Simple sample size calculation for cluster-randomized trials

Inequalities in maternity care and newborn outcomes: one-year surveillance of births in vulnerable slum communities in Mumbai

Background

Aggregate urban health statistics mask inequalities. We described maternity care in vulnerable slum communities in Mumbai, and examined differences in care and outcomes between more and less deprived groups.

Methods

We collected information through a birth surveillance system covering a population of over 280 000 in 48 vulnerable slum localities. Resident women identified births in their own localities and mothers and families were interviewed at 6 weeks after delivery. We analysed data on 5687 births over one year to September 2006. Socioeconomic status was classified using quartiles of standardized asset scores.

Results

Women in higher socioeconomic quartile groups were less likely to have married and conceived in their teens (Odds ratio 0.74, 95% confidence interval 0.69–0.79, and 0.82, 0.78–0.87, respectively). There was a socioeconomic gradient away from public sector maternity care with increasing socioeconomic status (0.75, 0.70–0.79 for antenatal care and 0.66, 0.61–0.71 for institutional delivery). Women in the least poor group were five times less likely to deliver at home (0.17, 0.10–0.27) as women in the poorest group and about four times less likely to deliver in the public sector (0.27, 0.21–0.35). Rising socioeconomic status was associated with a lower prevalence of low birth weight (0.91, 0.85–0.97). Stillbirth rates did not vary, but neonatal mortality rates fell non-significantly as socioeconomic status increased (0.88, 0.71–1.08).

Conclusion

Analyses of this type have usually been applied across the population spectrum from richest to poorest, and we were struck by the regularly stepped picture of inequalities within the urban poor, a group that might inadvertently be considered relatively homogeneous. The poorest slum residents are more dependent upon public sector health care, but the regular progression towards the private sector raises questions about its quality and regulation. It also underlines the need for healthcare provision strategies to take account of both sectors.

Home management of malaria with artemether-lumefantrine compared with standard care in urban Ugandan children: a randomised controlled trial

BACKGROUND

Home management of malaria-the presumptive treatment of febrile children with antimalarial drugs-is advocated to ensure prompt effective treatment of the disease. We assessed the effect of home delivery of artemether-lumefantrine on the incidence of antimalarial treatment and on clinical outcomes in children from an urban setting with fairly low malaria transmission.

METHODS

In Kampala, Uganda, 437 children aged between 1 and 6 years from 325 households were randomly assigned by a computer-generated sequence to receive home delivery of prepackaged artemether-lumefantrine for presumptive treatment of febrile illnesses (n=225) or current standard of care (n=212). Randomisation was done by household after a pilot period of 1 month. After randomisation, study participants were followed up for an additional 12 months and information on their health and treatment of illnesses was obtained by use of monthly questionnaires and household diaries, which were completed by the participants' carers. The primary outcome was treatment incidence density per person-year. Analysis of the primary outcome was done on the modified intention-to-treat population, which included all participants apart from those excluded before data collection. This trial is registered with ClinicalTrials.gov, number NCT00115921.

FINDINGS

Eight participants in the home management group and four in the standard care group were excluded before data collection; therefore, the primary analysis was done in 217 and 208 participants, respectively. The home management group received nearly twice the number of antimalarial treatments as the standard care group (4.66 per person-year vs 2.53 per person-year; incidence rate ratio [IRR] 1.72, 95% CI 1.43-2.06, p<0.0001), and nearly five times the number given to children with microscopically confirmed malaria in a comparable cohort of children (4.66 per person-year vs 1.03 per person-year, IRR 5.19, 95% CI 4.24-6.35, p<0.0001). Clinical data were available for 189 children in the home management group and 176 in the control group at study end; the main reasons for exclusion were movement out of the study area or loss to follow-up. The proportion of participants with parasitaemia at final assessment in the intervention group was lower than in the control group (four [2%] vs 17 [10%], p=0.006), but there were no other differences in standard malariometric indices, including anaemia. Serious adverse events were captured retrospectively. One child died in each group (home management-severe pneumonia and possible septicaemia; standard care-presumed respiratory failure).

INTERPRETATION

Although home management of malaria led to prompt treatment of fever, there was little effect on clinical outcomes. The substantial over-treatment suggests that artemether-lumefantrine provided in the home might not be appropriate for large urban areas or settings with fairly low malaria transmission.

FUNDING

Gates Malaria Partnership.

HPV screening for cervical cancer in rural India

BACKGROUND

In October 1999, we began to measure the effect of a single round of screening by testing for human papillomavirus (HPV), cytologic testing, or visual inspection of the cervix with acetic acid (VIA) on the incidence of cervical cancer and the associated rates of death in the Osmanabad district in India.

METHODS

In this cluster-randomized trial, 52 clusters of villages, with a total of 131,746 healthy women between the ages of 30 and 59 years, were randomly assigned to four groups of 13 clusters each. The groups were randomly assigned to undergo screening by HPV testing (34,126 women), cytologic testing (32,058), or VIA (34,074) or to receive standard care (31,488, control group). Women who had positive results on screening underwent colposcopy and directed biopsies, and those with cervical precancerous lesions or cancer received appropriate treatment.

RESULTS

In the HPV-testing group, cervical cancer was diagnosed in 127 subjects (of whom 39 had stage II or higher), as compared with 118 subjects (of whom 82 had advanced disease) in the control group (hazard ratio for the detection of advanced cancer in the HPV-testing group, 0.47; 95% confidence interval [CI], 0.32 to 0.69). There were 34 deaths from cancer in the HPV-testing group, as compared with 64 in the control group (hazard ratio, 0.52; 95% CI, 0.33 to 0.83). No significant reductions in the numbers of advanced cancers or deaths were observed in the cytologic-testing group or in the VIA group, as compared with the control group. Mild adverse events were reported in 0.1% of screened women.

CONCLUSIONS

In a low-resource setting, a single round of HPV testing was associated with a significant reduction in the numbers of advanced cervical cancers and deaths from cervical cancer.

Dying to count: mortality surveillance in resource-poor settings

Reliable cause-specific mortality data constitute a crucial resource for health monitoring, service planning and prioritisation. However, in the majority of the world's poorest settings, systematic health and vital event surveillance systems are weak or non-existent. As such, deaths are not counted and causes of death remain unregistered for more than two-thirds of the world's population.For researchers, health workers and policy makers in resource-poor settings, therefore, attempts to measure mortality have to be implemented from first principles. As a result, there is wide variation in mortality surveillance methodologies in different settings, and lack of standardisation and rigorous validation of these methods hinder meaningful comparison of mortality data between settings and over time.With a particular focus on Health and Demographic Surveillance Systems (HDSSs), this paper summarises recent research and conceptual development of certain methodological aspects of mortality surveillance stemming from a series of empirical investigations. The paper describes the advantages and limitations of various methods in particular contexts, and argues that there is no single methodology to satisfy all data needs. Rather, methodological decisions about mortality measurement should be a synthesis of all available knowledge relating to clearly defined concepts of why data are being collected, how they can be used and when they are of good enough quality to inform public health action.

An innovative telemedicine knowledge translation program to improve quality of care in intensive care units: protocol for a cluster randomized pragmatic trial

BACKGROUND

There are challenges to timely adoption of, and ongoing adherence to, evidence-based practices known to improve patient care in the intensive care unit (ICU). Quality improvement initiatives using a collaborative network approach may increase the use of such practices. Our objective is to evaluate the effectiveness of a novel knowledge translation program for increasing the proportion of patients who appropriately receive the following six evidence-based care practices: venous thromboembolism prophylaxis; ventilator-associated pneumonia prevention; spontaneous breathing trials; catheter-related bloodstream infection prevention; decubitus ulcer prevention; and early enteral nutrition.

METHODS AND DESIGN

We will conduct a pragmatic cluster randomized active control trial in 15 community ICUs and one academic ICU in Ontario, Canada. The intervention is a multifaceted videoconferenced educational and problem-solving forum to organize knowledge translation strategies, including comparative audit and feedback, educational sessions from content experts, and dissemination of algorithms. Fifteen individual ICUs (clusters) will be randomized to receive quality improvement interventions targeting one of the best practices during each of six study phases. Each phase lasts four months during the first study year and three months during the second. At the end of each study phase, ICUs are assigned to an intervention for a best practice not yet received according to a random schedule. The primary analysis will use patient-level process-of-care data to measure the intervention's effect on rates of adoption and adherence of each best practice in the targeted ICU clusters versus controls.

DISCUSSION

This study design evaluates a new system for knowledge translation and quality improvement across six common ICU problems. All participating ICUs receive quality improvement initiatives during every study phase, improving buy-in. This study design could be considered for other quality improvement interventions and in other care settings.

