Regression is one of the most widely used analytical tools in statistics and data science for understanding relationships between variables and explaining real-world phenomena. In health research, it plays a central role for students, junior researchers, and senior investigators when identifying factors associated with clinical or public-health outcomes.

In practice, beginners often struggle with regression because it is not always clear what makes a model “good.” Common difficulties include obtaining contradictory results after adding or removing variables, coefficients that change direction or magnitude unexpectedly, wide confidence intervals that limit interpretation, and models that are hard to explain despite appearing statistically sound. These problems usually arise not from incorrect calculations, but from limited understanding of variable selection, parsimony, and the goals of regression modeling.

Despite its importance, regression is frequently misused. A common mistake is the inclusion of too many explanatory variables in a single model, under the assumption that more variables necessarily lead to better results. In practice, overly complex models often perform poorly, are difficult to interpret, and may generalize badly to new data.

A good model is therefore not the one that includes the largest number of variables, but the one that explains reality accurately with the least necessary complexity. This principle is known as parsimony.

1. What is regression ?

Regression is a statistical method used to study the relationship between a dependent variable (the outcome) and one or more independent (explanatory) variables, while accounting for the influence of other factors, including potential confounders. In practical terms, regression allows a more accurate estimation of the association between an explanatory variable (X) and an outcome (Y), by adjusting for other variables that may influence this relationship.

For example, comparing the performance of two surgeons solely based on crude surgical success rates may lead to misleading conclusions. Such a comparison ignores important factors such as case complexity, surgical technique, and years of experience. Regression makes it possible to adjust for these variables, allowing a fairer and more informative comparison between surgeons.

Depending on the nature of the dependent variable, several types of regression are commonly used:

• Simple and multiple linear regression
• Binary, ordinal, and multinomial logistic regression
• Poisson regression
• Cox proportional hazards regression

2. Explanatory versus Predictive Regression: Distinct Goals

Before evaluating model quality, it is essential to distinguish between two broad purposes of regression modeling: explanatory and predictive.

Explanatory models aim to understand relationships between variables. Their primary goals are interpretation, estimation of associations, and insight into underlying mechanisms. In this context, emphasis is placed on coefficient stability, plausibility, and clarity of interpretation. Parsimony is critical because overly complex models obscure understanding and weaken inference.

Predictive models, by contrast, aim to accurately predict outcomes for new observations. Their success is judged mainly by predictive performance, often evaluated using validation datasets and prediction error metrics. Such models may include many variables and interactions if they improve predictive accuracy, even at the expense of interpretability.

This blog focuses primarily on explanatory regression modeling, where the goal is not prediction, but understanding and communicating meaningful associations. The principles discussed below are therefore framed around interpretability, stability, and parsimony rather than predictive optimization.

3. Evaluation of the Quality of a Regression Model

The quality of a regression model is primarily assessed by its ability to explain or predict the outcome while maintaining interpretability and stability. Several performance indicators exist, among which the most commonly used are the coefficients of determination: R² and adjusted R².

• R² (coefficient of determination) represents the proportion of variability in the dependent variable (Y) explained by the model. Its value ranges from 0 to 1, with higher values indicating better explanatory power. However, R² always increases when new variables are added, even if those variables contribute little meaningful information.

• Adjusted R² addresses this limitation by accounting for the number of explanatory variables included in the model. It penalizes unnecessary predictors and therefore allows a fairer comparison between models with different levels of complexity.

The objective is not to maximize the number of variables, but to identify the optimal combination of explanatory variables (predictors) (Xi) that explain the outcome (Y) adequately without unnecessary complexity. Such a model is referred to as parsimonious: simple, interpretable, and robust, with a reduced risk of overfitting.

What is overfitting in explanatory models? In the context of explanatory regression, overfitting refers to a situation in which a model includes too many predictors relative to the available information, leading to unstable estimates, inflated standard errors, and coefficients that are highly sensitive to small changes in the data. Rather than clarifying the relationships between predictors and the outcome, an overfitted explanatory model obscures interpretation and weakens inferential validity, making it difficult to draw reliable conclusions about associations or underlying mechanisms.

4. How to obtain a parsimonious model?

Before discussing specific techniques, it is useful to keep a simple workflow in mind. In most applied analyses, building a parsimonious regression model follows four broad steps: (1) prepare the data and ensure an adequate number of events, (2) identify plausible variables based on theory and prior evidence, (3) reduce redundancy and overfitting through selection or penalization, and (4) evaluate model performance and interpretability.

4.1. Data preparation and the events‑per‑variable principle

Before performing any regression analysis, careful data preparation is important. This includes checking for sparse categories, zero counts, and implausible values that may compromise the stability of parameter estimates.

An important consideration is the ratio between the number of observed events (for example, cases with a positive outcome) and the number of explanatory variables included in the model. Vittinghoff et al. (2007) and van Smeden et al. (2016), emphasize the importance of this ratio in limiting overfitting and improving model validity. A commonly recommended rule of thumb is a minimum of five events per variable.

Example: If a study includes 20 observed events, no more than four explanatory variables should be entered into the model (20 ÷ 5 = 4). Exceeding this threshold increases the risk of unstable estimates and poor generalizability.

4.2. Variable selection strategies

Once initial data checks are complete, variable selection becomes the next critical step. Three main approaches are commonly used.

1. Manual selection

Manual selection relies on subject‑matter knowledge, clinical reasoning, and evidence from the literature. Variables are chosen based on prior evidence of association with the outcome rather than purely statistical considerations.

For example, when studying factors associated with loss to follow‑up among patients receiving antiretroviral therapy, variables such as distance to the health facility, income level, or perceived stigma are more plausible candidates than biologically unrelated variables like blood type.

An initial bivariate screening is often conducted using appropriate statistical tests, such as:

• The Chi-square test – used to assess the association between categorical dependent and independent variables.
• Spearman, Pearson, or Kendall correlation tests – used to assess the relationship between quantitative or ordinal dependent and independent variables, depending on data distribution and measurement scale.

Because this step is exploratory, a more lenient significance threshold is typically used, often around 10% (p < 0.10) or using relaxed p-value cutoffs such as p < 0.20 or p < 0.25, following the purposeful selection strategy described by Hosmer and colleagues in Applied Logistic Regression. They recommend avoiding the premature exclusion of potentially important variables, as some variables that are not statistically significant in univariate analysis may become significant after adjustment for confounding factors in the multivariable model. They further recommend initially including variables with a p-value < 0.25 in univariate analysis, and then retaining in the multivariable model those that remain statistically significant at the conventional threshold of p < 0.05.

Table 1: An example of candidate variable selection.

Following this step, collinearity between explanatory variables must be assessed. Strong correlations indicate redundant information and may destabilize the model.

For example, if weight and BMI are included in the same model, their strong correlation can lead to multicollinearity. In this case, BMI might be preferred, as it already incorporates weight in its calculation.

Common tools include Pearson, Spearman, or Kendall correlation coefficients, as well as the Variance Inflation Factor (VIF). A VIF value above 10 generally suggests problematic multicollinearity. When strong collinearity is detected, one variable should be retained based on clinical or contextual relevance, or a composite variable may be constructed.

i. Automatic selection

Automatic selection relies on algorithmic procedures that evaluate the contribution of each variable to overall model fit, most often through the likelihood function or related information criteria.

• Forward selection starts with an empty model and adds variables one at a time, selecting at each step the variable whose inclusion leads to the greatest improvement in model fit.
• Backward selection begins with a full model that includes all candidate variables and then removes variables sequentially, eliminating those whose exclusion has the smallest impact on model fit.
• Stepwise selection combines both approaches, allowing variables to be added or removed iteratively based on predefined criteria.

For example, consider a logistic regression model aimed at identifying factors associated with loss to follow-up among patients on antiretroviral therapy. Suppose the initial set of candidate variables includes age, sex, education level, distance to the health facility, employment status, and perceived stigma.

• In a forward selection approach, the model may first include distance to the health facility because it provides the largest improvement in likelihood. In the next step, employment status may be added, followed by perceived stigma, until no additional variable meaningfully improves model fit. The final model may therefore include only three predictors, even though six were initially considered.
• In contrast, a backward selection approach would start with all six variables in the model and progressively remove those that contribute the least. For example, sex and age may be dropped early if their removal does not substantially worsen model fit, leading again to a more compact model.

These methods are widely implemented in statistical software packages such as SPSS and can be useful for exploratory analyses. However, because they rely heavily on statistical criteria and may ignore clinical or contextual relevance, their results should be interpreted cautiously and ideally complemented by subject-matter knowledge.

Figure 2 shows the SPSS dialog for binary logistic regression. After loading the dataset, the interface is accessed via Analyze → Regression → Binary Logistic Regression. The outcome is specified in the Dependent field and candidate predictors in Covariates. The Method menu defines the variable selection strategy. By default, Enter includes all selected variables simultaneously (no selection). Forward and backward options implement automatic selection based on likelihood ratio, Wald, or conditional criteria, corresponding to the procedures described in the text.

ii. Penalized regression methods

Advanced techniques such as Lasso, Ridge, and Elastic Net regression are designed to control model complexity by adding a penalty to the regression coefficients. In simple terms, these methods discourage the model from assigning large effects to many variables at the same time.

Why is this useful? In explanatory analyses, including too many variables or highly correlated predictors can destabilize coefficient estimates, inflate uncertainty, and complicate interpretation. Penalization helps limit unnecessary complexity, producing more stable and interpretable estimates, even when predictors are correlated.

• Ridge regression shrinks the coefficients of correlated variables toward zero, reducing instability without removing variables entirely.
• Lasso regression goes a step further by shrinking some coefficients exactly to zero, effectively performing automatic variable selection.
• Elastic Net combines both approaches, making it useful when many predictors are correlated.

These methods are used mainly in predictive modeling, but they can also be informative in exploratory explanatory analyses when the number of candidate predictors is large. However, because penalization alters coefficient estimates, results should be interpreted cautiously when the primary goal is causal explanation. While they often produce stable and accurate models, the resulting coefficients should be interpreted cautiously, especially when drawing causal conclusions.

Common Pitfalls to Avoid

When building regression models, several recurrent mistakes should be avoided:

• Including variables solely because they are statistically significant in bivariate analysis
• Ignoring collinearity between predictors
• Relying on automatic selection methods without theoretical justification
• Interpreting coefficients from heavily penalized models as causal effects

Conclusion

A high‑quality regression model is not one that fits the data perfectly, but one that explains reality accurately using the fewest necessary variables. Parsimony is therefore central to building models that are interpretable, stable, and generalizable.

When in doubt, a simpler model that can be clearly explained and defended is often preferable to a complex model whose assumptions and results are difficult to justify.

References

1. Chesneau, C. (2015, 7 novembre). Modèles de régression. Université de Caen Basse-Normandie. Consulté le 26/07/2025 sur http://www.math.unicaen.fr/~chesneau/
2. Wikistat. Sélection de modèle en régression linéaire. Consulté le 26/07/2025 sur  https://www.math.univ-toulouse.fr/~besse/Wikistat/pdf/st-m-app-linSelect.pdf
3. Vittinghoff E, McCulloch CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol. 2007 Mar 15;165(6):710-8. doi: 10.1093/aje/kwk052. Epub 2006 Dec 20. PMID: 17182981
4. van Smeden M, de Groot JA, Moons KG, Collins GS, Altman DG, Eijkemans MJ, Reitsma JB. No rationale for 1 variable per 10 events criterion for binary logistic regression analysis. BMC Med Res Methodol. 2016 Nov 24;16(1):163. doi: 10.1186/s12874-016-0267-3. PMID: 27881078; PMCID: PMC5122171.
5. Legrand P, Bories D. Le choix des variables explicatives dans les modèles de régression logistique. Communication présentée aux Journées de l’AIMS 2007, mai 2007. Disponible sur: https://www.researchgate.net/publication/281834969
6. Bursac, Z., Gauss, C. H., Williams, D. K., & Hosmer, D. W. (2008). Purposeful selection of variables in logistic regression. BMC Medical Research Methodology, 8, 17. https://doi.org/10.1186/1471-2288-8-17
7. Hosmer DW, Lemeshow S, Sturdivant RX. Applied Logistic Regression. 3rd ed. New York: Wiley; 2013.

