy of the model language, also posted in our reference documents.
What subject or sample identification code will the repository use?
The repository can use the study's assigned ID for each participant. In situations where this would be inappropriate, because the ID contains potential personal identifiers, the study should assign a different, masked, identifier.
Does the NIDDK Repository maintain a Certificate of Confidentiality?
No. The Repository does not have a Certificate of Confidentiality because we do not accept any personal identifying information.
When the DNA samples (or biorepository samples or data) are made available to requestors, will the actual name of the study clinic be released?
Who owns the samples submitted to the repository? Does the Institution submitting the samples maintain ownership or are they owned by NIDDK once they have been submitted?
The NIDDK Repository does not address its sample holdings in terms of "ownership", It is more a matter of control of the samples. The Repository is the custodian of samples from designated studies. The study group submitting the samples controls their distribution while studies are ongoing and during an agreed-upon proprietary period. Once the proprietary period ends, the Repository and the NIDDK Data Access Committee assume responsibility for distributing the samples in accordance with applicable NIH policies and procedures.
What measures are available for DCCT/EDIC?
All available measures can be viewed on the NIDDK Data Repository Website.
I find anomalies in the DCCT-EDIC datasets; can you offer any guidance on how to handle them?
Repository staff have fielded a large number of queries about such anomalies. We have posted a listing NEED LINK of these anomalies and our interpretation of how to handle them.
When ASCII datasets are generated there are numerous fields that contain either BLANKS (i.e., no data) or periods (.). What does this indicate?
These are missing data indicators. They occur because the measurement was not required (e.g., date of last menstrual period for male subjects) or was not made for any other reason (e.g., patient did not appear for one of the exams or did not answer an interview question).
For many variables in the DCCT-EDIC datasets, there are a few cases that have codes that are not within the allowed range. How does this happen and what should we do with these measurements?
These are usually erroneous data entries made by the person recording the data or the person entering it into the computer file. When fewer than 1% of the cases (e.g., 14 of 1,441 for measurements made once on the total cohort) have such errors, we recommend that these entries be treated as missing data.
What is meaning of variable DUPOK which occurs in several datasets?
The study Data Coordinating Center (DCC) reports that this variable was used for data management by DCC staff, and that it should be ignored by analysts.
Is there a unique "key" that identifies unique records in the DCCT adverse events analysis dataset (m9exevet)?
This is one of those datasets for which a unique key must be created, if needed. For most "adverse events", the combination of patient ID (mask_pat) plus date (eventday) will be unique for the vast majority of events. However, some patients will have multiple events on the same day. To make a unique key, you must construct a counter that identifies the 1st, 2nd, 3rd, etc. event on a given day.
On DCCT Form 13.3, variables acb5prob & acb6prob are derived from questions on a form that require text input. But the data fields are either blank or 1s. Why?
1s may indicate that text was entered. The DCC recommends that these variables be ignored.
What does code of 3 mean for variables bag3bi & bah5h in the DCCT Form 21.6?
What are codes for bhprep in DCCT Form 28?
0 = Not Prepared, 1 = Somewhat Prepared, 2 = well prepared.
What are codes for variables bhrepeat & bhvrpt in DCCT Form 28?
1 = no, 2 = yes.
In the DCCT Diet History, variable supplement has values other than documented values of 0 (no) and 1 (yes). Why?
There are 3,802 codes of zero, 1,424 codes of one, plus 7 codes of nine. We suggest that 9s be treated as missing or erroneous data entries.
In DCCT Form 2, how are the "ages" of diabetic relatives coded in the variables OBFAM4, 7, 11, 13, 15, 19, 21, 23?
Estimated age in years with 01 indicating age 1 or less. A code of 0 indicates age is unknown.
What does code of 99 mean for variable OBFAM9 (Number of diabetic relatives) in DCCT Form 2?
A code 99 indicates the number is unknown.
In DCCT Form 2, what does the undocumented code for OBGENIT mean?
There is one case with an undocumented code of 62; we suggest that it be treated as missing or an erroneous data entry.
What do the undocumented codes for OBIBW mean in DCCT Form 2?
There are two cases with undocumented codes of 7 and 186. We suggest that these codes be treated as missing or erroneous data entries.
What does the undocumented code for OCI6 mean in DCCT Form 3?
One case has an undocumented code of 7. We suggest that this code be treated as missing or an erroneous data entry.
What are units used for the WAIS measurements A0KK12, AOK2, AON2, AOB2, AOC2 in DCCT Form 10?
These are latency measurements so seconds or minutes would be the logical choice. Given
the values recorded, they would need to be seconds or else the times for testing would be extraordinarily long.
What is meant by the term "t-score" used for variable TSCDEP ("T-Score Depression") in the DCCT Baseline?
These are standardized scores normalizeded to have a mean of 50 and a standard deviation of 10.
In DCCT MS5exprt, FSCORE00 is Framingham risk score for coronary heart disease (chd; see Anderson et al., 1991: PMID: 1999037). How is it scored?
This appears to be the Framingham CVD risk score. But the DCCT scores have a mean of 0.01, so it does not appear to be using the published scale, see: http://www.nhlbi.nih.gov/guidelines/cholesterol/risk_tbl.htm#men.
In DCCT Form 21, for variable Bag3bi and bah5h, 320 and 432 cases respectively have undocumented codes of 3. What does this mean?