Designing phase III or IV trials for vaccines: choosing between individual or cluster randomised trial designs

The choice of design between individual randomisation, cluster or pseudo-cluster randomisation is often made difficult. Clear methodological guidelines have been given for trials in general practice, but not for vaccine trials. This article proposes a decisional flow-chart to choose the most adapted design for evaluating the effectiveness of a vaccine in large-scale studies. Six criteria have been identified: importance of herd immunity or herd protection, ability to delimit epidemiological units, homogeneity of transmission probability across sub-populations, population's acceptability of randomisation, availability of logistical resources, and estimated sample size. This easy to use decisional method could help sponsors, trial steering committees and ethical committees adopt the most suitable design.

Rabies control in rural Africa: evaluating strategies for effective domestic dog vaccination

Effective vaccination campaigns need to reach a sufficient percentage of the population to eliminate disease and prevent future outbreaks, which for rabies is predicted to be 70%, at a cost that is economically and logistically sustainable. Domestic dog rabies has been increasing across most of sub-Saharan Africa indicating that dog vaccination programmes to date have been inadequate. We compare the effectiveness of a variety of dog vaccination strategies in terms of their cost and coverage in different community settings in rural Tanzania. Central-point (CP) vaccination was extremely effective in agro-pastoralist communities achieving a high coverage (>80%) at a low cost (US$5/dog) and inadequate (<20% coverage); combined approaches using CP and either house-to-house vaccination or trained community-based animal health workers were most effective with coverage exceeding 70%, although costs were still high (>US$6 and >US$4/dog, respectively). No single vaccination strategy is likely to be effective in all populations and therefore alternative approaches must be deployed under different settings. CP vaccination is cost-effective and efficient for the majority of dog populations in rural Tanzania and potentially elsewhere in sub-Saharan Africa, whereas a combination strategy is necessary in remote pastoralist communities. These results suggest that rabies control is logistically feasible across most of the developing world and that the annual costs of effective vaccination campaigns in Tanzania are likely to be affordable.

Restricted randomization of ZAMSTAR: a 2 x 2 factorial cluster randomized trial

BACKGROUND

A small number of clusters and substantial variation between clusters increase the chance of unbalanced randomization in cluster randomized trials. Baseline imbalances between groups may distort intervention effects. When adjusting for imbalances in the cluster-level analysis, this results in loss of degrees of freedom. Variance reduction that can be achieved through stratification and blocking is limited. Restricted randomization is an alternative approach that ensures balanced allocation.

PURPOSE

We present the randomization scheme used in the ZAMSTAR trial of tuberculosis control interventions in Southern Africa.

METHODS

We used stratification and restriction to randomize 24 clusters (16 Zambian, 8 South African) into four intervention groups in a 2 x 2 factorial design. Stratification was by country and tuberculous infection prevalence and restriction by tuberculous infection prevalence, HIV prevalence, urban/rural, social context, and geographical location. Balance was defined in terms of covariate-specific tolerance thresholds for the measure of imbalance. For binary (0/1) covariates we defined imbalance = max(S(i)) - min(S(i)), where, S(i) was the number of 1s in group i = 1,2,3,4. For continuous covariates we defined imbalance = (max(M(i)) - min(M(i)))/ min(M(i) ), where, M( i) was the average in group i = 1,2,3,4.We used simulation to estimate the restriction factor (proportion of unacceptable allocations) both for individual covariates and overall. Simulation was also used to investigate the validity of the restricted randomization design, with the use of the validity matrix, by monitoring the probability that any given pair of clusters is allocated to the same intervention group.

RESULTS

There were 3 657 930 400 possible ways of allocating the 24 clusters to the four groups after stratification. With a combined restriction factor of 0.998 this still left 7 million acceptable allocations. The final allocation was selected at a public ceremony from a randomly-generated list of acceptable allocations. The design of the allocation process was observed to be valid.

LIMITATIONS

The restricted randomization scheme significantly decreased the total number of available allocations of clusters into intervention groups.

CONCLUSION

Our restricted randomization was successful in that it achieved good balance while preserving the impartiality and validity of the trial.

ZAMSTAR, The Zambia South Africa TB and HIV Reduction Study: design of a 2 x 2 factorial community randomized trial

BACKGROUND

TB and HIV form a deadly synergy in much of the developing world, especially Africa. Interventions to reduce the impact of these diseases at community level are urgently needed. This paper presents the design of a community randomised trial to evaluate the impact of two complex interventions on the prevalence of tuberculosis (TB) in high HIV prevalence settings in Zambia and South Africa.

METHODS

The interaction between TB and HIV is reviewed and possible interventions that could reduce the prevalence of TB in HIV-endemic populations are discussed. Two of these interventions are described in detail and the design of a 2 x 2 factorial community randomised trial to test these interventions is presented. The limitations and challenges of the design are identified and discussed.

CONCLUSION

There is an urgent need to reduce the prevalence of TB in communities highly affected by HIV. Potential interventions are complex and require innovative trial designs to provide the rigorous evidence needed to inform health policy makers and to ensure that resources are used optimally.

TRIAL REGISTRATION

Number: ISRCTN36729271.

Micronutrient supplementation has limited effects on intestinal infectious disease and mortality in a Zambian population of mixed HIV status: a cluster randomized trial

BACKGROUND

Diarrheal disease remains a major contributor to morbidity and mortality in Africa, but host defense against intestinal infection is poorly understood and may depend on nutritional status.

OBJECTIVE

To test the hypothesis that defense against intestinal infection depends on micronutrient status, we undertook a randomized controlled trial of multiple micronutrient supplementation in a population where there is borderline micronutrient deficiency.

DESIGN

All consenting adults (> or =18 y) living in a carefully defined sector of Misisi, Lusaka, Zambia, were included in a cluster-randomized (by household), double-blind, placebo-controlled trial with a midpoint crossover. There were no exclusion criteria. Participants were given a daily tablet containing 15 micronutrients at just above the recommended nutrient intake or placebo. The primary endpoint was the incidence of diarrhea; secondary endpoints were severe episodes of diarrhea, respiratory infection, nutritional status, CD4 count, and mortality.

RESULTS

Five hundred participants were recruited and followed up for 3.3 y (10,846 person-months). The primary endpoint, incidence of diarrhea (1.4 episodes/y per person), did not differ with treatment allocation. However, severe episodes of diarrhea were reduced in the supplementation group (odds ratio: 0.50; 95% CI: 0.26, 0.92; P = 0.017). Mortality was reduced in HIV-positive participants from 12 with placebo to 4 with supplementation (P = 0.029 by log-rank test), but this was not due to changes in CD4 count or nutritional status.

CONCLUSION

Micronutrient supplementation with this formulation resulted in only modest reductions in severe diarrhea and reduced mortality in HIV-positive participants. The trial was registered as ISRCTN31173864.

Integrated audit as a means to implement unit protocols: a randomized and controlled study

RATIONALE

Great emphasis is placed on the need to base assistance activities on efficacy evidence. Nevertheless, it is demonstrated that the mere knowledge of scientific recommendations is not sufficient to determine changes in practice. Many elements can influence the operators' behaviour, so that comprehensive strategies are needed to change wrong or not evidence-based practices.

OBJECTIVES

The aim was to compare the efficacy of two protocol implementation strategies: clinical-organizational integrated audits with feedback and presence of facilitators in the departments versus standard observation.

METHODS

Cluster-randomized, controlled and open trial methods were used. Implementation was investigated for protocols regarding the prevention of pressure lesions and the management of peripheral and central venous catheters. For both protocols, a checklist with the indicators needed for the evaluation was created. The study was divided into five time points. For the initial and final investigation, evaluators filled-in the checklists for both study groups; for the three intermediate investigations, the checklists were self-filled by the operators of the control group, while the operators of the experimental group received the intervention.

RESULTS

For almost all indicators, the data show an increment in the adoption of correct practices. Nevertheless, the last investigation shows a significant (P < 0.05) difference between the two groups in favour of the experimental group: 12 process indicators versus one for catheter management protocol and seven versus one for protocol on pressure lesions.

CONCLUSIONS

The main strong point of the study lies in carrying out an intervention aimed at improving the operators' adoption of two protocols as a whole, rather than of single procedures. The audit intervention allowed to highlight some more subjective criticalities important in determining the success or failure of the implementation of effective practices.

Effect of community-based behaviour change management on neonatal mortality in Shivgarh, Uttar Pradesh, India: a cluster-randomised controlled trial

BACKGROUND

In rural India, most births take place in the home, where high-risk care practices are common. We developed an intervention of behaviour change management, with a focus on prevention of hypothermia, aimed at modifying practices and reducing neonatal mortality.

