Causality assessment: A casualty of compensation?

Recent Indian rule on compensation for any injury or death of the subject in a clinical trial, has generated a lot of discussion on its impact on clinical trials in India. The legal process in the compensation rule is based on the causality assessment. However, it ignores the scientific basis of causality assessment. This could have far reaching consequences on (1) science of causality assessment and (2) safety assessment of a new drug under clinical development. There is a need to bring clarity on causality assessment process within the compensation rules to balance the ethical need of human subject protection and the scientific requirements of safety assessment.

INTRODUCTION

Recent Indian rule on compensation for any injury or death of the subject in a clinical trial,[1] has generated a lot of discussion on its impact on clinical trials in India. The compensation, in a clinical trial setting, reflects three important issues — morality, causality and legality. The morality emphasizes the right of a clinical trial participant for compensation in case of a serious adverse event (SAE) due to an investigational product (IP). The causality establishes the relationship between an SAE and IP. And the legality prescribes the legal framework for paying compensation in such cases. Although, the legal process in the compensation rule seems to be based on causality assessment, it ignores the scientific basis of causality assessment. Thus the legality of compensation rule, whilst trying to support the morality of human subject protection, could have far reaching consequences on (1) science of causality assessment and (2) safety assessment of a new drug under clinical development.

This article reviews some of the implications of the compensation rule on causality assessment.

INJURY RELATED TO CLINICAL PRACTICE OR CLINICAL TRIAL?

Adverse events (AE)/injuries caused of medical care are not uncommon in clinical practice. An AE is usually defined as an unintended injury or complication resulting in prolonged hospital stay, disability at the time of discharge or death and caused by health-care management rather than by the patient's underlying disease process.

A systematic review of eight studies, including a total of 74,485 patient reported that a median overall incidence of in-hospital AEs was 9.2%.[2] The median percentage of preventable AEs was 43.5%, permanent disability was 7% and death was 7.4%. The median percentage of AEs due to different medical practice areas was: Operation-related 39.6%; drug related 15.1%; diagnosis-related 7.5%; therapy-related 7% and procedure related 7.8%. There are hardly any Indian studies of AEs. However, the prevalence of such AEs in India is likely to be similar.

The AEs occurring in the clinical practice may be treated if such events are serious. As the AEs are caused by health-care management, some of these could be due to negligence on the part of the treating physician. These AEs are usually not compensated, unless the patient takes legal recourse.

Many of the AEs occurring in a clinical trial could be caused by health-care management issues, e.g., negligence of the investigator or medical procedure. The compensation rules seem to ignore this ground reality as it mandates that all clinical trial related should be treated and compensated.

CAUSALITY — CLINICAL TRIAL RELATED OR IP RELATED?

The compensation rules cover clinical trial injury of death. In contrast, US Food and Drug Administration (FDA) refer to the causal relationship between SAE and the IP.[3]

The current compensation rule includes several conditions beyond an AE due to IP to qualify an injury as clinical trial related. This means that even if an AE was not causally linked to IP, it could be labeled as clinical trial related injury. This could lead to a discrepancy between SAE reporting to Indian regulatory authorities and international regulatory agencies.

The quantitative impact could be even more alarming. As per Central Drugs Standard Control Organization (CDSCO) data, from January 2005 to June 2012, 57,303 patients were enrolled. There were 11,972 non-fatal SAEs, of which 506 were related to clinical trial.[4] There were 2644 deaths, of which 80 were related to clinical trials. The total number of patients who suffered from an SAE was 586 (1%) of the total enrolled patients 57,303.

These assessments of relationship between the clinical trial and SAEs were done by the investigators without any specific guidance on the criteria of assessment. If one were to assess these by the conditions listed in the compensation rules, there is a likelihood of many of these SAEs being re-classified as related to clinical trial. Furthermore as cited above, 9.2% of patients in clinical practice have an AE caused by health-care management, which is much higher than the 1% of patients reported by CDSCO to have a clinical trial related injury/death. Hence, percentage of clinical trial related SAEs is likely to be higher in post-compensation rule setting. This would have an impact on the safety profile of the IP.

WHO ASSESSES CAUSALITY — INVESTIGATOR, SPONSOR, ETHICS COMMITTEE (EC), EXPERT COMMITTEE OR CDSCO?

