Study design

Of the “SAN Case-control study”: The following summary provides the main points of the study design & protocol.

The full study protocol is attached in its initial version. A fewreversions were proposed and agreed by the ScientificAdvisory Committee. The final protocol was published.

Background:
In the absence of conclusive data on the risk of end-stage renal disease (ESRD) and substantial use of phenacetin-free analgesics , a new case-control study was recommended by an International Scientific Board (Potsdam, June 1999). The explicit objectives of the study are to examine the nephrotoxicity of newer (phenacetin-free) analgesic compounds and, if nephrotoxicity is ascertained, to determine the role of caffeine in analgesic drug consumption.

The upper age limit for the study population was defined as 50 years of age, to avoid contamination by previous phenacetin use, since phenacetin was gradually replaced by other analgesics in the early 1980's in Germany. Therefore younger patients would have had less opportunity for exposure to phenacetin. This age limit was deemed necessary, since there was concern that even with sophisticated reminders the recall of previous phenacetin use would be insufficient. . However, this low age limit makes the study logistically complicated and expensive, since three quarters of the rare incident cases of ESRD will not be usable for the study.

Objectives:
The primary study objective is to assess the risk of phenacetin-free analgesics with regard to the occurrence of ESRD. The main study questions are related to the risk produced by any phenacetin-free analgesics – either phenacetin-free combined analgesics or single substances.

Drug-specific analyses will be done to determine the risk for specific drugs, such as paracetamol with and without other analgesics, and with and without caffeine in the formulation.

The risk will be analyzed with regard to the magnitude of usage, such as dose, duration, dose times duration, and time since first or last, and peak use of analgesics, respectively.

If nephrotoxicity is shown for these analgesics, a secondary objective of the study will be to analyze user patterns of persons who overuse analgesics, and to determine the role of caffeine co-formulated with analgesics as compared with use of nutritional caffeine.


Design:
A case-control study design was chosen, because the outcome ESRD is a rare event, because many exposure groups will to be analyzed, and because reasonably prompt preliminary risk assessment is desired to meet the needs of the Drug Authorities.

Study Setting:
The case-control study to be conducted in Germany and Austria will be multicenter dialysis unit-based assembling as many co-operating centers for dialysis as necessary to reach the study size needed.

Cases:
Persons with end-stage-renal disease (ESRD) newly admitted to chronic dialysis programs (incident cases).


Controls:
Persons without ESRD, randomly selected from the population the case came from, and in the same 5-year age band as the case; four controls per case intended.

Index dates:
Four index dates will be considered:

(1) Time of entry into dialysis program for cases and time of interview for controls (Index Date 1 be used to define Index Date 3 and 4).

(2) Time when the diagnosis of a chronic renal condition was first mentioned to the patient; for controls the time of Index Date 2 of the corresponding case.

Additionally, two fixed index dates will be used in the analyses:

(3) Five years before the start of the dialysis program for the case, for controls the time of Index Date 3 of the corresponding case.

(4) Ten years prior to the start of dialysis for the case, for controls the time of Index Date 4 of corresponding case.

Any information after a selected index date will be ignored in the analyses


Target Variable:
End-stage Renal Disease (ESRD)


Exposures:
Exposure will be classified in three subgroups

(1) Main categories of exposure, such as any phenacetin-free analgesic, phenacetin-free combined and mono-analgesics, and non-use

(2) Drug-specific categories, such as paracetamol alone or in combination, with and without caffeine in the combination, ASA (aspirin) alone and in combination, ibuprofen and other drugs in similar groups

(3) Magnitude of usage of the above groups, i.e. dose, duration, dose times duration, and time since first or last, and peak use.


Co-Variables:
The following variables will also be considered in the statistical analyses (depending on the index dates): Age, sex, social indicators, collaborating center, selected chronic conditions, co-medication, health behaviour, life style, some psychosocial data, exposure to noxious agents at work or elsewhere. All data will be collected as time related (calendar method) to enable the construction of databases according to index dates.

