Randomisation methods in controlled trials

Study protocol: delayed intervention randomised controlled trial within the Medical Research Council (MRC) Framework to assess the effectiveness of a new palliative care service

BACKGROUND

Palliative care has been proposed to help meet the needs of patients who suffer progressive non-cancer conditions but there have been few evaluations of service development initiatives. We report here a novel protocol for the evaluation of a new palliative care service in this context.

METHODS/DESIGN

Using the MRC Framework for the Evaluation of Complex Interventions we modelled a new palliative care and neurology service for patients severely affected by Multiple Sclerosis (MS). We conducted qualitative interviews with patients, families and staff, plus a literature review to model and pilot the service. Then we designed a delayed intervention randomised controlled trial to test its effectiveness as part of phase II of the MRC framework. Inclusion criteria for the trial were patients identified by referring clinicians as having unresolved symptoms or psychological concerns. Referrers were advised to use a score of greater than 8 on the Expanded Disability Scale was a benchmark. Consenting patients newly referred to the new service were randomised to either receive the palliative care service immediately (fast-track) or after a 12-week wait (standard best practice). Face to face interviews were conducted at baseline (before intervention), and at 4-6, 10-12 (before intervention for the standard-practice group), 16-18 and 22-24 weeks with patients and their carers using standard questionnaires to assess symptoms, palliative care outcomes, function, service use and open comments. Ethics committee approval was granted separately for the qualitative phase and then for the trial.

DISCUSSION

We publish the protocol trial here, to allow methods to be reviewed in advance of publication of the results. The MRC Framework for the Evaluation of Complex Interventions was helpful in both the design of the service, methods for evaluation in convincing staff and the ethics committee to accept the trial. The research will provide valuable information on the effects of palliative care among non-cancer patients and a method to evaluate palliative care in this context.

Natural experiments: an underused tool for public health?

Policymakers and public health researchers alike have demanded better evidence of the effects of interventions on health inequalities. These calls have been repeated most recently in the UK in the final Wanless report, which spoke of the "almost complete lack of an evidence base on the cost-effectiveness of public health interventions", and pointed more generally to the limited evidence base for public health policy and practice. Wanless and others have suggested that the gaps may be partially filled by exploiting the opportunities offered by "natural experiments", such as changes in employment opportunities, housing provision, or cigarette pricing. Natural experiments have an important contributions to make within the health inequalities agenda. First, they can play an important role in investigating the determinants of health inequalities. Second, they can assist in the identification of effective interventions, an area where it is widely acknowledged that the evidence-base is currently sparsely populated. This paper discusses some of the benefits and limitations of using this type of evidence, drawing on two ongoing quasi-experimental studies as examples.

A randomised clinical trial to investigate bond failure rates using a self-etching primer

This clinical trial evaluated, over a 12-month period, the performance of brackets bonded to teeth etched and primed with Transbond Plus Self-Etching Primer (SEP) when compared with a conventional separate two-step etch and primer system. Thirty-nine randomly selected patients requiring fixed appliance therapy were entered into the study. Random allocation of each etching system, along with a 'split-mouth cross-quadrant' design was used. A total of 661 brackets were placed by two operators. The failure and survival rates of the brackets were determined for age and gender of the patients, each etching system, operator, mode of failure, tooth position in the dental arch, and number of manipulations prior to curing the adhesive. Statistical analysis showed that SEP had a significantly higher bond failure rate (11.2 per cent) than the conventional etch and primer system (3.9 per cent) at the P = 0.001 level. Cox's proportional hazards regression showed the conventional etch and primer system to have a 60 per cent reduced chance of bracket failure over a 12-month observation period, while males had a 2.4 times increased risk compared with females. The predominant mode of failure was at the composite enamel interface for the SEP, while for the conventional etch and primer system, it was within the composite adhesive. No statistically significant differences were found for the failure rate with respect to the age of the patient, operator, tooth location, or the number of manipulations of the bracket. This in vivo study showed that brackets bonded using SEP had an increased clinical bond failure rate compared with the conventional, separate, etch and prime system.

