Address for correspondence Debopam Samanta, MD, Child Neurology Section, Department of Pediatrics, University of Arkansas for Medical Sciences, 1 Children’s Way, Little Rock, AR 72202, United States (ude.smau@atnamasd).
The publisher's final edited version of this article is available at NeuropediatricsOver the last several decades, significant progress has been made in the discovery of appropriate therapy in the management of infantile spasms (IS). Based on several well-controlled studies, the American Academy of Neurology and the Child Neurology Society have published the current best practice parameters for the treatment of IS. However, dissemination and implementation of evidence-based guidelines remain a significant challenge. Though the number of well-performed controlled trials and systematic reviews is increasing exponentially, the proportion of valuable new information subsequently embedding into the routine clinical care is significantly lower. Planned and systematic implementation of evidence-based interventions in a given health care structure may outstrip the benefits of discovering a new insight, procedure, or drug in another controlled setting. Implementation problems can be broad-ranging to hinder effective, efficient, safe, timely, and patient-centered care without significant variation. The first part of this review article provides a detailed summary of some crucial comparative treatment studies of IS available in the literature. In the second part, practical challenges to mitigate the gap between knowledge and practice to improve outcomes in the management of IS has been explored, and a consolidated framework approach for systematic implementation research methodology has been discussed to implement evidence-based guidelines for the management of IS. Although large multicenter controlled studies will help gather quality evidence in the treatment of IS, a more comprehensive range of scientific methodologies, including qualitative research and mixed research methodologies, will hold the more considerable promise for implementing evidence-based practices in the health care system.
Keywords: implementation research, infantile spasms, epileptic spasms, guideline, ACTH, steroid, vigabatrin, qualitative research
Infantile spasms (IS), an age-specific severe epileptic encephalopathy, is characterized by epileptic spasms, neurodevelopmental disabilities, and the frequent presence of hypsarrhythmia in the electroencephalogram (EEG). A Cochrane systematic review, including 16 small randomized controlled trials (RCTs) (100 patients enrolled), evaluated a total of 916 patients with IS in 2013. 1 The authors concluded that hormonal treatments (prednisolone and tetracosactide) had been associated with a higher and faster response rate compared with vigabatrin. Moreover, it was also deduced that hormonal treatments were probably associated with better long-term developmental outcomes compared with vigabatrin in the subset of IS patients without a known cause. Additionally, the American Academy of Neurology (AAN)/Child Neurology Society (CNS) practice parameter (original version in 2004 and updated version in 2012) agreed on the effectiveness of adrenocorticotropic hormone (ACTH) for a short-term treatment of IS (level B). 2,3 The 2012 updated guideline recommended that there was not enough evidence to judge the equivalence of other forms of corticosteroids with ACTH. Despite national guidelines, considerable variation exists in the management of IS, including agents used, dosage regimen, and treatment duration. Although progress has been made regarding the use of standard therapy (hormone or vigabatrin), significant challenges remain in the implementation of hormonal therapy use over vigabatrin and improvement in treatment delays. Mytinger et al recently reported the implementation of an IS management guideline in a hospital (part of the national IS consortium). 4 Over 6 years of implementation, 114 of 115 consecutive patients received appropriate first-line therapy (hormone or vigabatrin); however, in cases of failed treatment, less than one-quarter patients were started on second- and third-line therapy within 2 weeks (a goal per the hospital guideline) of treatment failure. A detailed explanation of the success of the one goal and failure of the other had not been discussed in the paper. However, future goals have been set: increasing hormone therapy versus vigabatrin use as the first medicine (52 patients received vigabatrin as a first-line therapy compared with 32 patients with ACTH and 30 patients with prednisolone), use of hormonal therapy or vigabatrin in patients with failure to first-line treatment (nonstandard treatment was used in 18 patients after failure of first-line therapy and in 19 patients after second therapy), and reducing treatment delays after failure of the first treatment.
As Balas and Boren famously quoted in a seminal paper: “It takes 17 years to turn 14 percent of original research to the benefit of patient care.” 5 There is a significant lack of research regarding the formulation of a comprehensive framework to address challenges related to a health care delivery intervention. The first part of the article is a detailed summary of many relevant comparative treatment studies of IS available in the literature, where hormonal treatment (either ACTH or oral prednisolone) was compared with vigabatrin, whether in a blinded and randomized studies or in a naturalistic observational study. In the second part, the author discusses how to establish institutional evidence-based guidelines for IS treatment and provides a framework for future systematic implementation of a quality intervention for the treatment of IS.
There are several studies that evaluated the short-term effectiveness of ACTH and other forms of corticosteroids. Sorel and Dusaucy-Bauloye first reported the effectiveness of ACTH in the treatment of IS. 6,7 ACTH is available in natural and synthetic forms. The only available form of long-acting ACTH in the United States is the natural form, which is extracted from a bovine or porcine source and contains 39 amino acids. On the other hand, the only available form of ACTH in Europe is synthetic tetracosactide, containing 24 amino acids of the N-terminal fragment and has a much longer duration of action (24–72 hours) compared with the natural ACTH (18–24 hours), necessitating alternate day regimen. Head-to-head comparison between these two agents has not been performed to differentiate efficacy and safety differences between these products. In addition, 80 to 100 IU of natural ACTH has been estimated to be equivalent to 1 mg tetracosactide and more than 200 times expensive than the synthetic form.
For unknown reasons, a wide variability of ACTH dose has been used, particularly dependent on the country, with the lowest dose in Japan (10 U/day), highest in the United States (150 U/m 2 /day), and in the middle range in the United Kingdom and Finland (50 U every other day). Hrachovy et al compared high-dose, long-duration (150 U/m 2 , tapering over 12 weeks) therapy in 26 patients versus low-dose, short-duration (20–30 U/d, 2–6 weeks) ACTH regimens in 24 patients. 8 Approximately 50% of patients responded in the former group compared with 58% in the latter group without any statistically significant difference. Additionally, there was no difference in relapse between these two groups. However, the high-dose group was associated with a higher incidence of hypertension. One significant advantage of short-duration steroid therapy is the sparing of the hypothalamic–pituitary–adrenal axis suppression. However, suppressive effects between ACTH and other forms of steroids have not been systematically studied. AAN 2012 practice parameter suggested that low-dose ACTH can be considered as an alternative to high-dose ACTH based on one class I and one class II studies both by Hrachovy et al. 8,9 However, high-dose ACTH continues to be the most preferred regimen in the United States. Snead et al showed remarkable effectiveness of ACTH in two high-dose, long-duration regimen with 93 and 100% response rate with the cessation of IS and normalization of the EEG. 10,11
In 1964, Jeavons and Bower after reviewing 112 cases opined that oral corticosteroids could be similarly effective compared with ACTH. 12 Hrachovy et al performed a double-blind, placebo-controlled study to compare ACTH (20–30 U/day) with prednisolone (2 mg/kg/day) in 24 patients and found no statistically significant difference between these two agents. 9 Objective evaluation by video EEG confirmed the disappearance of both IS and hypsarrhythmia in 42% of patients in the ACTH group and 33% in the prednisolone group. During crossover in nonresponder patients, 50% responded to ACTH therapy and 43% to prednisolone therapy. It is commonly accepted that among oral corticosteroids, prednisolone is preferable than prednisone as prednisone needs conversion to the active form, prednisolone, in the body, and in early infancy, this conversion may not be optimum due to deficient hydroxysteroid dehydrogenase activity. Baram et al conducted a prospective, randomized, single-blind class II study to compare high-dose natural ACTH (150 U/m 2 ) with prednisolone over 2 weeks in 29 infants. 13 A significantly higher proportion of infants responded to ACTH (87%) compared with prednisolone (29%).
