Skip to main content
PMC home page
Bulletin of the World Health Organization logoLink to Bulletin of the World Health Organization
. 2024 Jun 4;102(8):588–599. doi: 10.2471/BLT.24.291783

National health examination surveys; a source of critical data

Enquêtes nationales de santé par examen: une source de données essentielles

Encuestas nacionales de salud: una fuente de datos fundamentales

مسوحات الفحص الصحي الوطني؛ مصدر للبيانات الحيوية

全国健康检查调查：关键数据的来源

Национальные опросы по изучению состояния здоровья: источник важнейших данных

Paula Margozzini a, Hanna Tolonen b, Antonio Bernabe-Ortiz c, Sarah Cuschieri d, Chiara Donfrancesco e, Luigi Palmieri e, Luz Maria Sanchez-Romero f, Jennifer S Mindell g,✉, Oyinlola Oyebode h
PMCID: PMC11276157  PMID: 39070597

Abstract

The aim of this paper is to contribute technical arguments to the debate about the importance of health examination surveys and their continued use during the post-pandemic health financing crisis, and in the context of a technological innovation boom that offers new ways of collecting and analysing individual health data (e.g. artificial intelligence). Technical considerations demonstrate that health examination surveys make an irreplaceable contribution to the local availability of primary health data that can be used in a range of further studies (e.g. normative, burden-of-disease, care cascade, cost and policy impact studies) essential for informing several phases of the health planning cycle (e.g. surveillance, prioritization, resource mobilization and policy development). Examples of the use of health examination survey data in the World Health Organization (WHO) European Region (i.e. Finland, Italy, Malta and the United Kingdom of Great Britain and Northern Ireland) and the WHO Region of the Americas (i.e. Chile, Mexico, Peru and the United States of America) are presented, and reasons why health provider-led data cannot replace health examination survey data are discussed (e.g. underestimation of morbidity and susceptibility to bias). In addition, the importance of having nationally representative random samples of the general population is highlighted and we argue that health examination surveys make a critical contribution to external quality control for a country’s health system by increasing the transparency and accountability of health spending. Finally, we consider future technological advances that can improve survey fieldwork and suggest ways of ensuring health examination surveys are sustainable in low-resource settings.

Introduction

National health examination surveys have been developed to gather important information that cannot be obtained from other sources. In these surveys, trained field staff take objective, biophysical measurements (e.g. of anthropometric variables or blood pressure) and collect biological samples (e.g. of blood or urine) for laboratory analysis. The data obtained complement the self-reported data collected, for example, by health interview surveys, which include only self-reported information. In addition, health examination surveys are the observational studies with the greatest external validity because they are based on randomized, representative household samples. Consequently, the information obtained is relevant for both population (i.e. public health) and individual health.1

Health examination surveys provide more accurate information than health interview surveys. For example, people tend to overestimate their height and underestimate their weight compared with measurements taken by trained staff, which results in underestimates of their body mass index – a measure widely used for assessing people for obesity and for predicting morbidity from several chronic noncommunicable diseases.2,3

Most high- and middle-income countries conduct national health interview surveys that use questionnaires to collect basic information about the general population. For European Union member states, European health interview surveys are mandatory.4 Far fewer countries regularly conduct health examination surveys. Nevertheless, many low- and middle-income countries have conducted at least one small health examination survey in accordance with the World Health Organization (WHO) STEPwise approach to noncommunicable disease risk factor surveillance, known as STEPS.5 In addition, many low- and middle-income countries conduct Demographic and Health Surveys (DHS), which are funded by the United States Agency for International Development, at least every 5 years and some (e.g. Peru) conduct them annually. These surveys include a small number of biophysical assessments, such as anthropometry and anaemia testing.6 However, few countries have a track record of frequent health examination surveys that include a wide range of biophysical measurements. The longest running health examination survey series in Europe has been conducted by Finland every 5 years since 1972 (i.e. the national FINRISK study).7 In the United States of America, National Health and Nutrition Examination Surveys started in the 1960s and have been run as a continuous programme since 1999.8

Existing national health examination surveys differ in the age groups covered, the range of measurements taken and the way they are organized. For example, most health examination surveys in mainland Europe and the United States make use of clinical examination centres, whereas those in the United Kingdom of Great Britain and Northern Ireland and Latin America involve visiting participants in their own homes.9,10 Despite some differences, they share many features, particularly sampling methods, survey questions, anthropometric measurements and some other biophysical measurement protocols.1,9,10

Although numerous reports and scientific papers have described findings based on health examination survey data, which implicitly suggests they are valuable data sources, few countries conduct regular, nationally representative, wide-ranging health examination surveys. Moreover, many countries that do conduct regular health examination surveys are struggling to secure long-term financing for successive surveys. Countries with small populations face additional challenges, including a lack of infrastructure and resources for conducting regular, nationally representative health examination surveys.11

Recently, several health examination surveys around the world have been delayed, survey field costs have risen and financing has been cut. In addition, advances in medical informatics and artificial intelligence have given rise to doubts about whether health examination surveys provide the most efficient means of collecting data. Apart from a few studies from Chile, England and Finland,12–15 little has been published on the usefulness of health examination surveys, their impact or factors influencing their effectiveness. In this context, it is possible to ask: (i) whether random general population samples are still needed for chronic disease surveillance and health planning; (ii) whether governments should still finance costly population health surveys; and (iii) whether new information technology could be used to generate chronic disease indicators that replace the field measurements conducted in health examination surveys.

The aim of this paper was to discuss these questions with reference to technical arguments and to examples drawn from health examination surveys conducted in Chile, England, Finland, Italy, Malta, Mexico, Peru and the United States.

Noncommunicable disease

Health examination surveys can provide primary data for use in different phases of the noncommunicable disease health planning cycle, including surveillance, prioritization, resource mobilization and policy formulation, implementation and impact assessment.