METHODS

We did a cluster-randomised controlled efficacy trial in Shivgarh, a rural area in Uttar Pradesh. 39 village administrative units (population 104,123) were allocated to one of three groups: a control group, which received the usual services of governmental and non-governmental organisations in the area; an intervention group, which received a preventive package of interventions for essential newborn care (birth preparedness, clean delivery and cord care, thermal care [including skin-to-skin care], breastfeeding promotion, and danger sign recognition); or another intervention group, which received the package of essential newborn care plus use of a liquid crystal hypothermia indicator (ThermoSpot). In the intervention clusters, community health workers delivered the packages via collective meetings and two antenatal and two postnatal household visitations. Outcome measures included changes in newborn-care practices and neonatal mortality rate compared with the control group. Analysis was by intention to treat. This study is registered as International Standard Randomised Control Trial, number NCT00198653.

FINDINGS

Improvements in birth preparedness, hygienic delivery, thermal care (including skin-to-skin care), umbilical cord care, skin care, and breastfeeding were seen in intervention arms. There was little change in care-seeking. Compared with controls, neonatal mortality rate was reduced by 54% in the essential newborn-care intervention (rate ratio 0.46 [95% CI 0.35-0.60], p<0.0001) and by 52% in the essential newborn care plus ThermoSpot arm (0.48 [95% CI 0.35-0.66], p<0.0001).

INTERPRETATION

A socioculturally contextualised, community-based intervention, targeted at high-risk newborn-care practices, can lead to substantial behavioural modification and reduction in neonatal mortality. This approach can be applied to behaviour change along the continuum of care, harmonise vertical interventions, and build community capacity for sustained development.

FUNDING

USAID and Save the Children-US through a grant from the Bill & Melinda Gates Foundation.

A provider intervention to improve colorectal cancer screening in county health centers

BACKGROUND

Publicly-funded health centers serve disadvantaged populations who underuse colorectal cancer screening (CRC). Because physicians play a key role in patient adherence to screening, provider interventions within health center practices could improve the delivery/utilization of CRC screening.

METHODS

A 2-group study design was used with 4 pairs of health centers randomized to the intervention or control condition. The provider intervention featured academic detailing of the small practice groups, followed by a strategic planning session with the entire health center staff using SWOT analysis. The outcome measure of provider endoscopy referral/fecal occult blood test dispensing and/or completion of CRC screening was determined by medical record audit (n = 2224). The intervention effect was evaluated using generalized estimating equations. Pre-post intervention patient surveys (n = 281) were conducted.

RESULTS

Chart audits of the 1 year period before and after the intervention revealed a 16% increase from baseline in CRC screening referral/dispensing/completion among intervention centers, compared with a 4% increase among controls, odds ratio (OR) = 2.25 (1.67-3.04) P < 0.001. Intervention versus control health center patient self-reports of lack of physician recommendation as a reason for not having CRC screening declined from baseline to follow-up (P = 0.04).

CONCLUSIONS

Provider referrals/dispensing/completion of CRC screening within health centers was significantly improved and barriers reduced through a provider intervention combining continuing medical education with a team building strategic planning exercise.

Implementing community-based perinatal care: results from a pilot study in rural Pakistan

OBJECTIVE

This pilot study investigated the feasibility of delivering a package of community-based interventions for improving perinatal care using lady health workers (LHWs) and traditional birth attendants (Dais) in rural Pakistan.

METHODS

The intervention was implemented in four of eight village clusters (315 villages, total population 138,600), while four served as a comparison group. The LHWs in intervention clusters received additional training focused on essential maternal and newborn care, conducted community education group sessions, and were encouraged to link up with local Dais. The intervention was delivered within the regular government LHW programme and was supported by the creation of voluntary community health committees.

FINDINGS

In intervention villages, there were significant reductions from baseline in stillbirth (from 65.9 to 43.1 per 1000 births, P < 0.001) and neonatal mortality rates (from 57.3 to 41.3 per 1000 live births, P < 0.001). The proportion of deliveries conducted by skilled attendants at public sector facilities also increased, from 18% at baseline to 30%, while the proportion of home births decreased from 79% to 65%. A household survey indicated a higher frequency of key behaviours (e.g. early and exclusive breastfeeding, delayed bathing and cord care) in intervention villages.

CONCLUSION

The improved stillbirth and neonatal mortality rates observed indicate that community health workers (i.e. LHWs and Dais) can be effective in implementing a community and outreach package that leads to improved home care practices by families, increased care-seeking behaviour and greater utilization of skilled care providers. These preliminary observations require confirmation in an adequately powered trial.

Impact of stepping stones on incidence of HIV and HSV-2 and sexual behaviour in rural South Africa: cluster randomised controlled trial

OBJECTIVE

To assess the impact of Stepping Stones, a HIV prevention programme, on incidence of HIV and herpes simplex type 2 (HSV-2) and sexual behaviour.

DESIGN

Cluster randomised controlled trial.

SETTING

70 villages (clusters) in the Eastern Cape province of South Africa.

PARTICIPANTS

1360 men and 1416 women aged 15-26 years, who were mostly attending schools.

INTERVENTION

Stepping Stones, a 50 hour programme, aims to improve sexual health by using participatory learning approaches to build knowledge, risk awareness, and communication skills and to stimulate critical reflection. Villages were randomised to receive either this or a three hour intervention on HIV and safer sex. Interviewers administered questionnaires at baseline and 12 and 24 months and blood was tested for HIV and HSV-2.

PRIMARY OUTCOME MEASURE

incidence of HIV. Other outcomes: incidence of HSV-2, unwanted pregnancy, reported sexual practices, depression, and substance misuse.

RESULTS

There was no evidence that Stepping Stones lowered the incidence of HIV (adjusted incidence rate ratio 0.95, 95% confidence interval 0.67 to 1.35). The programme was associated with a reduction of about 33% in the incidence of HSV-2 (0.67, 0.46 to 0.97; P=0.036)-that is, Stepping Stones reduced the number of new HSV-2 infections over a two year period by 34.9 (1.6 to 68.2) per 1000 people exposed. Stepping Stones significantly improved a number of reported risk behaviours in men, with a lower proportion of men reporting perpetration of intimate partner violence across two years of follow-up and less transactional sex and problem drinking at 12 months. In women desired behaviour changes were not reported and those in the Stepping Stones programme reported more transactional sex at 12 months.

CONCLUSION

Stepping Stones did not reduce incidence of HIV but had an impact on several risk factors for HIV-notably, HSV-2 and perpetration of intimate partner violence.

TRIAL REGISTRATION

Clinical Trials NCT00332878.

A simulation based technique to estimate intracluster correlation for a binary variable

Cluster randomized trials have become the design of choice for evaluating the effect of selected interventions on well-known health indicators such as neonatal mortality rate, episiotomy rate, and postpartum hemorrhage rate in a community setting. Determining the sample size of a cluster randomized trial requires a reliable estimate of cluster size and the intracluster correlation (ICC), because sample size can be substantially impacted by these parameters. During the design phase of a trial, the investigators may have estimates of the valid range of the health indicator which is the primary outcome variable. Furthermore, investigators often have an estimate of the average cluster size or range of cluster sizes that exist among the proposed samples they are planning to include in the trial. We present in this article a simulation technique to estimate the ICC value and its distribution for known binary outcome variables and a varying number of clusters and cluster sizes. We applied this technique to estimate ICC values and confidence intervals for a multi-country trial assessing the effect of neonatal resuscitation to decrease seven-day neonatal mortality, where communities within a country were clusters. This simulation technique can be used to estimate the possible ranges of the ICC values and to help to design an appropriately powered trial.

Effect of intermittent preventive treatment of malaria on health and education in schoolchildren: a cluster-randomised, double-blind, placebo-controlled trial

BACKGROUND

Malaria is a major cause of morbidity and mortality in early childhood, yet its consequences for health and education during the school-age years remain poorly understood. We examined the effect of intermittent preventive treatment (IPT) in reducing anaemia and improving classroom attention and educational achievement in semi-immune schoolchildren in an area of high perennial transmission.

METHODS

A stratified, cluster-randomised, double-blind, placebo-controlled trial of IPT was done in 30 primary schools in western Kenya. Schools were randomly assigned to treatment (sulfadoxine-pyrimethamine in combination with amodiaquine or dual placebo) by use of a computer-generated list. Children aged 5-18 years received three treatments at 4-month intervals (IPT n=3535, placebo n=3223). The primary endpoint was the prevalence of anaemia, defined as a haemoglobin concentration below 110 g/L. This outcome was assessed through cross-sectional surveys 12 months post-intervention. Analysis was by both intention to treat, excluding children with missing data, and per protocol. This study is registered with ClinicalTrials.gov, number NCT00142246.