The drug choice for IS has significant global variability, and before 2007, the vast majority of neurologists in the United States used natural ACTH as the initial therapy for IS. Unfortunately, the price of ACTH increased from US$40 a vial in 2000 to US$40,000 in 2020 (including a dramatic increase of 14-fold only in 2007). This raises the cost of a typical treatment course with ACTH to approximately US $150,000. As the synthetic forms are not available in the United States, neurologists in the United States who used natural ACTH previously started to consider an alternative cheaper but efficacious option such as oral corticosteroids. Early oral corticosteroid studies (two RCTs using prednisone, and other studies were class III prospective, open-label studies using prednisolone and prednisone) utilized 2 to 3 mg/kg/day of corticosteroids. However, UKISS (United Kingdom Infantile Spasms Study), a large class III RCT compared a high-dose oral prednisolone protocol (40–60 mg/day) with intramuscular (IM) ACTH. Subsequently, other studies used a high dose of prednisolone. For example, Kossoff et al reported that 10 (67%) of 15 infants became spasm-free within 2 weeks of high-dose oral prednisolone (40–60 mg/day). 14 Idiopathic patients did better than symptomatic cases (88 vs. 43%; p = 0.10). The authors also compared this outcome with a historical cohort of 15 infants receiving ACTH, among which 13 (87%) responded to therapy with no significant difference between these two groups (p = 0.16). The group treated with oral prednisolone had fewer adverse effects (53 vs. 80%; p = 0.10). There has also been a rare study that showed not only comparability but also actual superiority of oral prednisolone over ACTH injection. The Sri Lanka Infantile Spasms Study (SLISS) compared low-dose synthetic ACTH (40 to 60 IU/every other day) with oral prednisolone (40–60 mg/day) in 97 infants for immediate spasm control and resolution of hypsarrhythmia at 14 days in the single-blind, parallel-group, randomized clinical trial. 15,16 The findings showed a significantly higher rate of resolution of hypsarrhythmia as well as short-term control of spasms with oral prednisolone (58.3 vs. 36.7% in the ACTH group; p = 0.03). Furthermore, follow-up of these patients showed better spasm control at 3 months in the prednisolone group (64.6%) versus 38.8% in the ACTH (p = 0.01), with no significant difference at 6 and 12 months between the two groups, though there was a trend favoring prednisolone. 17
Chang et al published a systematic review/meta-analysis using five RCTs with a total of 239 individuals and detected no difference between corticosteroids and ACTH in the cessation of clinical spasms, hypsarrhythmia resolution, side effects, relapse rate, or rate of development of epilepsy subsequently. 18
In the 1990s, vigabatrin, a GABA (gamma-aminobutyric acid) transaminase inhibitor, was introduced in Europe, primarily for the treatment of IS associated with tuberous sclerosis complex (TSC). Elterman et al reported the treatment effect of low-dose (18–36 mg/kg/day) and high-dose (100–148 mg/kg/day) vigabatrin in 142 hormonal treatment-naive IS patients. 19 A significantly higher response rate (p > 0.001) was seen in the high-dose group with 2 weeks of treatment. The estimated proportion of responders increased gradually from 8% at 2 weeks, 42% at 1 month, 55% at 2 months, to 65% at 3 months. Time to response was also shorter in the high-dose group, and the TSC patients (out of 25 patients, 23 responded favorably within 3 months of treatment) responded better than other groups. Approximately 16% of patients had a relapse within the first 3 months. In general, only one-quarter to one-third of patients respond to vigabatrin within 2 weeks of therapy with a gradual increase in the responder rate to reach approximately two-thirds after 3 months of therapy.
Hormonal therapy has been noted to be more effective than vigabatrin for the short-term treatment of IS. The UKISS study compared vigabatrin (100 mg/kg/day) and hormonal therapy (40 mg/day of prednisolone or IM tetracosactide depot 0.5 mg [40 IU] on alternate days) in a large class III, multicenter, RCT. 20 Approximately 73% in the hormonal group remained spasm-free at 2 weeks compared with only 54% in the vigabatrin group (p = 0.043). However, clinical reporting was used for documentation of spasms rather than the current gold standard of using video EEG. Hypsarrhythmia was also resolved in a significantly higher number in the hormonal treatment group than vigabatrin (p = 0.024). In the hormonal group, there was no difference between the tetracosactide and the prednisolone group (the study was not powered to show a difference between these two groups). Importantly, this study also evaluated intermediate- to long-term outcomes related to neurodevelopment and future incidence of epilepsy. This cohort was followed till the age of 14 months to compare these two primary therapies again for development and epilepsy outcomes. 21 At that time, there was no difference in spasm control between these groups, as well as no difference in neurodevelopment as assessed by the Vineland adaptive behavioral scale (VABS). However, a particular subgroup of IS with no identified etiology had higher VABS scores (88.2 vs. 78.9; p = 0.025) if treated with hormonal therapy rather than with vigabatrin. These infants (77 of the original 107) further followed up till the age of 4 years, and again, no difference in epilepsy or developmental outcomes was noted between vigabatrin and hormonal treatment groups, with the exception of the subgroup with idiopathic etiology treated with hormonal agents. 22 VABS scores were higher in the hormonal treatment group (mean: 96) compared with the vigabatrin group (63; p = 0.033) in this subgroup, with no identified etiology. Interestingly, latency between spasm onset and initiation of treatment was associated with decreased VABS scores after controlling for the treatment effect and etiology.