Table 1 shows some traditional noncommunicable disease indicators that can be obtained directly from health examination survey measures, and describes the specific contribution each of these indicators can make to deriving estimates of variables in other studies, such as: (i) studies to establish population norms; (ii) studies of the care cascade and health system performance; (iii) studies of health inequalities; (iv) studies of the burden of disease and of the burden attributable to specific risk factors; (v) studies of health costs and avoidable health costs; and (vi) studies of changes in population trends. Health policies and laws usually cite these studies directly but do not cite health examination surveys, which may hide their importance. In practice, national ministries of health could justify their budgets for health examination surveys by highlighting these contributions, and by giving examples of how health examination survey data can be used in local policy formulation, implementation and assessment. In Chile, health examination survey data have been used to supplement the findings of all the different study types listed in Table 1.

Table 1. Contribution of national health examination survey data to other study types essential for noncommunicable disease health planning.

Study type Data contributed by national health examination surveys to other study types
Health examination survey indicator
Risk factor prevalence Self-reported diagnosis Objective disease prevalence Self-reported treatment and drug use Biophysical parameters and disease control status Perceived health and quality of life
Normative studies General population and percentile age and sex risk factor exposure curves National benchmarks for diagnostic coverage National denominators for prevalence calculations and age- and sex-adjusted prevalence estimates for local administrative divisions National benchmarks for treatment coverage Population distribution curves and normal cut-offs for biophysical parameters; and national benchmarks for effective disease control coverage National quality of life benchmarks; and local indexing of quality-of-life instruments
Care cascade and health system performance studies NA Diagnostic coverage NA Treatment coverage and inappropriate drug use Effective treatment coverage General population quality of life and disease-specific quality of life
Inequality studies Estimates of relative and absolute excess risk factor exposure in underserved population groups Gaps in diagnostic coverage in underserved population groups Unequal disease distribution Gaps in treatment coverage in underserved population groups Gaps in effective treatment coverage in underserved population groups Gaps in quality of life in underserved population groups
Studies of the global burden of disease and of the burden attributable to risk factors Population attributable fractions and population impact fractions NA Prevalence-based estimates of the global burden of disease (DALYs and YLDs) NA NA Estimates of country-specific disability weights
Health costs and avoidable costs studies Attributable costs and avoidable costs due to changes in risk factor exposure (i.e. counterfactuals) Avoidable costs attributable to increases in diagnostic coverage Total burden-of-disease costs Avoidable costs attributable to increases in treatment coverage Avoidable costs attributable to improvements in treatment quality Cost-effectiveness estimates (QALYs)
Changes in population trends Changes in population trends in risk factor exposure Changes in population trends in diagnostic coverage Changes in population trends in disease prevalence and burden Changes in population trends in treatment coverage Changes in population trends in effective treatment coverage and biophysical parameter distributions Changes in quality of life
Open in a new tab

DALY: disability-adjusted life-year; NA: not applicable; QALY: quality-adjusted life-year; YLD: year lived with disability.

Several noncommunicable disease preventive policies for nutrition, tobacco control and disease management (e.g. so-called WHO best buys) have been supported by, and their impact assessed, using health examination survey data. Examples of the influence of health examination surveys on policy in Chile, England, Finland, Italy, Malta, Mexico, Peru and the United States are described in Table 2 and Table 3.

Table 2. Examples of the influence of health examination surveys on dietary policy, Chile, England, Finland, Italy, Malta, Mexico, Peru and the United States, 1994–2020.

