Doing More With Less: Advancing a Contextualized Understanding of Human Biology With Minimally-Invasive Approaches to Capillary Blood Sampling

Abstract

In 2014 I published “Development and validation of assay protocols for use with dried blood spot samples” in the American Journal of Human Biology (McDade 2014). It appeared as part of the AJHB “Toolkit: Methods in Human Biology” series, a newly established mechanism for maintaining a virtual methods handbook that tracks new research directions, and provides up-to-date protocols for important, long-standing methods (Ellison and McDade 2012). I served as inaugural series editor, and took advantage of a lull in the Toolkit pipeline to contribute an article on the advantages and disadvantages of dried blood spot (DBS) sampling, and to share detailed “how-to” information that I hoped would encourage more colleagues to develop assays for use with DBS samples. With this commentary, I am grateful for the opportunity to reflect on developments over the past 10 years, but I also aim to highlight the critical role that field-friendly methods like DBS sampling play in advancing a more holistic and contextualized understanding of human biology and health.

I am reminded how important this role is each time I teach my introductory undergraduate course on social inequalities and health. Before the first class I ask the students to complete an online survey with the following question: “There can be many causes of problems with a person's health. What do you think are the three most important things that determine someone's health?” I compile the responses into a word cloud to generate discussion on the first day. It probably will not come as a surprise that “genes” and “genetics,” as well as “lifestyle choices” like “diet,” “exercise,” and “smoking” feature prominently in the responses. I point out that the students are privileging individual-level determinants that are either inherited and fixed, or health-related behaviors that imply personal responsibility. But human biology is a contingent biology, and most students are surprised to learn that the broader social and physical worlds we inhabit have powerful effects on our bodies, and that they activate multiple molecular, physiological, behavioral, and neurological pathways to influence our health.

There are many historical, political-economic, and cultural reasons why we—particularly in the United States—favor explanatory models of health that focus on individual action and responsibility (Lewontin and Levins 2007). There are epistemological ones as well, drawing on and reinforcing assumptions regarding appropriate study designs and measurement protocols for the production of knowledge about the causes of health and disease. Simply put, methods play a critical role in defining how we study and conceptualize human health. And if we only measure health-related systems in experimental animal models or in clinical settings attached to academic medical complexes, we will have a very narrow and de-contextualized perspective on the human body and how it (mal)functions (Stinson et al. 2012; Gurven and Lieberman 2020; Wiley 2021; McDade and Harris 2022). Of course, this is old news to biological anthropologists, for whom methodological innovation has been foundational to bridging the lab and the field, to conducting research in diverse community-based settings globally that foreground ecological variation as an important determinant of human biological variation (Ellison 1988; James 1991; Worthman and Stallings 1997; McDade, Stallings, and Worthman 2000; O'Connor et al. 2003; Leonard 2012; Salvante et al. 2012; Snodgrass 2022; Urlacher et al. 2022). The collection and analysis of DBS samples have been—and will continue to be—a central part of this effort.

Early in my graduate training (1993–1999) I recognized that methodological constraints would make it difficult for me to conduct field-based research on the human immune system. I had the incredible privilege of working closely with Carol Worthman and Joy Stallings in the Laboratory for Comparative Human Biology at Emory University during the time that they were systematically developing DBS assays to advance research in human reproductive ecology (Worthman and Stallings 1994, 1997). With considerable patience and wisdom, they oversaw my training in the lab and supported my efforts to validate DBS methods for C-reactive protein (CRP) and Epstein–Barr virus antibodies (McDade, Stallings, and Worthman 2000; McDade et al. 2004; McDade et al. 2000a), both of which featured in my dissertation and helped me launch the Laboratory for Human Biology Research at Northwestern in 2000.

Many people are not aware that DBS sampling has been a central part of public health infrastructure in the US since the 1960s, serving as the foundation of hospital-based newborn screening programs. Each year, more than 3.5 million DBS samples are collected from the heels of newborns and tested for congenital metabolic disorders (Centers for Disease Control and Prevention 2024). The filter papers (Whatman 903 Proteinsaver card and Ahlstrom 226 BioSample card) are specifically designed to meet performance standards for sample absorption and lot-to-lot consistency, and are subject to quality control monitoring by the CDC (Mei et al. 2001). In short, DBS sampling is a well-established and rigorously validated approach to collecting blood that has been widely used in the US for more than 60 years.

