Use of a Respiratory Care Model in Population Older Than 15 Years in Orizaba, Veracruz Mexico

The design was that of a prospective quasi-experimental pre-post study. A pre-survey and a follow-up survey were used.Investigators selected 4 health centers in the Sanitary Jurisdiction in Orizaba, Veracruz, two located in rural areas (Potrerillo and Tlilapan) and two located in urban areas (Río Blanco and Orizaba). The pre-survey (phase1) was conducted on all consenting individuals older than 15 years seeking health services in selected health centers. Before the physician consultation, researchers investigated sociodemographic, epidemiologic, clinical characteristics and reason for seeking health services. Immediately after consultation, researchers asked the patient his/her diagnoses. Investigators confirmed all diagnoses with treating physicians. If the patient had been diagnosed with respiratory disease, researchers collected information on prescribed treatment, and classified the type of respiratory disease and comorbidities based on the 10th International Classification of Diseases (ICD-10). Researchers conducted a home visit one month after the consultation to determine clinical evolution, occurrence of referral to a specialist, if the patient had obtained prescribed drugs at the health center or at a private drugstore and if the patients had been hospitalized.

During September 2014 researchers trained health personnel from the participating health centers on implementation of the proposed model based on the AIRE guidelines previously approved by the AIRE committee of the INER. The AIRE guidelines have been adapted from WHO "Practical Approach to Lung Health" and include management of asthma, acute respiratory infections (ARI, that comprises acute bronchitis and pneumonia), chronic obstructive pulmonary disease (COPD), infections of the upper respiratory airways, and preventive measures for smoking, tuberculosis and breathing disorders during sleep.

The model was presented to the authorities of the participating health services and of the local hospital (Regional Hospital in Río Blanco (HRRB, its acronym in Spanish) as well as to health personnel involved in treatment of respiratory diseases, to promote their participation, counseling and support for the referral and counter-referral of patients.

The implementation of the guidelines included: 1) availability of the guidelines for the management of respiratory diseases in primary care "AIRE", that were revised and modified based on the analysis of the first phase of the study. Physicians providing primary care in the health centers and at the local hospital revised the guidelines to ensure that language was understandable. 2) Distribution of educational materials to health personnel including printed and electronic versions of AIRE guidelines, algorithm for management of acute and severe respiratory disease, and operational and good clinical practices manuals; 3) Training at the INER of the pulmonologist of the local hospital on AIRE guidelines and certification in spirometry. 4) Development of two workshops for 97 physicians and nurses of the participating health centers on AIRE Guidelines, oximetry tests, spirometry, referral and counter-referral of patients and clinical cases with the participation of specialized physicians from the INER, HRRB and the Instituto Nacional de Salud Pública (INSP); 5) Creation of a respiratory health network between primary care facilities, hospital and public health offices.

Researchers considered that patients participated in one of the two phases according to the period in which they sought health services.

Researchers defined a patient with a respiratory disease as any patient who after consultation with a physician was diagnosed with a respiratory disease according to the ICD-10. The medical diagnosis was obtained from the patient and confirmed with the treating physician and classified according to ICD-10. Respiratory diseases were classified in the following categories: emphysema, asthma and bronchitis, allergic rhinitis, lung cancer, chronic bronchitis, COPD and TB. Researchers also used a second categorization as follows: viral infections, bacterial infections and chronic diseases following the national clinical practice guidelines . Comorbidities were classified in the following categories: diabetes, hypertension, HIV and heart disease. The reason for seeking medical attention was classified into 20 categories. Researchers defined timeliness in requesting of health services measuring the interval from the onset of symptoms to request of physician consultation both as a continuous variable and as a dichotomic variable using as cut off >3 days. Researchers obtained the physician prescription and classified prescribed drugs as antibiotics, corticosteroids, cough suppressants, expectorants, bronchodilators, analgesics and antihistamines. Researchers also grouped cough suppressants, expectorants, analgesics and antihistamines in one category.

Researchers categorized age in 3 groups: 15 to 44 years, 45 to 59 years and >60 years. The variable "type of floor" was classified in two categories (hard floor/earth floor) according to the material of the floor.

Smoking was categorized as whether the patient was a current smoker or not. Alcohol consumption was dichotomized in more or equal or less than 14 alcoholic drinks per week. Use of drugs was categorized as to whether the patient reported using drugs during the study period. Researchers considered a patient was exposed to dust or smoke if he/she had worked for more than a year in a place with dust, smoke or vapors. Researchers considered occupational exposure to asbestos if the patient had worked for more than a year in a place with asbestos. Biomass exposure was categorized as whether the patient was currently exposed or not.

Researchers compared sociodemographic, epidemiologic and clinical characteristics of patients demanding physician consultation and of patients who were diagnosed with respiratory disease according to study period. In the bivariate analysis researchers used U of Mann Whitney test for continuous variables that didn't follow a normal distribution, chi-squared of Pearson test for dichotomous variables and binomial test for categorical variables.

Researchers compared by bivariate analyses the following outcomes according to study phase: 1) Acquisition of prescribed drugs in the pharmacy of the health center; 2) Treatment based on guidelines; 3) Clinical outcome; 4) Referral to a specialist. Using multivariate unconditional logistic regression, we built three models to investigate the association between phase of the study and 1) Acquisition of prescribed drugs in the pharmacy of the health center; 2) Treatment based on guidelines; 3) Clinical outcome. Variables with p < 0.20 in the bivariate analysis and biological plausibility were included in multivariate models. Multivariate models were built considering patients as clusters given that each patient could have had more than one drug prescription. Researchers estimated the odds ratio (OR) and 95 per cent confidence intervals (95% CI), and identified the covariates that were independently associated with each outcome.

All statistical analyses were performed using the statistical package STATA V13.3 (StataCorp LP).