The post A Beginner’s Guide to Building Better Explanatory Regression Models appeared first on CRENC Learn.

1. Types of Missing Data

1.1 Missing Completely at Random (MCAR):

Data are said to be missing completely at random when the absence of a value depends neither on observed nor unobserved variables. In other words, the reasons for which certain data are missing are entirely independent of the characteristics of the individuals or the parameters being studied, and result purely from chance. When this condition is met, analyses performed on the remaining data are unbiased. However, this scenario is rarely encountered in practice. In the example (see Figure 1), we observe that missing values of Systolic Blood Pressure (SBP) are unrelated to any other characteristic, as both men and women, young and old alike, are affected.

1.2 Missing at Random (MAR):

This occurs when missing observations are not completely random but can be fully explained by variables for which complete information exists. For example, unlike MCAR, we may notice that missing SBP values tend to occur among individuals under 30 years of age.

1.3 Missing Not at Random (MNAR):

Also known as non-ignorable or non-response data, MNAR occurs when the value of the missing variable is related to the reason why it is missing. For instance, individuals without recorded SBP values might be those who arrived with high blood pressure in emergency situations, where the measurement could not be taken immediately.

2. Approaches for Handling Missing Data

Thus, to address the problem of missing values, we can use:

• Deletion Methods
• Imputation Methods

2.1 Deletion Methods (Complete Case Analysis)

This method, also called Complete Case Analysis, is one of the simplest. It involves identifying and removing observations (rows) that contain missing values (Figure 2). However, it is recommended only when the proportion of missing data is very small (less than 5% of the total population), to avoid biasing the dataset. For example, in a dataset of 100 individuals describing their sociodemographic profiles, if the ages of 25 participants (25% of the total) are missing, deleting these cases would lead to a significant loss of information. In such cases, imputation methods are preferred.

2.2 Simple Imputation Methods

Imputation methods involve replacing missing values with the mean or median of the series, or with values generated through more sophisticated techniques such as extrapolation or iterative Principal Component Analysis (PCA).

If the series is normally distributed, missing values are replaced by the mean; otherwise, by the median. For qualitative variables, the mode (most frequent category) is used. In our previous example, the distribution of the 75 observed ages is examined. If normally distributed, the 25 missing ages are replaced by the mean; if not, by the median. Although commonly used, this method tends to reduce variability by clustering values around the mean or median, potentially introducing bias. In Figure 3, missing “sex” values are replaced by the most frequent category (female), and quantitative variables like age and SBP are replaced by their respective means (38 and 113).

2.3 Multiple Imputation

This is currently the most widely used and recommended method in modern statistical analysis. It replaces missing values several times with plausible values generated from a probabilistic model. Each completed dataset is analyzed separately, and the results are combined to produce a more robust final estimate (Figure 3). This approach preserves the natural variability of the data and minimizes bias. Unlike simple imputation, it does not rely on a single fixed estimate but accounts for uncertainty in the estimation process.

2.4 Other imputation techniques

2.4.1 Interpolation and Extrapolation Methods

These are deterministic approaches that predict missing values based on relationships or correlations between variables. For example, if age correlates with height, taller individuals might be assigned higher ages, while shorter ones are assigned lower ages. The most common form is linear extrapolation. The main limitation is that for many variables, this process becomes tedious, requiring pairwise analysis this is where iterative PCA becomes useful..

2.4.2 Iterative Principal Component Analysis (PCA)

PCA is an exploratory data analysis technique that helps visualize relationships among several quantitative variables across different dimensions. Iterative PCA estimates missing values by repeating the PCA process several times, each time using results from the previous iteration to refine imputations, until the best dataset is obtained.

This process is available in several statistical software tools. Because the algorithm can iterate indefinitely, researchers must specify a maximum number of iterations.

3. Practical Example with SPSS

Once the data set is imported into SPSS and variables with completeness issues are identified, go to the Transform menu and find the option that allows you to perform this operation.

Choose the appropriate imputation method; in this case, we will choose mean imputation.

Select the variable to be processed and assign a name to the new variable.

The result is the following: all missing values have been replaced by the series mean, and this has been done in a new variable named exper_1.

Conclusion

Proper management of missing values is essential to ensure the validity and reliability of statistical analyses. A rigorous methodological approach that incorporates appropriate imputation techniques while accounting for data characteristics enhances research quality and increases confidence in conclusions. Although the choice of method depends on the nature of the missing data, Complete Case Analysis and Multiple Imputation remain the most widely used.

References

1. insightsoftware. How to handle missing data values while data cleaning. insightsoftware. 2023. https://insightsoftware.com/fr/blog/how-to-handle-missing-data-values-while-data-cleaning/. Accessed April 14, 2025.

2. Medistica. pvalue.io, a GUI of R statistical software for scientific medical publications. pvalue.io. 2019. https://www.pvalue.io. Accessed April 14, 2025.

3. Wikistat. Imputation of missing data.. https://www.math.univ-toulouse.fr/~besse/Wikistat/pdf/st-m-app-idm.pdf. 2025.

4. Expert. Handling missing data: best practices for 2024. Editverse. 2024. https://www.editverse.com/fr/meilleures-pratiques-de-gestion-des-donn%C3%A9es-manquantes-pour-les-chercheurs-en-2024 . Accessed 15 Apr 2025.

5. Ebasone, P.V, Peer N, Dzudie A, et al. (2025). Reporting and handling of missing data in published studies of co-morbid hypertension and diabetes among people living with HIV/AIDS: a systematic review. BMC Medical Research Methodology, 25, 180.
https://doi.org/10.1186/s12874-025-02630-1

6.Peter Ebasone (2025). Handling missing data in practice:Complete Case Analysis vs Multiple Imputation. CRENC [See Slides].

The post Procedure for Handling Missing Data in Statistical Analysis appeared first on CRENC Learn.

Understanding Implementation Science

Implementation science is defined as the study of methods to promote the adoption and integration of evidence-based practices, interventions, and policies into routine health care and public health settings to improve the impact on population health. Implementation science was borne out of a desire to address challenges associated with the use of research to achieve more evidence-based practice (EBP) in health care and other areas of professional practice. IS is dedicated to closing the gap between research findings and everyday practice, especially within public health. It examines how to apply proven interventions in real-world settings, focusing on understanding the factors that affect successful adoption, implementation, and sustainability of health programs. This field of study empowers health professionals to adapt and apply research-backed interventions to meet the unique needs of our communities, ensuring that impactful solutions reach those who need them most crucial for optimizing limited resources, strengthening healthcare systems, and improving health outcomes. 

As we continue to face global health challenges, from emerging infectious diseases to the growing burden of non-communicable diseases, the importance of implementation science cannot be overstated. And we cannot talk about implementation science without discussing the different research domains. Such as Implementation Research, Operational Research and Health System Research. The diagram below briefly summarises and explains these domains.

Figure 1: Different Research Domains

What is the role of Implementation Science in Public Health

• Healthcare settings often are plagued with lots of challenges including but not limited to: infrastructural challenges, limited resources, and unique cultural barriers that make implementing research-based interventions difficult. IS helps tackle limited resources, infrastructural gaps, and other barriers to implementing research-based interventions.
• Imagine you’re standing on one side of a wide river. On your   side is a wealth of public health research; studies, data, and evidence-based strategies. On the other side is the real world, where people live, work, and face everyday health challenges. Often, the bridge connecting these two sides is unstable, incomplete, or entirely missing. This is where IS comes in, the sturdy bridge that connects the vast knowledge we have in public health to the everyday realities people face. Without it, even the best research can fall short, leaving gaps in care, prevention, and health promotion. IS is important in closing the existing ‘KNOW-DO’ gap which takes an average of 17 years for only 14% of findings to be implemented.

• It ensures innovative solutions to integrate research into standard practice, crucial in resource-constrained countries like Cameroon.
• It is essential for overcoming barriers as it studies what works, for whom, and under what conditions, making interventions adaptable to local needs. Another practical example: is the use of insecticide-treated nets (ITNs) to prevent malaria. Scientific research has shown that the use of ITN (sleeping under an ITN, because many people ‘use’ ITNs for purposes they weren’t ideally invented for) can reduce malaria transmission by about 50%. However, despite this knowledge, many communities at high risk for malaria do not consistently use these nets. The question remains why? IS comes in, it explores questions like:
  • What are the barriers to ITN use? Are they cultural? Logistical? Economic?
  • How can these barriers be overcome? Would community education help? What about subsidizing the cost of nets?
  • How can we ensure long-term, sustainable use of ITNs?

By answering these questions, IS helps bridge the gap between knowing what works and actually making it work in the field.

Key Components of Implementation Science

In the context of implementation science, the key components can be succinctly categorized into “the how,” “the what,” and “the where.” These components focus on different aspects of implementing evidence-based interventions effectively:

The What: This component involves the intervention itself the specific practice, program, or policy being implemented. It focuses on the content and nature of the intervention, including: the objectives and goals of the intervention, the components and procedures involved and the expected outcomes and benefits.

The How: This component refers to the implementation strategies and processes used to integrate the intervention into practice. It encompasses the methods and techniques employed to facilitate the intervention’s uptake, adoption, and sustained use. Examples include: training and education for healthcare providers, development of guidelines or protocols and stakeholder engagement and communication strategies.

The Where: This refers to the context or setting in which the intervention is being implemented. It includes the environment, organizational, and systemic factors that can influence the implementation process. Such as the healthcare or public health setting (e.g., hospitals, clinics, community centres etc) which is referred to as an enabling environment.  Let’s dive straight into looking at these key components:

Evidence Based Interventions (EBIs)

Evidence-based interventions (EBIs) are the subject of implementation science, such as: programs, practices, principles, procedures, products, pills, or policies (7Ps) that have been demonstrated to improve health behaviours, health outcomes, or health-related environments. Evidence-based interventions are what is being implemented.

For example:

• Programs that provide HIV testing and counselling services in community settings to improve access and early diagnosis and testing.
• Use of ITNs for prevention malaria by pregnant women
• Providing free breast and cervical cancer screening for early detection and prevention of cervical cancer.

Implementation Strategies:

Now we know about EBIs but these EBIs need to be put to work/ practice, this can be done through best practices known as implementation strategies. These are the actions taken to enhance adoption, implementation, and sustainability of evidence-based interventions. Implementation strategies are how we seek to get evidence-based interventions into normal practice in clinical or community settings. This involves creating tailored approaches to overcome barriers and leverage facilitators for getting best results. 

There exist a list of 73 implementation strategies compiled by the ERIC project (found here) which provides detailed definitions and guidance on various methods to improve the adoption and effectiveness of evidence-based interventions. Some practical implementation strategies include:

• The training and education of CHWs for community dispensation of ART which is carried out in our context.
• The development of guidelines or protocols to improve ART adherence and health outcomes of people with HIV.
• The integration of HIV care into other health programs.

It is important to know what implementation strategies best works to suit the EBI being implemented. The diagram below serves as a guide for choosing the best fit HOW for the WHAT.

Implementation Outcomes

To assess the effectiveness of the implementation strategy there are what we call implementation outcomes. Implementation outcomes are key variables in assessing how well an intervention, policy, or program is integrated into real-world practice. These outcomes help to evaluate the effectiveness of strategies used to implement health interventions and are crucial in understanding the success of public health programs. The core implementation outcomes, as identified by Proctor et al. (2011), include acceptability, adoption, appropriateness, feasibility, fidelity, implementation cost, penetration, and sustainability. The table below provides vivid explanation of implementation outcomes and its applicability.

Table 1: Some Implementation Outcomes with real world examples

Enabling Environments:

These are the supportive infrastructures, policies, and resources that facilitate implementation. An enabling environment could include funding from the government or partnerships with non-governmental organizations (NGOs) that provide resources.