FINDINGS

2604 children in the IPT group and 2302 in the placebo group were included in the intention-to-treat analysis of the primary outcome; the main reason for exclusion was loss to follow-up. Prevalence of anaemia at 12 months averaged 6.3% in the IPT group and 12.6% in the placebo group (adjusted risk ratio 0.52, 95% CI 0.29-0.93; p=0.028). Significant improvements were also seen in two of the class-based tests of sustained attention, with a mean increase in code transmission test score of 6.05 (95% CI 2.83-9.27; p=0.0007) and counting sounds test score of 1.80 (0.19-3.41; p=0.03), compared with controls. No effect was shown for inattentive or hyperactive-compulsive behaviours or on educational achievement. The per-protocol analysis yielded similar results. 23 serious adverse events were reported within 28 days of any treatment (19 in the IPT group and four in the placebo group); the main side-effects were problems of balance, dizziness, feeling faint, nausea, and/or vomiting shortly after treatment.

INTERPRETATION

IPT of malaria improves the health and cognitive ability of semi-immune schoolchildren. Effective malaria interventions could be a valuable addition to school health programmes.

A cluster-randomized trial of provider-initiated (opt-out) HIV counseling and testing of tuberculosis patients in South Africa

OBJECTIVE

To determine whether implementation of provider-initiated human immunodeficiency virus (HIV) counseling would increase the proportion of tuberculosis (TB) patients who received HIV counseling and testing.

DESIGN

Cluster-randomized trial with clinic as the unit of randomization.

SETTING

Twenty, medium-sized primary care TB clinics in the Nelson Mandela Metropolitan Municipality, Port Elizabeth, Eastern Cape Province, South Africa.

SUBJECTS

A total of 754 adults (18 years and older) newly registered as TB patients in the 20 study clinics.

INTERVENTION

Implementation of provider-initiated HIV counseling and testing.

MAIN OUTCOME MEASURES

Percentage of TB patients HIV counseled and tested. SECONDARY: Percentage of patients with HIV test positive, and percentage of those who received cotrimoxazole and who were referred for HIV care.

RESULTS

: A total of 754 adults newly registered as TB patients were enrolled. In clinics randomly assigned to implement provider-initiated HIV counseling and testing, 20.7% (73/352) patients were counseled versus 7.7% (31/402) in the control clinics (P = 0.011), and 20.2% (n = 71) versus 6.5% (n = 26) underwent HIV testing (P = 0.009). Of those patients counseled, 97% in the intervention clinics accepted testing versus 79% in control clinics (P = 0.12). The proportion of patients identified as HIV infected in intervention clinics was 8.5% versus 2.5% in control clinics (P = 0.044). Fewer than 40% of patients with a positive HIV test were prescribed cotrimoxazole or referred for HIV care in either study arm.

CONCLUSIONS

Provider-initiated HIV counseling significantly increased the proportion of adult TB patients who received HIV counseling and testing, but the magnitude of the effect was small. Additional interventions to optimize HIV testing for TB patients urgently need to be evaluated.

Nurse-coordinated multidisciplinary, family-based cardiovascular disease prevention programme (EUROACTION) for patients with coronary heart disease and asymptomatic individuals at high risk of cardiovascular disease: a paired, cluster-randomised controlled trial

BACKGROUND

Our aim was to investigate whether a nurse-coordinated multidisciplinary, family-based preventive cardiology programme could improve standards of preventive care in routine clinical practice.

METHODS

In a matched, cluster-randomised, controlled trial in eight European countries, six pairs of hospitals and six pairs of general practices were assigned to an intervention programme (INT) or usual care (UC) for patients with coronary heart disease or those at high risk of developing cardiovascular disease. The primary endpoints-measured at 1 year-were family-based lifestyle change; management of blood pressure, lipids, and blood glucose to target concentrations; and prescription of cardioprotective drugs. Analysis was by intention to treat. The trial is registered as ISRCTN 71715857.

FINDINGS

1589 and 1499 patients with coronary heart disease in hospitals and 1189 and 1128 at high risk were assigned to INT and UC, respectively. In patients with coronary heart disease who smoked in the month before the event, 136 (58%) in the INT and 154 (47%) in the UC groups did not smoke 1 year afterwards (difference in change 10.4%, 95% CI -0.3 to 21.2, p=0.06). Reduced consumption of saturated fat (196 [55%] vs 168 [40%]; 17.3%, 6.4 to 28.2, p=0.009), and increased consumption of fruit and vegetables (680 [72%] vs 349 [35%]; 37.3%, 18.1 to 56.5, p=0.004), and oily fish (156 [17%] vs 81 [8%]; 8.9%, 0.3 to 17.5, p=0.04) at 1 year were greatest in the INT group. High-risk individuals and partners showed changes only for fruit and vegetables (p=0.005). Blood-pressure target of less than 140/90 mm Hg was attained by both coronary (615 [65%] vs 547 [55%]; 10.4%, 0.6 to 20.2, p=0.04) and high-risk (586 [58%] vs 407 [41%]; 16.9%, 2.0 to 31.8, p=0.03) patients in the INT groups. Achievement of total cholesterol of less than 5 mmol/L did not differ between groups, but in high-risk patients the difference in change from baseline to 1 year was 12.7% (2.4 to 23.0, p=0.02) in favour of INT. In the hospital group, prescriptions for statins were higher in the INT group (810 [86%] vs 794 [80%]; 6.0%, -0.5 to 11.5, p=0.04). In general practices in the intervention groups, angiotensin-converting enzyme inhibitors (297 [29%] INT vs 196 [20%] UC; 8.5%, 1.8 to 15.2, p=0.02) and statins (381 [37%] INT vs 232 [22%] UC; 14.6%, 2.5 to 26.7, p=0.03) were more frequently prescribed.

INTERPRETATION

To achieve the potential for cardiovascular prevention, we need local preventive cardiology programmes adapted to individual countries, which are accessible by all hospitals and general practices caring for coronary and high-risk patients.

Relationships between anaemia and parasitic infections in Kenyan schoolchildren: a Bayesian hierarchical modelling approach

Anaemia is multi-factorial in origin and disentangling its aetiology remains problematic, with surprisingly few studies investigating the relative contribution of different parasitic infections to anaemia amongst schoolchildren. We report cross-sectional data on haemoglobin, malaria parasitaemia, helminth infection and undernutrition among 1523 schoolchildren enrolled in classes 5 and 6 (aged 10–21 years) in 30 primary schools in western Kenya. Bayesian hierarchical modelling was used to investigate putative relationships. Children infected with Plasmodium falciparum or with a heavy Schistosoma mansoni infection, stunted children and girls were found to have lower haemoglobin concentrations. Children heavily infected with S. mansoni were also more likely to be anaemic compared with uninfected children. This study further highlights the importance of malaria and intestinal schistosomiasis as contributors to reduced haemoglobin levels among schoolchildren and helps guide the implementation of integrated school health programmes in areas of differing parasite transmission.

Cardiovascular Health Awareness Program (CHAP): a community cluster-randomised trial among elderly Canadians

OBJECTIVE

High blood pressure is an important and modifiable cardiovascular disease risk factor that remains under-detected and under-treated. Community-level interventions that address high blood pressure and other modifiable risk factors are a promising strategy to improve cardiovascular health in populations. The present study is a community cluster-randomised trial testing the effectiveness of CHAP (Cardiovascular Health Awareness Program) on the cardiovascular health of older adults.

METHODS

Thirty-nine mid-sized communities in Ontario, Canada were stratified by geographic location and size of the population aged >or=65 years and randomly allocated to receive CHAP or no intervention. In CHAP communities, residents aged >or=65 years were invited to attend cardiovascular risk assessment sessions held in pharmacies over 10 weeks in Fall, 2006. Sessions included blood pressure measurement and feedback to family physicians. Trained volunteers delivered the program with support from pharmacists, community nurses and local organisations.

RESULTS

The primary outcome measure is the relative change in the mean annual rate of hospital admission for acute myocardial infarction, congestive heart failure and stroke (composite end-point) among residents aged >or=65 years in intervention and control communities, using routinely collected, population-based administrative health data.

CONCLUSION

This paper highlights considerations in design, implementation and evaluation of a large-scale, community-wide cardiovascular health promotion initiative.

A simple approach to test for interaction between intervention and an individual-level variable in community randomized trials

OBJECTIVE

To develop a simple and robust approach for the test of interaction between community intervention and an individual-level variable suitable for use in typical situations of community randomized trials (CRTs), i.e. small number of communities but large number of subjects per community.

METHODS

We propose a method based on taking the difference between summary statistics from groups of individuals with and without an attribute within each community, then applying a two-sample t-test or Wilcoxon test to compare the distribution of within-community differences between trial arms. The method is evaluated using simulations and illustrated using data from a CRT of a health education intervention. Approximate sample size formulas are derived.

