Stigma Sta­tis­tics is tak­ing raw data and run­ning the num­bers to help save lives, writes Harry de Quet­teville

The Daily Telegraph - Business - - Front Page -

Seven years ago, Matthew Steans was work­ing in a high-rise build­ing when he saw a fig­ure atop the mul­ti­storey car park op­po­site his of­fice throw them­selves over the para­pet. Steans, a New Zealan­der now aged 33, was “go­ing through a pe­riod of de­pres­sion” him­self, and to­day de­scribes the in­ci­dent as “a trig­ger for me to go and get some help”. Once, af­ter emerg­ing from ther­apy, he found a se­ries of missed calls on his phone. They were from a friend who had killed them­selves.

“It was all very raw, and I guess it tipped me off on the cru­sade to do some­thing about the pain and suf­fer­ing as­so­ci­ated with sui­cide.”

Last year, sui­cide num­bers in Eng­land hit a record high. Across the UK, the lat­est fig­ures (for 2018) show that 6,859 peo­ple killed them­selves. By con­trast, just 1,770 were killed on Great Bri­tain’s roads in the same year.

The most at risk be­ing from sui­cide are men aged be­tween 45-49. In­deed al­most three quar­ters of those who com­mit sui­cide are men. “It might take a woman four years to de­cide, from low mood to want­ing to at­tempt sui­cide, whereas with a man it’s a lot less – maybe just four hours. An im­pulse and the means to do it.”

The im­por­tance of data

Data can show far more than vic­tims’ gen­der, how­ever. It can, for ex­am­ple, re­veal that 28pc of those who kill them­selves are al­ready known to men­tal health ser­vices, or that 10pc have them­selves been be­reaved by sui­cide. Be­cause, as Steans puts it, far from al­ways be­ing an af­flic­tion of lonely and des­per­ate in­di­vid­u­als in iso­la­tion, sui­cide can be catch­ing, with rashes of cases wreak­ing havoc among vul­ner­a­ble in­di­vid­u­als. And like the Covid-19 con­ta­gion that we are all fac­ing now, data are crit­i­cal to un­der­stand­ing out­breaks, to pre­dict­ing where they might emerge, iden­ti­fy­ing those most at risk and in­ter­ven­ing to save them.

The re­sult is Stigma Sta­tis­tics, an in­ter­net por­tal that col­lates in­for­ma­tion from the huge va­ri­ety of or­gan­i­sa­tions, from lo­cal au­thor­i­ties to coroners to GPs to – all too of­ten – rail­way op­er­a­tors – in­volved when a sui­cide oc­curs. The cru­cial point is that, with a sin­gle fam­ily per­mis­sion, the in­for­ma­tion can be col­lected and shared in real time. Tra­di­tion­ally, says Steans, it takes at least nine months for a death of­fi­cially to be con­firmed as sui­cide in this coun­try. That can de­lay sui­cide care ser­vices’ own ver­sion of track and trace, while anal­y­sis of real-time data can spot trends – geo­graphic, age re­lated, in a school district, say, or re­veal that sui­cides in one area oc­cur at times when a spe­cific men­tal health drop-in cen­tre is closed.

It might also al­low ef­fec­tive in­ter­ven­tions to block trans­mis­sion of the sui­cide “con­ta­gion”, by of­fer­ing timely im­me­di­ate sup­port to rel­a­tives, who them­selves be­come sta­tis­ti­cally far more likely to com­mit sui­cide af­ter a fam­ily mem­ber’s death.

“Pub­lic Health Eng­land have guid­ance that if there are three or more sui­cides in a cer­tain area in three months then there’s what’s called an ‘es­ca­la­tion plan’,” says Steans. “How­ever, prior to hav­ing a plat­form that was all done ret­ro­spec­tively. You had missed the boat.”

His ap­proach is not unique. In­deed, Steans de­vel­oped Stigma Sta­tis­tics, which is cur­rently be­ing used by two county coun­cils – Kent and Cam­bridgeshir­e – af­ter work­ing with Gary Slutkin, an Amer­i­can doc­tor who found that gun vi­o­lence in the US also fol­lowed the pat­terns of in­fec­tious dis­eases, and ar­gued that un­der­stand­ing, then cut­ting, “trans­mis­sion”, through swift data col­lec­tion was key to an ef­fec­tive re­sponse.

Three years ago Steans be­gan to try to find a sin­gle source of re­li­able, real-time data from the many agen­cies in­volved in a sui­cide, only to re­alise there wasn’t one.

Now he is not alone in the ef­fort. QES, a soft­ware com­pany based Glouces­ter­shire, has also de­vel­oped a real-time sui­cide sur­veil­lance plat­form.

‘A fac­tory closes, the foot­ball team lost and it’s a Tues­day in June. Th­ese events aren’t reg­u­lar, but you can build up a pic­ture’

Murky ethics

As with many ap­pli­ca­tions of data to sen­si­tive ar­eas of hu­man ex­is­tence, pre­vent­ing sui­cide by crunch­ing the num­bers is a murky eth­i­cal world. “We are de­vel­op­ing tech­nol­ogy around pre­dictabil­ity,” says Steans.

