- Open Access
Results of a sector-wide quality improvement initiative for substance-abuse care: an uncontrolled before-after study in Catalonia, Spain
Substance Abuse Treatment, Prevention, and Policy volume 5, Article number: 26 (2010)
The Health Department of the Regional Government of Catalonia, Spain, issued a quality plan for substance abuse centers. The objective of this paper is to evaluate the impact of a multidimensional quality improvement initiative in the field of substance abuse care and to discuss potentials and limitations for further quality improvement.
The study uses an uncontrolled, sector-wide pre-post design. All centers providing services for persons with substance abuse issues in the Autonomous Community of Catalonia participated in this assessment. Measures of compliance were developed based on indicators reported in the literature and by broad stakeholder involvement. We compared pre-post differences in dimension-specific and overall compliance-scores using one-way ANOVA for repeated measures and the Friedman statistic. We described the spread of the data using the inter-quartile range and the Fligner-Killen statistic. Finally, we adjusted compliance scores for location and size using linear and logistic regression models.
We performed a baseline and follow up assessment in 22 centers for substance abuse care and observed substantial and statistically significant improvements for overall compliance (pre: 60.9%; post: 79.1%) and for compliance in the dimensions 'care pathway' (pre: 66.5%; post: 83.5%) and 'organization and management' (pre: 50.5%; post: 77.2%). We observed improvements in the dimension 'environment and infrastructure' (pre: 81.8%; post: 95.5%) and in the dimension 'relations and user rights' (pre: 66.5%; post: 72.5%); however, these were not statistically significant. The regression analysis suggests that improvements in compliance are positively influenced by being located in the Barcelona region in case of the dimension 'relations and user rights'.
The positive results of this quality improvement initiative are possibly associated with the successful involvement of stakeholders, the consciously constructed feedback reports on individual and sector-wide performance and the support of evidence-based guidance wherever possible. Further research should address how contextual issues shape the uptake and effectiveness of quality improvement actions and how such quality improvements can be sustained.
Over the last decade a multitude of performance assessment and quality improvement activities have targeted the health care sector. In addition to regional and national initiatives, efforts led by the World Health Organization or the Organization for Economic Development and Cooperation aim at assessing and evaluating health care and health systems in an international comparative perspective [1–3]. Indeed, industrial approaches to quality improvement and performance assessment have been applied to health care since the 1960s, spearheaded by the works of Avedis Donabedian [4, 5]. While most attention has initially focused on the hospital setting and primary care, more and more attention is being been paid on other health and social care facilities in general, and mental health [6–8] and substance-abuse centers in particular [9–12].
Relevant initiatives to improve the quality in this sector are for example the treatment protocols and technical assistance publications (including models of good practice) that are being prepared by the US Center for Substance Addiction Treatment ; the Accreditation Manual for Behavioral Care issued by the Joint Commission , or the epidemiological studies and reports that are linked to evaluation and improvement efforts published by the European Monitoring Center for Drugs and Drug Addiction . Moreover, traditional quality management models, such as the European Foundation for Quality Management (EFQM) model, have been applied to substance abuse centers . In Spain, too, various Autonomous Communities have launched activities to improve the quality of substance-abuse centers, for example the evaluation based on the Common Assessment Framework carried out by the Government of Cantabria, the assessment of compliance of substance abuse centers with a set of standards conducted by the Institute for Addiction Research in Madrid and the quality plan and external evaluation system deployed by the Catalonian Government to improve substance abuse services . We described the conceptual basis of the latter system in a previous article .
In the following, we provide a more in-depth application of the evaluation framework widely used in health care to centers that provide ambulatory substance abuse services for people with drug dependencies. In order to address the quality issues of concern in these centers (adherence to treatment, continuity of care) and the multiple factors causing them, quality improvement efforts need to adapt a multi-professional approach that takes into consideration social, clinical, psychological factors as well as the user's views and satisfaction. The objective of this paper is thus to evaluate the impact of a multidimensional quality improvement initiative in the field of substance abuse care in the Autonomous Region of Catalonia, Spain, and to discuss potentials and limitations for further quality improvement.
