Table 1 Descriptions and summarised appraisal of included studies.
| Study | Participants Location Design | Aims (where possible verbatim) Appraisal of methods | Infant age Participation rate Attributable quotations | Setting Sampling Triangulation used | Key findings |
|---|---|---|---|---|---|
| Baughcum et al,15 1998 | 16 dieticians, 6 WIC mothers, 8 teenage WIC mothers | “To identify maternal beliefs and practices about child feeding that are associated with the development of childhood obesity” | 12–36 months old | WIC clinic and WIC nurses | Mothers were not concerned about overweight in their children. This was perceived as a problem by dieticians and study authors |
| Kentucky, USA | Design allowed for exploration of subjective experience | Not stated for mothers, 95% for dieticians | Risk of bias as sampling restricted to health clinic users | ||
| Qualitative (focus groups) | not attributable to individuals | Not stated | |||
| Baughcum et al,16 2001 | 454 mothers, 258 attending WIC and 196 attending private child health clinics | “To determine if the factor scores [from questionnaire under development] were associated cross‐sectionally with (1) the child being overweight at the time of the survey (2) maternal obesity, and (3) lower socio‐economic status.” | 11–24 months, but considering retrospectively to first year | Health clinics (WIC or private) | Mothers were more concerned about under eating and underweight, although where children were overweight there was concern about overeating and overweight |
| Cincinnati and Kentucky, USA | Design did not allow exploration of subjective views | 98% | Risk of bias as sampling restricted to health clinic users | ||
| Quantitative attitudinal (closed questionnaire) | Not attributable to individuals | Not stated | |||
| Birgeneau,17 2001 | 229 adults (129 women, 100 men) | Several hypotheses described in Chapter 2, summarised in abstract as “the present study examined adult weight‐related biases applied to infants in a sample of . . . adults” | Sitting infants | Not clear, but sample largely drawn from university population | Fatter infants rated as less attractive and less sociable than average weight infants |
| Boston, USA | Design did not allow subjective views | Not stated | Reporting did not give adequate details | ||
| Quantitative attitudinal (Likert scales) | Not attributable to individuals | Not stated | |||
| Brown,18 1981 | 93 mothers | “To investigate parental concerns about infant physical and behaviour development during first 9 months of life” | 0–9 months | University of California at Los Angeles newborn nursery plus other hospitals | 5/20 parents of full‐term babies recorded growth as concern across all ages. Physical growth was an increasing concern across time for parents of preterm infants |
| Los Angeles, USA | Categorisations of interview data. Relatively little exploration of subjective experience | Not stated | Not stated | ||
| Quantitative data | Not attributable to individuals | Not stated | |||
| Hall et al,19 2002 | 60 mothers | “To compare differences in maternal confidence and competence in mothers of preterm infants who were weighed before and after breastfeeding throughout hospitalisation and those who were not weighed with breastfeeds” | Postpartum—4 weeks after discharge. | NICU | Test weighing during hospitalisation did not affect maternal competence or confidence in a short‐term intervention, but both confidence and competence increased over time for both groups. Important influences from their social network were noted |
| Canada | Design did not allow exploration of subjective views | 82% (7 lost to follow‐up, 4 excluded) | Sampling appropriate for this population | ||
| Quantitative attitudinal | Not attributable to individuals | Not stated | |||
| Hewatt and Ellis,20 1986 | 40 mothers who had breast fed for at least 2 days | To explore “women's perceptions of their breastfeeding experiences” | 9–13 months | Home based, after delivery in single maternity hospital | Differences between mothers who breastfed for shorter and longer duration were noted; those who breastfed for longer were more likely to disregard health professionals comments about underweight |
| Vancouver, Canada | Design allowed exploration of subjective experience | 80% | Restricted to those who intended to breast feed, and criteria used for purposive element not clear | ||
| Qualitative (interviews) | Not attributable to individuals | Three interviewers used, who met regularly. Background of interviewers unknown and use of meetings not stated | |||
| Kramer et al,21 1983 | 50 mothers | To determine the importance of factors affecting obesity in the first 2 years of life | Newborn | Single hospital maternity ward | Older mothers and those with higher socio‐economic status were more likely to prefer leaner infants. No associations were found between preferences for body sizes and attitudes to feeding |
| Montreal, Canada | Design allowed limited exploration of subjective experience | 100% | Sampling not described in full, potential bias as consecutive births of full term, healthy hospital‐delivered infants | ||
| Quantitative (psychometric assessment tool) | Not attributable to individuals | Results from two questionnaires (both numerical) were compared | |||