Other than prednisolone, other corticosteroids have been rarely reported as an alternative to ACTH therapy. Haberlandt et al compared dexamethasone pulse therapy (20 mg/m 2 /day × 3 days, with an interval of 4 weeks between each cycle) with ACTH in a retrospective level IV study involving 18 patients. 23 In the dexamethasone group, seven patients received five cycles, with four patients becoming spasm-free. The resolution of spasms was better with ACTH (9 out of 11 patients) compared with dexamethasone. A class IV prospective study used 3 days of high-dose methylprednisolone followed by 2 months of a low dose of oral corticosteroid taper in 10 patients and detected a 50% response rate in 2 weeks. 24
The International Collaborative Infantile Spasms Study (ICISS) trial, the largest clinical trial of IS, compared hormonal therapy with a combination of vigabatrin and hormonal therapy in a multicenter, open-label, randomized trial. 25 Combination therapy of hormonal treatment with vigabatrin has been utilized with a hope of a synergistic effect, as well as based on a hypothesis that there may be two distinct populations of IS: one would respond to GABAergic manipulation and other by reducing epileptogenic corticotrophin-releasing hormone concentration. Parents were given options to choose a particular type of hormonal therapy: tetracosactide injection versus prednisolone. Both tetracosactide and prednisolone treatment options were available as some parents might prefer oral therapy versus an IM injection, and many physicians still have a preference to use ACTH over prednisolone, though not based on definitive evidence. Although parents had the option of choosing particular hormonal therapy, physicians were not given a choice. The primary outcome was freedom from spasms for 4 weeks between 2 to 6 weeks after the initiation of treatment. This criterion was more stringent than the UKISS trial, where spasm freedom rate only at 2 weeks of treatment was calculated. Among 377 infants, 72% had a positive outcome with the combination therapy compared with 57% (p = 0.002) in the only hormonal therapy group. When EEG data were included, electroclinical data similarly showed better effectiveness of the combination regimen. This study excluded infants with TSC, previously treated patients with IS, or patients with a previous use of hormonal therapy or vigabatrin. Moreover, to mitigate heterogeneity, infants with a recent clinical diagnosis of IS (diagnosis within 7 days) were only included in the study. Hormonal therapy was given for 29 days, and vigabatrin was for 3 months, followed by 4 weeks taper. The infants with spasms > 2 months prior to treatment had a significantly diminished response rate. Although in infants with a high risk of developmental impairment at randomization (chromosomal abnormality, established diagnosis of developmental delay, cerebral palsy, previous history of neonatal encephalopathy with seizures) had a significant outcome difference between two groups, in the infants with low risk for developmental impairment at randomization, the difference is highly substantial favoring combination therapy (p < 0.001). The treatment response rate was faster with combination therapy (median: 2 days) than hormonal therapy alone (4 days; p < 0.001). Approximately 20% of infants relapsed between 2 and 6 weeks of treatment. The most severe adverse events were related to infection, but no death was reported. Chance of electroclinical remission was less with prednisolone (151/263) than tetracosactide (76/111; p = 0.004); however, these treatments were not randomized and should be interpreted with caution.
The Pediatric Epilepsy Research Consortium (PERC) and the National Infantile Spasms Consortium brought together many large pediatric epilepsy programs to form a robust network and infrastructure to assess the evaluation, diagnosis, and treatment of IS. Several important prospective observational studies had been reported reflecting the current status of IS management in the United States. Knupp et al evaluated 232 infants with a new diagnosis of IS from 22 centers for prospective assessment of treatment response, that is, sustained clinical resolution and resolution of hypsarrhythmia in 3 months. 26 Overall, 46% of children responded to either hormonal or vigabatrin treatment (individual response rates: ACTH 55%, prednisolone 39%, and vigabatrin 36%). ACTH response rate was significantly higher than the vigabatrin group (p = 0.038) and marginally higher than the corticosteroid group (p = 0.06). In addition, 58% responded to high-dose ACTH group compared with 38% in the low/intermediate group (p = 0.14, statistically not significant and not powered to detect a difference between these groups). Patients treated with prednisolone had a higher relapse rate (24%) compared with other groups. A higher response rate was seen with ACTH in children with unknown etiology. Children with mild or no developmental issues were prescribed ACTH more often, suggesting possible selection bias, but the adjusted model for treatment also showed that etiology and developmental delay did not significantly change the response rate. Due to the observational nature of the study, the dosing regimen, the interval between medication changes, and developmental measures were not standardized. Prescribing bias had been tried to be controlled by multivariate logistic regression but limited by a potential underpowered sample size. Knupp et al evaluated in a prospective, observational, multicenter study of 118 infants that spasms responded better if the second therapy had a different mechanism of action (hormonal therapy to vigabatrin or vice versa; 55 vs. 25%; p < 0.001). 27 The overall response rate to the second therapy was 37%. Moreover, IS responded better to second therapy if the first treatment was initiated within 4 weeks, but the timing of initiation of second-line therapy did not affect the outcome.
During implementing a standard guideline, the biggest challenge is choosing the preferred first-line agent. Most rigorous guidelines give strength of evidence of each recommendation. The adoption of the guideline recommendations that are based on robust evidence is relatively easier than adopting less well-supported recommendations. For example, the 2012 AAN and CNS guideline endorsed low-dose ACTH as an alternative to high-dose ACTH with a level B recommendation. Level B evidence may not be compelling enough, and therefore other choices are still considered within reason by the practitioners and further creates complexity in the resistance to applying guidelines in the clinical practice. Having the option of choosing one agent among ACTH, prednisolone, and vigabatrin may create significant variation in the practice. If the goal of the health system to use hormonal therapy over vigabatrin, the guideline should state that explicitly, and indication for vigabatrin use (e.g., patients with TSC or immune-suppression) as the primary agent should also be documented. On the other hand, if a particular institute is trying to implement a guideline to increase the use of approved therapy (vigabatrin or hormones) rather than nonstandard treatments, all three agents with dosing and duration recommendations should be in the guideline. Individual intervention should be systemically analyzed for facilitators and barriers in the local context that help or hinder clinicians in adopting that agent in the clinical care pathway as a preferable agent in practice (► Table 1 ). For example, availability of vigabatrin or ACTH starter pack in one institute may function as a significant facilitator for the use of these agents, and, on the other hand, if one particular state’s Medicaid does not approve ACTH in drug-naive IS patients, that will act as a significant barrier for adoption of ACTH as first-line therapy in that practice. Similarly, team leaders’ opinions and existing practice of using one agent may significantly promote or hinder the implementation of scientific evidence in clinical practice. The family choice remains an essential factor for the implementation of any guideline, particularly family preference based on side effects, such as possible visual field restriction with vigabatrin or risk of death from sepsis with ACTH. Choosing the preferable first-line agent is the subject of most studies, and how to implement reduction of treatment delays has not been systemically studied. Moreover, the latter issue is a more complex problem to deal with, and controlled trials might not provide a straight forward answer to solve this problem. Rather than quantitative research, a comprehensive implementation research methodology (discussed in the later section) needs to be employed to find a better solution in the local context.
