You are here

Article Text

Article menu

Download PDFPDF

XML
General practice / Family practice
Research
Impact of primary care nursing workforce characteristics on the control of high-blood pressure: a multilevel analysis
  1. Ana Parro-Moreno1,
  2. Pilar Serrano-Gallardo2,
  3. Antonio Díaz-Holgado3,
  4. Jose L Aréjula-Torres3,
  5. Victor Abraira4,
  6. Isolina M Santiago-Pérez5,
  7. Jose M Morales-Asencio6
  1. 1Department of Nursing, Department of Preventive Medicine and Public Health, Department of Surgery, School of Medicine, Universidad Autónoma de Madrid/IISPHM, Madrid, Spain
  2. 2Department of Nursing, School of Medicine, Universidad Autónoma de Madrid/IDIPHIM/INAECU, Madrid, Spain
  3. 3Information System Unit, Directorate for Public Health, Health Service of Madrid, Spain
  4. 4Clinical Biostatistics of Ramón y Cajal University Hospital/IRYCIS/Centre for Biomedical Research on Epidemiology and Public Health (CIBERESP), Madrid, Spain
  5. 5Epidemiology Unit, Galician Directorate for Public Health, Galician Health Authority, Santiago de Compostela, Spain/IBIMA, A Coruña, Santiago de Compostela, Spain
  6. 6Faculty of Health Sciences, Universidad de Málaga, Málaga, Spain
  1. Correspondence to Dr Ana Parro-Moreno; anaisabel.parro{at}uam.es

Abstract

Objective To determine the impact of Primary Health Care (PHC) nursing workforce characteristics and of the clinical practice environment (CPE) perceived by nurses on the control of high-blood pressure (HBP).

Design Cross-sectional analytical study.

Setting Administrative and clinical registries of hypertensive patients from PHC information systems and questionnaire from PHC nurses.

Participants 76 797 hypertensive patients in two health zones within the Community of Madrid, North-West Zone (NWZ) with a higher socioeconomic situation and South-West Zone (SWZ) with a lower socioeconomic situation, and 442 reference nurses. Segmented analyses by area were made due to their different socioeconomic characteristics. Primary outcome measure: Poor HBP control (adequate figures below the value 140/90 mm Hg) associated with the characteristics of the nursing workforce and self-perceived CPE.

Results The prevalence of poor HBP control, estimated by an empty multilevel model, was 33.5% (95% CI 31.5% to 35.6%). In the multilevel multivariate regression models, the perception of a more favourable CPE was associated with a reduction in poor control in NWZ men and SWZ women (OR=0.99 (95% CI 0.98 to 0.99)); the economic immigration conditions increased poor control in NWZ women (OR=1.53 (95% CI 1.24 to 1.89)) and in SWZ, both men (OR=1.89 (95% CI 1.43 to 2.51)) and women (OR=1.39 (95% CI 1.09 to 1.76)). In all four models, increasing the annual number of patient consultations was associated with a reduction in poor control (NWZ women: OR=0.98 (95% CI0.98 to 0.99); NWZ men: OR=0.98 (95% CI 0.97 to 0.99); SWZ women: OR=0.98 (95% CI 0.97 to 0.99); SWZ men: OR=0.99 (95% CI 0.97 to 0.99).

Conclusions A CPE, perceived by PHC nurses as more favourable, and more patient–nurse consultations, contribute to better HBP control. Economic immigration condition is a risk factor for poor HBP control. Health policies oriented towards promoting positive environments for nursing practice are needed.

  • PRIMARY CARE
  • Nursing workforce
  • Nursing Work Index

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/

http://dx.doi.org/10.1136/bmjopen-2015-009126

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • This is the largest study in Spain to determine how the characteristics of the Primary Health Care nursing workforce, including the self-perceived working environment, exert influence on the control of high-blood pressure of adult patients.

    • The cross-sectional and observational nature of the study limits the establishment of causal relationships, but the analytical approach, linked to the fact that the vast majority of independent variables are temporarily located prior to the outcome variable, minimises such limitation. Anyhow, future longitudinal analytical and experimental studies are necessary to confirm a few of the outcomes.

    • The utilisation of documental sources can lead to information bias, although the current information systems, such as the ones used in this study, are demonstrating good validity and consistency as a source of information on epidemiological studies.

    Introduction

    The results on the health of the population are influenced by multiple factors, among which the healthcare system is an intermediate determinant and, within it, the actions of healthcare professionals.1 Any healthcare system is not solely sustained by medical assistance, but also depends in a determinant manner on the care given by nursing staff.2 In recent times, studies have been made on nursing staff characteristics, from the viewpoint of their impact on patients’ health results, such as mortality, adverse effects or avoidable complications.3–6 Most of these studies have been conducted in the hospital environment, while only Griffiths et al7 ,8 performed their study in Primary Health Care (PHC) where they found better results with regard to control of chronic diseases where the user/nurse ratio was more favourable, that is to say, when the number of users assigned to each nurse was smaller.

