Stats #53: Signal Detection Strategies for Paediatric Treatments

Content:  Signal detection and pharmacovigilance are already highly regulated and challenging fields, but once you factor in children as your subject group these challenges become even greater. There are physiologic, ethical, and statistical questions that you must consider for some (but not all) efforts in post marketing surveillance.

Discuss openly with your peers the issues that complicate drug safety studies in children and recognize when these issues apply and when they don't apply. Look at and debate the merits of alternative data sources, research designs, and statistical analyses to balance the sometimes conflicting needs of regulators, drug companies, and ethics review boards.

Teaching strategies: Didactic lectures and small group exercises.

Objectives:  In this class you will learn how to

• understand the medical differences in children that make pediatric research unique.
• be able to describe the controversies over informed consent and monetary compensation in pediatric studies.
• be able to identify the unique statistical challenges associated with pediatric research.

Teaching strategies:  Didactic lectures and small group exercises.

An important qualifier: There is an internet acronym used a lot in email and instant messaging: IANAL. It stands for "I Am Not A Lawyer" and it is used when someone is commenting on a legal issue but wants to remind the readers that this informal advice is not intended as a substitute for formal legal advice by a qualified lawyer. In the same vein, I want to emphasize IANAD. I do work closely with doctors, nurses, and other health care professionals but as you listen to my advice, keep in mind that "I Am Not A Doctor". I certainly want to comment on medical controversies to make you aware of their importance, but I am not the final arbiter of any medical controversy. My role is more to raise questions than to answer them.

Outline of this talk.

Introduction

• Where can you find this handout?
• Why don't I use PowerPoint?
• Information about my book.
• Therapeutic orphans.
• What, exactly, is a child?

How are children different?

• Medical concerns
• Ethical concerns
• Statistical concerns
• Searching for pediatric studies
• Case studies

Conclusion

---

Where can you find this handout?

This handout and the handouts that I use for all of my seminars and training classes are a compilation of individual web pages at www.childrensmercy.org/stats. I use the "Include Page" feature of Microsoft FrontPage to combine these into a single page. You can always find the most recent version of this compilation by going to the web address listed at the bottom of this page. Links for the handouts for other seminars and classes appear at www.childrensmercy.org/stats/training.asp.

Why don't I use PowerPoint?

I stopped using PowerPoint for my presentations in the mid 1990's. This was based on Edward Tufte's advice that presenting information in a paper handout is more effective than presenting the information on a projected screen. I found this to be excellent guidance. I enjoy talking when I don't have to wrestle with a laptop computer. I look at my audience more and interact with them better. I elaborate on this in greater detail at www.childrensmercy.org/stats/weblog2004/powerpoint.asp.

---

Information about my book, Statistical Evidence in Medical Trials

---

Therapeutic orphans (November 3, 2006).

As an introduction to a talk I am giving about research concerns involving children, I have to start off by pointing out that children are "therapeutic orphans." Historically, they have been, to a large extent, left behind by the medical research community, forcing pediatricians to make difficult choices for the care of their patients. As a simple example of this problem, a group of researchers in Australia examined the product information (PI) for 1,497 prescription medications and found that most of the PIs were unhelpful for children.

The proportions, for each age group, of PIs with inadequate paediatric dosing information were: < 1 month (80.5%), 1–3 months (79.1%), 3 months–2 years (77.5%), 2–6 years (73.2%), and 6–12 years (71.6%). The proportions, for each age group, of PIs that gave specific paediatric dosing information but did not provide a paediatric dosage form were: < 1 month (26.5%), 1–3 months (25.1%), 3 months–2 years (23.3%), 2–6 years (21.9%), and 6–12 years (24.0%). (Source: Dosing information for paediatric patients: are they really “therapeutic orphans”? Tan E, Cranswick NE, Rayner CR, Chapman CB. MJA 2003; 179 (4): 195-198. [Full text] [PDF]).

The reasons for this are complex and no one party is entirely to blame (Editorial Comment: Therapeutic Orphans. Shirkey H. Pediatrics 104(3): 583-584. [Full text] [PDF]). There are some proposed solutions, which I support, but these solutions are not without controversy. But one thing I want to dispel is the notion that research in children is impossible or even difficult.

Research in children is not impossible. You need to respect the differences. You can't blindly force-fit an adult research model onto children. But I want to encourage everyone to pay close attention to the pediatric field and leave my talk thinking "Yes, I understand the special issues associated with research in children and I'm ready to get to work."

This webpage was written by Steve Simon on 2006-11-03, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research

---

What, exactly, is a child? (November 19, 2007).

I'm updating a talk on research issues associated with pediatric treatments. Before I start throwing around terms like "paediatric" and "child," I should take some time to note that these terms have some ambiguity in them. A child is typically defined as a person with an age less than 18 years. This is not a hard and fast definition, though, and certainly I would not quibble if you used 16 years or 21 years as the boundary between a child and an adult. There are certain activities, such as marrying, living on one's own, becoming a father or mother, that will sometimes change the status of a child in legal or social settings.

The age range 0 to 18 years is a very broad range, and it makes sense at times to limit the discussion to certain age subgroups. This again is an arbitrary choice, but the International Conference on Harmonisation offers the following categories for consideration.

• preterm newborn infants,
• term newborn infants (0 to 27 days),
• infants and toddlers (28 days to 23 months),
• children (2 to 11 years), and
• adolescents (12 years to adult).

(Source: ICH Topic E11: Clinical Investigation of Medicinal Products in the Paediatric Population, www.emea.eu.int/pdfs/human/ich/271199en.pdf).

Other definitions of child and other categorizations of age groups may certainly be appropriate. A lot depends on the context of the particular research problem.

This webpage was written by Steve Simon on 2007-11-19, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research

---

Medical concerns about research in children (November 3, 2006).

Gray Syndrome. Alternate Names : Chloramphenicol Toxicity in Newborns, Gray Baby Syndrome. Definition Chloramphenicol is an antibacterial medication used against gram-positive and gram-negative bacteria and is often used for meningitis. If given to a newborn, however, it can be toxic (poisonous) and fatal. Overview, Causes, & Risk Factors "Gray syndrome" occurs if newborns (especially premature babies) are given chloramphenicol for a bacterial infection. Babies at that age do not have the necessary enzymes that allow the liver to be able to metabolize this drug appropriately. The chloramphenicol accumulates in the baby's blood stream, causing hypotension (low blood pressure), cyanosis (blue coloring of lips, nail beds, and skin from lack of oxygen in the blood), and often death. (Source: health.allrefer.com/health/gray-syndrome-info.html)

Children are not little adults and you need to respect the important differences between them. Some of the differences make research more difficult and some make the research easier.