Examples of Intervention assessment focusing on facilitators and barriers
Perceived as the most effective agent Highest reported short-term benefit A short course of therapy is only necessaryThe extremely high cost of a typical course of natural ACTH in the United States that may exceed US$100,000
Synthetic ACTH not available in the United States Caregivers need to provide injection High rates of insurance denial Dosage confusion: if a low dose is enough or high dose is necessarySignificant adverse effect profile: immunosuppression, hypertension, congestive heart failure hyperglycemia, hypokalemia, HPA axis suppression
Close follow-up necessary during acute stages: antibiotic prophylaxis, screening for hyperglycemia, gastrointestinal bleeding, serum potassium level, and monitoring for adrenal or pituitary insufficiency
Oral formulation Several studies suggested (but not definitely proven) equal efficacy to ACTH Much cheaper than ACTH (a typical course of prednisolone costs less than US$100) The U.S. national guideline equivocal about its efficacy compared with ACTHHigh-dose ACTH has not been compared with high-dose prednisolone in an adequately powered randomized controlled trial
Similar side effects and monitoring needed as ACTH Oral formulation Better tolerance Monitoring of different parameters are not needed in the acute stagesWith respect to long-term outcomes, the superiority of hormonal therapies over vigabatrin has not been proven other than cryptogenic subtype
Response substantially higher among patients with tuberous sclerosis Speculation that visual field loss may be less common in children Perceived safety as there is a lack of standardized visual field testing in infants,Lack of knowledge of how visual field loss affects patient functioning or quality of life after infantile exposure.
Reversibility of MRI findings and possibly asymptomatic in nature provides comfort Short-term response rates are considerably lower Enrollment in the REMS program necessary, A severe side effect of permanent bilateral concentric peripheral visual field defectsMRI changes high T2 signal and restricted diffusion in the thalami, basal ganglia, brainstem tegmentum, and cerebellar dentate nuclei
Abbreviations: ACTH, adrenocorticotropic hormone; HPA, hypothalamic–pituitary–adrenal; MRI, magnetic resonance imaging; REMS, Risk Evaluation and Mitigation Strategy.
Hormonal therapy has been suggested to be better in long-term developmental outcomes compared with vigabatrin in IS with unknown etiology. However, the definition of unknown etiology is an ever-changing: In a recent study, whole-exome sequencing unraveled 15% of additional de novo mutations in IS patients with unknown etiologies. 28 Moreover, prospective diagnosis is not possible as some work-up, particularly genetic testing, may take months to complete. In this context, to exclude unknown etiologies, the following parameters can be used: established developmental delays before the development of IS, neonatal seizures, known genetic diagnosis, the presence of dysmorphic features, a history of hypoxic–ischemic encephalopathy, and documented history of large cortical malformation. The absence of all of these may indicate an idiopathic IS (this is not the same as IS with “unknown etiology”) and favors treatment by hormonal therapy over vigabatrin.
The attitude related to the use of standard therapy (hormonal therapies or vigabatrin) is determined by the desired outcome of spasm freedom in the short term and better neurodevelopmental outcome in the long run. Successful adherence to treatments may be further influenced by perceived difficulties and feelings of worth associated with the use of standard therapies. Role of the team leader or opinion leader (IS clinic director, chief of child neurology, chief of epilepsy, or a well-respected senior neurologist) is particularly valuable in the initial stages of implementation. The team leaders can use role-modeling skills and facilitate the adoption of new changes over the old, outdated practices. It is the responsibility of the team leader to generate confidence among team members that change is urgent, essential, feasible, and achievable. If the leader or most clinicians are following the standard guideline, the guideline throughout the institute efficiently disseminates among all providers. However, a concrete plan should be made for different stages of IS, such as new IS, treatment-refractory IS, relapse after successful treatment, to achieve better outcomes in the implementation process.
As described previously, the PERC and the National Infantile Spasms Consortium reported several relevant prospective observational studies reflecting the current status of IS management in the United States. However, the dissemination of the critical work needs to move outside the network in all centers in the United States and globally. Smaller clinics and practices may not have access to the electronic medical record (EMR) data and the ability to perform research. The lack of standardization across EMR systems prevents the comparison of data across different health systems and inhibits research. Although diagnosis and procedure codes of IS are the same for all health care systems, the measurement of outcome research involving many U.S. or international centers remains difficult. Moreover, access to non-EMR databases such as pharmacy and EEG databases is necessary to get the complete picture. Documentation of clinical information, such as initiation of spasms, developmental history, previous history of seizures, and cessation of spasms, is highly variable among providers and, in most situations, only partially documented. Drug adverse effects are not commonly entered in the EMR. Nursing communication, bedside notes, and phone communication after discharge from the inpatient stay are not frequently extracted from the EMR. In addition, clinicians enter data in the EMR, mostly in a free-text format that they find convenient and useful for patient care. However, this free-text format is not easily extractable for research purposes. The variability of the data can be reduced by capturing only essential data with appropriate coding. During the analysis of EMR-based data, appropriate statistical measures to mitigate selection bias should be employed. In that regard, propensity scores, instrumental variable methods, and selection models for adjustments of selection bias are more suitable than commonly used techniques such as logistic regression, matching, and stratification.
As proposed by the complexity theory, a broader system-based implementation (e.g., a comprehensive improvement plan for the epilepsy program) may be more effective rather than the implementation of a single guideline pertinent to IS. On the other hand, successful implementation of IS guidelines may have a cascading effect to impact overall epilepsy care in the institute or the health system. The successful implementation may involve care providers receiving an incentive for following guidelines properly. Financial decisions may sustain the implementation process by allocating money to recruit IS educators, ACTH injection teachers, and IS clinic coordinators. Existing organizational, national, and international guidelines provide rich resources for implementation. Providers need to be interviewed individually and in group- settings to obtain a robust set of qualitative data. The interviews can be digitally recorded, transcribed, coded, and clustered for analysis in terms of individual, organization, and system-level contexts. Qualitative data analysis and coding can help to compare interviews, checking for constancy and contrasting perspectives of different stakeholders. A complete picture of the implementation barrier can be drawn after the triangulation of qualitative data sources with existing baseline data available in the organization. 29–31
Implementation science, methods and strategies to improve evidence-based practices in routine care, is relatively unexplored in the realm of neurology (essential terminology has been reviewed in ► Table 2 ). Mindful consideration of a comprehensive framework significantly increases the success rate of implementation of any guidelines, including IS (► Table 3 ). 32
Terminology related to implementation science
| Implementation science | The study of methods to promote the adoption and integration of evidence-based practices, interventions, and policies into routine health care and public health settings |
| Four distinct phases of implementation | Exploration, preparation, implementation, and sustainment |
| Five major domains of Consolidated Framework for Advancing Implementation Science | Intervention characteristics, outer setting, inner setting, characteristics of the individuals involved, and the process of implementation |
| Intervention characteristics | Features of an intervention that might influence implementation Constructs included in intervention characteristics: if the intervention locally developed versus adopted from outside, quality and validity of the evidence supporting the intervention, stakeholders’ perceptions about the relative advantage of implementing the intervention, ease in adaptation, ability to start in small scale, perceived difficulty in implementation, how the intervention is presented, and cost of the intervention |