Stakeholders:

 In Cameroon, stakeholders in HIV and malaria programs range from patients and caregivers to healthcare providers, policymakers, and international partners. Their roles and level of engagement significantly impact implementation success.

Implementation Context:

This refers to physical, economic, political, and cultural environment in which the intervention is deployed. It could be in the community, at the health facility or school. What works in one context may not be effective in another due to variations in resources, culture, or infrastructure. The diagram below provides a detailed picture of how the different concepts of implementation science are applied.

Conclusion

Implementation science serves as a crucial bridge between research and practice, particularly in the context of resource-limited settings like Cameroon. By systematically exploring methods to integrate evidence-based interventions into real-world healthcare settings, it plays an instrumental role in strengthening health systems and improving health outcomes. With a focus on understanding local needs, overcoming barriers, and adapting to specific cultural and logistical challenges, implementation science offers a strategic approach for maximizing the impact of public health programs, especially those addressing high-priority areas like HIV.

References

1. Eccles, M.P., Mittman, B.S. Welcome to Implementation Science . Implementation Sci 1, 1 (2006). https://doi.org/10.1186/1748-5908-1-1
2. About Implementation Science (IS). National Cancer Institute: Division of Cancer Control & Population Sciences. https://cancercontrol.cancer.gov/IS/about.html. March 8, 2018.
3. Cabinet Implementation Unity Toolkit: Engaging Stakeholders. Australian Government. https://www.pmc.gov.au/sites/default/files/files/pmc/implementation-toolkit-3-engagingstakeholders.pdf. June 2013.
4. Brownson RC, Kreuter MW, Arrington BA, True WR. Translating scientific discoveries into public health action: how can schools of public health move us forward? Public Health Rep. 2006 Jan-Feb;121(1):97-103. doi: 10.1177/003335490612100118. PMID: 16416704; PMCID: PMC1497798.
5. Yapa, H.M., Bärnighausen, T. Implementation science in resource-poor countries and communities. Implementation Sci 13, 154 (2018). https://doi.org/10.1186/s13012-018-0847-1
6. World Health Organization. WHO Guideline on when to start antiretroviral therapy and on pre-exposure prophylaxis for HIV September 2015 [27th Dec, 2019]. https://apps.who.int/iris/bitstream/handle/10665/186275/9789241509565_eng.pdf;jsessionid=D95F5690AC992DCB82B673779D9EF5D9?sequence=1
7. Proctor, E., Silmere, H., Raghavan, R. et al. Outcomes for Implementation Research: Conceptual Distinctions, Measurement Challenges, and Research Agenda. Adm Policy Ment Health 38, 65–76 (2011). https://doi.org/10.1007/s10488-010-0319-7
8. Proctor, E.K., Bunger, A.C., Lengnick-Hall, R. et al. Ten years of implementation outcomes research: a scoping review. Implementation Sci 18, 31 (2023). https://doi.org/10.1186/s13012-023-01286-z
9. Powell, B.J., Waltz, T.J., Chinman, M.J. et al. A refined compilation of implementation strategies: results from the Expert Recommendations for Implementing Change (ERIC) project. Implementation Sci 10, 21 (2015). https://doi.org/10.1186/s13012-015-0209-1
10. Rogers Ajeh, Halle Ekane, Egbe O. Thomas, Anastase Dzudie, and Assob N. Jules, “Perceived Patients’ Satisfaction, Barriers and Implications on Engagement in Antiretroviral Treatment Services in Cameroon within the HIV Test and Treat Context.” American Journal of Public Health Research, vol. 7, no. 2 (2019): 73-80. doi: 10.12691/ajphr-7-2-5.
11. Cameroon AIDS Month for campaigns [2024]; https://cnls.cm/site/en/mois-camerounais-mc#:~:text=Cameroon%20AIDS%20Month%20is%20the,AIDS%20Control%20Committee%20(NACC).
12. Ndenkeh JJN, Bowring AL, Njindam IM, Folem RD, Fako GCH, Ngueguim FG, Gayou OL, Lepawa K, Minka CM, Batoum CM, Georges S, Temgoua E, Nzima V, Kob DA, Akiy ZZ, Philbrick W, Levitt D, Curry D, Baral S. HIV Pre-exposure Prophylaxis Uptake and Continuation Among Key Populations in Cameroon: Lessons Learned From the CHAMP Program. J Acquir Immune Defic Syndr. 2022 Sep 1;91(1):39-46. doi: 10.1097/QAI.0000000000003012. Epub 2022 May 10. PMID: 35536113; PMCID: PMC9377496.
13. Johns Hopkins School of Public Health and Metabiota Cameroon. Feasibility and acceptability of HIV self-testing among female sex workers and men who have sex with men in Yaoundé, Cameroon. Washington, DC: LINKAGES; 2018.
14. Decentralized Drug Distribution in Cameroon: Final Report. https://www.fhi360.org/wp-content/uploads/2024/02/epic-ddd-cameroon-english.pdf
15. Nilsen, P. Making sense of implementation theories, models and frameworks. Implementation Sci 10, 53 (2015). https://doi.org/10.1186/s13012-015-0242-0
16. Ogundahunsi O, Kamau EM (eds). Implementation research toolkit, second edition. Geneva: World Health Organization; 2018. Licence: CC BY-NC-SA 3.0 IGO.

The post Unveiling the Fundamentals of Implementation Science appeared first on CRENC Learn.

5. Comparing Questionnaires with Other Data Collection Methods

Understanding how questionnaires compare to other research methods helps researchers choose the most appropriate approach for their study objectives. While questionnaires offer many advantages, they are just one tool in a researcher’s toolkit. Let’s explore how questionnaires compare to other common data collection methods:

5.1 Questionnaires vs. Focus Groups

Questionnaires are particularly effective when you need to collect standardized data from a large number of participants. They allow for statistical analysis and generalization of findings to broader populations. Participants can complete questionnaires at their convenience, and the anonymity they provide can encourage honest responses to sensitive questions.

Focus Groups, in contrast, bring together small groups of participants (typically 6-10) for guided discussions. This method excels at uncovering the “why” behind opinions and exploring how views are formed through social interaction. While questionnaires capture individual responses, focus groups reveal how ideas develop through conversation and can generate insights that might not emerge in isolation.

When to choose which method:

• Choose questionnaires when: you need a large sample size, statistical data, or to test specific hypotheses
• Choose focus groups when: you want to explore emerging topics, understand group dynamics, or generate new ideas through discussion

5.2 Questionnaires vs. Observation

Questionnaires rely on self-reported information, which is valuable for understanding participants’ perceptions, beliefs, and reported behaviors. They can cover events that occurred in the past or hypothetical scenarios, and they make it possible to gather data across different locations simultaneously.

Observation involves systematically watching and recording behavior in natural settings. This approach captures what people actually do rather than what they say they do. Observation is particularly valuable for studying behaviors that participants may not be aware of or might not report accurately. It provides rich contextual information about environments and nonverbal behaviors. You can have passive observation (where you mostly “observe” with no interaction) and active observation (where you can interact and observe their habits according to particular situations).

When to choose which method:

• Choose questionnaires when: self-reported information is sufficient, you need data on attitudes or opinions, or when direct observation is impractical
• Choose observation when: actual behavior is more important than reported behavior, context is crucial, or when participants may not be able to articulate their experiences

5.3 Questionnaires vs. Medical Records

Questionnaires can collect data on a wide range of topics, including subjective experiences like pain, satisfaction, or quality of life. They can be deployed quickly to gather current information and can include questions about various aspects of health that might not be documented in medical records.

Medical Records provide verified clinical data documented by healthcare professionals. They offer objective measurements, diagnoses, and treatment details that are often more accurate than recalled information. Medical records can provide longitudinal data collected over time without the bias of retrospective reporting.

When to choose which method:

• Choose questionnaires when: subjective experiences are important, you need information not typically documented in records, or when accessing records is not feasible
• Choose medical records when: accuracy of clinical information is critical, you need longitudinal data, or when recall bias is a significant concern

5.4 Combining Methods for Stronger Research

Many strong research designs incorporate multiple data collection methods to capitalize on the strengths of each while offsetting their limitations. For example:

• Using questionnaires to identify broad patterns, then conducting focus groups to explore the reasons behind those patterns
• Combining observational data with questionnaires to compare actual behaviors with self-reported behaviors
• Supplementing medical record data with questionnaires about quality of life or patient experiences

When designing your research, consider how different methods might complement each other to provide a more complete understanding of your research question. While this approach requires more resources, the depth and richness of the resulting data often justify the additional investment.

6. Advantages and Limitations of Questionnaires

Understanding both the strengths and weaknesses of questionnaires helps researchers make informed decisions about when to use them and how to design them effectively. Here we explore the key advantages that make questionnaires popular and the limitations that researchers should consider.

Addressing Questionnaire Limitations

For each limitation of questionnaires, there are strategies researchers can employ to minimize their impact:

To address limited depth:

• Include some open-ended questions to allow for more detailed responses
• Consider complementing questionnaires with qualitative methods like interviews or focus groups
• Use follow-up questionnaires to explore interesting findings in more depth

To reduce response bias:

• Ensure anonymity and confidentiality to encourage honest responses
• Avoid leading questions that suggest “correct” answers
• Use indirect questioning techniques for sensitive topics
• Include validation questions to check for consistency in responses

To improve response rates:

• Keep questionnaires concise and focused
• Clearly communicate the purpose and importance of the study
• Send reminders to non-respondents
• Consider incentives for participation when appropriate
• Design user-friendly formats (whether paper or digital)

To minimize misinterpretation:

• Use clear, simple language
• Pilot test the questionnaire with representative participants
• Provide examples or definitions for potentially confusing terms
• Ensure consistent formatting and response options
• Consider cultural and linguistic factors in question wording

Understanding these advantages and limitations helps researchers make informed decisions about when questionnaires are most appropriate and how to design them effectively to maximize their strengths while mitigating their weaknesses.

7. Aligning Questionnaire Design with Research Objectives

A well-crafted questionnaire begins with a clear understanding of what you want to achieve. Every question should serve a purpose and directly contribute to answering your research questions. Here’s a structured approach that flows naturally from your research objectives:

7.1 Define Your Research Objectives

Start by clearly identifying what you need to learn. Ask yourself which key issues or questions your study must address. Once you’ve outlined these goals, prioritize them so that you focus on the most important areas. This initial step ensures that you have a strong foundation for designing your questionnaire.

Example: If your research aims to improve a healthcare service, your objectives might include:

• Assess patient satisfaction with current services
• Identify barriers to accessing care
• Determine which aspects of service most need improvement
• Understand patient preferences for service delivery

7.2 Map Objectives to Questions

After setting your research objectives, create a plan to link each objective with specific questionnaire items. Consider making a simple list or table where each research aim is paired with one or more potential questions. This item mapping process helps you verify that every question is relevant and necessary, while also revealing any areas where additional questions might be needed.

Example Mapping:

• Objective: Assess patient satisfaction
  • Questions: “How would you rate your overall experience?” “Would you recommend our service to others?”
• Objective: Identify access barriers
  • Questions: “How long did you wait for an appointment?” “What challenges did you face in reaching our facility?”

7.3 Develop a Logical Flow

With your objectives and potential questions outlined, think about how to organize your questionnaire to flow naturally. Group related questions together and arrange them in an order that feels logical, often starting with broader topics before moving into more detailed areas. For instance, begin with general background questions before addressing sensitive topics later in the survey. This sequencing not only builds rapport with respondents but also helps ensure that earlier answers inform later questions where appropriate.

7.4 Involve Peers in the Review Process

Before finalizing your questionnaire, share your draft with colleagues or experts in your field. Discuss how each question aligns with your research objectives and whether the flow of topics makes sense. Feedback from peers can help you identify unclear or redundant questions and offer suggestions for improvement. This collaborative step is important for refining your design.

