Article, Emergency Medicine

Identifying and minimizing abuse of emergency call center services through technology

a b s t r a c t

Background: To identify and minimize unnecessary calls to emergency numbers and to assess the effectiveness of call-tracking technology in addressing the problem.

Methods: A retrospective, interventional study was conducted of all emergency calls made to Magen David Adom (MDA), Israel’s national Emergency Medicine Service (EMS) during years 2012-2016. In 2015 a tiered technolog- ical intervention was developed and implemented by MDA. The call-tracking technology self-identified harass- ment calls by call duration and frequency. The system automatically diverted harassing calls to a non- emergency number system in order not to lose any call. The rates of harassment calls were analyzed by shift, re- gion, and season. Trends were compared before and after intervention.

Results: During the years 2012-2016, 53,527 shifts took place, and 8.2% (4277) of shifts identified as receiving in- coming harassment calls. The evening shift (11.5%), the Jerusalem region (16.9%), and the summer season (9.6%) were most prone to harassment calls. After implementing an intervention using specialized call-tracking technol- ogy, the prevalence of harassment calls decreased significantly (from 10.9% to 2.9% p b .001). The Jerusalem re- gion showed the greatest decrease of 92% (from 26.5%-2.0% p b .001).

Conclusions: MDA’s call tracking technology has been shown to identify and minimize harassment calls and can be implemented by emergency organizations to reduce abuse of emergency call services.

(C) 2019

Introduction

The World Health Organization has established that emergency medical service (EMS) systems are an inseparable part of any well- functioning, effective healthcare system [1]. EMS services are usually the first point of contact for most people in times of medical crisis or in- jury [2]. Contact with EMS systems is established through calls to desig- nated emergency numbers, which are answered in an emergency dispatch center. These numbers are toll-free and are widely publicized through every form of media.

Emergency centers all over the world receive non-emergency calls on a regular basis, a problem which impedes their ability to assist civil- ians who are truly in need of emergency help [2-4]. Identifying and responding to these calls remains an unresolved worldwide problem. The diversity in the nature and intent of unnecessary calls to emergency numbers is reflected in the high rates of these calls all over the world. In the USA, England, the European Union, Pakistan, Australia and South American countries, up to 95% of calls to EMS centers are non- emergency calls [3,8-10]. Israel’s national emergency service, Magen

* Corresponding author at: 42 Yermiyahu, Kiryar Ono 6706210, Israel.

E-mail address: [email protected] (M. Siman-Tov).

David Adom (MDA), faces similar difficulties regarding unnecessary calls to the national pre-hospital emergency number 101.

Defining and identifying unnecessary calls to an emergency number is problematic. Non-malicious calls and technical glitches make up a large number of unnecessary calls to emergency centers. Unnecessary calls are often categorized as intentional or unintentional [2-5]. Inten- tional non-emergency calls are designed to harass, while unintentional non-emergency calls are misplaced (should be re-directed to appropri- ate sources), misdialed (wrong numbers) or technical miscalls (pocket calls, automatic calls, redials, etc.). Another category includes non- emergency calls made by lonely, needy, or unstable individuals, rarely for the purpose of malicious harassment [6,7]. These special cases re- quire intervention by social services or community resources. Another important area of concern relates to calls easily assumed to be harass- ment calls (heavy breathing or unclear speech), which are later revealed to be true emergencies, placed by individuals who are very young or hard to understand [12].

Mitigating the dangerous phenomena of phone abuse has proven difficult. The danger of missing a real emergency versus the danger of unnecessarily tying up lines and resources is difficult to resolve. Legisla- tion regarding harassment calls is lacking or unenforced, enabling ha- rassers to continue with little fear of retribution [11]. Harassment calls

https://doi.org/10.1016/j.ajem.2019.07.015

0735-6757/(C) 2019

O. Blushtein et al. / American Journal of Emergency Medicine 38 (2020) 916919 917

abuse essential medical and human resources necessary for saving lives and pose a danger to people in need of lifesaving assistance. The objec- tive of this study was to present the way that Israel’s national emer- gency system (MDA) identifies and minimizes harassment calls through call tracking technology.

