Recent years have brought a host of new initiatives relating to adolescents' health. The Lancet Commission on Adolescent Health and Wellbeing [[2]Patton G.C. Sawyer S.M. Santelli J.S. et al.Our future: A Lancet commission on adolescent health and wellbeing.], the Global Accelerated Action for the Health of Adolescents [[1]World Health Organization
Global Accelerated action for the health of adolescents (AA-HA!): Guidance to support country implementation.], the Global Strategy for Women's, Children's, and Adolescents' Health (2016–2030) [[9]Every Woman Every Child
The global Strategy for Women’s, Children’s and adolescents’ health (2016-2030).], and Countdown to 2030 [[10]Countdown to 2030. Women's, Children's and adolescents' health.] all represent a critical step forward in emphasizing the significance of the adolescent period. The importance of measurement has been highlighted in these initiatives. The Lancet Commission, for example, emphasized the role of indicators in aligning response to needs and ensuring accountable actions. Each initiative, however, has proposed its own set of indicators for tracking progress.These efforts, together with numerous others, address what has been perceived as a long-standing deficit of indicators related to adolescents globally [[11]Guthold R. Moller A.B. Azzopardi P. et al.The global action for measurement of adolescent health (GAMA) initiative-Rethinking adolescent Metrics.] by highlighting critical topics and advancing adolescent-specific indicators such as those relating to life satisfaction and mental health. Although these developments have helped to advance the long-neglected adolescent measurement field, the increase in both proposed indicators and measurement initiatives, however, has come about with little coordination, as previously noted [[12]Guthold R. Moller A.B. Adebayo E. et al.Priority areas for adolescent health measurement.,[13]Gaps in global monitoring and evaluation of adolescent and youth reproductive health.]. The result has been inconsistent across initiatives, including long-standing indicators, as well as a substantial number of new indicators, some of which are poorly defined and difficult to measure, whereas gaps in the topics addressed remain.To strengthen coordination, technical standards, and capacity for adolescent health measurement, the World Health Organization (WHO), in collaboration with United Nations (UN) H6+ partner agencies (Joint UN Program on HIV/AIDS, UN Educational, Scientific and Cultural Organization, UN Population Fund, UN Children's Fund, UN Women, the World Bank Group, and the World Food Program), established the Global Action for Measurement of Adolescent Health (GAMA) Advisory Group, consisting of 17 global adolescent health experts [[11]Guthold R. Moller A.B. Azzopardi P. et al.The global action for measurement of adolescent health (GAMA) initiative-Rethinking adolescent Metrics.,[14]World Health Organization
The global action for measurement of adolescent health (GAMA).]. A primary goal of the GAMA Advisory Group is to identify a set of priority adolescent health indicators, on which robust measurement standards will be built and harmonized guidance will be developed. Two work streams were determined as necessary first steps toward this goal: (1) identify measurement domains and identify core areas for adolescent health measurement globally [[12]Guthold R. Moller A.B. Adebayo E. et al.Priority areas for adolescent health measurement.] and (2) compile available indicators related to these core measurement areas. The purpose of this article is to map existing indicators for adolescent health, assess what they measure, and highlight overlap across initiatives and measurement gaps.ResultsThe compilation comprised a total of 413 separately listed indicators identified across 16 initiatives, including 44 (11%) categorized as primarily relating to the domain determinants of health, 136 (33%) relating to health behaviors and risks, 49 (12%) categorized under policies, programs, and laws, 22 (5%) under systems performance and interventions, and 162 (39%) primarily pertaining to health outcomes and conditions [[21]Newby H. Marsh A.D. Moller A.-B. et al.GAMA adolescent health indicator mapping.]. Indicators spanned 32 of the 33 core measurement areas as previously identified by GAMA, with no indicator pertaining to the core measurement area disability (Table 2).The majority of the reviewed measurement initiatives covered multiple topical areas, and yet, none encompassed more than two-thirds of the 33 GAMA core adolescent health measurement areas (Table 3). The core measurement areas commonly addressed across initiatives were HIV/AIDS and adolescent fertility rate (represented by 12 initiatives each) as well as weight status, reproductive health, cause-specific mortality, and interpersonal violence (represented by 11 initiatives each). On the other hand, numerous core measurement areas were addressed by fewer than five initiatives, including income level and poverty, bullying, and immunization (represented by four initiatives each); population, system for monitoring and surveillance of adolescent health, and anxiety disorders (represented by three initiatives each); health service quality (represented by two initiatives); and disability, which no initiative addressed.