“A fac­tory closes, and the foot­ball team lost, and it’s a Tues­day in June. Th­ese events aren’t reg­u­lar, but you can build up a pic­ture of that.” With more and more data, how­ever, pre­dic­tions could be re­fined, un­til in­di­vid­u­als – who may not them­selves even re­alise they are at risk – are au­to­mat­i­cally flagged by the sys­tem.

Steans has no prob­lem with the con­cept. “On the ethics, I’m re­ally sup­port­ive. It’s bet­ter to ask a well worded ques­tion than not do it. And if we’ve got all the data, there’s no rea­son why we can’t ask that ques­tion to the right per­son at the right time.”

Yet if the sum­mer’s exam grading fi­asco re­veals any­thing, it is that ac­cu­racy does not al­ways make al­go­rith­mic pre­dic­tion pub­licly ac­cept­able. Nor are such meth­ods guar­an­teed to be ac­cu­rate. Pre­dic­tive polic­ing tools – widely used Amer­ica – iden­tify crime hotspots. But they also tar­get in­di­vid­u­als es­ti­mated to be likely to com­mit of­fences in fu­ture, which has been found open to abuse and led to ha­rass­ment. One such sys­tem, de­scribed by the Tampa Bay Times, de­scribed how a po­lice force gen­er­ated a list of sus­pects, many un­der 18, then sent “deputies to find and in­ter­ro­gate any­one whose name ap­pears, of­ten with­out prob­a­ble cause, a search war­rant or ev­i­dence of a spe­cific crime”.

Yet from sui­cide pre­ven­tion to the dispatch of SWAT teams, the im­pact of so-called “pre­dic­tive an­a­lyt­ics” is only likely to in­crease, par­tic­u­larly un­der the data-hun­gry lead­er­ship of the Prime Min­is­ter’s chief ad­viser Do­minic Cum­mings. Just two days ago, the

‘On the ethics, I’m re­ally sup­port­ive. It’s bet­ter to ask a well worded ques­tion than not do it’

Gov­ern­ment launched a Na­tional Data Strat­egy to “kick-start the data revo­lu­tion across the UK” with 500 data-sci­ence an­a­lysts, work­ing un­der a new gov­ern­ment chief data of­fi­cer, promis­ing to “drive ef­fi­ciency and im­prove pub­lic ser­vices”.

The hu­man touch

Fol­low­ing on from World Sui­cide Pre­ven­tion Day yes­ter­day, Steans is hope­ful the data revo­lu­tion will save lives. The ex­ist­ing cross-gov­ern­ment sui­cide pre­ven­tion Work­plan, pub­lished in Jan­uary 2019, ac­knowl­edges the need to “look at in­no­va­tive ways to im­prove lo­cal data col­lec­tion and sui­cide au­dits through projects such as real-time sur­veil­lance of sui­cides”.

“It’s a very con­ser­va­tive mar­ket,” Steans says. “Pub­lic health and lo­cal au­thor­i­ties are wary of data shar­ing. But we need to know what’s go­ing on. I want to turn the innovation dial right up.”

In that he has been helped by Covid, which has ex­posed the prime im­por­tance of ac­cu­rate, real-time data and may be soft­en­ing such of­fi­cial re­luc­tance.

The ar­gu­ments are eco­nomic as well as eth­i­cal. Sui­cides are ex­pen­sive. Last year the De­part­ment for Trans­port es­ti­mated sui­cides caused 847,000 min­utes of de­lays on the rail net­work – at a cost of £68m. In 2004, a study put the cost of sui­cide to fam­i­lies and the wider econ­omy at £1.29m per death, the equiv­a­lent of £2m to­day – £13.7bn for all 6,859 deaths in 2018. And no mat­ter how au­to­mated, how ap­par­ently in­tru­sive, sui­cide pre­ven­tion be­comes in fu­ture (and for the mo­ment Stigma Sta­tis­tics’ an­a­lyt­ics merely no­ti­fies flesh and blood ad­min­is­tra­tors to fol­low up, or not) it will, says Steans, al­ways re­quire a hu­man touch at the very sharpest end.

“Some­one still has to go to that bridge, to walk up to that per­son, and ask: ‘How are you?’. Be­cause you need some­one to ask. I had no idea that it wasn’t OK to feel like that. A lot of peo­ple don’t. Un­til it’s too late.”

If you are hav­ing sui­ci­dal thoughts or are wor­ried about some­one else, you will find help here: The Sa­mar­i­tans 24-hour helpline: 116 123. You can also contact them by email­ing jo@ sa­mar­i­

If you or a young per­son you know is strug­gling to cope, you can also contact Pa­pyrus, the young per­son sui­cide pre­ven­tion line. Call on 0800 068 4141, text 07860 039 967, or email


Newspapers in English

Newspapers from UK

© PressReader. All rights reserved.