This study was conducted in the context of the quality plan issued by the Health Department of the Autonomous Government of Catalonia. The plan states the need for quality improvement efforts directed at centers for substance-abuse service and follow up (CAS), with particular attention on existence of assessments of quality as perceived by the users of the centers. CAS provide services by multi-professional teams including physicians, psychologists, social works and nurses that review on a case by case basis the needs and treatment forms for each patient and assess further health and social care needs, such as the use of therapeutic communities or hospitalization in an detoxication unit. These centers can offer different treatment options such as methadone treatment or treatment with narcotic antagonists and a like. In addition, some CAS are linked to small peripheral centers that facilitate for example medication dispensing. Performance of these small units is not included in the analysis presented here.
Measures and data sources
Considering that developing performance measures can be very resource demanding, we developed the measures for external evaluation based on reviews of the literature, consensus methods including subject experts and involving a broad range of stakeholders. The key premise in this process is to address issues that are of concern of practicing health and social care professionals, to include the latest scientific evidence and expert knowledge in the definition of indicators and to ensure practicality and timeliness of the work. The measurement development would first draw on previously reported and validated survey measures, data collection instruments and indicators. Multi-disciplinary stakeholder groups then meet five to seven times to discuss how the measures reflect the main problems in the sector, identify objectives that can be met within three years, consent on the key measurement area and review, prioritize and technically define the measures. The development and psychometric testing of new indicators that require complex surveys or clinical scores is not the intention and is beyond the scope of our work. Indicators are defined using a standardized template. For each indicator, standards are set based on a review of the literature and consensus in the group. Standards are set to encourage improvement as considered to be achievable by stakeholders within the next 2 or 3 years, after which they will be measured again and a new set is developed. Indicators for which a majority of centers are already performing very well will be discarded. The process of measure development is described in more detail elsewhere [18, 19].
Data sources used for the external evaluation depended on the criteria and dimension of care assessed. This includes revisions of clinical records for care processes and follow-up indicators, direct observation for structural indicators, harm reduction program and assessment of systems for user evaluation, and revision of specific documents such as program descriptions, protocols and minutes for the remaining indicators. For standardization and quality assurance purposes, all data are recorded in a specifically developed program in MS ACCESS.
For indicators based on user records we took a random sample of 100 records based on the user lists provided by the centers. All patients are included at the sampling stage. In the instance of small centers not reaching this number we assessed all cases. Data extraction is based on an algorithm taking into account the type of indicator, non-applicability of measurable elements and level of compliance. These measures are then aggregated at indicator level (an indicator is typically based on around four measurable elements) and in a subsequent step at dimension level. The original data format specifies for each center, indicator and assessment period (pre-post) whether the predetermined standards were met. Based on the indicator-specific dichotomous compliance scores we calculated center-level compliance scores for each performance dimension. The resulting scores can be expressed as follows:
Overall compliance: for each centre, the total number of indicators that are in compliance divided by the total number of indicators evaluated (where "Compliant/NotCompliant" is a dichotomous variable(1 = compliant, 0 = not compliant) and m is the number of indicators evaluated.
For each of the 4 dimensions: for each centre, the total number of indicators related to the respective dimension divided by the total number of indicators evaluated for the same dimension (where "Compliant/NotCompliant" is a dichotomous variable(1 = compliant, 0 = not compliant) and "m Indicadors Evaluated in the dimension" is the number of indicators evaluated):
These resulting scores are continuous (except for the dimension 'environment and infrastructure' as this is composed of only one indicator) and allow for the comparison of pre-post compliance scores. While the aggregate measures account for applicability of the item assessed and for missing data, we did not weight individual indicators. For example, the correct registration of drug prescriptions is not assigned more weight than, say, providing vaccination against Hepatitis B to users at risk. The rationale for this is that given the different views in the multi-disciplinary measure development stakeholder group (involving users) we could not consent on criteria that would justify weighting of indicators.