Table 3GAMA core measurement areas covered by reviewed initiatives

Dark blue = The initiative includes one or more indicators with a primary focus on the listed measurement area. Light blue = The initiative includes no indicators with a primary focus on the listed measurement area but includes one or more indicators that are cross-listed with this area. Abbreviations: ADH = Adolescent health; GAMA = Global Action for Measurement of Adolescent Health; M&E = Monitoring and evaluation; STI = sexually transmitted infection.

Analysis in accordance with the age group was hampered by insufficient details in the underlying metadata. The compilation included a total of 352 entries based on individuals as the unit of measurement (e.g., as opposed to schools or policies ). Based on available information, we were able to classify only 221 indicators by age, which is less than two-thirds. Thus, although it was clear from the initiative or the indicator name that the indicator was related to the adolescent age range, we were unable to classify 126 individual-level indicators as pertaining to a specific age group because the underlying metadata did not include sufficient information regarding age.

For those 221 indicators that we were able to classify by age, the analysis revealed substantial variation in the age ranges different indicators were intended to measure. Less than one-quarter – 51 indicators – pertained specifically to the full adolescent age group of 10–19 years. In addition, just one indicator was specific to only the entire young adolescent age range (10–14 years) and 27 captured the entire older adolescent age range (15–19 years). Beyond the indicators pertaining to specific adolescent age groups, many indicators encompassed years well beyond the adolescent age range. Notably, 43 indicators were based on ages 15–49 years, reflecting the reality that many indicators suggested for older adolescents were derived from measures mainly focusing on individuals of reproductive age. Similarly, there were indicators overlapping with adolescence that were intended to measure aspects of children's lives, such as the 11 indicators based on ages 0–17 years (Figure 1).

Figure 1Age coverage across adolescent health indicators. Note: This figure presents the 11 most common age ranges covered by indicators in the GAMA mapping. The figure presents all age ranges that occurred at least five times, accounting for 196 of the 221 indicators (87%) where a specific age range was provided. The ordering and width of each line present the relative frequency with which the age range occurred (least common = thinnest line, bottom of figure; most common = thickest line, top of figure). GAMA, Global Action for Measurement of Adolescent Health

After identifying overlapping indicators, a subset of 236 indicators of the total 413 (57%) were deemed distinct entries. Most core measurement areas contained numerous overlapping indicators, demonstrating that two or more initiatives had identified identical, or at least similar, indicators. There were some notable exceptions, however, such as those core measurement areas falling under the policies, programs, law domain, and the systems performance and intervention domain, which contained indicators that, even if related to some extent, were nonetheless different enough to be considered distinct. The core measurement areas of dietary behavior and road injury were similarly characterized by a lack of overlapping indicators across the initiatives.