We have requested clarification from DCC.
Many of the variables on DCCT Form 33 are difficult to understand. What should I do?
This form includes the variables that were used by reviewers to provide summary retinopathy scale scores from the DCCT fundus photography. Users should focus on ETDRS summary scores. The ETDRS scale was developed by the "Early Treatment Diabetic Retinopathy Study"; see: http://clinicaltrials.gov/ct2/show/NCT00000151. You may also refer to the analysis dataset and documentation associated with the primary DCCT retinopathy paper (Arch Opthalmol 113:36-51 ), available in the DCCT archive.
In DCCT Form #36, variables impa, impp, wor, word, wora, glo, are not shown on the questionnaire, what are they?
These are the GLOBAL and SUBSCALE Quality of Life scores described in http://care.diabetesjournals.org/cgi/content/abstract/11/9/725.
In DCCT Form 108, what is the meaning of the alphabetic codes for variable JHCODE?
FA=Father; MO=Mother; PT=Patient; HU=Husband; WI=Wife; C1-C9=Child 1 thru Child 9; S1=Oldest sibling; S2=2nd oldest sibling
What is the metric for variables tpain and tpercept in DCCT Form 128?
Based on page 1 of documentation, these appear to be PERCENTILE scores. Analysis indicates this is so. One negative value for tpain (-11.1) and two values for tpercept (-20, -8) are anomalous. We suggest that you treat these values as missing.
What is the meaning of the codes for origin (source of event data) in DCCT MS5event?
CALCULATED = Derived from automated comparison of multiple reporting forms; HARDCODE = Coded explicitly in setup program; used for events not captured by variables on the usual reporting forms (for example, silent MIs reported by letter from the Central ECG Reading Unit); MASTER = Master data set of DCCT patient status (principally used for deaths); 0034, 0033, 0032 = Successive versions of the DCCT annual clinic visit form (Forms 003.4, 003.3, and 003.2); 0204 = DCCT intercurrent event form (Form 020.4); 0217, 0216 = Successive versions of the DCC'T quarterly visit form (Forms 021.7 and 021.6); 0241 = Automated review of ECG results reported by- the CERU using the Minnesota codes (DCCT Form 024.1).
In DCCT MS7exf10, what do the "half-point" codes used in changex2 and globalx2 mean?
For ratings that seem to fall between categories, a plus (+) sign is used to indicate that the rating is closer to the next higher category (somewhat impaired) and a minus (-) is used to indicate that the rating is` closer to the next lower category (somewhat less impaired). To accommodate these intermediate ratings, the following "extended scale" is used in the final dataset: 1='1'; 2='1+'; 3='2-'; 4='2'; 5='2+'; 6= '3-'; 7='3'; 8='3+'; 9='4=4-'; 10='4'; 11='4+'; 12='5-'; 13='5'; 14='5+'; 15=6-'; 16='6'; 17='6+'; 18='7'.
Is there specific wording that needs to be used to acknowledge the NIDDK DCCT/EDIC samples?
Yes. Please see https://repository.niddk.nih.gov/pages/acknowledgements/.
Are there special procedures for access to DPP samples?
The procedures for applying for access are identical for all studies. However, if non-renewable samples are requested from the DPP study, a DPP study representative is added to the review panel.
How many DPT-1 subjects that were free of disease have gone on to develop T1D?
I am interested in IBDGC genetic data. Is this available?
For investigators wishing to do their own genotyping, the cell lines for the first 5 years worth of sample collection (n = 4,761) are now
available, with the corresponding phenotype data available through the data repository. For those interested in the genotype data from the GWA
studies, the data from the CD GWA are at dbGaP, and may be requested from them. The UC GWA data were only recently collected, and a timeline for
distribution has not yet been finalized. If you are interested in those data specifically, you may contact
Dr. Phil Schumm with your specific question.
How do I construct the variable "percent of peginterferon dosage in the first 24 weeks?" from the raw variables in the VirahepC legacy datasets? (as reported in the publication by Conjeevaram HS, et al  "Peginterferon and Ribavirin Treatment in African American and Caucasian American Patients with Hepatitis C Genotype 1")
The VirahepC Data Coordinating Center (DCC) did not submit analysis datasets to the Repository. For the DSIC, Repository statisticians recreated a limited number of analysis measures using the "raw" datasets. Percent of interferon taken in the first 24 weeks (from Table 2, p.474) was not replicated for this analysis. We found the data to be well-organized and have confidence that this measure may be recreated from the raw measures.
The formula for "percent of peginterferon dosage" was given on p.471 of the publication: "The amount of drug taken was monitored using the Electronic Medication Event Management System (MEMS)...By multiplying the number of cap openings by the dose prescribed per opening, the estimated total dose of each drug taken was calculated. The proportion of maximum dose taken for 24 weeks of treatment (maximum value = 1.0) was calculated as the total estimated dose taken divided by the maximum dose possible had there been no dose reductions or discontinuations".
Forms datasets which will provide the necessary data for the calculation include:
- MEMSINT (contains one record per PEG-interferon cap opening. Non-events from form TW were removed.)
- MEMSRIB (contains one record per ribavirin cap opening. Non-events from form TW were removed.)
Other datasets which have been incorporated:
- DC (Dose Change; tracks physician prescribed dose changes).
- AD (Adherence Questionnaire)