7.5 Pilot Test and Refine

Conducting a pilot test with a small, representative group of respondents can provide valuable insights into how well your questionnaire works in practice. Observe whether participants understand the questions as intended and whether their responses yield the information you need. Use this feedback to fine-tune the wording, adjust the order of questions, or even remove items that don’t add value.

Pilot testing checklist:

• Are the instructions clear?
• Do respondents understand all questions as intended?
• Are any questions consistently skipped or answered inconsistently?
• How long does it take to complete the questionnaire?
• Do the responses provide the data needed to address your research objectives?

7.6 Maintain an Ongoing Review

Finally, recognize that aligning your questionnaire with research objectives is an ongoing process. Even after your initial design and testing phases, be prepared to review and update your questionnaire as needed, especially if your research focus shifts or if new insights emerge during data collection. Keeping a checklist that ties each question back to a specific research objective can help ensure that your tool remains focused and effective throughout the study. This is particularly important for ongoing longitudinal studies with open enrollment and follow up, like the International Epidemiology Databases to Evaluate AIDS (IeDEA) project.

By following these steps, you can create a questionnaire that effectively gathers the information you need while minimizing unnecessary questions that might burden respondents and complicate analysis.

8. Conclusion

Designing an effective questionnaire is essential for collecting reliable and relevant data that aligns with your study’s objectives, ultimately achieving impactful research outcomes. Throughout this guide, we have explored the fundamental aspects of questionnaire design, from understanding basic principles to implementing practical strategies.

Key takeaways from this guide include:

• Purpose-driven design: Every element of your questionnaire should serve your research objectives. By carefully mapping questions to objectives, you ensure that the data collected will be relevant and useful.
• Respondent-centered approach: Consider your participants’ characteristics, comfort, and engagement when designing questions. Clear language, logical flow, and appropriate question types enhance response quality and completion rates.
• Balance of structure and flexibility: Different question types and questionnaire formats offer varying advantages. Choose the approach that best serves your research needs while considering resource constraints.
• Ethical considerations: Respect for participants’ privacy, autonomy, and time is not just an ethical imperative but also improves the quality and reliability of your data.
• Iterative refinement: Questionnaire design benefits from collaboration, pilot testing, and ongoing review. Be prepared to adapt your design based on feedback and preliminary results.

While questionnaires excel in large-scale data collection, they may benefit from being paired with qualitative methods such as interviews or focus groups to address their limitations in capturing depth and context. This mixed-methods approach can provide both breadth and richness to your research findings.

Next steps for novice questionnaire designers:

1. Practice with small projects: Before launching a major study, develop your skills by creating questionnaires for smaller, lower-stakes projects.
2. Seek mentorship: Connect with experienced researchers who can review your designs and offer guidance based on their practical experience.
3. Study exemplars: Examine well-designed questionnaires in your field to understand how established researchers structure their instruments.
4. Stay current: Research methodologies evolve, so continue to learn about best practices in questionnaire design through current literature.

Remember that questionnaire design is both a science and an art. While following the principles outlined in this guide will provide a strong foundation, developing expertise comes through practice, reflection, and continuous improvement. With time and experience, you will develop an intuitive sense for crafting questions that effectively capture the information you need while providing an engaging experience for your respondents.

References

1. Floyd J. Fowler, Jr. Survey research methods. SAGE Publications; 2014.

2. Dillman, D. A., Smyth, J. D., Christian, L. M. Internet, phone, mail, and mixed mode surveys: The tailored design method. John Wiley & Sons, Inc; 2014.

3. John W. Creswell. Research Design: Qualitative, Quantitative and Mixed Methods Approaches. SAGE Publications; 2014.

4. Alan Bryman. Social Research Method. Oxford University Press; 2016.

5. Michael Quinn Patton. Qualitative Research & Evaluation Methods. SAGE Publications; 2015.

 6. Groves, R.M., et al. Survey Methodology.2009

7. McCoy, L. P. Advantages and Disupsides of Surveys. Academic Journal of Public Health, 7(2), 1-5. 2008.

8. Bradburn, N. M., Sudman, S., & Wansink, B. (2004). Answering Questions: Methodology for Determining Cognitive and Communicative Processes in Survey Research.

The post How to Design Effective Questionnaires: A Complete Beginner’s Guide – Part 2 appeared first on CRENC Learn.

In this blog, we’ll explore key elements to consider when designing an effective questionnaire, guiding you through the process—from selecting the right question types to enhancing clarity and relevance. By the end, you’ll be able to create a suitable questionnaire that meets your research objectives and design requirements.

1. What exactly is a questionnaire?

A questionnaire is a structured set of well-thought-out questions designed to collect information from respondents about a specific research topic. In some contexts, such as clinical research, questionnaires may be referred to as Case Report Forms (CRFs). Questionnaires can be administered in various formats, including paper-based, online, and face-to-face. No matter the format, their purpose is to collect data that can be analyzed to draw conclusions, test hypotheses, or inform decision-making. Designing an effective questionnaire is crucial for obtaining reliable and valid data that will enhance the overall quality of your research.

2. Why you should design an effective questionnaire

By carefully considering your questionnaire design, you add significant value not only to your final research output but also to the broader research community. A well-designed questionnaire leads to better data, more reliable conclusions, and a smoother research process. Here are several key reasons to invest time and thought into getting your questionnaire design right:

The language should be clear and simple. A well-structured questionnaire avoids ambiguity by using simple, direct language, which enhances data accuracy, consistency, completeness, reliability and validity by reducing response bias from the participants. The more participants understand your questions, the more they can provide complete and reliable responses. Design your questionnaire with the participant in mind, consider their level of education, language proficiency and preferences, and sociocultural contexts. A good questionnaire should therefore be sufficiently comprehensible to the participant. One way to ensure this is to pre-test earlier versions with several typical model participants and adjust accordingly. Also consider engaging colleagues or experts who understand the context in the development phase of the questionnaire design.

Design for Purpose: A good questionnaire aligns with the goals of your study, ensuring that all responses are purposeful, relevant and meaningful. Avoid trying to capture too much information that might not be relevant in answering your research question(s). By staying focused on your research goals, you save your time but also, the participants’ time and avoid poor questionnaire completion rates.

Strive to engage the participant: A well-organized questionnaire captures the respondent’s interest, leading to more thoughtful answers and detailed responses. The structure, phrasing and flow of your questionnaire design should be such that the participant finds it interesting and engaging enough to follow through till the end. So, a well-designed questionnaire will anticipate questions that can disrupt rapport with the participant. For example, push sensitive questions like sexual behaviour and finances to the end, and bring forth less triggering questions ahead.

Make statistical analysis easier: Well-designed questionnaires have structured questions leading to organized answers, simplifying the statistical analysis to draw clearer conclusions. By collecting data in the appropriate format, you ensure that the data are suitable for applying the suitable statistical analysis methods. Involve a statistician in the early stages of the design using a well statistical analysis plan.

Example: If you want to measure attitudes or opinions, using a Likert scale (e.g., “Strongly Disagree” to “Strongly Agree”) allows you to quantify responses. For instance, a question like “How satisfied are you with the training?” with a 5-point Likert scale (1 = Very Dissatisfied to 5 = Very Satisfied) generates numerical data that can be easily analyzed using statistical methods like mean, median, or regression analysis (linear for continuous variables and you can categorize the responses to perform logistic regression).

Uphold Ethical standards always: Design your question with ethical considerations in mind. In study procedures ensure all responses are preceded by informed consent and that the questions do not violate ethical principles. Therefore, a good questionnaire should ensure the respect of participants’ confidentiality, privacy and autonomy; which builds trust with respondents and ensures compliance with ethical research standards.

A good questionnaire becomes a model for future studies: If you design your questionnaire effectively with validity and reliability, it can serve as a model for future studies who need a similar tool to assess the aspect of interest. Other researchers can refine your questionnaire and build on its successes, enhancing the quality of subsequent research efforts for the production of new data.

3. Essential Parts of a Good Questionnaire

A questionnaire is made up of three key parts:

3.1 The Introduction:

This is a brief explanation of the study’s purpose and objectives. It gives instructions on how to complete the questionnaire and information about confidentiality and how the data will be used. A consent statement is necessary.
Example: “Thank you for participating in our research on healthcare accessibility. This survey will take approximately 15 minutes to complete. Your responses will remain confidential and will be used only for research purposes. By proceeding, you consent to participate in this study.”

3.2 Demographic questions:

These questions are designed to gather information about the respondent’s background. Sociodemographic questions are typically asked in a straightforward and non-intrusive manner to ensure respondents feel comfortable providing accurate information. Here’s how they are commonly asked:

Age: Can be asked in ranges (e.g., “What is your age group? 18-24, 25-34, 35-44, etc.”) to avoid making respondents uncomfortable about disclosing their exact age.

Gender: Typically includes options like “Male,” and “Female,” in our context.

Education Level: Presented as a multiple-choice question with options such as “No education,” “Primary,” “Secondary,” and “University”.

Employment status: Often asked with predefined categories (e.g., “Student,” “Employed Full-Time,” “Self-Employed,” “Unemployed,” “Retired,” etc.).

3.3 Content questions:

There are 3 types of questions: closed–ended questions where the respondents choose from a list of predetermined options (Examples: multiple-choice, Likert scale questions), open-ended questions where respondents provide answers in their own words, and a combination of both formats to capture a wide range of information while maintaining some consistency in responses.

4. Types of Questionnaires

Questionnaires can be classified into different types based on several characteristics. Understanding these different types will help you choose the most appropriate format for your research needs. Let’s explore the main categories:

4.1 Questionnaire Format Types

4.2 Questionnaire Administration Methods

4.3 Questionnaire Question Types

By understanding these different questionnaire types, you can select the most appropriate approach based on your research objectives, target population, and available resources.

Now that you’ve got the basics of designing effective questionnaires under your belt—from understanding their purpose to mastering question types like open-ended and multiple-choice—you’re well on your way to collecting meaningful data. But how do questionnaires stack up against other tools like interviews or surveys? Curious to find out? Don’t miss How to Design Effective Questionnaires: A Complete Beginner’s Guide – Part 2 , where we’ll compare methods and tie it all together with practical insights.

END OF PART 1

>>> CONTINUE READING PART 2

————————————————-

References

1. Floyd J. Fowler, Jr. Survey research methods. SAGE Publications; 2014.

2. Dillman, D. A., Smyth, J. D., Christian, L. M. Internet, phone, mail, and mixed mode surveys: The tailored design method. John Wiley & Sons, Inc; 2014.

3. John W. Creswell. Research Design: Qualitative, Quantitative and Mixed Methods Approaches. SAGE Publications; 2014.

4. Alan Bryman. Social Research Method. Oxford University Press; 2016.

5. Michael Quinn Patton. Qualitative Research & Evaluation Methods. SAGE Publications; 2015.

 6. Groves, R.M., et al. Survey Methodology.2009

7. McCoy, L. P. Advantages and Disupsides of Surveys. Academic Journal of Public Health, 7(2), 1-5. 2008.

8. Bradburn, N. M., Sudman, S., & Wansink, B. (2004). Answering Questions: Methodology for Determining Cognitive and Communicative Processes in Survey Research.

The post How to Design Effective Questionnaires: A Complete Beginner’s Guide – Part 1 appeared first on CRENC Learn.

Developing a research question can be a challenging task, especially when you are new to the field of research. You may have too many ideas, or none at all. You may feel overwhelmed by the scope, complexity, or novelty of the research methods. You may wonder if your question is interesting, relevant, original, or feasible.

Selecting an appropriate research question is however critical to the success of your project, even though it may seldom be explicitly stated in the final paper. First of all, choosing the right research question can help you avoid frustration and disappointment. If you pick a topic or a problem that is too broad, too narrow, too complex or too simple, you may encounter difficulties in finding relevant sources, defining your research objectives, designing your methodology or presenting your findings. Secondly, developing the right research question can help you achieve your academic or professional goals more effectively. It guides you on choosing the appropriate research design and methods. You should develop a question that aligns with your objectives and expectations, thus maintaining your focus. You should also consider the feasibility, relevance and contribution of your question to your field of study or practice. Finally, developing a good research question can help you enjoy the research process and learn new things. If you are passionate about your question, you will develop more curiosity, creativity and persistence in exploring it. You will find research more rewarding and satisfying if you choose a question that sparks your interest and challenges your skills.