Health topic and country Stage at which health examination surveys influenced dietary policya
Setting the agenda Policy formation Policy evaluation
Obesity
Chile The increases in obesity, an unhealthy diet and high urinary sodium levels reported in the 2010 Encuesta Nacional de Salud led to further studies of the local burden of disease and of the disease burden attributable to specific risk factors and to an obesity costs study16 These studies were decisive in establishing support for an increase in taxes on sugary beverages, front-of-package labelling for foods, the banning of unhealthy food sales in schools and the marketing regulations introduced in the 2014 food law. Based on Encuesta Nacional de Salud data,16 a new National Nutritional Policy was published in 2018, nutritional guidelines were updated and waist circumference cut-off points were defined for Chile17 NA
England, United Kingdom General and abdominal obesity were monitored in adults and children using annual Health Survey for England data The 2020 obesity strategy for England included: (i) a ban on television and online advertising of foods high in fat, sugar and salt before 21:00; (ii) restrictions on price promotions for these foods; (iii) restrictions on where shops may display these foods and unhealthy drinks; and (iv) a mandate for energy labelling of food in restaurants. These proposals were assessed during development of the obesity strategy using Health Survey for England anthropometric data in a Department of Health and Social Care calorie model that explored the likely impact of these and other possible policy interventions18 NA
Malta Anthropometric data from health examination surveys were used to accurately estimate obesity19 These data helped in estimating the national cost of this disease and led to a range of stakeholders working together in a timely fashion to prevent obesity NA
Diet
Chile Low serum folic acid and vitamin D levels and urinary iodine levels measured in the 2016–2017 Encuesta Nacional de Salud In reaction to the observed ongoing nutritional transition, a national micronutrient strategy was developed in 2021 that updated requirements for the mandatory fortification of food16 Data from Encuesta Nacional de Salud surveys helped in monitoring the effect of changes in mandatory food fortification programmes20
Mexico A DHS identified the double burden of malnutrition (i.e. undernutrition and overnutrition)21 NA Health examination survey data demonstrated that social food assistance programmes reduced food insecurity;22 and DHSs were used to monitor the double burden of malnutrition and the effect of policies to address this burden21
Vitamin D level
Finland Data from the 2002 national FINRISK study showed low population levels of vitamin D14 These data supported the fortification of milk and fat spreads with vitamin D, which began in 200314 Mean serum 25-hydroxyvitamin D levels rose from 48 nmol/L in 2000 to 65 nmol/L in 2011 (P < 0.001), with a higher rise of 20 nmol/L in survey participants not taking supplements23
Salt intake
England, United Kingdom Health examination survey data showed excessive sodium excretion, a marker for salt intake24 The Department for Health in England set targets in 2006 and worked with the food industry to reduce the salt content of the processed food products that contributed most to the population’s salt intake, as identified through health examination survey data (the initial focus was therefore on bread) Significant falls in salt intake were observed from 2005–2006 to 2008–2009 but no further falls occurred by 2018–2019
Italy 24-hour urine samples taken in an health examination survey showed higher than recommended daily salt consumption in the adult population25 As a result, the health ministry, in collaboration with NGOs, undertook health promotion actions aimed at increasing awareness of the risks associated with excessive salt intake and reformulated a wide range of products to reduce their salt content in voluntary agreements with the food industry and artisan bakeries 24-hour urine samples showed a reduction in sodium excretion of around 12% between 2008–2012 and 2018–2019
Nutritional status
Peru A DHS identified high levels of maternal anaemia and childhood stunting These findings informed the 2006 Children Malnutrition Initiative, which was accompanied by a National Strategy for Poverty Reduction and Economic Opportunities in 2007;26 and DHS data demonstrated the need to target social and nutritional programmes on rural areas27 These policies, alongside a cash transfer programme, were credited with the improvements seen in nutritional status28
United States NA National Health and Nutrition Examination Survey measurements provided the basis for the 2000 CDC growth charts for children, which updated the 1977 National Center for Health Statistics charts and created new body mass index-for-age charts29 – such population-level data can be used to identify outliers NA
Neural tube defects
United States National Health and Nutrition Examination Survey data for 1988–1994 showed that the median red blood cell folate level in women aged 15–45 years was 160 ng/mL (substantially below the recommended 220 ng/mL); 38% of women had levels < 220 mg/mL30 In response, mandatory fortification of cereal grain products was implemented in January 1998 In the 1999–2000 National Health and Nutrition Examination Survey, the median red blood cell folate level in women aged 15–45 years was 264 ng/mL;30 the birth rate of babies born with neural tube defects fell in the 2 years before the mandatory fortification of cereal grain products, probably due to increased use of folate supplementation and prenatal diagnosis; the rate fell further in the 2 years after mandatory fortification; and, although the median folate level fell after 2000, the birth rate of babies born with neural tube defects continued to fall, though more slowly
Lead intake
United States National Health and Nutrition Examination Surveys have involved measuring population blood lead levels and a blood lead reference value is used to identify children who have blood lead levels above the 97.5th percentile for children aged 1–5 years31 Based on National Health and Nutrition Examination Survey data, federal policies and a range of public health interventions were introduced to reduce exposure to lead National Health and Nutrition Examination Survey measurements of blood lead levels demonstrated falls over time in both adults and children, including those in high-risk areas
Open in a new tab

CDC: Centers for Disease Control and Prevention; DHS: Demographic Health Survey; NA: not applicable; NGO: nongovernmental organization.

a Demographic Health Surveys, Encuestas Nacionales de Salud (National Health Surveys), the Health Survey for England, National Health and Nutrition Examination Surveys and national FINRISK studies are all health examination surveys.

Table 3. Examples of the influence of health examination surveys on health-care policy-making, Chile, England, Finland, Italy, Malta, Mexico, Peru and the United States, 1994–2020.

Health topic and country Stage at which health examination surveys influenced policy-makinga
Setting the agenda Policy formation Policy evaluation
Hypertension
Chile Underdiagnosis and inadequate detection and treatment of hypertension identified in Encuesta Nacional de Salud surveys32 Antihypertensive drugs became free in 2014 (although Encuesta Nacional de Salud data provided a possible justification for this development, there was no direct link) Reduction in salt intake; reductions in undiagnosed and untreated hypertension; and improvements in blood pressure control
England, United Kingdom Underdiagnosis and inadequate detection and treatment of hypertension identified in Health Surveys for England33 General practitioner contracts were changed to incentivize blood pressure control Reductions in undiagnosed and untreated hypertension; and improvements in blood pressure control
Peru Underdiagnosis and inadequate detection and treatment of hypertension identified in Encuesta Nacional de Salud surveys34 NA Reductions in undiagnosed and untreated hypertension; and improvements in blood pressure control
Mexico In 2016, a health examination survey found that 40% of adults with probable hypertension were undiagnosed35 NA NA
Italy Between 2008 and 2012, CUORE Project health examination surveys found that 39% of men and 34% of women were unaware they had raised blood pressure36 NA The 2018–2019 CUORE Project health examination survey provided evidence of improvements in awareness of raised blood pressure and in the control of hypertension37
Diabetes
England, United Kingdom Underdiagnosis of diabetes observed in 2009 Health Survey for England38 NA NA
Finland NA FINRISK data used to validate the FINDRISC tool for identifying high diabetes risk39 NA
Italy Between 2008 and 2012, CUORE Project health examination surveys found that 39% of men and 29% of women were unaware of probable diabetes36 NA NA
Malta The prevalence of diabetes in adults observed in the 2014–2015 health examination survey was 10.3%, with 4% of cases newly identified; the prevalence in a 2015 health interview survey was 8.5%40 NA NA
Peru Objective anthropometric data from health examination surveys were used to predict the likely numbers of cases of diabetes41 Weaknesses in diabetes screening programmes were identified by comparing body mass indices derived from measurements with responses to questions about screening for diabetes42 NA
Hypercholesterolaemia
Italy Between 2008 and 2012, CUORE Project health examination surveys found that 38% of men and 42% of women were unaware of probable hypercholesterolaemia36 NA NA
Cardiovascular risk
Chile NA Health examination survey data on the prevalence of chronic multimorbidity linked to national hospital data on discharges and deaths were used to assess the predictive validity of a new multimorbidity stratification that is now being introduced into primary care at the national level (P. Margozzini, personal communication) NA
Finland NA Data from national FINRISK studies linked to national health-care data sets were used to create the FINRISK calculator for identifying individuals at a high risk of cardiovascular disease and the CAIDE risk score for predicting the subsequent development of dementia43,44 NA
United States Accelerometer data from 2003 to 2006 National Health and Nutrition Examination Surveys linked to data on deaths showed that the length of physical activity bouts was not associated with the mortality benefits of accumulated moderate-to-vigorous physical activity45 2018 physical activity guidelines differed from 2008 guidance in dropping the requirement for aerobic activity to be performed in episodes of at least 10 minutes46 NA
Chronic kidney disease
England, United Kingdom47 In 2009–2010, 3% of adults aged ≥ 75 years reported having chronic kidney disease in the Health Survey for England interview component but, according to the Health Survey for England measurements, 32% of adults had stage 3 to 5 chronic kidney disease NA NA
Mexico NA Modelling based on Chilean health examination survey data was used by the Mexican health ministry to persuade the country’s treasury to expand the range of treatments covered for end-stage chronic kidney disease (M Walbaum, personal communication) NA
Thyroid disease
Chile48 The 2009–2010 Encuesta Nacional de Salud found that the median thyroid-stimulating hormone level in the population was high, which prompted research into local causes This finding resulted in the development of normal values for thyroid-stimulating hormone serum levels in the country and influenced Chilean hypothyroidism clinical guidelines NA
Open in a new tab

