Rational use of medicines : safety, challenges and potentials of electronic mobile devices for implementing a new algorithm for management of childhood illness (ALMANACH) in low and middle income countries

Nearly 7 million children die each year before the age of 5 despite the availability of effective low-cost interventions. Integrated Management of Childhood Illness (IMCI) guidelines have proven, when used correctly, to improve quality of care and reduce under-5 mortality. However, the actual impact of IMCI worldwide has been less than anticipated due to limited uptake of the intervention, and poor compliance to algorithms by clinicians. A recent study in Tanzania showed poor compliance to the IMCI protocols, resulting to low quality of care. Indeed 78% of the patients attending outpatient clinics with fever who tested negative for malaria were prescribed antibiotics although this was appropriate in less than 20% of the case. Such procedures lead to poor health outcomes, huge wastage of medicines and rapid spread of bacterial resistance. The rapidly changing patterns of disease, the rapid spread of resistance to antimicrobial drugs, the poor compliance to clinical guidelines using paper-based algorithms, and the increasing availability of new technologies that can improve both diagnoses accuracy and compliance to evidence-based clinical algorithms were the basis of this PhD thesis.
The objectives of the thesis were:
 To develop a new algorithm on paper and on electronic support for the management of childhood illness (based on a literature review, results of a fever study conducted in Tanzania and discussions with IMCI experts and which aims to improve clinical outcome and rational use of antimalarial and antibiotics.
 To assess whether the new ALgorithm for MANAgement of CHildhood illness (ALMANACH), which was derived from IMCI and based on the new evidence about disease patterns and drug resistance algorithms, is as safe as routine practice (IMCI) and lead to a more appropriate use of antibiotics
 To describe the distribution of acutely ill children in the various pathways of the new algorithm for the management of childhood illness in order to find out potential branches (symptoms and signs) that could be modified or dropped to facilitate its use, and hence uptake, but keeping it suitable to the epidemiology and burden of paediatric illnesses in outpatients departments of primary health care settings.
 To assess potential barriers and facilitators for the uptake of the new algorithm on electronic support in pragmatic conditions in order to recommend ways for its sustainable use and optimal compliance by clinicians in primary health care settings.
Methods
These four objectives were addressed in three different studies with the following designs and methodologies:
(i) A structured literature review in Medline, Embase and the Cochrane Database for Systematic Reviews (CDSR) searching for available evidence on (a) disease prevalence in paediatric outpatients, (b) accuracy of clinical predictors and (c) performance of point of care tests on targeted diseases (Clotilde Rambaud-Althaus et al, submitted).
(ii) A controlled non-inferiority trial to compare the clinical outcome and antibiotic prescriptions of children managed according to ALMANACH or to standard practice (IMCI). Consecutive children aged 2-59 months with acute illness were managed using ALMANACH (2 intervention facilities), or standard practice (2 control facilities) in Dar es Salaam and Morogoro in Tanzania (urban and rural settings). Children enrolled in the intervention arm were managed by two study clinicians (one for each setting) who were trained to strictly comply with the ALMANACH algorithm. Primary outcomes were proportion of children cured at day 7 and who received antibiotics on day 0.
(iii) A descriptive analysis of the data collected in the intervention arm of the study described under ii). Outcomes were i) proportions of children presenting with danger signs, with main symptoms and signs, ii) proportions of children who were prescribed antimicrobials and iii) clinical outcomes.
(iv) A qualitative study using in-depth interviews and focus group discussions among 40 primary health care workers from 6 public primary health facilities in the three municipalities of Dar es Salaam, Tanzania where the electronic clinical algorithm ALMANACH was used. Health workers’ perceptions related to factors facilitating or constraining the uptake of the electronic device were identified.
Results
i) Development of new algorithm for the Integrated Management of Childhood Illness to improve rational use of antimicrobials (ALMANACH) while maintaining health outcome (Clotilde Rambaud-Althaus et al, submitted).