Methods

Setting

A retrospective interventional study was conducted of all emergency calls made to Israel’s national emergency service MDA emergency num- ber (101), between the years 2012-2016. MDA is accessed by its emer- gency number and all dispatch centers are manned by trained paramedics. Based on the caller’s location, calls are diverted to regional dispatch centers. If all lines in the regional dispatch are occupied, the call automatically diverts to the nearest region or to National Headquarters Dispatch. For the purpose of this study MDA’s 11 regional dispatch cen- ters were divided into four greater regions: North, South, Central, and Jerusalem.

Technological setting and descriptive

MDA’s approach to emergency health services is built on a synthesis of medical resources and innovative technology. Since the late 1990s, MDA has incorporated a sophisticated computerized Command and Control (CC) system that monitors, tracks, archives, and analyzes all MDA activity thus enabling the organization to audit and improve its ac- tivities, thereby optimizing use of its human and technological re- sources. The entire CC platform is based on Microsoft tools and written in Windows Presentation Foundation (WPF). Using standard tools such as these enables ease of availability and application to any system worldwide.

MDA National Headquarters houses two divisions that work in full compatibility and include the medical dispatch division and the com- puter technology division. All aspects of call intake, dispatch, vehicle monitoring, medical records, virtual diagnosis and treatment, video and audio recording, and all archiving and auditing are based on ongo- ing in-house computer technology and support. On staff at Headquar- ters are computer developers and IT specialists working full time. All development is in physical and chronological proximity to the actual call intake and dispatch events enabling an in-depth first-hand under- standing of on-site application of the technological solutions to the dis- patch needs.

Intervention design

The highly developed CC system is the technological platform that services the tiered call Tracking system developed in 2015 by MDA to re- duce malicious harassment calls. The entire intervention is automatic and no human resource or assessment is required. MDA defined harass- ment as calls disconnected by the caller within a very brief time. Com- puter algorithms embedded in the incoming emergency telephone system (101), identify calls abandoned within 10 s and count the aban- doned calls emerging from the same number. Although most malicious calls come from unidentified numbers, Israeli regulations demand that cell phone providers automatically display the phone number of any call to an emergency number.

When the same abandoned number appears on MDA’s incoming calls register N15 times within 24 h, or 20 times within 48 h, it automat- ically redirects to a phone message offering a non-toll-free phone num- ber. This avoids blocking the number completely, allowing access to help in the case of a real emergency from this number. The algorithm excludes landlines in order to protect institutions such as nursing homes and hospitals who commonly place multiple legitimate emer- gency calls daily.

Big data analysts working in the technology division are constantly appraising the data regarding the harassing calls and this enables the re- search department of MDA to identify patterns and consequences. These analyses assist MDA management in refining the needs of the call and dispatch systems in alignment with technological solutions.

To quantify the rate of harassment calls, the number of harassment calls per daily shift was analyzed. Shifts were defined by periods: morn- ing, evening, and night. The prevalence of harassment calls per shift was analyzed dichotomously using a value of none or at least one per shift.

Data analysis

Data for this study were retrieved from the MDA Control and Com- mand system. The number of abandoned calls per shift was analyzed and central and spread tendencies were presented. Due to abnormal distribution the analysis was based on dichotomous values of none or one and more. Differences in the prevalence of harassment calls were analyzed by shift time, region, season, and year presenting the odds for harassment calls and 95% confidence interval (CI). The change in the prevalence of harassment calls in total and by regional divisions was analyzed comparing two periods; before the intervention during years 2012-2014 and after the intervention during years 2015-2016 using chi square tests. Statistical analysis was conducted using SPSS ver- sion 25. A p-value b0.05 was considered statistically significant.

Results

During the years 2012-2016 in all MDA emergency call centers the number of harassment calls per shift recognized by the technological call system ranged from zero to 420 with a mean of 3.36 and standard deviation of 14.66 (median of zero).

Due to abnormal distribution with kurtosis of 86.6 and skewness of 7.5, the distribution of harassment calls was recoded into two catego- ries: No harassment calls during shift and at least one harassment call per shift. During the years 2012-2016, 53,527 shifts took place, and 8.2% (4277) of shifts identified as receiving incoming harassment calls. Among the shifts defined as having at least one incoming harassment call (4277), the actual Number of calls received per shift ranged from 20 to 420, with a mean of 42.0 and standard deviation of 32.6 (median of 29).