In this blog post, we will provide you with some tips on how to develop a research question that is interesting, relevant, feasible, and most importantly, that works for you. We will also provide you with some examples of effective research questions. We will cover the following aspects in detail:

• How to brainstorm and identify research needs
• How to conduct a literature review and identify gaps
• How to formulate a research question
• How to evaluate and refine your ideas

By the end of this blog post, you should have a better understanding of how to develop a research question that will help you achieve your research goals.

2. Steps to develop a good research question

21. Identify your field of interest and pick a topic that interests you.

i. Start with Curiosity

• Start with what you’re curious about. What topics or questions interest you? What are the gaps in this field of study or unresolved issues? What field or fact do you want to explore? For example, you may be curious about the impact of nutrition on maternal and child health in Cameroon. You may want to solve the problem of malnutrition and its consequences among pregnant women and children under five.

ii. Exploration Techniques

Mind Mapping Technique:

• Write your general topic in the center of a page
• Branch out with related concepts, questions, and sub-topics
• Look for connections between branches that might suggest unique angles

The Five Ws Approach:

• Who is affected by this topic/problem?
• What exactly is the issue or phenomenon?
• When does this occur (historically, seasonally, etc.)?
• Where is this most relevant geographically or contextually?
• Why does this matter to the field and broader society?

iii. Journal Browsing:

• Explore different sources of information. Read books, articles, blogs, podcasts, videos, etc. that spark your interest. See what other researchers are doing or have done and what gaps or opportunities they identify, such as the lack of data, the need for more evaluation, the potential for innovation, or the importance of context-specific solutions. You will usually identify these aspects under the discussion, limits/limitations sections in most research papers. Exploring these sources, a particular theme or field might feel inspiring to you or may capture your curiosity

iv. Focus and Feasibility

• Narrow down your focus. Once you have a general idea of your field of interest, try to define it more specifically. What aspect or angle do you want to explore? What can be the main question or goal of your research? What are the sub-questions or objectives that support it?
• Consider the feasibility and relevance of your research. How realistic is your research question? Do you have the time, resources, and skills to conduct it? How will your research contribute to the existing knowledge or practice in your field? How will it benefit you or others?

2.2 Review the existing literature to understand the background and find the gaps!

This step is crucial to the success of any research project. You should always review the existing literature and find the gaps in the knowledge that your study can address. A literature review is not just a summary of what has already been done, but a critical analysis of how your research question fits into the current state of knowledge. It contextualizes your research question. Here are some ways on how you can conduct a literature review and identify the gaps in the knowledge:

• Start Broad, Then Narrow: Make a broad search of your topic and narrow it down as you go. Use keywords, and apply them in databases, journals, books, and other sources that are relevant to your field and topic. For example, search databases such as Google Scholar, PubMed, Scopus, Web of Science, AJOL, etc.
• Example Search Process: Suppose your research idea is about investigating the impact of climate change on infectious diseases in Africa. You may start by using keywords such as “climate change”, “infectious diseases”, and “Africa” to search for relevant sources on Google Scholar or PubMed. You may then narrow down your search by adding more specific keywords, such as “malaria”, “dengue”, “cholera”, or “vector-borne diseases”. You may also filter your search by publication date, language, or type of source.

Digital Tools to Enhance Your Literature Review

Quality Assessment

• Evaluate the quality and credibility of the sources you find. Check the author’s credentials, the publication date, the publication journal, the methodology and the results. Avoid sources that are outdated, biased, or unsupported by evidence. For example, you may check the author’s affiliation, qualifications, and publications to see if they are experts in the field of your topic. You may also check the publication date to see if the source is current and up-to-date. Also ensure to check the publication journal to see if it is peer-reviewed, reputable, and has a high impact factor. You may also check the methodology and the results to see if they are valid, reliable, and relevant to your research question.

Organization and Analysis

• Organize your literature review into themes or categories that reflect the main aspects of your research question. For each theme or category, summarize the main findings, compare and contrast different perspectives, and highlight the strengths and weaknesses of the existing literature. This can easily be done with the use of referencing managers such as Zotero or Mendeley.
• Identify areas where further exploration is needed or that previous research studies have not addressed. These can be theoretical, methodological, empirical, or practical gaps or challenges that have not been adequately explored or resolved. Reflect on which research can fill these gaps or challenge these assumptions and contribute to the advancement of knowledge in that field.

2.3. Brainstorm some possible research questions that address the gaps!

Here are some tips to help you out brainstorming on possible research questions after your thorough literature review and exploration:

i. Initial Question Development

• Start by stating a broad research question that you think can help filling the gap you noticed while doing the literature search. For example, you may have noticed that there is a lack of studies on the mental health impacts of the COVID-19 pandemic in Africa, especially among vulnerable groups such as refugees, internally displaced persons (IDPs) or HIV patients. Therefore, you may state a broad research question like this: Does the COVID-19 pandemic affects the mental health of vulnerable populations in Africa?
• Do some preliminary research on your research question to see what has been done before and what are the current issues or debates. You can use online search strategies with keywords. For example, you may use keywords such as “COVID-19”, “mental health”, “refugees”, “IDPs”, “PLHIV” and “Africa” to search for relevant articles, reports, and websites. You may find some sources that provide background information, statistics, and evidence on the topic.

ii. Advanced Brainstorming Techniques

The Contrarian Approach:

• Take existing research conclusions and ask “What if the opposite is true?”
• Examine underlying assumptions in the field and question them
• Look for contexts where established principles might not apply

Interdisciplinary Integration:

• Identify concepts from other disciplines that could inform your field
• Consider methodologies from different fields that could offer new insights
• Look for parallel problems in other domains that might have transferable solutions

Problem-Solution Mapping:

• List all the problems or challenges identified in your literature review
• For each problem, brainstorm potential solutions or approaches
• Consider which problems align with your interests and expertise
• Develop questions that examine the efficacy of your proposed solutions

iii. Formulate specific and focused questions that address the gaps that you identified.

• These questions should be clear, concise, and answerable with data. They should also be relevant and meaningful to your field of interest and your audience. You may formulate some specific and focused questions like: What are the most common and severe mental health disorders among IDPs in Douala Cameroon, and how do they vary by age, gender, and location? Or What are the barriers and facilitators to accessing and utilizing mental health services and resources for refugees, IDPs and PLHIV in Africa, and how can they be addressed or enhanced?

These questions are not yet refined but they help you enumerate and focus on the aspects you would like to explore.

2.4 Select the Most Suitable Research Question and Make it SMART

You may now want to refine your research question to a specific issue, population, or context by revising, expanding, or narrowing it based on your research findings. You can use different strategies to refine your research question, such as adding or removing variables, specifying the population or context, comparing, or contrasting different aspects, or using different types of questions (such as descriptive, explanatory, evaluative or predictive). When applicable, your research question should answer all or some of the following key questions: “What”, “how”, “why”, “who/where/when”.

The PICOTS framework is widely recommended in medical and epidemiological research for defining research questions. PICO stands for Population P, Intervention I, Comparison C, Outcome O, Time T and Setting S.

Population: It also includes the patient or the problem of interest. In order to define this aspect of PICOTS, you can ask yourself the following questions:

• Who is the population of interest of my study?
• Who do I define as a patient?
• What are the characteristics of the population (age, gender, health condition…)?
• What are the eligibility criteria for participants (age, gender, health condition…)?

Intervention or the exposure:

• What is the intervention, exposure or factor being studied? Interventions such as drug therapy, behavioral therapy, diagnostic tests, etc or if there is no intervention, what is the exposure (smoking, alcohol intake…)?

Comparison

• What is the alternative or control group that is compared with the intervention? For example, in clinical trials we usually compare treatment groups with Placebo groups or Gold and standard treatments. Similarly, we can compare groups of people exposed to a particular factor with unexposed groups.

Outcome

• What result/outcome are you measuring? How will you measure it? It can be a clinical or epidemiological outcomes such as prevalence, incidence, mortality, quality of life, etc.

Time

• When will the study take place? What period? Will there be a follow up period?

Setting

• Where will the study take place? A whole country, a specific city, hospital or facility? You should be specific about the setting of your study.

It is also important to mention that this framework varies with the type of research question (such as descriptive, explanatory, evaluative or predictive). The table below summarizes the application of the PICO components per question type.

Table. PICO by type of Research Question

Source: Formulating the Research Question PICO Framework, CRENC (2025)

For example, we have a broad causal research question: What are the effects of antidepressants on the anxiety levels in Yaounde?

Breakdown of the PICOTS components:

• P (Population): Individuals with anxiety disorders in Yaoundé
• I (Intervention): Antidepressant treatment (e.g., SSRIs, SNRIs)
• C (Comparison/Control): No treatment or an alternative treatment (e.g., psychotherapy, placebo)
• O (Outcome): Change in anxiety levels (measured via standardized scales, e.g., GAD-7, HAM-A)
• T (Time): Over a 12-week period (or another appropriate time frame)
• S (Setting): Clinical and outpatients in Jamot Hospital in Yaounde

Now the refined question will look like this:

“In individuals diagnosed with anxiety disorders in Yaoundé, how does treatment with antidepressants (e.g., SSRIs or SNRIs) compared to no treatment or psychotherapy affect anxiety levels over a 12-week period in clinical and outpatients of Jamot hospital?”

This version makes the research question more specific, measurable, and applicable to research design, relevant and is additionally time bound (SMART).

2.5 Evaluate the feasibility, originality or novelty and significance of your research questions

When you will have a list of research questions you want to pursue, you will ask yourself “how do I decide which ones are worth your time and effort?” The following points will help you evaluate the feasibility, originality and significance of your research questions.

• Feasibility: Suppose you want to research the effectiveness of a mobile app for improving the adherence to antiretroviral therapy (ART) among people living with HIV in Cameroon. To answer this question, you need to consider the feasibility of your research project. Can you realistically answer the question with the resources, data skills and time that you have? Do you have access to the data, methods and tools that you need? Are there any ethical, legal or practical barriers that might prevent you from conducting the research?
• Originality: Is the question new or novel in your field of study? Does it address a gap in the existing literature or challenge a dominant paradigm? Does it offer a new perspective or a different angle on a familiar topic? For example: How does exposure to air pollution affect the cognitive development of children in urban areas? This question is original and novel because it explores a relatively under-researched topic in public health, namely the impact of air pollution on cognitive outcomes. It addresses a gap in the existing literature by focusing on a specific population (children in urban areas) and a specific outcome (cognitive development) that have not been extensively studied in relation to air pollution.
• Significance: Does the question matter to your discipline, society or institution? Does it have implications for theory, policy or practice? Does it contribute to the advancement of knowledge or the solution of a problem? For example, what are the effects of community-based interventions on the prevention and control of malaria in sub-Saharan Africa? This question is significant because it matters to the discipline of public health, as malaria is one of the leading causes of morbidity and mortality in the region. It contributes to the advancement of knowledge by providing evidence-based recommendations for the optimal design and implementation of community-based interventions for malaria prevention and control.

Note that your research question should be ethical, meaning that it does not harm or exploit any individuals or groups involved in your research.

To narrow down your list of research questions, you can use these criteria to rank them from high to low priority. You can also ask for feedback from your peers, mentors or supervisors to get their opinions and suggestions. Remember that you can always revise your research question as you go along.

• Test Your Question

Like mentioned above, your final research question must be specific, measurable, achievable, relevant and time bounded.  You must ask yourself whether it can be answered using research methods.