NA: not applicable.

a Demographic Health Surveys, Encuestas Nacionales de Salud (National Health Surveys), the Health Survey for England, National Health and Nutrition Examination Surveys and national FINRISK studies are all health examination surveys.

Data from health examination surveys can also be used to generate local hypotheses on the association between the population’s exposure to risk factors and chronic disease, to predict the future disease burden and to influence clinical and lifestyle guideline development (Table 3).

Health policy

Data from health examination surveys can also support public health decision-making, including deciding on where policies should be targeted.

Monitoring and establishing norms

Repeated health examination measures can be used to monitor changes in the health status of the population over time, thereby making it possible to determine whether implemented policies are having the desired effects (Table 2 and Table 3).

Even where health examination survey measurements are not used directly, they can contribute to improving other survey data. For example, Public Health England used survey data on height and weight to correct self-reported anthropometric data from the Active People Survey in England and to create an adult excess weight indicator for the Public Health Outcomes Framework.49 The Active People Survey included 1000 telephone participants from each local government area; that level of coverage would be far too expensive for an health examination survey.

Data from health examination surveys have also been used to determine population norms and reference values for diagnostic and laboratory tests (e.g. blood tests for thyroid hormones, liver enzymes and vitamins and urine tests for iodine excretion). Examples are listed in Table 2 and Table 3.

Health spending

Health examination surveys can help improve the transparency and accountability of government health spending. Countries without health examination survey data on basic health indicators often base policy decisions on weak data or on imprecise estimates borrowed from other countries. Moreover, a lack of data makes it almost impossible to evaluate and justify budgets. Among the different branches of government, the health sector has always had the greatest difficulty in demonstrating the social impact of interventions and the efficiency of investments. This lack of evidence may be one reason why some countries have underfinanced health budgets. Serial cross-sectional health examination surveys can help assess trends in health indicators and the impact of policy. Without this information, health spending may lack transparency and accountability, which could affect the population’s confidence in the health authority.

Data quality

As the use of health informatics and artificial intelligence to analyse health-care records and administrative health data is increasing, decision-makers could ask, why not replace health examination surveys with these or other data sources? However, the data obtained by health providers are usually captured using non-standardized formats: for example, different definitions of risk factor exposure or disease may be used and not everyone may be asked the same questions, as is done in surveys. Moreover, people may be receiving treatment simultaneously from different providers. In the noncommunicable disease era in particular, an individual’s journey across health and disease states is complex and difficult to register.

Health provider data usually underestimate noncommunicable disease morbidity because they exclude people who are: (i) asymptomatic; (ii) at risk of a disease but unaware of their status; (iii) symptomatic but have not sought help; (iv) have been diagnosed with a disease but have not attended regular check-ups; or (v) have died without being diagnosed. Table 3 shows examples of where health examination survey data has been used: (i) to provide accurate estimates of disease prevalence at the population level; (ii) to guide the diagnosis, treatment and control of hypertension, diabetes and hypercholesterolaemia; and (iii) to assist in the formulation and evaluation of consequent policies.

In addition, health provider-led data sources may introduce bias. The probability of being systematically recorded in a health-care database, or coded in a standardized registry, is influenced by variables that affect access to care, such as age, sex and socioeconomic and geographical characteristics – variables that are also known health determinants. Consequently, administrative data and individual health records may be biased in ways that make health inequalities appear less severe. Moreover, on occasion, provider-based financial incentives may introduce additional biases.50

One could argue that implementing a mandatory national notification system for chronic diseases could attenuate some of these problems. In fact, this type of surveillance has been used for communicable diseases (e.g. acute meningitis infections or chronic HIV infection) for decades with good results. However, mandatory registries are useful when disease events are infrequent but not when a disease is highly prevalent or during a pandemic. In addition, mandatory health notification systems and registries are expensive and time-consuming to maintain and validate. Although all medical specialists advocate national registries for their specific diseases, common sense tells us that doctors will not be able to complete standardized notifications for all noncommunicable diseases. Moreover, mandatory registries would not solve the problem of undiagnosed disease.