External evaluation and quality improvement
Centers were informed before the visit about the evaluation schedule and the documents required during the evaluation. Baseline evaluation was carried out in 2001 and follow up evaluation was performed in 2004. Between the two evaluations, centers received specific performance reports including targeted recommendations for quality improvement. At the same time a sector-wide report was published including the baseline assessment (without identifying individual providers) and recommendations for sector-wide improvements. Evaluators received a 40 hour training for the assessment exercise including a presentation of the indicators, data sources, the assessment instrument and overall methodology. A pilot test was carried out to test the assessment material and the reliability of the assessment by different evaluators. The assessment of inter-rater reliability itself is outsourced to another scientific institute (the Ibero-American Cochrane Collaboration, Barcelona) to limit any biases results from personal acquaintance with the evaluators.
Subsequent to the evaluation, and in order to facilitate quality improvement initiatives in the participating centers, each center received an individualized quality improvement report. This report indicates the results of the assessment, offers concrete proposals for quality improvement and provides anonymized comparative data on other center's performance. In addition, a sector wide report is prepared to address quality issues beyond individual center's research and is discussed at sector level, involving health policy and other stakeholders. Based on the later report, quality improvement actions are consented at sector level.
We computed indicator-, dimension-specific and overall compliance scores based on the calculation of compliance with measurable elements. Using these scores we present overall and dimension-specific results for baseline and follow up evaluation with descriptive statistics of tendency and variation. We performed one-way analysis of repeated measures (ANOVA) and the Friedman statistic to assess the level of statistical significance for differences between baseline and follow-up evaluation. To test whether sector wide changes in performance are associated with reduced spread in individual center's performance we assessed changes in inter-quartile range and the Fligner-Killeen statistic for the performance dimensions between baseline and follow-up assessment. Finally, in order to control for the effect of the structural center-characteristics 'location' and 'size', we entered these variables into multiple linear and logistic regression models. Data analysis was performed using SPSS 17 and the R statistical software package (version 2.10.1).
We assessed and followed up all 22 Centers for Attention and Follow Up participating in the quality plan and providing services to persons addicted to drugs and other substances in the Autonomous Community of Catalonia, Spain (reflecting the whole sector). In Table 1, we provide an overview on the distribution of centers by geographic territory as well as descriptive statistics on the number and characteristics of their users (Table 1).
All Centers for Attention and Follow Up were evaluated twice using consensus indicators developed jointly with the stakeholders. These indicators were designed to be applicable to all centers. Table 2 gives an overview on the indicators, their relation to the multidimensional evaluation framework and their distribution across dimensions, themes, indicators, and measurable elements. Overall, the evaluation framework includes 4 dimensions, 20 themes, 35 indicators and 169 measurable elements (Table 2).
The number of measurable elements for each indicator was determined based on the indications from stakeholders regarding key aspects to be assessed. The median number of measurable elements per indicator is 4, but some indicators are assessed by only one measurable element while others are assessed by up to 16. Evaluators assessing the indicators are specifically trained in data collection and use standardized algorithms and data entry sheets for data collection.
The number of cases that forms the basis of the calculation of the indicators differs substantially as some indicators are based on the random review of a sample 100 user records, others are based on a full review of user records (in case the number of users for this indicator is <100 at center level) and again others are based on assessment at center level, for example, the external evaluators' assessment of the availability of certain organizational procedures.
For some indicators we observed negative changes in compliance ratings (indicators 2, 7, 16, 17, 19, 20, 25). In case of indicators 2 and 25, these negative changes can be explained by a substantial increase in demand that coincided with the evaluation period and centers were not prepared for this increase. For indicators 16, 17, 19 and 20 other institutions or third parties were established during the evaluation period and took over some responsibilities that were initially held at the level of the center (such as the complaint handling system). Finally, indicators 7 on the follow up of users proved to be very difficult to achieve due to the reluctance of those users already on methadone to establish contact with the CAS.