An investigation of minimisation criteria

Background

Minimisation can be used within treatment trials to ensure that prognostic factors are evenly distributed between treatment groups. The technique is relatively straightforward to apply but does require running tallies of patient recruitments to be made and some simple calculations to be performed prior to each allocation. As computing facilities have become more widely available, minimisation has become a more feasible option for many. Although the technique has increased in popularity, the mode of application is often poorly reported and the choice of input parameters not justified in any logical way.

Methods

We developed an automated package for patient allocation which incorporated a simulation arm. We here demonstrate how simulation of data can help to determine the input parameters to be used in a subsequent application of minimisation.

Results

Several scenarios were simulated. Within the selected scenarios, increasing the number of factors did not substantially adversely affect the extent to which the treatment groups were balanced with respect to the prognostic factors. Weighting of the factors tended to improve the balance when factors had many categories with only a slight negative effect on the factors with fewer categories. When interactions between factors were included as minimisation factors, there was no major reduction in the balance overall.

Conclusion

With the advent of widely available computing facilities, researchers can be better equipped to implement minimisation as a means of patient allocation. Simulations prior to study commencement can assist in the choice of minimisation parameters and can be used to justify those selections.

Understanding randomised controlled trials

The hierarchy of evidence in assessing the effectiveness of interventions or treatments is explained, and the gold standard for evaluating the effectiveness of interventions, the randomised controlled trial, is discussed. Issues that need to be considered during the critical appraisal of randomised controlled trials, such as assessing the validity of trial methodology and the magnitude and precision of the treatment effect, and deciding on the applicability of research results, are discussed. Important terminologies such as randomisation, allocation concealment, blinding, intention to treat, p values, and confidence intervals are explained.

Blood-pressure control for renoprotection in patients with non-diabetic chronic renal disease (REIN-2): multicentre, randomised controlled trial

BACKGROUND

In chronic nephropathies, inhibition of angiotensin-converting enzyme (ACE) is renoprotective, but can further renoprotection be achieved by reduction of blood pressure to lower than usual targets? We aimed to assess the effect of intensified versus conventional blood-pressure control on progression to end-stage renal disease.

METHODS

We undertook a multicentre, randomised controlled trial of patients with non-diabetic proteinuric nephropathies receiving background treatment with the ACE inhibitor ramipril (2.5-5 mg/day). We randomly assigned participants either conventional (diastolic <90 mm Hg; n=169) or intensified (systolic/diastolic <130/80 mm Hg; n=169) blood-pressure control. To achieve the intensified blood-pressure level, patients received add-on therapy with the dihydropyridine calcium-channel blocker felodipine (5-10 mg/day). The primary outcome measure was time to end-stage renal disease over 36 months' follow-up, and analysis was by intention to treat.

FINDINGS

Of 338 patients who were randomised, three (two assigned intensified and one allocated conventional blood-pressure control) never took study drugs and they were excluded. Over a median follow-up of 19 months (IQR 12-35), 38/167 (23%) patients assigned to intensified blood-pressure control and 34/168 (20%) allocated conventional control progressed to end-stage renal disease (hazard ratio 1.00 [95% CI 0.61-1.64]; p=0.99).

INTERPRETATION

In patients with non-diabetic proteinuric nephropathies receiving background ACE-inhibitor therapy, no additional benefit from further blood-pressure reduction by felodipine could be shown.

Conservative treatment in patients with an acute lumbosacral radicular syndrome: design of a randomised clinical trial [ISRCTN68857256]

Background

The objective is to present the design of randomised clinical trial (RCT) on the effectiveness of physical therapy added to general practitioners management compared to general practitioners management only in patients with an acute lumbosacral radicular syndrome (also called sciatica).