    Moreover, evidence points to the practice environment as one of the variables with the greatest impact on the quality of care, with clear repercussions on health results, such as mortality, average hospital stay and patient satisfaction.9–11 This construct was defined by Lake12 in 2002 as the set of organisational characteristics at the workplace that hinder or aid professional practice.

    Factors inherent to progress (ageing, obesity and unhealthy lifestyles) are contributing to an increasing prevalence of high-blood pressure (HBP),13 ,14 which is becoming one of the principal reasons for consultation in PHC. According to the latest National Health Survey 2011–2012, a rise in chronic pathologies can be appreciated and 18.5% of the population suffers from hypertension.15 The chief objective of the treatment and control of HBP is to reduce cardiovascular morbimortality associated with HBP.16 ,17

    Regarding the impact on health of Primary Care nursing, in the review by Keleher et al,18 the results observed when the care was provided by nurses were similar to when such care was provided by physicians, except in relation to the acquisition of knowledge, adherence to treatment and user satisfaction, in which areas the results were more favourable for nurses. Recent works, which examine the nurse's role as leader in monitoring patients with HBP (with major responsibility, with possibility of prescribing or adjusting treatment in concordance with guidelines) as opposed to the traditional model (nurse as an assistant to the doctor), have shown better results in controlling HBP,19–21 and therefore an estimated reduction of up to 30% of the risk of cardiovascular events.19

    In view of the above and owing to the absence of studies on PHC, the aim of this paper was to determine how the characteristics of the PHC nursing workforce, and the working environment as perceived by this professional group, impact on the control of HBP of adult patients.

    Methods

    Design

    A transversal analytical study, conducted in 2010, based on data obtained from Primary Care information systems and through the validated questionnaire Practice Environment Scale Nursing Work Index (PES-NWI).22–24

    Subjects

    The original population was composed of all patients over 14 years diagnosed with HBP—128 193—at the 45 health centres in two health zones within the Community of Madrid, North-West Zone (NWZ), with a better socioeconomic situation, and South-West Zone (SWZ), with a worse socioeconomic situation; and the 507 nurses employed at these health centres and referred to the said patients. The following were excluded:

    Patients under 14 years of age (n=143), patients who had not registered any BP values in their clinic records in 2010 (n=44.295), those who had not identified the pair doctor–nurse (n=6.958). The patient population was finally established at 76 797. A further inclusion criterion was that nurses should have held their current post for at least 6 months (the final sample of nurses was 442).

    All analyses were made separately in each zone due to their different socioeconomic situation.

    Variables

    Dependent variable

    Poor HBP control (not adequate): patients were considered to be under adequate control when the systolic blood pressure (SBP) and diastolic blood pressure (DBP) (arithmetic mean of measurements in 1 year) were below the value 140/90 mm Hg.13

    Independent variables

    Patient variables: age; sex; economic immigration condition (to render this variable operable, the country of origin was taken into account; if this was without the Organisation for Economic Cooperation and Development, the patient was considered an economic immigrant). The variable time from diagnosis, despite its importance, could not be included due to the low reliability of these data in information systems.

    Nursing staff variables: age; sex; contract type (permanent or temporary post); professional experience and experience in the current post (years); educational level (bachelors, higher degree in nursing or studies in other disciplines); professional category (clinical nurse and nurse manager); users/nurse ratio of the general population and patients over 65 years of age; average nursing consultations per day; patients’ average nursing consultations per year; percentage of economic immigrant population within the assigned quota to each nurse; perception of the workplace climate through the questionnaire Practice Environment Scale Nursing Work Index (PES-NWI) validated for the Spanish environment and for Primary Care,22–24 which measures 31 items, grouped into five factors: Participation in Hospital affairs (maximum score 36), Nursing Foundations for Quality of Care (maximum score 40), Nurse Managers ability and support for nurses (maximum score 20), Staffing adequacy (maximum score 16), Nurse–physician Relationship (maximum score 12). The global score can range from 31 to 124. The procedure of questionnaires administration has been described previously.25

    Other study variables: users/physician ratio, proportion of patients with HBP whose hypertension is not measured at the centre (ie, they had not registered any BP values) (HBPnotM) (aggregate variable related to health centre), deprivation index (DI) per basic health zone (aggregate variable related to the geographic zone where the health centre is located), drawn from four basic indicators in the 2011 census: manual workers, unemployment, temporary personnel and total insufficient training. This index allows the population’s socioeconomic situation to be detected (the higher the score, the more unfavourable the situation). It has been built following the index methodology drawn up for the MEDEA project,26 ,27 and values were categorised in quartiles (Q1: most favourable socioeconomic situation).