| Inner setting | Features of the implementing organization that might influence implementation Constructs included in the inner setting: characteristics of the organization, networks and communication within the organization, culture of the organization, implementation climate (this is more tangible than culture and assessed by stakeholders desire for change in the sphere of intervention), leadership engagement, and available resources |
| Outer setting | Features of the external context or environment (system-level factors) that might influence implementation Constructs included in the outer setting: external policy and incentives, need of the patients, networking with other external organizations, and competitive pressure from competing organizations |
| Characteristics of individuals | Individuals involved in implementation that might influence implementation Constructs related to characteristics of individuals: knowledge and beliefs about the intervention, belief in own capabilities to execute implementation process, individual identification with the organization, and personal attributes such as motivation, intelligence, innovativeness, and learning styles |
| Implementation process | Strategies to influence implementation Constructs related to the implementation process: planning, engaging appropriate individuals in the implementation, executing the intervention, reflecting, evaluating, and intermittently revisiting the process, and expanding, refining, and reevaluating the process |
An implementation framework for quality initiatives in the management of infantile spasms
| Characteristic of the initiative |
| Reducing treatment lag time to treatment |
| Usage of hormone or vigabatrin vs. nonstandard therapy |
| Use of ACTH rather than vigabatrin |
| Use of low-dose ACTH vs. high-dose ACTH |
| Reducing treatment lag with alternative standard therapy after failure of the first therapy |
| Standardized developmental testing |
| External environment and context |
| Availability of national or international guideline |
| Cost of the medicine |
| Part of a national/international consortium |
| Insurance coverage of the proposed medicine |
| Availability of starter packs (samples) of the medicines in the hospital |
| Difficulty in getting therapy approved by insurance providers |
| Faster referrals for video EEGs in patients with suspected spasm |
| Increasing awareness of IS among pediatricians |
| High out-of-pocket cost for families |
| High cost to health care |
| Patient assistance program |
| Support from the drug manufacturer |
| Ease of access(available only at a specialty pharmacy or most pharmacies) |
| Internal context and setting of the practice |
| Organizational commitment |
| Prior experience with quality improvement efforts |
| Robust electronic health record technology integrated with IS registries |
| Trained staff |
| Amount of support available for parents/caregivers |
| Parents/caregivers not comfortable with the mode of administration |
| Characteristics and attitudes of practice staff and clinicians |
| Availability of opinion leader |
| Clinician skepticism regarding the value of guideline-based care |
| Need for inpatient admission for initiation of therapy |
| Tolerance of hassles related to efficacy and adverse effects of the drug and getting insurance approval |
| Implementation process within the practice |
| Qualitative interviews to identify facilitators and barriers |
| Use of established quality improvement process |
| Pilot testing |
| Implementation meeting |
| Standardized/uniform workflow |
| Knowledge sharing among stakeholders |
Abbreviations: ACTH, adrenocorticotropic hormone; EEG, electroencephalogram; IS, infantile spasms.
Complexity can be a barrier to successful implementation, for example, having more options without clear guidance to choose one over another. Physicians all over the world continue to struggle to choose between vigabatrin and hormonal therapy as the primary treatment option. As described previously, the ICISS trial compared hormonal treatment with a combination of vigabatrin and hormonal therapy in a multicenter, open-label, randomized trial. The use of a combination of hormonal and vigabatrin treatment at the time of the diagnosis can decrease the complexity associated with choosing one or other first-line medicine. However, prior to the widespread implementation of combination therapy, several limitations of this study should be considered. Parents and clinicians were not masked to treatment allocation due to the use of IM injection of tetracosactide. However, EEG reviewers were blinded to the particular treatment. The combination therapy of hormonal treatment and vigabatrin could be compared with hormonal treatment and placebo, but placebo was not used due to prohibitive cost issues in this noncommercial trial. Another limitation of the study was that 377 infants out of 766 screened were included in the study, and reasons for exclusion were not adequately documented. This study does not answer the potential utility of sequential therapy versus combination therapy from the outset, as well as the most appropriate third-line therapy. Although acute electroclinical remission is the standard assessment variable for IS management, long-term neurodevelopmental outcome is the more meaningful criterion. As a follow-up study, 362 of the original 377 infants were evaluated for developmental status at 18 months. 33 Mean VABS scores were not different between the combination and only hormonal treatment groups. Approximately 30% of infants had epilepsy in both groups, with around 15% showed persistence of IS. Prompt spasm control (within 6 weeks from the treatment initiation) was associated with higher VABS scores (16-point difference in mean scores) and lower likelihood (17 vs. 53%) of seizures at 18 months. Delayed treatment was noted to be associated with lower VABS scores and poor epilepsy outcomes. A large portion of the only hormonal therapy patients received additional vigabatrin therapy later (74% infants received vigabatrin at the end of the third month if not responding to hormonal therapy) and might receive the benefit of combination therapies, diluting any difference in developmental outcome between two groups. Another hypothesis, though unlikely, is that developmental gain achieved by the faster and higher rate of acute remission from combination therapy might be counteracted by vigabatrin-induced poor neurodevelopment. Additionally, regular video EEG was not performed after the initial treatment period, with possible misses of subtle spasms. A subtle variant of spasms has been reported within 2 weeks of initiation of vigabatrin. 34 At 18-month follow-up, neither of these hormonal therapies was associated with better developmental or epilepsy outcome. Contrary to the UKISS trial, early resolution of spasms in infants with unknown causes was not associated with better development at 18 months.
Although better acute electroclinical remission prompted many institutes to adopt the combination therapy, many other institutes decided against using the combination therapy due to similar neurodevelopmental outcomes between two groups at 18 months. 4 However, 18 months may be too early to assess the differences in the developmental impairments between two groups, and developmental testing at 3.5 years of age has been planned by the researchers. Moreover, children with a high risk of developmental impairment had a longer lead time to treatment due to various reasons such as difficulty in recognition of IS, less urgency in the diagnosis, and the absence of obvious developmental regression. Regarding the testing tool, VABS is not a global or comprehensive assessment of behavior and only a measurement of adaptive behavior. More detailed standardized tests such as Bayley Scales of Infant Development may detect subtle differences missed by a less comprehensive method.
There is also a strong interest in sequential therapy with the strict time-bound guideline to switch to an alternative form of approved therapy rapidly. Eliyan et al estimated utility of a standardized protocol where ACTH had reserved for high-dose prednisolone nonresponders. 35 In this cohort of 102 infants with IS, 60 (59%) patients responded to prednisolone (8 mg/kg/day, 60 mg divided into three equal doses). Approximately 33% of nonresponders responded to ACTH, suggesting substantial efficacy after prednisolone failure. This might represent a higher efficacy of ACTH or superior efficacy of longer duration of combined hormonal therapy. Failure in the cessation of spasms and/or hypsarrhythmia or relapse of spasms after initial response may trigger an alternative form of therapy. Response at that time usually indicates the effectiveness of the alternative form of treatment, but the delayed response of the original treatment (particularly with vigabatrin) or effectiveness from the combination regimen cannot be excluded. Treatment-naive patients responded at a higher rate to prednisolone therapy. However, the rate of sustained response progressively decreases over time to reach 40% at 18 months. There was a trend toward a higher relapse rate in infants who responded to ACTH after prednisolone failure (may suggest a refractory subpopulation) rather than the infants who responded to prednisolone. The cumulative response rate to prednisolone followed by ACTH when needed in treatment-naive patients was 84% (total over 28 days) and almost equivalent to the highest reported response rate from the primary ACTH therapy over 24 days.