DiscussionIncreased focus on adolescent health in recent years has led to welcome advances in the development, promotion, and use of indicators related to this population group, as evidenced by the number of measurement initiatives included in this review. On the other hand, the sheer number of indicators being promoted by these initiatives is overwhelming and impossible to track in a meaningful way. Although many of the separately listed 413 indicators in our review were overlapping, the 236 indicators that were deemed distinct still represent a huge monitoring burden. This finding underscores the need for prioritization, which is one of GAMA's primary goals [[14]World Health Organization
The global action for measurement of adolescent health (GAMA).].Our scoping review revealed deficiencies, gaps, and substantial variations across measurement initiatives' metadata. For instance, specific information regarding age was provided for less than two-thirds of the separately listed individual-based indicators in the compilation, and where information on age was provided, it often varied across indicators. The adolescent birth rate provides an illustrative example of the challenges: of the 12 separately listed indicators, five referred only to ages 15–19 years; 1 referred to ages 10–14 years; three referred to both ages 10–14 and 15–19 years; two were internally inconsistent, meaning the age-pertinent details in the name and definition were not aligned; and 1 indicator was simply named “adolescent birth rate” and lacked any definitional information. This is not only confusing but can also result in inconsistent measurement and hinder comparability. To resolve these issues, work on standardizing age groups in data collection, analysis, and reporting of health data across the life course is underway. This work is led by the WHO and supported not only by GAMA but also by similar advisory groups for other periods of the life course [[20]Moller A.B. Newby H. Hanson C. et al.Measures matter: A scoping review of maternal and newborn indicators.,[22]Strong K. Requejo J. Agweyu A. et al.Child health accountability tracking-extending child health measurement.]. Furthermore, in developing detailed and harmonized measurement guidance for selected priority indicators, GAMA aims at improving consistency in measurement of the most relevant adolescent health indicators [[14]World Health Organization
The global action for measurement of adolescent health (GAMA).].This scoping review also demonstrated that measurement in some of GAMA's previously defined domains and core areas is more advanced than in others. The policies, programs, laws domain, and the systems performance and interventions domain, for example, lack long-established, standardized indicators, and thus, there was little convergence in the indicators recommended by different initiatives. Notably, measurement of policies and systems has long lagged behind health outcomes and conditions or even health behaviors and risks, measurement of which has often been captured in both routine health information systems as well as population-based surveys [[23]Marsh A.D. Muzigaba M. Diaz T. et al.Effective coverage measurement in maternal, newborn, child, and adolescent health and nutrition: Progress, future prospects, and implications for quality health systems.]. As adolescent health measurement advances, it will be important to give special consideration to areas where measurement has historically lagged to ensure proper coordination of relevant tasks and conduction of necessary research as a basis for indicator development.However, measurement domains and areas that are more advanced and contain a large number of sometimes long-standing indicators bring different challenges. Our scoping review suggests that in at least some cases, the inclusion of indicators in initiatives was driven more by the supply of existing indicators – frequently those that are easy to measure – than demand for measurement specific to adolescence. For example, adolescent health indicators in the areas of sexual and reproductive health and contraception, HIV/AIDS, and sexually transmitted infections are characterized by an imbalance: numerous indicators currently in widespread use are simply the disaggregation of those based primarily on women of reproductive age 15–49 years, whereas there are only few indicators pertaining to younger adolescents aged 10–14 years. Similarly, in other areas, indicators that are used in adolescent health may be derived from those originally developed for children younger than 18 years of age and not for adolescents specifically [[18]Azzopardi P. Kennedy E. Patton G. Data and indicators to measure adolescent health, social development and well-being.]. Although an indicator originally developed for a broader age group may be perfectly fit for adolescent measurement, this is a point that warrants consideration both in terms of the indicator itself and its operationalization. Generally, in areas where measurement is advanced, there is often a need to review indicators, ensuring they are adequate to capture the adolescent experience, and then to harmonize methodologies, important tasks that GAMA will pay special attention to as its work is being advanced.Several important measurement gaps were revealed through our scoping review. Interestingly, they were not limited to domains and areas lagging behind in measurement but were also present in areas containing a large number of long-standing indicators. This finding is consistent with a recent review showing that more than 800 indicators including variations of supposedly the same indicator were used in adolescent reproductive health [[13]Gaps in global monitoring and evaluation of adolescent and youth reproductive health.], but there were still data deficiencies for unmarried youth, adolescent boys, and very young adolescents and in specific areas such as abortion, nonheterosexual behavior, or fertility intentions. Core measurement areas that appear to lack a sufficient number of indicators to broadly reflect the main concepts include, for instance, substance use other than alcohol and tobacco, health service quality, and, in particular, disability, for which none of the initiatives included an indicator. Furthermore, we also found a general lack of indicators for young adolescents and those not attending school [[18]Azzopardi P. Kennedy E. Patton G. Data and indicators to measure adolescent health, social development and well-being.].The identified measurement gaps need to be filled immediately, understanding that gaps in indicator availability may be driven by a variety of factors. In certain cases, gaps may exist owing to long-standing measurement practices. For example, much available adolescent data are commonly collected in household surveys that focus on interviewing women aged 15–49 years, and thus data pertaining to women aged 10–14 years are not collected. Another common source of data is school-based surveys, which miss the out-of-school population. Complexity of measurement can present a different sort of methodological challenge resulting in a gap: robust measurement of disability, for example, requires a lengthy set of questions. Finally, the long-standing deficit of attention given to adolescent health means that some topics require more methodological work to develop robust indicators and measurement approaches. Unless these gaps can be systematically addressed, program managers and policy makers will necessarily base decisions on a partial – and potentially biased – picture of the adolescent population. The way forward must include innovative thinking around different options for collecting data on all adolescents – for example, stand-alone surveys or specialized questionnaires integrated into existing questionnaires [[24]UNICEF
Home - UNICEF MICS.] – and validation of existing indicators together with methodological work to develop new measurement approaches.This study highlighted numerous challenges in trying to compile and classify indicators in a standardized way, and there are limitations to our results. First, given imprecise, contradictory, or missing metadata, categorization of indicators is necessarily imperfect. Repeated listing of what may have initially appeared to be the same indicator resulted in numerous differences between entries, starting with inconsistency in the name and, importantly, the specified age groups. This was true even for Sustainable Development Goals indicators [[25]Inter-Agency and Expert Group on Sustainable Development Goal Indicators Revised list of global sustainable development goal indicators.], which are supported by detailed metadata and we originally assumed would be standardized across initiatives. Second, it was sometimes difficult to fully understand what the indicator was intended to measure based on the information provided by the initiative. In cases where no definitional information was provided, our classification had to solely rely on the indicator names, which varied greatly in detail. Thus, identifying overlapping indicators – those that are either true duplicates or have the same measurement intent but varying slightly in terms of the name and/or definition – was an inherently fraught exercise, although one we still felt was crucial to undertake. Entering all indicators into the compilation as we did, instead of trying to reconcile and synthesize during compilation, has the important advantage that differing definitions and operationalization for the same measurement concept can be reviewed and considered during the selection of priority indicators, which will in turn result in a more robust set of measures and guidance.Third, this indicator compilation reflects a subset of existing indicators – those that were contained within the initiatives included in this scoping review – and thus presents a partial picture of the measurement landscape at a particular point of time. We identified initiatives via expert consultation and did not conduct a full search of the literature or include any initiatives at the country level. These global-level measurement initiatives may not reflect the full range of measures that are available. Notably, there are rapid developments in the field, evidenced by the recently debuted, United Nations International Children’s Emergency Fund–led Measurement of Mental Health among Adolescents at the Population Level initiative [[26]UNICEF
Measurement of mental health among adolescents at the population level (MMAP).], which was subsequently added to our initial list of 15 initiatives. Furthermore, in terms of topical scope, we only considered 33 previously defined core measurement areas. Although this may have skewed our findings, it enabled us to focus our work. Notably, the 33 measurement areas were systematically identified by adolescent health experts considering not only the adolescent mortality and morbidity burden [[12]Guthold R. Moller A.B. Adebayo E. et al.Priority areas for adolescent health measurement.] but also input from young people and countries and we therefore believe that our mapping indeed includes the most relevant adolescent health issues.This indicator compilation serves as an important basis for GAMA's continued work of reviewing existing indicator recommendations, systematically selecting priority indicators for each core measurement area, and ensuring that each selected priority indicator is supported by clear, consistent, and comprehensive metadata. This indicator selection process is currently underway and draws on existing guidance [[18]Azzopardi P. Kennedy E. Patton G. Data and indicators to measure adolescent health, social development and well-being.,27Benova L. Moller A.-B. Moran A.C. “What gets measured better gets done better”: The landscape of validation of global maternal and newborn health indicators through key informant interviews., 28MEASURE Evaluation
Selection of indicators., 29World Health Organization
Health in sustainable development planning : The role of indicators/Yasmin von Schirnding.]. As part of this work, measurement gaps are being identified and will be put forward for further research, including exploring any potential barriers to the widespread adoption of these indicators. The GAMA priority indicators will then be promoted for use in countries and regularly revisited as new evidence emerges and as measurement gaps are filled. This long-term process will require continued investment in advocacy and careful consideration of both existing measurement systems and the country-specific contextual factors. To that end, the selection of GAMA priority indicators is undertaken in collaboration with a broad range of stakeholders, including UN H6+ partnership agencies, key measurement groups, country-level implementing partners, and the public.

