Factors which encourage compliance

(4)

• Minimize the complexity of the drug regimen.  This is the most important factor in determining patient compliance; once per day regimens should be used whenever possible.   
• CASE STUDY:  Nurses in a hospital ward were directly observed to determine compliance in a study by Vere (Lancet 1:370, 1965).  It was determined that errors in drug administration increased sharply with increasing numbers of drug orders.
• Adequate one-to-one education of the patient is the second most important factor in encouraging drug compliance.  In a study by Hecht (Nurs. Forum XIII: 112, 1974), the rate of serious errors in self-administration of drugs was decreased from 45% to 17% by intense training of the patients in a personalized way. This will not work if the health care worker does not take the time to develop an appropriate therapeutic relationship with the patient.
• Doctors, nurses, or pharmacists can all be successfully used for training.  In contrast, it has been shown that providing a patient with a slide show and booklet on, for example, hypertension and antihypertensive drugs improves understanding of the disease but does not foster compliance.

Factors which encourage compliance

Education Must include a discussion of the Following Topics

(7)

• What benefits will the patient derive from taking the drug(s)?
• How can the progress of therapy be monitored?
• What are the consequences of noncompliance?
• What side effects and drug interactions might occur?
• How should side effects and drug interactions be handled if they occur?
• Will the patient become “addicted” to the drug(s)?
• Prescription parameters:  how, how much and how long?

Other factors which mitigate against compliance

(10)

• Lack of discernible progress in the therapy (ex: hypertension or hyperlipidemia).
• Unpleasant taste.
• Cost of medication.
• Poor past treatment experience(s).
• Peer or family pressure (especially with mental health drugs).
• Special patient factors (ex: elderly patients and/or children).
• poor eyesight or inability to read
• confusion or inability to understand instructions
• difficulty remembering to take the medicine
• dependence on others for medication

Methods of assessing compliance other than direct observation include the following:

(6)

• Physician’s Estimation of Patient Compliance.  
• Patient’s reports of drug use always overestimate compliance.  
•  CASE STUDY:  Children with strep throat were assigned to take a 10-day course of oral penicillin.  Urine samples were collected throughout the study and parents were questioned after the treatment period.  Most parents (83%) claimed that all doses had been administered.  However, measurement of penicillin in the urine revealed that 55% and 70% of the children took no drug after the 3^rd and the 6^th days, respectively.   (Bergman et. al., NEJM, 268:1334, 1963.)
• Pill counting.  
• Mechanical medication monitors keep a log of times when a bottle is opened or medicine packets are removed from a box. 
• Measurement of Drug Concentrations in Biological Fluids.  

Methods of assessing compliance other than direct observation include the following:

Physician’s Estimation of Patient Compliance.

Methods of assessing compliance other than direct observation include the following:

Patient’s reports

Methods of assessing compliance other than direct observation include the following:

Pill counting.  

Methods of assessing compliance other than direct observation include the following:

Mechanical medication monitors 

Methods of assessing compliance other than direct observation include the following:

Measurement of Drug Concentrations in Biological Fluids.  

• This technique is the most reliable.  It is most useful if the drug has a long half-life as this allows estimation of drug use over a longer period.  Alternatively, an easily assayable tracer such as riboflavin that has a long half-life can be added to the normal medication and the tracer can be measured.

Consequences of Inadequate Compliance.

(6)

• Inadequate compliance might jeopardize the intended therapy.
• If the therapy fails, the doctor might be misled.  
• He or she might inaccurately conclude that either the initial diagnosis was incorrect or that the selection or dose of the drug was wrong.  
• In either case, the physician might then choose to switch to a different drug that might not work as well or might be more toxic or to increase the drug dosage with a consequent increase in the risk of toxicity.  
• Therefore, if a physician does not have an objective measure of patient compliance and if initial therapy fails, noncompliance should always be considered as a possible explanation for the lack of therapeutic success.
• If the drug has a narrow therapeutic index, a noncompliant patient might be at risk when suddenly forced to be compliant.  For example, a poorly compliantpatient on digitalis might develop toxicity after receiving the “normal” dose for several days in the hospital. 