RESULTS

Analyses based on the t-test give power very close to expected level in a variety of situations, including when the summary statistics are not symmetrically distributed across communities, the covariate is not distributed as planned, and the number of communities per intervention arm ranges from 8 to 20. Even in the situation with as few as four communities per arm, the power is only slightly lower than expected. Type I error rates always closely follow 5% as required, whether the distributional assumption is correct or not. The application of the Wilcoxon test appears too conservative.

CONCLUSIONS

The proposed approach to test for interaction is valid and easy to use. The application of the t-test in this setting is robust.

Effect of newborn resuscitation training on health worker practices in Pumwani Hospital, Kenya

BACKGROUND

Birth asphyxia kills 0.7 to 1.6 million newborns a year globally with 99% of deaths in developing countries. Effective newborn resuscitation could reduce this burden of disease but the training of health-care providers in low income settings is often outdated. Our aim was to determine if a simple one day newborn resuscitation training (NRT) alters health worker resuscitation practices in a public hospital setting in Kenya.

METHODS/PRINCIPAL FINDINGS

We conducted a randomised, controlled trial with health workers receiving early training with NRT (n = 28) or late training (the control group, n = 55). The training was adapted locally from the approach of the UK Resuscitation Council. The primary outcome was the proportion of appropriate initial resuscitation steps with the frequency of inappropriate practices as a secondary outcome. Data were collected on 97 and 115 resuscitation episodes over 7 weeks after early training in the intervention and control groups respectively. Trained providers demonstrated a higher proportion of adequate initial resuscitation steps compared to the control group (trained 66% vs control 27%; risk ratio 2.45, [95% CI 1.75-3.42], p<0.001, adjusted for clustering). In addition, there was a statistically significant reduction in the frequency of inappropriate and potentially harmful practices per resuscitation in the trained group (trained 0.53 vs control 0.92; mean difference 0.40, [95% CI 0.13-0.66], p = 0.004).

CONCLUSIONS/SIGNIFICANCE

Implementation of a simple, one day newborn resuscitation training can be followed immediately by significant improvement in health workers' practices. However, evidence of the effects on long term performance or clinical outcomes can only be established by larger cluster randomised trials.

TRIAL REGISTRATION

Controlled-Trials.com ISRCTN92218092.

Cluster-randomised controlled trial of community mobilisation in Mumbai slums to improve care during pregnancy, delivery, postpartum and for the newborn

BACKGROUND

The United Nations Millennium Development Goals look to substantial improvements in child and maternal survival. Morbidity and mortality during pregnancy, delivery and the postnatal period are prime obstacles to achieving these goals. Given the increasing importance of urban health to global prospects, Mumbai's City Initiative for Newborn Health aims to improve maternal and neonatal health in vulnerable urban slum communities, through a combination of health service quality improvement and community participation. The protocol describes a trial of community intervention aimed at improving prevention, care seeking and outcomes.

OBJECTIVE

To test an intervention that supports local women as facilitators in mobilising communities for better health care. Community women's groups will build an understanding of their potential to improve maternal and infant health, and develop and implement strategies to do so.

DESIGN

Cluster-randomized controlled trial.

METHODS

The intervention will employ local community-based female facilitators to convene groups and help them to explore maternal and neonatal health issues. Groups will meet fortnightly through a seven-phase process of sharing experiences, discussion of the issues raised, discovery of potential community strengths, building of a vision for action, design and implementation of community strategies, and evaluation.The unit of allocation will be an urban slum cluster of 1000-1500 households. 48 clusters have been randomly selected after stratification by ward. 24 clusters have been randomly allocated to receive the community intervention. 24 clusters will act as control groups, but will benefit from health service quality improvement. Indicators of effect will be measured through a surveillance system implemented by the project. Key distal outcome indicators will be neonatal mortality and maternal and neonatal morbidity. Key proximate outcome indicators will be home care practices, uptake of antenatal, delivery and postnatal care, and care for maternal and neonatal illness. Data will be collected through a vital registration system for births and deaths in the 48 study clusters. Structured interviews with families will be conducted at about 6 weeks after index deliveries. We will also collect both quantitative and qualitative data to support a process evaluation.

TRIAL REGISTRATION

Current controlled trials ISRCTN96256793.

Improving the outcomes of primary care attenders with common mental disorders in developing countries: a cluster randomized controlled trial of a collaborative stepped care intervention in Goa, India

BACKGROUND AND OBJECTIVE

Common mental disorders (CMD) are a leading global burden of disease. Up to 30% of primary care attenders suffer from these disorders but most do not receive evidence-based drug or psychological treatments. There are no trials of interventions which attempt to integrate these treatments into routine primary care in developing countries. The aims of this trial (the MANAS Project) are to evaluate the clinical and cost-effectiveness of a collaborative stepped-care intervention for the treatment of CMD in India.

STUDY DESIGN

A cluster randomized controlled trial will be implemented in the state of Goa, on the west coast of India. Twenty-four primary care facilities, 12 from the government sector and 12 from the private sector, will be enrolled in two consecutive phases. For each sector, facilities will be randomly allocated within strata defined by urban/rural location, population size and presence of a visiting psychiatrist. Facilities will be randomly allocated to receive the collaborative stepped care intervention or the enhanced usual care control intervention. Both arms share two components of the intervention, viz., routine screening, and in the government clinics provision of antidepressants. In addition, the collaborative stepped care arm also provides a range of psychosocial treatments delivered by a specially trained Health Counselor, and supervision by a visiting Psychiatrist. A total of 3600 primary care attenders who are detected to suffer from a CMD based on a validated screening questionnaire will be recruited. The primary outcome is the proportion of subjects who recover from an ICD10 defined CMD at baseline by 6 months. Additional endpoints at 2 and 12 months will assess the speed and sustainability of achieving the primary outcomes. Other outcomes will include recovery from ICD10 defined depression and incidence of ICD-10 among individuals who were sub-threshold cases at baseline. Economic and disability outcomes will be assessed to estimate incremental cost-effectiveness ratios.

IMPLICATIONS

This will be the first trial of the effectiveness of a complex intervention aiming to integrate efficacious treatments for CMD into routine primary care in a developing country. If effective, its findings will have relevance to policy makers who wish to scale up treatments for CMD in primary care across the world, but mostly in those countries where specialist mental health services are few.

STUDY REGISTRATION

The MANAS project is registered through the National Institutes of Health sponsored clinical trials registry and has been assigned the identifier: NCT00446407.

Training farmers in rational drug-use improves their management of cattle trypanosomosis: A cluster-randomised trial in south Mali

We carried out a stratified, cluster-randomised, controlled trial in south Mali in 2004 to evaluate the impact of providing information on the diagnosis and treatment of bovine trypanosomosis by farmers. We recruited cattle farmers (444) in 46 villages and used stratified, restricted-randomisation to assign villages to either the test or control group. Farmers in the test group received an information leaflet designed to address gaps in farmer knowledge likely to lead to inadequate treatment; their knowledge was assessed before the intervention, and at 2 weeks and 5 months after the intervention. We assessed the quality of farmer treatments by measuring clinical outcomes in cattle 2 weeks after selection and treatment. As an indicator of herd health, we assessed the mean hematocrit of the village herd before, and 5 months after, the intervention. To account for clustering, we analysed data using generalised estimating equations. Improvements in farmer knowledge of trypanosomosis diagnosis and treatment at 2 weeks and 5 months in the group receiving information were 23% and 14% greater at 2 weeks and 5 months, respectively. In the test group, 84% of farmer treatments were successful, compared to 73% in the control group. Giving rational drug-use information to farmers improved their knowledge and management of trypanosomosis as well as clinical outcomes in cattle they treated and had no discernible negative impacts.

A cluster-randomised trial to compare home-based with health facility-based antiretroviral treatment in Uganda: study design and baseline findings

The scale-up of antiretroviral therapy is progressing rapidly in Africa but with a limited evidence-base. We report the baseline results from a large pragmatic cluster-randomised trial comparing different strategies of ART delivery. The trial is integrated in normal health service delivery. 1453 subjects were recruited into the study. Significantly more women (71%) than men (29%) were recruited. The WHO HIV clinical stage at presentation did not differ significantly between men and women: 58% and 53% respectively were at WHO stage III or IV (p=0.9). Median CD4 counts (IQR) x 106cells/l were 98 (28, 160) among men and 111 (36, 166) among women. Sixty-four percent of women and 61% men had plasma viral load ≥100,000 copies. Baseline characteristics did not change over time. Considerably fewer men than women presented for treatment. Both men and women presented at an advanced stage with very low median CD4 count and high plasma viral load.