Methods/Design

Patients in general practice diagnosed with an acute (less than 6 weeks) lumbosacral radicular syndrome and an age above 18 years are eligible for participation. The general practitioners treatment follows their clinical guideline. The physical therapy treatment will consist of patient education and exercise therapy. The primary outcome measure is patients reported global perceived effect. Secondary outcome measures are severity of complaints, functional status, health status, fear of movement, medical consumption, sickness absence, costs and treatment preference. The follow-up is 52 weeks.

Discussion

Treatment by general practitioners and physical therapists in this study will be transparent and not a complete "black box". The results of this trial will contribute to the decision of the general practitioner regarding referral to physical therapy in patients with an acute lumbosacral radicular syndrome.

Effectiveness of behavioural graded activity compared with physiotherapy treatment in chronic neck pain: design of a randomised clinical trial [ISRCTN88733332]

Background

Chronic neck pain is a common complaint in the Netherlands with a point prevalence of 14.3%. Patients with chronic neck pain are often referred to a physiotherapist and, although many treatments are available, it remains unclear which type of treatment is to be preferred.

The objective of this article is to present the design of a randomised clinical trial, Ephysion, which examines the clinical and cost effectiveness of behavioural graded activity compared with a physiotherapy treatment for patients with chronic non-specific neck pain.

Methods

Eligible patients with non-specific neck pain persisting longer than 3 months will be randomly allocated to either the behavioural graded activity programme or to the physiotherapy treatment. The graded activity programme is based on an operant approach, which uses a time-contingent method to increase the patient's activity level. This treatment is compared with physiotherapy treatment using a pain-contingent method.

Primary treatment outcome is the patient's global perceived effect concerning recovery from the complaint. Global perceived effect on daily functioning is also explored as primary outcome to establish the impact of treatment on daily activity. Direct and indirect costs will also be assessed. Secondary outcomes include the patient's main complaints, pain intensity, medical consumption, functional status, quality of life, and psychological variables. Recruitment of patients will take place up to the end of the year 2004 and follow-up measurement will continue until end 2005.

How many do I need? Basic principles of sample size estimation

BACKGROUND

In conducting randomized trials, formal estimations of sample size are required to ensure that the probability of missing an important difference is small, to reduce unnecessary cost and to reduce wastage. Nevertheless, this aspect of research design often causes confusion for the novice researcher.

AIM

This paper attempts to demystify the process of sample size estimation by explaining some of the basic concepts and issues to consider in determining appropriate sample sizes.

METHOD

Using a hypothetical two group, randomized trial as an example, we examine each of the basic issues that require consideration in estimating appropriate sample sizes. Issues discussed include: the ethics of randomized trials, the randomized trial, the null hypothesis, effect size, probability, significance level and type I error, and power and type II error. The paper concludes with examples of sample size estimations with varying effect size, power and alpha levels.

CONCLUSION

Health care researchers should carefully consider each of the aspects inherent in sample size estimations. Such consideration is essential if care is to be based on sound evidence, which has been collected with due consideration of resource use, clinically important differences and the need to avoid, as far as possible, types I and II errors. If the techniques they employ are not appropriate, researchers run the risk of misinterpreting findings due to inappropriate, unrepresentative and biased samples.

Quality of RCTs in periodontology--a systematic review

Randomization, concealment of treatment allocation, blinding, and patient follow-up are key quality components of randomized controlled trials (RCTs). The objective of this study was to assess the quality of RCTs in periodontology using these evidence-based components. Following a detailed search, screening and quality assessments of RCTs were conducted in duplicate and independently. The results showed that although 91% of trials were described as randomised, adequate methods for randomization and allocation concealment were found in 17% and 7% of studies, respectively. Blinding was adequate for the caregiver in 17% and for the examiner in 55% of studies. A clear accounting of all participants was present in 56% of reports. This rigorous systematic review revealed that the quality of RCTs in periodontology, judged by their publications, frequently does not meet recommended standards. If this quality is reflected in actual study conduct, fundamental errors could have a significant impact on the outcomes of these trials.