3. Conclusion

Choosing a research question is one of the most important and challenging steps in any research project. It requires curiosity, creativity, and critical thinking. A good research question should be feasible, original, significant, and ethical. It should also reflect your own interests and goals, as well as the needs and expectations of your field. To choose a research question, you should start with a broad topic, review the existing literature, identify the gaps or challenges, and formulate specific and focused questions. You should also seek feedback from your peers, mentors, or supervisors, and be ready to revise your question as you progress with your research. By following these directories, you will be able to find a research question that is both interesting and meaningful to you and your field of study.

References

How to Write About Your Research Interests. Accepted Admissions Blog (2023, December 8). https://blog.accepted.com/writing-about-research-interests/

Brainstorming and Deciding On Questions—Mann Learning Technologies Committee—Dashboard. (n.d.). Retrieved January 24, 2024, from https://confluence.cornell.edu/display/mannltc/Brainstorming+and+Deciding+On+Questions

CRENC. (2025, January 30). Formulating the research question PICO Framework [Video]. YouTube. https://youtu.be/rkfaR3xmaTM

DeCarlo, M. (2018). 8.5 Feasibility and importance. https://pressbooks.pub/scientificinquiryinsocialwork/chapter/8-5-feasibility-and-importance/

How do I identify a research gap during the literature review? (2021, January 29). Editage Insights. https://www.editage.com/insights/how-do-i-identify-a-research-gap-during-the-literature-review

Identify Your Research Interests | Undergraduate Research | University of Arizona. (n.d.). Retrieved January 24, 2024, from https://ur.arizona.edu/content/identify-your-research-interests

McCombes, S. (2022, October 30). 10 Research Question Examples to Guide Your Research Project. Scribbr. https://www.scribbr.com/research-process/research-question-examples/

McCombes, S. (2023, January 2). How to Write a Literature Review | Guide, Examples, & Templates. Scribbr. https://www.scribbr.com/dissertation/literature-review/

Researching Programs: Profiling Your Research Interests—Purdue OWL®—Purdue University. (n.d.). Retrieved January 24, 2024, from https://owl.purdue.edu/owl/general_writing/graduate_school_applications/graduate_school_applications_researching_programs/research_programs_profiling_your_research_interests.html

The post How to Develop a Research Question appeared first on CRENC Learn.

This guide is meant to demystify the p-value, breaking down its purpose, how to interpret it, and its advantages and limits. By the end, you should feel confident in your ability to not only understand the p-value but also to use it effectively in your research.

So, what is the p-value?

Essentially, the p-value helps us answer one important question: If there were no true effect (i.e., the null hypothesis is true), how likely is it that I would see the effect observed in my study just by chance? Think of it this way: Suppose you’re studying whether a new drug improves patient outcomes compared to the standard treatment. If there truly were no difference between the two (this is what we call the null hypothesis), the p-value tells you how likely it is that the differences you observe in your study are just random noise. A smaller p-value means it’s less likely that what you observed happened by chance, suggesting there’s a genuine effect.  

Here’s a more formal definition: The p-value is the probability that the observed data would occur if the null hypothesis were true. The lower the p-value, the less compatible your data are with the null hypothesis.

Understanding the Null Hypothesis Significance Testing (NHST)

To understand the p-value better, you need to understand what a hypothesis is and what a hypothesis test is all about. In research, a “Hypothesis is a formal and testable statement or claim about the relationship between variables (or observable phenomenon). For example, we can hypothesize that there are more women in the world than men. This is a claim that is testable.

Null Hypothesis Significance Testing (NHST), also known as Statistical Hypothesis Testing (SHT), is a fundamental approach in statistical inference used to make decisions about parameters in the broader population based on a sample of data from it. This method involves formulating and testing two competing hypotheses: the null hypothesis and the alternative hypothesis.

• Null Hypothesis (H₀): It is a statement of no effect, no difference, or no association or no change in your study, —think of it as the “status quo”.  For example, if you’re testing whether a new treatment is better than the current one, the null hypothesis would claim they are equally effective.  It is the hypothesis that researchers try to reject or disprove. It is statistically tested under the assumption that it is true.
• Alternative Hypothesis (H₁): This is the opposite of the null hypothesis. In our example, this would claim the new treatment is better than the current one. It typically represents the research hypothesis or the effect that researchers are looking for. It is accepted if there is sufficient evidence to reject the null hypothesis.

Your goal in NHST is to collect data and use statistical tests to decide whether to reject the null hypothesis in favor of the alternative. If your data provide enough evidence (i.e., if the p-value is low enough), you reject the null hypothesis and accept that the effect is real.

How Do You Use the P-value in Hypothesis Testing?

Let’s walk through an example. Imagine you’re investigating whether smoking increases the risk of lung cancer.

• The null hypothesis (H₀) would be: Smoking is not associated with lung cancer.
• The alternative hypothesis (H₁) would be: Smoking is associated with lung cancer.

To test this hypothesis pair, collect data from your study participants, then you run a statistical test—let’s say a t-test—and the software calculates a p-value for you. If the p-value is small (commonly set at a threshold of 0.05), it suggests that the observed association between smoking and lung cancer is unlikely to be due to chance alone. You would then reject the null hypothesis, concluding that smoking is indeed associated with lung cancer. However, if the p-value is greater than 0.05, you wouldn’t have enough evidence to reject the null hypothesis. This doesn’t necessarily mean the null hypothesis is true—it just means your data aren’t strong enough to say otherwise. Findings yielding p-values smaller than the cut-off value (<0.05) are typically considered as ‘statistically significant’, while findings with p-values equal to or larger than the cutoff are described as ‘non-statistically significant’.

Procedure of NHST

The NHST process typically involves these steps:

1. Clearly formulate the null and alternative hypotheses, based on the research question.
2. Set Type I error rate  (α or critical value), which defines the probability of rejecting the null hypothesis when it is actually true (False Positive).
3. Choose a significance level (commonly α = 0.05 or 0.01), which represents the threshold for determining statistical significance. The smaller the α, the stricter the test, reducing the chance of a Type I error but increasing the chance of a Type II error.
4. Choose an appropriate statistical test, based on your study design, data type, and research question. Be mindful of test assumptions (e.g., normal distribution for t-tests) and select statistical software to compute the test. Establish your decision rule, considering whether to use a one-tailed or two-tailed test depending on your hypothesis.
5. Then compute the test statistic and p-value, ensuring proper data handling and cleaning processes are followed beforehand. .
6. Compare the p-value to the significance level and make a decision based on the rule for rejecting H0:

If p ≤ α, reject the null hypothesis and accept the alternative hypothesis.

If p > α, fail to reject the null hypothesis.

Remember, failing to reject does not prove the null hypothesis is true—it only suggests the data do not strongly contradict it. Also, consider the practical significance of your results alongside the statistical significance. Also, interpret the results in the context of the research question.

Interpreting P-values in Research

The conventional threshold for p-values in most fields is 0.05. This means that if your p-value is less than or equal to 0.05, the results are considered statistically significant. In other words, you’re saying there’s only a 5% (or lower) chance that you would see these results if the null hypothesis were true.

• A p-value of 0.05 means there’s a 5% probability of observing your data, or something more extreme, if the null hypothesis is true.
• A p-value of 0.01 means there’s only a 1% chance of seeing such extreme data under the null hypothesis.

On the other hand, if the p-value is greater than 0.05, it suggests that the observed results are not unusual under the null hypothesis. This doesn’t mean the null hypothesis is true, but it does mean that you lack strong evidence to reject it.

How to Report P-values

There are a couple of ways to report p-values in your research:

1. General reporting: You could report it as, for example, “p > 0.05” or “p < 0.01.”
2. Exact reporting: You can also report the exact p-value, at 2 or 3 decimal places, like “p = 0.023.” If the p-value is less than 0.001, it’s often conventional to simply write “p < 0.001.”

Figure 1 below depicts one way to present p-values. P-values are denoted with asterisks and the corresponding threshold added to the footer. In simple terms, this manuscript presents p-values not in the table but add asterisks to the corresponding Odd ratios (95%CI) and specify in the legend of the table the p-value thresholds that are statistically significant according to the baseline characteristics, helping readers understand the strength of the evidence without adding many numbers in the main table.

Figure 01: The results section of a manuscript (Predictors and Consequences of HIV Status Disclosure in Adolescents living with HIV in Eastern Cape, South Africa: A Prospective Cohort Study) presenting p-values significance as footer notes

Meanwhile, in figure 2, P-values are presented directly after the OR and the 95%CI. It’s one of the most common ways of presenting it for the reader to directly visualize all the statistical tests performed per baseline characteristics and obverse their significance across the row. Here, the statistical significance for each variable is observed on the same line.

Figure 02: The results section of a manuscript (Hypertension among People living with HIV/AIDS in Cameroon: A Cross-sectional Analysis from Central Africa International Epidemiology Databases to Evaluate AIDS) presenting p-values significance in the main table.

Advantages of using the p-value in medical research

Why are p-values so popular in research? Here are some reasons:

• Simplicity: It can be easily interpreted and computed.
• Decision Making tool: It helps determine whether the results are likely due to chance, guiding researchers on whether to pursue a hypothesis further..
• Versatility: P-values can be used with a wide range of statistical tests and models, making them applicable across different research designs.
• The p-value provides a continuous measure of evidence against the null hypothesis.: Unlike binary decision-making (e.g., reject or fail to reject the null hypothesis), the p-value gives a range of evidence. Example: A p-value of 0.001 suggests stronger evidence against the null hypothesis than a p-value of 0.04. This continuous scale allows researchers to gauge the strength of the evidence instead of making a strict yes/no decision.
• When used correctly, p-values help control the rate of false positive results (Type I errors) in research: If a researcher set the p-value threshold at 0.05, it means there’s a 5% chance of concluding there is an effect when there actually isn’t. This helps limit false positives, maintaining the integrity of scientific findings.

Limitations of P-value

Despite their usefulness, p-values are not without their shortcomings:

• A small p-value doesn’t “prove” the alternative hypothesis is true. It only indicates that the null hypothesis should be rejected and does not otherwise indicate for which reason the alternative hypothesis may be true.
• P-values are sensitive to sample size—a small sample can lead to a large p-value, even if there is a real effect.
• Researchers may misinterpret the p-value, thinking it tells the probability of the null hypothesis being true or false, which it does not. This is particularly the case when there is poor replication of p-values.
• Encourages p-hacking (Researchers may be tempted to manipulate data to achieve p < 0.05. This can lead to false positive results and publication bias).
•  

How to Address the Limitations of P-values

To get a more complete picture, you should complement p-values with other statistical measures. You should:

• Report effect sizes along with p-values is more meaningful than just knowing whether the effect exists or not.
•  Report confidence intervals, to give a range of plausible values for your estimate, helping you gauge the precision of your results.
• Consider bayesian methods, which offer an alternative approach, incorporating prior knowledge into the analysis.
• Consider practical or clinical significance, not just statistical significance.
• Pre-register studies (publishing protocols and data analysis plans) to prevent p-hacking st the data analysis stage.

Conclusion

The p-value remains a cornerstone of statistical inference in medical and public health research, offering a standardized measure to assess the strength of evidence against a null hypothesis. While it provides valuable insights, researchers must be aware of its limitations and potential for misinterpretation. By addressing the limitations of p-values through pre-registration of studies, transparent reporting, and a more holistic interpretation of results, researchers can improve the reliability and reproducibility of their work. To enhance the robustness of research findings, it’s crucial to complement p-values with effect sizes, confidence intervals, and, where appropriate, Bayesian methods. Additionally, researchers should focus on practical and clinical significance alongside statistical significance.