Absorption, distribution, and elimination issues.

The tortuous path that drugs take is quite amazing when you think about it. The path is different between children and adults, especially for infants in their first few months of life.

There is a definitive resource for absorption, distribution, and elimination issues in children:

• Developmental pharmacology--drug disposition, action, and therapy in infants and children. G. L. Kearns, S. M. Abdel-Rahman, S. W. Alander, D. L. Blowey, J. S. Leeder, R. E. Kauffman. N Engl J Med 2003: 349(12); 1157-67. [Medline]

and all of the material discussed in this section is taken directly from this reference.

Stomach/Intestines. If a drug is ingested orally, it travels through the stomach and the intestines. A young infact will have a higher stomach pH. This can lead to differences in bioavailability of acid-labile drugs like penicillin G. Gastric emptying increases during the first week of life. Absorption of lipophilic drugs is dependent on the transport of bile salts into the intestinal lumen. This process is less efficient in infants. Intestinal motility increases from birth through the first four months of life. There are also changes in the splanchnic blood flow and intestinal microflora during the first few weeks of life. Some metabolism of drugs occurs in the intestine, and there are important age related differences in the activity of two key enzymes, CYP1A1 and glutathione-S-transferase.

Skin. Infants (especially pre-term infants) have an immature skin barrier, which leads to substantial increases in percutaneous absorption of topically applied drugs. This increased absorption is due to a thinner stratum corneum and greater hydration of the epidermis.

Muscles. The physiology of the muscles has an important impact on intramuscular injections. Infants have reduced blood flow and inefficient muscular contractions. This may be more than offset, though, by a higher density of skeletal-muscle capillaries.

Lungs. Most drugs that are administered by inhalation are intended to stay in the lungs, but some portion of these drugs will get absorbed into the blood stream leading to various side effects. The lungs of infants and children are different in the vital capacity and the respiratory rate.

Rectum. For drugs that are administered rectally, there are no apparent differences in mucosal absorption, but the infant has a greater number of high-amplitude pulsatile contractions which enhances the expulsion of solid drugs.

Body composition. Relative to their size, infants have a greater amount of extracellular water and total-body water. In infants, there is a larger ratio of water to lipid in the fat tissues. This can change the apparent volume of distribution for some drugs.

Blood. For some drugs, the ratio of bound to unbound drugs is critical for bioavailability. In young infants, there are fewer plasma proteins (especially albumin) which will often leave a large proportion of the drug unbound. Infants also have a greater degree of permeability of the blood-brain barrier, which can produce some serious side effects that are not present in adults.

Liver. A variety of enzymes located predominantly in the liver are responsible for biotransformation of drugs. Many of these enzymes (CYP3A7, CYP2C9, UGT1A6) are absent or have low activity in the infant and competing metabolic pathways may predominate. Other metabolic pathways (CYP1A2, UGT2B7) show substantial age-related variations.

Kidney. The kidneys play an important role in the elimination of certain drugs from the bloodstream. There are rapid changes during the first few weeks of life and more gradual changes during the remainder of childhood in the glomerular filtration rate and in tubular secretion.

The Kearns et al review article stresses that most of the changes and differences in absorption, distribution, and elimination occur in the first few weeks and months of life. The differences can be even more exaggerated in pre-term infants. There is some continual growth during the first few years of life that require disproportionate dosing, but by the time children reach their eighth birthday, there is sufficient similarity between children and adults to justify dosing proportional to body weight or body surface area. This reference had relatively little discussion of the effect of puberty on absorption, distribution, and elimination, and apparently this is not a major factor.

Compliance issues

I have two medication situations at home that remind me about the complexities of compliance

The picture above is Newton, a 17 year old cat with a hyperthyroid condition. Down near her front paws are her daily doses of Tapezole. She gets a half pill in the morning and a quarter pill in the evening. She does not take these pills herself, but relies on me to open her mouth and thrust the pill in. If she recognizes that I am about to pill her, she runs and hides in the basement. When I catch her and try to pill her, she will try to spit out the pill. Sometimes it takes three or four attempts before the pill goes down. Sometimes I think it goes down but I find later that she had spit out the pill while I was not watching.

The second picture is Nicholas, a four year old boy that we adopted from Russia. He is getting ready to brush his teeth with his Dora the Explorer toothbrush. Near the back of the sink are two different types of children's toothpaste: a fluoride version for older children and a fluoride-free version for younger children.

A good definition of compliance appears in the Wikipedia:

Compliance (or Adherence) in a medical context refers to a patient both agreeing to and then undergoing some part of their treatment program as advised by their doctor or other healthcare worker. Most commonly it is whether a patient takes their medication (Drug compliance), but may also apply to use of surgical appliances (e.g. compression stockings), chronic wound care, self-directed physiotherapy exercises, or attending for a course of therapy (e.g. counselling). (Source: en.wikipedia.org/wiki/Compliance_%28medicine%29

Researchers in this field have suggested that the word "compliance" be replaced with "adherence."

The term "compliance" suggests a restricted medical-centered model of behavior, while the alternative "adherence" implies that patients have more autonomy in defining and following their medical treatments. (Source: Beyond "compliance" is "adherence". Improving the prospect of diabetes care. KE Lutfey and WJ Wishner, Diabetes Care; 22(4): 635-639. [Abstract] [PDF])

There are many reasons for poor compliance

• Prescription not collected or not dispensed
• Purpose of medicine not clear
• Perceived lack of efficacy
• Real or perceived side-effects
• Patients’ perception of the risk and severity of side-effects may differ from that of the prescriber
• Instructions for administration not clear
• Physical difficulty in taking medicines (e.g. with swallowing the medicine, with handling small tablets, or with opening medicine containers)
• Unattractive formulation (e.g. unpleasant taste)
• Complicated regimen

(Source: British National Formulary 52)

There is another reason not listed above, "I forgot." Researchers today are making a distinction between "accidental noncompliance" (I forgot and related reasons) and "volitional noncompliance" (I didn't want to and related reasons). These distinctions are helpful in understanding how to improve compliance

A nice summary of the causes of poor compliance in children is

• How Do You Improve Compliance? Sheldon Winnick, David O. Lucas, Adam L. Hartman, and David Toll. Vol. 115 No. 6 June 2005, pp. e718-e724. [Abstract] [Full text] [PDF]

The article lists the various barriers to compliance in a paediatric patient.