After the baseline evaluation using these indicators, a bundle of quality improvement actions were initiated both at the level of the sector and at the level of individual centers. Key to these quality improvement actions were the performance reports provided to each center which depict the center's performance in relation to the sectors performance. Thus, each center could identify targeted quality improvement actions in areas where the center's performance was below standard. In addition to these center-specific performance reports, a sector-wide performance report was published focusing on quality improvement actions that require consorted actions or changes in the normative framework. The following table gives an example of a performance report for one of the indicators (Table 3).
The performance report provides information on comparative benchmarking for each indicator and pinpoints at measurable elements that were evaluated particularly low. Target audience of the report is the center's management and involved professionals. It should be noted that the performance report sent to each center includes such a detailed assessment for each of the 35 indicators, accompanied by graphical illustrations of relative performance.
After the publication of the performance reports, local quality improvement actions were established, implemented and evaluated in the follow-up evaluation. For example, centers participated in the public presentation of the results and in the elaboration of the document on good practices. The good-practice document provides for all indicators that were evaluated as below target performance detailed rationales, improvement steps and reference documents in order to facilitate the use of this information for quality improvement activities. Specific quality improvement actions initiated on the basis of the good-practice document may for example be the administration of a user satisfaction survey, the implementation of a new protocol or the introduction of an organizational policy towards collaboration with other providers of substance abuse services. Based on individual performance reports, some centers also requested specific advice from the research coordination team on how to improve the quality of services.
Major improvements between baseline and follow up evaluation can be observed for most of the 20 themes evaluated (Figure 1, Results at the level of individual indicators are reported in Table 2). Exceptions can be observed in the four areas: adequate space, prevention, community relations and overall satisfaction. In the remaining 16 areas, major improvements were made at sector level.
In order to assess statistical significance for the differences observed, we carried out comparisons between baseline and follow up evaluation and performed a one-way ANOVA analysis for repeated measures and the Friedman statistic to assess the improvements in overall and dimension-specific performance (Table 4).
Global improvement in mean compliance between baseline assessment and follow up evaluation is substantial (29.9%; 95% Confidence Interval (CI): 22.4-37.3; ANOVA, F1, 21 = 47.4). Since we were interested not only in the improvements of individual centres but also in the improvement of the sector as a whole, we assessed whether improvements observed are also accompanied by diminished variance between centers. The assessments using the inter-quartile range (IQR) yielded reductions for the dimensions 'overall compliance' (IQRpre: 0.179; IQRpost: 0.142) and 'care pathway' (IQRpre: 0.18, IQRpost: 0.14), but not for the dimensions 'relations and user rights' (IQRpre: 0.18, IQRpost: 0.30) and 'organization and management' (IQRpre: 0.16, IQRpost: 0.24). Given the dichotomous nature of the indicator for the dimension environment and infrastructure, we could not compute this statistic for this dimension.
In order to formally analyze the reduction in variance between pre-post assessments, we performed the Fligner-Killen test for changes in overall and dimension-specific compliance which yielded the following p-values: 'overall compliance' = 0.418 (Fligner-Killeen Test = 0.67; df = 1), 'care pathway' = 0.043 (Fligner-Killeen Test = 4.10; df = 1), 'relations and users rights' = 0.165 (Fligner-Killeen Test = 1.93; df = 1) and 'organization and management' = 0.963 (Fligner-Killeen Test = 0.002; df = 1). According the results of the Fligner-Killen test, the reduction in variance between pre and post assessment is only significant for the dimension 'care pathway'.