References

• Goodman S. A dirty dozen: twelve P-value misconceptions. Semin Hematol. 2008;45:135–140. 7.
• Lytsy P. P in the right place: Revisiting the evidential value of P-values. JEvidBasedMed.
• 2018;11:288291. https://doi.org/10.1111/jebm.12319
• Cohen J. The earth is round (p < .05). Am Psychol. 1994;49:997–1003. 8.
• Szucs D, Ioannidis JPA. When null hypothesis significance testing is unsuitable for research: a reassessment. Front Hum Neurosci. 2017;11:390
• Greenland S, Senn SJ, Rothman KJ, et al. Statistical tests, P values, confidence intervals, and power: a guide to misinterpretations. Eur J Epidemiol. 2016;31:337–350.
• Daniël Lakens, The Practical Alternative to the p Value Is the Correctly Used p Value. Perspectives on Psychological Science 2021 Sep 4;23(1):578. doi: 10.1186/s12879-023-08544-x. PMID: 37667182; PMCID: PMC10478445.
• Chén OY, Bodelet JS, Saraiva RG, Phan H, Di J, Nagels G, Schwantje T, Cao H, Gou J, Reinen JM, Xiong B, Zhi B, Wang X, de Vos M. The roles, challenges, and merits of the p-value. Patterns (N Y). 2023 Dec 8;4(12):100878. doi: 10.1016/j.patter.2023.100878. PMID: 38106615; PMCID: PMC10724370.
• Zhu W. p < 0.05, < 0.01, < 0.001,  < 0.0001, < 0.00001, < 0.000001, or < 0.0000001 …. J Sport Health Sci. 2016 Mar;5(1):77-79. doi: 10.1016/j.jshs.2016.01.019. Epub 2016 Jan 21. PMID: 30356881; PMCID: PMC6191982.
• Daniël Lakens, The Practical Alternative to the p-value Is the Correctly Used p-value. Perspectives on Psychological Science 2021

https://www.investopedia.com/terms/p/p-value.asp

The post The P-value: A Beginner’s Guide for Medical and Public Health Researchers appeared first on CRENC Learn.

What is Mortality?

Before we get into the details, let’s be clear on what we are talking about. There are different measures of mortality, and each offers a unique perspective on death rates and their implications, providing a more detailed understanding of public health challenges. Let’s define some key terms before we get technical:

• Death: Refers to the permanent disappearance of all evidence of life at any time after birth.
• Mortality: The average risk of a person dying during a time span
• Mortality Rate: The measure of the frequency of occurrence of death in a defined population during a specified interval.

Importance of Measuring Mortality in Public Health

As crucial metrics in public health, mortality measures gauge the overall health status of populations, assess the impact of diseases, and help in designing effective health policies. Specifically, they are:

• Useful in projecting future population size as demographers can predict future population changes, such as whether a population is likely to grow, shrink, or age. This information is vital for planning future healthcare needs, infrastructure, and social services.
• Some mortality measures like infant mortality rates and age-specific death rates, help identify population groups at high risks and in need of health services. This can tailor services and interventions to meet their specific needs, ensuring that those who need help the most receive it.
• It indicates the quality of life and life expectation at birth helping to gauge the effectiveness of public health efforts. For instance, a lower life expectancy at birth might indicate poor living conditions, inadequate healthcare, or high disease prevalence.
• Indicates priority areas for health actions, including the allocation of resources and health interventions. Making room for more strategic allocation of resources, directing funds, and interventions to where they will have the most significant impact.

Determinants or Factors Influencing Death Rates

Mortality is influenced by a complex mix of socioeconomic, environmental, healthcare-related, behavioural, biological, and social factors. Addressing these determinants through targeted public health interventions and policies is important for public health efforts aimed at reducing mortality rates and improving overall population health. The table below provides a summary of some factors that determine death rates.

Table 1: Summary of factors influencing death rates

Key Mortality Indicators

1. Crude Death Rate (CDR)

This is the total number of deaths in a calendar year per 1,000 people in a population. i.e The crude mortality rate is the mortality rate from all causes of death for a population. CDR provides a general overview of the mortality level within a population, making it useful for assessing overall health and comparing mortality rates over time within the same population. It can be calculated as follows:

• CDR is easy to calculate and provides a quick snapshot of mortality levels.
• One problem with CDR is that it does not consider the age, sex, or other demographic factors, due to which it is called a crude measure.
• Weakness for international comparisons as it makes no allowance for differential age and sex composition.

Worked Example

 The CDR for Cameroon for year 2019 may be computed as follows: Total number of deaths during the year 2019 = 70,000

Total population at the middle of the year i.e., midyear population during 2019 = 12,500,000 Therefore:

2. Cause Specific Mortality Rate

The cause-specific mortality rate is the mortality rate from a specified cause for a population (eg HIV, Cancer, accidents, NCDs etc). CSMR focuses on mortality from specific causes, such as diseases or accidents, which helps in identifying public health priorities and targeting interventions. Thus calculated as:

• Provides insights into specific health issues, allowing for targeted public health strategies.
• However, it does not provide a comprehensive view of overall mortality and may require detailed cause-of-death data, which is not always available.

Worked example:

10,000 deaths due to Disease A occurred in Town R (midyear population 100,000) in 2004. Therefore:

 3. Age-Specific Death Rate (ASDR)

Age-Specific Death Rates are mortality rates within specific age groups. ASDR provides insights into the mortality risk for different age groups, crucial for identifying vulnerable populations and targeting public health interventions where needed the most. It can be calculated as follows:

• It plays a vital role in understanding the health risks faced by different age groups, allowing for more effective age-targeted interventions.
• Though important in understanding health risks, it requires detailed age-specific data, which may not always be readily available.

Some specific types of age-specific mortality rates are neonatal, postneonatal, and infant mortality rate.

Worked Example

For the age group 0-4 years in 2019, there were 12,000 deaths following a cholera outbreak in the Littoral Region from a total population of about 2,000,000 people. The ASDR can be calculated as follows:

4. Infant Mortality Rate (IMR)

The Infant Mortality Rate (IMR) Defined as the number of infant deaths (under 1 year of age) occurring in an area within a specified calendar year per 1000 live births in the same community during the same calendar year

IMR is a key indicator of the overall health status and well-being of a population, reflecting the quality of healthcare, maternal health, and socioeconomic conditions.

Advantage: IMR is a sensitive indicator of both current and long-term public health conditions, guiding policy and resource allocation.

 Disadvantage: It focuses only on infant mortality, so it does not provide information about the health of other age groups or the population as a whole.

It can be calculated as follows:

Worked Example

200,000 live births were recorded in 2020 and 1,500 infant deaths reported in the same year. By using this formula, the IMR can be calculated as follows:

5. Neonatal Mortality Rate(NMR)

 NMR reflects the number of deaths among newborns (within the first 28 days of life) and is a crucial indicator of the quality of care provided during pregnancy, childbirth, and the immediate postnatal period. It helps in assessing the overall healthcare system’s capacity to care for newborns.

It is given by: The number of neonatal deaths (infants under 28 days of age) in a given year divided by the total number of live births in the same year, then multiplied by 1,000.

• NMR provides specific insights into early life mortality, which can guide targeted interventions to improve neonatal care and reduce preventable deaths.
• However, NMR only covers a narrow timeframe (the first 28 days), so it doesn’t capture mortality risks that occur later in infancy or childhood.

6. Post-Neonatal Mortality Rate (PNMR)

Measures the number of deaths among infants aged 28 days to under one year. This indicator helps in understanding the health risks that affect infants after the neonatal period, such as infections, malnutrition, or environmental hazards.

It is calculated as follows: The number of post-neonatal deaths (infants aged 28 days to under one year) each year is divided by the total number of live births in the same year, then multiplied by 1,000.

• PNMR allows for a focus on health risks and interventions needed after the neonatal period, contributing to better infant survival rates.
• Like NMR, it only focuses on a specific age group, missing broader health trends that might affect other age groups.

 7. Peri-Natal Death Rate (PNDR)

 PNDR encompasses both late fetal deaths (stillbirths) and early neonatal deaths (within the first seven days of life). This measure is critical for evaluating the effectiveness of maternal and newborn healthcare during late pregnancy and the immediate post-birth period.

It is given by the number of peri-natal deaths (late fetal deaths plus early neonatal deaths) in a given year divided by the total number of live births plus late fetal deaths, then multiplied by 1,000.

• PNDR provides a comprehensive view of mortality around the time of birth, highlighting the need for improved maternal and newborn care.
• PNDR can be difficult to calculate accurately in settings where data on stillbirths and early neonatal deaths are incomplete or unreliable.

Below is a flow diagram with time reference for mortality through infancy to help you understand the different time durations.

Figure 1: Time reference for mortality in Childhood and infancy

8. Maternal Mortality Ratio/Rate (MMR)

Maternal death is defined as, “the death of a woman while pregnant or within 42 days of termination of pregnancy irrespective of the duration and site of the pregnancy, from any causes related to, or aggravated by the pregnancy or its management, but not from accidental or incidental causes.”

• It is a primary indicator of the health status or QOL of a country or geographic area.
• MMR highlights the risk of death due to pregnancy or childbirth-related causes, indicating the quality of maternal healthcare and access to medical services.

Advantages: Provides critical data for assessing and improving maternal health services.

 Disadvantages: The data collection can be challenging, especially in areas with weak health information systems, and it only focuses on maternal deaths, not on other aspects of maternal health.

Note: For Maternal Mortality Rate, the numerator is a part of the denominator, Whereas, in Maternal Mortality Ratio, the numerator is not a part of the denominator.

9. Death-to-case ratio

 The death-to-case ratio is the number of deaths attributed to a particular disease during a specified time period divided by the number of new cases of that disease identified during the same time period.

It is important in its ability to help in understanding the lethality of a particular disease, which is crucial for assessing the severity of an outbreak or the effectiveness of treatment measures. It can be derived as follows:

Worked Example

 Between 1940 and 1949, a total of 143,497 incident cases of diphtheria were reported. During the same decade, 11,228 deaths were attributed to diphtheria. Calculate the death-to-case ratio.

• It is useful for tracking disease outbreaks and evaluating the effectiveness of healthcare responses.
• However, it can be affected by the accuracy of case reporting and does not account for differences in population sizes or characteristics thus one of its limitations.

10. Case-fatality rate(CFR)

Refers to the proportion of persons with a particular condition (cases) who die from that condition. CFR is a measure of the severity of a disease, showing the proportion of individuals diagnosed with a disease who die from it, which is vital for understanding the virulence and impact of an infectious disease.

It is given by the number of deaths due to a particular disease  divided by the number of incident cases of the disease, then multiplied by 100. As follows:

• It represents the ratio of death to cases
• Establish the virulence and killing power of a disease
• Useful in acute infectious diseases, like cholera, for quickly assessing the deadliness of an outbreak.
• It’s major disadvantage is the fact that the CFR can vary over time and may not reflect the long-term mortality risk if the disease has a long duration.

Worked Example

 In an epidemic of a cholera outbreak in Yaounde, 555 cases were identified and confirmed. Three of the case patients died as a result of their infections. Calculate the case-fatality rate. 

References

• Centres for Disease Control and Prevention (CDC) 2019. Age-Specific Death Rates for Selected Age Groups, by Race and Hispanic Origin and Sex: United States, 2019. Retrieved from [CDC](https://www.cdc.gov/nchs/data/nvsr/nvsr70/nvsr70-08-508.pdf).
• WHO. 2017. Mortality, Health Topics. Available online at: http://www.who.int/topics/mortality/en/
• United Nations. 2017. Death, Mortality, Demographic and Social Statistics. Available online at: https://unstats.un.org/unsd/demographic/sconcerns/mortality/mortmethods.htm
• Principles of Epidemiology in Public Health Practice; 3rd Edition An Introduction to Applied Epidemiology and Biostatistics, center for disease control, 2012.
• World Health Organisation. 2018. Maternal Mortality Ratio (per 100,000 live births), Health Statistics and Information Systems. Available online at: http://www.who.int/healthinfo/statistics/indmaternalmortality/en/

The post Understanding and Quantifying Measures of Mortality appeared first on CRENC Learn.

What is coercion and undue influence?

Coercion

The Belmont Report defines coercion as ‘occurring when an overt threat of harm is intentionally presented by one person to another to obtain compliance’.

Coercion also involves threats of something bad or unfair or withholding something participants have a right to if they do not act as desired. Therefore, if any research carried out involves even a statement to cause harm or inflict pain on the potential participant for compliance, the researcher or investigator is practicing coercion which must be avoided at all stages of the study.

