Secondary data plays an important role in population health research by allowing healthcare professionals to investigate disease patterns, identify health disparities, and evaluate factors associated with patient outcomes. Unlike primary research, which requires investigators to collect new information, secondary data analysis uses information previously gathered through surveys, surveillance systems, clinical studies, or administrative databases. This approach can reduce research costs, shorten investigation timelines, and support evidence-based public health decisions. However, the usefulness of secondary data depends on the quality of the original information, the appropriateness of the research variables, and the methods used to interpret findings (Areco et al., 2021).
Chronic obstructive pulmonary disease (COPD) provides an important example of a population health problem that can be investigated through secondary data. COPD contributes to respiratory disability, hospital utilization, premature mortality, and healthcare expenditures. Understanding its distribution across populations requires information about smoking, socioeconomic conditions, environmental exposures, and healthcare access. This assessment examines COPD as a population health concern and evaluates three potential secondary data resources: the Centers for Disease Control and Prevention (CDC) Behavioral Risk Factor Surveillance System (BRFSS), the World Health Organization (WHO) Global Health Observatory (GHO), and the Inter-university Consortium for Political and Social Research (ICPSR). The discussion considers their strengths, limitations, validity, relevant variables, and implications for nursing research and population health improvement.
Chronic obstructive pulmonary disease is a progressive respiratory condition characterized by persistent airflow limitation, respiratory symptoms, and structural or functional abnormalities of the lungs. Individuals with COPD commonly experience shortness of breath, chronic coughing, sputum production, and reduced exercise tolerance. Cigarette smoking remains an important risk factor, although occupational dust, chemical exposure, household air pollution, and other environmental conditions also contribute to disease development.
COPD represents a substantial population health concern because its consequences extend beyond individual respiratory symptoms. Patients may experience repeated exacerbations, emergency department visits, hospital admissions, and limitations in daily activities. These complications create financial and organizational challenges for healthcare systems. Research examining the economic burden of COPD demonstrates the importance of understanding disease-related costs and the factors contributing to healthcare utilization (Villegas et al., 2021).
The burden of COPD is not distributed equally across populations. Older adults, individuals with limited financial resources, and people exposed to tobacco smoke or occupational pollutants may experience increased vulnerability. Differences in access to preventive services, diagnostic testing, medications, and specialist care can further influence disease outcomes.
From a nursing perspective, examining COPD at the population level helps identify communities that may benefit from smoking cessation programs, respiratory health education, early screening for symptoms, and improved chronic disease management. Secondary data can support these efforts by identifying patterns that may not be visible through individual clinical encounters.
Selecting appropriate variables is essential when examining COPD through existing datasets. The primary outcome variable for a population-based investigation could be the prevalence of diagnosed COPD. Depending on the available data, additional outcomes might include respiratory-related hospitalization, mortality, functional limitations, or healthcare utilization.
Smoking status represents an important explanatory variable because tobacco exposure is strongly associated with COPD development and progression. Other relevant variables include age, sex, educational attainment, household income, employment status, health insurance coverage, and geographic location. Environmental indicators, including exposure to air pollution and occupational respiratory hazards, may provide additional insight into differences in disease risk.
The relationships among these variables require careful interpretation. For example, an observed association between lower income and increased COPD prevalence may reflect several interconnected factors, including smoking exposure, working conditions, housing environments, and barriers to healthcare. An analysis that considers income without accounting for age and smoking status could produce misleading conclusions.
Researchers should therefore distinguish between dependent variables, independent variables, and potential confounding factors. Appropriate statistical adjustments can improve the interpretation of observed relationships, although observational secondary data generally cannot establish causation independently.
The Behavioral Risk Factor Surveillance System is a major population health surveillance resource administered through the CDC in collaboration with state health departments. It collects information about health-related behaviors, chronic conditions, preventive services, and healthcare access among noninstitutionalized adults in the United States.
BRFSS is particularly relevant to COPD research because its questionnaires include information about self-reported COPD diagnoses, smoking behaviors, demographic characteristics, and other health indicators. Researchers can use these variables to investigate how COPD prevalence differs across population groups and geographic locations.
Strengths of BRFSS
One significant strength of BRFSS is its extensive population coverage. Its large survey samples allow researchers to examine health patterns across states and demographic groups. Standardized questions also support comparisons across reporting periods when question wording and survey methods remain sufficiently consistent.
Another advantage is the availability of information about multiple behavioral and socioeconomic factors within the same survey. Researchers can examine associations between COPD and smoking, education, income, and healthcare access without collecting new information from participants.
BRFSS data have been used in peer-reviewed investigations of health behaviors and service utilization, demonstrating their broader value for population health research (Powers et al., 2024). Public access to many BRFSS datasets further supports cost-effective research and timely public health assessments.