The features that make DBS sampling well suited to newborn screening translate over to research settings where venous blood collection is problematic (McDade et al. 2007; Worthman and Stallings 1994). The costs and burdens of DBS sampling are minimal: A sterile, single-use lancet is used to prick the heel or finger, and up to 5 drops of capillary blood are absorbed onto filter paper. The procedure can be safely implemented by nonmedically trained research assistants in a participant's home or a centralized site in the community, or in some cases, by participants themselves. Once applied to filter paper, the blood dries and is stabilized during transport and storage, eliminating the need for sample processing (e.g., centrifuging, separating, aliquoting of serum/plasma) in the field. In addition, DBS samples can be stored at room temperature for several days, weeks, or even years (depending on the specific analytes of interest), and a cold chain is not needed between the site of collection and the lab where samples will be stored and analyzed.

These attributes greatly reduce logistical hassles, costs, and biohazard risks associated with sample collection, handling, and transport. More concretely, the supplies needed to collect a DBS sample currently cost approximately $5. In comparison, the clinical research unit at Northwestern estimates $88 for basic outpatient phlebotomy and blood sample processing. In addition to cost savings, the ability to safely collect blood in the home makes it possible to orient blood collection around the research participant, rather than a medical research facility. For example, in focus groups discussing the acceptability of blood testing, urban, low-income women expressed a preference for in-home, finger-stick blood collection implemented by trained community-based interviewers, as opposed to arranging visits to phlebotomy clinics (Borders et al. 2007).

But these advantages in the field need to be weighed against complications associated with analyzing DBS samples in the lab (McDade 2014). First, the volume of blood is much lower than what is generally collected through venipuncture, and the sample itself is different: Whole blood contains red and white blood cells, whereas processing of venous blood removes these fractions to isolate serum or plasma. The limited sample volume and more complex matrix of DBS samples can impose challenges in the lab, and for some analytes, quantification in DBS may not be possible. We were reminded of this recently when we validated a new multiplex assay for inflammatory cytokines (McDade et al. 2021). For IL6, TNFα, and IL10, results from DBS and plasma samples corresponded nicely, whereas the assay broke down for IL8. As it turns out, red blood cells are a reservoir for IL8 (Darbonne et al. 1991; Karsten et al. 2018), and as the cells lyse on filter paper, they flood the sample with IL8 and obscure quantification of the IL8 fraction typically measured in plasma.

A broader issue is that clinical testing of health-related biomarkers centers on venous blood, and results from serum or plasma are generally considered the gold standard. The number of validated assays for DBS samples is relatively low in comparison, and the intent of my 2014 article was to encourage more researchers to develop, validate, and disseminate protocols for DBS samples. I offered a “how-to” guide, with detailed information on selecting reagents, preparing calibration and quality control materials, optimizing elution protocols, and evaluating assay performance. With more labs working with DBS, my hope was that we could build out a collective “toolkit” for field-based research that would help us keep pace with developments in the biomedical sciences and expand the impact of research in human biology/biological anthropology.

Despite these efforts, I have been surprised by the reticence of many experienced laboratory scientists to engage with DBS. In my mind, a DBS sample is not that different from plasma: After some upfront processing to elute the blood off the paper, the skills, equipment, and laboratory space needed to assay DBS and plasma samples are effectively the same. The availability of validated DBS protocols and labs willing to run them is currently a rate limiter for the field. Fortunately, we have a strong group of investigators in human biology working with DBS samples (among other sample types), but it is a challenge for small academic labs to access the resources needed to develop assay protocols and implement high-throughput analyses.

Early in my career, I set up my lab at Northwestern as a collaborative resource for the interdisciplinary health research community, and I discussed the potential opportunities afforded by DBS sampling with many colleagues in epidemiology, psychology, sociology, economics, and demography. All were interested in illuminating the social and developmental determinants of health, and generally had limited knowledge of biological mechanisms and measurement. For the most part, they were survey researchers who relied on participant reports or administrative records to measure aspects of mental and physical health. Probably because they were trained as social scientists and not clinicians, I found them to be open to thinking more broadly about health and its measurement, and they quickly realized that DBS sampling could advance their research agendas by identifying biological pathways through which contexts and experiences “get under the skin” (Weinstein et al. 2007).

Survey researchers, however, are very concerned about sampling bias and attrition, particularly in prospective cohort studies. I was often asked, with some skepticism, whether research participants would be willing to provide a DBS sample. I generally do not assume that asking people for a few drops of blood is more off-putting than asking them questions about their sex lives and drug habits, but I recognize that this position is not universally held. In this light, it is interesting to consider the case of the National Longitudinal Study of Adolescent to Adult Health (Add Health), a large, ongoing, NIH-funded population-based survey in the United States that added DBS sampling in wave 4 (Harris et al. 2019). The study began in 1994 with the recruitment of a nationally representative sample of 20,745 adolescents in grades 7–12, with in-home interviews gathering information on self-reported health and health-related behaviors, access to health care, peer and family contexts, psychosocial stressors, and socioeconomic resources. In 2007–2008, interviewers also asked participants to provide a DBS sample, and 94% agreed. In comparison, 93% of participants answered questions on income. In other words, young adults in the United States were more willing to provide a finger stick blood sample than reveal how much money they made.