Perhaps the most important barrier is the limited amount of time that a physician has to spend with his/her patient. This lack of time is no different than with adult patients, but the problem is that a pediatrician has to establish lines of communication with both the patient and the parents/caregivers. Having the time to communicate with the parents alone (without a restless child to distract their attention from the doctor) can be very important.

If a family unit is dysfunctional, this can be a serious barrier to compliance. If someone other than the primary caregiver brings the child in for an office visit, the pediatrician has to rely on an intermediate source for relaying important information to the primary caregiver. Just as bad is when there are multiple caregivers and complex medical instructions have to be coordinated among these caregivers. Children who live in different houses on the various days of the week because of shared custody arrangements may also have problems with keeping medication readily available. Finally, rebellious adolescents (what adolescent isn't rebellious?) can also cause trouble with compliance.

Medicines that taste bad or are hard to swallow are more of an issue in children than adults. The differing dose requirements in children may require the splitting or crushing of pills. It is unclear what impact splitting or crushing will have. Does the presence of flavoring and coloring agents in mediation influence the safety and efficacy profile? Frequency of administration can also create problems, especially if it requires the cooperation of the child's school.

Although most of the barriers listed above are plausible and have anecdotal support, the authors of this publication point out that there has been relatively little quantitative research on these barriers and their effect on compliance.

Differential susceptibilities

The injuries and illnesses that children are susceptible to are quite different than the susceptibilities in adults. An Institute of Medicine report on emergency medical care for children,

• Emergency Care for Children--Growing Pains, www.iom.edu/CMS/3809/16107/35002.aspx

highlights many of these issues in Table 1.1, starting on page 18. Children have a greater body surface area to body mass ratio. This increases their heat loss and places them at greater risk for hypothermia. They have less protective muscle around internal organs, less fat,  and a more pliable skeleton, so trauma is much more likely to cause internal organ damage. The relatively large head size also contributes to hypothermia and traumatic injuries. The increased respiratory rate makes children more susceptible to effects of air pollution.

There are additional concerns in children. Children have a greater tendency for placing foreign objects in their mouths, placing them at greater risk for choking and accidental poisoning. Their inability to distinguish candy and medicine also places them at greater risk for accidental poisoning.

Most of us have stopped growing (vertically, anyway), but the rapid growth in children raises special concerns about drugs that might interfere with this growth. Intellectual growth is just as important as physical growth. Any damage to the brain cells or the nervous system can lead to serious developmental delays. While loss of hearing or vision is traumatic at any age, the sensory deficit caused by hearing or vision loss during the critical phase of brain development can produce serious deficits in intellectual development.

The premature infant has many susceptibilities because of their immature organ systems: cranial hemorrhages, bronchopulmonary dysplasia, and retinopathy of prematurity.

The character and nature of heart disease and cancer are quite different in children. Most heart problems are congenital. Cancer is rarer in children than adults, but the types of cancers in children are quite different.

On the positive side, children are less susceptible to diseases that are caused by abuse of adult vices, such as cirrhosis of the liver and lung cancer. They also do not have the burden of the aging process and have little or no risk for degenerative diseases like Alzheimer's.

Another positive feature of paediatric diseases is their simplicity. Children, unlike adults and especially unlike elderly patients, will have fewer co-medications and fewer concurrent diseases. This is ideal from a research perspective because extra medications and diseases can complicate the analysis and interpretation of research results.

This webpage was written by Steve Simon on 2006-11-03, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research

---

Ethical concerns about research in children (October 17, 2006).

Pediatric research must balance two important social and ethical goals. First, scientific progress that includes issues important to children will by necessity require that children be involved in research. Second, because children cannot themselves consent to participation in research and must depend on adults to protect their interests, it is essential that the well-being of the individual child who participates in a research project be protected. Although these two goals are compatible, they will at times conflict. Achieving the appropriate balance between these two goals becomes the focus of discourse among those who seek to do good research. Good research is ethical research, and both require investigators who take seriously the importance of participant welfare, meaningful informed consent, and respect for research participants. Current controversies in pediatric research ethics: Proceedings introduction. D. S. Diekema, F.Bruder Stapleton. The Journal of Pediatrics 2006: 149(1); S1-S2. [Abstract]

I am giving a talk in London about the differences in research when children are involved. One major aspect of these differences is that the ethical and regulatory requirements change. I do not plan to talk about regulatory issues for two reasons.

First, the regulatory environment is quite diverse. At my hospital, which straddles the border between the states of Kansas and Missouri, we have different legal requirements for patients depending on which state they come from. How much more different is it when you consider regulations in the international arena!

(Source: www.idcide.com/citydata/mo/kansas-city.htm)

Second, a focus on just the regulations leads to a narrowly drawn perspective: what can I get away with. Rather than thinking about your research from a narrow legalistic perspective, I'd encourage you to think about what you are comfortable with. What type of research can you do and still sleep well at night. The regulatory and legal constraints are important, of course, because a small number of people have no conscience and sleep well at night after doing terrible things during the day. These people cause all the problems for the rest of us.

Informed consent requirements were derived in response to Nazi medical experiments

Ethical research requires the full and free voluntary participation of your research subjects. This principle was first elaborated in the Nuremberg code, a set of principles developed in response to abusive medical experiments conducted by the Nazis at concentration camps and prisons. There are other abusive medical experiments, such as the Tuskegee syphilis study, that also raised our awareness of the importance of informed consent.

There is a nasty tendency in today's society to attack research that you do not like by comparing it to the research done by the Nazis, and I want to discourage those sorts of analogies. They are not helpful. But it also important to understand the abuses that led to the first elaboration of principles of ethical research.