Plant based insect repellent and insecticide treated bed nets to protect against malaria in areas of early evening biting vectors: double blind randomised placebo controlled clinical trial in the Bolivian Amazon

OBJECTIVE

To determine the effectiveness in reducing malaria of combining an insect repellent with insecticide treated bed nets compared with the nets alone in an area where vector mosquitoes feed in the early evening.

DESIGN

A double blind, placebo controlled cluster-randomised clinical study.

SETTING

Rural villages and peri-urban districts in the Bolivian Amazon.

PARTICIPANTS

4008 individuals in 860 households.

INTERVENTIONS

All individuals slept under treated nets; one group also used a plant based insect repellent each evening, a second group used placebo.

MAIN OUTCOME MEASURE

Episodes of Plasmodium falciparum or P vivax malaria confirmed by rapid diagnostic test or blood slide, respectively.

RESULTS

We analysed 15,174 person months at risk and found a highly significant 80% reduction in episodes of P vivax in the group that used treated nets and repellent (incidence rate ratio 0.20, 95% confidence interval 0.11 to 0.38, P<0.001). Numbers of P falciparum cases during the study were small and, after adjustment for age, an 82% protective effect was observed, although this was not significant (0.18, 0.02 to 1.40, P=0.10). Reported episodes of fever with any cause were reduced by 58% in the group that used repellent (0.42, 0.31 to 0.56, P<0.001).

CONCLUSIONS

Insect repellents can provide protection against malaria. In areas where vectors feed in the early evening, effectiveness of treated nets can be significantly increased by using repellent between dusk and bedtime. This has important implications in malaria vector control programmes outside Africa and shows that the combined use of treated nets and insect repellents, as advocated for most tourists travelling to high risk areas, is fully justified.

REGISTRATION

NCT 00144716.

Are FPCIs a source of increased risk for children? Results of a multicenter, experimental study comparing children's behaviour with FPCIs and toys

INTRODUCTION

Food Products Containing Inedibles (FPCIs) are believed to represent a source of higher choking risk in children. The aim of this study was to set up a controlled study, conducted on children aged 3-6 in a laboratory setting, in order to understand their behavior when interacting with FPCIs (with reference to mouthing activities, double nature recognition, and toy assembling ability).

METHOD

The experimental phase was divided into two sessions: a FPCI session and a Toy session, to which 247 children were randomly assigned. During these sessions children were observed in order to catch their mouthing activity according to the two types of objects available to them (FPCIs and Toys).

RESULTS

This study shows that: (a) children's behavior with respect to toys contained in FPCIs and toys presented alone is not significantly different; (b) children's ability to distinguish between the edible and non-edible part of the FPCI was very high; and (c) mouthing episodes of the inedible parts were negligible and comparable between FPCIs and toys presented alone. This strongly suggests that, with respect to choking risk, FPCIs are not per se distinguishable from toys containing small parts.

IMPACT ON INDUSTRY

Restrictions on the sale of FPCIs with small toys exist in the U.S. market. In Europe, FPCIs are allowed to be on sale, under the condition that, in case, they will follow the general regulatory requirements of small toys packaged and sold alone. In this case, they must provide age warnings and labels. Our findings do not justify the different attention that toys in FPCIs are at times afforded by regulators when compared to "stand alone" toys.

Intracluster correlation coefficient in multicenter childhood trauma studies

OBJECTIVE

To calculate the intracluster correlation coefficient (ICC) for emergency department (ED) shock rate, early trauma death (ie, death during the first 24 h after arrival at hospital), and in-hospital trauma death rate for multicenter childhood injuries.

METHODS

The National Trauma Data Bank (5th revision), the largest multicenter trauma registry in the US, was used. Data from 80 trauma centers were used to calculate the ICC for in-hospital trauma death rate. Thirty three states provided data for calculation of the ICC for ED shock and early trauma death rate.

RESULTS

From 2000 to 2004, 13% of the 952 242 patients in the National Trauma Data Bank were <15 years old. Approximately 17 000 of these children had injuries with an injury severity score >15, of whom 84% (14 095 subjects) were hospitalized at 80 level I or II trauma centers in 33 states. The ICCs for ED shock rate, early trauma death rate, and in-hospital death rate were 0.005 (95% CI 0.000 to 0.010), 0.014 (95% CI 0.004 to 0.024), and 0.023 (95% CI 0.013 to 0.033), respectively. These ICCs were calculated for boys and girls and also for blunt and penetrating injuries.

CONCLUSION

Clustered childhood trauma studies that aim to compare different aspects of pre-hospital and hospital trauma care should incorporate these ICCs for sample calculation. When cluster randomized clinical trials are mounted, if sample sizes are calculated without adjustment for ICC, then the planned trial is likely to be seriously underpowered.

Intraclass correlation coefficients from three cluster randomised controlled trials in primary and residential health care

OBJECTIVE

This paper provides intraclass correlation coefficients (ICCs) for estimation of sample size inflation required in future cluster randomised trials in primary or residential care.

METHODS

Three cluster randomised trials were conducted among middle-aged and older adults in primary care and residential care in Australia and New Zealand between 1995 and 2002. Baseline means or proportions, mean change, and ICCs with their standard errors and 95% confidence intervals are reported for outcome variables used in the three studies. The ICCs were estimated from a one-way random effects model using the analysis of variance method.

RESULTS

ICCs for quality of life and psychological variables in the primary care studies were low (below 0.018). ICCs for clinical and physical activity variables ranged from 0 to 0.08. ICCs for health and functional status in residential care for the elderly were high, ranging from 0.025 to 0.514.

CONCLUSIONS

The magnitude of the intraclass correlation varies with the venue of the trial, the outcome variables used, and the expected effect of the intervention. However, the intraclass correlations provided will be useful for more appropriate planning of residential and primary care-based trials in the future.

Sample size considerations when groups are the appropriate unit of analysis

This article discusses issues to be considered by nurse researchers when groups should be used as the unit of randomization. Advantages and disadvantages are presented, with statistical calculations needed to determine the effective sample size. Examples of these concepts are presented using data from the Black Cosmetologists Promoting Health Program. Different hypothetical scenarios and their impact on sample size are also presented. Given the complexity of calculating the sample size when using groups as the unit of randomization, it is advantageous for researchers to work closely with statisticians when designing and implementing studies that anticipate the use of groups as the unit of randomization.

Effect of visual screening on cervical cancer incidence and mortality in Tamil Nadu, India: a cluster-randomised trial

BACKGROUND

Cervical cancer is the most common cancer among women in developing countries. We assessed the effect of screening using visual inspection with 4% acetic acid (VIA) on cervical cancer incidence and mortality in a cluster randomised controlled trial in India.

METHODS

Of the 114 study clusters in Dindigul district, India, 57 were randomised to one round of VIA by trained nurses, and 57 to a control group. Healthy women aged 30 to 59 years were eligible for the study. Screen-positive women had colposcopy, directed biopsies, and, where appropriate, cryotherapy by nurses during the screening visit. Those with larger precancerous lesions or invasive cancers were referred for appropriate investigations and treatment. Cervical cancer incidence and mortality in the study groups were analysed and compared using Cox regression taking the cluster design into account, and analysis was by intention to treat. The primary outcome measures were cervical cancer incidence and mortality.

RESULTS

Of the 49,311 eligible women in the intervention group, 31,343 (63.6%) were screened during 2000-03; 30,958 control women received the standard care. Of the 3088 (9.9%) screened positive, 3052 had colposcopy, and 2539 directed biopsy. Of the 1874 women with precancerous lesions in the intervention group, 72% received treatment. In the intervention group, 274,430 person years, 167 cervical cancer cases, and 83 cervical cancer deaths were accrued compared with 178,781 person-years, 158 cases, and 92 deaths and in the control group during 2000-06 (incidence hazard ratio 0.75 [95% CI 0.55-0.95] and mortality hazard ratio 0.65 [0.47-0.89]).

INTERPRETATION

VIA screening, in the presence of good training and sustained quality assurance, is an effective method to prevent cervical cancer in developing countries.

Limitations of the randomized controlled trial in evaluating population-based health interventions

Population- and systems-based interventions need evaluation, but the randomized controlled trial (RCT) research design has significant limitations when applied to their complexity. After some years of being largely dismissed in the ranking of evidence in medicine, alternatives to the RCT have been debated recently in public health and related population and social service fields to identify the trade-offs in their use when randomization is impractical or unethical. This review summarizes recent debates and considers the pragmatic and economic issues associated with evaluating whole-population interventions while maintaining scientific validity and credibility.