In terms of dimension-specific results, improvements ranging from 9.0% to 52.9% can be observed. Improvements were statistically significant in the dimensions 'care pathway' (pre: 66.5%, post: 83.5%; 95% CI for improvement in compliance: 15.0-36.1, Chi21 = 9.8) and 'organization and management' (pre: 50.5%, post: 77.2%; 95% CI for improvement in compliance: 37.8-67.9, Chi21 = 11.64). We observed improvements for the dimension environment and infrastructure' (pre: 81.8%; post: 95.5%) and the dimension 'relations and user rights' (pre: 66.5%; post: 72.5%), however, these improvements were not statistically significant.
In a final step, and in order to assess whether improvements in compliance are influenced by the location of the center or its size, we entered the following variables into a regression model: change in compliance between baseline and follow up evaluation, location and size. To this end we created dummy variables for location of the center (Barcelona vs. elsewhere) and for its size (small, medium, large; based on the distribution of the mean number of users per centers in terciles). Table 5 reports the improvement in overall and dimension-specific compliance by location and size of the center using multiple linear and logistic regression models. According to the model for 'overall compliance' (R2 = 0.19, F3, 18 = 1.42), smaller centers reach higher improvements (B-value for 'big' compared to 'small' = -0.39) and centers in Barcelona improve more than others (B-value for 'Barcelona' = 0.65), however, these factors are not significant in the model (p-value = 0.27). Thus, it seems that the improvements presented in Table 4 are not confounded by these factors and centers appear to improve on the whole, irrespective of location and size.
In the linear regression models for the dimensions 'care pathway' and 'organization and management' the factors location and size are not significant. However, the factor 'location = Barcelona' is significant in the model for the dimension 'relations and user rights' (p = 0.03, F3, 18 = 3.80, R2 = 0.39). For the dimension 'environment and infrastructure' we fitted a logistic regression model, given the dichotomous nature of this dimension based on only one indicator. The odds-ratios and corresponding 95% confidence intervals do not suggest an influence of size and location on probability to improve.
We assessed and facilitated quality improvement activities in 22 Centers for Attention and Follow Up in the Autonomous Community of Catalonia, Spain. Using consensus indicators developed with the involvement of key stakeholders we carried out a baseline assessment, prepared targeted performance reports and performed a follow-up assessment. Overall, we observed substantial and statistically significant improvements in the whole sector. The results of the regression analysis in general do not suggest that centers in Barcelona and those being larger differ substantially in their compliance scores. An exception is the dimension 'relations and user rights' for which the regression analysis suggest a positive effect of a center being located in Barcelona.
We interpret the improvements as follows: first, the broad involvement of professionals in the design of consensus indicators and the participation of a large number of centers from the whole sector creates common sense and motivates improvement. Moreover, it allows setting challenging but realistic targets. Secondly, the combination of targeted feedback - to individual centers on the one hand and to the whole sector on the other - allows addressing confidentially those providers with low performance and, in collaboration with all stakeholders, bringing about change in the whole sector for those issues that require changes at the regulatory level. Both feedback mechanisms are consciously constructed, elaborated in collaboration with the participants and are supported by evidence-based guidance wherever possible. Moreover, center's feedback reports are addressed to the center's Chief Executive rather than individual professionals, and prepared in a way to support management in implementing change. Finally, the involvement of the whole sector and a wide range of stakeholders led to the development of small networks and collaborations between centers that promote the exchange of best-practice models and improvement projects.