The new algorithm (ALMANACH) differs from IMCI 2008 version in the following aspects:
(a) 10 general danger signs are assessed; (b) the non-severe children are classified in to febrile and non-febrile patients, the latter not being recommended to receive antibiotics; (c) pneumonia is based on a respiratory rate threshold of ≥ 50 breaths per minute that is assessed twice for children aged 12-59 months; (d) all febrile patients are tested with malaria rapid diagnostic test; when no identified source of fever at the end of assessment, then (e) urine dipstick is performed in febrile children <2 years to identify urinary tract infection (UTI), f) abdominal tenderness is performed in febrile children >2 years to ‘classify possible typhoid’; (g) classification of likely viral infection is made in cases where nothing specific is found.
ii) Clinical outcome and antibiotic prescriptions using ALMANACH
130/842 (15∙4%) in ALMANACH and 241/623 (38∙7%) in control arm were diagnosed with an infection in need for antibiotics, while 3∙8% and 9∙6% had malaria. 815/838 (97∙3%;96∙1-98.4%) were cured at day 7 using ALMANACH versus 573/623 (92∙0%;89∙8-94∙1%) using standard practice (p<0∙001). Of 23 children not cured at day 7 using ALMANACH, 44% had skin problems, 30% pneumonia, 26% upper respiratory infection and 13% likely viral infection at day 0. At day 0, antibiotics were prescribed to 15∙4% (12∙9-17∙9%) using ALMANACH versus 84∙3% (81∙4-87∙1%) using standard practice (p<0∙001) corresponding to a reduction of 82%. 2∙3% (1∙3-3.3%) and 3∙2% (1∙8-4∙6%) respectively received an antibiotic secondarily. Secondary hospitalization occurred for one child using ALMANACH; one child died at day 4 in the standard practice arm.
iii) Description of the distribution of acutely ill children in the various branches of the new algorithm for the management of childhood illness (ALMANACH)
842 consecutive patients were assessed. 0.1% (1/842) of children had general danger signs. 67.8% (571/842) entered the fever, 59.1% the cough, 21.9% the diarrhoea and 14.5% the skin problems branches. 0.3% (1/351) of patients with fever and cough had lower chest indrawing. 71.5 % (251/351) of them had a final classification of upper respiratory tract infection and 28.5% (100/351) of pneumonia. 5.6% (32/570) patients, for whom a malaria rapid diagnostic test was performed, had malaria. 7.1% (6/85) patients for whom a urine dipstick was performed had a positive result. None of the 39 patients, in whom abdominal tenderness (considered as predictive for typhoid fever) was searched for, presented this sign. On day 0, 1.9% (5/271) of non-febrile patients received antibiotics compared to 21.9% (125/571) of febrile patients. Day 7 cure rate in febrile and non-febrile patients was 97.2% and 97.4% respectively.
iv) Potential barriers and facilitators for the uptake of the new algorithm on electronic support (mobile devices) in pragmatic conditions
In general, the ALMANACH was assessed positively. The majority of the respondents felt comfortable to use the devices during consultations and stated that patient’s trust was not affected. Most health workers said that the ALMANACH simplified their work, reduced antibiotic prescription and gave correct classification and treatment for common causes of childhood illnesses.
Few reported technical challenges using the mobile devices and complained about having had difficulties typing. Majority of the respondents stated that the devices increased the consultation duration compared to routine practice. In addition, health system barriers such as lack of staff, lack of medicine and lack of financial motivation were identified as key reasons for the low uptake of the devices.
Conclusion
The new algorithm has the potential to improve clinical outcome of pediatric patients presenting at primary care facilities and drastically reduce antibiotic prescription. This was achieved through more accurate diagnoses and hence better identification of children in need of antibiotic treatment or not. A detailed count of patients with defined symptoms, signs, laboratory test results, and clinical outcome when managed by ALMANACH allowed to precisely assessing the value of each of the proposed branches of the algorithm in a primary care setting. This helped to decide on the relevance of keeping, modifying or dropping each of the algorithm components. A vast majority of children had mild illnesses and most of them did not require antibiotics to be cured. ALMANACH building on mobile technology allowed easy access, rapid update of the decision chart and has a potential to improve quality and timely use of clinical data collected at the point of care for planning at all levels of health care. Further studies are recommended to assess feasibility of phased scale up of an improved ALMANACH on mobile devices in order to improve rational use of antimicrobials and the quality of health care for children in developing countries.