Limitations of BRFSS
Despite these advantages, BRFSS relies substantially on self-reported information. Participants may inaccurately recall health behaviors or provide responses influenced by social desirability. For example, some respondents may underestimate cigarette consumption or inaccurately describe their smoking history.
Another concern involves the measurement of COPD itself. A self-reported diagnosis does not necessarily confirm airflow obstruction through spirometry. Individuals with undiagnosed COPD may be classified as not having the disease, potentially underestimating its actual prevalence.
BRFSS also excludes certain populations, including institutionalized adults. Researchers must consider sampling methods, survey weights, nonresponse, and changes in questionnaire design when interpreting results.
Consequently, BRFSS is useful for estimating patterns of diagnosed COPD among community-dwelling adults but is less suitable for establishing clinically confirmed COPD prevalence.
The WHO Global Health Observatory provides access to health indicators collected and compiled from countries and territories worldwide. Its resources include information about mortality, disease burden, health systems, environmental risks, and population health trends.
For COPD research, the GHO can help investigators examine international differences in chronic respiratory disease burden and associated health indicators. This perspective is valuable because tobacco exposure, environmental conditions, healthcare infrastructure, and population demographics vary considerably between countries.
Strengths of GHO
The principal advantage of GHO is its international scope. Researchers can investigate patterns across countries and regions, identify broad differences in disease burden, and examine how respiratory health outcomes relate to population-level conditions.
The resource also supports public health planning by providing access to indicators developed through established international reporting processes. Its information can help policymakers compare national progress and identify areas requiring additional attention.
Global comparisons are particularly relevant when examining differences in life expectancy and health outcomes. Garmany and Terzic (2024), for example, investigated international differences in healthspan and lifespan, illustrating the value of multinational health indicators for identifying population health inequalities.
Limitations of GHO
An important limitation involves differences in national data collection systems. Countries may vary in diagnostic capacity, disease registration practices, reporting completeness, and the availability of healthcare information.
Some GHO indicators are also derived from statistical estimates rather than direct population measurements. Such estimates may be necessary when complete national information is unavailable, but they introduce uncertainty that must be acknowledged.
Additionally, many international indicators are reported at the population or country level. These aggregated measures cannot automatically establish relationships between individual risk factors and individual health outcomes.
For example, a country with high tobacco consumption and elevated COPD mortality does not necessarily demonstrate that every person with COPD has a history of smoking. Interpreting national associations as individual-level relationships would introduce an ecological fallacy.
Therefore, GHO is especially useful for examining international population health patterns, while its suitability for individual-level COPD risk analysis is more limited.
The Inter-university Consortium for Political and Social Research is a research data archive that provides access to datasets covering social, behavioral, economic, and health-related topics. Unlike BRFSS and GHO, which represent defined surveillance or indicator systems, ICPSR contains collections developed through different research projects and methodologies.
Its relevance to COPD depends on the specific studies available within the archive. Potentially useful datasets may contain information about smoking behavior, socioeconomic status, health service utilization, chronic illness, or environmental exposures.
Strengths of ICPSR
A major strength of ICPSR is the diversity of research materials it preserves. Investigators can identify datasets containing detailed social and behavioral information that may not be available through broader surveillance systems.
The archive also emphasizes documentation and data preservation. Study descriptions, codebooks, variable information, and methodological records can help researchers understand how information was originally collected.
Bratt (2023) examined research data archiving practices associated with ICPSR, highlighting the importance of infrastructure and documentation in making existing research materials available for future use.
For COPD investigations, a suitable ICPSR dataset could support an examination of socioeconomic factors associated with chronic disease outcomes. Longitudinal datasets, when available and appropriate, may also permit analysis of changes over time.
Limitations of ICPSR
The primary limitation is that ICPSR does not represent a single standardized population health dataset. Each archived study has its own sampling procedures, eligibility criteria, variables, and measurement instruments.
Consequently, the validity and generalizability of findings must be evaluated separately for each selected dataset. Researchers cannot assume that every study within the archive provides representative information about adults with COPD.
Some datasets also require restricted-access approval because they contain sensitive participant information. Other studies may lack important clinical variables, including spirometry results or detailed respiratory disease histories.
ICPSR is therefore most useful when investigators identify a specific dataset whose research design and variables closely match the intended COPD research question.
The credibility of secondary data analysis depends on whether the selected information accurately represents the concepts being investigated. Validity concerns whether a dataset measures the intended phenomenon, while reliability concerns the consistency of measurements and procedures.