    Data analysis

    First, a descriptive analysis was conducted (measure of central tendency and frequency distribution) of patient and nurse characteristics.

    Multilevel logistic regression models were adjusted to estimate the prevalence of poor HBP control and to identify the patient and nurse characteristics that are associated with poor HBP control. The multilevel models are particularly appropriate when individuals clustered within groups and these groups share characteristics. To take into account this hierarchical structure, random effects terms are introduced into the model to estimate the effect of the different levels. Fixed effect variables can also be included.

    In our models, the response variable was poor HBP control; the random effect variables were the health centre and the physician–nurse pair and the fixed effect variables were the patient and nurse characteristics as well as the group variables.

    The prevalence rates of poor HBP were estimated by means of adjusting a multilevel logistic regression model without fixed effects (called ‘empty model’) and previously mentioned random effects: health centre and physician–nurse pair. For each health zone, prevalence rates were reported by sex and immigration status, with 95% CIs.

    Subsequently, univariate multilevel regression models were adjusted for each fixed effect variable in order to explore its association with poor HBP control taking into account the hierarchical structure of the data. The variables with p<0.20 were considered statistically significant and included in the multivariate analysis.

    To select the final multilevel multivariate models, comparisons were made among models by means of the likelihood-ratio test. For each zone, the final multivariate model was also estimated for men and women separately, in order to understand the cultural construction of gender.28 ,29

    The random effect was quantified through the mean Median OR (MOR).30 MOR can be interpreted as the increase (in median) in the odds to poor control of a patient if he or she changes from one group to another of higher risk. The random effect of the centre on slope of the PES-NWI variable was analysed. CIs were calculated with a confidence of 95%, considering a significance level of 0.05. For this analysis, the STATA V.12 statistical package was used.

    Results

    Description of the population in the study

    A total of 76797 subjects were studied. The participants’ characteristics were: average age 65.9 years (SD 12.8); 55.4% were women; 4.9% were economic immigrants. 32.2% (95% CI 31.8% to 32.5%) had poor HBP control. Nurses: average age was 47.1 years (SD 10.3); 83% were women; average time in post was from 8.3 (SD 7.5) to 24 years (SD 10.5) regarding professional practice. In total 88.5% were clinical nurses; 76% had a permanent post and 82.2% held a bachelor’s degree ; the PES-NWI was responded to by 268 nurses (response rate: 60.6%); the average PES-NWI score was 81.2 (95% CI 79.2 to 8.4) ; the average users/nurse ratio was 2213.2 (SD 548.5); the ratio of patients over 65 years was 280.6 (SD 128.9); the average percentage of immigrant population per nurse was 8.5% (95% CI 7.9% to 9.1%); the average number of patient consultations per day was 16.18 (SD 3.0); the annual average number of consultations per patient with HBP was 7.3 (SD 7.0). We may highlight that in the SWZ there is no population in the Q1 in a better socioeconomic situation (table 1).

    View this table:
    Table 1

    Sociodemographic characteristics of nurses and patients

    Control of HBP

    Prevalence of poor HBP control

    The prevalence of poor HBP control among the total population, estimated by means of a multilevel empty model, was 31.5% (95% CI 31.5% to 35.6%). By sex, 34.5% (95% CI 32.5% to 36.6%) of native men had poor control against 43.4% (95% CI 40.7% to 46.2%) of immigrant men and 31.4% (95% CI 29.5% to 33.4%) of native women against 39.9% (95% CI 37.4% to 42.7%) of immigrant women; all prevalences were calculated taking into account centre and physician–nurse pair as random effect variables. With regard to random effect variables, the centre variable obtained an MOR of 1.29, or 29% more (on average) of poor control if the patient changed centre, and for the physician–nurse pair variable an MOR of 1.51 was obtained, that is, 51% more of poor control if the patient changed pair, in both health zones (table 2).

    View this table:
    Table 2

    Prevalence of poor high-blood pressure (HBP) Control according to place of origin, sex and health zone

    Multilevel univariate analysis

    In the multilevel univariate analysis, in the NWZ, poor HBP control decreased significantly (p≤0.05): with age; in women; when the nurse held a temporary contract; with the nurse’s favourable perception of the environment; with higher rates of nursing consultations per day; with higher rates of patient consultations per year. However, the condition of being an economic immigrant; nurse’s professional experience exceeding 20 years; being a nurse manager; older nurses; a higher ratio of patients over the age of 65 and a higher proportion of HBPnotM patients, all contributed to a significant increase (p≤0.05) in poor HBP control. In the SWZ: age; being a woman; higher general population ratio; higher PES-NWI score; higher rate of patient consultations per year and a deprivation index in the quartile 3, significantly reduced (p≤0.05) poor HBP control. However, the condition of being an immigrant; nurse’s professional experience exceeding 20 years; older nurses; and a higher proportion of HBPnotM, all contributed to a significant increase (p≤0.05) in poor HBP control (table 3).