Eva Mozes-Kor tells the story of her family's experience

My thoughts were interrupted by the sound of the cattle car door as it opened. "Schnell, schnell." The SS soldiers were ordering everybody out. As soon as we stepped out onto the cement platform, my mother grabbed my sister and me by the hand, hoping somehow to protect us. Everything was moving very fast. I suddenly realized that my father and my two older sisters, Edit and Aliz, were gone. I never saw them again. I think the whole thing took 10 minutes; they were lost in the crowd as Miriam and I clutched my mother's hand. The SS soldiers walked by, shouting louder. Suddenly, they stopped my mother and looked at my twin sister and me, because we dressed alike and looked very much alike. "Are they twins?" one soldier asked my mother. My poor mother was bewildered. What was this place? she must have thought. What was happening here? What were the rules? What was a good answer and what was bad? She asked the SS soldier if being a twin was good. The guard nodded his head. My mother said very hesitantly, "Yes, they are." Without further explanation, the officer grabbed Miriam and me, and another SS soldier grabbed my mother and pulled her in the opposite direction. We screamed and pleaded as we were separated. I remember looking back and seeing my mother's arm stretched in despair as she was being pulled away. I never even said goodbye to her. I did not know that was the last time we would see our mother. The Mengele Twins and Human Experimentation: A Personal Account. Mozes-Kor E. In: Annas GJ and Grodin MA ed. The Nazi Doctors and the Nuremberg Code. 1992; Vol. New York NY: Oxford University Press; 53-59.

It was the Spring of 1944. Eva and her twin sister were at Birkenau. Dr. Josef Mengele was conducting medical research, and twins were highly prized for this research. No one told Eva or Miriam what was going on, what was being done to them or what their ultimate fate would be.

Eva was infected with a germ-she doesn't know which one-and developed a fever. She has put in a hospital, but did not receive any care at all. Dr. Mengele and a team of doctors reviewed Eva's fever chart but did not otherwise intervene. They gave her no medicine, no food. She had to crawl by herself to a water faucet to drink.

Eva realized that she would never get out until she convinced these doctors that she no longer had a fever. She manipulated the thermometers so that her fever gradually disappeared. After three weeks, she was re-united with her sister.

Upon my return, Miriam told me that during the first 2 weeks of my hospitalization, someone had stayed with her continually. She was not told of my condition, but it was clear that had I died in the hospital, Miriam would have been taken immediately to Mengele's lab to be killed

In very cold and cruel terms, Miriam would have been a matched control subject. Her healthy organs at autopsy would be compared to Eva's disease ravaged organs.

The public's response to the horrors of stories like this led to the development of the Nuremberg Code. The ideas behind the Nuremberg Code have been refined over time, but it is valuable to read these original principles. Here are the ten principles.

1. The voluntary consent of the human subject is absolutely essential.

This means that the person involved should have legal capacity to give consent; should be so situated as to be able to exercise free power of choice, without the intervention of any element of force, fraud, deceit, duress, over-reaching, or other ulterior form of constraint or coercion; and should have sufficient knowledge and comprehension of the elements of the subject matter involved as to enable him to make an understanding and enlightened decision. This latter element requires that before the acceptance of an affirmative decision by the experimental subject there should be made known to him the nature, duration, and purpose of the experiment; the method and means by which it is to be conducted; all inconveniences and hazards reasonable to be expected; and the effects upon his health or person which may possibly come from his participation in the experiment.

The duty and responsibility for ascertaining the quality of the consent rests upon each individual who initiates, directs or engages in the experiment. It is a personal duty and responsibility which may not be delegated to another with impunity.

2. The experiment should be such as to yield fruitful results for the good of society, unprocurable by other methods or means of study, and not random and unnecessary in nature.

3. The experiment should be so designed and based on the results of animal experimentation and a knowledge of the natural history of the disease or other problem under study, that the anticipated results will justify the performance of the experiment.

4. The experiment should be so conducted as to avoid all unnecessary physical and mental suffering and injury.

5. No experiment should be conducted, where there is an a priori reason to believe that death or disabling injury will occur; except, perhaps, in those experiments where the experimental physicians also serve as subjects.

6. The degree of risk to be taken should never exceed that determined by the humanitarian importance of the problem to be solved by the experiment.

7. Proper preparations should be made and adequate facilities provided to protect the experimental subject against even remote possibilities of injury, disability, or death.

8. The experiment should be conducted only by scientifically qualified persons. The highest degree of skill and care should be required through all stages of the experiment of those who conduct or engage in the experiment.

9. During the course of the experiment, the human subject should be at liberty to bring the experiment to an end, if he has reached the physical or mental state, where continuation of the experiment seemed to him to be impossible.

10. During the course of the experiment, the scientist in charge must be prepared to terminate the experiment at any stage, if he has probable cause to believe, in the exercise of the good faith, superior skill and careful judgement required of him, that a continuation of the experiment is likely to result in injury, disability, or death to the experimental subject.

The desire to insure that research participation is truly voluntary has led to a huge framework for the production and review of the informed consent process.

There are several notable areas where the need for informed consent can SOMETIMES be waived. It is a bit dangerous to summarize these areas with just a few bullet points because each of these areas could probably demand its own three hour training class.

• Quality assurance programs,
• Publicly available records,
• Anonymous patient records,
• Retrospective chart reviews,
• Emergency situations where consent is impractical.

The rules are different in the United States and in Europe, but in general you need to consider the degree of risk in the research and the practical barriers that obtaining informed consent might raise. The barriers imposed by obtaining informed consent need to be more than just an inconvenience to the researcher before you can waive the informed consent requirement.

There are other areas where informed consent has sometimes been waived. For example, there have been many studies in which intercessory prayer has been studied in a double blind randomized trial. In most of these studies, subjects were not asked for their consent prior to randomization so as to avoid selection bias.

Informed consent is impossible in children

The whole issue of informed consent gets tossed out the window when it comes to research involving children. Children do not have the maturity and intellectual capability to make an informed decision to participate. There is a vigorous debate in the research community about what this implies, and some have taken the position that non-therapeutic research (research that does not benefit the patient) is always unethical in children.

• Nontherapeutic research with children: the Ramsey versus McCormick debate. A. R. Jonsen. J Pediatr 2006: 149(1 Suppl); S12-4. [Medline] [Full text] [PDF]
• Children in clinical research: a conflict of moral values. Vera Hassner Sharav. American Journal of Bioethics 2003: 3(1); W12-W59. [PDF]

There is also vigorous debate about the definition of non-therapeutic research

• Phase I research and the meaning of direct benefit. L. Ross. J Pediatr 2006: 149(1 Suppl); S20-4. [Medline] [Full text] [PDF] (Ethics, Children)

Among those who believe that non-therapeutic research in children can be ethical, the conditions under which the research can be considered ethical center on three carefully defined terms:

• minimal risk,
• permission, and
• assent.

Minimal risk. I could not find a European (e.g., EMEA) definition of minimal risk. The U.S. Code of Federal Regulations (45CFR46.102) defines minimal risk as "the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests." Note that there are two dimensions, probability and magnitude. If either dimension is too large, then the research is not minimal risk. Research is possible if the risk is greater than minimal risk, but extra safeguards are needed. The U.S. regulations use some rather convoluted language here, such as "minor increase over minimal risk."