| May,22 1997 | 14 mothers of infants admitted to NICU | “What is the process mothers use to seek help when providing care to low birth weight infants at home?” | Mean 9 months, range 4–12 months | Home based, after admission to NICU | Mothers were concerned that babies should be ‘normal', compared with other babies, and monitored for signs of progress |
| USA | Design allowed exploration of subjective experience | Not stated | Good sampling for this population | ||
| Qualitative (semistructured interviews) | Identified by interview number | Triangulation used as well as inquiry audit, peer debriefing and reflexive journal | |||
| McCann et al,23 1994 | 26 mothers of children with poor weight gain | “To determine whether disturbed eating habits and attitudes to body shape, weight, and food are more common among mothers of children with non‐organic failure to thrive than among mothers from the general population.” | Mean age 3.8, range 0.9–9.6 years considering previous experience | Purposive sample from one region of UK | Mothers were concerned about possibility of underlying organic causes of poor growth. Some mothers restricted intake of food they viewed as unhealthy |
| Oxfordshire, UK | Design may allow exploration of subjective experience, but interview methodology not clear | Not stated | Recruitment methods not stated | ||
| Quantitative (psychometric assessment tool) and qualitative (interview) | Not attributable to individuals | Not stated | |||
| Pridham*,24 1984 | 62 mothers | “To explore mothers' specification and description of issues on which they are working and the nature of the infant care task and the types of help and stress/support they experienced.” | 0–90 days | Single health clinic | Data reduced substantially in reporting, eg, 10 000 logged problems reduced to 7 categories. Mothers sought help from books, relatives, partners and friends. Most common problems were never mentioned to a healthcare professional |
| Madison, USA | Categorisations of participant's diary and log. Little exploration of subjective experience | Not stated | Sample restricted to participants at university department teaching clinic | ||
| Quantitative | Not attributable to individuals | Not stated | |||
| Pridham*,24 1984 | 22 mothers | “To examine the types of goals and decision‐making rules that mothers applied to 2 simulated infant care problems.” | 0–90 days | Single health clinic | Grandparents were the most commonly used source of information. Mothers' problem solving behaviour conformed to other models of problem solving by lay people |
| Madison, USA | Categorisations of telephone interview. Little exploration of subjective experience | Not stated | Sample restricted to participants at university department teaching clinic | ||
| Quantitative | Not attributable to individuals | Not stated | |||
| Rand and Wright†,13,14 2000, 2001 | 1317 (303 elementary school children; 427 high school adolescents; 261 university students; 326 middle‐age adults) | Paper one: an “evaluation of ideal and acceptable body sizes across a wide subject age span” | Sitting up babies | Varied settings in one region of USA | Preference for mid‐range body size across groups. A wider range of sizes was seen as acceptable in infants than in older children |
| Appalachia, USA | Paper two: “to examine the possibility of a more restrictive standard for females than males across a wide age span” | Not stated | Recruitment not described, except that university students gained credit points for taking part | ||
| Quantitative attitudinal. (Likert scales) | Design did not allow exploration of subjective views | Not attributable to individuals | Some triangulation, in that different groups responses to the same question are compared | ||
| Rajan and Oakley,25 1990 | 467 mothers | “We are concerned with . . . how the women themselves felt abut having a small baby, why they thought it happened, and how it had affected their life” | 6 weeks | Participants own homes | Different experiences presented. “Two thirds of all women did not see low birth weight as a problem in itself”, instead many thought that “prematurity, rather than low birth weight alone, was the major problem”. Where support was available to mothers from family or health professionals it was seen as vital |
| Midlands and South of England, UK | Qualitative (postal questionnaire with some open‐ended questions). Design allows exploration of subjective experience | 94% | Sample taken from previous study, of new mothers who had previously had a low birthweight baby | ||
| Qualitative (postal questionnaire with some open ended questions) | Yes, brief descriptions of participants | Somewhat. Statistical and some qualitative data (generated from open‐ended questionnaire) presented, and previous study referred to, but not different points of view | |||
| Reifsinder et al,26 2000 | 22 low‐income mothers (WIC) (13 with previous children with faltering growth) | “What are the explanatory models of growth held by mothers of growth‐deficient children?” | Age range 22.5–51 months, considering growth in previous 2 years | Data collection took place in participants own homes, but all data collected by nurses | Mothers were concerned about growth and assessed it through growth measurements, clothing sizes and comparisons to other children. Although size and growth were important, mothers viewed them as “natural phenomena” that were unproblematic when other factors (feeding, care taking, inherited characteristics) were accounted for |