    View this table:
    Table 3

    OR of poor high-blood pressure control according to Univariate analysis by health Zone and Sex*

    Multilevel multivariate models

    In the multilevel multivariate model for men in NWZ, the variables that remained significant (p≤0.05), with a protective effect (reduction of poor HBP control), were: age of the patient (OR: 0.99; 95% CI 0.98 to 0.99); perceived working environment (OR: 0.99; 95% CI 0.98 to 0.99), average patient consultations per year (OR: 0.98; 95% CI 0.97 to 0.98), deprivation index (OR: 0.34; 95% CI 0.13 to 0.88). Nevertheless, the following showed a risk effect (increase in poor HBP control): being a nurse manager (OR: 1.41; 95% CI 1.007 to 1.46) and a higher proportion of HBPnotM patients (OR: 5.31; 95% CI 1.58 to 17.83). With respect to random effect variables, the physician–nurse pair obtained an MOR of 1.43, that is, on average, 43% more of poor control if the patient changed to another physician–nurse pair. In the case of women, the number of patient consultations per year had a protective effect (OR: 0.98; 95% CI 0.97 to 0.99) and age proved to have a risk effect (OR: 1.01; 95% CI 1.007 to 1.01), as did being an immigrant (OR: 1.53; 95% CI 1.24 to 1.89) and the patient being a nurse manager (OR: 1.35; 95% CI 0.97 to 1.88). The centre random effect variable obtained an MOR of 1.14, and the physician–nurse pair obtained an MOR of 1.41 (table 4).

    View this table:
    Table 4

    Poor high-blood pressure control according to multivariate multilevel logistic regression models in North-West Zone by sex*

    In the multilevel multivariate model for men in SWZ, a protective effect was shown by patient age (OR: 0.98; 95% CI 0.98 to 0.99); number of patient consultations per year (OR: 0.98; 95% CI 0.97 to 0.99), the deprivation index (OR: 0.60; 95% CI 0.40 to 0.90): and a risk effect was seen in being an immigrant (OR: 1.89; 95% CI 1.43 to 2.51) and having between 11 and 20 years’ work experience (OR: 1.65; 95% CI 1.42 to 2.62). MOR obtained from the physician–nurse pair variable was 1.45. As for women, a protective effect was gained from the nurses’ perception of the working environment (OR:0.98; 95% CI 0.98 to 0.99), patient average number of consultations per year (OR: 0.98; 95% CI 0.97 to 0.99), deprivation index (OR: 0.55; 95% CI 0.32 to 0.97); and a risk effect was caused by: patient age (OR: 1.005; 95% CI 1.001 to 1.009) and being an immigrant (OR: 1.39; 95% CI 1.09 to 1.76). MOR obtained for the centre and physician–nurse pair random variables was 1.15 and 1.62, respectively (table 5).

    View this table:
    Table 5

    Poor high-blood Pressure control according to multivariate multilevel logistic regression models in South-West Zone by sex*

    The analysis of centre random effect on slope of the PES-NWI variable was not statistically significant.

    Discussion and conclusions

    In total 33.54% of the HBP patient population had poor control. Prior studies have shown poorer degrees of control than in this work; it must be underlined that in these studies the general population is analysed and samples also include people with HBP who had not been diagnosed before and therefore had not been monitored.31 ,32

    Moreover, a working environment perceived as more favourable by nurses, and patients with HBP having more consultations with a nurse, are factors that contribute to keeping a better control of the disease. Working environment quality has been an important focus in health research, as there is evidence that it is directly or indirectly associated with professionals, patients and results within health organisations.4 ,9 ,11

    This study shows that when the average nursing consultation per year was increasing, the poor control of hypertension was decreasing. In the interpretation of these results, and taking into account the transversal nature of the study, we must consider the possibility of ‘inverse care’ consisting in giving more attention to those who demand more attention and who may be patients whose blood pressure data are better than others who are in need of closer attention and whose HBP data are poorer. Diverse clinical tests on the impact of nursing care on controlling hypertension in which good results are yielded20 ,33 may lead us to believe that the greater the number of consultations with the nurse, the greater the control over HBP figures, thanks to the nurse’s influence in changing the patient’s lifestyle. In any case, consolidation of these findings will require further longitudinal studies.

    Regarding other variables relating to the characteristics of nursing staff, the following proved significant in the multilevel univariate analysis: age, professional experience, contract type, professional category, workload (nursing consultations per day), users/nurse ratio for all users and users over 65. However, most of them lost their statistical significance in the multilevel multivariate analysis, including users/nurse ratio. These results do not coincide with other studies carried out in the hospital setting3–6 and the work performed in PHC by Griffiths,7 where the users/nurse ratios were far higher than in this work. Also, it is striking to note that the older age of nurses contributes to poorer control. This finding has been reported previously in other studies in Spain, in Hospital and Primary Heath Care.34 ,35 Senior staff reflect a poorer perception of its working environment, as well as a reduced ability to incorporate evidence into clinical practice. These contradictory results could be explained by the lack of a professional career which encourages nurses to make progress on their own development, and consequently a progressive professional exhaustion could contribute to this perception. Additionally, younger professionals are more likely to have received a pregrade education on evidence-based practice.