Table 1 presents the frequency and odds for harassment calls per shift during varying shift times, regions, seasons, and year. Evening

Table 1

Harassment calls in different shift times and regions.

Harassment calls during shift

Total

Yes

No

OR 95% CI

p-value

Total

52,537

8.0%

92.05%

Region

Central

19,794

6.5%

93.5%

1

North

17,423

6.7%

93.3%

1.04 (0.95-1.12)

0.399

South

10,897

8.4%

91.6%

1.31 (1.20-1.43)

b0.001

Jerusalem

5413

16.9%

83.1%

2.92 (2.67-3.20)

b0.001

Year

2012

11,849

12.3%

87.7%

1

2013

11,640

12.7%

87.3%

1.03 (0.96-1.12)

0.395

2014

10,598

7.4%

92.6%

0.57 (0.52-0.62)

b0.001

2015

10,584

3.0%

97.0%

0.22 (0.19-0.25)

b0.001

2016

8856

2.8%

97.2%

0.21 (0.18-0.24)

b0.001

Season

Summer

13,572

9.6%

90.4%

1

Fall

12,728

7.3%

92.7%

0.74 (0.68-0.81)

b0.001

Winter

13,405

7.3%

92.7%

0.74 (0.68-0.81)

b0.001

Spring

13,822

7.7%

92.3%

0.78 (0.72-0.85)

b0.001

Shift

Morning

18,397

6.9%

93.1%

1

Evening

17,383

11.5%

88.5%

1.76 (1.64-1.90)

b0.001

Night

17,747

5.5%

94.5%

0.79 (0.73-0.86)

b0.001

918 O. Blushtein et al. / American Journal of Emergency Medicine 38 (2020) 916919

shifts were more likely to receive harassment calls vs. morning or night shifts (11.5%, 6.9% and 5.5% respectively p b .001). Summer was the sea- son most prone to harassment calls (9.6% vs. 7.5% approx. in other sea- sons). The Jerusalem region had the highest rate of harassment calls per shift (16.9%), almost three times higher than the Central region (OR 2.92 95% CI 2.67-3.20). During 2012-2013 the prevalence of harassment calls per shift was 12%, decreasing to b3% in 2016. The odds for harass- ment calls during 2016 are 79% less likely compared to 2012 (OR 0.21

95%CI 0.18-0.24).

Further analysis was based on an aggregate of the years before (years 2012-2014) and after (2015-2016) the intervention. Fig. 1 pre- sents the flux in the prevalence of harassment calls per shifts, according to region, before and after the intervention. In total, we found a decrease of 73% in the prevalence of harassment calls before and after the inter- vention (from 10.9% to 2.9% p b .001). A significant decrease of 92% was observed after the intervention especially in the Jerusalem region (from 26.5%-2.0% p b .001).

Discussion

A country’s ability to provide effective emergency care is the primary sign of a robust and functioning EMS system. It is critical to discover and reduce the source of any element that undermines this goal in order to prevent damage to public emergency health systems. Unnecessary calls to EMS centers consume valuable time and resources required for pa- tients in genuine need of treatment and transport. The impact on EMS services as a result harassing emergency EMS numbers is a worldwide concern [2-4].

When emergency numbers are unnecessarily tied up, patients re- quiring life-saving assistance and transport are denied full access to pre-hospital medical resources, potentially resulting in greater mortal- ity and morbidity [2-4]. Prior to intervention, high rates of unnecessary calls were observed in MDA, Israel’s national EMS organization, call cen- ters. Our data analysis showed that the location with the disproportion- ately highest rates of harassment calls (prior to intervention) was the Jerusalem region. While our research did not extend to analyzing causes of prevalence, we assumed that the complex status of any major me- tropolis, and the unique characteristics of Jerusalem as the capitol city of Israel in particular, expose the emergency system to greater chal- lenges. The data regarding the prevalence of harassing calls during the evening shift and during the summer season raised the possibility that free times of day and vacation times invite more abuse.