| Texas, USA | Design allows exploration of subjective experience | 56% | Participants recruited from previous research with WIC mothers. | ||
| Qualitative (interviews) | Not attributable to individuals | Not stated | |||
| Sherratt et al,27 1991 | 228 mothers (113 of high‐risk infants, 115 low‐risk infants) | “Sought parents' views on the available services and compared the concerns of parents of six‐month old infants who had been at high risk neonatally with those of parents whose infants did not have these risk factors” | 6 months | Participants own homes | Lack of concern about growth patterns and overall developmental progress in both groups. However, child health clinics were ‘overwhelmingly perceived as places to weigh babies'. No tests of significance performed. |
| Buckinghamshire, UK | Design offered limited exploration of subjective views | 88% | Recruited from hospital nursery and health visitor records, appropriate for sample of health service users | ||
| Quantitative attitudinal (questionnaire, closed questions) | Not attributable to individuals | Not stated | |||
| Smith,28 1989 | 41 parents (possibly all mothers) 19 primaparae, 22 multiparae. | “The purpose of this study was to determine the major concerns of primiparae and multiparae 1 month post‐delivery and the resources used in meeting identified concerns.” | Contacted 4 weeks postpartum but return dates not stated | Single large maternity hospital | 32 categories of concerns, in which growth and development came 15th, with 7/19 primaparae and 6/22 multiparae including growth among concerns. Sources of help identified; partners, pamphlets, books, help from health professionals more common in primaparae |
| Vancouver, Canada | Design likely to offer limited exploration of subjective views | 68% for postal survey, not stated for recruitment by ward. | Appropriate to recruit English speaking mothers of healthy babies after spontaneous vaginal delivery in hospital | ||
| Quantitative attitudinal (questionnaire) | Not attributable to individuals | Not stated | |||
| Sturm et al,29 1997 | 132 mothers (50 with babies with faltering growth) | “Tested for defensive bias in mothers' causal attributions for infant (2–12.5 months) growth deficiency” | 2–12.5 months | Participants at health clinic or outpatients. Data collection took place in waiting rooms which may have created bias in responses | Findings difficult to use because responses to individual questions were not reported. Mothers seemed to place most weight on medical explanations of their child's growth |
| Location of study not stated, assumed USA | Design did not allow exploration of subjective views | Not stated | Restricted to users of healthcare services | ||
| Quantitative attitudinal. (Likert scales) | Not attributable to individuals | Some, inasmuch as different groups of mothers to same questions, but no differences in interpretation of results. | |||
| Thomlinson,30 2002 | 12 families (including 21 participants: 11 mothers, 1 stepmother, 6 fathers, 3 grandmothers). | To explore the experience of families with children who were failing to thrive; aim was not to generalise but to generate a rich description of the phenomenon of living with children who were not growing as expected | Not stated, asked for retrospective account | Participants' homes | Anxiety among family members when children had very poor growth strongly linked to anxiety about the health of the child |
| Canada | Design allows exploration of subjective experience | Not stated | Sampling appropriate for this population | ||
| Qualitative (interviews) | Not attributable to individuals | Yes, different family members were spoken to, although no input from other researchers/analysts | |||
| Vehvilainen‐Julkunen,31 1994 | 263 Public health nurses working in primary care and 323 of their clients | “The following two question were addressed | Age not stated, but still under care of nurses after birth | Very wide geographical area | Nurses interpreted function of home visits as providing support and encouragement for parents. For parents the most important function of the home visits related to the examination of the newborn including weighing |
| Finland | How do public health nurses and clients describe the options they have with regard to home visits? | 87% of eligible nurses, not stated for patients | Random sample of all public health nurses working in prenatal and child welfare clinics. Nurses sent out questionnaires to last three clients, which may have introduced selection bias | ||
| Quantitative attitudinal (questionnaire) | How do public health nurses and clients describe the functions and meanings of home visits?” | Not attributable to individuals | Not in terms of analysis or interpretation, although client and nurse views compared | ||
| Design likely to offer limited exploration of subjective views |
NICU, neonatal intensive care unit; WIC, Special Supplemental Nutrition Program for Women, Infants and Children.
*Two studies were presented in this paper and are described separately here.
†A single study was described across these papers.