It is impossible to know with certainty, but I estimate that well over 100,000 research participants have provided DBS samples over the past 20 years as part of community- and population-based studies in the United States and globally. Add Health, the Health and Retirement Study, and the National Social Life, Health, and Aging Project were early, large-scale adopters in the United States (Williams and McDade 2009; Harris et al. 2019; Kim et al. 2024), while the Study on Global Ageing and Health mounted an ambitious effort in 2007 to implement DBS sampling in nationally representative samples of older adults in China, Ghana, India, Mexico, Russia, and South Africa, comprising a total sample of more than 40,000 individuals (Kowal et al. 2012).

The COVID-19 pandemic turned out to be an important inflection point: It showcased the value of DBS sampling as an alternative to venipuncture while orienting researchers—and research participants—to the possibilities of self-sampling in the home. In the initial phase of the pandemic, before vaccines and at-home COVID tests were available, when schools were closed and many states locked down, there was an urgent need to track the community spread of SARS-CoV-2. Severe cases resulted in hospitalization, but how many people had milder or asymptomatic cases? What behaviors and policies reduced the risk of transmission? Does seroconversion provide immunity against re-infection? Antibody testing was recognized as an important tool for addressing these questions since it could be used to identify exposed individuals even in the absence of clinical symptoms (Abbasi 2020; McDade and Sancilio 2020).

But there was a problem: Clinics and hospitals were overwhelmed, personal protective equipment was in short supply, and lockdowns were keeping people at home. Venous blood collection was simply not an option for the large-scale, community-based seroprevalence studies that were needed to inform pandemic responses. In April 2020, I recognized that DBS sampling could fill a critical gap, and I gained access to my lab and connected with colleagues at our medical school who were developing a plasma assay for clinical testing. They shared a key reagent (SARS-CoV-2 receptor binding domain antigen) and I sat at the bench, alone in the lab, and developed the first DBS method for SARS-CoV-2 antibodies (McDade et al. 2020). This assay, along with at-home DBS collection kits and a web-based “no contact” platform for engaging participants, served as the bases for a fully remote seroprevalence study—among the first and largest in the United States. By December 2020, we had collected and analyzed approximately 8000 DBS samples, with initial results revealing widespread exposure to SARS-CoV-2 (the infection rate was seven times greater than estimated from direct viral testing) and high frequencies of mild and asymptomatic infections (Demonbreun et al. 2021). Other groups also deployed DBS sampling for seroprevalence studies in the US and globally (Abbasi 2020; Mulchandani et al. 2021; Ojji et al. 2023), and the lessons learned regarding the utility and flexibility of DBS as a blood collection platform are carrying forward into the post-pandemic research landscape. DBS sampling has become an established part of the population health research toolkit because it combines precise biomarker quantification in the lab with a low-cost, low-burden approach to collecting blood samples in the field.

When I talk to colleagues who are designing studies that include blood-based biomarkers, I always recommend they collect venous blood if possible—it provides abundant sample and access to widely available, gold-standard analytic protocols. But for those of us conducting research outside the clinic, there are clear tradeoffs here as the costs and logistics of venipuncture can consume project budgets, complicate operations in the field, and hinder the recruitment of hard-to-reach participants. On the other hand, capillary blood sampling can be deployed at scale in a wide range of remote settings, outside of the clinic. But it poses challenges on the back end, with a small sample volume and a complex matrix that most labs are not accustomed to handling.

Advantages and disadvantages of various approaches to blood collection need to be weighed on a case-by-case basis, but it is worth highlighting how technological advances over the past 10–15 years are shifting the balance. For example, recent developments in highly sensitive analytic platforms have expanded the range of biomarkers that can be accurately quantified in small sample volumes. Multiplex immunoassay platforms (e.g., Meso Scale Discovery, Luminex) leverage novel capture and signal amplification techniques to simultaneously measure multiple analytes in a single aliquot of sample, with exceptional specificity and lower limits of detection. Mass spectrometry has been widely used in newborn screening since the 1970s, but it is increasingly being applied to quantify toxins as well as endocrine and metabolic markers in DBS (Funk et al. 2015; Zakaria et al. 2016; Jacobson et al. 2023). Sequencing technologies increasingly provide deeper information on the genome, epigenome, and transcriptome at lower cost and with smaller sample volumes (Wong et al. 2008; McDade et al. 2016).