The Placebo Effect

(4)

• The term “placebo” has as its etymological origin the Latin verb meaning “I shall please”.  In its pharmacological context, a placebo is a drug that lacks therapeutic efficacy and is given merely to satisfy (or “please”) a patient.  
• The placebo effect is “any effect attributable to a pill, potion, or procedure, but not to its pharmacodynamic or specific properties” (Wolf, Pharm. Rev., 11:689, 1959).
• Placebos primarily act by calming a patient’s anxieties;    the greater the mental component of a condition, the greater the likelihood that a placebo will provide relief.  
• As a result, it is hardly surprising that “impure” placebos (i.e., those lacking therapeutic effect but capable of causing an unrelated action such as sugar pills containing niacin) seem to produce a greater placebo effect than “pure” placebos (i.e., those without any discernable effect such as plain sugar pills).

The Placebo Effect

How Powerful Really is the Placebo

• Beecher (“The Powerful Placebo”; 1955) reported that 35% of patients treated with placebos receive relief from conditions including:
• Postoperative wound pain  
• Cough
• Mood changes
• Angina pectoris
• Headache
• Seasickness
• Anxiety
• Tension

• Others reported over the last fifty years that a smaller fraction (20-25%) of patients having status asthmaticus, hypertension, and mental depression also benefit from placebos. 

The Placebo Effect

How Powerful Really is the Placebo

Recent Study

• In contrast, a recent retrospective analysis of 114 prior studies challenged the efficacy of placebos (Is the Placebo Powerless? NEJM 344:1594, 2001).   
• The study established a moderate placebo effect for relief of pain but none for any other disorder.  The authors concluded that placebos should ONLY be used in clinical trials and an editor of NEJM went so far as to call for a reduction in use of placebos (but not for their elimination).
•  However, this analysis was confounded because the data was pooled either
• for different types of placebos (pharmacological (50%), physical, psychological) or 
• for different types of disorders (pain, obesity, asthma, hypertension, insomnia, anxiety, nausea, smoking, depression).  Also, it didn’t differentiate between active and inactive placebos.  

The Placebo Effect

• “Placebo reactors” are individuals who obtain positive responses from use of a placebo.  
• As with compliance, it is impossible to identify these individuals by psychological testing either in advance or retrospectively.  
• Doctors sometimes regard placebo reactors as individuals with hysterical tendencies, but this is not so.  
• Depending on current emotional state and psychological expectations, it is thought that every individual is potentially a placebo reactor.  
• This point was driven home in the following study.

The Placebo Effect

• CASE STUDY:  Medical students were told that they were taking part in an evaluation of a mild sedative (blue pills) and a mild stimulant (pink pills). After placebo administration, drug-associated changes were reported in 30% of the students and were severe in several individuals.  
• Two pills produced more noticeable changes than one.  
• Among the reactors,
• blue pills were reliably associated with sedation while pink pills were associated with stimulation.  
• All of the students rated the experiment highly as a learning experience but strong responders felt humiliated. 

The Placebo Effect

Toxicities

• Toxicities of the placebo would be presumed to be nonexistent.  However, taken in a broad sense, subtle but potentially serious toxicities are associated with the use of placebos.
• Any existing doctor-patient relationship will be destroyed if a physician is caught prescribing placebos.
• They may divert patients from seeking more effective treatment for a real disease.
• The patient’s belief that symptoms are due to an underlying physical illness will be strengthened.
• Patients may become psychologically addicted to a placebo.
• A physician might inappropriately give a patient a placebo when the diagnosis is not apparent or determination of what is actually wrong would be difficult.

NONCOMPLIANCE AND THE PLACEBO EFFECT IN DRUG TESTING

• An understanding of the effects of noncompliance on clinical trials is extremely important.
• First, if you are comparing the effects of the currently accepted drug to a new drug, then anuneven distribution of noncompliers among test groups can bias the results and lead to either a false positive (due to too many noncompliers in the standard treatment group) or a false negative (due to too many noncompliers in the group receiving the new drug).  In the first case, an ineffective drug would erroneously be judged to be effective while, in the latter case, an effective drug would appear to be ineffective.  In clinical testing, any noncompliers must be equally distributed among treatment groups.
• Second, the presence of noncompliers, even if they are uniformly distributed among treatment groups, diminishes the sensitivity of a clinical trial and, third, leads to an underestimation of drug toxicity.

NONCOMPLIANCE AND THE PLACEBO EFFECT IN DRUG TESTING

Read the rest of the handout