Effective means of reducing the prevalence of this abuse have not been developed on a large scale due both to the difficulty of defining a harassing call and to the inherent obligation of emergency services to

provide pre-hospital medical assistance to all. A particular obstacle to obstructing offenders results from the common practice of making abu- sive calls from unidentified numbers. Research has shown that many harassment calls are untraceable, with callers using readily available household technology to block their phone numbers. Israel has sidestepped this issue to a great degree by regulating calls to emergency numbers in collaboration with cellphone providers. Legal regulation al- lows MDA to apply technology, automatically exposing the phone num- ber of every caller to its emergency number.

In Israel the potential danger resulting from abusing pre-hospital services has lead MDA to seek solutions offering observable results in minimizing harassment calls. MDA is highly committed to providing full and equal access to all patients requiring assistance, and the organi- zation has adopted as its motto the Talmudic dictum “One who saves a single life, is as one who saves an entire world.” Minimizing unnecessary calls to MDA’s emergency number supports the goal of allowing ill and injured patients maximum access to EMS resources. In Israel, EMS phone abuse has particular repercussions, as all MDA emergency lines are manned by trained paramedics able to offer full medical phone in- struction, including births, CPR, critical injuries etc. Prior research has demonstrated that immediate resuscitation efforts significantly affect outcomes, [13] and easily available paramedic phone instruction sup- ports positive outcomes.

While legislation and enforcement have been used to fight abuse of emergency calls [3] in Israel, MDA has developed a call tracker technol- ogy that has proven to be extremely effective. The technology that en- ables identifying harassing calls and numbers was implemented in 2015. The system is based on an automatic protocol that identifies calls that fit a predetermined set of characteristics based on duration and frequency and identifies the cellphone numbers that fit the criteria. Calls that fit this category are subsequently blocked from the regular emergency dispatch number system and redirected to an automated toll number, which in a real emergency will still enable service.

The use of call duration as an objective warning parameter, rather than the traditional human assessment of call content, reduces the chance of overlooking a call involving a real medical emergency. The ad- ditional use of high frequency as a further defining parameter confirming abuse automatically differentiates non-malicious unneces- sary calls from abusive calls intended to inflict damage and harm lives. It must be stressed that the present study in no way bypasses Israel’s 2016 law forbidding false calls to emergency numbers. Technology pro- vides an immediate response to an immediate threat to the system without waiting for lengthy bureaucratic procedures. This study shows that a smart algorithm to block a harasser is more effective

than are complicated legal procedures.

Fig. 1. Trends by regions1 in harassment call rates before and after the intervention.2 1Based on chi square test. 2Before intervention years 2012-2014, after intervention years 2015-2016.

O. Blushtein et al. / American Journal of Emergency Medicine 38 (2020) 916919 919

In contrast to our results, a retrospective study in Punjab revealed that although technological solutions such as software-based blacklisting and auto-blocking for habitual callers were implemented, they were not sufficient. The authors suggested that a mass awareness campaign would help educate the public regarding the danger and sen- sitivity of the issue [14]. An American in depth study of fraudulence in crime report concludes that the law is the best way to combat abuse of emergency numbers, however the editors concur that enforcement and judicial power are lacking. [15] Further studies of abusive emer- gency number use, including the “swatting” phenomenon, claim that legislation cannot be properly enforced due to emergency services lack of technology capable of correctly identifying the caller’s phone number and location. [16,17]

On principle, in MDA’s technology landlines are not identified as po- tential harassers nor re-directed based on frequency. This is to protect health institutions that regularly make multiple legitimate calls to MDA for care and transport. However, call register data shows that in Israel the majority of harassment calls do come from cellphones.

A possible limitation to this technology is that is based on frequency and duration parameters for defining harassment and not on call con- tent. Countries experiencing primarily content abuse need further solu- tions. In MDA data showed that abandoned calls made with malicious intent was a primary form of harassment.

Conclusions

We found that the most effective way to combat the dangerous phe- nomenon of EMS phone abuse is by the use of technology, rather than by the traditional means of legislation. Call tracking technology, devel- oped and implemented by MDA for over four years, has minimized, and almost eliminated, phone abuse without increasing the need for human resources.

Healthcare professionals and stakeholders everywhere are con- cerned about the misuse of emergency numbers which strains service provision and may jeopardize patient care. The technological solution presented here can be implemented by rescue organizations to reduce abusive phone use without necessitating special training, which further stresses limited resources, or relying on human assessment in situations when time is vital.