Increasing cost and side effects can be barriers in the implementation of the combined therapy from the outset, as well as the concern that combination therapy may not be cost-effective. However, the combination therapy is suitably justified if it improves long-term developmental outcomes, prevents the development of epilepsy, enhances the quality of life, and facilitates potential cost savings for the health care system over the lifetime of the patient.
Lack of resources (time, money) and physician’s perception that IS outcome depends purely on etiology rather than treatment can be a potential barrier to the implementation. On the other hand, concerned parents of the patient, team leader, and other change champions may consider the implementation of evidence-based standard IS guidelines fundamental and act as facilitators. Societal and peer pressure can act as a positive or negative factor for successful implementation. The health system’s increasing interest in productivity formula based on work relative value units or pay-for-performance system may potentially act as a barrier as any nonfunded research/implementation activities might be considered less valuable than direct clinical activities, but this is only relevant for certain countries such as the United States. Neurologists with change-averse perspectives as well as non-EMR savvy providers can be a barrier to implement a standard guideline. Implementation of computer-based intervention and translation of clinical informatics into routine clinical care may need significant preparation and ground-work to assess the feasibility of that particular intervention. 36–39 Physician perspectives (e.g., optimum treatment is already given without any standard guideline, implementation of the standard guidelines will generate duplication of effort, severely affected IS patients do not need standard therapies) can be a substantial barrier and need to be addressed during qualitative research and exploration phase of the implementation process.
The relevance of more extensive environmental influence over the successful implementation of a particular guideline is immense. Outside influences on implementation of the guideline relevant to IS may include national and international policies, guidelines, research findings, evidence, regulation, legislation, mandates, directives, recommendations, political stability, public reporting, benchmarking, and organizational networks. For example, price control and ready availability of ACTH injection may have a considerable influence on the adoption of the guideline. More significant endorsement of the pay-for-performance system may also accelerate the acceptance of evidence-based guidelines. Institutes with a membership of a particular consortium such as PERC may embrace evidence-backed guidelines promptly rather than other centers with less outside control.
Significant work needs to be done on the inner context of the organization to accelerate the implementation of the evidence-based guideline. Structure and culture of the organization are critical in the diffusion process of guidelines throughout the organization. Consolidated Framework for Implementation Research has provided five constructs (structural characteristics, network and communications, culture, implementation climate, and readiness for implementation) and several subconstructs to assess the inner setting of the organization. 40 Although it is still unclear how these constructs and subconstructs influence implementation, further development and standardized testing are ongoing to detect the essential constructs. Simple organizational context such as availability of ACTH/vigabatrin starter pack in the hospital may make one particular implementation process much more straightforward (starting ACTH or vigabatrin) or difficult (staring prednisoloneas a primary therapy). The use of concept mapping to understand the importance and feasibility of a particular strategy in the local context can be extremely valuable. 41
More studies are needed to investigate the team leaders’ and stakeholders’ roles in the implementation of the standard guideline. The team leader’s role is to the ability to support and motivate all team players and stakeholders such as nurses, coordinators, and EEG technicians. 42 As the implementation process becomes more complex, more time should be deployed to understand the characteristics of the stakeholders to modify the guideline in a way that would be not only implementable acutely but also sustainable in the long run.
A clear guideline is important, as well as the ability to adapt and modify the guideline for effective implementation. Locally produced and owned protocols have a higher chance of adoption compared with national guidelines and practice parameters. 43–45 Rather than forcing an intervention (e.g., start using prednisolone for all patients as primary therapy), collaborative conversations early in the implementation process can increase the desire of the providers to participate. To bring everyone on the same page, the team leader should engage not only neurologists and epileptologists but also residents, fellows, nurses (both bedside and specialty), administrators, and managers early in the implementation process. 46 Each subgroup in the section has its discipline, agenda, and culture, and frequent conversations are necessary to sustain the altered workflow associated with the change. Providers rarely adopt a national guideline in their practice without any alteration. They modify (known as “tinkering”) the guideline using practice-based knowledge to suit their particular situation. The ultimate adoption rate of a guideline may depend on characteristics of adopters (innovators, early adopters, early majority, late majority, or laggards) in the group. Rather than directly implementing the national guideline, a meeting among stakeholders to discuss existing practices and agreement on priorities to align explicit evidence with the local needs can be an important first step. High-quality decision support systems and computerized prompts can be far more effective in improving implementation than didactic educational approach.
Typical Plan-Do-Study-Act cycles and audit and feedback approaches should be employed during the implementation process. 47,48 To maintain provider motivation and commitment, various other methods can be vital: the establishment of a clear goal, provision of regular feedback, and the institution of small-scale change. 49 Emphasis on the exploration and preparation phase of implementation is strongly recommended for the successful subsequent processes associated with the desired change. Moreover, special attention should be given to the sustenance of the implemented changes. A combination of lean thinking and Six Sigma can potentially tackle complex problems better. Lean thinking, invented by Toyota, is a value-added activity with prioritization of principles, practices, tools, and techniques to reduce waste, synchronize workflow, and diminish variability in the production. This provides standard solutions by excluding redundant steps and bottlenecks. For complex problems, Six Sigma methodologies can be more useful. With a thoughtful analysis of the value stream map (flowchart with detailed information about the work), the Six Sigma method can help with the quality intervention in the context of organizational infrastructure, deployment plan, analytics, and control method. 50
Although further studies are needed to determine the most cost-effective therapy of IS that provides short-term benefit and long-term favorable neurodevelopmental outcome, implementation and dissemination of existing knowledge are of paramount importance. Systematic implementation research methodology should be used to engage stakeholders of all levels to generate actionable items and deliver quality intervention. Although large multicenter RCTs will help gather quality evidence in the treatment of IS, a wider range of scientific methodologies, including qualitative research and a combination of different methods, will hold a more considerable promise for implementing evidence-based practices in the health care system.
As an implementation research scholar, the author thanks Dr. Geoffrey M Curran, the Director of the Center for Implementation Research (University of Arkansas for Medical Sciences) for his mentorship, and Cindy L. Mosley for her immense help with the management of implementation research program.
Dr. Samanta reports grants from the Translational Research Institute and the National Center for Advancing Translational Sciences of the National Institutes of Health (grant UL1 TR003107) during the conduct of the study.
Conflict of Interest
The authors declare no potential conflicts of interest with respect to the research, authorship, and publication of this article.