School-based indicated prevention: a randomised trial of group therapy

BACKGROUND

One hundred and twenty-two children identified by teachers as at risk for behavioural or emotional problems were randomly allocated to drama-group therapy or to a curriculum-studies control, based in school.

METHODS

One hundred and seventeen completed the intervention phase of the trial, which comprised 12 hour-long sessions. Post-intervention self-reports showed significant effects associated with both interventions.

RESULTS

However, there was a clear advantage of group therapy over both a waiting list control and curriculum studies, according to teacher reports. This was true also of categorical analyses focusing on those with the most severe symptoms.

CONCLUSIONS

These analyses confirmed sustained teacher-reported improvement over a year-long follow-up period.

A randomized controlled clinical trial of intravitreous fomivirsen for treatment of newly diagnosed peripheral cytomegalovirus retinitis in patients with AIDS

PURPOSE

To assess the efficacy of intravitreous fomivirsen sodium, an antisense oligonucleotide, for newly diagnosed peripheral cytomegalovirus (CMV) retinitis in patients with acquired immunodeficiency syndrome (AIDS).

DESIGN

Randomized treatment intervention clinical trial.

METHODS

A multicenter, prospective, and randomized clinical trial compared immediate treatment of CMV retinitis with fomivirsen (165 microg administered intravitreously) to deferral of treatment until CMV retinitis lesions progressed by standard definitions. Included were patients with CMV retinitis lesions at least 750 microm outside of zone 1. Fomivirsen was injected weekly for three doses as induction therapy, followed by injection every other week as maintenance therapy. All patients were examined regularly until evidence of retinitis progression. Time to first progression was determined by two independent masked fundus photography reading centers (principal outcome) and by clinician investigators based on indirect ophthalmoscopy. Patients in the deferral of treatment group were offered fomivirsen therapy at the time of clinically determined retinitis progression.

RESULTS

Patients in the immediate treatment group (n = 18) and the deferral of treatment group (n = 10) were comparable at baseline. Median time to first progression of disease for the immediate treatment group was 71 days (95% confidence interval [CI] 28 days-not determinable) and for the deferral of treatment group was 13 days (95% CI 9-15 days; P =.0001, Wilcoxon rank sum test). Progression occurred in 44% of patients in the immediate treatment group during the study compared with 70% of patients in the deferral of treatment group during the study. There were no retinal detachments among eyes treated with fomivirsen.

CONCLUSIONS

Fomivirsen is an effective treatment for CMV retinitis in patients with AIDS that utilizes a mechanism of action different than that of ganciclovir, foscarnet, and cidofovir.

A clinical trial of the anticaries efficacy of casein derivatives complexed with calcium phosphate in patients with salivary gland dysfunction

OBJECTIVE

The purpose of this study was to compare the caries preventive efficacy of a mouthrinse solution containing casein derivatives coupled with calcium phosphate (CD-CP) with that of a 0.05% sodium fluoride mouthrinse among individuals with dry mouth.

STUDY DESIGN

A randomized control trial design was used. Participants included individuals who had had radiotherapy for head and neck cancer (n = 82) and others with Sjögren's syndrome (n = 56). Baseline data collection was followed by reexamination 12 months later. Posterior bite-wing radiographs were taken on both occasions.

RESULTS

A total of 124 participants, 61 (49.2%) in the sodium fluoride group and 63 (50.8%) in the CD-CP group, completed the 12-month examination. The baseline characteristics of the 2 groups did not differ. Coronal caries incidence was higher in the sodium fluoride group than in the CD-CP group (34.4% and 27%, respectively), but the difference was not statistically significant. Similarly, the small difference in coronal caries increment between the 2 groups was not statistically significant (0.4 and 0.3 surfaces, respectively). There was insufficient root surface caries experience between the 2 groups observed for differences to be examined. Proportionately more of the CD-CP group lost 1 or more teeth, and the mean number of tooth loss was higher. The participants with the highest incidence and increment were those with Sjögren's syndrome in the CD-CP group. Some of that difference was accounted for by differences in baseline caries status.