    The results presented here underline the impact of nurses on health and the effects of their surveillance activities and monitoring of chronic diseases such as hypertension, given that the time dedicated to each patient may prove essential to enabling therapeutic communication. Recent studies have shown the efficiency of a nurse-led clinic as opposed to usual care on reduction in HBP figures in the general population33 and the immigrant population suffering from HBP without control.20 Moreover, Clark et al19 conducted a systematic review and meta-analysis of trials on nursing intervention in the control of patients with HBP, in which they found that a nurse-led intervention reduces 3 mm Hg in SPB and 4 mm Hg in DBP, which in turn translated into a reduction of up to 30% in cardiovascular events.

    It must be highlighted that major differences exist in the prevalence of poor HBP control among men and women, both among natives and in the immigrant population, the highest levels occurring among men. Being a woman is associated with up to 14% lower risk in the SWZ (worse socioeconomic level). This is consistent with the findings from other works in the Spanish context,32 ,36 which imply the need to seek more effective actions to control hypertension in men, grounded on studies that identify the factors leading to poor hypertension control in men.

    The condition of being an economic immigrant, moreover, raises the likelihood of poor HBP control for men in the SWZ up to 89%. Aerny et al,37 who studied the health of the immigrant population in the Community of Madrid, pointed to HBP screening below the recommended standards, especially among men (16% less); likewise, the immigrant population with shorter terms of residence made less frequent visits to the health services than did the native population. Palacios-Soler et al38 studied the differences in the degree of control of HBP among the immigrant and native populations, and found that fewer diagnostic tests were conducted on immigrants and that the pharmacological treatment they received was inferior to that administered to the native population.

    Among the results from random effect variables, it is especially significant that an average increase is observed of poor HBP control, reaching up to 15% if the patients receive care at one or another health centre and up to 62% depending on the physician–nurse pair treating them. Multilevel analyses in health sciences are becoming increasingly more numerous as it is recognised that health results are conditioned by the variability in healthcare structures and in health professionals. Fuste and Rue39 applied multilevel analysis to the differences in the performance of preventive activities among PHC teams, and found that variability existed among these PHC teams that was not explained by individual characteristics, and that was associated in part with the workload of the team.

    One of the strengths of this study is that it includes socioeconomic variables such as the deprivation index or the percentage of the immigrant population per nurse, which undoubtedly has an impact on the care given to the population.40 This study shows that under worse socioeconomic conditions, the HBP figures are better; this finding is contrary to what is described in the literature, pointing to an increase in blood pressure to lower socioeconomic status.41 However, in our context, the frequency of consultations made in PHC is higher in manual workers or people with smaller incomes while people with high incomes tend to use private healthcare,42 ,43 which could explain the data. As limitations, should be noted that the cross-sectional and observational nature of the study, limits establishing causal relationships; however, the analytical approach, linked to the fact that the vast majority of independent variables are temporarily located prior to the outcome variable, minimises the aforementioned limitation. To collect the data from the questionnaire PES-NWI, it was an essential requisite that the nurse had a postoccupancy, with the same quota of patients, for more than 6 months. Since the questionnaire was administered in the period from June to October, we could guarantee, in part, that the findings could be due to the nurse's performance. In addition, it should be noted that the structure of the PHC system in Spain is characterised by the allocation of professionals, like the nurse, to each patient, ensuring continuity of care, and would, at least partially, attribute the impact of the professional on the patient. Anyway, future longitudinal analytical and experimental studies are necessary. On the other hand, the use of documentary sources can lead to information bias, although the current information systems, such as the ones used in this study, are demonstrating good validity and consistency to be used as a source of information on epidemiological studies.44 ,45 Finally, it may happen that the index of deprivation, which is not a single variable but is centre aggregated, could lead us into an ecological fallacy, so it requires further research to deepen this issue.

    The current situation requires that nurses move towards taking a leading role in PHC, owing to the increase in age and chronic conditions among the population. This study provides evidence of the roles played by Primary Care nurses and the impact they have on patients’ health. A working environment perceived by Primary Care nurses as more favourable contributes to better HBP control. It is fundamental to promote a favourable organisational climate in order to achieve better health results.

    Acknowledgments

    The authors acknowledge the help of Ana Gandarillas Grande, Medical Epidemiologist, Health Promotion and Prevention of Primary Care in the Community of Madrid, for providing data relating to the Deprivation Index of 2011 and for her contribution to the revision of the report. She did not receive any compensation.