The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
1. Hancock EC, Osborne JP, Edwards SW. Treatment of infantile spasms . Cochrane Database Syst Rev 2013;( 06 ):CD001770. [PMC free article] [PubMed] [Google Scholar]
2. Mackay MT, Weiss SK, Adams-Webber T American Academy of Neurology Child Neurology Society, et al.; Practice parameter: medical treatment of infantile spasms: report of the American Academy of Neurology and the Child Neurology Society . Neurology 2004; 62 ( 10 ):1668–1681 [PMC free article] [PubMed] [Google Scholar]
3. Go CY, Mackay MT, Weiss SK Child Neurology Society American Academy of Neurology, et al.; Evidence-based guideline update: medical treatment of infantile spasms. report of the Guideline Development Subcommittee of the American Academy of Neurology and the Practice Committee of the Child Neurology Society . Neurology 2012; 78 ( 24 ):1974–1980 [PMC free article] [PubMed] [Google Scholar]
4. Mytinger JR, Albert DV, Twanow JD, et al. Compliance with standard therapies and remission rates after implementation of an infantile spasms management guideline . Pediatr Neurol 2020; 104 :23–29 [PMC free article] [PubMed] [Google Scholar]
5. Balas EA, Boren SA. Managing clinical knowledge for health care improvement . Yearb Med Inform 2000; 9 ( 01 ):65–70 [PubMed] [Google Scholar]
6. Jeavons PM, Harper JR, Bower BD. Long-term prognosis in infantile spasms: a follow-up report on 112 cases . Dev Med Child Neurol 1970; 12 ( 04 ):413–421 [PubMed] [Google Scholar]
7. Jeavons PM, Bower BD. The natural history of infantile spasms . Arch Dis Child 1961; 36 :17–22 [PMC free article] [PubMed] [Google Scholar]
8. Hrachovy RA, Frost JD Jr, Glaze DG. High-dose, long-duration versus low-dose, short-duration corticotropin therapy for infantile spasms . J Pediatr 1994; 124 ( 5 Pt 1 ):803–806 [PubMed] [Google Scholar]
9. Hrachovy RA, Frost JD Jr, Kellaway P, Zion TE. Double-blind study of ACTH vs prednisone therapy in infantile spasms . J Pediatr 1983; 103 ( 04 ):641–645 [PubMed] [Google Scholar]
10. Snead OC III, Benton JW Jr, Hosey LC, et al. Treatment of infantile spasms with high-dose ACTH: efficacy and plasma levels of ACTH and cortisol . Neurology 1989; 39 ( 08 ):1027–1031 [PubMed] [Google Scholar]
11. Snead OC III. Treatment of infantile spasms . Pediatr Neurol 1990; 6 ( 03 ):147–150 [PubMed] [Google Scholar]
12. Jeavons P, Bower B. Infantile spasms. A review of the literature and a study of 112 cases. Clinics in Developmental Medicine No. 15. The Spastics Society Medical Education and Information Unit in Association with William Heinemann Medical Books Limited Mac Keith Press [Google Scholar]
13. Baram TZ, Mitchell WG, Tournay A, Snead OC, Hanson RA, Horton EJ. High-dose corticotropin (ACTH) versus prednisone for infantile spasms: a prospective, randomized, blinded study . Pediatrics 1996; 97 ( 03 ):375–379 [PMC free article] [PubMed] [Google Scholar]
14. Kossoff EH, Hartman AL, Rubenstein JE, Vining EP. High-dose oral prednisolone for infantile spasms: an effective and less expensive alternative to ACTH . Epilepsy Behav 2009; 14 ( 04 ):674–676 [PubMed] [Google Scholar]
15. Wanigasinghe J, Arambepola C, Sri Ranganathan S, Sumanasena S, Attanapola G. Randomized, single-blind, parallel clinical trial on efficacy of oral prednisolone versus intramuscular corticotropin on immediate and continued spasm control in west syndrome . Pediatr Neurol 2015; 53 ( 03 ):193–199 [PubMed] [Google Scholar]
16. Wanigasinghe J, Arambepola C, Sri Ranganathan S, Sumanasena S, Muhandiram EC. The efficacy of moderate-to-high dose oral prednisolone versus low-to-moderate dose intramuscular corticotropin for improvement of hypsarrhythmia in West syndrome: a randomized, single-blind, parallel clinical trial . Pediatr Neurol 2014; 51 ( 01 ):24–30 [PubMed] [Google Scholar]
17. Wanigasinghe J, Arambepola C, Ranganathan SS, Sumanasena S. Randomized, single-blind, parallel clinical trial on efficacy of oral prednisolone versus intramuscular corticotropin: a 12-month assessment of spasm control in west syndrome . Pediatr Neurol 2017; 76 :14–19 [PubMed] [Google Scholar]
18. Chang YH, Chen C, Chen SH, Shen YC, Kuo YT. Effectiveness of corticosteroids versus adrenocorticotropic hormone for infantile spasms: a systematic review and meta-analysis . Ann Clin Transl Neurol 2019; 6 ( 11 ):2270–2281 [PMC free article] [PubMed] [Google Scholar]
19. Elterman RD, Shields WD, Mansfield KA, Nakagawa JUS Infantile Spasms Vigabatrin Study Group. Randomized trial of vigabatrin in patients with infantile spasms . Neurology 2001; 57 ( 08 ):1416–1421 [PubMed] [Google Scholar]
20. Lux AL, Edwards SW, Hancock E, et al. The United Kingdom Infantile Spasms Study comparing vigabatrin with prednisolone or tetracosactide at 14 days: a multicentre, randomised controlled trial . Lancet 2004; 364 ( 9447 ):1773–1778 [PubMed] [Google Scholar]
21. Lux AL, Edwards SW, Hancock E United Kingdom Infantile Spasms Study, et al.; The United Kingdom Infantile Spasms Study (UKISS) comparing hormone treatment with vigabatrin on developmental and epilepsy outcomes to age 14 months: a multicentre randomised trial . Lancet Neurol 2005; 4 ( 11 ):712–717 [PubMed] [Google Scholar]
22. Darke K, Edwards SW, Hancock E trial steering committee on behalf of participating investigators, et al.; Developmental and epilepsy outcomes at age 4 years in the UKISS trial comparing hormonal treatments to vigabatrin for infantile spasms: a multicentre randomised trial . Arch Dis Child 2010; 95 ( 05 ):382–386 [PubMed] [Google Scholar]
23. Haberlandt E, Weger C, Sigl SB, et al. Adrenocorticotropic hormone versus pulsatile dexamethasone in the treatment of infantile epilepsy syndromes . Pediatr Neurol 2010; 42 ( 01 ):21–27 [PubMed] [Google Scholar]
24. Mytinger JR, Quigg M, Taft WC, Buck ML, Rust RS. Outcomes in treatment of infantile spasms with pulse methylprednisolone . J Child Neurol 2010; 25 ( 08 ):948–953 [PubMed] [Google Scholar]
25. O’Callaghan FJ, Edwards SW, Alber FD participating investigators, et al.; Safety and effectiveness of hormonal treatment versus hormonal treatment with vigabatrin for infantile spasms (ICISS): a randomised, multicentre, open-label trial . Lancet Neurol 2017; 16 ( 01 ):33–42 [PubMed] [Google Scholar]