Author contributions statement

OB, MST, and RM conceived and designed the study. OB supervised the data collection and managed the data, including quality control. MST provided statistical advice on study design and analyzed the data. OB and MST drafted the manuscript. All authors contributed substan- tially to its revision and responsibility for the paper as a whole.

Disclosure of interest

The authors report no conflict of interest.

Ethics approval

Not applicable.

Clinical trial registration

Not applicable.

Funding statement

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Data sharing

The data belong to the MDA organization. Data will be made avail- able from the corresponding author upon reasonable written request.

Declaration of Competing Interest

None declared.

Acknowledgments

The authors would like to acknowledge Mrs. Fay Schreiber and Mrs.

Timna Podolsky for their valuable assistance in editing the article.

References

  1. Mock C, Lormand JD, Goosen J, Joshipura M, Peden M. Guidelines for essential trauma care. Geneva: World Health Organization; 2004.
  2. Sampson R. COPS, U.S. Department of Justice, Office of Community Oriented Policing Services. Problem-oriented guides for police series guide no. 19, misuse and abuse of 911. https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=196149; 2004.
  3. Rashford S, Isoardi K. Optimizing the appropriate use of the emergency call system, and dealing with hoax callers. Emerg Med Australas 2010;22(5):366-7. https://doi. org/10.1111/j.1742-6723.2010.01325.x.
  4. EENA operations document. False emergency calls; 2011.
  5. EENA operations document. Emergency silent, hang-up and abandoned 112 calls; 2011.
  6. Kirkby HM, Roberts LM. Inappropriate 999 calls: an online pilot survey. Emerg Med J

    2012;29:141-6. https://doi.org/10.1136/emj.2010.092346.

    Rosenbaum N. Identification and strategies for diffusing 911 abuse caused by psychi- atric symptoms. J Police Crisis Negotiations 2010;10:181-90. https://doi.org/10. 1080/15332581003757354.

  7. Brumfield E. Chapter 284: Deterring and paying for prank 911 calls that generate a SWAT team response. 45 McGeorge L. Rev, vol. 585. ; 2014. p. 571-80 Available at: http://digitalcommons.mcgeorge.edu/mlr/vol45/iss3/17.
  8. Luchtemberg MN, de Pires DE, Schoeller SD, Possamai FP. Analysis of calls to the Mo- bile first-aid medical Services in a Brazilian capital city. Revista da Rede de Enfermagem do Nordeste. Universidade Federal do Ceara; 2015 2015 Feb 25. 6.
  9. Waseem H, Durrani M, Naseer R. Emerg Med Australas 2010;22(5):480.
  10. Jaffe EM. Swatting: the new cyberbullying frontier after Elonis v. United States. Drake L Rev 2016;64:455.
  11. Hazinski MF, Markenson D, Neish S, Gerardi M, Hootman J, Nichol G, et al. Response to cardiac arrest and selected life-threatening Medical emergencies: the medical emergency response plan for schools. A statement for healthcare providers, policymakers, school administrators, and community leaders. Ann Emerg Med 2004;43(1):83-99 https://www.ncbi.nlm.nih.gov/pubmed/14707021.
  12. Dami F, Golay C, Pasquier M, Fuchs V, Carron PN, Hugli O. prehospital triage accuracy in a criteria based dispatch centre. BMC Emerg Med 2015;15:32. https://doi.org/10. 1186/s12873-015-0058-x.
  13. Tahir MN, Akbar AH, Kayani A, Al Ramadhani S, Naseer R. 605 nonemergency calls at an emergency setting: mass awareness needed. Inj Prev 2016;22:A217. https://doi. org/10.1136/injuryprev-2016-042156.605.
  14. Turvey BE, Savino JO, Baeza JJ (Eds.). False allegations: Investigative and forensic issues in fraudulent reports of crime. Elsevier. 2017. ISBN 0128013605, 9780128013601.
  15. Enzweiler MJ. Swatting political discourse: a domestic terrorism threat. Notre Dame Law Rev 2015;90:2001.
  16. Garrett R. Sighting in swatting: agencies place prank-calling perpetrators in the crosshairs. Law Enforce Technol 2009;36(5):26 28-32. https://www.ncjrs.gov/App/ Publications/abstract.aspx?ID=249062.

Leave a Reply

Your email address will not be published. Required fields are marked *