Another challenge for data collection is that health care is increasingly being provided outside traditional care structures, for example by public–private partnerships or nongovernmental organizations (NGOs) or through technology-based self-management. The result is the absence of integrated care and a lack of data interoperability. There is a need, therefore, for universal integrated surveillance of the entire population that would provide data for the design and assessment of population-wide preventive policies and for determining the coverage and performance of individual-based programmes. No matter where the population is receiving care, which health provider is involved (e.g. public, private, a NGO or culturally diverse social services) and whether individuals have adopted a self-help approach, decision-makers should be able to identify the characteristics of the people who are at risk or in need of health services. In addition, governments and health providers must be able to interpret trends in indicators (e.g. coverage of a preventive health intervention or use of a pharmacological treatment for a noncommunicable disease) and to determine whether these trends reflect a real change in people’s behaviour or risk, or are explained by changes in the places where they seek help or are registered.

External quality control

Data from a prescription information system on dispensed drugs do not tell us which drugs have actually been taken or whether they are working effectively. In contrast, health examination survey data can provide real information about compliance and disease control. Today, the opinions of individuals are more important than ever for guiding health services. As health examination surveys draw information directly from a random sample of both users and non-users of health services, they can act as external quality controls for the entire health system. Moreover, as data are not obtained from health providers, confidence in, and the accountability of, the health authority is increased.

Recently there has been some concern that the decreasing response rates observed in health examination surveys in high-income countries may affect their external validity. However, even though response rates are decreasing, they remain higher than for mail or online surveys. Moreover, if the participants’ characteristics and the participation rate are known, study weights can be used to derive nationally representative statistics. When data are obtained instead from several health centres, data quality can be affected by a lack of standardization, interobserver variability or inter-laboratory variations in analytical processes.

Random samples

Although health surveys may change their way of collecting information (e.g. they may become multimodal) and technological advances may improve electronic data capture, the concept of the random sample (associated with a known survey response rate) remains important. Calculations that involve denominators based on the general population are essential for guiding health policy.

Forms of data collection such as open internet surveys and random mobile phone sampling can reach millions of people at a low cost, but response rates cannot be estimated adequately. Moreover, uptake is low (response rates are generally below 30%) and these methods are prone to several selection biases associated, for example, with digital literacy, access to digital technologies or socioeconomic factors. In one project forecasting obesity in Europe,51 the researchers noted that data quality across the continent was highly variable and that fewer than half the countries involved possessed nationally representative, objective data. In addition, random sampling is not possible in open internet surveys because participants are all volunteers. As a result, information on event frequency is obtained only from respondents and true population prevalence rates remain unknown.

The most important characteristic of health examination surveys, in contrast, is their external validity, which enables their findings to be directly applied to the population from which their sample was drawn. When adequately powered, health examination surveys can reveal true heterogeneities in risk factor exposure and inequalities in disease burden throughout the population, which can guide the targeting of subgroups during policy development and implementation. Moreover, as health examination surveys use standardized methods, comparative analyses across countries can be performed, including analyses for Global Burden of Disease studies.

Although information technology and artificial intelligence can help develop new sampling frames and better electronic data capture systems, they cannot fix a non-random sample or non-standardized measurements of risk exposure (e.g. different definitions of a smoker) or disease prevalence (e.g. different definitions of hypertension). Consequently, non-standardized data cannot be adjusted for use in time-series analyses.

Future directions

Governance

To aid quality assurance and transparency, health examination surveys should be outsourced by government health authorities, and executed by independent institutions or academic groups whose unedited final reports and databases should be openly shared with the community. Some governments (e.g. in the United Kingdom) use a tendering process for a series of successive health examination surveys to promote long-term investment by universities, survey providers and laboratories. This type of health examination survey governance stimulates positive developments, including: (i) additional fund-raising by, and collaboration between, universities; (ii) continuous training of field teams and standardization of processes and procedures; (iii) the avoidance of precarious working conditions for survey staff; (iv) career development for research assistants and field staff; (v) investment in technological support for surveys; (vi) investment in the validation of new disease screening instruments (e.g. laboratory tests, examination devices and questionnaires); and (vii) opportunities for teaching and research development.

Legislative support for health examination surveys is also needed. Few countries have a track record of regular wide-ranging health examination surveys, and even fewer have secured their continuity by law. Laws should link health examination surveys directly to the iterative process used by health authorities to establish 5- or 10-year health objectives.

Using some health examination survey indicators to assess government objectives in different sectors (e.g. transport, employment, sports and physical activity, social welfare or the environment) could help promote intersectoral policy coordination and a systems-for-health approach. In addition, such linkages could help justify the financing of health examination surveys.

Financing

Objective assessment of the impact of noncommunicable disease surveillance on health policy development, and of the avoidable costs of noncommunicable diseases could help governments justify their budget for health examination surveys. To date, however, little evidence has been disseminated about the usefulness or impact of health examination surveys or about factors influencing the effectiveness of health examination surveys. Global health initiatives should advocate that every low-resource country has its own health examination survey, and should include financing for these surveys. Initiatives like those developed for global vaccine availability during the coronavirus disease 2019 (COVID-19) pandemic could be established to combat the noncommunicable disease burden and could include: (i) international training for health examination survey project leaders; (ii) the translation of video training material for fieldworkers; and (iii) the low-cost availability of devices used in the field, laboratory materials and the point-of-care analytics used for biophysical examinations.

Methods

Technical support and efforts to standardize surveillance methods, such as WHO’s STEPS model and DHS initiatives, are useful and should strengthen low- and middle-income countries’ ability to collect data on noncommunicable disease indicators. The availability of guidelines on using the results of health examination surveys to support the additional studies needed in the health planning cycle (Table 1) could help. Decisions on whether health examination surveys should be conducted annually or less frequently, or become a continuous programme must be made on a country-by-country basis and take financial restrictions into account.

There is space to improve, and invest in, the technologies used in the field, such as digital data capture; telephone surveys; digital disease screening (including the use of images); laboratory techniques for easier biological sampling; home-based self-sampling or self-examination; and point-of-care laboratory analytics. In the future, health examination surveys could make use of new cancer risk biomarkers, of easier and less expensive biochemical assessments of nutritional status in the general population, and of new environmental exposure biomarkers.