The trend is clear: The options for accurately quantifying small and large molecules in DBS samples have expanded substantially, with no end in sight. Recently, the flagship journal Clinical Chemistry published a scoping review entitled “State of the Science in Dried Blood Spots” (Freeman et al. 2018). The authors state: “We identified 2018 analytes measured in DBS and found every common analytic method applied to traditional liquid samples had been applied to DBS samples” (p. 1). They also note: “Technological advancements will likely continue to minimize constraints around DBS adoption” (p. 1). The article reflects a shift toward broader acceptance of DBS beyond newborn screening, and the authors highlight current and emerging applications in disease surveillance, biomarker and drug discovery, veterinary science, forensics, toxicology, personalized medicine, and epidemiology. These are useful points to include in grant applications in case it is necessary to alleviate concerns of a naïve or skeptical reviewer.

In addition to technical developments in the lab, new products are making it easier to collect capillary blood in the home. The Tasso M20 (Tasso Inc., Seattle, WA) and the OneDraw (Drawbridge Health Inc., New York, NY) are self-contained devices that attach to the upper arm or leg, release a sterile lancet with the push of a button, and then draw whole blood into a cartridge where it dries on filter paper. In my own studies, I have found that having trained research assistants implement the finger stick protocol yields the best quantity and quality of DBS samples. Results are more variable when participants are asked to prick their own fingers. The Tasso and OneDraw devices cost more and collect less blood, but they are a good option for studies where participants will self-collect in the home.

If it is easy and cheap to collect a few drops of capillary blood, and we have the technical ability to quantify most biomarkers in micro-samples of blood, why don't we? This was the question I asked myself one afternoon almost 20 years ago, after reading my wife's lab report following a visit to the phlebotomy clinic where they pulled three different tubes of blood from her arm. The sheet in front of me listed concentrations of four hormones, each of which I could have measured in my lab in a total of three drops of finger stick blood. But clinical testing is not oriented around the needs of the patient. Rather, it is beholden to capital-intensive high-throughput instrumentation, a complex regulatory landscape, and clinical accreditations that rarely question the assumption that large volumes of venous blood are the foundation of clinical testing and laboratory medicine.

Fortunately, those of us working in the research space are less constrained. We have the opportunity to think more creatively about bringing the lab into the field. We can orient around our research participants and reduce the hassles of blood collection, while also generating cutting-edge knowledge on the function of important biological systems. In short, only a lack of will or imagination can prevent us from leveraging technical developments in the lab to advance our understanding of human biology and health in community-based settings around the world.

At the end of the quarter, I ask the students in my health inequalities course the same question as the first week: “What do you think are the three most important things that determine someone's health?” I show them the new word cloud, and to my great relief it now features terms like “education,” “social,” “discrimination,” and “environment,” while “genes” and terms referencing “lifestyle choices” recede into the background. Obviously, they all matter, but we need to be mindful of what explanations we privilege, and how the methods we deploy to measure health-related processes and outcomes may constrain us from thinking beyond genetic and individual levels of causation.

Minimally invasive approaches like DBS sampling make it possible to collect a blood sample from just about anyone, anywhere in the world. Other tissues, including saliva, urine, breastmilk, and hair, also provide access to physiological information without the constraints of venous blood collection. Additionally, there is a large and expanding toolkit of methods for remotely capturing aspects of cardiovascular, metabolic, and neurological activity, much of which is greatly enabled by emerging smartphone and wearable technologies (Fischer and Kleen 2021; McDade and Harris 2022).

Who has the imagination and motivation to get outside of the clinic, to develop and validate methods that “get under the skin” in field-based research settings in the United States, and globally? Who is going to do this work? It requires training that is not typically built into PhD, MD, or MPH programs. It requires the ability to think beyond existing measurement paradigms. It takes flexibility and creativity, and a willingness to take risks.

I can think of no intellectual community better positioned to do this work than biological anthropologists trained in human biology. Methodological innovation has played a critical role in field-based research on the causes and consequences of human biological variation, but it matters much more broadly: It serves as a catalyst for knowledge production that places contextual factors on an equal footing with “genes” and “lifestyle” as determinants of human health. It generates an empirical understanding of the mechanisms through which social and physical environments shape the human body, thereby bringing them into the foreground as we seek solutions to pressing global health challenges. Building out a toolkit of minimally invasive, field-friendly methods is essential to this effort.

T.M. conceptualized the study and drafted the manuscript.

The author has nothing to report.

T.M. is a scientific advisor to Salimetrics, a company that conducts laboratory analyses for the research community.