26. Knupp KG, Coryell J, Nickels KC Pediatric Epilepsy Research Consortium, et al.; Response to treatment in a prospective national infantile spasms cohort . Ann Neurol 2016; 79 ( 03 ): 475–484 [PMC free article] [PubMed] [Google Scholar]
27. Knupp KG, Leister E, Coryell J Pediatric Epilepsy Research Consortium, et al.; Response to second treatment after initial failed treatment in a multicenter prospective infantile spasms cohort . Epilepsia 2016; 57 ( 11 ):1834–1842 [PMC free article] [PubMed] [Google Scholar]
28. Yuskaitis CJ, Ruzhnikov MRZ, Howell KB, et al. Infantile spasms of unknown cause: Predictors of outcome and genotype-phenotype correlation . Pediatr Neurol 2018; 87 :48–56 [PMC free article] [PubMed] [Google Scholar]
29. Palinkas LA, Aarons GA, Horwitz S, Chamberlain P, Hurlburt M, Landsverk J. Mixed method designs in implementation research . Adm Policy Ment Health 2011; 38 ( 01 ):44–53 [PMC free article] [PubMed] [Google Scholar]
30. Aarons GA, Fettes DL, Sommerfeld DH, Palinkas LA. Mixed methods for implementation research: application to evidence-based practice implementation and staff turnover in community-based organizations providing child welfare services . Child Maltreat 2012; 17 ( 01 ):67–79 [PMC free article] [PubMed] [Google Scholar]
31. Aarons GA, Green AE, Willging CE, et al. Mixed-method study of a conceptual model of evidence-based intervention sustainment across multiple public-sector service settings . Implement Sci 2014; 9 ( 01 ):183. [PMC free article] [PubMed] [Google Scholar]
32. Hulscher ME, Laurant MG, Grol RP. Process evaluation on quality improvement interventions . Qual Saf Health Care 2003; 12 ( 01 ): 40–46 [PMC free article] [PubMed] [Google Scholar]
33. O’Callaghan FJK, Edwards SW, Alber FD International Collaborative Infantile Spasms Study (ICISS) investigators, et al.; Vigabatrin with hormonal treatment versus hormonal treatment alone (ICISS) for infantile spasms: 18-month outcomes of an open-label, randomised controlled trial . Lancet Child Adolesc Health 2018; 2 ( 10 ):715–725 [PubMed] [Google Scholar]
34. Gaily E, Liukkonen E, Paetau R, Rekola R, Granström ML. Infantile spasms: diagnosis and assessment of treatment response by video-EEG . Dev Med Child Neurol 2001; 43 ( 10 ):658–667 [PubMed] [Google Scholar]
35. Eliyan Y, Heesch J, Alayari A, Rajaraman RR, Sankar R, Hussain SA. Very-high-dose prednisolone before ACTH for treatment of infantile spasms: evaluation of a standardized protocol . Pediatr Neurol 2019; 99 :16–22 [PubMed] [Google Scholar]
36. Fiks AG, Grundmeier RW, Biggs LM, Localio AR, Alessandrini EA. Impact of clinical alerts within an electronic health record on routine childhood immunization in an urban pediatric population . Pediatrics 2007; 120 ( 04 ):707–714 [PubMed] [Google Scholar]
37. Richardson JE, Abramson EL, Pfoh ER, Kaushal RHITEC Investigators. Bridging informatics and implementation science: evaluating a framework to assess electronic health record implementations in community settings . AMIA Annu Symp Proc 2012; 2012 :770–778 [PMC free article] [PubMed] [Google Scholar]
38. Herzberg S, Rahbar K, Stegger L, Schäfers M, Dugas M. Concept and implementation of a computer-based reminder system to increase completeness in clinical documentation . Int J Med Inform 2011; 80 ( 05 ):351–358 [PubMed] [Google Scholar]
39. Bakken S, Ruland CM. Translating clinical informatics interventions into routine clinical care: how can the RE-AIM framework help? J Am Med Inform Assoc 2009; 16 ( 06 ):889–897 [PMC free article] [PubMed] [Google Scholar]
40. Damschroder LJ, Aron DC, Keith RE, Kirsh SR, Alexander JA, Lowery JC. Fostering implementation of health services research findings into practice: a consolidated framework for advancing implementation science . Implement Sci 2009; 4 ( 01 ):50. [PMC free article] [PubMed] [Google Scholar]
41. Waltz TJ, Powell BJ, Matthieu MM, et al. Use of concept mapping to characterize relationships among implementation strategies and assess their feasibility and importance: results from the Expert Recommendations for Implementing Change (ERIC) study . Implement Sci 2015; 10 ( 01 ):109. [PMC free article] [PubMed] [Google Scholar]
42. Beidas RS, Stewart RE, Adams DR, et al. A multi-level examination of stakeholder perspectives of implementation of evidence-based practices in a large urban publicly-funded mental health system . Adm Policy Ment Health 2016; 43 ( 06 ):893–908 [PMC free article] [PubMed] [Google Scholar]
43. Grol R, Dalhuijsen J, Thomas S, Veld C, Rutten G, Mokkink H. Attributes of clinical guidelines that influence use of guidelines in general practice: observational study . BMJ 1998; 317 ( 7162 ): 858–861 [PMC free article] [PubMed] [Google Scholar]
44. Onion CW, Dutton CE, Walley T, Turnbull CJ, Dunne WT, Buchan IE. Local clinical guidelines: description and evaluation of a participative method for their development and implementation . Fam Pract 1996; 13 ( 01 ):28–34 [PubMed] [Google Scholar]
45. Feder G, Eccles M, Grol R, Griffiths C, Grimshaw J. Clinical guidelines: using clinical guidelines . BMJ 1999; 318 ( 7185 ):728–730 [PMC free article] [PubMed] [Google Scholar]
46. Borbas C, Morris N, McLaughlin B, Asinger R, Gobel F. The role of clinical opinion leaders in guideline implementation and quality improvement . Chest 2000; 118 ( 02 ):24S–32S [PubMed] [Google Scholar]
47. Leis JA, Shojania KG. A primer on PDSA: executing plan-do-study-act cycles in practice, not just in name . BMJ Qual Saf 2017; 26 ( 07 ): 572–577 [PubMed] [Google Scholar]
48. Stikes R, Barbier D. Applying the plan-do-study-act model to increase the use of kangaroo care . J Nurs Manag 2013; 21 ( 01 ): 70–78 [PubMed] [Google Scholar]
49. Ivers NM, Sales A, Colquhoun H, et al. No more ‘business as usual’ with audit and feedback interventions: towards an agenda for a reinvigorated intervention . Implement Sci 2014; 9 ( 01 ):14. [PMC free article] [PubMed] [Google Scholar]
50. de Koning H, Verver JP, van den Heuvel J, Bisgaard S, Does RJ. Lean six sigma in healthcare . J Healthc Qual 2006; 28 ( 02 ):4–11 [PubMed] [Google Scholar]