All health examination surveys should be designed to include consent for data linkage with vital statistics and health records. In addition, to increase analytical capacity, ministries of health should develop adequate anonymization techniques and ensure data sets are freely available. The use of standardized questions and protocols in surveys will enable comparative analyses to be conducted over time and across countries. Response rates in household surveys could be increased through social marketing and by improving community literacy, both activities that could also be supported by global initiatives.

Conclusions

In this paper, we have added technical arguments to the debate about the importance of health examination surveys and their continued use during the post-pandemic health financing crisis, and in the context of a technological innovation boom that offers new ways of collecting and analysing individual health data. These technical arguments demonstrate that health examination surveys make an irreplaceable contribution to the local availability of primary health data that can be used in a range of further studies (e.g. burden-of-disease, cost and policy impact studies) which are essential for informing several phases of the health planning cycle (e.g. surveillance, prioritization, resource mobilization and policy development). There are solid reasons why health provider-led data sources cannot replace health examination surveys: they may underestimate morbidity and are susceptible to several types of bias. Moreover, the use of nationally representative, random samples of the general population is crucial for maintaining the external validity of any survey. In addition, health examination surveys can provide an external quality control for a country’s health system, thereby helping to ensure that the health authority’s expenditure is transparent and accountable. In the future, fieldwork for health examination surveys will probably be improved by incorporating technological innovations. Global efforts are required to help low-income countries develop the health examination surveys they need to guide policy development, implementation and assessment.

Acknowledgements

We thank J Jaime Miranda, all individuals who provided examples to the research team and all participants in the health examination surveys, the staff who conducted them and the commissioners who funded them. PM and HT are joint first authors. JSM and OO are joint last authors. PM was supported by the Centro para la Prevención y el Control del Cáncer CECAN (FONDAP 152220002).

Competing interests:

None declared.

