Health

I looked for every health data system that shares features with my Medical Snapshot proposal, something I’ve been working on since 2007. Such programmes medically measure people over time, store what they find, and don’t usually tell the individuals what their own data says. This article summarises what I have learned by looking at 25 existing systems, covering 90 years from the Tuskegee Syphilis Study ↗ (1932) to Our Future Health ↗ (2022). I have tried to indicate my biases and assumptions but the purpose of these notes is to inform my own thoughts about the Snapshot concept. ...

This page is the working behind the Health Observer Systems comparative review of the Medical Snapshot system: feature definitions, search and source notes, and the full scoring table. The R source reproduces the plots. Comments on measurements This analysis is an exploration of how existing systems compare to my hypothetical Snapshot system. The matrix measures how close each system comes to the Snapshot’s architecture, asking the narrow question: which of the Snapshot’s features has each existing system implemented? ...

R source for the quadrant plots in the comparative review of the Medical Snapshot system. Data is embedded in the script. Save as snapshot-comparison.R and run using Rscript snapshot-comparison.R. # Medical Snapshot System: Quadrant Analysis # # Two scatter plots positioning 25 health data systems, plus the Snapshot itself, on # research mechanics vs. Snapshot-feature axes. # # Supporting exhaustive review at https://shearer.org/research/medical-observer-systems-review/ and # the overview of the Snapshot proposal at https://shearer.org/research/medical-snapshot/ # # and various works by Dan Shearer 2007-2026 # # Output: combined snapshot_quadrant.pdf and snapshot_quadrant.png # also snapshot_quadrant_p1.png and _p2.png separated out # # Dependencies: tidyverse, ggrepel, patchwork # install.packages(c("tidyverse", "ggrepel", "patchwork")) options(repos = c(CRAN = "https://cran.r-project.org")) if (!requireNamespace("ggrepel", quietly = TRUE)) install.packages("ggrepel") if (!requireNamespace("patchwork", quietly = TRUE)) install.packages("patchwork") library(tidyverse) library(ggrepel) library(patchwork) # -- 1. Data ------------------------------------------------------------------ # mech : research mechanics score A1-B7 (max 7) # core : Snapshot participant-benefit features D13,F16,C8-C10,D11 (max 6) # voluntary consent + public data release + individual clinical benefit # full : Snapshot governance features C8-F18 (max 11) # adds fiduciary duty, state exclusion, multi-jurisdictional, benefit-sharing # category : display group # label : display name (NA = unlabelled in plot) systems <- tribble( ~id, ~label, ~mech, ~core, ~full, ~category, # -- isolated / labelled points --------------------------------------------- # Evidence tiers: A = Chinese govt source, B = Citizen Lab/HRW primary doc, # C = HRW interpretive inference only. ASPI excluded entirely. # "Physicals for All": mechanics (A1-B7) A/B-confirmed; D12+F17 inversions # A-confirmed via Article 16, 2019 Human Genetic Resources Regulations; # D13 inversion C-only (single HRW interview, culturally contested). # MPS database: A2+D13 inversions B-confirmed via Citizen Lab. # Composite scores (core, full) are unchanged from pre-exclusion values # because all inversions count zero regardless of confidence tier. "snapshot", "Medical Snapshot", 7.0, 6.0, 11.0, "Target", "physicals", "Physicals for All", 7.0, 0.0, 0.0, "Coercive", "dodsr", "DoD Serum Repository", 6.0, 2.0, 2.0, "Military", "tuskegee", "Tuskegee", 2.0, 0.0, 0.0, "Coercive", "nz", "NZ Experiment", 1.0, 0.0, 0.0, "Coercive", "ndnad", "NDNAD / CODIS", 3.0, 0.0, 0.0, "Coercive", "decode", "deCODE", 4.0, 0.0, 1.0, "Coercive", "guthrie", "Guthrie Card", 3.5, 2.0, 2.0, "Other", # -- research biobank cluster (exact scores; jitter applied in geom) -------- "epic", NA, 5.5, 2.0, 2.0, "Research", "ckb", NA, 5.0, 2.0, 2.0, "Research", "ukbb", NA, 5.0, 2.0, 2.5, "Research", "taizhou", NA, 5.0, 2.0, 2.0, "Research", "cnhbm", NA, 5.0, 1.0, 1.0, "Research", "framingham", NA, 4.0, 2.0, 2.0, "Research", "whitehall", NA, 4.0, 2.0, 2.0, "Research", "nurses", NA, 4.0, 2.0, 2.0, "Research", "alspac", NA, 4.0, 2.0, 2.5, "Research", "bbjapan", NA, 4.0, 2.0, 2.0, "Research", "allofus", NA, 4.0, 2.0, 3.0, "Research", "genscot", NA, 3.5, 2.0, 2.0, "Research", "estonian", NA, 3.0, 2.0, 2.0, "Research", "finngen", NA, 3.0, 1.0, 2.0, "Research", "ofh", NA, 3.0, 1.0, 1.0, "Research", "majengo", NA, 2.5, 0.5, 0.5, "Other", # -- surveillance ----------------------------------------------------------- "cngb", NA, 4.0, 0.0, 0.0, "Surveillance", "mps", NA, 3.0, 0.0, 0.0, "Surveillance" ) %>% mutate( category = factor(category, levels = c("Target","Research","Coercive", "Military","Surveillance","Other")), labelled = !is.na(label) ) # -- 2. Aesthetics ------------------------------------------------------------ pal <- c( "Target" = "#16a34a", "Research" = "#2563eb", "Coercive" = "#dc2626", "Military" = "#7c3aed", "Surveillance" = "#d97706", "Other" = "#6b7280" ) shp <- c( "Target" = 23, # diamond (filled) "Research" = 21, # circle (filled) "Coercive" = 24, # triangle up (filled) "Military" = 22, # square (filled) "Surveillance" = 25, # triangle down (filled) "Other" = 21 # circle (filled) ) base_theme <- theme_minimal(base_size = 13) + theme( legend.position = "bottom", legend.title = element_blank(), legend.key.size = unit(0.45, "cm"), legend.text = element_text(size = 9), panel.grid.major = element_line(colour = "grey92", linewidth = 0.35), panel.grid.minor = element_blank(), axis.title = element_text(size = 10, colour = "grey40"), axis.text = element_text(size = 9, colour = "grey50"), plot.title = element_text(size = 11, face = "plain", margin = margin(b = 2)), plot.subtitle = element_text(size = 9, colour = "grey40", lineheight = 1.3, margin = margin(b = 6)), plot.margin = margin(6, 12, 6, 6) ) # -- 3. Plot function --------------------------------------------------------- make_quad <- function(dat, y_col, y_max, q_y, y_label, title, subtitle, x_max = 7.8, gap_bracket = FALSE) { y_step <- if (y_max == 6) 1 else 2 p <- ggplot(dat, aes(x = mech, y = .data[[y_col]], colour = category, fill = category, shape = category)) + # Quadrant shading --------------------------------------------------------- annotate("rect", xmin = 3.5, xmax = Inf, ymin = -Inf, ymax = q_y, fill = "#fef2f2", alpha = 0.6) + annotate("rect", xmin = 3.5, xmax = Inf, ymin = q_y, ymax = Inf, fill = "#eff6ff", alpha = 0.28) + # Research-biobank cluster box -------------------------------------------- annotate("rect", xmin = 3.25, xmax = 5.75, ymin = 1.15, ymax = 3.10, fill = "#dbeafe", colour = "#93c5fd", alpha = 0.4, linewidth = 0.45) + annotate("text", x = 4.5, y = 3.35, label = "research biobanks", size = 3, colour = "#1d4ed8", hjust = 0.5) + # Quadrant dividers ------------------------------------------------------- geom_vline(xintercept = 3.5, linetype = "dashed", colour = "grey55", linewidth = 0.45) + geom_hline(yintercept = q_y, linetype = "dashed", colour = "grey55", linewidth = 0.45) + # Points: jitter unlabelled to reveal cluster density -------------------- geom_jitter(data = filter(dat, !labelled), position = position_jitter(width = 0.13, height = 0.08, seed = 42), size = 2.2, alpha = 0.8, stroke = 0.4) + # Points: exact positions for labelled ------------------------------------ geom_point(data = filter(dat, labelled), size = 3.5, stroke = 0.65) + # Labels (ggrepel handles overlap) ---------------------------------------- geom_label_repel( data = filter(dat, labelled), aes(label = label), colour = "grey15", fill = "white", size = 3.5, label.size = 0.3, label.padding = unit(0.20, "lines"), box.padding = unit(0.6, "lines"), point.padding = unit(0.4, "lines"), min.segment.length = 0.2, show.legend = FALSE ) + # Scales ------------------------------------------------------------------ scale_colour_manual(values = pal, drop = FALSE) + scale_fill_manual(values = pal, drop = FALSE) + scale_shape_manual(values = shp, drop = FALSE) + scale_x_continuous(limits = c(-0.2, x_max), breaks = 0:7, expand = expansion(0)) + scale_y_continuous(limits = c(-0.6, y_max + 1.6), breaks = seq(0, y_max, y_step), expand = expansion(0)) + labs(x = "Research mechanics (0 - 7)", y = y_label, title = title, subtitle = subtitle, colour = NULL, fill = NULL, shape = NULL) + base_theme # Bracket showing unclaimed gap (plot 2 only) ------------------------------ if (gap_bracket) { p <- p + annotate("segment", x = 7.3, xend = 7.3, y = 3.05, yend = 10.9, colour = "grey50", linewidth = 0.55, arrow = arrow(ends = "both", type = "open", length = unit(0.12, "cm"))) + annotate("text", x = 7.42, y = 7.0, label = "8 unachieved\nfeatures", size = 3.2, colour = "grey40", hjust = 0, lineheight = 1.2) } p } # -- 4. Build plots ----------------------------------------------------------- p1 <- make_quad( dat = systems, y_col = "core", y_max = 6, q_y = 1, y_label = "Snapshot participant-benefit features (0 - 6)", title = "Plot 1 -- Snapshot participant-benefit features", subtitle = paste0( "Voluntary consent + public data release + individual clinical benefit.\n", "Research biobanks and DoD Serum Repository reach score 2 by different routes." ) ) p2 <- make_quad( dat = systems, y_col = "full", y_max = 11, q_y = 1, y_label = "Snapshot governance features (0 - 11)", title = "Plot 2 -- Snapshot governance features", subtitle = paste0( "Adds fiduciary duty, state-access exclusion, multi-jurisdictional\n", "architecture, and benefit-sharing. Research biobanks barely move." ), x_max = 8.9, gap_bracket = TRUE ) # -- 5. Combine and save ------------------------------------------------------ combined <- (p1 | p2) + plot_layout(guides = "collect") & theme(legend.position = "bottom") ggsave("snapshot_quadrant.pdf", combined, width = 18, height = 8, device = cairo_pdf) ggsave("snapshot_quadrant.png", combined, width = 18, height = 8, dpi = 300, bg = "white") ggsave("snapshot_quadrant_p1.pdf", p1, width = 9, height = 8, device = cairo_pdf) ggsave("snapshot_quadrant_p1.png", p1, width = 9, height = 8, dpi = 300, bg = "white") ggsave("snapshot_quadrant_p2.pdf", p2, width = 9, height = 8, device = cairo_pdf) ggsave("snapshot_quadrant_p2.png", p2, width = 9, height = 8, dpi = 300, bg = "white") message("Saved: combined, p1, and p2 as PDF and PNG")