The adolescent health measurement landscape is vast and includes substantial variation among indicators purportedly assessing the same concept. Important gaps persist, with a lack of robust, standardized indicators measuring systems performance and interventions, policies and laws, and the needs of men, those with disabilities, and younger adolescents, specifically. Addressing these gaps and harmonizing measurement approaches will be fundamental to improving policy and program implementation as well as accountability for adolescent health globally. Margaret Chan, the former Director-General of the WHO, has said repeatedly: “what gets measured gets done.” Consistent measurement is the first step to understanding the needs of adolescents across countries and over time, thus providing policy makers with actionable information that will enable them to improve the health of adolescents in their countries.

Article InfoPublication History

Published online: July 13, 2021

Accepted: April 18, 2021

Received: December 18, 2020

Publication stageIn Press Corrected ProofFootnotes

Conflicts of Interest: The authors have no conflicts of interest to disclose.

Author statements: HN, ADM, ABM, and RG are members of the World Health Organization. LC is a member of the United Nations Children's Fund. HSF is a member of the United Nations Population Fund. AM is a member of the World Bank Group. The authors alone are responsible for the views expressed in this publication, and they do not necessarily represent the decisions, policy, or views of the World Health Organization, of the United Nations Children's Fund , of the United Nations Population Fund , or of the World Bank Group.

Identification

DOI: https://doi.org/10.1016/j.jadohealth.2021.04.026

Copyright

© 2021 Society for Adolescent Health and Medicine. Published by Elsevier Inc.

User License Creative Commons Attribution (CC BY 4.0) | ScienceDirectAccess this article on ScienceDirect