Permission. Most research involving children will obtain permission from the child's parent or legal guardian prior to the research. Just like informed consent can sometimes be waived, the need for parental permission can also sometimes be waived. The criteria for waiver examine the practicality of getting permission as well as the degree of risk. Sometimes permission is called "proxy consent" but you need to be careful how you use this term.

Assent. Although children do not have the intellectual capacity to provide informed consent, you should still seek their assent before starting the research. Assent should be an active agreement rather than a passive failure to raise an objection. When you should seek assent and the type of assent that you get depends on the age and intellectual development of the child. There is no hard and fast rule, but many sources have suggested that children at the age of seven have sufficient capacity to be able to offer assent. The process of seeking assent, of course, should be more detailed and involved for a teenager than a seven year old.  Keep in mind that some pediatric diseases can cause developmental delays.

A nice review of the issues associated with assent can be found in

• A developmental approach to child assent for nontherapeutic research. V. A. Miller, R. M. Nelson. J Pediatr 2006: 149(1 Suppl); S25-30. [Medline] [Full text] [PDF]

Please note that many of the controversies involving minimal risk, permission, and assent are irrelevant if your pharmacovigilance efforts involve the use of existing databases. There are indeed privacy risks for these types of studies, but no need for you to seek permission or assent, unless you are the one developing the database yourself.

Financial payments can sometimes undermine informed consent

When a person volunteers to participate in research, they are offering a gift to the researchers. They usually suffer some level of inconvenience. Sometimes they endure painful procedures like venipuncture. In some cases, they take on additional risk in order to participate. It seems only natural to offer financial compensation in return for their sacrifices. But you need to be careful that any financial payment does not become so large as to become an undue influence. This influence could lead people to discount the risks of the research and to hide potential disqualifying information from the researchers. To better understand the controversies behind these issues, you should recognize the distinctions among four different types of financial payments:

1. Reimbursement payments, which cover direct expenditures such as transportation costs and time lost from work,
2. Compensation payments, which cover indirect expenses such as the inconvenience of not being able to sleep late on a Saturday morning,
3. Appreciation payments, which represent an effort to offer thanks to the research participants.
4. Incentive payments, which represent an effort to encourage enrollment in the research study.

Financial compensation in a pediatric study is controversial and there are those who argue that ANY payments in a pediatric research study are unethical. In children, there are two problems. First, even a small amount of money may seem large to a child and cloud their judgment. For some children, the amount of money they receive in exchange for participating in a research study might be more money at one time than they have ever had before. Second, money that is given to the parent may potentially encourage the parent to exploit their children. Very few parents would do something that would benefit them at the expense of their children. Indeed, almost every parent will make significant sacrifices on behalf of their children and would be horrified at even the thought of doing anything against the interests of their children. Sadly, though, there are enough parents who don't have their children's best interests at heart that you need to be careful.

Discussion of what level is appropriate for these four types of payments appears in an Institute of Medicine report, Ethical Conduct of Research Involving Children (available at www.nap.edu/catalog/10958.html) and is summarized in

• Appropriate compensation of pediatric research participants: thoughts from an Institute of Medicine committee report. B. W. Ramsey. J Pediatr 2006: 149(1 Suppl); S15-9. [Medline] [Full text] [PDF]

Incentive payments raise the most concerns and the committee strongly discouraged their use. Incentives intended to overcome recruitment problems for painful and risky procedures are especially to be avoided. Appreciation payments are acceptable if they are "age-appropriate and of nominal monetary value."

There is little controversy, according to the IOM report, on reimbursement payments or compensation payments, as long as they are not unrealistically large. You can offer to buy someone lunch, but not lunch at a five star restaurant, for example. Defining what is unrealistic for indirect expenses is tricky and depends a lot on the economic circumstances of the child and parents. Bonnie Ramsey, the author of the above article shares two interesting anecdotes.

The first described a teenager who visited a clinic with multiple research studies open for volunteers who announced "Tell me all the studies going on and exactly how much each pays!" Clearly there is a problem with undue financial influence in this situation.

The second involved a discussion of a particular research project between Dr. Ramsey and a teenager with Cystic Fibrosis as well as that teenager's parents.

The family had faced a recent job loss by the father and change in employment by the mother. They lived a significant distance from the medical center. I began discussions regarding the study rationale, risks versus benefits, and study procedures but had not addressed the issue of compensation for participation. Compensation was clearly defined in the consent form, which the parents had not yet read. The adolescent expressed significant interest, but the parents were reluctant and claimed "lack of time" and a "long trip over the mountains." Going against my usual policy to not emphasize compensation, I noted that we would cover the costs of gas for the trip and compensate them for their time. The parents' faces lit up and both stated that they had always wanted to have both of their children  with CF participate in research trials but were too proud and embarrassed to admit that they could not afford the money for gas.

Dr. Ramsey makes an important point here, that compensation to remove barriers is certainly reasonable and goes on to suggest that other things beyond money (such as scheduling appointments at times convenient to the family rather than at times convenient to the researcher) might be worth considering.

Another important issue is the timing and distribution of payments in a longitudinal study. Certain studies have an increasing scale of compensation over time that reflects the real value of getting full information across the time spectrum on as many patients as possible. This can include a "balloon payment" at the end of the study, a large sum paid only to patients that complete all of the required visits. In general, the use of an increasing scale or a balloon payment is not consistent with reimbursement or compensation. These payments can sometimes be coercive, because a patient may be reluctant to withdraw from the study early when faced with the loss of a substantial portion of the financial compensation.

Again it is worth noting with data collected by someone else (a frequent source of information in pharmacovigilance trials) that you do not have to worry about any compensation issues.

Other resources:

• Psychological screening of children for participation in nontherapeutic invasive research. McCarthy AM, Richman LC, Hoffman RP, Rubenstein L. Arch Pediatr Adolesc Med 2001: 155(11); 1197-203. [Medline] [Abstract] [Full text] [PDF]

This webpage was written by Steve Simon on 2006-10-17, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research

---

Statistical concerns in research studies involving children (November 2, 2006).

The statistical design and analysis issues for research involving children are not really that much different than for adults, but there are three areas that you need to pay special attention to: sample size issue, subgroup analysis, and validity/reliability of measurements. I want to summarize some of these issues and offer some concrete examples. The two statistical issues boil down to validity and sample size.