The Rule-based Epidemic Modelling (RBEM) Project was funded by the UK Medical Research Council project (grant X/011658/1) to make rule-based modelling methodology more accessible to the infectious disease modelling community. The rule-based approach is instead of writing differential equations, and there can be human as well as technical advantages to this. Our pre-print (draft) paper illustrates how the approach can be applied to classical problems, in addition to our previous work (cited below) on novel problems during pandemic pressures. ...

This is about an Observer-only Medical system, which can be immensely useful to society and governments, helping understand health and reducing suffering. There are many such systems but all of them have a common principle: individuals have their health monitored and are not told the results. So what’s in it for the individual who gives up the right to their own health data? I started thinking about this while working in health systems (the Internet Archive found a version from 2010 ↗). Over the years since then I have been variously employed in privacy, data sharing/donation, safety in health data movements, causality and risk assessment and my original background of cybersecurity - all of which are combined in observer health systems. When I realised that versions of my idea have already been implemented, I made a comparison of 26 existing systems spanning 90 years, complete with a scoring system and graphical analysis. ...

The Active Heat Exchanger (AHE) research programme started as an engineering development project, and discovered first an engineering research question and then a health research question. The AHE is a retrofit ventilation device which solves the problem of there being no large-scale, long-term environmental epidemiology data for indoor air quality, despite most humans spending most their lives indoors. The device addresses the ventilation component of sick housing while generating the continuous surveillance data the field needs. ...

💡 Advised not to have a scan due to radiation risk? If you or your children are advised against having a medical scan due to radiation risks, you are probably hearing ‘radiophobia’. Health services often confuse protecting patients receiving radiation with protecting radiation workers who may be in these environments for decades. It will take a while for practices within health services to catch up to current understanding. Meanwhile, there is no need to be hesitant of any scans you are offered, despite some lurid warnings coming as standard. (The clinical term ‘radiophobia’ refers to an individual’s unfounded fear. Here I use it to mean when misinformation creates fear in patients.) When a medical person told me recently You can do without the PET scan ↗ you have had recommended, because we limit the number of scans people have so we don’t cause more problems than we find, it seemed a bit strange. I am in the fortunate position of working at a university where I have access to people who understand these things. So I made enquiries and learned that this was radiophobia, and it can cause people to miss out on healthcare. But I still wasn’t offered a scan. ...

While working with the Rule-based Epidemic Modelling group at the University of Southampton ↗ I began to consider the wider context, outside our core area of studying malaria. On the one hand, the techniques of epidemiology save lives at scale, but on the other, emerging diseases and newer health-related epidemics are accelerating. It seemed as if the field of epidemiology was really struggling, and indeed that turns out to be true. Most of the world has now adopted a new, holistic and systemic approach to healthcare, called One Health. One Health treats ecology, animals and humans as a system of systems across dozens of science fields, using the language of epidemiology. ...