For COPD research, validity is especially important because the identification of disease can differ between datasets. A survey asking whether a healthcare professional has diagnosed COPD measures reported diagnosis, whereas a clinical investigation using spirometry may provide a more objective assessment of airflow obstruction. These measurements are related but should not be treated as interchangeable.
BRFSS benefits from established survey procedures, standardized questions, and statistical weighting methods. Nevertheless, its reliance on self-reported information creates potential measurement error. Researchers should review questionnaire wording, response categories, sampling procedures, and available methodological documentation before conducting an analysis.
The validity of GHO indicators depends partly on the underlying national reporting systems and the methods used to generate estimates. Researchers should examine indicator definitions, reporting periods, data sources, and uncertainty estimates when available. Differences in healthcare infrastructure and disease identification practices may influence comparisons between countries.
ICPSR requires a more individualized evaluation. Because it archives studies produced by different investigators, researchers must examine the methodology of each selected study. Relevant considerations include sample size, participant recruitment, measurement instruments, missing information, and data collection procedures.
Baldwin et al. (2022) emphasize the importance of addressing researcher bias when analyzing existing datasets. Investigators may unintentionally select variables, statistical models, or outcomes that favor an expected conclusion. Establishing research questions and analytical procedures before examining results can reduce this concern.
A practical approach would involve developing a data quality assessment before beginning COPD analysis. This assessment should evaluate population coverage, measurement validity, missing data, variable definitions, and the compatibility of information across sources. Such preparation strengthens confidence in findings and reduces the likelihood of inappropriate conclusions.
One challenge in secondary data analysis is that existing datasets were usually created to answer questions different from those of the current investigator. As a result, researchers may find that important variables are missing or insufficiently detailed.
For example, BRFSS may identify whether a respondent reports having COPD but may not provide objective lung function measurements or detailed clinical information about disease severity. Similarly, a dataset containing smoking status may not include sufficient information about cumulative tobacco exposure.
These limitations can restrict the types of conclusions that researchers can reasonably draw. Investigators must ensure that the available variables adequately represent the intended research concepts rather than selecting convenient measures without considering their clinical meaning.
Combining information from multiple sources introduces methodological challenges. BRFSS primarily provides survey-based information about adults in the United States, whereas GHO frequently presents national or international health indicators. ICPSR may contain either individual-level or aggregated information, depending on the archived study.
These differences affect the comparability of findings. For instance, COPD prevalence estimated through self-reported diagnoses cannot automatically be compared with disease burden estimates developed through statistical modeling.
Researchers must evaluate whether datasets use compatible definitions, reporting periods, age categories, and population characteristics. Without these considerations, differences in measurement may be mistaken for meaningful differences in disease occurrence.
Selection bias occurs when individuals included in a dataset differ systematically from those who are excluded or do not participate. In COPD research, individuals experiencing severe illness, unstable housing, or limited communication access may be underrepresented in certain surveys.
Missing responses can also influence statistical findings, particularly when missingness is associated with health status or socioeconomic characteristics.
Appropriate statistical weighting, sensitivity analyses, and transparent reporting of missing data can help address these concerns. However, statistical adjustments cannot fully compensate for every weakness in the original data collection process.
Secondary data analysis also creates ethical responsibilities concerning confidentiality, informed consent, and responsible information management.
Although many public health datasets contain de-identified information, researchers must still consider the possibility of identifying individuals through combinations of demographic or geographic characteristics.
Restricted ICPSR datasets may require formal authorization, secure storage, and compliance with data use agreements. Investigators must also determine whether institutional review board oversight or an exemption determination is required.
Protecting participant information is particularly important when examining vulnerable populations or small communities. Ethical research requires balancing the value of population health findings against the obligation to preserve privacy and prevent misuse of sensitive information.
The three secondary data resources offer different opportunities for investigating COPD and informing nursing practice. Their usefulness depends on the research question and the level of analysis required.
Consider a proposed population health investigation examining the relationship between smoking status, household income, and diagnosed COPD among adults in the United States. BRFSS would be an appropriate primary resource because it contains relevant individual-level survey variables and supports population-based comparisons.
The investigator could identify respondents reporting COPD and examine whether disease prevalence differs across smoking categories and income groups. Age, sex, education, and other available characteristics could be included as potential confounding variables.
A weighted multivariable logistic regression model could estimate associations between these factors and reported COPD diagnosis. However, because BRFSS is observational and generally cross-sectional, the findings would describe associations rather than establish causal relationships.
GHO could provide additional international context by illustrating how COPD burden differs between countries or regions. This information could help explain why respiratory disease prevention is a broader public health priority, although it should not be directly combined with BRFSS individual-level records without a methodologically appropriate design.
A carefully selected ICPSR study could complement the analysis by providing information about social determinants, healthcare utilization, or changes in health behaviors over time. Before selecting such a study, researchers would need to verify that its population, variables, and methodology support the intended investigation.