Establishing validity in a study involving children

Every discipline has its own definition of validity, but in general it can be thought of as truth-in-packaging. A measurement is valid if it actually measures what it is claiming to measure. If it is measuring something else, it is not a valid measurement.

For example, there are a lot of studies looking at stress and its relationship to health. So how do you measure stress?

1. You can look at physiologic signs.
   
   academic.cuesta.edu/wholehealth/Level2/Lecpages/str07.htm
    
2. You can hook someone up to a meter that measures stress.
   
    www.cliving.org/stresstools.htm
    
3. You can ask people to fill out a survey.
   
   www.agnet.org/library/image/pt2003033t1.html

Just because you attach the label "stress" to a questionnaire or an electronic device does not mean that the label is accurate. How do you know, for example, that you are not measuring anxiety instead of stress?

How do you establish validity? It's too much for me to talk about today. There's an excellent book,

• Health Measurement Scales A Practical Guide to Their Development and Use. David L. Streiner, Geoffrey R. Norman (1989) New York: Oxford University Press, Inc.

which offers a lot of practical advice.

One point I want to emphasize is that establishing validity is not a one shot deal. You don't run a single statistical test and conclude valid/invalid on that basis. Validity is the slow and careful accumulation of information across multiple studies. Think of validity as a journey and not a destination.

The other point that I need to make is that a measurement that is valid in one population is not necessarily valid in another population. There are physiologic, cultural, and language differences between various ethnic and racial groups for example, and these differences could ruin a measurement that is nicely validated on one population.

Children represent a special group, and measurements validated on adults should not be considered as being automatically valid in children. There are certain things we take for granted in adults that are not true for children or which are only true for children above a certain age threshold.

1. The ability to speak.
2. The ability to read.
3. The ability to write.
4. The ability to understand abstract concepts.

By "abstract concepts" I don't mean "beauty" or "truth". Much simpler tasks, like marking a line on a visual analog scale to represent the degree of pain intensity or selecting a value from one to ten, require an understanding of numbers, order, and distance. A child that is not capable of counting from one to ten would not be able to perform a task like this. The situation is hopeless, though. For a child that has only limited verbal skills, the following graphic can be used to measure pain.

(Source: painsourcebook.ca/pdfs/pps92.pdf)

What about infants, who can't even understand an instruction about which face to point to? The issues are complex,

• Reflections on measuring pain in infants: dissociation in responsive systems and "honest signalling". Barr RG. Arch Dis Child Fetal Neonatal Ed 1998;79:F152-F156 (September). [Full text] [PDF]

but pain measurements are indeed possible, even in premature infants.

Another dimension to validity is the concept of a normal range. What is considered normal (and therefore safe) in adults is quite different from what is normal and safe in children. Heart rates, lung function, and blood pressure vary by age and by gender in children.

Assessing an appropriate BMI is a good example of the problems you face. In adults, there are well accepted standards: above 25 and you are overweight, 30 and higer and you are obese. In children, the values vary markedly by age and gender. Here's a  chart of the percentiles of BMI for girls ages 2 to 20 (reduced in size to make it fit better here).

(Source: www.cdc.gov/nchs/data/nhanes/growthcharts/set1clinical/cj41c024.pdf)

So a BMI of 20, for example, is troublesome in a 4 year old girl, but not in a 16 year old girl.

A third problem is the reliance on proxy reports. Younger children do not have sufficient language skills to be able to describe quality of life issues, so we rely on reports from parents and teachers. Are these people in a position to provide an informed opinion about the child?

Sample size concerns in a study involving children

For many diseases, the number of affected children is a small fraction of the number of adults who are affected. There are at least two reasons for this.

First, children represent a small fraction of the entire demographic population, with the possible exception of developing countries that still have very high birthrates. Here's the age distribution in England and Wales. The median age in England and Wales is 37 years, and less than 25% of the population is under the age of 20.

Source: www.statistics.gov.uk/census2001/pyramids/pages/727.asp

Second, some diseases are the result of decades of abuse that we heap upon ourselves through sedentary lifestyles, overeating, smoking, abuse of alcohol and other drugs. Children don't have easy access to tobacco and alcohol. Lack of exercise and too much television are indeed a problem for children, but they will not reap many of the consequences of this behavior until they become adults. Some medical conditions are consequences of the aging process and appear outside of adults only in rare cases.

There are some diseases and conditions, of course, which are unique to children, and others that occur. But as a general rule, children represent a small subgroup.

There are two statistical issues here associated with the adjective "small" and the noun "subgroup."

Small sample size problems represent the biggest crisis in medical research today. But with pharmacovigilance, it is that much worse, because you are  quite often for a "needle in a haystack."  Also remember that you are looking for something that went undetected with the typical sample size in a Phase III study.

So what sort of sample size do you need in a pharmacovigilance study? There are plenty of formulas, but one rule that I have found useful is the "rule of 50". It was originally developed by Gerald Van Belle. It appeared in the first draft of his excellent book, Statistical Rules of Thumb (ISBN: 0471402273), which was available on the web. Somehow, the "rule of 50" did not make it into the final draft of the book.

The rule of fifty applies when you are comparing two binomial proportions. A reasonable goal for your sample size is to have adequate power to detect a doubling or halving of the proportion. The sample size has to be large enough so that you observe 50 events in one group and 25 events in the other group.

The derivation of the rule of 50 is quite simple. The classic formula for sample size is

The notation here is fairly standard

	the probability of success in group 1,
	the probability of success in group 2,
	the probability of failure in group 1,
	the probability of failure in group 2,
	the number of patients in group 1,
	the number of patients in group 2,
	the average of p1 and p2,
	the average of q1 and q2,
	the ratio of n2 to n1,
	the difference in proportions, or p1-p2
	the percentile of a standard normal distribution associated with a probability of a Type I error of alpha (two sided test).
	the percentile of a standard normal distribution associated with a probability of Type II error of beta

Assume that the probability of success is small in both groups, so that the probability of failure is close to 1. Also assume that we are trying to detecting a 50% decline in the probability of success from group 1 to group 2. Assume that the ratio of the two sample sizes is 1. Finally, assume that we want an alpha level of 0.05 for a two sided test and a beta level of 0.20. With these assumptions, we get the following relationships.

Substitute these values into the formula for sample size to get

which simplifies to

We round this up to 50.