    References

    1. ↵
      1. Solar O,
      2. Irwin A
      . A conceptual framework for action on the social determinants of health. Genove: WHO Comission on Social Determinants of Health, 2007.
    2. ↵
      1. Salido Moreno M,
      2. Gálvez Toro A
      . Calidad=Enfermera. ¿Hay relación causal? Evidentia 2007;4(17):15/07/2014. http://www.index-f.com/evidentia/sumarion17.php
    3. ↵
      1. Aiken L,
      2. Sloane D,
      3. Bruyned L, et al
      . Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study. Lancet 2014;383:1824–30. doi:10.1016/S0140-6736(13)62631-8
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. Aiken LH,
      2. Cimiotti JP,
      3. Sloane DM, et al
      . Effects of nurse staffing and nurse education on patient deaths in hospitals with different nurse work environments. Med Care 2011;49:1047–53. doi:10.1097/MLR.0b013e3182330b6e
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Manojlovich M,
      2. Antonakos CL,
      3. Ronis DL
      . Intensive care units, communication between nurses and physicians, and patients’ outcomes. Am J Crit Care 2009;18:21–30. doi:10.4037/ajcc2009353
      OpenUrlAbstract/FREE Full Text
    6. ↵
      1. Kane RL,
      2. Shamliyan TA,
      3. Mueller C, et al
      . The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care 2007;45:1195–204. doi:10.1097/MLR.0b013e3181468ca3
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      1. Griffiths P,
      2. Murrells T,
      3. Maben J, et al
      . Nurse staffing and quality of care in UK general practice: cross-sectional study using routinely collected data. Br J Gen Pract 2010;60:36–48. doi:10.3399/bjgp10X482086
      OpenUrlCrossRefPubMed
    8. ↵
      1. Griffiths P,
      2. Maben J,
      3. Murrells T
      . Organisational quality, nurse staffing and the quality of chronic disease management in primary care: observational study using routinely collected data. Int J Nurs Stud 2011;48:1199–210. doi:10.1016/j.ijnurstu.2011.03.011
      OpenUrlCrossRefPubMed
    9. ↵
      1. Aiken LH,
      2. Clarke SP,
      3. Sloane DM, et al
      . Effects of hospital care environment on patient mortality and nurse outcomes. J Nurs Adm 2008;38:223–9. doi:10.1097/01.NNA.0000312773.42352.d7
      OpenUrlCrossRefPubMedWeb of Science
    10. ↵
      1. Kazanjian A,
      2. Green C,
      3. Wong J, et al
      . Effect of the hospital nursing environment on patient mortality: a systematic review. J Health Serv Res Policy 2005;10:111–17. doi:10.1258/1355819053559100
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Kirwan M,
      2. Matthews A,
      3. Scott PA
      . The impact of the work environment of nurses on patient safety outcomes: a multi-level modelling approach. Int J Nurs Stud 2013;50:253–63. doi:10.1016/j.ijnurstu.2012.08.020
      OpenUrlCrossRefPubMed
    12. ↵
      1. Lake ET
      . Development of the practice environment scale of the Nursing Work Index. Res Nurs Health 2002;25:176–88. doi:10.1002/nur.10032
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Mancia G,
      2. Fagard R,
      3. Narkiewicz K, et al
      . 2013 ESH/ESC practice guidelines for the management of arterial hypertension. Blood Press 2014;23:3–16. doi:10.3109/08037051.2014.868629
      OpenUrlCrossRefPubMed
    14. ↵
      1. Kushner RF,
      2. Ryan DH
      . Assessment and lifestyle management of patients with obesity: clinical recommendations from systematic reviews. JAMA 2014;312:943–52. doi:10.1001/jama.2014.10432
      OpenUrlCrossRefPubMed
    15. ↵
      Instituto Nacional de estadística. Encuesta Nacional de Salud 2011–2012. 2013. http://www.ine.es/jaxi/tabla.do (accessed 11/2014, 2014).
    16. ↵
      1. Sicras-Mainar A,
      2. Velasco-Velasco S,
      3. Llopart-Lopez JR, et al
      . Relationship among the degree of control of arterial hypertension, comorbidity and costs in individuals over age 30 during 2006. Rev Esp Salud Publica 2008;82:315–22. doi:10.1590/S1135-57272008000300007
      OpenUrlPubMed
    17. ↵
      1. James PA,
      2. Oparil S,
      3. Carter BL, et al
      . 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507–20. doi:10.1001/jama.2013.284427
      OpenUrlCrossRefPubMedWeb of Science
    18. ↵
      1. Keleher H,
      2. Parker R,
      3. Abdulwadud O, et al
      . Systematic review of the effectiveness of primary care nursing. Int J Nurs Pract 2009;15:16–24. doi:10.1111/j.1440-172X.2008.01726.x