References

  • 1.Tolonen H, Koponen P, Al-Kerwi A, Capkova N, Giampaoli S, Mindell J, et al. ; EHES Network. European health examination surveys – a tool for collecting objective information about the health of the population. Arch Public Health. 2018. Jun 28;76(1):38. 10.1186/s13690-018-0282-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Flegal KM, Ogden CL, Fryar C, Afful J, Klein R, Huang DT. Comparisons of self-reported and measured height and weight, BMI, and obesity prevalence from national surveys: 1999–2016. Obesity (Silver Spring). 2019. Oct;27(10):1711–9. 10.1002/oby.22591 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Maukonen M, Männistö S, Tolonen H. A comparison of measured versus self-reported anthropometrics for assessing obesity in adults: a literature review. Scand J Public Health. 2018. Jul;46(5):565–79. 10.1177/1403494818761971 [DOI] [PubMed] [Google Scholar]
  • 4.European health interview survey. Luxembourg: Eurostat; 2024. Available from: https://ec.europa.eu/eurostat/web/microdata/european-health-interview-survey [cited 2024 Mar 13].
  • 5.STEPwise approach to NCD risk factor surveillance (STEPS). Geneva: World Health Organization; 2024. Available from: https://www.who.int/teams/noncommunicable-diseases/surveillance/systems-tools/steps [cited 2021 Aug 13].
  • 6.The DHS Program [internet]. Rockville: ICF; 2024. Available from: https://dhsprogram.com/ [cited 2024 Mar 13].
  • 7.Borodulin K, Tolonen H, Jousilahti P, Jula A, Juolevi A, Koskinen S, et al. Cohort profile: the national FINRISK study. Int J Epidemiol. 2018. Jun 1;47(3):696–696i. 10.1093/ije/dyx239 [DOI] [PubMed] [Google Scholar]
  • 8.About the National Health and Nutrition Examination Survey. Hyattsville: National Center for Health Statistics: 2024. Available from: https://www.cdc.gov/nchs/nhanes/about_nhanes.htm [cited 2024 May 20].
  • 9.Mindell JS, Giampaoli S, Goesswald A, Kamtsiuris P, Mann C, Männistö S, et al. ; HES Response Rate Group. Sample selection, recruitment and participation rates in health examination surveys in Europe – experience from seven national surveys. BMC Med Res Methodol. 2015. Oct 5;15(1):78. 10.1186/s12874-015-0072-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Mindell JS, Moody A, Vecino-Ortiz AI, Alfaro T, Frenz P, Scholes S, et al. Comparison of health examination survey methods in Brazil, Chile, Colombia, Mexico, England, Scotland, and the United States. Am J Epidemiol. 2017. Sep 15;186(6):648–58. 10.1093/aje/kwx045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Cuschieri S, Pallari E, Terzic N, Alkerwi A, Sigurðardóttir ÁK. Mapping the burden of diabetes in five small countries in Europe and setting the agenda for health policy and strategic action. Health Res Policy Sys. 2021. Mar 29;19(1):43. 10.1186/s12961-020-00665-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Oyebode O, Mindell J. Use of data from the Health Survey for England in obesity policy making and monitoring. Obes Rev. 2013. Jun;14(6):463–76. 10.1111/obr.12024 [DOI] [PubMed] [Google Scholar]
  • 13.Oyebode O, Mindell JS. A review of the use of health examination data from the Health Survey for England in government policy development and implementation. Arch Public Health. 2014. Jul 15;72(1):24. 10.1186/2049-3258-72-24 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Kilpeläinen K, Koponen P, Tolonen H, Koskinen S, Borodulin K, Gissler M. From monitoring to action: utilising health survey data in national policy development and implementation in Finland. Arch Public Health. 2019. Nov 13;77(1):48. 10.1186/s13690-019-0374-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Margozzini P, Passi Á. Encuesta Nacional de Salud, ENS 2016–2017: un aporte a la planificación sanitaria y políticas públicas en Chile. ARS Medica Revista de Ciencias Médicas. 2018;43(1):30–4. Spanish. 10.11565/arsmed.v43i1.1354 [DOI] [Google Scholar]
  • 16.Política nacional de alimentación y nutrición. Santiago: Ministerio de Salud; 2018. Spanish. Available from: http://www.bibliotecaminsal.cl/politica-nacional-de-alimentacion-y-nutricion [cited 2024 Mar 13].
  • 17.Petermann-Rocha F, Martínez-Sanguinetti MA, Ho FK, Celis-Morales C, Pizarro A; ELHOC Research Group. Optimal cut-off points for waist circumference in the definition of metabolic syndrome in Chile. Public Health Nutr. 2020. Nov;23(16):2898–903. 10.1017/S1368980020001469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Technical consultation document. Department of Health and Social Care (DHSC) calorie model. London: Department of Health and Social Care; 2018. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/736417/dhsc-calorie-model-technical-document.pdf [cited 2024 Feb 8].
  • 19.Weighing the costs of obesity in Malta. Valletta: PwC; 2017:60. Available from: https://www.pwc.com/mt/en/publications/weighing-the-costs-of-obesity.html [cited 2023 May 30].
  • 20.Busso D, Echeverría G, Passi-Solar A, Morales F, Farías M, Margozzini P. Folate status in women of childbearing age in the urban metropolitan region of Chile: results from the National Health Survey 2016–2017. Public Health Nutr. 2021. Feb;24(3):385–92. 10.1017/S1368980020002608 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Rivera-Dommarco JÁ, Cuevas-Nasu L, González de Cosío T, Shamah-Levy T, García-Feregrino R. Desnutrición crónica en México en el último cuarto de siglo: análisis de cuatro encuestas nacionales. Salud Publica Mex. 2013;55 Suppl 2:S161–9. Spanish. 10.21149/spm.v55s2.5112 [DOI] [PubMed] [Google Scholar]
  • 22.Morales-Ruán MDC, Shamah-Levy T, Mundo-Rosas V, Cuevas-Nasu L, Lozada-Tequeanes AL, Romero-Martínez M. Evolución de los programas de ayuda alimentaria en México a través de información de Ensanut MC 2016. Salud Publica Mex. 2018 May-Jun;60(3):319–27. Spanish. 10.21149/8818 10.21149/8818 [DOI] [PubMed]
  • 23.Jääskeläinen T, Itkonen ST, Lundqvist A, Erkkola M, Koskela T, Lakkala K, et al. The positive impact of general vitamin D food fortification policy on vitamin D status in a representative adult Finnish population: evidence from an 11-y follow-up based on standardized 25-hydroxyvitamin D data. Am J Clin Nutr. 2017. Jun;105(6):1512-20. 10.3945/ajcn.116.151415 [DOI] [PubMed] [Google Scholar]
  • 24.Ashford R, Jones K, Collins D, Earl K, Moore S, Koulman A, et al. National Diet and Nutrition Survey. Assessment of salt intake from urinary sodium in adults (aged 19 to 64 years) in England, 2018 to 2019. London: Public Health England; 2020. Available from: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/876252/Report_England_Sodium_Survey_2018-to-2019__3_.pdf [cited 2021 Aug 13].
  • 25.Donfrancesco C, Lo Noce C, Russo O, Minutoli D, Di Lonardo A, Profumo E, et al. Trend of salt intake measured by 24-h urine collection in the Italian adult population between the 2008 and 2018 CUORE project surveys. Nutr Metab Cardiovasc Dis. 2021. Mar 10;31(3):802–13. 10.1016/j.numecd.2020.10.017 [DOI] [PubMed] [Google Scholar]
  • 26.Plan nacional para la reducción y control de la anemia materno-infantil y la desnutrición crónica infantil en el Perú: 2017–2021 (Documento técnico). Lima: Ministerio de Salud; 2017. Spanish. [Google Scholar]