A factorial-design cluster randomised controlled trial investigating the cost-effectiveness of a nutrition supplement and an exercise programme on pneumonia incidence, walking capacity and body mass index in older people living in Santiago, Chile: the CENEX study protocol

BACKGROUND

Chile is currently undergoing a period of rapid demographic transition which has led to an increase in the proportion of older people in the population; the proportion aged 60 years and over, for example, increased from 8% of the population in 1980 to 12% in 2005. In an effort to promote healthy ageing and preserve function, the government of Chile has formulated a package of actions into the Programme of Complementary Feeding for the Older Population (PACAM) which has been providing a nutritional supplement to older people since 1998. PACAM distributes micronutrient fortified foods to individuals aged 70 years and over registered at Primary Health Centres and enrolled in the programme. The recommended serving size (50 g/day) of these supplements provides 50% of daily micronutrient requirements and 20% of daily energy requirements of older people. No information is currently available on the cost-effectiveness of the supplementation programme.

AIM

The aim of the CENEX cluster randomised controlled trial is to evaluate the cost-effectiveness of an ongoing nutrition supplementation programme, and a specially designed physical exercise intervention for older people of low to medium socio-economic status living in Santiago, Chile.

METHODS

The study has been conceptualised as a public health programme effectiveness study and has been designed as a 24-month factorial cluster-randomised controlled trial conducted among 2800 individuals aged 65.0-67.9 years at baseline attending 28 health centres in Santiago. The main outcomes are incidence of pneumonia, walking capacity and change in body mass index over 24 months of intervention. Costing data (user and provider), collected at all levels, will enable the determination of the cost-effectiveness of the two interventions individually and in combination. The study is supported by the Ministry of Health in Chile, which is keen to expand and improve its national programme of nutrition for older people based on sound science-base and evidence for cost-effectiveness. :

TRIAL REGISTRATION

ISRCTN48153354.

Sizing a trial to alter the trajectory of health behaviours: methods, parameter estimates, and their application

Group-randomized trials often involve repeat observations on the same participants. When there are no more than two observations from each participant, standard mixed-model regression methods for a pretest-posttest design can be used. When there are more than two observations from each participant, random coefficients models may be useful. This paper describes the random coefficients analysis appropriate to data from an extended nested cohort design and presents the methods for power analysis and sample size calculations based on that analysis. We provide estimates for the parameters required for those calculations for a number of adolescent health behaviours. We also show how the estimates can be used to plan a future trial.

Understanding Hierarchical Linear Models

Nurses practice within hierarchical organizations and occupational structures. Hence, data emanating from nursing environments are structured, often inherently, hierarchically. From the perspective of ordinary regression, such structuring constitutes a statistical problem because this violates the assumption that we have observed independent and identical cases. A preferable approach is to employ analytical methods that mesh with the kinds of natural aggregations present in nursing environments. Consequently, there has been increasing interest in applying hierarchical, or multilevel, linear models to nursing contexts because this powerful analytical tool recognizes and accommodates naturally hierarchical data structures. The purpose of this article is to foster an understanding of both the strengths and limitations of hierarchical models. A hypothetical nursing example is progressively extended from the most basic hierarchical linear model toward a full two-level model. The structural similarities between two-level and three-level models are pointed out while focusing on the hierarchical nature of models rather than statistical technicalities. The limitations of hierarchical models are discussed also.

Equitable utilisation of Indian community based health insurance scheme among its rural membership: cluster randomised controlled trial

OBJECTIVE

To evaluate alternative strategies for improving the uptake of benefits of a community based health insurance scheme by its poorest members.

DESIGN

Prospective cluster randomised controlled trial.

SETTING

Self Employed Women's Association (SEWA) community based health insurance scheme in rural India. Participants 713 claimants at baseline (2003) and 1440 claimants two years later among scheme members in 16 rural sub-districts.

INTERVENTIONS

After sales service with supportive supervision, prospective reimbursement, both packages, and neither package, randomised by sub-district.

MAIN OUTCOME MEASURES

The primary outcome was socioeconomic status of claimants relative to members living in the same sub-district. Secondary outcomes were enrolment rates in SEWA Insurance, mean socioeconomic status of the insured population relative to the general rural population, and rate of claim submission.

RESULTS

Between 2003 and 2005, the mean socioeconomic status of SEWA Insurance members (relative to the rural population of Gujarat) increased significantly. Rates of claims also increased significantly, on average by 21.6 per 1000 members (P<0.001). However, differences between the intervention groups and the standard scheme were not significant. No systematic effect of time or interventions on the socioeconomic status of claimants relative to members in the same sub-district was found.

CONCLUSIONS

Neither intervention was sufficient to ensure that the poorer members in each sub-district were able to enjoy the greater share of the scheme benefits. Claim submission increased as a result of interventions that seem to have strengthened awareness of and trust in a community based health insurance scheme. Trial registration Clinical trials NCT00421629.

A cluster randomized clinical trial to improve prescribing patterns in ambulatory pediatrics

OBJECTIVES

Having shown previously that an electronic prescription writer and decision support system improved pediatric prescribing behavior for otitis media in an academic clinic setting, we assessed whether point-of-care delivery of evidence could demonstrate similar effects for a wide range of other common pediatric conditions.

DESIGN

Cluster randomized controlled trial.

SETTING

A teaching clinic/clinical practice site and a primary care pediatric clinic serving a rural and semi-urban patient mix.

PARTICIPANTS

A total of 36 providers at the teaching clinic/practice site and eight providers at the private primary pediatric clinic.

INTERVENTION

An evidence-based message system that presented real-time evidence to providers based on prescribing practices for acute otitis media, allergic rhinitis, sinusitis, constipation, pharyngitis, croup, urticaria, and bronchiolitis.

OUTCOME MEASURES

The proportion of prescriptions dispensed in accordance with evidence.

RESULTS

The proportion of prescriptions dispensed in accordance with evidence improved four percentage points, from 38% at baseline to 42% following the intervention. The control group improved by one percentage point, from 39% at baseline to 40% at trial's conclusion. The adjusted difference between the intervention and control groups was 8% (95% confidence interval 1%, 15%). Intervention effectiveness did not decrease with time.

CONCLUSION

For common pediatric outpatient conditions, a point-of-care evidence-based prescription writer and decision support system was associated with significant improvements in prescribing practices.

Use of over-the-counter malaria medicines in children and adults in three districts in Kenya: implications for private medicine retailer interventions

Background

Global malaria control strategies highlight the need to increase early uptake of effective antimalarials for childhood fevers in endemic settings, based on a presumptive diagnosis of malaria in this age group. Many control programmes identify private medicine sellers as important targets to promote effective early treatment, based on reported widespread inadequate childhood fever treatment practices involving the retail sector. Data on adult use of over-the-counter (OTC) medicines is limited. This study aimed to assess childhood and adult patterns of OTC medicine use to inform national medicine retailer programmes in Kenya and other similar settings.

Methods

Large-scale cluster randomized surveys of treatment seeking practices and malaria parasite prevalence were conducted for recent fevers in children under five years and recent acute illnesses in adults in three districts in Kenya with differing malaria endemicity.

Results

A total of 12, 445 households were visited and data collected on recent illnesses in 11, 505 children and 19, 914 adults. OTC medicines were the most popular first response to fever in children with fever (47.0%; 95% CI 45.5, 48.5) and adults with acute illnesses (56.8%; 95% CI 55.2, 58.3). 36.9% (95% CI 34.7, 39.2) adults and 22.7% (95% CI 20.9, 24.6) children using OTC medicines purchased antimalarials, with similar proportions in low and high endemicity districts. 1.9% (95% CI 0.8, 4.2) adults and 12.1% (95% CI 16.3,34.2) children used multidose antimalarials appropriately. Although the majority of children and adults sought no further treatment, self-referral to a health facility within 72 hours of illness onset was the commonest pattern amongst those seeking further help.

Conclusion

In these surveys, OTC medicines were popular first treatments for fever in children or acute illnesses in adults. The proportions using OTC antimalarials were similar in areas of high and low malaria endemicity. In all districts, adults were more likely to self-treat with OTC antimalarial medicines than febrile children were to receive them, and less likely to use them in recommended ways. Government health centres were the most common second resort for treatment and were often used within 72 hours. In view of these practices, more research is needed to assess the impact on the popularity of private medicine sellers of strengthened public sector policies on access to malaria treatment and insecticide-treated bed nets. Improved targeting of OTC antimalarials to high risk groups, better communication strategies regarding adult as well as children's dosages, and facilitating more rapid referral to trained health workers where needed are important challenges to private medicine seller programmes.