      OpenUrlCrossRefPubMed
    19. ↵
      1. Clark CE,
      2. Smith LF,
      3. Taylor RS, et al
      . Nurse led interventions to improve control of blood pressure in people with hypertension: systematic review and meta-analysis. BMJ 2010;341:c3995. doi:10.1136/bmj.c3995
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Hebert PL,
      2. Sisk JE,
      3. Tuzzio L, et al
      . Nurse-led disease management for hypertension control in a diverse urban community: a randomized trial. J Gen Intern Med 2012;27:630–9. doi:10.1007/s11606-011-1924-1
      OpenUrlCrossRefPubMed
    21. ↵
      1. Perez-Cuevas R,
      2. Reyes Morales H,
      3. Doubova SV, et al
      . Comprehensive diabetic and hypertensive patient care involving nurses working in family practice. Rev Panam Salud Publica 2009;26:511–17. doi:10.1590/S1020-49892009001200006
      OpenUrlPubMed
    22. ↵
      1. De Pedro-Gomez J,
      2. Morales-Asencio JM,
      3. Sese-Abad A, et al
      . Psychometric testing of the Spanish version of the practice environment scale of the nursing work index in a primary healthcare context. J Adv Nurs 2012;68:212–21. doi:10.1111/j.1365-2648.2011.05730.x
      OpenUrlCrossRefPubMed
    23. ↵
      1. de Pedro Gomez J,
      2. Morales-Asencio JM,
      3. Sese-Abad A, et al
      . Validación y adaptación al español de la escala del entorno de práctica enfermera del ursing work index. Metas de Enferm 2009;12:65–73.
      OpenUrl
    24. ↵
      1. Rivero Sánchez C,
      2. López Alonso SR
      . ¿Es válida la adaptación española de la Escala de Entorno de Práctica Enfermera del Nursing Work Index para Atención Primaria? Index de Enfermería 2014;23:188–90. doi:10.4321/S1132-12962014000200016
      OpenUrl
    25. ↵
      1. Parro Moreno A,
      2. Serrano Gallardo P,
      3. Ferrer Arnedo C, et al
      . Influence of socio-demographic, labour and professional factors on nursing perception concerning practice environment in Primary Health Care. Aten Primaria 2013;45:476–85. doi:10.1016/j.aprim.2012.12.015
      OpenUrl
    26. ↵
      1. Dominguez-Berjón MF,
      2. Borrel C,
      3. Cano-Serral G, et al
      . Construcción de un índice de privación a partir de datos censales en grandes ciudades españolas (Proyecto MEDEA). Gac Sanit 2008;22:179–87. doi:10.1157/13123961
      OpenUrlPubMedWeb of Science
    27. ↵
      1. Gandarillas A,
      2. Dominguez-Berjón MF,
      3. Soto MJ, et al
      . Desigualdades en mortalidad: II Atlas de mortalidad y desigualdades socio-económicas en la Comunidad de Madrid (2001–2007). Servicio Madrileño de Salud D G de Atención Primaria, 2012.
    28. ↵
      1. Kunkel SR,
      2. Atchley RC
      . Why gender matters: being female is not the same as not being male. Am J Prev Med 1996;12:294–6.
      OpenUrlPubMedWeb of Science
    29. ↵
      1. Ruiz-Cantero MT,
      2. Vives-Cases C,
      3. Artazcoz L, et al
      . A framework to analyse gender bias in epidemiological research. J Epidemiol Community Health 2007;61(Suppl 2):ii46–53. doi:10.1136/jech.2007.062034
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Merlo J,
      2. Chaix B,
      3. Ohlsson H, et al
      . A brief conceptual tutorial of multilevel analysis in social epidemiology: using measures of clustering in multilevel logistic regression to investigate contextual phenomena. J Epidemiol Community Health 2006;60:290–7. doi:10.1136/jech.2004.029454
      OpenUrlAbstract/FREE Full Text
    31. ↵
      1. Barrios V,
      2. Escobar C,
      3. Llisterri JL, et al
      . Blood pressure and lipid control and coronary risk in the hypertensive population attended in Primary Care setting in Spain. The PRESCOT study. Rev Clin Esp 2007;207:172–8. doi:10.1157/13101845
      OpenUrlPubMed
    32. ↵
      1. Ortiz Marron H,
      2. Vaamonde Martin RJ,
      3. Zorrilla Torras B, et al
      . Prevalence, degree of control and treatment of hypertension in the adult population of Madrid, Spain. Rev Esp Salud Publica 2011;85:329–38. doi:10.1590/S1135-57272011000400002
      OpenUrlPubMed
    33. ↵
      1. Dean SC,
      2. Kerry SM,
      3. Khong TK, et al
      . Evaluation of a specialist nurse-led hypertension clinic with consultant backup in two inner city general practices: randomized controlled trial. Fam Pract 2014;31:172–9. doi:10.1093/fampra/cmt074
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. de Pedro-Gomez J,
