Emanuela Sala

Correlates of obtaining informed consent to data linkage: respondent, interview and interviewer characteristics

In the United Kingdom, in order to link individual-level administrative records to survey responses, respondents need to give their consent. Using an unprecedented set of respondent, interview, and interviewer characteristics derived from the British Household Panel Survey (BHPS) matched with an interviewer survey, this research investigates which characteristics influence consent to adding health and social security records to the survey responses. We find that consent is related to respondents’ attitudes to privacy, community mindedness and data linkage salience as well as to some interview features such as the “household contagion effect” and the survey “fidelity.” Interviewer characteristics, including their personality, attitudes to persuading respondents, and job experience, are not associated with consent. Interviewers’ survey experience in the current wave and their task-specific experience, however, do influence consent. Implications of the findings are discussed and areas for future research are identified.

Older carers in the UK: are there really gender differences? New analysis of the Individual Sample of Anonymised Records from the 2001 UK Census

The aim of this paper is to disentangle the role of gender and partnership status in the caring commitments of older people (age 65 and over). Logistic and interval regression models are applied to individual records from the 2001 UK Census to estimate: (1) the impact of gender on the likelihood of being a career; (2) the impact of gender on the hours of care provided; and (3) the impact of gender on the likelihood of being a career for different groups defined by marital status. In the general population the share of women who provide care is higher than the corresponding share of men, but men have a higher probability of being carers among people aged 65 or above. This phenomenon is largely explained by gender differences in marital status. As older men are more likely to be married, and married people are more likely to be carers, we observe higher levels of caring among older men. Once differences in marital status are accounted for, the relationship between gender and care provision among older people is overturned. In particular, we find that, without controlling for household size, limiting long-term illness or marital status, the odds of being an informal career are lower for older women than men [odds ratio (OR): 0.85; 95% confidence interval (CI): 0.83-0.87]. Once these factors are accounted for, older women have higher odds of caring than older men (OR: 1.12; 95% CI: 1.09-1.15). Restricting the sample to care providers, and controlling for the same factors, it is shown that older women supply on average 3.77 (95% CI: 3.14-4.40) more hours of care per week than older men. Gender differences in the provision of care among older people disappear only when considering married individuals and adjusting for the presence of other household residents affected by a limiting long-term illness.

Etiology of acute otitis media in human immunodeficiency virus-infected children

BACKGROUND Acute otitis media (AOM) is one of the most common infections that are implicated as significant contributors to morbidity in HIV-infected children. To establish the optimal antibiotic therapy tympanocentesis is indicated as the first line diagnostic procedure, because unusual pathogens may play a role in advanced stages of deficient humoral or cellular immunity. METHODS The microbiology of 60 episodes of AOM diagnosed in 21 symptomatic HIV-infected children (ages 9 months to 12 years) was compared with that of 121 episodes of AOM occurring in 113 immunocompetent HIV-negative children (ages 6 months to 12 years) in the last 5 years. RESULTS The prevalence of the three most common pathogens (Streptococcus pneumoniae, Haemophilus influenzae and group A beta-hemolytic Streptococcus) was similar in HIV-infected and in normal children (56.5% vs. 54.9% of the ears). Staphylococcus aureus was significantly more frequent in AOM diagnosed in severely immunosuppressed stages. A significantly lower proportion of middle ear effusions obtained in HIV-infected children yielded no bacteria compared with normal children. Beta-lactamase production among isolates of H. influenzae was a rare phenomenon, both in HIV-infected and in normal children. No penicillin-resistant S. pneumoniae was found. CONCLUSIONS In HIV-infected children with absent or moderate immunosuppression empiric antibiotic therapy should be based on the recommendations given for immunocompetent children of the same geographic area. In severe immunosuppressed stages, given the possible role of Staph. aureus, extended spectrum antibiotics should be considered.