CONCLUSION

It appears that CD-CP preparations hold promise as caries preventive agents for individuals with dry mouth, although confirmation of this study's findings in other settings is warranted before a definitive conclusion can be reached.

Splinting or surgery for carpal tunnel syndrome? Design of a randomized controlled trial [ISRCTN18853827]

BACKGROUND

Carpal tunnel syndrome is a common disorder, which can be treated with surgery or conservative options. However, there is insufficient evidence and no consensus among physicians with regard to the preferred treatment for carpal tunnel syndrome. Therefore, a randomized controlled trial is conducted to compare the short- and long-term efficacy of surgery and splinting in patients with carpal tunnel syndrome. An attempt is also made to avoid the (methodological) limitations encountered in earlier trials on the efficacy of various treatment options for carpal tunnel syndrome.

METHODS

Patients of 18 years and older, with clinically and electrophysiologically confirmed idiopathic carpal tunnel syndrome, are recruited by neurologists in 13 hospitals. Patients included in the study are randomly allocated to either open carpal tunnel release or wrist splinting during the night for at least 6 weeks. The primary outcomes are general improvement, waking up at night and severity of symptoms (main complaint, night and daytime pain, paraesthesia and hypoesthesia). Outcomes are assessed up to 18 months after randomization.

Placebos, placebo effect, and the response to the healing situation: the evolution of a concept

In spite of its impressive progress, medicine has been strongly criticized for relying on its modern biomedical tradition to the neglect of the psychosocial aspects of health. This neglect may account for patients' dissatisfaction and eventual use of alternative health approaches. The concept of placebo has sustained dramatic "protean" metamorphoses through the ages. For centuries, placebos have been regarded as powerful deceptive therapies. From the middle of the twentieth century, however, conventional medicine has used placebos as methodologic tools to distinguish the specific from the nonspecific ingredients in treatments. In modern medical research, the double-blind, placebo-controlled, randomized clinical trial has been established as the gold standard for the assessment of any new treatment. Recently a new trend regarding placebos seems to have emerged. The placebo and other nonspecific effects elicited by the "healing situation" have been independently subjected to scientific study. Progress in this area may promote useful clinical applications, enabling physicians to broaden their perspectives on the healing process. We present the historical changes of the concept of placebo and the ethical issues raised by their use.

Information and advice to patients with back pain can have a positive effect. A randomized controlled trial of a novel educational booklet in primary care

STUDY DESIGN

A double-blind, randomized controlled trial of a novel educational booklet compared with a traditional booklet for patients seeking treatment in primary care for acute or recurrent low back pain.

OBJECTIVE

To test the impact of a novel educational booklet on patients' beliefs about back pain and functional outcome.

SUMMARY OF BACKGROUND DATA

The information and advice that health professionals give to patients may be important in health care intervention, but there is little scientific evidence of their effectiveness. A novel patient educational booklet, The Back Book, has been developed to provide evidence-based information and advice consistent with current clinical guidelines.

METHODS

One hundred sixty-two patients were given either the experimental booklet or a traditional booklet. The main outcomes studied were fear-avoidance beliefs about physical activity, beliefs about the inevitable consequences of back trouble, the Roland Disability Questionnaire, and visual analogue pain scales. Postal follow-up response at 1 year after initial treatment was 78%.

RESULTS

Patients receiving the experimental booklet showed a statistically significant greater early improvement in beliefs which was maintained at 1 year. A greater proportion of patients with an initially high fear-avoidance beliefs score who received the experimental booklet had clinically important improvement in fear-avoidance beliefs about physical activity at 2 weeks, followed by a clinically important improvement in the Roland Disability Questionnaire score at 3 months. There was no effect on pain.

CONCLUSION

This trial shows that carefully selected and presented information and advice about back pain can have a positive effect on patients' beliefs and clinical outcomes, and suggests that a study of clinically important effects in individual patients may provide further insights into the management of low back pain.