      2. Morales-Asencio JM,
      3. Sese Abad A, et al
      . Nursing practice settings and competence to incorporate evidence into decisions: analysis of the situation in the Balearic Islands (Spain). Gac Sanit 2011;25:191–7. doi:10.1016/j.gaceta.2010.11.007
      OpenUrlPubMed
    35. ↵
      1. Gonzalez-Torrente S,
      2. Pericas-Beltran J,
      3. Bennasar-Veny M, et al
      . Perception of evidence-based practice and the professional environment of primary health care nurses in the Spanish context: a cross-sectional study. BMC Health Serv Res 2012;12:227. doi:10.1186/1472-6963-12-227
      OpenUrlPubMed
    36. ↵
      1. Abellán-Alemán J,
      2. División-Garrote J,
      3. Prieto-Díaz M, et al
      . Tratamiento y grado de control de los hipertensos diabéticos atendidos en atención primaria en España. Estudio Brand I. Hipertens riesgo vasc 2010;27:195–202. doi:10.1016/j.hipert.2010.04.002
      OpenUrl
    37. ↵
      1. Aerny-Perreten N,
      2. Ramasco-Gutierrez M,
      3. Cruz-Maceín J, et al
      . La salud y sus determinantes en la población inmigrante de la Comunidad de Madrid. Gac Sanit 2010;24:136–44. doi:10.1016/j.gaceta.2009.10.007
      OpenUrlCrossRefPubMed
    38. ↵
      1. Palacios Soler L,
      2. Camps Vila L,
      3. Fabregas Escurriola M, et al
      . Follow-up and monitoring of hypertension. Is this performed the same in the autochthonous and immigrant populations? Aten Primaria 2007;39:651–4. doi:10.1157/13113958
      OpenUrlPubMed
    39. ↵
      1. Fuste J,
      2. Rue M
      . Variability in preventive activities among primary care teams in Catalonia. Application of a multilevel analysis. Gac Sanit 2001;15:118–27. doi:10.1016/S0213-9111(01)71531-2
      OpenUrlPubMed
    40. ↵
      1. Rechel B,
      2. Mladovsky P,
      3. Ingleby D, et al
      . Migration and health in an increasingly diverse Europe. Lancet 2013;381:1235–45. doi:10.1016/S0140-6736(12)62086-8
      OpenUrlCrossRefPubMedWeb of Science
    41. ↵
      1. Chaix B,
      2. Bean K,
      3. Leal C, et al
      . Individual/neighborhood social factors and blood pressure in the RECORD Cohort Study: which risk factors explain the associations? Hypertension 2010;55:769–75. doi:10.1161/HYPERTENSIONAHA.109.143206
      OpenUrlCrossRef
    42. ↵
      Observatorio de Salud de la Mujer. Ministerio de Salud. Desigualdades sociales en la salud, los estilos de vida y la utilización de los recursos sanitarios en las comunidades autónomas 1993–2003. 2006.
    43. ↵
      Consejería de Sanidad de la Comunidad de Madrid. Informe sobre desigualdades en salud a partir de la Encuesta Regional de Salud de Madrid de 2007. 2010.
    44. ↵
      1. Gil Montalban E,
      2. Ortiz Marron H,
      3. Lopez-Gay Lucio-Villegas D, et al
      . Validity and concordance of electronic health records in primary care (AP-Madrid) for surveillance of diabetes mellitus. PREDIMERC study. Gac Sanit 2014;28:393–6. doi:10.1016/j.gaceta.2014.03.001
      OpenUrl
    45. ↵
      1. Tannen RL,
      2. Weiner MG,
      3. Xie D
      . Use of primary care electronic medical record database in drug efficacy research on cardiovascular outcomes: comparison of database and randomised controlled trial findings. BMJ 2009;338:b81. doi:10.1136/bmj.b81
      OpenUrlAbstract/FREE Full Text
    View Abstract

    Footnotes

    • Contributors AP-M, PS-G and JMM-A were responsible for study concept and design. AD-H, JLA-T, VA, IMS-P, PS-G, AP-M and JMM-A were responsible for acquisition, analysis or interpretation of data. AP-M, PS-G and JMM-A were responsible for drafting of the manuscript. All authors participated in critical revision of the manuscript for important intellectual content. VA, IMS-P, PS-G and AP-M took part in statistical analysis. AD-H, JLA-T, IMS-P and VA were responsible for administrative, technical or material support. AP-M and PS-G had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    • Funding The results presented here form part of a study that has been funded partially with the First Prize for National Research in Nursing (12th edition) from Hospital Universitario Marqués de Valdecilla (Santander) in 2010.

    • Competing interests None declared.

    • Ethics approval Clinical Research Ethical Committee of University Hospital Puerta de Hierro Majadahonda (Madrid).

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement No additional data are available.