However, a number of limitations of the study should be noted. First, we did not perform systematic psychometric validations of the measures used. Most of the indicators are based on existing measures previously validated and reported the literature. However, some measures also stem from the discussions and ratings by the consensus panels. While these measures have high face validity in the stakeholder group they may lack the presumed associations with outcomes, which should be addressed by future research . Secondly, several factors might have confounded the improvements observed. As a starting point, this is the first evaluation of the sector that was carried out and it might be argued that centers started from a rather low baseline level. The improvements observed might thus reflect quality improvement actions that are easy to implement. Another factor explaining the improvements observed might be the three year time lag between baseline and follow up assessment. This time lag is on the one hand justified given the efforts to perform the assessment, provide center-specific and sector-wide performance feedback, implement quality improvement actions and prepare the follow up evaluation. On the other hand, this time lag also means that changes or evolutions in delivery, organization, or regulatory measures of social care might have influenced the results. Controlling for these potential confounding factors; however, would be in contradiction to our methodological approach which aims at involving all sector stakeholders and policy makers in addressing those quality problems that are beyond the control of an individual center. In this sense, our approach puts higher emphasis on working with the sector's stakeholders in addressing real life quality improvement actions (ecological validity) than on reducing threats to linking specific interventions to specific outcomes measures (internal validity) . This is in line with what Berwick and quality improvement experts calls the 'pragmatic science' approach which gives importance to the context that shape the design and implementation of quality improvement strategies, rather than assuming that a 'best-model' exists that can be applied irrespective of contextual characteristics [22, 23]. While maintaining the main orientations of our approach we plan for future evaluations to allow the applications of multi-level statistical models, which was not possible given the current algorithms for data collection . Such an approach would allow a more in-depth analysis of the variations in improvements between and within centers, accounting for characteristics of users and organizational context.
In conclusion, we report on an evaluation and improvement method for centers for attention and follow up of drugs users in Catalonia, Spain, in which importance is given to involve stakeholders in the development of performance measures and in which assessments are performed externally by independent evaluators. This method is designed to involve and motivate a whole sector in a given area. The results indicate that major improvements between baseline and follow up evaluation using these performance measure that are developed to 'make sense' for those being evaluated and that use standards that users themselves consider to be motivated and attainable. We believe that the broad involvement of stakeholders and the non-punishing nature of our approach to evaluation and improvement are key factors that might explain the substantial improvements observed. Further research should address how contextual issues shape the uptake and effectiveness of quality improvement actions, and how such quality improvements can be sustained.
WHO: World health report: Improving health systems. 2000, Geneva: World Health Organization
Groene O, Skau JK, Froelich A: An international review of projects on hospital performance assessment. Int J Qual Health Care. 2008, 20: 162-71. 10.1093/intqhc/mzn008.
OECD: Health at a glance 2008. 2008, Paris: Organization for Economic Cooperation and Development
Donabedian A: The quality of medical care. Science. 1978, 200: 856-64. 10.1126/science.417400.
Donabedian A: Quality assessment and monitoring. Retrospect and prospect. Eval Health Prof. 1983, 6: 363-75. 10.1177/016327878300600309.
Committee on Quality Assurance and Accreditation Guidelines for Managed Behavioral Health Care: Managing Managed Care. Quality Improvement in Behavioral Health. 1997, Institute of Medicine: The National Academies Press, Washington, DC
Committee on Crossing the Quality Chasm: Adaptation to Mental Health and Addictive Disorders: Improving the Quality of Health Care for Mental and Substance-Use Conditions: Quality Chasm Series. 2006, Institute of Medicine: The National Academies Press, Washington, DC
Department of Veterans Affairs: VHA Handbook 1160.01. Uniform Mental Health Services in VA Medical Centers and Clinics DOV, Washington 2008http://www1.va.gov/vhapublications/ViewPublication.asp?pub_ID=1762
Garnick DW, Lee MT, Horgan CM, Acevedo A, Washington Circle Public Sector W: Adapting Washington Circle performance measures for public sector substance abuse treatment systems. J Subst Abuse Treat. 2009, 36: 265-77. 10.1016/j.jsat.2008.06.008.
Harris SK, Sherritt L, Van Hook S, Wechsler H, Knight JR: Alcohol policy enforcement and changes in student drinking rates in a statewide public college system: a follow-up study. Subst Abuse Treat Prev Policy. 2010, 4: 18-10.1186/1747-597X-5-18.
Hoffman KA, Ford JH, Choi D, Gustafson DH, McCarty D: Replication and sustainability of improved access and retention within the Network for the Improvement of Addiction Treatment. Drug Alcohol Depend. 2008, 98: 63-9. 10.1016/j.drugalcdep.2008.04.016.