For nursing practice, these findings could inform community respiratory health assessments and guide prevention programs. If an analysis identified a high prevalence of reported COPD among adults with tobacco exposure and limited healthcare access, nursing leaders could prioritize smoking cessation education, symptom assessment, referrals for diagnostic evaluation, and improved access to chronic respiratory care.
These interventions would require additional local assessment before implementation. National or international data can identify potential priorities, but local clinical resources, community characteristics, and patient preferences must also inform intervention planning.
A rigorous approach to secondary data research should begin with a clearly defined research question and a systematic assessment of available information. Researchers investigating COPD should select datasets according to their ability to measure the population, exposures, and outcomes of interest.
BRFSS should be considered when the objective involves examining reported COPD prevalence and associated behavioral or demographic characteristics among U.S. adults. GHO is more appropriate for examining international disease patterns and population-level health indicators. ICPSR should be used when a specific archived study provides additional variables or a research design that directly addresses the investigation.
Researchers should also establish a documented analysis plan before conducting statistical tests. Baldwin et al. (2022) identify researcher bias as an important concern in secondary analysis, making transparency in variable selection and analytical decisions particularly valuable.
Another recommendation is to conduct sensitivity analyses when possible. For example, investigators could evaluate whether observed associations remain consistent after adjusting for age, smoking status, and socioeconomic characteristics. Such analyses help determine whether findings depend heavily on particular modeling assumptions.
Finally, nursing researchers should communicate findings with appropriate caution. Limitations involving self-reported diagnoses, missing clinical measurements, selection bias, and differences in data collection must be clearly described. Research conclusions should reflect what the data can reasonably demonstrate.
By following these practices, researchers can produce findings that support evidence-based nursing decisions without overstating the certainty or generalizability of their results.
Secondary data analysis provides valuable opportunities to investigate chronic obstructive pulmonary disease and identify factors associated with population health outcomes. BRFSS, WHO GHO, and ICPSR offer complementary resources for examining disease prevalence, health behaviors, socioeconomic conditions, and broader patterns of respiratory disease.
Each resource also presents limitations that influence its suitability for particular research questions. BRFSS relies on self-reported information, GHO includes aggregated indicators and modeled estimates, and ICPSR contains studies with varying methodologies and access requirements. Recognizing these differences is necessary for selecting appropriate variables and interpreting results accurately.
For population health nursing, effective secondary data analysis can strengthen community assessments, guide respiratory disease prevention, and support decisions about healthcare resource allocation. Its greatest value emerges when investigators combine credible information, appropriate analytical methods, ethical research practices, and careful interpretation to develop meaningful evidence for improving population health.
Areco, K. N., Konstantyner, T., Paiva, P., Balda, R., Nobre, D. T., Sanudo, A., Kiffer, V., Kawakami, M. D., Miyoshi, M. H., Marinonio, A. S. S., Freitas, R., Morais, L., Teixeira, M., Waldvogel, B., Almeida, M., & Guinsburg, R. (2021). Operational challenges in the use of structured secondary data for health research. Frontiers in Public Health, 9, Article 642163. https://doi.org/10.3389/fpubh.2021.642163
Baldwin, J. R., Pingault, B., Schoeler, T., Sallis, H. M., & Munafò, M. R. (2022). Protecting against researcher bias in secondary data analysis: Challenges and potential solutions. European Journal of Epidemiology, 37(1), 1–10. https://doi.org/10.1007/s10654-021-00839-0
Bratt, S. (2023). “Routine infrastructuring”: How social scientists appropriate resources to deposit qualitative data to ICPSR and implications for FAIR and CARE. Proceedings of the Association for Information Science and Technology, 60(1), 61–72. https://doi.org/10.1002/pra2.769
Garmany, A., & Terzic, A. (2024). Global healthspan-lifespan gaps among 183 World Health Organization member states. JAMA Network Open, 7(12), Article e2450241. https://doi.org/10.1001/jamanetworkopen.2024.50241
Powers, A. M., Patel, D., DeAngelis, M. M., Feng, C., & Allison, K. (2024). Risk factors affecting the utilization of eye care services were evaluated by the CDC’s Behavioral Risk Factor Surveillance System from 2018 to 2021. Frontiers in Public Health, 12, Article 1335427. https://doi.org/10.3389/fpubh.2024.1335427
Villegas, C., Zulueta, M., Montes, M., Bravo, P., & Pérez, M. (2021). Cost analysis of chronic obstructive pulmonary disease (COPD): A systematic review. Health Economics Review, 11(1), 1–12. https://doi.org/10.1186/s13561-021-00329-