What does this mean in a practical sense? The quantity n₁p₁ represents the expected number of events in group 1, and you want to select n₁ sufficiently large so that n₁p₁ is at least 50. Alternately, select  n₂ sufficiently large so that n₂p₂ is at least 25 So if you are studying a side effect that occurs in about 1% of the untreated patients, and you want to be able to detect a doubling of this rate to 2% in the treated group, then select the number of treated patients, n, so that 0.02n=50 (or equivalently 0.01n=25). This tells you that your sample size should be approximately 2,500.

A real world application of the rule of 50. An article by Schwartz et al proposes an interesting scenario for a research study (N Engl J Med. 1998;338:1709-1714). These authors noticed an association between prolonged QT interval and Sudden Infant Death Syndrome. In the discussion of these findings, the authors raise the possibility of screening all newborn infants using electrocardiography and the placing those infants with prolonged QT intervals on a beta blocker. The authors discuss the complexity of the cost benefit issues, which is beyond the scope of this web page. It is interesting, however, to speculate on how to test whether beta blockers would be effective as a therapy to prevent SIDS in those infants with long QT intervals.

The paper provides much interesting data to help you calculate an appropriate sample size for this study. The risk of SIDS in infants with prolonged QT intervals is 1.5%. Suppose that a beta blocker could cut this risk in half (to 0.75%). What sample size would you have to collect in order to have adequate power?

The rule of 50 tells us that we would need 50 SIDS events in the placebo arm of the trial. At a rate of 1.5% that translates into recruiting 50 / 0.015 = 3,333 infants with prolonged QT interval for the placebo arm. You would recruit a similar number of infants for the beta-blocker arm of the study.

Not every infant, however, will have a prolonged QT interval. The cutoff used in this paper for a prolonged interval represented the 97.5 percentile. So only 2.5% of the infants screened could qualify to be in the study. In order to recruit 6,666 infants who qualify for the study, you would have to screen 6,666 / 0.025 = 266,640 normal infants.

Subgroup analyses. Subgroup analyses raise troubling issues. The number of possible subgroups that could be analyzed is large. But if you adjust for multiple comparisons, you lose a lot of power. And this loss of power is exacerbated by the sample size of the subgroups which is always less (sometimes much less) than the total sample size.

Another big problem is that many researchers apply the wrong statistical test to the subgroup by testing each subgroup independently from the overall results. So if you have an overall p-value of 0.06, but you compute a slightly smaller p-value, 0.04, in children, there are some who would say that the side effect is not present in adults, but is present in children. The only appropriate test for subgroups is a formal test of interaction.

So what should you do? There is a nice overview on subgroup analysis in clinical trials,

• Subgroup analysis in clinical trials. Cook DI, Gebeski VJ, Keech AC, MJA 2004; 180 (6): 289-291. [Full text] [PDF]

and the discussion can be easily applied to pharmacovigilance trials. This paper offers the following checklist:

Box 2. Checklist for subgroup analyses.

Design

• Are the subgroups based on pre-randomisation characteristics?
• What is the impact of patient misallocation on the subgroup analysis?
• Is the intention-to-treat population being used in the subgroup analysis?
• Were the subgroups planned a priori?
• Were they planned in response to existing trial or biological data?
• Was the expected direction of the subgroup effect stated a priori?
• Was the trial designed to have adequate power for the proposed subgroup analysis?

Reporting

• Is the total number of subgroup analyses undertaken declared?
• Are relevant summary data, including event numbers and denominators, tabulated?
• Are analyses decided on a priori clearly distinguished from those decided on a posteriori?

Statistical analysis

• Are the statistical tests appropriate for the underlying hypotheses?
• Are tests for heterogeneity (ie, interaction) statistically significant?
• Are there appropriate adjustments for multiple testing?

Interpretation

• Is appropriate emphasis being placed on the primary outcome of the study?
• Is the validity of the findings of the subgroup analysis discussed in the light of current biological knowledge and the findings from similar trials?

Many pharmacovigilance studies are not randomized, and many of the findings in these studies are post hoc. This makes subgroup analysis even more troubling, but you don't really have a choice. I have found the criteria for causation developed in a 1965 article by Sir Austin Bradford Hill "The Environment and Disease: Association or Causation" to be very helpful. He mentions nine factors:

1. Strength (is the risk of a side effect so large that we can easily rule out other factors)
2. Consistency (have the results have been replicated by different researchers and under different conditions)
3. Specificity (is the exposure associated with a very specific side effect as opposed to a wide range of side effects)
4. Temporality (did the drug treatment precede the side effect)
5. Biological gradient (are increasing drug dosages associated with increasing risks of the side effect)
6. Plausibility (is there a credible scientific mechanism that can explain the association)
7. Coherence (is the association consistent with the natural history of the side effect)
8. Experimental evidence (does a physical intervention show results consistent with the association)
9. Analogy (is there a similar result that we can draw a relationship to)

There is no magic rule, such as 7 out of 9 will guarantee causation. Instead, presence of any of the nine factors will strengthen the quality of the evidence and absence of any factor will dilute the quality of the evidence.

Data entry issues in a study involving children.

There are some important data elements in a pediatric research study that are not normally included in an adult research study. Not all of these elements should be collected in every research study, of course. The context of the problem you are working on is critical.

More detailed information about age:

Most studies report age truncated to the lower integer. So an age of 51 means that the patient has experienced his/her 51st birthday, but not the 52nd birthday. In children, you may need a greater level of detail. If you wish to use a growth chart, you need to know age in months. For infants, age in days may be necessary. If you are studying pre-term births, you should also specify the post-conceptual age. In some studies where the onset of puberty is critical, you might need to include a Tanner stage.

Characteristics of the delivery and pre-natal exposures:

Although the effects in older children is less apparent, for younger children, and especially infants, you might need to collect information about the birth process, such as birthweight, length of stay in the birth hospital, and apgar scores. The type of delivery (vaginal or C-section) might be important. You might also note prenatal exposure to potentially harmful substances, like alcohol and tobacco.

Characteristics of the child's environment:

In adult study, you establish socioeconomic status (SES) by examining the level of education and the type of job held by the patient. You can't do this for a child, of course, so you have to assess SES by education level of the parents and their job category. If you are assessing developmental delays, then should probably measure factors about the richness of the child's environment, such as the number and types of toys available and the vocabulary skills of the primary caregiver. Where does the child spend most of the day? Childcare arrangements, such as enrollment in a daycare center can sometimes be very important. For older children, where they go to school is important, especially with the rise of home schooling.

Characteristics of the child's diet.