Nabitz U, Walburg J: Addicted to quality–winning the Dutch Quality Award based on the EFQM Model. Int J Health Care Qual Assur Inc Leadersh Health Serv. 2000, 13: 259-65.
Substance Abuse & Mental Health Services Administration (SAMHSA) and Center for Substance Abuse Treatment (CSAT) Treatment Improvement Protocols:http://0-www.ncbi.nlm.nih.gov.brum.beds.ac.uk/bookshelf/br.fcgi?book=hssamhsatip
Joint Commission: Standards for Behavioral Health Care 2004-2005. 2004, Oakbrook Terrace: Joint Commission Resources
European Monitoring Center for Drugs and Drug Addiction: National Report to the EMCDDA by the Retox National Focal Point. New developments, Trends and in-depth information on selected issues Spain 2008http://www.emcdda.europa.eu/attachements.cfm/att_86714_EN_NR_2008_ES.pdf
Nabitz U, Van den Brink W, Waburg J: A quality framework for addiction treatment. Addictive Behaviours. 2005, 30: 1254-1260. 10.1016/j.addbeh.2004.12.006.
Divar Conde JM: Sectoral planning for quality of care based on consensus indicators. Approximation to ambulatory care for drug dependent users in Aragon. 2009, Doctoral Thesis. University of Zaragoza: Zaragoza, [in Spanish]
Hilarion P, Suñol R, Groene O, Vallejo P, Herrera E, Saura RM: Making performance indicators work: the experience of using consensus indicators for external assessment of health and social services at regional level in Spain. Health Policy. 2009, 90: 94-103. 10.1016/j.healthpol.2008.08.002.
Hilarion P, Miro M, Salva A, Suñol R, Minguell C, Caja C: Social/health care services: indicators to assess the quality of patient and family centerd services. 2006, Barcelona: Generalitat de Catalunya Departament de Salut, [in Catalan]
Harris AH, Kivlahan DR, Bowe T, Finney JW, Humphreys K: Developing and validating process measures of health care quality: an application to alcohol use disorder treatment. Med Care. 2009, 47: 1244-1250. 10.1097/MLR.0b013e3181b58882.
Brewer M: Research Design and Issues of Validity. Handbook of Research Methods in Social and Personality Psychology. Edited by: Reis H, Judd C. 2000, Cambridge: Cambridge University Press
Berwick D: Broadening the view of evidence-based medicine. Qual Saf Health Care. 2005, 14: 315-316. 10.1136/qshc.2005.015669.
Davidoff F, Batalden P, Stevans D, Ogrinc G, Mooney S, SQUIRE Development Group: Publication Guidelines for Improvement Studies in Health Care: Evolution of the SQUIRE Project. Qual Saf Health Care. 2008, 149: 670-676.
Leyland AH, Groenewegen PP: Multilevel modelling and public health policy. Scand J Public Health. 2003, 31: 267-74. 10.1080/14034940210165028.
We acknowledge the support of all stakeholders involved in the design and execution of this study, without which the quality improvement efforts would not have been so substantial. We also acknowledge the help of Elisenda Grau in the data handling and Nuria Mora in the statistical analysis. Finally, we would like to thank the editor and the reviewers for the constructive remarks, corrections and suggestions for the analytical strategy.
The authors declare that they have no competing interests.
PH, RS, JC, RML and OG conceived of the study. OG performed the statistical analysis. OG, PH and RS drafted the manuscript. All authors critically read, revised and approved the final manuscript.
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
About this article
Cite this article
Hilarion, P., Groene, O., Colom, J. et al. Results of a sector-wide quality improvement initiative for substance-abuse care: an uncontrolled before-after study in Catalonia, Spain. Subst Abuse Treat Prev Policy 5, 26 (2010) doi:10.1186/1747-597X-5-26
- Quality Improvement
- Care Pathway
- Performance Report
- Measurable Element
- Autonomous Community