  • 27.Loret de Mola C, Quispe R, Valle GA, Poterico JA. Nutritional transition in children under five years and women of reproductive age: a 15-years trend analysis in Peru. PLoS ONE. 2014. Mar 18;9(3):e92550. 10.1371/journal.pone.0092550 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Pérez-Lu JE, Cárcamo C, Nandi A, Kaufman JS. Health effects of ‘Juntos’, a conditional cash transfer programme in Peru. Matern Child Nutr. 2017. Jul;13(3):e12348. 10.1111/mcn.12348 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kuczmarski R, Ogden C, Guo S, Grummer-Strawn L, Flegal KM, Mei Z, et al. 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11. 2002. May;(246):1-190. . [PubMed] [Google Scholar]
  • 30.Centers for Disease Control and Prevention (CDC). Folate status in women of childbearing age, by race/ethnicity – United States, 1999–2000. MMWR Morb Mortal Wkly Rep. 2002. Sep 13;51(36):808–10. [PubMed] [Google Scholar]
  • 31.Childhood lead poisoning prevention. National Health and Nutrition Examination Survey (NHANES). Blood lead levels in the U.S. population. Atlanta: Centers for Disease Control and Prevention; 2021. Available from: https://www.cdc.gov/nceh/lead/data/nhanes.htm [cited 2024 Feb 8].
  • 32.Passi-Solar Á, Margozzini P, Mindell JS, Ruiz M, Valencia-Hernandez CA, Scholes S. Hypertension care cascade in Chile: a serial cross-sectional study of national health surveys 2003–2010–2017. BMC Public Health. 2020. Sep 14;20(1):1397. 10.1186/s12889-020-09483-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Falaschetti E, Mindell J, Knott C, Poulter N. Hypertension management in England: a serial cross-sectional study from 1994 to 2011. Lancet. 2014. May 31;383(9932):1912–9. 10.1016/S0140-6736(14)60688-7 [DOI] [PubMed] [Google Scholar]
  • 34.Carrillo-Larco RM, Guzman-Vilca WC, Bernabe-Ortiz A. Mean blood pressure according to the hypertension care cascade: analysis of six national health surveys in Peru. Lancet Reg Health Am. 2021. Sep;1:None. 10.1016/j.lana.2021.100016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Campos-Nonato I, Hernández-Barrera L, Pedroza-Tobías A, Medina C, Barquera S. Hipertensión arterial en adultos mexicanos: prevalencia, diagnóstico y tipo de tratamiento. Ensanut MC 2016. Salud Publica Mex. 2018 May-Jun;60(3):233-43. Spanish. 10.21149/8813 . 10.21149/8813 [DOI] [PubMed]
  • 36.Giampaoli S, Palmieri L, Donfrancesco C, Lo Noce C, Pilotto L, Vanuzzo D; Osservatorio Epidemiologico Cardiovascolare/Health Examination Survey Research Group. Cardiovascular health in Italy. Ten-year surveillance of cardiovascular diseases and risk factors: Osservatorio Epidemiologico Cardiovascolare/health examination survey 1998–2012. Eur J Prev Cardiol. 2015. Sep;22(2 Suppl):9–37. 10.1177/2047487315589011 [DOI] [PubMed] [Google Scholar]
  • 37.Donfrancesco C, Di Lonardo A, Lo Noce C, Buttari B, Profumo E, Vespasiano F, et al. Trends of blood pressure, raised blood pressure, hypertension and its control among Italian adults: CUORE Project cross-sectional health examination surveys 1998/2008/2018. BMJ Open. 2022. Nov 14;12(11):e064270. 10.1136/bmjopen-2022-064270 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Nazroo JY, Falaschetti E, Pierce M, Primatesta P. Ethnic inequalities in access to and outcomes of healthcare: analysis of the Health Survey for England. J Epidemiol Community Health. 2009. Dec;63(12):1022–7. 10.1136/jech.2009.089409 [DOI] [PubMed] [Google Scholar]
  • 39.Lindström J, Tuomilehto J. The diabetes risk score: a practical tool to predict type 2 diabetes risk. Diabetes Care. 2003. Mar;26(3):725–31. 10.2337/diacare.26.3.725 [DOI] [PubMed] [Google Scholar]
  • 40.Cuschieri S. The diabetes epidemic in Malta. S East Eur J Public Health. 2020. Feb 19;XIII:2020. 10.4119/seejph-3322 [DOI] [Google Scholar]
  • 41.Carrillo-Larco RM, Pearson-Stuttard J, Bernabe-Ortiz A, Gregg EW. The Andean Latin-American burden of diabetes attributable to high body mass index: a comparative risk assessment. Diabetes Res Clin Pract. 2020. Feb;160:107978. 10.1016/j.diabres.2019.107978 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Carrillo-Larco RM, Guzman-Vilca WC, Bernabe-Ortiz A. Who is getting screened for diabetes according to body mass index and waist circumference categories in Peru? A pooled analysis of national surveys between 2015 and 2019. PLoS One. 2021. Aug 27;16(8):e0256809. 10.1371/journal.pone.0256809 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Vartiainen E, Laatikainen T, Peltonen M, Puska P. Predicting coronary heart disease and stroke: the FINRISK calculator. Glob Heart. 2016. Jun 1;11(2):213–6. 10.1016/j.gheart.2016.04.007 [DOI] [PubMed] [Google Scholar]
  • 44.Kivipelto M, Ngandu T, Laatikainen T, Winblad B, Soininen H, Tuomilehto J. Risk score for the prediction of dementia risk in 20 years among middle aged people: a longitudinal, population-based study. Lancet Neurol. 2006. Sep;5(9):735–41. 10.1016/S1474-4422(06)70537-3 [DOI] [PubMed] [Google Scholar]
  • 45.Saint-Maurice PF, Troiano RP, Matthews CE, Kraus WE. Moderate‐to‐vigorous physical activity and all‐cause mortality: Do bouts matter? J Am Heart Assoc. 2018. Mar 22;7(6):e007678. 10.1161/JAHA.117.007678 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Physical activity guidelines for Americans, 2nd edition. Washington, DC: United States Department of Health and Human Services; 2018. Available from: https://health.gov/sites/default/files/2019-09/Physical_Activity_Guidelines_2nd_edition.pdf [cited 2024 Feb 8].
  • 47.Roth MA, Roderick PJ, Mindell J. Kidney disease and renal function. In: Craig R, Mindell J, editors. Health Survey for England, 2010. Leeds: The Health and Social Care Information Centre; 2011. [Google Scholar]
  • 48.Mosso L, Margozzini P, Trejo P, Domínguez A, Solari S, Valdivia G, et al. Aproximación al patrón de normalidad de TSH para la población chilena según Encuesta Nacional de Salud 2009–2010. Rev Med Chil. 2013. Jan;141(1):95–103. Spanish. 10.4067/S0034-98872013000100013 [DOI] [PubMed] [Google Scholar]
  • 49.Public health outcomes framework. Indicator definitions and supporting information. London: Office for Health Improvement & Disparities; 2024. Available from: https://fingertips.phe.org.uk/profile/public-health-outcomes-framework/data#page/6/gid/1000042/pat/15/par/E92000001/ati/502/are/E09000002/iid/93088/age/168/sex/4/cat/-1/ctp/-1/yrr/1/cid/4/tbm/1/page-options/car-do-0 [cited 2024 May 20].
  • 50.Mansilla A C, Montero L J, Majluf S N, Rojas V MP. Comparación de datos regionales de efectividad de tratamiento de HTA según la ENS e informes REM. Rev Med Chil. 2013. Sep;141(9):1222–3. Spanish. 10.4067/S0034-98872013000900020 [DOI] [PubMed] [Google Scholar]
  • 51.Pineda E, Sanchez-Romero LM, Brown M, Jaccard A, Jewell J, Galea G, et al. Forecasting future trends in obesity across Europe: the value of improving surveillance. Obes Facts. 2018;11(5):360–71. 10.1159/000492115 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Bulletin of the World Health Organization are provided here courtesy of World Health Organization

RESOURCES