The dietary information for a child is a bit different than of an adult, because a young child cannot fill out a food diary and must rely on a proxy report of diet. For infants, of course, you might need details about the duration of breastfeeding and the age at which solid foods were introduced.

What if you can't measure these things?

In many research situations, but especially in pharmacovigilance studies, you will not have access to many of these child-specific variables. In some situations, the loss is trivial. What you do lose when you fail to collect this extra information is the ability to make risk adjustments. If these variables are imbalanced across comparison groups and these variables are strongly associated with a particular safety indicator, you have the most cause for concern.

Summary

In a research study involving children, you have several statistical concerns. First, you may need to take extra efforts to establish validity in measurements in children. Second, you have to deal with sample size concerns and subgroup analysis issues. Finally, you may need to collect a greater level of detail about the child's age, information about the pre-natal exposures and delivery and possibly special data on the child's environment and diet.

This webpage was written by Steve Simon on 2006-11-02, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research

---

Searching for pediatric articles on Medline (October 26, 2006).

A recent publication

• Age-Specific Search Strategies for Medline. Monika Kastner, Nancy L Wilczynski, Cindy Walker-Dilks, Kathleen Ann McKibbon, Brian Haynes. J Med Internet Res 2006 (Oct 25); 8(4):e25. [Full text]

examines search strategies for articles relevant to geriatric medicine, adult medicine, pediatric medicine, neonatal medicine, and obstetrics. For studies of pediatric medicine, the most sensitive search used the following terms:

• child:.mp. OR adolescent.mp. OR infan:.mp.

which had a sensitivity of 98% and a specificity of 81%. The colon after "child" tells MedLine to search for any word beginning with "child." This allows you to search for either "child" or "children". The .mp is a PubMed tag for multiple posting. It searches for terms that appear in the title, abstract, or subject heading. The search that maximized specificity was

• children.tw

which had a specificity of 97% but a sensitivity of only 58%. The .tw is a PubMed tag for text word, and will search for a word appearing in a variety of locations.

Includes all words and numbers in the title, abstract, other abstract, MeSH terms, MeSH Subheadings, Publication Types, Substance Names, Personal Name as Subject, MEDLINE Secondary Source, and Other Terms (see Other Term [OT] above) typically non-MeSH subject terms (keywords), including NASA Space Flight Mission, assigned by an organization other than NLM. www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=helppubmed.section.pubmedhelp.Search_Field_Descrip 

The best compromise was

• adolescent.tw. OR children.tw. OR child, preschool.sh.

which produced 89% sensitivity and 87% specificity. The .sh is a PubMed tag for MeSH subheading. MeSH is an acronym for Medical Subject Heading and represents an effort by the coders at PubMed to classify the medical specialties associated with the article being listed.

NLM's Medical Subject Headings controlled vocabulary of biomedical terms that is used to describe the subject of each journal article in MEDLINE. MeSH contains more than 23,000 terms and is updated annually to reflect changes in medicine and medical terminology. MeSH terms are arranged hierarchically by subject categories with more specific terms arranged beneath broader terms. PubMed allows you to view this hierarchy and select terms for searching in the MeSH Database. Skilled subject analysts examine journal articles and assign to each the most specific MeSH terms applicable - typically ten to twelve. Applying the MeSH vocabulary ensures that articles are uniformly indexed by subject, whatever the author's words. www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=helppubmed.section.pubmedhelp.Search_Field_Descrip 

What does sensitivity and specificity mean in the context of a Medline search. The JMIR article explains it thusly:

Sensitivity for a given age-specific topic is defined as the proportion of relevant articles (ie, articles with the desired age-specific content) that are retrieved; specificity is the proportion of nonrelevant articles (ie, articles that are outside the desired age-specific content) not retrieved.

This webpage was written by Steve Simon on 2006-10-26, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research,  Category: Information searching

---

Case studies for differences in research in children (October 16, 2006).

I want to include several case studies about issues relating to the differences in research in children. Here are some articles with full free text on the web which I think might generate a lot of interesting discussion.

• Safety and efficacy of the nicotine patch and gum for the treatment of adolescent tobacco addiction. E. T. Moolchan, M. L. Robinson, M. Ernst, J. L. Cadet, W. B. Pickworth, S. J. Heishman, J. R. Schroeder. Pediatrics 2005: 115(4); e407-14. [Medline] [Abstract] [Full text] [PDF]
• A double-blind, randomized, placebo-controlled trial of acupuncture for the treatment of childhood persistent allergic rhinitis. D. K. Ng, P. Y. Chow, S. P. Ming, S. H. Hong, S. Lau, D. Tse, W. K. Kwong, M. F. Wong, W. H. Wong, Y. M. Fu, K. L. Kwok, H. Li, J. C. Ho. Pediatrics 2004: 114(5); 1242-7. [Medline] [Abstract] [Full text] [PDF]
• The safety and efficacy of inhaled dry powder mannitol as a bronchial provocation test for airway hyperresponsiveness: a phase 3 comparison study with hypertonic (4.5%) saline. J. D. Brannan, S. D. Anderson, C. P. Perry, R. Freed-Martens, A. R. Lassig, B. Charlton. Respir Res 2005: 6; 144. [Medline] [Abstract] [Full text] [PDF]
• Vagal nerve stimulation in refractory epilepsy: the first 100 patients receiving vagal nerve stimulation at a pediatric epilepsy center. J. V. Murphy, R. Torkelson, I. Dowler, S. Simon, S. Hudson. Arch Pediatr Adolesc Med 2003: 157(6); 560-4. [Medline] [Abstract] [Full text] [PDF]

This webpage was written by Steve Simon on 2006-10-16, edited by Steve Simon, and was last modified on 2008-07-08. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page. Category: Children in research,  Category: Information searching

---

Conclusion

There are indeed important issues that you need to consider when conducting research in children. I'm hoping that this presentation will encourage you to tackle these problems though rather than view this as a series of insurmountable obstacles. We desperately need more work in this area.

Every practicing physician, especially pediatricians and pediatric surgeons, departments of pediatrics, and departments of pharmacology should closely examine their own capacities and performance in this area of greatly needed activity. If we are to have drugs of better efficacy and safety for children, those responsible for child care will have to assume this responsibility for developing active programs of clinical pharmacology and drug testing in infants and children. The alternative is to accept the status of "Therapeutic Orphans" for their patients. (Editorial Comment: Therapeutic Orphans. Shirkey H. Pediatrics 104(3): 583-584. [Full text] [PDF])