Example 1: Imagine being a patient suffering from a potentially fatal disease who has been denied a medication that is the confirmed cure for your disease by your attending physician because he wants you to participate in the research study, he is conducting on patients of the same disease he is treating. This is blatantly coercive and should be avoided.

Example 2: A research investigator who threatens to pay a participant less than what was promised at the initial consent unless he continues into a distinct extension study although the participant has adhered to all requirements and stated conditions on receiving the full payment.

Undue Influence

Undue influence is defined as occurring through an offer of an excessive, unwarranted, inappropriate, or improper reward to obtain compliance from a participant. It can also be the offer of something good, such as money, and may have the effect of motivating people to enroll or remain in a study. Undue influence is centered around tempting a participant with something that will make him or her comply with the request of participating in the research. In this case, that participant will not comply with the research in the absence of what has been used to lure him/her to participate.

Example 1: If you are carrying out a study on severely poor individuals, a grocery voucher of 50,000 FCFA may be considered undue influence. This amount is way too enticing to compensate for participation especially because the study is carried out on economically disadvantaged population.

Example 2: If a physician offers special services or benefits to patients which are inappropriate with the intention of tempting the patients to enroll in his study without the patients evaluating the risk/benefit situation to them, it is considered undue influence. These services could include better hospital private wards and treatment, offering to pay the patients’ bills, and taking care of their needs throughout their hospital stay.

Differences between coercion and undue influence  

The table below describes the differences between coercion and undue influence.

How can Coercion and Undue Influence be prevented in Research?

Generally, it is recommended that when recruiting participants in clinical studies, the investigator should seek for consent under any circumstances that provide the prospective subject or the Legally Authorised Representative (LAR) sufficient time, information and opportunity to consider participation thereby, minimizing the possibility of coercion and undue influence. The following precautions are required when recruiting participants into studies;

• Ethical Review: Regular review by an ethical committee should be done. Therefore, the research protocol should be submitted to an IRB for approval before implementation. 
• Transparent Communication: Provide clear, jargon-free information about risks and benefits.
• Appropriate Compensation: Ensure fair compensation that doesn’t overly incentivize participation as it will lead to undue influence.
• Independent Consent Process: An independent party for consent to avoid influence.
• Researcher Training: Educate staff on ethical recruitment and voluntary participation. All staff involved in the study should undergo proper training on GCP and ethics.
• Voluntary Participation: Emphasize the right to refuse or withdraw without penalty. This point is very important because it clearly informs the potential participant of free consent and rights which keeps the study free from coercion and/or undue influence.  
• Careful Recruitment: Avoid targeting vulnerable populations; ensure voluntary participation. According to the International Conference for Harmonisation (ICH), these are a group of people whose willingness to participate may be unduly influenced by the expectation of benefit or fear of reprisal for noncompliance. These groups are; infants and very young children, adults with mental or emotional disorders, patients with dementia, unconscious patients and prisoners. We can also have many others.
• Monitoring and Feedback: Continuously monitor recruitment and consent by providing feedback channels. The researcher needs feedback at all intervals to keep informed of how the study is being done at the site. This helps to identify any research misconduct and malpractices as a whole.  
• Cultural Sensitivity: Tailor consent processes to cultural and social contexts. The investigator needs to be aware that cultural differences and similarities between people exist. Therefore, the consent processes should be made in a manner that will avoid coercion and /or undue influence on the participants.
• Distinguish Care from Research: Clearly differentiate between medical care and research participation. The purpose of clinical research is to generate knowledge which will be useful for patients in the future while medical care aims to promote the wellbeing of individual patients. 

Below are a few examples of vulnerable populations that if research should be conducted on these precautions should be considered:

1. Prisoners: This is a group of people that are considered vulnerable because their freedom has been grounded due to maybe a crime they have committed or something against the law. Therefore, enrolment of this group of people should not provide participants with advantages in terms of living conditions, quality of food, amenities, opportunities etc. because participants’ ability to evaluate the risks is minimal since they hunger for freedom.
2. People living with terminal diseases: They can be easily influenced because of their health situations and they might not clearly evaluate the risks that the study can expose them to. So, the informed consent should clearly state that whether or not they agree to participate, it will not in any way affect their treatment (medication) and routine appointments. It is advised that even if consent is obtained from them, their legal representative should also give his/her consent to ensure that the study has been thoroughly assessed by both parties.
3. Elderly patients in group healthcare settings: elderly patients may have reduced mental capacity which is temporary, progressive or even permanent due to an ongoing disease process, neurological disorders like stroke, or dementia. Therefore, their decision capacity is a big area of concern. A legally authorized representative must consent to such participants.
4. Students: In the case where the professor has to issue his survey to be completed by students, the study team should not include the professor as a member since his presence alone can cause coercion or undue influence. Also, the team can arrange to have someone other than the professor to obtain consent from participants to avoid any words of threat or promises that can tempt the students into compliance. E.g. promises of glowing recommendation letters at course completion.

Conclusion

Summarily, researchers should incorporate in their study designs additional protection for participants, especially the vulnerable ones. Researchers should respect the Belmont’s principles by selecting participants fairly without bias, respecting their autonomy and providing them with a favourable risk-benefit ratio. they should be respected and neither coerced nor unduly influenced. For the vulnerable participants, the consenting process should be obtained with the help of a legally authorized representative.

References

• https://www.advara.com. The Many Faces of “Coercion” and “Undue Influence” by Luke Gelinas, PhD, senior chair director December 2, 2020
• https://ps.psychiatry online.org/doi/abs/10.1176/appi.ps.202000427digital technologies and coercion in psychiatry
• https://www.acrpnet.org/2019/03/paying- subjects- to- take-part-in-research-a-new perspective-on-coercion-and-undue-influence- march 12,2019 (volume 33, issue 3)
•  http://testbook.com /amp/key-differences/difference-between-coercion-and undue-influence
• https://research.uoregon.edu.com/manage/research-integrity-compliance/human-subjects-research/compensation-participation-research
• https://www.statistics.com/compensation-for research-participants-avoiding-undue-influence
• https://testbook.com/key-differences-between-coercion-and-undue-influence-know-all-the-differences-between-coercion-and-undue-influence
• https://www.washington.edu/research/glossary/undue/influence
• https://blog.petrieflom.law.harvard.edu/2015/11/23/vulbnerability-coercion-and-undue-influence-from-themud-into-the-muck

The post Coercion and Undue Influence in Health Research Ethics appeared first on CRENC Learn.

What is confidentiality?

Confidentiality involves keeping an individual’s information private and undisclosed. This principle is crucial in various fields, particularly in medicine, research, and law.

Confidential information refers to details about a person’s private life that they wish to keep undisclosed, except for the purposes of the study. This is distinct from public information, which is accessible to everyone. The right of research participants to control their personal information is increasingly recognized and protected by extensive legislation. In research ethics, confidentiality is about respecting the dignity and autonomy of the participant and ensuring that the use of their information does not violate their interests. This right is recognized in international and national bioethical guidelines, including the Helsinki Declaration, the General Data Protection Regulation (GDPR), and the law governing biomedical research in Cameroon. Practically, safeguarding confidentiality means:

• Keeping research participants’ data anonymous and de-identified.
• Obtaining private data only when necessary.
• Informing participants about the research’s purpose and who has access to their data.
• Ensuring participants give active consent and retain the right to withdraw at any moment.
• Allowing participants to review and correct their data.

Fictional Case Study to Elucidate the Concept of Confidentiality in Cameroon

Consider a case study involving a medical research project in the North-West region of Cameroon, an area known for its cultural and linguistic diversity. The project aimed to study the prevalence of a certain genetic disorder among different ethnic groups in the region. To ensure confidentiality, the research team faced the challenge of communicating effectively with participants who spoke different local languages and had varying levels of understanding and trust in medical research.

The team employed local translators and cultural mediators who were fluent in the predominant local languages and dialects. They also conducted community engagement sessions to explain the purpose of the research, how data would be collected, and the measures in place to protect participants’ confidentiality. Consent forms were translated into multiple local languages, ensuring that participants fully understood their rights and the confidentiality of their data.

Despite these measures, the team encountered a situation where a participant’s genetic data, which had implications for his extended family, was inadvertently disclosed within his community. This breach not only caused distress to the participant and his family but also raised concerns among other participants and threatened the credibility of the research project.

The incident led to a review of data handling procedures. The research team reinforced their data security protocols and provided additional training to staff on the importance of maintaining strict confidentiality. They also held community meetings to address the incident, reassure participants, and restore trust in the research process.

This case study highlights the complexities of ensuring confidentiality in a diverse setting like Cameroon and underscores the need for continuous vigilance, cultural sensitivity, and community engagement in medical research.

Tips on Ensuring Confidentiality and Pitfalls to Avoid

Ensuring confidentiality in medical research involves a multifaceted approach. Here, we explore several key strategies and the pitfalls to avoid:

Legal Implications of a Confidentiality Breach in Cameroon

In Cameroon, the legal framework provides clear guidelines and penalties for breaches of confidentiality in medical research and practice. Cameroon’s law on medical research outlines specific guidelines for data use and sharing. Key articles mandate informed consent for data reuse and sharing, and the obligation to maintain data confidentiality. Non-compliance can lead to legal action. Section 62 of the law states that “whoever, involved in a medical research project, discloses confidential information without the prior consent of its owner, shall be punished with imprisonment of from 3 (three) months to 3 (three) years and a fine of from 20,000 (twenty thousand) to 100,000 (one hundred thousand) CFA francs”. This provision is important to protect the privacy and confidentiality of participants in medical research. It also helps to ensure that research data is not misused or exploited. Example: A researcher is conducting a medical study on a new drug to treat HIV/AIDS. The researcher collects confidential information from the study participants, such as their medical history, HIV status, and treatment response. If the researcher discloses this confidential information to anyone without the prior consent of the participants, they could be punished with imprisonment and/or a fine.

The law also dictates that medical practitioners adhere to professional secrecy. Breaches, as per the Penal Code, can result in imprisonment, fines, and professional sanctions such as suspension or removal from the medical register.

References

• Confidentiality – Definition, Examples, Cases (legaldictionary.net)
• Bos, Jaap. “Confidentiality,” 149–73, 2020. https://doi.org/10.1007/978-3-030-48415-6_7. (10) (PDF) Confidentiality. Available from: https://www.researchgate.net/publication/ 344372529_Confidentiality [accessed Jun 13 2023].
• Avoiding Preventable Breaches of Confidentiality  Mary Alice Fisher https://doi.org/10.1093/med:psych/9780199752201.003.0009 Step 5: Avoiding Preventable Breaches of Confidentiality | The Ethics of Conditional Confidentiality: A Practice Model for Mental Health Professionals | Oxford Academic (oup.com)
• The 7 Most Common Causes of Data Breaches (makeuseof.com)
• Journal Officiel de la République du Cameroun; portant Code de Procédure Pénale N°2005/007 27 Juillet 2005, https://bit.ly/3r07gMG
• Law No. 2022/008 of April 27, 2022, relating to Medical Research Involving Human Subjects in Cameroon, LOI Nº 2022/008 DU 27 AVRIL 2022 RELATIVE A LA RECHERCHE MEDICALE IMPLIQUANT LA PERSONNE HUMAINE AU CAMEROUN | Centre de Documentation Numérique du Secteur Santé (minsante.cm)
• Cameroun: Décret N° 83-166 Du 12 Avril 1983 Portant Code De Déontologie Des Médecins. Cameroun: Décret n° 83-166 du 12 Avril 1983 portant code de déontologie des médecins – Camerlex
• Loi no 92/007 du 14 août 1992, portant Code du Travail au Cameroun https://www.medcamer.org/wp-content/uploads/2014/10/Code-du-Travail-au-Cameroun.pdf

Image by DCStudio on Freepik

The post Ensuring Patients’ Confidentiality in Medical Research in Cameroon appeared first on CRENC Learn.