Community-randomized trial of lambdacyhalothrin-treated hammock nets for malaria control in Yanomami communities in the Amazon region of Venezuela

We conducted a community-randomized controlled trial in an area of moderate malaria transmission in the Amazon region, southern Venezuela, home of the Yanomami indigenous ethnic group. The aim was to compare the malaria incidence rate in villages with lambdacyhalothrin-treated hammock nets (ITHN) or with placebo-treated hammock nets (PTHN). In both arms of the study, intensive surveillance for early case detection was maintained and prompt malaria treatment was administered. Baseline data were collected before the intervention and a population of around 924 Yanomami was followed for 2 years. Despite the recent introduction of nets in the Yanomami villages and the adverse natural conditions in the area, the nets were accepted enthusiastically by the study population, used conscientiously and looked after carefully. The malaria incidence rate per thousand person-years at risk was 114.6 in the IHTN group and 186.8 in the PTHN group. The adjusted rate ratios indicated that ITHN prevent 56% [IRR: 0.44, 95% confidence interval (CI): 52-59%] of new malaria cases. ITHN reduced the prevalence of parasitaemia by 83% [relative risks (RR): 0.17, 95% CI: 47-100%], according to a cross-sectional survey carried out during the high transmission season. The prevalence of splenomegaly and anaemia was too low to detect any possible reduction as a result of ITHN. The main conclusion of the present study is that ITHN can reduce malaria incidence in the area and it is the most feasible method for malaria control in a forested area where indigenous villages are scattered over a large territory. This is the first community-level epidemiological trial to show that ITHN are highly effective against malaria transmitted by Anopheles darlingi.

Insecticide-treated bednets to prevent anthroponotic cutaneous leishmaniasis in Aleppo Governorate, Syria: results from two trials

The increasing incidence of anthroponotic cutaneous leishmaniasis (ACL) in Syria, coupled with the inefficacy of residual insecticide spraying to control the disease, have led to the further evaluation of the preventive efficacy of insecticide-treated bednets (ITNs) through large-scale field studies. Two trials were performed in the historical ACL-endemic areas of the Aleppo Governorate. In 1997-1999, a matched-cluster randomized trial was conducted in five intervention (ITNs) and five control (untreated bednets) villages, which involved a population of 10354 in 1321 households. In 2001-2003, a second study was performed in four villages (a population of 9325 in 858 households), which allowed the evaluation of the impact of the interruption of ITN intervention on ACL incidence rates. Both studies not only confirmed the high efficacy of ITNs in preventing ACL during 1 year post-intervention [about 85% (95% CL 76-98%) in the 1997-1999 trial], but also suggested that the interruption of this control measure might restore the pre-intervention disease incidence within 1-2 years. These findings underline the need for a sustainable and durable implementation of ITN-based control of the disease.

Community-level micronutrient fortification of a food supplement in India: a controlled trial in preschool children aged 36-66 mo

BACKGROUND

Children participating in the Integrated Child Development Service (ICDS) in India have high rates of iron and vitamin A deficiency.

OBJECTIVE

The objective was to assess the efficacy of a premix fortified with iron and vitamin A and added at the community level to prepared khichdi, a rice and dal mixture, in increasing iron and vitamin A stores and decreasing the prevalence of iron deficiency, anemia, and vitamin A deficiency.

DESIGN

This cluster, randomized, double-blind, controlled trial was initiated in 30 Anganwadi centers (daycare centers) in West Bengal state, India. Children aged 36-66 mo (n = 516) attending village-based ICDS centers were randomly assigned to receive either a fortified or a nonfortified premix for 24 wk. Blood was drawn at 0 and 24 wk by venipuncture for the measurement of hemoglobin, serum ferritin, and serum retinol.

RESULTS

The change in the hemoglobin concentration of anemic children was significantly different between fortified and nonfortified khichdi groups (P < 0.001). Prevalence rates of anemia, iron deficiency, and iron deficiency anemia were significantly lower after 24 wk in the fortified-khichdi group than in the nonfortified-khichdi group (P < 0.001). There were no significant differences in serum retinol concentrations or in the prevalence of vitamin A deficiency between the fortified- and nonfortified-khichdi groups.

CONCLUSION

A premix fortified with iron, vitamin A, and folic acid and added to supplementary food at the community level can be effective at increasing iron stores and reducing the prevalence of iron deficiency and anemia.

Impact and process evaluation of integrated community and clinic-based HIV-1 control: a cluster-randomised trial in eastern Zimbabwe

BACKGROUND

HIV-1 control in sub-Saharan Africa requires cost-effective and sustainable programmes that promote behaviour change and reduce cofactor sexually transmitted infections (STIs) at the population and individual levels.

METHODS AND FINDINGS

We measured the feasibility of community-based peer education, free condom distribution, income-generating projects, and clinic-based STI treatment and counselling services and evaluated their impact on the incidence of HIV-1 measured over a 3-y period in a cluster-randomised controlled trial in eastern Zimbabwe. Analysis of primary outcomes was on an intention-to-treat basis. The income-generating projects proved impossible to implement in the prevailing economic climate. Despite greater programme activity and knowledge in the intervention communities, the incidence rate ratio of HIV-1 was 1.27 (95% confidence interval [CI] 0.92-1.75) compared to the control communities. No evidence was found for reduced incidence of self-reported STI symptoms or high-risk sexual behaviour in the intervention communities. Males who attended programme meetings had lower HIV-1 incidence (incidence rate ratio 0.48, 95% CI 0.24-0.98), and fewer men who attended programme meetings reported unprotected sex with casual partners (odds ratio 0.45, 95% CI 0.28-0.75). More male STI patients in the intervention communities reported cessation of symptoms (odds ratio 2.49, 95% CI 1.21-5.12).

CONCLUSIONS

Integrated peer education, condom distribution, and syndromic STI management did not reduce population-level HIV-1 incidence in a declining epidemic, despite reducing HIV-1 incidence in the immediate male target group. Our results highlight the need to assess the community-level impact of interventions that are effective amongst targeted population sub-groups.

Design considerations in a clinical trial of a cognitive behavioural intervention for the management of low back pain in primary care: Back Skills Training Trial

BACKGROUND

Low back pain (LBP) is a major public health problem. Risk factors for the development and persistence of LBP include physical and psychological factors. However, most research activity has focused on physical solutions including manipulation, exercise training and activity promotion.

METHODS/DESIGN

This randomised controlled trial will establish the clinical and cost-effectiveness of a group programme, based on cognitive behavioural principles, for the management of sub-acute and chronic LBP in primary care. Our primary outcomes are disease specific measures of pain and function. Secondary outcomes include back beliefs, generic health related quality of life and resource use. All outcomes are measured over 12 months. Participants randomised to the intervention arm are invited to attend up to six weekly sessions each of 90 minutes; each group has 6-8 participants. A parallel qualitative study will aid the evaluation of the intervention.

DISCUSSION

In this paper we describe the rationale and design of a randomised evaluation of a group based cognitive behavioural intervention for low back pain.

HIV incidence during a cluster-randomized trial of two strategies providing voluntary counselling and testing at the workplace, Zimbabwe

OBJECTIVE

To investigate HIV incidence during a trial of two voluntary counselling and testing (VCT) strategies. Counselling may promote beneficial behavioural change, although knowledge of negative status does not appear to contribute further benefit.

DESIGN

The parent cluster-randomized trial demonstrated much greater uptake of VCT when counselling and rapid testing were available on-site (intensive VCT) than through pre-paid vouchers to an external provider (standard VCT). Anonymous HIV tests had been requested from all employees at enrolment and after 2 years intervention.

METHODS

The study setting was 22 businesses in Harare, Zimbabwe. Participants were 3146 HIV-negative individuals remaining in employment at the end of intervention, of whom 2966 (94.3%) consented to repeat testing. VCT linked to basic HIV care was provided and the main outcome measures were HIV incidence under each study arm, as a retrospective secondary analysis.

RESULTS

Mean VCT uptake in this cohort was 70.7 and 5.2%, respectively, in the intensive and standard arms. Crude HIV incidence was 1.21 per 100 person-years, with non-significantly higher rates in the intensive VCT arm [mean site incidence 1.37 and 0.95 per 100 person-years, respectively; adjusted rate ratio 1.49 (95% confidence interval 0.79-2.80).

CONCLUSIONS

Highly acceptable VCT did not reduce HIV incidence in this predominantly male cohort. HIV incidence was highest in the high uptake VCT arm, lending support to a US trial in which rapid testing appeared to have adverse behavioural consequences in some HIV-negative clients. Careful comparison of outcomes under different counselling and testing strategies is needed to maximize HIV prevention from global scale-up of VCT.

Design and analysis of stepped wedge cluster randomized trials

Cluster randomized trials (CRT) are often used to evaluate therapies or interventions in situations where individual randomization is not possible or not desirable for logistic, financial or ethical reasons. While a significant and rapidly growing body of literature exists on CRTs utilizing a "parallel" design (i.e. I clusters randomized to each treatment), only a few examples of CRTs using crossover designs have been described. In this article we discuss the design and analysis of a particular type of crossover CRT - the stepped wedge - and provide an example of its use.