As per FDA regulations, the investigators are required to provide a causality assessment for each SAE reported to the sponsor. For serious events that are unexpected, the sponsor considers the investigator's causality assessment, but submits an investigational new drug safety report only for those events for which the sponsor determines that there is a reasonable possibility that the drug caused the event, regardless of the investigator's causality assessment.[3]

The investigator's view is important for the sponsor to consider because the investigator is aware of the subject's clinical state and hence, can distinguish between events that may be related to the drug versus those due to the underlying disease process and/or concomitant therapies. However, the sponsor is better positioned than the investigator to assess the overall safety of the IP as the sponsor has access to all the analyzed SAE reports. Furthermore, the sponsor is more familiar with the drug's mechanism of action, class effects and other information.[3]

As per the compensation rule, the causality assessment made by the investigator and the sponsor, would be reviewed by the EC, the Expert Committee and the CDSCO. This means the EC, the Expert Committee and the CDSCO may not agree with the causality assessment made by the investigator and the sponsor. If the SAE is considered clinical trial related by the investigator and the sponsor, it is likely to remain unchanged. However, if the SAE is not considered clinical trial related by the investigator and sponsor, the EC, the Expert Committee and the CDSCO may change the status to clinical trial related. This would have an impact on the overall safety profile of the IP.

PROCESS OF CAUSALITY ASSESSMENT

The investigator is supposed forward the SAE reports after due analysis. The rule does not define, what is meant by due analysis. The analysis would be similar to an SAE case narrative, which is a summary all relevant clinical and related information, including patient characteristics, therapy details, medical history, clinical course of the events, diagnosis and adverse drug reactions including the outcome, laboratory evidence and any other information that supports or refutes a causality assessment. This kind of analysis would be more important when the SAE is considered not related to clinical trial.

In Indian scenario, the lack of training on causality assessment, the lack of time for causality assessment and the likelihood of questions from the EC/CDSCO in case the SAE is considered not related to clinical trial, may pressurize the investigator to report most SAEs as clinical trial related.

CAUSALITY ASSESSMENT — CHANGING UNCERTAINTY TO CERTAINTY

The system of pre-screening for submission of reports of SAEs mandates categorization of causality into related or unrelated.[5] This categorization ignores the ground reality of causality assessment. Adverse reactions are rarely specific for the drug, diagnostic tests are often not available and a rechallenge is rarely ethically justified.[6] In practice few adverse reactions can be classified as “certain” or “unlikely;” most fall between these extremes, i.e., “possible” or “probable.”[6] As per World Health Organization WHO,[6] the causality assessment cannot prove the connection between the drug and the event and change uncertainty into certainty. The approach is of having two categories, is likely to result in overestimating the real incidence of SAEs causally linked to a drug.

SAFETY DATABASE FOR ANALYSIS

There are several situations (as discussed above), which can influence the final assessment of a causal relationship between the clinical trial, the IP and the SAE. This can lead to different documents reflecting different causality assessments. For example, the investigator reported causality is “not related” to clinical trial, but is changed by the EC/expert committee as “related” to clinical trial. This discrepancy would create problems in international safety reporting and overall safety data base of IP.

One issue in reporting would be: Whether an SAE considered “related” to clinical trial would be considered “Suspected Unexpected Serious Adverse Reaction?” If so, it would have to be reported to the international regulatory authorities and all the global clinical trial sites.

If there are more SAEs considered “related” in the final CDSCO order, the overall incidence of SAE related to the clinical trial in Indian database would be higher compared to the incidence of suspected ADR in international safety database. For an Indian Research based pharma company, which would have a large proportion of Indian patients in its clinical trials, a high percentage of clinical related SAE cases in its safety database could invite a lot of regulatory questions, when a new drug application is filed.

CONCLUSION

The current compensation rule is likely to have significant impact on the scientific aspects of causality hence, there is a need to clearly define the terms used in the rule e.g. clinical related the term clinical trial related injury needs to be defined clearly.

Clinical trial related injury could mean:

SAE was attributable to IP, i.e., reasonable causal relationship is suspected between SAE and IP

SAE was attributable to any clinical trial intervention or procedure required by the protocol, which would not have occurred but for the inclusion of the patient in a clinical trial.

Such clarity is vital in balancing the ethical need of human subject protection and the scientific requirements of safety assessment.