Consenting to health record linkage: evidence from a multi-purpose longitudinal survey of a general population

BackgroundThe British Household Panel Survey (BHPS) is the first long-running UK longitudinal survey with a non-medical focus and a sample covering the whole age range to have asked for permission to link to a range of administrative health records. This study determines whether informed consent led to selection bias and reflects on the value of the BHPS linked with health records for epidemiological research.MethodsMultivariate logistical regression is used, with whether the respondent gave consent to data linkage or not as the dependent variable. Independent variables were entered as four blocks; (i) a set of standard demographics likely to be found in most health registration data, (ii) a broader set of socio-economic characteristics, (iii) a set of indicators of health conditions and (iv) information about the use of health services.ResultsParticipants aged 16-24, males and those living in England were more likely to consent. Consent is not biased with respect to socio-economic characteristics or health. Recent users of GP services are underrepresented among consenters.ConclusionsWhilst data could only be linked for a minority of BHPS participants, the BHPS offers a great range of information on peoples life histories, their attitudes and behaviours making it an invaluable source for epidemiological research.

Propensity to consent to data linkage: experimental evidence on the role of three survey design features in a UK longitudinal panel

When performing data linkage, survey respondents need to provide their informed consent. Since not all respondents agree to this request, the linked data-set will have fewer observations than the survey data-set alone and bias may be introduced. By focusing on the role that survey design features play in gaining respondents’ consent, this paper provides an innovative contribution to the studies in this field. Analysing experimental data collected in a nationally representative household panel survey of the British population, we find that interview features such as question format (dependent/independent questions) and placement of the consent question within the questionnaire have an impact on consent rates.

The feasibility of linking household survey and administrative record data: New evidence for Britain

Linkage of household survey responses with administrative data is increasingly on the agenda. Unique individual identifiers have clear benefits for making linkages but are also subject to problems of survey item non-response and measurement error. Our experimental study that linked survey responses to UK government agency records on benefits and tax credits elucidates this trade-off. We compare five linkage criteria: one based on a respon-dent-supplied National Insurance Number (NINO) and the other four using different combinations of sex, name, address and date of birth. As many linkages were made using non-NINO-based matches as were made using matches on NINO and the former were also relatively accurate when assessed in terms of false-positive and false-negative linkage rates. The potential returns from hierarchical and pooled matching are also examined.

It is time computers do clever things! The impact of dependent interviewing on interviewer burden

Despite the increased interest in assessing the effects of dependent interviewing (DI) on data quality, the impact of DI on interviewer burden and interviewer—respondent interaction remains currently unexplored and undocumented. Drawing on Japec’s work, this article attempts to evaluate the impact of DI on interviewer burden and describes the mechanisms by which DI affects perceived interviewer burden and ultimately data quality. We use a diverse set of data including qualitative data collected from a survey carried out in 2006 on the British Household Panel Survey (BHPS) interviewers. Three main findings stand out: DI is usually welcomed by interviewers; under condition of no change in respondent circumstances, DI eases interviewer burden; the mechanisms by which DI eases interviewer burden are complex as interviewer burden factors are often strictly related to respondent burden; and there are strong interrelationships among the different factors that constitute interviewer burden.

When Change Matters: An Analysis of Survey Interaction in Dependent Interviewing on the British Household Panel Study

The authors examine how questionnaire structure affects survey interaction in the context of dependent interviewing (DI). DI is widely used in panel surveys to reduce observed spurious change in respondent circumstances. Although a growing literature generally finds beneficial measurement properties, little is known about how DI functions in interviews. The authors systematically observed survey interaction using behavior coding and analyzed an application of DI to obtain respondent employment characteristics. The authors found respondents indicated change in circumstances through a number of verbal machinations, including mismatch answers and explanations. Assessing whether these behaviors influenced subsequent question administration, the authors found qualitative evidence that the information disclosed when negating a DI question leads to subsequent interviewing errors. Quantitative analyses supported this evidence, suggesting that standardized interviewing deteriorates as respondents struggle to identify change in their circumstances. This analysis suggests that the reliability of detail about changed circumstances may not be improved using DI.

Personalized SMS, Survey Participation and Data quality – The Italian Case:

A number of papers have found that personalization of salutations in e-mail communications may lead to higher response rates. Given their widespread use, Short Text Messages (SMS) may also be useful contact modes in surveys. Using experimental data from an online survey on Italian graduates, this work intends to evaluate the impact of personalized SMS salutations on response and measurement error. We find evidence that personalization has a positive impact on response rates and some indication that it may lead to better survey data. Limitations and implications of the study are also discussed.