Smartphones to improve health workers performance and rational drug use for management of childhood illnesses in low resource settings.

Author(s) Clotilde Rambaud-Althaus1, Amani Shao2, Ndeniria Swai3, Seneca Perri 4, Marc Mitchell 5, Judith Kahama-Maro6, Valerie D'acremont7, Blaise Genton8
Affiliation(s) 1Epidemiology and Public Health department, Swiss Tropical and Public Health institute, Basel, Switzerland, Geneva, Switzerland, 2Amani Research Center, National Institute of Medical Research, Dar es Salaam, Tanzania, 3City Medical Office of Health, Dar es Salaam City council, Dar es Salaam, Tanzania, Dar es Salaam, Tanzania, 4Nursing informatics, University of Utah, Salt Lake city, USA, Salt Lake City, United States, 5Department of Global Health and Population, Harvard School of Public Health, Boston, United States, 6City Medical Office of Health, Dar es Salaam City council, Dar es Salaam, Tanzania, Dar es Salaam, Tanzania, 7Infectious Diseases Service and Department of Ambulatory care and Community Medicine, University Hospital, Lausanne, Switzerland, Lausanne, Switzerland, 8Epidemiology and Public Health department, Swiss Tropical and Public Health department, Lausanne, Switzerland
Country - ies of focus Tanzania
Relevant to the conference tracks Innovation and Technologies
Summary In low resource settings, where childhood mortality is high, health workers (HWs)’ compliance to guidelines is essential to improve management of childhood illness. We developed paper and electronic versions of an algorithm for the management of childhood illness (ALMANACH) combining clinical elements with rapid tests for malaria and urinary infection. We assessed HWs’ compliance to guidelines when using paper versus electronic ALMANACH. The electronic clinical decision supports improved HWs performance in terms of symptoms assessment and rational use of drugs. However, it did not improve the accuracy of bacterial diseases diagnoses, which is essential to ensure quality of child care.
Background In low resource settings, where qualified health workers (HWs) are scarce and childhood mortality due to infectious diseases is high, the overuse of antimicrobial drugs is fastening the emergence of drug resistance. HWs’ compliance to evidence-based guidelines has the potential to improve the rational use of drugs and ensure quality of health care for children. Available new mobile technology, through smartphones and tablets, offers new perspectives in the delivery of evidence-based clinical guidelines to remote ambulatory health workers.
Objectives Our aim was to assess whether the use of smartphones to deliver a clinical ALgorithm for the MANAgement of CHildhood illnesses (ALMANACH) improves health workers diagnostic performance and rational use of drugs through a better compliance to the clinical chart.
Methodology ALMANACH, a decision chart combining clinical elements with a rapid test for malaria and a urine dipstick, was developed both on paper and electronic version. Nine Primary Health Care Facilities (PHCF) in Dar es Salaam, Tanzania (United Republic of), were randomized into 3 arms. Three PHCF were allocated an intervention using Paper ALMANACH, 3 other PHCF used Smartphone ALMANACH, the remaining 3 followed routine practice. The intervention, in Paper and Smartphone arms, consisted in a 2 days training on ALMANACH for all HWs attending to children below 5 years of age in outpatient departments (OPDs). The training was followed for each HW by one day of on-site supervision, and monthly supervision visits to the PHCF. Smartphones running ALMANACH algorithm were provided to the PHCF in the phone arm. The same ALMANACH algorithm printed as a paper booklet was provided to the PHCF in the Paper arm. Control PHCF did not receive any algorithm nor training or supervision and thus managed patients as usual. From September to November 2011, we conducted a cross sectional consultation process observational survey in the 9 PHCFs. Children aged 2-59 months, coming for the first visit for an acute medical ailment, were enrolled. All consultations were observed by a surveyor who collected data on a standardized form. An expert re-examined each child to establish the ‘gold standard’ assessment and treatments needed. We compared HWs’ performance between the 3 arms using indicators of child’s assessment, and appropriateness of diagnosis and treatment as outcome variables. Indicators (proportions) were compared using the Fischer’s exact test and 95% confidence interval (95%CI). The expert consultation was the reference.
Results A total of 504 children were enrolled. The mean age was 19.7 months (interquartile range 8-29) and 261 (52%) were males. The most frequent symptoms at presentation were: fever (71%), cough (66%), diarrhoea (20%) and skin problems (8%). The use of smartphones was associated with better child assessment indicators. The proportion of children checked for 3 dangers signs (unable to drink, vomits everything, and history of convulsion) was 74% (95%CI, 67-81%) when using smartphones, 41% (34-49%) when using paper ALMANACH and only 3% (0.4-6%) in the control arm. The proportion of children checked for main symptoms (fever, cough, and diarrhoea) was 99% (98-100%) when using smartphones, 75% (68-81%) when using paper ALMANACH and 77% (71-84%) in controls. The proportion of children with reported fever tested for malaria was 72% (63-80%), 92% (77-90%), and 84% (77-90%) in smartphones, paper and control arm respectively. The proportions of children with reported fever and cough who had their respiratory rate measured (for diagnose pneumonia) were 38% (27-49%), 53% (42-65%) and 3% (0-7%) in smartphones, paper and control arm respectively. With regards to diagnostic appropriateness, the proportion of expert-validated viral diseases identified by HWs (diagnostic sensitivity) was 74% (65-82%) with smartphones, 35% (26-44%) with paper and 45% (37-56%) in control arm. On the other hand, identification of expert-validated bacterial diseases (pneumonia and dysentery clinically diagnosed, and urinary tract infection using dipstick urinalysis) remained low in all 3 arms, with the proportion of validated bacterial infections identified by HWs at 41% (21-61%), 33% (20-65%), and 36% (15-58%) respectively. Nevertheless, the use of ALMANACH was associated with an important decrease in unnecessary antibiotic prescriptions. The proportion of children not needing antibiotics who were still prescribed antibiotics by HWs was: 20% (13-27%) in the smartphone arm, 25% (19-33%) in the paper arm and 66% (58-74%) in controls. Seven among 12 children in the paper arm and 14 among 26 children in the phone arm who needed antibiotics according to the expert, were not prescribed antibiotics by the HW. All these 21 cases were diagnosed as non-severe pneumonia by the expert, based on a respiratory rate above 50 breaths/min. For 10 (1 in paper, 9 in phone arm) of these patients, HWs did not measure the respiratory rate and did not consider pneumonia diagnosis. In the remaining 11 patients, HWs’ respiratory rate measures were below the threshold of 50 breaths/min. This difference could be due to inaccurate measurement by the HW or alternatively, to natural fluctuation of the respiratory rate between the two assessments.
Conclusion The use of ALMANACH compared to control arm was associated with better child’s assessment and considerable reduction of unnecessary antibiotic prescription. The use of smartphones running ALMANACH software versus the use of the paper ALMANACH was associated with better symptoms assessment and better identification of viral diseases. However, the discriminative clinical and lab tests available (i.e. respiratory rate for pneumonia, and malaria test) remained underutilized after our intervention, with no significant difference between smartphones and paper arm. This resulted in missed antibiotic prescription for about half of the children in need of it according to ALMANACH, both in the smartphone and the paper arm. The use of smartphones, a powerful tool to guide HWs during the consultation, is not enough to convince them to use the key clinical and lab tests in all children. Devices that would directly measure the respiratory rate, or point-of-care tests identifying pneumonia or severe disease, are highly desirable. A supportive working environment, with iterative supervision is necessary to ensure the rational use of antimicrobials by HWs.

, , , , , , ,

No comments yet.

Leave a Reply