Sample size:
The study currently aims to identify at least 1000 cases (800 to be included) in the study, with 3200 included controls to detect a two-fold difference between comparison groups, if it exists, at an alpha of 0.05 (i.e. the chance of obtaining a false-positive result is 5%), and a beta of 0.2 (i.e. the chance of obtaining a false-negative result is 20%. The power of the study to detect a difference, if one exists, is 80%, which conforms to the usual standards. An interim analysis after the first year will help to check the assumptions made in the sample size calculation.

Data collection & protection:
Two documentation forms will be used: an interview questionnaire (personal interview) and a form for abstracting medical data from medical records. Study participants will be asked to sign the informed consent form. Administrative steps will be taken to assure data confidentiality. Some anonymous data will be collected in a logbook for non-responders.

Quality assurance:
These measures include training (and re-training) of interviewers; logbooks will be kept to check if all procedures meet the criteria prepared in the protocol and manual, and to identify potential differences across centers, i.e., in addition to site visits. A sophisticated checking system for incoming data, together with a "query system", will be implemented to guarantee data quality.

Data management and statistical analyses:
Locally collected data will be transferred to the Data Management and Coordination Centre (DMCC) in Berlin according to specified time lines and responsibilities. After application of various checking procedures the clean database will be “locked”.

Frequency tabulations and appropriate multivariable models will be applied, i.e. adjusting for major confounders. The envisaged analyses are described in an "analyses plan" and will be discussed in depth with the international Scientific Advisory Committee (SAC).


Study management:
Nephrological aspects of the study will be managed by Prof. Van der Woude (Mannheim/Heidelberg) and Prof. Graf (Vienna). Prof. Heinemann (ZEG Berlin) will be responsible for the study and evaluation of the data. The German Kidney Foundation will play the key role in supporting the study in Germany, and the Austrian Association for Nephrology will do so in Austria.

Advisory Committees:
An international Scientific Advisory Committee (SAC) was jointly nominated by Drug Regulators, sponsors, and study team to approve all scientific matters of the study and to resolve any conflicts that may arise.

There is a Managing Committee (three Principal Investigators - PI's) to support a high-quality execution of the study.

A Steering Committee was established, consisting of delegates of all the concerned Nephrological Associations of Germany and Austria to help overview and supervise the study.


Communication:
Periodic reports of study progress will be made as requested to the International Scientific Advisory Committee (SAC) .

Analyses will be performed and submitted every 6 months to the SAC for purposes of checking for intra- and inter-center differences in response and quality control criteria. In addition, annual interim analyses of selected research questions will be discussed by the SAC to meet the needs of the Drug Regulators.

The publication policy has been arranged in the agreement with collaborating study centers.


Time table:
The study protocol and materials were prepared during 11/1999 to 09/2000. The revision of protocol and materials after discussion with the authorities as well as preparation of the study logistics are planned from 05/2001 to 12/2001 (including identification of first incident cases). It will be followed by the full fieldwork until 08/2004 and 12/2004, respectively for cases and controls. Thereafter, the final statistical analysis will be done in 10/2005 and circulated 12/2005; the draft of the main publication will be finished by 03/2006.

This plan should be satisfactory since annual interim analyses will be performed to meet the needs of the Drug Regulators and also to adjust for unforeseen circumstances.

The SAC will discontinue the study if it turns out that the study is not feasible or if the accrual of cases and controls does not meet realistic expectations.


Funding:
The SAN Case-Control Study is scientifically independent and governed by an independent Scientific Advisory Council. A group of sponsoring companies provided an unconditional grant. The SAN study group is accountable to the Scientific Advisory Council in all scientific matters. The members of the AC are outstanding international experts of relevant scientific fields. The sponsoring companies are invited as observers in open meetings of the SAC as are the Drug Authorities of the three countries (D,A,CH).
The members of the council receive remuneration of expenses and a honorarium to compensate a loss of potential earnings during their work for the advisory council. The members will not be involved in or paid for the operational conduct of the study.